[llvm] [AMDGPU] Overload resource descriptor in image intrinsics. (PR #107255)
via llvm-commits
llvm-commits at lists.llvm.org
Thu Sep 26 10:40:50 PDT 2024
https://github.com/sstipano updated https://github.com/llvm/llvm-project/pull/107255
>From 88fd9229045a50f2b0e85ff3389c2b10b757fed2 Mon Sep 17 00:00:00 2001
From: Stefan Stipanovic <Stefan.Stipanovic at amd.com>
Date: Wed, 4 Sep 2024 17:13:49 +0200
Subject: [PATCH 1/3] [AMDGPU] Overload resource descriptor in image
intrinsics.
---
llvm/include/llvm/IR/IntrinsicsAMDGPU.td | 4 +-
.../AMDGPU/llvm.amdgcn.image.atomic.ll | 86 ++---
.../AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll | 300 ++++++++--------
...mdgcn-demanded-vector-elts-inseltpoison.ll | 180 +++++-----
.../AMDGPU/amdgcn-demanded-vector-elts.ll | 180 +++++-----
.../AMDGPU/amdgcn-intrinsics-gfx8.ll | 12 +-
.../InstCombine/AMDGPU/amdgcn-intrinsics.ll | 340 +++++++++---------
.../amdgcn-simplify-image-buffer-stores.ll | 30 +-
.../InstCombine/AMDGPU/image-d16.ll | 52 +--
llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll | 42 +--
10 files changed, 613 insertions(+), 613 deletions(-)
diff --git a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
index 4cd32a0502c66d..a03a92b5a97f78 100644
--- a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
+++ b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
@@ -862,8 +862,8 @@ class AMDGPUImageDimIntrinsic<AMDGPUDimProfile P_,
!foreach(arg, P_.DataArgs, arg.Type), // vdata(VGPR) -- for store/atomic
!if(P_.IsAtomic, [], [llvm_i32_ty]), // dmask(imm)
P_.AddrTypes, // vaddr(VGPR)
- [llvm_v8i32_ty], // rsrc(SGPR)
- !if(P_.IsSample, [llvm_v4i32_ty, // samp(SGPR)
+ [llvm_any_ty], // rsrc(SGPR); Valid types: v4i32 and v8i32
+ !if(P_.IsSample, [llvm_any_ty, // samp(SGPR);
llvm_i1_ty], []), // unorm(imm)
[llvm_i32_ty, // texfailctrl(imm; bit 0 = tfe, bit 1 = lwe)
llvm_i32_ty]), // auxiliary/cachepolicy(imm):
diff --git a/llvm/test/Analysis/UniformityAnalysis/AMDGPU/llvm.amdgcn.image.atomic.ll b/llvm/test/Analysis/UniformityAnalysis/AMDGPU/llvm.amdgcn.image.atomic.ll
index f6ac8fbd99c589..e2b743e5f514aa 100644
--- a/llvm/test/Analysis/UniformityAnalysis/AMDGPU/llvm.amdgcn.image.atomic.ll
+++ b/llvm/test/Analysis/UniformityAnalysis/AMDGPU/llvm.amdgcn.image.atomic.ll
@@ -1,131 +1,131 @@
; RUN: opt -mtriple amdgcn-mesa-mesa3d -passes='print<uniformity>' -disable-output %s 2>&1 | FileCheck %s
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(
define float @image_atomic_swap(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32.v8i32(
define float @image_atomic_add(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32.v8i32(
define float @image_atomic_sub(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32.v8i32(
define float @image_atomic_smin(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32.v8i32(
define float @image_atomic_umin(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32.v8i32(
define float @image_atomic_smax(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32.v8i32(
define float @image_atomic_umax(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32.v8i32(
define float @image_atomic_and(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32.v8i32(
define float @image_atomic_or(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32.v8i32(
define float @image_atomic_xor(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32.v8i32(
define float @image_atomic_inc(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32.v8i32(
define float @image_atomic_dec(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32.v8i32(i32 %data, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(
define float @image_atomic_cmpswap(<8 x i32> inreg %rsrc, i32 inreg %addr, i32 inreg %data, i32 inreg %cmp) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %data, i32 %cmp, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %data, i32 %cmp, i32 %addr, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32(
+;CHECK: DIVERGENT: %orig = call i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32.v8i32(
define float @image_atomic_add_2d(<8 x i32> inreg %rsrc, i32 inreg %s, i32 inreg %t, i32 inreg %data) #0 {
main_body:
- %orig = call i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32(i32 %data, i32 %s, i32 %t, <8 x i32> %rsrc, i32 0, i32 0)
+ %orig = call i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32.v8i32(i32 %data, i32 %s, i32 %t, <8 x i32> %rsrc, i32 0, i32 0)
%r = bitcast i32 %orig to float
ret float %r
}
-declare i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32(i32, i32, <8 x i32>, i32, i32) #0
-declare i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32, i32, i32, <8 x i32>, i32, i32) #0
-
-declare i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32(i32, i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.add.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.sub.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.smin.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.umin.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.smax.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.umax.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.and.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.or.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.xor.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.inc.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.dec.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32, i32, i32, <8 x i32>, i32, i32) #0
+
+declare i32 @llvm.amdgcn.image.atomic.add.2d.i32.i32.v8i32(i32, i32, i32, <8 x i32>, i32, i32) #0
attributes #0 = { nounwind }
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll
index 5ffdbb0f8c5b07..854a9ed4e01738 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll
@@ -7,10 +7,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1(<8 x i32> inreg %rsrc, i3
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0:[0-9]+]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -20,7 +20,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1(<8 x i32> inreg %rsrc, i3
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0:[0-9]+]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -47,10 +47,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask2(<8 x i32> inreg %rsrc, i3
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask2(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -60,7 +60,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask2(<8 x i32> inreg %rsrc, i3
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask2(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -87,10 +87,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask4(<8 x i32> inreg %rsrc, i3
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask4(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -100,7 +100,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask4(<8 x i32> inreg %rsrc, i3
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask4(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -127,10 +127,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask8(<8 x i32> inreg %rsrc, i3
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask8(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -140,7 +140,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask8(<8 x i32> inreg %rsrc, i3
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask8(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -167,10 +167,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_reverse(<8 x i32> inreg %rsrc, i
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_reverse(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -180,7 +180,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_reverse(<8 x i32> inreg %rsrc, i
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_reverse(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 3
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
@@ -208,10 +208,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_vaddr(<8 x i32> inreg %rsrc, i32
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_vaddr(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S0:%.*]], i32 [[T0:%.*]], i32 [[S1:%.*]], i32 [[T1:%.*]], i32 [[S2:%.*]], i32 [[T2:%.*]], i32 [[S3:%.*]], i32 [[T3:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S0]], i32 [[T0]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S1]], i32 [[T1]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S2]], i32 [[T2]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S3]], i32 [[T3]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S0]], i32 [[T0]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S1]], i32 [[T1]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S2]], i32 [[T2]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S3]], i32 [[T3]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -221,10 +221,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_vaddr(<8 x i32> inreg %rsrc, i32
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_vaddr(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S0:%.*]], i32 [[T0:%.*]], i32 [[S1:%.*]], i32 [[T1:%.*]], i32 [[S2:%.*]], i32 [[T2:%.*]], i32 [[S3:%.*]], i32 [[T3:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S0]], i32 [[T0]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S1]], i32 [[T1]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S2]], i32 [[T2]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S3]], i32 [[T3]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S0]], i32 [[T0]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S1]], i32 [[T1]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S2]], i32 [[T2]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S3]], i32 [[T3]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -247,14 +247,14 @@ define amdgpu_ps [8 x float] @load_2dmsaa_v8f32(<8 x i32> inreg %rsrc, i32 %s, i
; NO-MSAA-LABEL: define amdgpu_ps [8 x float] @load_2dmsaa_v8f32(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I4:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I5:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I6:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I7:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I4:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I5:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I6:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I7:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I8:%.*]] = insertvalue [8 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I9:%.*]] = insertvalue [8 x float] [[I8]], float [[I1]], 1
; NO-MSAA-NEXT: [[I10:%.*]] = insertvalue [8 x float] [[I9]], float [[I2]], 2
@@ -268,7 +268,7 @@ define amdgpu_ps [8 x float] @load_2dmsaa_v8f32(<8 x i32> inreg %rsrc, i32 %s, i
; MSAA-LABEL: define amdgpu_ps [8 x float] @load_2dmsaa_v8f32(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -311,20 +311,20 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_interleaved(<8 x i32> inreg %rsr
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_interleaved(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I1:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I3:%.*]] = insertvalue [4 x float] [[I1]], float [[I2]], 1
-; NO-MSAA-NEXT: [[I4:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I4:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I3]], float [[I4]], 2
-; NO-MSAA-NEXT: [[I6:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I6:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I7:%.*]] = insertvalue [4 x float] [[I5]], float [[I6]], 3
; NO-MSAA-NEXT: ret [4 x float] [[I7]]
;
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_interleaved(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I4:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -351,8 +351,8 @@ define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId01(<8 x i32> inreg %rsrc,
; NO-MSAA-LABEL: define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId01(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I2:%.*]] = insertvalue [2 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I3:%.*]] = insertvalue [2 x float] [[I2]], float [[I1]], 1
; NO-MSAA-NEXT: ret [2 x float] [[I3]]
@@ -360,7 +360,7 @@ define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId01(<8 x i32> inreg %rsrc,
; MSAA-LABEL: define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId01(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = insertvalue [2 x float] undef, float [[I]], 0
@@ -379,8 +379,8 @@ define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId23(<8 x i32> inreg %rsrc,
; NO-MSAA-LABEL: define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId23(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I2:%.*]] = insertvalue [2 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I3:%.*]] = insertvalue [2 x float] [[I2]], float [[I1]], 1
; NO-MSAA-NEXT: ret [2 x float] [[I3]]
@@ -388,7 +388,7 @@ define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId23(<8 x i32> inreg %rsrc,
; MSAA-LABEL: define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId23(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 3
; MSAA-NEXT: [[I2:%.*]] = insertvalue [2 x float] undef, float [[I]], 0
@@ -409,8 +409,8 @@ define amdgpu_ps [2 x <2 x float>] @load_2dmsaa_v2v2f32_dmask3(<8 x i32> inreg %
; NO-MSAA-LABEL: define amdgpu_ps [2 x <2 x float>] @load_2dmsaa_v2v2f32_dmask3(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [2 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [2 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: ret [2 x <2 x float>] [[I5]]
@@ -418,8 +418,8 @@ define amdgpu_ps [2 x <2 x float>] @load_2dmsaa_v2v2f32_dmask3(<8 x i32> inreg %
; MSAA-LABEL: define amdgpu_ps [2 x <2 x float>] @load_2dmsaa_v2v2f32_dmask3(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I4:%.*]] = insertvalue [2 x <2 x float>] undef, <2 x float> [[I]], 0
; MSAA-NEXT: [[I5:%.*]] = insertvalue [2 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; MSAA-NEXT: ret [2 x <2 x float>] [[I5]]
@@ -439,10 +439,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask3(<8 x i32> inreg %
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask3(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -452,8 +452,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask3(<8 x i32> inreg %
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask3(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -495,10 +495,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask5(<8 x i32> inreg %
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask5(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -508,8 +508,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask5(<8 x i32> inreg %
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask5(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -551,10 +551,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask6(<8 x i32> inreg %
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask6(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -564,8 +564,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask6(<8 x i32> inreg %
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask6(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -607,10 +607,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask9(<8 x i32> inreg %
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask9(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -620,8 +620,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask9(<8 x i32> inreg %
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask9(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -663,10 +663,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask10(<8 x i32> inreg
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask10(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -676,8 +676,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask10(<8 x i32> inreg
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask10(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -719,10 +719,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask12(<8 x i32> inreg
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask12(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -732,8 +732,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask12(<8 x i32> inreg
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask12(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 8, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -772,8 +772,8 @@ define amdgpu_ps [2 x half] @load_2dmsaa_v2f16_fragId01(<8 x i32> inreg %rsrc, i
; NO-MSAA-LABEL: define amdgpu_ps [2 x half] @load_2dmsaa_v2f16_fragId01(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I2:%.*]] = insertvalue [2 x half] undef, half [[I]], 0
; NO-MSAA-NEXT: [[I3:%.*]] = insertvalue [2 x half] [[I2]], half [[I1]], 1
; NO-MSAA-NEXT: ret [2 x half] [[I3]]
@@ -781,7 +781,7 @@ define amdgpu_ps [2 x half] @load_2dmsaa_v2f16_fragId01(<8 x i32> inreg %rsrc, i
; MSAA-LABEL: define amdgpu_ps [2 x half] @load_2dmsaa_v2f16_fragId01(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x half> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f16.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x half> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f16.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x half> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x half> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = insertvalue [2 x half] undef, half [[I]], 0
@@ -800,10 +800,10 @@ define amdgpu_ps [4 x float] @load_2darraymsaa_v4f32_dmask1(<8 x i32> inreg %rsr
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2darraymsaa_v4f32_dmask1(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SLICE:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -813,7 +813,7 @@ define amdgpu_ps [4 x float] @load_2darraymsaa_v4f32_dmask1(<8 x i32> inreg %rsr
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2darraymsaa_v4f32_dmask1(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SLICE:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2darraymsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2darraymsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -843,10 +843,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2darraymsaa_v4v2f32_dmask3(<8 x i32> in
; NO-MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2darraymsaa_v4v2f32_dmask3(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SLICE:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <2 x float>] undef, <2 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <2 x float>] [[I4]], <2 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <2 x float>] [[I5]], <2 x float> [[I2]], 2
@@ -856,8 +856,8 @@ define amdgpu_ps [4 x <2 x float>] @load_2darraymsaa_v4v2f32_dmask3(<8 x i32> in
; MSAA-LABEL: define amdgpu_ps [4 x <2 x float>] @load_2darraymsaa_v4v2f32_dmask3(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SLICE:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2darraymsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2darraymsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2darraymsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2darraymsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 [[SLICE]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP2:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP3:%.*]] = insertelement <2 x float> undef, float [[TMP2]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -901,10 +901,10 @@ define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7(<8 x i32> inreg %
; NO-MSAA-LABEL: define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <3 x float>] undef, <3 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <3 x float>] [[I4]], <3 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <3 x float>] [[I5]], <3 x float> [[I2]], 2
@@ -914,9 +914,9 @@ define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7(<8 x i32> inreg %
; MSAA-LABEL: define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP3:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = insertelement <3 x float> undef, float [[TMP3]], i64 0
; MSAA-NEXT: [[TMP5:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -968,10 +968,10 @@ define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7_group1(<8 x i32>
; NO-MSAA-LABEL: define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7_group1(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 5, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 6, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 [[S]], i32 [[T]], i32 7, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 5, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 6, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S]], i32 [[T]], i32 7, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x <3 x float>] undef, <3 x float> [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x <3 x float>] [[I4]], <3 x float> [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x <3 x float>] [[I5]], <3 x float> [[I2]], 2
@@ -981,9 +981,9 @@ define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7_group1(<8 x i32>
; MSAA-LABEL: define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7_group1(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 2, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 4, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 2, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 4, i32 [[S]], i32 [[T]], i32 4, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[TMP3:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[TMP4:%.*]] = insertelement <3 x float> undef, float [[TMP3]], i64 0
; MSAA-NEXT: [[TMP5:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
@@ -1030,11 +1030,11 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_sections(<8 x i32> inreg %rsrc,
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_sections(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], float [[VDATA:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: call void @llvm.amdgcn.image.store.2dmsaa.f32.i32(float [[VDATA]], i32 1, i32 [[S]], i32 [[T]], i32 [[FRAGID]], <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: call void @llvm.amdgcn.image.store.2dmsaa.f32.i32.v8i32(float [[VDATA]], i32 1, i32 [[S]], i32 [[T]], i32 [[FRAGID]], <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -1044,11 +1044,11 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_sections(<8 x i32> inreg %rsrc,
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_sections(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], float [[VDATA:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
-; MSAA-NEXT: call void @llvm.amdgcn.image.store.2dmsaa.f32.i32(float [[VDATA]], i32 1, i32 [[S]], i32 [[T]], i32 [[FRAGID]], <8 x i32> [[RSRC]], i32 0, i32 0)
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: call void @llvm.amdgcn.image.store.2dmsaa.f32.i32.v8i32(float [[VDATA]], i32 1, i32 [[S]], i32 [[T]], i32 [[FRAGID]], <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP1]], i64 2
; MSAA-NEXT: [[I3:%.*]] = extractelement <4 x float> [[TMP1]], i64 3
; MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
@@ -1074,10 +1074,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(<8 x i32> inreg %rsrc, i3
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[COND:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -1085,20 +1085,20 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(<8 x i32> inreg %rsrc, i3
; NO-MSAA-NEXT: [[I8:%.*]] = trunc i32 [[COND]] to i1
; NO-MSAA-NEXT: br i1 [[I8]], label [[IF_EQUAL:%.*]], label [[IF_UNEQUAL:%.*]]
; NO-MSAA: if_equal:
-; NO-MSAA-NEXT: [[I9:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I10:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I11:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I12:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I9:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I10:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I11:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I12:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I13:%.*]] = insertvalue [4 x float] undef, float [[I9]], 0
; NO-MSAA-NEXT: [[I14:%.*]] = insertvalue [4 x float] [[I13]], float [[I10]], 1
; NO-MSAA-NEXT: [[I15:%.*]] = insertvalue [4 x float] [[I14]], float [[I11]], 2
; NO-MSAA-NEXT: [[I16:%.*]] = insertvalue [4 x float] [[I15]], float [[I12]], 3
; NO-MSAA-NEXT: br label [[MERGE:%.*]]
; NO-MSAA: if_unequal:
-; NO-MSAA-NEXT: [[I17:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I18:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I19:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I20:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I17:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I18:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I19:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 2, <8 x i32> [[RSRC]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I20:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 3, <8 x i32> [[RSRC]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I21:%.*]] = insertvalue [4 x float] undef, float [[I17]], 0
; NO-MSAA-NEXT: [[I22:%.*]] = insertvalue [4 x float] [[I21]], float [[I18]], 1
; NO-MSAA-NEXT: [[I23:%.*]] = insertvalue [4 x float] [[I22]], float [[I19]], 2
@@ -1111,7 +1111,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(<8 x i32> inreg %rsrc, i3
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(
; MSAA-SAME: <8 x i32> inreg [[RSRC:%.*]], i32 [[S:%.*]], i32 [[T:%.*]], i32 [[COND:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP0]], i64 2
@@ -1123,7 +1123,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: [[I8:%.*]] = trunc i32 [[COND]] to i1
; MSAA-NEXT: br i1 [[I8]], label [[IF_EQUAL:%.*]], label [[IF_UNEQUAL:%.*]]
; MSAA: if_equal:
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I9:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
; MSAA-NEXT: [[I10:%.*]] = extractelement <4 x float> [[TMP1]], i64 1
; MSAA-NEXT: [[I11:%.*]] = extractelement <4 x float> [[TMP1]], i64 2
@@ -1134,7 +1134,7 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: [[I16:%.*]] = insertvalue [4 x float] [[I15]], float [[I12]], 3
; MSAA-NEXT: br label [[MERGE:%.*]]
; MSAA: if_unequal:
-; MSAA-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC]], i32 0, i32 0)
; MSAA-NEXT: [[I17:%.*]] = extractelement <4 x float> [[TMP2]], i64 0
; MSAA-NEXT: [[I18:%.*]] = extractelement <4 x float> [[TMP2]], i64 1
; MSAA-NEXT: [[I19:%.*]] = extractelement <4 x float> [[TMP2]], i64 2
@@ -1188,10 +1188,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1_different_rsrc(<8 x i32>
; NO-MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1_different_rsrc(
; NO-MSAA-SAME: <8 x i32> inreg [[RSRC1:%.*]], <8 x i32> inreg [[RSRC2:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; NO-MSAA-NEXT: main_body:
-; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC1]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC1]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC2]], i32 0, i32 0)
-; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC2]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC1]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I1:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC1]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I2:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC2]], i32 0, i32 0)
+; NO-MSAA-NEXT: [[I3:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 1, <8 x i32> [[RSRC2]], i32 0, i32 0)
; NO-MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
; NO-MSAA-NEXT: [[I5:%.*]] = insertvalue [4 x float] [[I4]], float [[I1]], 1
; NO-MSAA-NEXT: [[I6:%.*]] = insertvalue [4 x float] [[I5]], float [[I2]], 2
@@ -1201,10 +1201,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1_different_rsrc(<8 x i32>
; MSAA-LABEL: define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1_different_rsrc(
; MSAA-SAME: <8 x i32> inreg [[RSRC1:%.*]], <8 x i32> inreg [[RSRC2:%.*]], i32 [[S:%.*]], i32 [[T:%.*]]) #[[ATTR0]] {
; MSAA-NEXT: main_body:
-; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC1]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP0:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC1]], i32 0, i32 0)
; MSAA-NEXT: [[I:%.*]] = extractelement <4 x float> [[TMP0]], i64 0
; MSAA-NEXT: [[I1:%.*]] = extractelement <4 x float> [[TMP0]], i64 1
-; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC2]], i32 0, i32 0)
+; MSAA-NEXT: [[TMP1:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.2dmsaa.v4f32.i32.v8i32(i32 1, i32 [[S]], i32 [[T]], i32 0, <8 x i32> [[RSRC2]], i32 0, i32 0)
; MSAA-NEXT: [[I2:%.*]] = extractelement <4 x float> [[TMP1]], i64 0
; MSAA-NEXT: [[I3:%.*]] = extractelement <4 x float> [[TMP1]], i64 1
; MSAA-NEXT: [[I4:%.*]] = insertvalue [4 x float] undef, float [[I]], 0
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll
index 7e9d12241ee362..0a1e5dc054ad59 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll
@@ -4085,7 +4085,7 @@ declare <4 x i32> @llvm.amdgcn.struct.ptr.tbuffer.load.v4i32(ptr addrspace(8), i
define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4096,7 +4096,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32(float %vaddr, <8
; Check that the intrinsic remains unchanged in the presence of TFE or LWE
define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_tfe(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_1d_v4f32_f32_tfe(
-; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 1, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 1, i32 0)
; CHECK-NEXT: [[DATA_VEC:%.*]] = extractvalue { <4 x float>, i32 } [[DATA]], 0
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
@@ -4110,7 +4110,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_tfe(float %vaddr,
; Check that the intrinsic remains unchanged in the presence of TFE or LWE
define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_lwe(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_1d_v4f32_f32_lwe(
-; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 2, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 2, i32 0)
; CHECK-NEXT: [[DATA_VEC:%.*]] = extractvalue { <4 x float>, i32 } [[DATA]], 0
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
@@ -4123,7 +4123,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_lwe(float %vaddr,
define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %t, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.2d.f32.f32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.2d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4133,7 +4133,7 @@ define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %
define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float %s, float %t, float %r, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 0, float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -4144,7 +4144,7 @@ define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(float %s, float %slice, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1darray.f32.f32(i32 1, float [[S:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1darray.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 1, float %s, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4154,7 +4154,7 @@ define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(f
define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 2, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 2, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4164,7 +4164,7 @@ define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 4, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 4, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4174,7 +4174,7 @@ define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 8, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 8, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 8, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4184,7 +4184,7 @@ define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 9, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4194,7 +4194,7 @@ define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4204,7 +4204,7 @@ define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4214,7 +4214,7 @@ define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(float
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <2 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <2 x float> [[SHUF]]
;
@@ -4225,7 +4225,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4235,7 +4235,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4245,7 +4245,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4255,7 +4255,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <3 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
@@ -4266,7 +4266,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
@@ -4277,7 +4277,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
@@ -4288,7 +4288,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4298,7 +4298,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_1111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_1111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4318,7 +4318,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32, float, floa
define amdgpu_ps float @extract_elt1_image_sample_cl_2darray_v4f32_f32(float %s, float %t, float %slice, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_image_sample_cl_2darray_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.2darray.f32.f32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.2darray.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32 15, float %s, float %t, float %slice, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4334,7 +4334,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32, float, f
define amdgpu_ps float @extract_elt2_image_sample_d_cube_v4f32_f32_f32(float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt2_image_sample_d_cube_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cube.f32.f32.f32(i32 4, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cube.f32.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32 15, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4350,7 +4350,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32, float, f
define amdgpu_ps float @extract_elt3_image_sample_d_cl_1d_v4f32_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt3_image_sample_d_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.1d.f32.f32.f32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4366,7 +4366,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32, float,
define amdgpu_ps float @extract_elt1_dmask_0110_image_sample_l_1d_v2f32_f32(float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_dmask_0110_image_sample_l_1d_v2f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.1d.f32.f32(i32 4, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32 6, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4382,7 +4382,7 @@ declare <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32, float, float,
define amdgpu_ps float @extract_elt1_dmask_1001_image_sample_b_1d_v4f32_f32_f32(float %bias, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_dmask_1001_image_sample_b_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.1d.f32.f32.f32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 9, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4398,7 +4398,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32, float, flo
define amdgpu_ps <2 x float> @extract_elt1_elt2_dmask_1101_image_sample_b_cl_1d_v4f32_f32_f32(float %bias, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_elt2_dmask_1101_image_sample_b_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.b.cl.1d.v2f32.f32.f32(i32 12, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.b.cl.1d.v2f32.f32.f32.v8i32.v4i32(i32 12, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 13, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4414,7 +4414,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32, float,
define amdgpu_ps <2 x float> @extract_elt1_elt3_image_sample_lz_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_elt3_image_sample_lz_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.1d.v2f32.f32(i32 10, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.1d.v2f32.f32.v8i32.v4i32(i32 10, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4430,7 +4430,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32, float, <8 x i
define amdgpu_ps <3 x float> @extract_elt1_elt2_elt3_image_sample_cd_1d_v4f32_f32_f32(float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_elt2_elt3_image_sample_cd_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.cd.1d.v3f32.f32.f32(i32 14, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.cd.1d.v3f32.f32.f32.v8i32.v4i32(i32 14, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4446,7 +4446,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32, float, fl
define amdgpu_ps half @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4456,7 +4456,7 @@ define amdgpu_ps half @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
define amdgpu_ps half @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 4, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4466,7 +4466,7 @@ define amdgpu_ps half @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
define amdgpu_ps half @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 2, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 2, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4476,7 +4476,7 @@ define amdgpu_ps half @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
define amdgpu_ps <4 x half> @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v3f16.f32.f32(i32 7, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v3f16.f32.f32.v8i32.v4i32(i32 7, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[RES:%.*]] = shufflevector <3 x half> [[DATA]], <3 x half> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
@@ -4487,7 +4487,7 @@ define amdgpu_ps <4 x half> @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32
define amdgpu_ps <4 x half> @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v2f16.f32.f32(i32 3, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v2f16.f32.f32.v8i32.v4i32(i32 3, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[RES:%.*]] = shufflevector <2 x half> [[DATA]], <2 x half> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
@@ -4498,7 +4498,7 @@ define amdgpu_ps <4 x half> @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32
define amdgpu_ps <4 x half> @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[RES:%.*]] = insertelement <4 x half> poison, half [[DATA]], i64 0
; CHECK-NEXT: ret <4 x half> [[RES]]
;
@@ -4509,7 +4509,7 @@ define amdgpu_ps <4 x half> @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32
define amdgpu_ps half @extract_elt0_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4525,7 +4525,7 @@ declare <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32, float,
define amdgpu_ps float @extract_elt0_image_sample_c_1d_v4f32_f32(float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4541,7 +4541,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32, float, float,
define amdgpu_ps float @extract_elt0_image_sample_c_cl_1d_v4f32_f32(float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cl_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4557,7 +4557,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32, float, floa
define amdgpu_ps float @extract_elt0_image_sample_c_d_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4573,7 +4573,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32, float, f
define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4589,7 +4589,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32, float
define amdgpu_ps float @extract_elt0_image_sample_c_l_1d_v4f32_f32(float %zcompare, float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_l_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4605,7 +4605,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32, float, float
define amdgpu_ps float @extract_elt0_image_sample_c_b_1d_v4f32_f32_f32(float %bias, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.1d.f32.f32.f32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4621,7 +4621,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32, float, f
define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_1d_v4f32_f32_f32(float %bias, float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.1d.f32.f32.f32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4637,7 +4637,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32, float
define amdgpu_ps float @extract_elt0_image_sample_c_lz_1d_v4f32_f32(float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_lz_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4653,7 +4653,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32, float, floa
define amdgpu_ps float @extract_elt0_image_sample_c_cd_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4669,7 +4669,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32, float,
define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4685,7 +4685,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32, floa
define amdgpu_ps float @extract_elt0_image_sample_o_1d_v4f32_f32(i32 %offset, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4701,7 +4701,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32, i32, float, <8
define amdgpu_ps float @extract_elt0_image_sample_cl_o_1d_v4f32_f32(i32 %offset, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cl_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4717,7 +4717,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_d_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_d_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4733,7 +4733,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_sample_d_cl_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_d_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4749,7 +4749,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_l_o_1d_v4f32_f32(i32 %offset, float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_l_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4765,7 +4765,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_b_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_b_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4781,7 +4781,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_sample_b_cl_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_b_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4797,7 +4797,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_lz_o_1d_v4f32_f32(i32 %offset, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_lz_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.lz.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4813,7 +4813,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_cd_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cd_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4829,7 +4829,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32, i32, fl
define amdgpu_ps float @extract_elt0_image_sample_cd_cl_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cd_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4845,7 +4845,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_c_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4861,7 +4861,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_c_cl_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cl_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4877,7 +4877,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32, i32, floa
define amdgpu_ps float @extract_elt0_image_sample_c_d_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4893,7 +4893,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32, i32, f
define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4909,7 +4909,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32, i32
define amdgpu_ps float @extract_elt0_image_sample_c_l_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_l_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4925,7 +4925,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32, i32, float
define amdgpu_ps float @extract_elt0_image_sample_c_b_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4941,7 +4941,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32, i32, f
define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4957,7 +4957,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32, i32
define amdgpu_ps float @extract_elt0_image_sample_c_lz_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_lz_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4973,7 +4973,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32, i32, floa
define amdgpu_ps float @extract_elt0_image_sample_c_cd_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4989,7 +4989,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -5007,7 +5007,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32, i3
define amdgpu_ps float @extract_elt0_image_gather4_2d_v4f32_f32(float %s, float %t, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5024,7 +5024,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32, float, float, <
define amdgpu_ps float @extract_elt0_image_gather4_cl_2d_v4f32_f32(float %s, float %t, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_cl_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5041,7 +5041,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32, float, float
define amdgpu_ps float @extract_elt0_image_gather4_l_2d_v4f32_f32(float %s, float %t, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_l_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32 4, float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5058,7 +5058,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32, float, float,
define amdgpu_ps float @extract_elt0_image_gather4_b_2darray_v4f32_f32_f32(float %bias, float %s, float %t, float %slice, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_2darray_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32.v8i32.v4i32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5075,7 +5075,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32, floa
define amdgpu_ps float @extract_elt0_image_gather4_b_cl_cube_v4f32_f32_f32(float %bias, float %s, float %t, float %face, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_cl_cube_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32 1, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5092,7 +5092,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32, floa
define amdgpu_ps float @extract_elt0_image_gather4_lz_2d_v4f32_f16(half %s, half %t, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_lz_2d_v4f32_f16(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16.v8i32.v4i32(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5109,7 +5109,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32, half, half,
define amdgpu_ps float @extract_elt0_image_gather4_o_2d_v4f32_f32(i32 %offset, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5126,7 +5126,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32, i32, float, f
define amdgpu_ps float @extract_elt0_image_gather4_cl_o_2d_v4f32_f32(i32 %offset, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_cl_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5143,7 +5143,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32, i32, float
define amdgpu_ps float @extract_elt0_image_gather4_l_o_2d_v4f32_f32(i32 %offset, float %s, float %t, float %lod, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_l_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5160,7 +5160,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_gather4_b_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5177,7 +5177,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32, i32, fl
define amdgpu_ps float @extract_elt0_image_gather4_b_cl_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_cl_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5194,7 +5194,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_gather4_lz_o_2d_v4f32_f32(i32 %offset, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_lz_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5211,7 +5211,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32, i32, float
define amdgpu_ps float @extract_elt0_image_gather4_c_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5228,7 +5228,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_gather4_c_cl_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_cl_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5245,7 +5245,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_gather4_c_l_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, float %lod, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_l_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5262,7 +5262,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32, i32, floa
define amdgpu_ps float @extract_elt0_image_gather4_c_b_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_b_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5279,7 +5279,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_gather4_c_b_cl_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_b_cl_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5296,7 +5296,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32, i3
define amdgpu_ps float @extract_elt0_image_gather4_c_lz_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_lz_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5313,7 +5313,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_getlod_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_getlod_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getlod.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getlod.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -5329,7 +5329,7 @@ declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32, float, <8 x i32>
define amdgpu_ps float @extract_elt0_image_load_2dmsaa_v4f32_i32(i32 %s, i32 %t, i32 %sample, <8 x i32> inreg %sampler) #0 {
; CHECK-LABEL: @extract_elt0_image_load_2dmsaa_v4f32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SAMPLE:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SAMPLE:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %sample, <8 x i32> %sampler, i32 0, i32 0)
@@ -5345,7 +5345,7 @@ declare <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32, i32, i32, i32,
define amdgpu_ps float @extract_elt0_image_load_mip_1d_v4f32_i32(i32 %s, i32 %mip, <8 x i32> inreg %sampler) #0 {
; CHECK-LABEL: @extract_elt0_image_load_mip_1d_v4f32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.mip.1d.f32.i32(i32 1, i32 [[S:%.*]], i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.mip.1d.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
@@ -5361,7 +5361,7 @@ declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32, i32, i32, <8 x
define amdgpu_ps float @extract_elt0_image_getresinfo_1d_v4f32_i32(i32 %mip, <8 x i32> inreg %sampler) #0 {
; CHECK-LABEL: @extract_elt0_image_getresinfo_1d_v4f32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getresinfo.1d.f32.i32(i32 1, i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getresinfo.1d.f32.i32.v8i32(i32 1, i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32 15, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
@@ -5377,7 +5377,7 @@ declare <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32, i32, <8 x i3
define amdgpu_ps float @extract_elt0_tfe_image_load_1d_v4f32i32_i32(i32 %s, <8 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_tfe_image_load_1d_v4f32i32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 1)
+; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 1)
; CHECK-NEXT: [[RGBA:%.*]] = extractvalue { <4 x float>, i32 } [[DATA]], 0
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[RGBA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
@@ -5392,7 +5392,7 @@ declare {<4 x float>, i32} @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32, i32,
define amdgpu_hs float @tfe_check_assert() #0 {
; CHECK-LABEL: @tfe_check_assert(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2d.f32.i32(i32 1, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2d.f32.i32.v8i32(i32 1, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call nsz <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 15, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1) #2
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll
index af12367b9759ce..28403ffd33b99c 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll
@@ -4084,7 +4084,7 @@ declare <4 x i32> @llvm.amdgcn.struct.ptr.tbuffer.load.v4i32(ptr addrspace(8), i
define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4095,7 +4095,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32(float %vaddr, <8
; Check that the intrinsic remains unchanged in the presence of TFE or LWE
define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_tfe(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_1d_v4f32_f32_tfe(
-; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 1, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 1, i32 0)
; CHECK-NEXT: [[DATA_VEC:%.*]] = extractvalue { <4 x float>, i32 } [[DATA]], 0
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
@@ -4109,7 +4109,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_tfe(float %vaddr,
; Check that the intrinsic remains unchanged in the presence of TFE or LWE
define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_lwe(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_1d_v4f32_f32_lwe(
-; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 2, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 2, i32 0)
; CHECK-NEXT: [[DATA_VEC:%.*]] = extractvalue { <4 x float>, i32 } [[DATA]], 0
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
@@ -4122,7 +4122,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_lwe(float %vaddr,
define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %t, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.2d.f32.f32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.2d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4132,7 +4132,7 @@ define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %
define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float %s, float %t, float %r, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 0, float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -4143,7 +4143,7 @@ define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(float %s, float %slice, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1darray.f32.f32(i32 1, float [[S:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1darray.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 1, float %s, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4153,7 +4153,7 @@ define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(f
define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 2, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 2, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4163,7 +4163,7 @@ define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 4, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 4, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4173,7 +4173,7 @@ define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 8, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 8, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 8, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4183,7 +4183,7 @@ define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 9, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4193,7 +4193,7 @@ define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4203,7 +4203,7 @@ define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(float
define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4213,7 +4213,7 @@ define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(float
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <2 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <2 x float> [[SHUF]]
;
@@ -4224,7 +4224,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4234,7 +4234,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4244,7 +4244,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32
define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4254,7 +4254,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <3 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
@@ -4265,7 +4265,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
@@ -4276,7 +4276,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
@@ -4287,7 +4287,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4297,7 +4297,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_
define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_1111_image_sample_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_1111_image_sample_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4317,7 +4317,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32, float, floa
define amdgpu_ps float @extract_elt1_image_sample_cl_2darray_v4f32_f32(float %s, float %t, float %slice, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_image_sample_cl_2darray_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.2darray.f32.f32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.2darray.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32 15, float %s, float %t, float %slice, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4333,7 +4333,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32, float, f
define amdgpu_ps float @extract_elt2_image_sample_d_cube_v4f32_f32_f32(float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt2_image_sample_d_cube_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cube.f32.f32.f32(i32 4, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cube.f32.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32 15, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4349,7 +4349,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32, float, f
define amdgpu_ps float @extract_elt3_image_sample_d_cl_1d_v4f32_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt3_image_sample_d_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.1d.f32.f32.f32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4365,7 +4365,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32, float,
define amdgpu_ps float @extract_elt1_dmask_0110_image_sample_l_1d_v2f32_f32(float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_dmask_0110_image_sample_l_1d_v2f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.1d.f32.f32(i32 4, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32 6, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4381,7 +4381,7 @@ declare <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32, float, float,
define amdgpu_ps float @extract_elt1_dmask_1001_image_sample_b_1d_v4f32_f32_f32(float %bias, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_dmask_1001_image_sample_b_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.1d.f32.f32.f32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 9, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4397,7 +4397,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32, float, flo
define amdgpu_ps <2 x float> @extract_elt1_elt2_dmask_1101_image_sample_b_cl_1d_v4f32_f32_f32(float %bias, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_elt2_dmask_1101_image_sample_b_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.b.cl.1d.v2f32.f32.f32(i32 12, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.b.cl.1d.v2f32.f32.f32.v8i32.v4i32(i32 12, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 13, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4413,7 +4413,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32, float,
define amdgpu_ps <2 x float> @extract_elt1_elt3_image_sample_lz_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_elt3_image_sample_lz_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.1d.v2f32.f32(i32 10, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.1d.v2f32.f32.v8i32.v4i32(i32 10, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4429,7 +4429,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32, float, <8 x i
define amdgpu_ps <3 x float> @extract_elt1_elt2_elt3_image_sample_cd_1d_v4f32_f32_f32(float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_elt2_elt3_image_sample_cd_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.cd.1d.v3f32.f32.f32(i32 14, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.cd.1d.v3f32.f32.f32.v8i32.v4i32(i32 14, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4445,7 +4445,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32, float, fl
define amdgpu_ps half @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4455,7 +4455,7 @@ define amdgpu_ps half @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
define amdgpu_ps half @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 4, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4465,7 +4465,7 @@ define amdgpu_ps half @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
define amdgpu_ps half @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 2, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 2, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4475,7 +4475,7 @@ define amdgpu_ps half @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
define amdgpu_ps <4 x half> @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <3 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v3f16.f32.f32(i32 7, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <3 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v3f16.f32.f32.v8i32.v4i32(i32 7, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[RES:%.*]] = shufflevector <3 x half> [[DATA]], <3 x half> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
@@ -4486,7 +4486,7 @@ define amdgpu_ps <4 x half> @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32
define amdgpu_ps <4 x half> @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <2 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v2f16.f32.f32(i32 3, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <2 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v2f16.f32.f32.v8i32.v4i32(i32 3, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[RES:%.*]] = shufflevector <2 x half> [[DATA]], <2 x half> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
@@ -4497,7 +4497,7 @@ define amdgpu_ps <4 x half> @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32
define amdgpu_ps <4 x half> @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[RES:%.*]] = insertelement <4 x half> poison, half [[DATA]], i64 0
; CHECK-NEXT: ret <4 x half> [[RES]]
;
@@ -4508,7 +4508,7 @@ define amdgpu_ps <4 x half> @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32
define amdgpu_ps half @extract_elt0_image_sample_cd_cl_1d_v4f16_f32_f32(float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cd_cl_1d_v4f16_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
%data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4524,7 +4524,7 @@ declare <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32, float,
define amdgpu_ps float @extract_elt0_image_sample_c_1d_v4f32_f32(float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4540,7 +4540,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32, float, float,
define amdgpu_ps float @extract_elt0_image_sample_c_cl_1d_v4f32_f32(float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cl_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4556,7 +4556,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32, float, floa
define amdgpu_ps float @extract_elt0_image_sample_c_d_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4572,7 +4572,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32, float, f
define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4588,7 +4588,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32, float
define amdgpu_ps float @extract_elt0_image_sample_c_l_1d_v4f32_f32(float %zcompare, float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_l_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4604,7 +4604,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32, float, float
define amdgpu_ps float @extract_elt0_image_sample_c_b_1d_v4f32_f32_f32(float %bias, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.1d.f32.f32.f32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4620,7 +4620,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32, float, f
define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_1d_v4f32_f32_f32(float %bias, float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.1d.f32.f32.f32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4636,7 +4636,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32, float
define amdgpu_ps float @extract_elt0_image_sample_c_lz_1d_v4f32_f32(float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_lz_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.1d.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4652,7 +4652,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32, float, floa
define amdgpu_ps float @extract_elt0_image_sample_c_cd_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4668,7 +4668,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32, float,
define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_1d_v4f32_f32_f32(float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_cl_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.1d.f32.f32.f32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4684,7 +4684,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32, floa
define amdgpu_ps float @extract_elt0_image_sample_o_1d_v4f32_f32(i32 %offset, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4700,7 +4700,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32, i32, float, <8
define amdgpu_ps float @extract_elt0_image_sample_cl_o_1d_v4f32_f32(i32 %offset, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cl_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4716,7 +4716,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_d_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_d_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4732,7 +4732,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_sample_d_cl_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_d_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4748,7 +4748,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_l_o_1d_v4f32_f32(i32 %offset, float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_l_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4764,7 +4764,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_b_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_b_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4780,7 +4780,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_sample_b_cl_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_b_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4796,7 +4796,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_lz_o_1d_v4f32_f32(i32 %offset, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_lz_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.lz.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4812,7 +4812,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_cd_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cd_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4828,7 +4828,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32, i32, fl
define amdgpu_ps float @extract_elt0_image_sample_cd_cl_o_1d_v4f32_f32_f32(i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_cd_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4844,7 +4844,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_c_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4860,7 +4860,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_sample_c_cl_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cl_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4876,7 +4876,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32, i32, floa
define amdgpu_ps float @extract_elt0_image_sample_c_d_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4892,7 +4892,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32, i32, f
define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_d_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4908,7 +4908,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32, i32
define amdgpu_ps float @extract_elt0_image_sample_c_l_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_l_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4924,7 +4924,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32, i32, float
define amdgpu_ps float @extract_elt0_image_sample_c_b_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4940,7 +4940,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32, i32, f
define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_o_1d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_b_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4956,7 +4956,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32, i32
define amdgpu_ps float @extract_elt0_image_sample_c_lz_o_1d_v4f32_f32(i32 %offset, float %zcompare, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_lz_o_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.o.1d.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4972,7 +4972,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32, i32, floa
define amdgpu_ps float @extract_elt0_image_sample_c_cd_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -4988,7 +4988,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_o_1d_v4f32_f32_f32(i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_sample_c_cd_cl_o_1d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.o.1d.f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -5006,7 +5006,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32, i3
define amdgpu_ps float @extract_elt0_image_gather4_2d_v4f32_f32(float %s, float %t, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5023,7 +5023,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32, float, float, <
define amdgpu_ps float @extract_elt0_image_gather4_cl_2d_v4f32_f32(float %s, float %t, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_cl_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5040,7 +5040,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32, float, float
define amdgpu_ps float @extract_elt0_image_gather4_l_2d_v4f32_f32(float %s, float %t, float %lod, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_l_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32 4, float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5057,7 +5057,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32, float, float,
define amdgpu_ps float @extract_elt0_image_gather4_b_2darray_v4f32_f32_f32(float %bias, float %s, float %t, float %slice, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_2darray_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32.v8i32.v4i32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5074,7 +5074,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32, floa
define amdgpu_ps float @extract_elt0_image_gather4_b_cl_cube_v4f32_f32_f32(float %bias, float %s, float %t, float %face, float %clamp, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_cl_cube_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32 1, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5091,7 +5091,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32, floa
define amdgpu_ps float @extract_elt0_image_gather4_lz_2d_v4f32_f16(half %s, half %t, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_lz_2d_v4f32_f16(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16.v8i32.v4i32(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5108,7 +5108,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32, half, half,
define amdgpu_ps float @extract_elt0_image_gather4_o_2d_v4f32_f32(i32 %offset, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5125,7 +5125,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32, i32, float, f
define amdgpu_ps float @extract_elt0_image_gather4_cl_o_2d_v4f32_f32(i32 %offset, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_cl_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5142,7 +5142,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32, i32, float
define amdgpu_ps float @extract_elt0_image_gather4_l_o_2d_v4f32_f32(i32 %offset, float %s, float %t, float %lod, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_l_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5159,7 +5159,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_gather4_b_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5176,7 +5176,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32, i32, fl
define amdgpu_ps float @extract_elt0_image_gather4_b_cl_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_b_cl_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5193,7 +5193,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_gather4_lz_o_2d_v4f32_f32(i32 %offset, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_lz_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5210,7 +5210,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32, i32, float
define amdgpu_ps float @extract_elt0_image_gather4_c_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5227,7 +5227,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32, i32, float,
define amdgpu_ps float @extract_elt0_image_gather4_c_cl_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_cl_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5244,7 +5244,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_gather4_c_l_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, float %lod, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_l_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5261,7 +5261,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32, i32, floa
define amdgpu_ps float @extract_elt0_image_gather4_c_b_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_b_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5278,7 +5278,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32, i32,
define amdgpu_ps float @extract_elt0_image_gather4_c_b_cl_o_2d_v4f32_f32_f32(i32 %offset, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_b_cl_o_2d_v4f32_f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5295,7 +5295,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32, i3
define amdgpu_ps float @extract_elt0_image_gather4_c_lz_o_2d_v4f32_f32(i32 %offset, float %zcompare, float %s, float %t, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_gather4_c_lz_o_2d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[GATHER4R:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
@@ -5312,7 +5312,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32, i32, flo
define amdgpu_ps float @extract_elt0_image_getlod_1d_v4f32_f32(float %s, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_image_getlod_1d_v4f32_f32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getlod.1d.f32.f32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getlod.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
@@ -5328,7 +5328,7 @@ declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32, float, <8 x i32>
define amdgpu_ps float @extract_elt0_image_load_2dmsaa_v4f32_i32(i32 %s, i32 %t, i32 %sample, <8 x i32> inreg %sampler) #0 {
; CHECK-LABEL: @extract_elt0_image_load_2dmsaa_v4f32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SAMPLE:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SAMPLE:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %sample, <8 x i32> %sampler, i32 0, i32 0)
@@ -5344,7 +5344,7 @@ declare <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32, i32, i32, i32,
define amdgpu_ps float @extract_elt0_image_load_mip_1d_v4f32_i32(i32 %s, i32 %mip, <8 x i32> inreg %sampler) #0 {
; CHECK-LABEL: @extract_elt0_image_load_mip_1d_v4f32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.mip.1d.f32.i32(i32 1, i32 [[S:%.*]], i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.mip.1d.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
@@ -5360,7 +5360,7 @@ declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32, i32, i32, <8 x
define amdgpu_ps float @extract_elt0_image_getresinfo_1d_v4f32_i32(i32 %mip, <8 x i32> inreg %sampler) #0 {
; CHECK-LABEL: @extract_elt0_image_getresinfo_1d_v4f32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getresinfo.1d.f32.i32(i32 1, i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getresinfo.1d.f32.i32.v8i32(i32 1, i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32 15, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
@@ -5376,7 +5376,7 @@ declare <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32, i32, <8 x i3
define amdgpu_ps float @extract_elt0_tfe_image_load_1d_v4f32i32_i32(i32 %s, <8 x i32> inreg %rsrc) #0 {
; CHECK-LABEL: @extract_elt0_tfe_image_load_1d_v4f32i32_i32(
-; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 1)
+; CHECK-NEXT: [[DATA:%.*]] = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 1)
; CHECK-NEXT: [[RGBA:%.*]] = extractvalue { <4 x float>, i32 } [[DATA]], 0
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[RGBA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
@@ -5391,7 +5391,7 @@ declare {<4 x float>, i32} @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32, i32,
define amdgpu_hs float @tfe_check_assert() #0 {
; CHECK-LABEL: @tfe_check_assert(
-; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2d.f32.i32(i32 1, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1)
+; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2d.f32.i32.v8i32(i32 1, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1)
; CHECK-NEXT: ret float [[DATA]]
;
%data = call nsz <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 15, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1) #2
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll
index 0cb9b4b84a5568..fc9dd99c4803a8 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll
@@ -15,7 +15,7 @@ declare <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32, float, floa
define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float [[S32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -29,7 +29,7 @@ define amdgpu_kernel void @image_sample_a16_2d(ptr addrspace(1) %out, <8 x i32>
; CHECK-LABEL: @image_sample_a16_2d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float [[S32]], float [[T32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S32]], float [[T32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -45,7 +45,7 @@ define amdgpu_kernel void @image_sample_a16_3d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
; CHECK-NEXT: [[R32:%.*]] = fpext half [[R:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 15, float [[S32]], float [[T32]], float [[R32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 15, float [[S32]], float [[T32]], float [[R32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -63,7 +63,7 @@ define amdgpu_kernel void @image_sample_a16_cube(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
; CHECK-NEXT: [[FACE32:%.*]] = fpext half [[FACE:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32(i32 15, float [[S32]], float [[T32]], float [[FACE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32.v8i32.v4i32(i32 15, float [[S32]], float [[T32]], float [[FACE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -79,7 +79,7 @@ define amdgpu_kernel void @image_sample_a16_1darray(ptr addrspace(1) %out, <8 x
; CHECK-LABEL: @image_sample_a16_1darray(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[SLICE32:%.*]] = fpext half [[SLICE:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 15, float [[S32]], float [[SLICE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32 15, float [[S32]], float [[SLICE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -95,7 +95,7 @@ define amdgpu_kernel void @image_sample_a16_2darray(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
; CHECK-NEXT: [[SLICE32:%.*]] = fpext half [[SLICE:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32 15, float [[S32]], float [[T32]], float [[SLICE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32.v8i32.v4i32(i32 15, float [[S32]], float [[T32]], float [[SLICE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
index fabf8ab51764b9..e886dbc509e1d5 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
@@ -3216,7 +3216,7 @@ declare <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32(i32, i
define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3228,7 +3228,7 @@ define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32>
define amdgpu_kernel void @image_sample_a16_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3241,7 +3241,7 @@ define amdgpu_kernel void @image_sample_a16_2d(ptr addrspace(1) %out, <8 x i32>
define amdgpu_kernel void @image_sample_a16_3d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %r) {
; CHECK-LABEL: @image_sample_a16_3d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3256,7 +3256,7 @@ define amdgpu_kernel void @image_sample_a16_3d(ptr addrspace(1) %out, <8 x i32>
define amdgpu_kernel void @image_sample_a16_cube(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %face) {
;
; CHECK-LABEL: @image_sample_a16_cube(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[FACE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[FACE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3270,7 +3270,7 @@ define amdgpu_kernel void @image_sample_a16_cube(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_1darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %slice) {
; CHECK-LABEL: @image_sample_a16_1darray(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f16(i32 15, half [[S:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3283,7 +3283,7 @@ define amdgpu_kernel void @image_sample_a16_1darray(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_2darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %slice) {
; CHECK-LABEL: @image_sample_a16_2darray(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3297,7 +3297,7 @@ define amdgpu_kernel void @image_sample_a16_2darray(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s) {
; CHECK-LABEL: @image_sample_a16_c_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3309,7 +3309,7 @@ define amdgpu_kernel void @image_sample_a16_c_1d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_c_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_c_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3322,7 +3322,7 @@ define amdgpu_kernel void @image_sample_a16_c_2d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f16(i32 15, half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3335,7 +3335,7 @@ define amdgpu_kernel void @image_sample_a16_cl_1d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_a16_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3349,7 +3349,7 @@ define amdgpu_kernel void @image_sample_a16_cl_2d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_a16_c_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3362,7 +3362,7 @@ define amdgpu_kernel void @image_sample_a16_c_cl_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3376,7 +3376,7 @@ define amdgpu_kernel void @image_sample_a16_c_cl_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_b16_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, half %s) {
; CHECK-LABEL: @image_sample_a16_b16_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3390,7 +3390,7 @@ define amdgpu_kernel void @image_sample_a16_b16_1d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_b32_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, half %s) {
; CHECK-LABEL: @image_sample_a16_b32_1d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3402,7 +3402,7 @@ define amdgpu_kernel void @image_sample_a16_b32_1d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_b16_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_b16_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3418,7 +3418,7 @@ define amdgpu_kernel void @image_sample_a16_b32_2d(ptr addrspace(1) %out, <8 x i
; CHECK-LABEL: @image_sample_a16_b32_2d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S32]], float [[T32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S32]], float [[T32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3431,7 +3431,7 @@ define amdgpu_kernel void @image_sample_a16_b32_2d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_c_b16_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, float %zcompare, half %s) {
; CHECK-LABEL: @image_sample_a16_c_b16_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3445,7 +3445,7 @@ define amdgpu_kernel void @image_sample_a16_c_b16_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_b32_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %zcompare, half %s) {
; CHECK-LABEL: @image_sample_a16_c_b32_1d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3457,7 +3457,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_b16_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, float %zcompare, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_c_b16_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3473,7 +3473,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_2d(ptr addrspace(1) %out, <8 x
; CHECK-LABEL: @image_sample_a16_c_b32_2d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], float [[T32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], float [[T32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3486,7 +3486,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_b16_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_b16_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3502,7 +3502,7 @@ define amdgpu_kernel void @image_sample_a16_b32_cl_1d(ptr addrspace(1) %out, <8
; CHECK-LABEL: @image_sample_a16_b32_cl_1d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[CLAMP32:%.*]] = fpext half [[CLAMP:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3515,7 +3515,7 @@ define amdgpu_kernel void @image_sample_a16_b32_cl_1d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_b16_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_b16_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3533,7 +3533,7 @@ define amdgpu_kernel void @image_sample_a16_b32_cl_2d(ptr addrspace(1) %out, <8
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
; CHECK-NEXT: [[CLAMP32:%.*]] = fpext half [[CLAMP:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S32]], float [[T32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S32]], float [[T32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3547,7 +3547,7 @@ define amdgpu_kernel void @image_sample_a16_b32_cl_2d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_c_b16_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, float %zcompare, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_b16_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3563,7 +3563,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_cl_1d(ptr addrspace(1) %out, <
; CHECK-LABEL: @image_sample_a16_c_b32_cl_1d(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[CLAMP32:%.*]] = fpext half [[CLAMP:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3576,7 +3576,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_cl_1d(ptr addrspace(1) %out, <
define amdgpu_kernel void @image_sample_a16_c_b16_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %bias, float %zcompare, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_b16_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f16.f16(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[BIAS:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3594,7 +3594,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_cl_2d(ptr addrspace(1) %out, <
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[T32:%.*]] = fpext half [[T:%.*]] to float
; CHECK-NEXT: [[CLAMP32:%.*]] = fpext half [[CLAMP:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], float [[T32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S32]], float [[T32]], float [[CLAMP32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3608,7 +3608,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_cl_2d(ptr addrspace(1) %out, <
define amdgpu_kernel void @image_sample_a16_d_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, half %s) {
; CHECK-LABEL: @image_sample_a16_d_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3622,7 +3622,7 @@ define amdgpu_kernel void @image_sample_a16_d_1d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_d_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_d_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3639,7 +3639,7 @@ define amdgpu_kernel void @image_sample_a16_d_2d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_d_3d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %drdh, half %dsdv, half %dtdv, half %drdv, half %s, half %t, half %r) {
; CHECK-LABEL: @image_sample_a16_d_3d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DRDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[DRDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DRDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[DRDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3659,7 +3659,7 @@ define amdgpu_kernel void @image_sample_a16_d_3d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_c_d_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, half %s) {
; CHECK-LABEL: @image_sample_a16_c_d_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3673,7 +3673,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_1d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_c_d_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_c_d_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3690,7 +3690,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_2d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_d_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_d_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3705,7 +3705,7 @@ define amdgpu_kernel void @image_sample_a16_d_cl_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_d_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_d_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3723,7 +3723,7 @@ define amdgpu_kernel void @image_sample_a16_d_cl_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_d_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_d_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3738,7 +3738,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_cl_1d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_c_d_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_d_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3756,7 +3756,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_cl_2d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_cd_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, half %s) {
; CHECK-LABEL: @image_sample_a16_cd_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3770,7 +3770,7 @@ define amdgpu_kernel void @image_sample_a16_cd_1d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_a16_cd_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_cd_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3787,7 +3787,7 @@ define amdgpu_kernel void @image_sample_a16_cd_2d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_a16_c_cd_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, half %s) {
; CHECK-LABEL: @image_sample_a16_c_cd_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3801,7 +3801,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_cd_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_c_cd_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3818,7 +3818,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_cd_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_cd_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3833,7 +3833,7 @@ define amdgpu_kernel void @image_sample_a16_cd_cl_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_cd_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_cd_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f16.f16(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3851,7 +3851,7 @@ define amdgpu_kernel void @image_sample_a16_cd_cl_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_cd_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, half %s, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_cd_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], half [[S:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3866,7 +3866,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_cl_1d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_c_cd_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t, half %clamp) {
; CHECK-LABEL: @image_sample_a16_c_cd_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f16.f16(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f16.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3884,7 +3884,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_cl_2d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_l_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %lod) {
; CHECK-LABEL: @image_sample_a16_l_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f16(i32 15, half [[S:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3897,7 +3897,7 @@ define amdgpu_kernel void @image_sample_a16_l_1d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %lod) {
; CHECK-LABEL: @image_sample_a16_l_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3911,7 +3911,7 @@ define amdgpu_kernel void @image_sample_a16_l_2d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_a16_c_l_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s, half %lod) {
; CHECK-LABEL: @image_sample_a16_c_l_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3924,7 +3924,7 @@ define amdgpu_kernel void @image_sample_a16_c_l_1d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_c_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s, half %t, half %lod) {
; CHECK-LABEL: @image_sample_a16_c_l_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], half [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3938,7 +3938,7 @@ define amdgpu_kernel void @image_sample_a16_c_l_2d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_a16_lz_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_lz_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f16(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3950,7 +3950,7 @@ define amdgpu_kernel void @image_sample_a16_lz_1d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_a16_lz_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_lz_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3963,7 +3963,7 @@ define amdgpu_kernel void @image_sample_a16_lz_2d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_a16_c_lz_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s) {
; CHECK-LABEL: @image_sample_a16_c_lz_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3975,7 +3975,7 @@ define amdgpu_kernel void @image_sample_a16_c_lz_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_lz_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_c_lz_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f16(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f16.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -3988,7 +3988,7 @@ define amdgpu_kernel void @image_sample_a16_c_lz_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_V1(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t, half %slice) {
; CHECK-LABEL: @image_sample_a16_c_d_o_2darray_V1(
-; CHECK-NEXT: [[RES:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f16.f16(i32 4, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f16.f16.v8i32.v4i32(i32 4, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store float [[RES]], ptr addrspace(1) [[OUT:%.*]], align 4
; CHECK-NEXT: ret void
;
@@ -4006,7 +4006,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_V1(ptr addrspace(1) %o
define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_V2(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %t, half %slice) {
; CHECK-LABEL: @image_sample_a16_c_d_o_2darray_V2(
-; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f16(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f16.v8i32.v4i32(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <2 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 8
; CHECK-NEXT: ret void
;
@@ -4024,7 +4024,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_V2(ptr addrspace(1) %o
define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_const(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, half %s, half %slice) {
; CHECK-LABEL: @image_sample_a16_c_d_o_2darray_const(
-; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f16(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half 0xH3400, half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f16.v8i32.v4i32(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[S:%.*]], half 0xH3400, half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <2 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 8
; CHECK-NEXT: ret void
;
@@ -4043,7 +4043,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_const_noopt(ptr addrsp
; CHECK-LABEL: @image_sample_a16_c_d_o_2darray_const_noopt(
; CHECK-NEXT: [[S32:%.*]] = fpext half [[S:%.*]] to float
; CHECK-NEXT: [[SLICE32:%.*]] = fpext half [[SLICE:%.*]] to float
-; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f32(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S32]], float 1.000000e+10, float [[SLICE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f32.v8i32.v4i32(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S32]], float 1.000000e+10, float [[SLICE32]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <2 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 8
; CHECK-NEXT: ret void
;
@@ -4060,7 +4060,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_const_noopt(ptr addrsp
define amdgpu_kernel void @image_load_a16_mip_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i16 %s) {
; CHECK-LABEL: @image_load_a16_mip_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4073,7 +4073,7 @@ define amdgpu_kernel void @image_load_a16_mip_1d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_load_a16_mip_1d_noopt(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i16 %s) {
; CHECK-LABEL: @image_load_a16_mip_1d_noopt(
; CHECK-NEXT: [[S32:%.*]] = sext i16 [[S:%.*]] to i32
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 [[S32]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 [[S32]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4085,7 +4085,7 @@ define amdgpu_kernel void @image_load_a16_mip_1d_noopt(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_load_a16_mip_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i16 %s, i16 %t) {
; CHECK-LABEL: @image_load_a16_mip_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i16(i32 15, i16 [[S:%.*]], i16 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i16.v8i32(i32 15, i16 [[S:%.*]], i16 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4098,7 +4098,7 @@ define amdgpu_kernel void @image_load_a16_mip_2d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_load_a16_mip_2d_const(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i16 %s) {
; CHECK-LABEL: @image_load_a16_mip_2d_const(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i16(i32 15, i16 [[S:%.*]], i16 -1, <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i16.v8i32(i32 15, i16 [[S:%.*]], i16 -1, <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4111,7 +4111,7 @@ define amdgpu_kernel void @image_load_a16_mip_2d_const(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_load_a16_mip_2d_const_noopt(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i16 %s) {
; CHECK-LABEL: @image_load_a16_mip_2d_const_noopt(
; CHECK-NEXT: [[S32:%.*]] = zext i16 [[S:%.*]] to i32
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 15, i32 [[S32]], i32 65536, <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32.v8i32(i32 15, i32 [[S32]], i32 65536, <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4127,7 +4127,7 @@ define amdgpu_kernel void @image_load_a16_mip_2d_const_noopt(ptr addrspace(1) %o
define amdgpu_kernel void @image_sample_g16_d_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, float %s) {
; CHECK-LABEL: @image_sample_g16_d_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4140,7 +4140,7 @@ define amdgpu_kernel void @image_sample_g16_d_1d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_g16_d_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t) {
; CHECK-LABEL: @image_sample_g16_d_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4155,7 +4155,7 @@ define amdgpu_kernel void @image_sample_g16_d_2d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_g16_d_3d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %drdh, half %dsdv, half %dtdv, half %drdv, float %s, float %t, float %r) {
; CHECK-LABEL: @image_sample_g16_d_3d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DRDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[DRDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DRDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], half [[DRDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4172,7 +4172,7 @@ define amdgpu_kernel void @image_sample_g16_d_3d(ptr addrspace(1) %out, <8 x i32
define amdgpu_kernel void @image_sample_g16_c_d_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, float %s) {
; CHECK-LABEL: @image_sample_g16_c_d_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4185,7 +4185,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_1d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_g16_c_d_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t) {
; CHECK-LABEL: @image_sample_g16_c_d_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4200,7 +4200,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_2d(ptr addrspace(1) %out, <8 x i
define amdgpu_kernel void @image_sample_g16_d_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @image_sample_g16_d_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4213,7 +4213,7 @@ define amdgpu_kernel void @image_sample_g16_d_cl_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_g16_d_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @image_sample_g16_d_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4228,7 +4228,7 @@ define amdgpu_kernel void @image_sample_g16_d_cl_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_g16_c_d_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @image_sample_g16_c_d_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4241,7 +4241,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_cl_1d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_g16_c_d_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @image_sample_g16_c_d_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4256,7 +4256,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_cl_2d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_g16_cd_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, float %s) {
; CHECK-LABEL: @image_sample_g16_cd_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4269,7 +4269,7 @@ define amdgpu_kernel void @image_sample_g16_cd_1d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_g16_cd_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t) {
; CHECK-LABEL: @image_sample_g16_cd_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4284,7 +4284,7 @@ define amdgpu_kernel void @image_sample_g16_cd_2d(ptr addrspace(1) %out, <8 x i3
define amdgpu_kernel void @image_sample_g16_c_cd_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, float %s) {
; CHECK-LABEL: @image_sample_g16_c_cd_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4297,7 +4297,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_g16_c_cd_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t) {
; CHECK-LABEL: @image_sample_g16_c_cd_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4312,7 +4312,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_g16_cd_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @image_sample_g16_cd_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4325,7 +4325,7 @@ define amdgpu_kernel void @image_sample_g16_cd_cl_1d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_g16_cd_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @image_sample_g16_cd_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f16.f32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4340,7 +4340,7 @@ define amdgpu_kernel void @image_sample_g16_cd_cl_2d(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_g16_c_cd_cl_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @image_sample_g16_c_cd_cl_1d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4353,7 +4353,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_cl_1d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_g16_c_cd_cl_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @image_sample_g16_c_cd_cl_2d(
-; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f16.f32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f16.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4368,7 +4368,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_cl_2d(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_g16_c_d_o_2darray_V1(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t, float %slice) {
; CHECK-LABEL: @image_sample_g16_c_d_o_2darray_V1(
-; CHECK-NEXT: [[RES:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f16.f32(i32 4, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f16.f32.v8i32.v4i32(i32 4, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store float [[RES]], ptr addrspace(1) [[OUT:%.*]], align 4
; CHECK-NEXT: ret void
;
@@ -4383,7 +4383,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_o_2darray_V1(ptr addrspace(1) %o
define amdgpu_kernel void @image_sample_g16_c_d_o_2darray_V2(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, half %dsdh, half %dtdh, half %dsdv, half %dtdv, float %s, float %t, float %slice) {
; CHECK-LABEL: @image_sample_g16_c_d_o_2darray_V2(
-; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f32(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f16.f32.v8i32.v4i32(i32 6, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[DSDH:%.*]], half [[DTDH:%.*]], half [[DSDV:%.*]], half [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <2 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 8
; CHECK-NEXT: ret void
;
@@ -4402,7 +4402,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_o_2darray_V2(ptr addrspace(1) %o
define amdgpu_kernel void @image_sample_a16_1d_nnan(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d_nnan(
-; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4414,7 +4414,7 @@ define amdgpu_kernel void @image_sample_a16_1d_nnan(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_1d_nnan_ninf_nsz(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d_nnan_ninf_nsz(
-; CHECK-NEXT: [[RES:%.*]] = call nnan ninf nsz <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan ninf nsz <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4426,7 +4426,7 @@ define amdgpu_kernel void @image_sample_a16_1d_nnan_ninf_nsz(ptr addrspace(1) %o
define amdgpu_kernel void @image_sample_a16_1d_fast(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d_fast(
-; CHECK-NEXT: [[RES:%.*]] = call fast <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call fast <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4438,7 +4438,7 @@ define amdgpu_kernel void @image_sample_a16_1d_fast(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_2d_nnan(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t) {
; CHECK-LABEL: @image_sample_a16_2d_nnan(
-; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4451,7 +4451,7 @@ define amdgpu_kernel void @image_sample_a16_2d_nnan(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_3d_nnan(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %r) {
; CHECK-LABEL: @image_sample_a16_3d_nnan(
-; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[R:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4466,7 +4466,7 @@ define amdgpu_kernel void @image_sample_a16_3d_nnan(ptr addrspace(1) %out, <8 x
define amdgpu_kernel void @image_sample_a16_cube_nnan(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %face) {
;
; CHECK-LABEL: @image_sample_a16_cube_nnan(
-; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[FACE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[FACE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4480,7 +4480,7 @@ define amdgpu_kernel void @image_sample_a16_cube_nnan(ptr addrspace(1) %out, <8
define amdgpu_kernel void @image_sample_a16_1darray_nnan(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %slice) {
; CHECK-LABEL: @image_sample_a16_1darray_nnan(
-; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f16(i32 15, half [[S:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4493,7 +4493,7 @@ define amdgpu_kernel void @image_sample_a16_1darray_nnan(ptr addrspace(1) %out,
define amdgpu_kernel void @image_sample_a16_2darray_nnan(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t, half %slice) {
; CHECK-LABEL: @image_sample_a16_2darray_nnan(
-; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f16(i32 15, half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[RES:%.*]] = call nnan <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], half [[T:%.*]], half [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[RES]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4523,7 +4523,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2darray.v4f32.f32(i32, i32,
define amdgpu_kernel void @sample_l_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %lod) {
; CHECK-LABEL: @sample_l_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4536,7 +4536,7 @@ main_body:
define amdgpu_kernel void @sample_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t, float %lod) {
; CHECK-LABEL: @sample_l_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4549,7 +4549,7 @@ main_body:
define amdgpu_kernel void @sample_c_l_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %lod) {
; CHECK-LABEL: @sample_c_l_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4562,7 +4562,7 @@ main_body:
define amdgpu_kernel void @sample_c_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t, float %lod) {
; CHECK-LABEL: @sample_c_l_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4575,7 +4575,7 @@ main_body:
define amdgpu_kernel void @sample_l_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %s, float %lod) {
; CHECK-LABEL: @sample_l_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4588,7 +4588,7 @@ main_body:
define amdgpu_kernel void @sample_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %s, float %t, float %lod) {
; CHECK-LABEL: @sample_l_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4601,7 +4601,7 @@ main_body:
define amdgpu_kernel void @sample_c_l_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s, float %lod) {
; CHECK-LABEL: @sample_c_l_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4614,7 +4614,7 @@ main_body:
define amdgpu_kernel void @sample_c_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s, float %t, float %lod) {
; CHECK-LABEL: @sample_c_l_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4627,7 +4627,7 @@ main_body:
define amdgpu_kernel void @gather4_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t, float %lod) {
; CHECK-LABEL: @gather4_l_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4640,7 +4640,7 @@ main_body:
define amdgpu_kernel void @gather4_c_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t, float %lod) {
; CHECK-LABEL: @gather4_c_l_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4653,7 +4653,7 @@ main_body:
define amdgpu_kernel void @gather4_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %s, float %t, float %lod) {
; CHECK-LABEL: @gather4_l_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4666,7 +4666,7 @@ main_body:
define amdgpu_kernel void @gather4_c_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s, float %t, float %lod) {
; CHECK-LABEL: @gather4_c_l_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4679,7 +4679,7 @@ main_body:
define amdgpu_kernel void @gather4_c_l_o_2darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s, float %t, float %slice, float %lod) {
; CHECK-LABEL: @gather4_c_l_o_2darray(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2darray.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2darray.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4696,7 +4696,7 @@ main_body:
define amdgpu_kernel void @load_mip_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i32 %s) {
; CHECK-LABEL: @load_mip_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4709,7 +4709,7 @@ main_body:
define amdgpu_kernel void @load_mip_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i32 %s, i32 %t) {
; CHECK-LABEL: @load_mip_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4722,7 +4722,7 @@ main_body:
define amdgpu_kernel void @load_mip_3d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %u) {
; CHECK-LABEL: @load_mip_3d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.3d.v4f32.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.3d.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4735,7 +4735,7 @@ main_body:
define amdgpu_kernel void @load_mip_1darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i32 %s, i32 %t) {
; CHECK-LABEL: @load_mip_1darray(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1darray.v4f32.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1darray.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4748,7 +4748,7 @@ main_body:
define amdgpu_kernel void @load_mip_2darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %u) {
; CHECK-LABEL: @load_mip_2darray(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2darray.v4f32.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.2darray.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4761,7 +4761,7 @@ main_body:
define amdgpu_kernel void @load_mip_cube(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %u) {
; CHECK-LABEL: @load_mip_cube(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.cube.v4f32.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.load.cube.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4775,7 +4775,7 @@ main_body:
define amdgpu_kernel void @store_mip_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x float> %vdata, i32 %s) {
; CHECK-LABEL: @store_mip_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret void
;
main_body:
@@ -4786,7 +4786,7 @@ main_body:
define amdgpu_kernel void @store_mip_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x float> %vdata, i32 %s, i32 %t) {
; CHECK-LABEL: @store_mip_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: call void @llvm.amdgcn.image.store.2d.v4f32.i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.amdgcn.image.store.2d.v4f32.i32.v8i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret void
;
main_body:
@@ -4797,7 +4797,7 @@ main_body:
define amdgpu_kernel void @store_mip_3d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x float> %vdata, i32 %s, i32 %t, i32 %u) {
; CHECK-LABEL: @store_mip_3d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: call void @llvm.amdgcn.image.store.3d.v4f32.i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.amdgcn.image.store.3d.v4f32.i32.v8i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret void
;
main_body:
@@ -4808,7 +4808,7 @@ main_body:
define amdgpu_kernel void @store_mip_1darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x float> %vdata, i32 %s, i32 %t) {
; CHECK-LABEL: @store_mip_1darray(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: call void @llvm.amdgcn.image.store.1darray.v4f32.i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.amdgcn.image.store.1darray.v4f32.i32.v8i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret void
;
main_body:
@@ -4819,7 +4819,7 @@ main_body:
define amdgpu_kernel void @store_mip_2darray(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x float> %vdata, i32 %s, i32 %t, i32 %u) {
; CHECK-LABEL: @store_mip_2darray(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: call void @llvm.amdgcn.image.store.2darray.v4f32.i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.amdgcn.image.store.2darray.v4f32.i32.v8i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret void
;
main_body:
@@ -4830,7 +4830,7 @@ main_body:
define amdgpu_kernel void @store_mip_cube(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x float> %vdata, i32 %s, i32 %t, i32 %u) {
; CHECK-LABEL: @store_mip_cube(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: call void @llvm.amdgcn.image.store.cube.v4f32.i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.amdgcn.image.store.cube.v4f32.i32.v8i32(<4 x float> [[VDATA:%.*]], i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[U:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret void
;
main_body:
@@ -4875,7 +4875,7 @@ declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32(i32, i32, floa
define amdgpu_kernel void @sample_b_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s) {
; CHECK-LABEL: @sample_b_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4888,7 +4888,7 @@ main_body:
define amdgpu_kernel void @sample_b_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; CHECK-LABEL: @sample_b_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4901,7 +4901,7 @@ main_body:
define amdgpu_kernel void @sample_c_b_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s) {
; CHECK-LABEL: @sample_c_b_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4914,7 +4914,7 @@ main_body:
define amdgpu_kernel void @sample_c_b_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @sample_c_b_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4927,7 +4927,7 @@ main_body:
define amdgpu_kernel void @sample_b_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %s) {
; CHECK-LABEL: @sample_b_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4940,7 +4940,7 @@ main_body:
define amdgpu_kernel void @sample_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %s, float %t) {
; CHECK-LABEL: @sample_b_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4953,7 +4953,7 @@ main_body:
define amdgpu_kernel void @sample_c_b_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s) {
; CHECK-LABEL: @sample_c_b_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4966,7 +4966,7 @@ main_body:
define amdgpu_kernel void @sample_c_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @sample_c_b_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4979,7 +4979,7 @@ main_body:
define amdgpu_kernel void @gather4_b_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; CHECK-LABEL: @gather4_b_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -4992,7 +4992,7 @@ main_body:
define amdgpu_kernel void @gather4_c_b_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @gather4_c_b_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5005,7 +5005,7 @@ main_body:
define amdgpu_kernel void @gather4_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %s, float %t) {
; CHECK-LABEL: @gather4_b_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5018,7 +5018,7 @@ main_body:
define amdgpu_kernel void @gather4_c_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @gather4_c_b_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5031,7 +5031,7 @@ main_body:
define amdgpu_kernel void @sample_c_b_o_a16_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, i32 %offset, float %zcompare, half %s, half %t) {
; CHECK-LABEL: @sample_c_b_o_a16_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f16(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f16.v8i32.v4i32(i32 15, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5045,7 +5045,7 @@ main_body:
define amdgpu_kernel void @sample_b_1d_pos(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s) {
; CHECK-LABEL: @sample_b_1d_pos(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 15, float 1.000000e+00, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float 1.000000e+00, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5059,7 +5059,7 @@ main_body:
define amdgpu_kernel void @sample_b_1d_neg(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s) {
; CHECK-LABEL: @sample_b_1d_neg(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 15, float -1.000000e+00, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float -1.000000e+00, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5073,7 +5073,7 @@ main_body:
define amdgpu_kernel void @sample_b_1d_a16(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @sample_b_1d_a16(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f16.v8i32.v4i32(i32 15, half [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5091,7 +5091,7 @@ main_body:
define amdgpu_kernel void @offset_sample_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s) {
; CHECK-LABEL: @offset_sample_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5104,7 +5104,7 @@ main_body:
define amdgpu_kernel void @offset_sample_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; CHECK-LABEL: @offset_sample_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5117,7 +5117,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s) {
; CHECK-LABEL: @offset_sample_c_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5130,7 +5130,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @offset_sample_c_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5143,7 +5143,7 @@ main_body:
define amdgpu_kernel void @offset_sample_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f32(i32 15, float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5156,7 +5156,7 @@ main_body:
define amdgpu_kernel void @offset_sample_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5169,7 +5169,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_c_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5182,7 +5182,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_c_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5195,7 +5195,7 @@ main_body:
define amdgpu_kernel void @offset_sample_b_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %s) {
; CHECK-LABEL: @offset_sample_b_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5208,7 +5208,7 @@ main_body:
define amdgpu_kernel void @offset_sample_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %s, float %t) {
; CHECK-LABEL: @offset_sample_b_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5221,7 +5221,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_b_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %zcompare, float %s) {
; CHECK-LABEL: @offset_sample_c_b_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5234,7 +5234,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @offset_sample_c_b_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5247,7 +5247,7 @@ main_body:
define amdgpu_kernel void @offset_sample_b_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_b_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5260,7 +5260,7 @@ main_body:
define amdgpu_kernel void @offset_sample_b_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_b_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5273,7 +5273,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_b_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %zcompare, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_c_b_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5286,7 +5286,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_b_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %bias, float %zcompare, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_c_b_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5299,7 +5299,7 @@ main_body:
define amdgpu_kernel void @offset_sample_d_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dsdv, float %s) {
; CHECK-LABEL: @offset_sample_d_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5312,7 +5312,7 @@ main_body:
define amdgpu_kernel void @offset_sample_d_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t) {
; CHECK-LABEL: @offset_sample_d_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5325,7 +5325,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_d_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dsdv, float %s) {
; CHECK-LABEL: @offset_sample_c_d_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5338,7 +5338,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_d_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t) {
; CHECK-LABEL: @offset_sample_c_d_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5351,7 +5351,7 @@ main_body:
define amdgpu_kernel void @offset_sample_d_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_d_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5364,7 +5364,7 @@ main_body:
define amdgpu_kernel void @offset_sample_d_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_d_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5377,7 +5377,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_d_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_c_d_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5390,7 +5390,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_d_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_c_d_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5403,7 +5403,7 @@ main_body:
define amdgpu_kernel void @offset_sample_cd_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dsdv, float %s) {
; CHECK-LABEL: @offset_sample_cd_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5416,7 +5416,7 @@ main_body:
define amdgpu_kernel void @offset_sample_cd_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t) {
; CHECK-LABEL: @offset_sample_cd_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5429,7 +5429,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_cd_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dsdv, float %s) {
; CHECK-LABEL: @offset_sample_c_cd_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5442,7 +5442,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_cd_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t) {
; CHECK-LABEL: @offset_sample_c_cd_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5455,7 +5455,7 @@ main_body:
define amdgpu_kernel void @offset_sample_cd_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_cd_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5468,7 +5468,7 @@ main_body:
define amdgpu_kernel void @offset_sample_cd_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_cd_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5481,7 +5481,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_cd_cl_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp) {
; CHECK-LABEL: @offset_sample_c_cd_cl_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5494,7 +5494,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_cd_cl_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp) {
; CHECK-LABEL: @offset_sample_c_cd_cl_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[CLAMP:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5507,7 +5507,7 @@ main_body:
define amdgpu_kernel void @offset_sample_l_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %lod) {
; CHECK-LABEL: @offset_sample_l_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32(i32 15, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5520,7 +5520,7 @@ main_body:
define amdgpu_kernel void @offset_sample_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t, float %lod) {
; CHECK-LABEL: @offset_sample_l_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5533,7 +5533,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_l_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %lod) {
; CHECK-LABEL: @offset_sample_c_l_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5546,7 +5546,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t, float %lod) {
; CHECK-LABEL: @offset_sample_c_l_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], float [[LOD:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5559,7 +5559,7 @@ main_body:
define amdgpu_kernel void @offset_sample_lz_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s) {
; CHECK-LABEL: @offset_sample_lz_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5572,7 +5572,7 @@ main_body:
define amdgpu_kernel void @offset_sample_lz_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; CHECK-LABEL: @offset_sample_lz_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5585,7 +5585,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_lz_o_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s) {
; CHECK-LABEL: @offset_sample_c_lz_o_1d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
@@ -5598,7 +5598,7 @@ main_body:
define amdgpu_kernel void @offset_sample_c_lz_o_2d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %zcompare, float %s, float %t) {
; CHECK-LABEL: @offset_sample_c_lz_o_2d(
; CHECK-NEXT: main_body:
-; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT: [[V:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: store <4 x float> [[V]], ptr addrspace(1) [[OUT:%.*]], align 16
; CHECK-NEXT: ret void
;
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll
index 65c89618e20ba0..92d11a2bbd7aca 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll
@@ -7,15 +7,15 @@
define amdgpu_ps void @image_store_1d_store_all_zeros(<8 x i32> inreg %rsrc, i32 %s) #0 {
; GCN-LABEL: @image_store_1d_store_all_zeros(
-; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32(float 0.000000e+00, i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32.v8i32(float 0.000000e+00, i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GCN-NEXT: ret void
;
; GFX12-LABEL: @image_store_1d_store_all_zeros(
-; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32(float 0.000000e+00, i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32.v8i32(float 0.000000e+00, i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX12-NEXT: ret void
;
; GFXUNKNOWN-LABEL: @image_store_1d_store_all_zeros(
-; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> zeroinitializer, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> zeroinitializer, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFXUNKNOWN-NEXT: ret void
;
call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> zeroinitializer, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
@@ -24,17 +24,17 @@ define amdgpu_ps void @image_store_1d_store_all_zeros(<8 x i32> inreg %rsrc, i32
define amdgpu_ps void @image_store_1d_store_insert_zeros_at_end(<8 x i32> inreg %rsrc, float %vdata1, i32 %s) #0 {
; GCN-LABEL: @image_store_1d_store_insert_zeros_at_end(
-; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32(float [[VDATA1:%.*]], i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32.v8i32(float [[VDATA1:%.*]], i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GCN-NEXT: ret void
;
; GFX12-LABEL: @image_store_1d_store_insert_zeros_at_end(
; GFX12-NEXT: [[NEWVDATA4:%.*]] = insertelement <4 x float> <float poison, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00>, float [[VDATA1:%.*]], i64 0
-; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> [[NEWVDATA4]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> [[NEWVDATA4]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX12-NEXT: ret void
;
; GFXUNKNOWN-LABEL: @image_store_1d_store_insert_zeros_at_end(
; GFXUNKNOWN-NEXT: [[NEWVDATA4:%.*]] = insertelement <4 x float> <float poison, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00>, float [[VDATA1:%.*]], i64 0
-; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> [[NEWVDATA4]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> [[NEWVDATA4]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFXUNKNOWN-NEXT: ret void
;
%newvdata1 = insertelement <4 x float> undef, float %vdata1, i32 0
@@ -49,19 +49,19 @@ define amdgpu_ps void @image_store_mip_1d_store_insert_zeros_at_end(<8 x i32> in
; GCN-LABEL: @image_store_mip_1d_store_insert_zeros_at_end(
; GCN-NEXT: [[TMP1:%.*]] = insertelement <3 x float> <float 0.000000e+00, float poison, float poison>, float [[VDATA1:%.*]], i64 1
; GCN-NEXT: [[TMP2:%.*]] = insertelement <3 x float> [[TMP1]], float [[VDATA2:%.*]], i64 2
-; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.v3f32.i32(<3 x float> [[TMP2]], i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.v3f32.i32.v8i32(<3 x float> [[TMP2]], i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GCN-NEXT: ret void
;
; GFX12-LABEL: @image_store_mip_1d_store_insert_zeros_at_end(
; GFX12-NEXT: [[TMP1:%.*]] = insertelement <3 x float> <float 0.000000e+00, float poison, float poison>, float [[VDATA1:%.*]], i64 1
; GFX12-NEXT: [[TMP2:%.*]] = insertelement <3 x float> [[TMP1]], float [[VDATA2:%.*]], i64 2
-; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.v3f32.i32(<3 x float> [[TMP2]], i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.v3f32.i32.v8i32(<3 x float> [[TMP2]], i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX12-NEXT: ret void
;
; GFXUNKNOWN-LABEL: @image_store_mip_1d_store_insert_zeros_at_end(
; GFXUNKNOWN-NEXT: [[TMP1:%.*]] = insertelement <4 x float> <float 0.000000e+00, float poison, float poison, float 0.000000e+00>, float [[VDATA1:%.*]], i64 1
; GFXUNKNOWN-NEXT: [[NEWVDATA4:%.*]] = insertelement <4 x float> [[TMP1]], float [[VDATA2:%.*]], i64 2
-; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> [[NEWVDATA4]], i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> [[NEWVDATA4]], i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFXUNKNOWN-NEXT: ret void
;
%newvdata1 = insertelement <4 x float> undef, float 0.0, i32 0
@@ -148,17 +148,17 @@ define amdgpu_ps void @struct_tbuffer_store_insert_undefs(<4 x i32> inreg %a, fl
define amdgpu_ps void @image_store_1d_store_shufflevector_same(<8 x i32> inreg %rsrc, <4 x float> %vdata1, i32 %s) #0 {
; GCN-LABEL: @image_store_1d_store_shufflevector_same(
; GCN-NEXT: [[DATA:%.*]] = shufflevector <4 x float> [[VDATA1:%.*]], <4 x float> poison, <4 x i32> zeroinitializer
-; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> [[DATA]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> [[DATA]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GCN-NEXT: ret void
;
; GFX12-LABEL: @image_store_1d_store_shufflevector_same(
; GFX12-NEXT: [[TMP1:%.*]] = extractelement <4 x float> [[VDATA1:%.*]], i64 0
-; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32(float [[TMP1]], i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.f32.i32.v8i32(float [[TMP1]], i32 1, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX12-NEXT: ret void
;
; GFXUNKNOWN-LABEL: @image_store_1d_store_shufflevector_same(
; GFXUNKNOWN-NEXT: [[DATA:%.*]] = shufflevector <4 x float> [[VDATA1:%.*]], <4 x float> poison, <4 x i32> zeroinitializer
-; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> [[DATA]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> [[DATA]], i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFXUNKNOWN-NEXT: ret void
;
%data = shufflevector <4 x float> %vdata1, <4 x float> poison, <4 x i32> <i32 0, i32 0, i32 0, i32 0>
@@ -168,15 +168,15 @@ define amdgpu_ps void @image_store_1d_store_shufflevector_same(<8 x i32> inreg %
define amdgpu_ps void @image_store_1d_store_shufflevector(<8 x i32> inreg %rsrc, float %vdata1, i32 %s) #0 {
; GCN-LABEL: @image_store_1d_store_shufflevector(
-; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> <float 2.000000e+00, float 2.000000e+00, float 5.000000e+00, float 2.000000e+00>, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GCN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> <float 2.000000e+00, float 2.000000e+00, float 5.000000e+00, float 2.000000e+00>, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GCN-NEXT: ret void
;
; GFX12-LABEL: @image_store_1d_store_shufflevector(
-; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.v3f32.i32(<3 x float> <float 2.000000e+00, float 2.000000e+00, float 5.000000e+00>, i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX12-NEXT: call void @llvm.amdgcn.image.store.1d.v3f32.i32.v8i32(<3 x float> <float 2.000000e+00, float 2.000000e+00, float 5.000000e+00>, i32 7, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX12-NEXT: ret void
;
; GFXUNKNOWN-LABEL: @image_store_1d_store_shufflevector(
-; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> <float 2.000000e+00, float 2.000000e+00, float 5.000000e+00, float 2.000000e+00>, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> <float 2.000000e+00, float 2.000000e+00, float 5.000000e+00, float 2.000000e+00>, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFXUNKNOWN-NEXT: ret void
;
%data = shufflevector <4 x float> <float 2.0, float 1.0, float 2.0, float 5.0>, <4 x float> poison, <4 x i32> <i32 0, i32 0, i32 3, i32 2>
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll b/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll
index d8d3c25688367e..6d40062a3fda7e 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll
@@ -7,13 +7,13 @@
define amdgpu_ps half @image_sample_2d_fptrunc_to_d16(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; GFX7-LABEL: @image_sample_2d_fptrunc_to_d16(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[TEX:%.*]] = call float @llvm.amdgcn.image.sample.lz.2d.f32.f32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX7-NEXT: [[TEX:%.*]] = call float @llvm.amdgcn.image.sample.lz.2d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX7-NEXT: [[TEX_HALF:%.*]] = fptrunc float [[TEX]] to half
; GFX7-NEXT: ret half [[TEX_HALF]]
;
; GFX81PLUS-LABEL: @image_sample_2d_fptrunc_to_d16(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[TEX:%.*]] = call half @llvm.amdgcn.image.sample.lz.2d.f16.f32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX81PLUS-NEXT: [[TEX:%.*]] = call half @llvm.amdgcn.image.sample.lz.2d.f16.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX81PLUS-NEXT: ret half [[TEX]]
;
main_body:
@@ -25,7 +25,7 @@ main_body:
define amdgpu_ps half @image_sample_2d_v2f32(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; GFX7-LABEL: @image_sample_2d_v2f32(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[TEX:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.2d.v2f32.f32(i32 3, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX7-NEXT: [[TEX:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.2d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX7-NEXT: [[TEX_2_HALF:%.*]] = fptrunc <2 x float> [[TEX]] to <2 x half>
; GFX7-NEXT: [[TEX_HALF_0:%.*]] = extractelement <2 x half> [[TEX_2_HALF]], i64 0
; GFX7-NEXT: [[TEX_HALF_1:%.*]] = extractelement <2 x half> [[TEX_2_HALF]], i64 1
@@ -34,7 +34,7 @@ define amdgpu_ps half @image_sample_2d_v2f32(<8 x i32> inreg %rsrc, <4 x i32> in
;
; GFX81PLUS-LABEL: @image_sample_2d_v2f32(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[TEX:%.*]] = call <2 x half> @llvm.amdgcn.image.sample.lz.2d.v2f16.f32(i32 3, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX81PLUS-NEXT: [[TEX:%.*]] = call <2 x half> @llvm.amdgcn.image.sample.lz.2d.v2f16.f32.v8i32.v4i32(i32 3, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX81PLUS-NEXT: [[TEX_HALF_0:%.*]] = extractelement <2 x half> [[TEX]], i64 0
; GFX81PLUS-NEXT: [[TEX_HALF_1:%.*]] = extractelement <2 x half> [[TEX]], i64 1
; GFX81PLUS-NEXT: [[ADDF_SUM_0:%.*]] = fadd half [[TEX_HALF_0]], [[TEX_HALF_1]]
@@ -52,7 +52,7 @@ main_body:
define amdgpu_ps half @image_sample_2d_v3f32(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; GFX7-LABEL: @image_sample_2d_v3f32(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[TEX:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.lz.2d.v3f32.f32(i32 7, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX7-NEXT: [[TEX:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.lz.2d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX7-NEXT: [[TEX_3_HALF:%.*]] = fptrunc <3 x float> [[TEX]] to <3 x half>
; GFX7-NEXT: [[TEX_HALF_0:%.*]] = extractelement <3 x half> [[TEX_3_HALF]], i64 0
; GFX7-NEXT: [[TEX_HALF_1:%.*]] = extractelement <3 x half> [[TEX_3_HALF]], i64 1
@@ -63,7 +63,7 @@ define amdgpu_ps half @image_sample_2d_v3f32(<8 x i32> inreg %rsrc, <4 x i32> in
;
; GFX81PLUS-LABEL: @image_sample_2d_v3f32(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[TEX:%.*]] = call <3 x half> @llvm.amdgcn.image.sample.lz.2d.v3f16.f32(i32 7, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX81PLUS-NEXT: [[TEX:%.*]] = call <3 x half> @llvm.amdgcn.image.sample.lz.2d.v3f16.f32.v8i32.v4i32(i32 7, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX81PLUS-NEXT: [[TEX_HALF_0:%.*]] = extractelement <3 x half> [[TEX]], i64 0
; GFX81PLUS-NEXT: [[TEX_HALF_1:%.*]] = extractelement <3 x half> [[TEX]], i64 1
; GFX81PLUS-NEXT: [[TEX_HALF_2:%.*]] = extractelement <3 x half> [[TEX]], i64 2
@@ -85,7 +85,7 @@ main_body:
define amdgpu_ps half @image_sample_2d_v4f32(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t) {
; GFX7-LABEL: @image_sample_2d_v4f32(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[TEX:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX7-NEXT: [[TEX:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX7-NEXT: [[TEX_4_HALF:%.*]] = fptrunc <4 x float> [[TEX]] to <4 x half>
; GFX7-NEXT: [[TEX_HALF_0:%.*]] = extractelement <4 x half> [[TEX_4_HALF]], i64 0
; GFX7-NEXT: [[TEX_HALF_1:%.*]] = extractelement <4 x half> [[TEX_4_HALF]], i64 1
@@ -98,7 +98,7 @@ define amdgpu_ps half @image_sample_2d_v4f32(<8 x i32> inreg %rsrc, <4 x i32> in
;
; GFX81PLUS-LABEL: @image_sample_2d_v4f32(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[TEX:%.*]] = call <4 x half> @llvm.amdgcn.image.sample.lz.2d.v4f16.f32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX81PLUS-NEXT: [[TEX:%.*]] = call <4 x half> @llvm.amdgcn.image.sample.lz.2d.v4f16.f32.v8i32.v4i32(i32 15, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX81PLUS-NEXT: [[TEX_HALF_0:%.*]] = extractelement <4 x half> [[TEX]], i64 0
; GFX81PLUS-NEXT: [[TEX_HALF_1:%.*]] = extractelement <4 x half> [[TEX]], i64 1
; GFX81PLUS-NEXT: [[TEX_HALF_2:%.*]] = extractelement <4 x half> [[TEX]], i64 2
@@ -124,7 +124,7 @@ main_body:
define amdgpu_ps half @image_gather4_2d_v4f32(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s, half %t) {
; GFX7-LABEL: @image_gather4_2d_v4f32(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[TEX:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f16(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX7-NEXT: [[TEX:%.*]] = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f16.v8i32.v4i32(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX7-NEXT: [[TEX_4_HALF:%.*]] = fptrunc <4 x float> [[TEX]] to <4 x half>
; GFX7-NEXT: [[TEX_HALF_0:%.*]] = extractelement <4 x half> [[TEX_4_HALF]], i64 0
; GFX7-NEXT: [[TEX_HALF_1:%.*]] = extractelement <4 x half> [[TEX_4_HALF]], i64 1
@@ -137,7 +137,7 @@ define amdgpu_ps half @image_gather4_2d_v4f32(<8 x i32> inreg %rsrc, <4 x i32> i
;
; GFX81PLUS-LABEL: @image_gather4_2d_v4f32(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[TEX:%.*]] = call <4 x half> @llvm.amdgcn.image.gather4.2d.v4f16.f16(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
+; GFX81PLUS-NEXT: [[TEX:%.*]] = call <4 x half> @llvm.amdgcn.image.gather4.2d.v4f16.f16.v8i32.v4i32(i32 1, half [[S:%.*]], half [[T:%.*]], <8 x i32> [[RSRC:%.*]], <4 x i32> [[SAMP:%.*]], i1 false, i32 0, i32 0)
; GFX81PLUS-NEXT: [[TEX_HALF_0:%.*]] = extractelement <4 x half> [[TEX]], i64 0
; GFX81PLUS-NEXT: [[TEX_HALF_1:%.*]] = extractelement <4 x half> [[TEX]], i64 1
; GFX81PLUS-NEXT: [[TEX_HALF_2:%.*]] = extractelement <4 x half> [[TEX]], i64 2
@@ -162,12 +162,12 @@ main_body:
define amdgpu_ps half @load_1d(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-LABEL: @load_1d(
-; GFX7-NEXT: [[S_FLOAT:%.*]] = call float @llvm.amdgcn.image.load.1d.f32.i16(i32 1, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[S_FLOAT:%.*]] = call float @llvm.amdgcn.image.load.1d.f32.i16.v8i32(i32 1, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[S_HALF:%.*]] = fptrunc float [[S_FLOAT]] to half
; GFX7-NEXT: ret half [[S_HALF]]
;
; GFX81PLUS-LABEL: @load_1d(
-; GFX81PLUS-NEXT: [[S_FLOAT:%.*]] = call half @llvm.amdgcn.image.load.1d.f16.i16(i32 1, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[S_FLOAT:%.*]] = call half @llvm.amdgcn.image.load.1d.f16.i16.v8i32(i32 1, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: ret half [[S_FLOAT]]
;
%s_float = call float @llvm.amdgcn.image.load.1d.f32.i16(i32 1, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
@@ -177,7 +177,7 @@ define amdgpu_ps half @load_1d(i16 %s, <8 x i32> inreg %rsrc) {
define amdgpu_ps half @load_1d_v2(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-LABEL: @load_1d_v2(
-; GFX7-NEXT: [[V2_FLOAT:%.*]] = call <2 x float> @llvm.amdgcn.image.load.1d.v2f32.i16(i32 3, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[V2_FLOAT:%.*]] = call <2 x float> @llvm.amdgcn.image.load.1d.v2f32.i16.v8i32(i32 3, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[V2_HALF:%.*]] = fptrunc <2 x float> [[V2_FLOAT]] to <2 x half>
; GFX7-NEXT: [[S0:%.*]] = extractelement <2 x half> [[V2_HALF]], i64 0
; GFX7-NEXT: [[S1:%.*]] = extractelement <2 x half> [[V2_HALF]], i64 1
@@ -185,7 +185,7 @@ define amdgpu_ps half @load_1d_v2(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-NEXT: ret half [[ADDF_SUM_0]]
;
; GFX81PLUS-LABEL: @load_1d_v2(
-; GFX81PLUS-NEXT: [[V2_FLOAT:%.*]] = call <2 x half> @llvm.amdgcn.image.load.1d.v2f16.i16(i32 3, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[V2_FLOAT:%.*]] = call <2 x half> @llvm.amdgcn.image.load.1d.v2f16.i16.v8i32(i32 3, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: [[S0:%.*]] = extractelement <2 x half> [[V2_FLOAT]], i64 0
; GFX81PLUS-NEXT: [[S1:%.*]] = extractelement <2 x half> [[V2_FLOAT]], i64 1
; GFX81PLUS-NEXT: [[ADDF_SUM_0:%.*]] = fadd half [[S0]], [[S1]]
@@ -201,7 +201,7 @@ define amdgpu_ps half @load_1d_v2(i16 %s, <8 x i32> inreg %rsrc) {
define amdgpu_ps half @load_1d_v3(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-LABEL: @load_1d_v3(
-; GFX7-NEXT: [[V3_FLOAT:%.*]] = call <3 x float> @llvm.amdgcn.image.load.1d.v3f32.i16(i32 7, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[V3_FLOAT:%.*]] = call <3 x float> @llvm.amdgcn.image.load.1d.v3f32.i16.v8i32(i32 7, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[V3_HALF:%.*]] = fptrunc <3 x float> [[V3_FLOAT]] to <3 x half>
; GFX7-NEXT: [[S0:%.*]] = extractelement <3 x half> [[V3_HALF]], i64 0
; GFX7-NEXT: [[S1:%.*]] = extractelement <3 x half> [[V3_HALF]], i64 1
@@ -211,7 +211,7 @@ define amdgpu_ps half @load_1d_v3(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-NEXT: ret half [[ADDF_SUM_1]]
;
; GFX81PLUS-LABEL: @load_1d_v3(
-; GFX81PLUS-NEXT: [[V3_FLOAT:%.*]] = call <3 x half> @llvm.amdgcn.image.load.1d.v3f16.i16(i32 7, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[V3_FLOAT:%.*]] = call <3 x half> @llvm.amdgcn.image.load.1d.v3f16.i16.v8i32(i32 7, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: [[S0:%.*]] = extractelement <3 x half> [[V3_FLOAT]], i64 0
; GFX81PLUS-NEXT: [[S1:%.*]] = extractelement <3 x half> [[V3_FLOAT]], i64 1
; GFX81PLUS-NEXT: [[S2:%.*]] = extractelement <3 x half> [[V3_FLOAT]], i64 2
@@ -231,7 +231,7 @@ define amdgpu_ps half @load_1d_v3(i16 %s, <8 x i32> inreg %rsrc) {
define amdgpu_ps half @load_1d_v4(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-LABEL: @load_1d_v4(
-; GFX7-NEXT: [[V4_FLOAT:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[V4_FLOAT:%.*]] = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[V4_HALF:%.*]] = fptrunc <4 x float> [[V4_FLOAT]] to <4 x half>
; GFX7-NEXT: [[S0:%.*]] = extractelement <4 x half> [[V4_HALF]], i64 0
; GFX7-NEXT: [[S1:%.*]] = extractelement <4 x half> [[V4_HALF]], i64 1
@@ -243,7 +243,7 @@ define amdgpu_ps half @load_1d_v4(i16 %s, <8 x i32> inreg %rsrc) {
; GFX7-NEXT: ret half [[ADDF_SUM_2]]
;
; GFX81PLUS-LABEL: @load_1d_v4(
-; GFX81PLUS-NEXT: [[V4_FLOAT:%.*]] = call <4 x half> @llvm.amdgcn.image.load.1d.v4f16.i16(i32 15, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[V4_FLOAT:%.*]] = call <4 x half> @llvm.amdgcn.image.load.1d.v4f16.i16.v8i32(i32 15, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: [[S0:%.*]] = extractelement <4 x half> [[V4_FLOAT]], i64 0
; GFX81PLUS-NEXT: [[S1:%.*]] = extractelement <4 x half> [[V4_FLOAT]], i64 1
; GFX81PLUS-NEXT: [[S2:%.*]] = extractelement <4 x half> [[V4_FLOAT]], i64 2
@@ -268,13 +268,13 @@ define amdgpu_ps half @load_1d_v4(i16 %s, <8 x i32> inreg %rsrc) {
define amdgpu_ps half @load_2dmsaa(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %fragid) {
; GFX7-LABEL: @load_2dmsaa(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[S_FLOAT:%.*]] = call float @llvm.amdgcn.image.msaa.load.x.2dmsaa.f32.i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[S_FLOAT:%.*]] = call float @llvm.amdgcn.image.msaa.load.x.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[S_HALF:%.*]] = fptrunc float [[S_FLOAT]] to half
; GFX7-NEXT: ret half [[S_HALF]]
;
; GFX81PLUS-LABEL: @load_2dmsaa(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[S_FLOAT:%.*]] = call half @llvm.amdgcn.image.msaa.load.x.2dmsaa.f16.i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[S_FLOAT:%.*]] = call half @llvm.amdgcn.image.msaa.load.x.2dmsaa.f16.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: ret half [[S_FLOAT]]
;
main_body:
@@ -286,7 +286,7 @@ main_body:
define amdgpu_ps half @load_2dmsaa_v2(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %fragid) {
; GFX7-LABEL: @load_2dmsaa_v2(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[V2_FLOAT:%.*]] = call <2 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v2f32.i32(i32 3, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[V2_FLOAT:%.*]] = call <2 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v2f32.i32.v8i32(i32 3, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[V2_HALF:%.*]] = fptrunc <2 x float> [[V2_FLOAT]] to <2 x half>
; GFX7-NEXT: [[S0:%.*]] = extractelement <2 x half> [[V2_HALF]], i64 0
; GFX7-NEXT: [[S1:%.*]] = extractelement <2 x half> [[V2_HALF]], i64 1
@@ -295,7 +295,7 @@ define amdgpu_ps half @load_2dmsaa_v2(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32
;
; GFX81PLUS-LABEL: @load_2dmsaa_v2(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[V2_FLOAT:%.*]] = call <2 x half> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v2f16.i32(i32 3, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[V2_FLOAT:%.*]] = call <2 x half> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v2f16.i32.v8i32(i32 3, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: [[S0:%.*]] = extractelement <2 x half> [[V2_FLOAT]], i64 0
; GFX81PLUS-NEXT: [[S1:%.*]] = extractelement <2 x half> [[V2_FLOAT]], i64 1
; GFX81PLUS-NEXT: [[ADDF_SUM_0:%.*]] = fadd half [[S0]], [[S1]]
@@ -313,7 +313,7 @@ main_body:
define amdgpu_ps half @load_2dmsaa_v3(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %fragid) {
; GFX7-LABEL: @load_2dmsaa_v3(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[V3_FLOAT:%.*]] = call <3 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v3f32.i32(i32 7, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[V3_FLOAT:%.*]] = call <3 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v3f32.i32.v8i32(i32 7, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[V3_HALF:%.*]] = fptrunc <3 x float> [[V3_FLOAT]] to <3 x half>
; GFX7-NEXT: [[S0:%.*]] = extractelement <3 x half> [[V3_HALF]], i64 0
; GFX7-NEXT: [[S1:%.*]] = extractelement <3 x half> [[V3_HALF]], i64 1
@@ -324,7 +324,7 @@ define amdgpu_ps half @load_2dmsaa_v3(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32
;
; GFX81PLUS-LABEL: @load_2dmsaa_v3(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[V3_FLOAT:%.*]] = call <3 x half> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v3f16.i32(i32 7, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[V3_FLOAT:%.*]] = call <3 x half> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v3f16.i32.v8i32(i32 7, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: [[S0:%.*]] = extractelement <3 x half> [[V3_FLOAT]], i64 0
; GFX81PLUS-NEXT: [[S1:%.*]] = extractelement <3 x half> [[V3_FLOAT]], i64 1
; GFX81PLUS-NEXT: [[S2:%.*]] = extractelement <3 x half> [[V3_FLOAT]], i64 2
@@ -346,7 +346,7 @@ main_body:
define amdgpu_ps half @load_2dmsaa_v4(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %fragid) {
; GFX7-LABEL: @load_2dmsaa_v4(
; GFX7-NEXT: main_body:
-; GFX7-NEXT: [[V4_FLOAT:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v4f32.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX7-NEXT: [[V4_FLOAT:%.*]] = call <4 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v4f32.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX7-NEXT: [[V4_HALF:%.*]] = fptrunc <4 x float> [[V4_FLOAT]] to <4 x half>
; GFX7-NEXT: [[S0:%.*]] = extractelement <4 x half> [[V4_HALF]], i64 0
; GFX7-NEXT: [[S1:%.*]] = extractelement <4 x half> [[V4_HALF]], i64 1
@@ -359,7 +359,7 @@ define amdgpu_ps half @load_2dmsaa_v4(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32
;
; GFX81PLUS-LABEL: @load_2dmsaa_v4(
; GFX81PLUS-NEXT: main_body:
-; GFX81PLUS-NEXT: [[V4_FLOAT:%.*]] = call <4 x half> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v4f16.i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
+; GFX81PLUS-NEXT: [[V4_FLOAT:%.*]] = call <4 x half> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v4f16.i32.v8i32(i32 15, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[FRAGID:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: [[S0:%.*]] = extractelement <4 x half> [[V4_FLOAT]], i64 0
; GFX81PLUS-NEXT: [[S1:%.*]] = extractelement <4 x half> [[V4_FLOAT]], i64 1
; GFX81PLUS-NEXT: [[S2:%.*]] = extractelement <4 x half> [[V4_FLOAT]], i64 2
diff --git a/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll b/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll
index 13d13bca56de6c..3de0ac17d840fc 100644
--- a/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll
+++ b/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll
@@ -40,22 +40,22 @@ declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>
define void @invalid_image_sample_1d_v4f32_f32(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc, i32 %dmask, i1 %bool, i32 %arg) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %dmask
- ; CHECK-NEXT: %data0 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 %dmask, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ ; CHECK-NEXT: %data0 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 %dmask, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%data0 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 %dmask, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i1 %bool
- ; CHECK-NEXT: %data1 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 %bool, i32 0, i32 0)
+ ; CHECK-NEXT: %data1 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 %bool, i32 0, i32 0)
%data1 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 %bool, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %arg
- ; CHECK-NEXT: %data2 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 %arg, i32 0)
+ ; CHECK-NEXT: %data2 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 %arg, i32 0)
%data2 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 %arg, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %arg
- ; CHECK-NEXT: %data3 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 %arg)
+ ; CHECK-NEXT: %data3 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 %arg)
%data3 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 %arg)
ret void
}
@@ -340,17 +340,17 @@ declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32, i32, <8 x i32>, i3
define amdgpu_ps void @load_1d(<8 x i32> inreg %rsrc, i32 %s, i32 %var) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 %var, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ ; CHECK-NEXT: %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 %var, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
%val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 %var, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %var, i32 0)
+ ; CHECK-NEXT: %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %var, i32 0)
%val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %var, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 %var)
+ ; CHECK-NEXT: %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 %var)
%val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 %var)
ret void
}
@@ -359,17 +359,17 @@ declare {<4 x float>,i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32, i32, <8 x
define amdgpu_ps void @load_1d_tfe(<8 x i32> inreg %rsrc, ptr addrspace(1) inreg %out, i32 %s, i32 %val) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val0 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 %val, i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
+ ; CHECK-NEXT: %val0 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 %val, i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
%val0 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 %val, i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val1 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
+ ; CHECK-NEXT: %val1 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
%val1 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val2 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 1, i32 %val)
+ ; CHECK-NEXT: %val2 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 1, i32 %val)
%val2 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 1, i32 %val)
ret void
}
@@ -378,22 +378,22 @@ declare {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32, float,
define amdgpu_ps void @sample_1d_tfe(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, ptr addrspace(1) inreg %out, float %s, i32 %var, i1 %bool) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val0 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 0)
+ ; CHECK-NEXT: %val0 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 0)
%val0 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i1 %bool
- ; CHECK-NEXT: %val1 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 %bool, i32 1, i32 0)
+ ; CHECK-NEXT: %val1 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 %bool, i32 1, i32 0)
%val1 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 %bool, i32 1, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val2 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 %var, i32 0)
+ ; CHECK-NEXT: %val2 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 %var, i32 0)
%val2 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 %var, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val3 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 %var)
+ ; CHECK-NEXT: %val3 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 %var)
%val3 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 %var)
ret void
}
@@ -402,17 +402,17 @@ declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32, i16, <8 x i32>, i3
define amdgpu_ps void @load_1d_a16(<8 x i32> inreg %rsrc, <2 x i16> %coords, i16 %s, i32 %var) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 %var, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ ; CHECK-NEXT: %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 %var, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
%val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 %var, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 %var, i32 0)
+ ; CHECK-NEXT: %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 %var, i32 0)
%val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 %var, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
- ; CHECK-NEXT: %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 %var)
+ ; CHECK-NEXT: %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 %var)
%val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 %var)
ret void
}
@@ -430,12 +430,12 @@ declare i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32, i32, <8 x i32>, i32,
define amdgpu_ps void @atomic_swap_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %val) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val0 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
+ ; CHECK-NEXT: %val0 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
%val0 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val1 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
+ ; CHECK-NEXT: %val1 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
%val1 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
ret void
}
@@ -444,12 +444,12 @@ declare i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32, i32, i32, <8 x i32
define amdgpu_ps void @atomic_cmpswap_1d(<8 x i32> inreg %rsrc, i32 %cmp, i32 %swap, i32 %s, i32 %val) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val0 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
+ ; CHECK-NEXT: %val0 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
%val0 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
- ; CHECK-NEXT: %val1 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
+ ; CHECK-NEXT: %val1 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
%val1 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
ret void
}
>From 222c0a19cfcf4032bab140e7e134d11aeba568dc Mon Sep 17 00:00:00 2001
From: Stefan Stipanovic <Stefan.Stipanovic at amd.com>
Date: Tue, 24 Sep 2024 14:34:19 +0200
Subject: [PATCH 2/3] fix the called intrinsic name to match in the tests
---
.../AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll | 224 +++----
...mdgcn-demanded-vector-elts-inseltpoison.ll | 316 +++++-----
.../AMDGPU/amdgcn-demanded-vector-elts.ll | 316 +++++-----
.../AMDGPU/amdgcn-intrinsics-gfx8.ll | 24 +-
.../InstCombine/AMDGPU/amdgcn-intrinsics.ll | 586 +++++++++---------
.../amdgcn-simplify-image-buffer-stores.ll | 38 +-
.../InstCombine/AMDGPU/image-d16.ll | 26 +-
llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll | 56 +-
8 files changed, 793 insertions(+), 793 deletions(-)
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll
index 854a9ed4e01738..4df6fcd584ff61 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.load.2dmsaa.ll
@@ -32,10 +32,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -72,10 +72,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask2(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 2, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 2, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -112,10 +112,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask4(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 4, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 4, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -152,10 +152,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask8(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 8, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 8, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -192,10 +192,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_reverse(<8 x i32> inreg %rsrc, i
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -232,10 +232,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_vaddr(<8 x i32> inreg %rsrc, i32
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s0, i32 %t0, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s1, i32 %t1, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s2, i32 %t2, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s3, i32 %t3, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s0, i32 %t0, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s1, i32 %t1, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s2, i32 %t2, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s3, i32 %t3, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -288,14 +288,14 @@ define amdgpu_ps [8 x float] @load_2dmsaa_v8f32(<8 x i32> inreg %rsrc, i32 %s, i
; MSAA-NEXT: ret [8 x float] [[I15]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
- %i4 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i5 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i6 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i7 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i4 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i5 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i6 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i7 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i8 = insertvalue [8 x float] undef, float %i, 0
%i9 = insertvalue [8 x float] %i8, float %i1, 1
%i10 = insertvalue [8 x float] %i9, float %i2, 2
@@ -336,13 +336,13 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_interleaved(<8 x i32> inreg %rsr
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
%i1 = insertvalue [4 x float] undef, float %i, 0
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
%i3 = insertvalue [4 x float] %i1, float %i2, 1
- %i4 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i4 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
%i5 = insertvalue [4 x float] %i3, float %i4, 2
- %i6 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i6 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i7 = insertvalue [4 x float] %i5, float %i6, 3
ret [4 x float] %i7
}
@@ -368,8 +368,8 @@ define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId01(<8 x i32> inreg %rsrc,
; MSAA-NEXT: ret [2 x float] [[I3]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
%i2 = insertvalue [2 x float] undef, float %i, 0
%i3 = insertvalue [2 x float] %i2, float %i1, 1
ret [2 x float] %i3
@@ -396,8 +396,8 @@ define amdgpu_ps [2 x float] @load_2dmsaa_v2f32_fragId23(<8 x i32> inreg %rsrc,
; MSAA-NEXT: ret [2 x float] [[I3]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i2 = insertvalue [2 x float] undef, float %i, 0
%i3 = insertvalue [2 x float] %i2, float %i1, 1
ret [2 x float] %i3
@@ -425,8 +425,8 @@ define amdgpu_ps [2 x <2 x float>] @load_2dmsaa_v2v2f32_dmask3(<8 x i32> inreg %
; MSAA-NEXT: ret [2 x <2 x float>] [[I5]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [2 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [2 x <2 x float>] %i4, <2 x float> %i1, 1
ret [2 x <2 x float>] %i5
@@ -477,10 +477,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask3(<8 x i32> inreg %
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -533,10 +533,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask5(<8 x i32> inreg %
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 5, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 5, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -589,10 +589,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask6(<8 x i32> inreg %
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 6, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 6, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -645,10 +645,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask9(<8 x i32> inreg %
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 9, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 9, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -701,10 +701,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask10(<8 x i32> inreg
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 10, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 10, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -757,10 +757,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2dmsaa_v4v2f32_dmask12(<8 x i32> inreg
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32 12, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32 12, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -789,8 +789,8 @@ define amdgpu_ps [2 x half] @load_2dmsaa_v2f16_fragId01(<8 x i32> inreg %rsrc, i
; MSAA-NEXT: ret [2 x half] [[I3]]
;
main_body:
- %i = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call half @llvm.amdgcn.image.load.2dmsaa.f16.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
%i2 = insertvalue [2 x half] undef, half %i, 0
%i3 = insertvalue [2 x half] %i2, half %i1, 1
ret [2 x half] %i3
@@ -825,10 +825,10 @@ define amdgpu_ps [4 x float] @load_2darraymsaa_v4f32_dmask1(<8 x i32> inreg %rsr
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 %slice, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -881,10 +881,10 @@ define amdgpu_ps [4 x <2 x float>] @load_2darraymsaa_v4v2f32_dmask3(<8 x i32> in
; MSAA-NEXT: ret [4 x <2 x float>] [[I7]]
;
main_body:
- %i = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 %slice, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <2 x float>] undef, <2 x float> %i, 0
%i5 = insertvalue [4 x <2 x float>] %i4, <2 x float> %i1, 1
%i6 = insertvalue [4 x <2 x float>] %i5, <2 x float> %i2, 2
@@ -948,10 +948,10 @@ define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7(<8 x i32> inreg %
; MSAA-NEXT: ret [4 x <3 x float>] [[I7]]
;
main_body:
- %i = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <3 x float>] undef, <3 x float> %i, 0
%i5 = insertvalue [4 x <3 x float>] %i4, <3 x float> %i1, 1
%i6 = insertvalue [4 x <3 x float>] %i5, <3 x float> %i2, 2
@@ -1015,10 +1015,10 @@ define amdgpu_ps [4 x <3 x float>] @load_2dmsaa_v4v3f32_dmask7_group1(<8 x i32>
; MSAA-NEXT: ret [4 x <3 x float>] [[I7]]
;
main_body:
- %i = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 4, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 5, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 6, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 7, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 4, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 5, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 6, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 7, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x <3 x float>] undef, <3 x float> %i, 0
%i5 = insertvalue [4 x <3 x float>] %i4, <3 x float> %i1, 1
%i6 = insertvalue [4 x <3 x float>] %i5, <3 x float> %i2, 2
@@ -1058,11 +1058,11 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_sections(<8 x i32> inreg %rsrc,
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
call void @llvm.amdgcn.image.store.2dmsaa.f32.i32(float %vdata, i32 1, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -1149,10 +1149,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_blocks(<8 x i32> inreg %rsrc, i3
; MSAA-NEXT: ret [4 x float] [[I25]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -1160,20 +1160,20 @@ main_body:
%i8 = trunc i32 %cond to i1
br i1 %i8, label %if_equal, label %if_unequal
if_equal:
- %i9 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i10 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i11 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i12 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i9 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i10 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i11 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i12 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i13 = insertvalue [4 x float] undef, float %i9, 0
%i14 = insertvalue [4 x float] %i13, float %i10, 1
%i15 = insertvalue [4 x float] %i14, float %i11, 2
%i16 = insertvalue [4 x float] %i15, float %i12, 3
br label %merge
if_unequal:
- %i17 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
- %i18 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
- %i19 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
- %i20 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
+ %i17 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %i18 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc, i32 0, i32 0)
+ %i19 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 2, <8 x i32> %rsrc, i32 0, i32 0)
+ %i20 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 3, <8 x i32> %rsrc, i32 0, i32 0)
%i21 = insertvalue [4 x float] undef, float %i17, 0
%i22 = insertvalue [4 x float] %i21, float %i18, 1
%i23 = insertvalue [4 x float] %i22, float %i19, 2
@@ -1214,10 +1214,10 @@ define amdgpu_ps [4 x float] @load_2dmsaa_v4f32_dmask1_different_rsrc(<8 x i32>
; MSAA-NEXT: ret [4 x float] [[I7]]
;
main_body:
- %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc1, i32 0, i32 0)
- %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc1, i32 0, i32 0)
- %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc2, i32 0, i32 0)
- %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc2, i32 0, i32 0)
+ %i = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc1, i32 0, i32 0)
+ %i1 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc1, i32 0, i32 0)
+ %i2 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc2, i32 0, i32 0)
+ %i3 = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 1, <8 x i32> %rsrc2, i32 0, i32 0)
%i4 = insertvalue [4 x float] undef, float %i, 0
%i5 = insertvalue [4 x float] %i4, float %i1, 1
%i6 = insertvalue [4 x float] %i5, float %i2, 2
@@ -1225,16 +1225,16 @@ main_body:
ret [4 x float] %i7
}
-declare float @llvm.amdgcn.image.load.2dmsaa.f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare <2 x float> @llvm.amdgcn.image.load.2dmsaa.v2f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare <3 x float> @llvm.amdgcn.image.load.2dmsaa.v3f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare float @llvm.amdgcn.image.load.2darraymsaa.f32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare <3 x float> @llvm.amdgcn.image.load.2darraymsaa.v3f32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare float @llvm.amdgcn.image.load.2darraymsaa.f32.i32.v8i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare <2 x float> @llvm.amdgcn.image.load.2darraymsaa.v2f32.i32.v8i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare <3 x float> @llvm.amdgcn.image.load.2darraymsaa.v3f32.i32.v8i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare half @llvm.amdgcn.image.load.2dmsaa.f16.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare half @llvm.amdgcn.image.load.2dmsaa.f16.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.2dmsaa.f32.i32(float, i32, i32, i32, i32, <8 x i32>, i32, i32)
+declare void @llvm.amdgcn.image.store.2dmsaa.f32.i32.v8i32(float, i32, i32, i32, i32, <8 x i32>, i32, i32)
attributes #0 = { nounwind readonly willreturn }
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll
index 0a1e5dc054ad59..ea6422e5ed8c2e 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll
@@ -4088,7 +4088,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32(float %vaddr, <8
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4101,7 +4101,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_tfe(float %vaddr,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 1, i32 0)
+ %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 1, i32 0)
%data.vec = extractvalue {<4 x float>,i32} %data, 0
%elt0 = extractelement <4 x float> %data.vec, i32 0
ret float %elt0
@@ -4115,7 +4115,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_lwe(float %vaddr,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 2, i32 0)
+ %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 2, i32 0)
%data.vec = extractvalue {<4 x float>,i32} %data, 0
%elt0 = extractelement <4 x float> %data.vec, i32 0
ret float %elt0
@@ -4126,7 +4126,7 @@ define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.2d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4137,7 +4137,7 @@ define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float %s, float %t, float %r, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 0, float %s, float %t, float %r, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4147,7 +4147,7 @@ define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(f
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1darray.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 1, float %s, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32 1, float %s, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4157,7 +4157,7 @@ define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 2, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 2, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4167,7 +4167,7 @@ define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 4, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 4, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4177,7 +4177,7 @@ define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 8, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 8, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 8, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4187,7 +4187,7 @@ define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 9, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 9, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4197,7 +4197,7 @@ define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4207,7 +4207,7 @@ define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4218,7 +4218,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <2 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <2 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4228,7 +4228,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4238,7 +4238,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4248,7 +4248,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4259,7 +4259,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <3 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4270,7 +4270,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4281,7 +4281,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4291,7 +4291,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_
; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4301,16 +4301,16 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_1111_image_sample_1d_
; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cl
@@ -4321,12 +4321,12 @@ define amdgpu_ps float @extract_elt1_image_sample_cl_2darray_v4f32_f32(float %s,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.2darray.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32 15, float %s, float %t, float %slice, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32.v8i32.v4i32(i32 15, float %s, float %t, float %slice, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 1
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d
@@ -4337,12 +4337,12 @@ define amdgpu_ps float @extract_elt2_image_sample_d_cube_v4f32_f32_f32(float %ds
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cube.f32.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32 15, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 2
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d.cl
@@ -4353,12 +4353,12 @@ define amdgpu_ps float @extract_elt3_image_sample_d_cl_1d_v4f32_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 3
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.l
@@ -4369,12 +4369,12 @@ define amdgpu_ps float @extract_elt1_dmask_0110_image_sample_l_1d_v2f32_f32(floa
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32 6, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32.v8i32.v4i32(i32 6, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <2 x float> %data, i32 1
ret float %elt0
}
-declare <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b
@@ -4385,12 +4385,12 @@ define amdgpu_ps float @extract_elt1_dmask_1001_image_sample_b_1d_v4f32_f32_f32(
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 9, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 9, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 1
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b.cl
@@ -4401,12 +4401,12 @@ define amdgpu_ps <2 x float> @extract_elt1_elt2_dmask_1101_image_sample_b_cl_1d_
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.b.cl.1d.v2f32.f32.f32.v8i32.v4i32(i32 12, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 13, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 13, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 1, i32 2>
ret <2 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.lz
@@ -4417,12 +4417,12 @@ define amdgpu_ps <2 x float> @extract_elt1_elt3_image_sample_lz_1d_v4f32_f32(flo
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.1d.v2f32.f32.v8i32.v4i32(i32 10, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 1, i32 3>
ret <2 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd
@@ -4433,12 +4433,12 @@ define amdgpu_ps <3 x float> @extract_elt1_elt2_elt3_image_sample_cd_1d_v4f32_f3
; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.cd.1d.v3f32.f32.f32.v8i32.v4i32(i32 14, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 1, i32 2, i32 3>
ret <3 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd.cl
@@ -4449,7 +4449,7 @@ define amdgpu_ps half @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 3
ret half %elt0
}
@@ -4459,7 +4459,7 @@ define amdgpu_ps half @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 2
ret half %elt0
}
@@ -4469,7 +4469,7 @@ define amdgpu_ps half @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 2, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 1
ret half %elt0
}
@@ -4480,7 +4480,7 @@ define amdgpu_ps <4 x half> @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32
; CHECK-NEXT: [[RES:%.*]] = shufflevector <3 x half> [[DATA]], <3 x half> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%res = shufflevector <4 x half> %data, <4 x half> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
ret <4 x half> %res
}
@@ -4491,7 +4491,7 @@ define amdgpu_ps <4 x half> @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32
; CHECK-NEXT: [[RES:%.*]] = shufflevector <2 x half> [[DATA]], <2 x half> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%res = shufflevector <4 x half> %data, <4 x half> poison, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
ret <4 x half> %res
}
@@ -4502,7 +4502,7 @@ define amdgpu_ps <4 x half> @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32
; CHECK-NEXT: [[RES:%.*]] = insertelement <4 x half> poison, half [[DATA]], i64 0
; CHECK-NEXT: ret <4 x half> [[RES]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%res = shufflevector <4 x half> %data, <4 x half> poison, <4 x i32> <i32 0, i32 4, i32 5, i32 6>
ret <4 x half> %res
}
@@ -4512,12 +4512,12 @@ define amdgpu_ps half @extract_elt0_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 0
ret half %elt0
}
-declare <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c
@@ -4528,12 +4528,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_1d_v4f32_f32(float %zcompare
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cl
@@ -4544,12 +4544,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cl_1d_v4f32_f32(float %zcomp
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d
@@ -4560,12 +4560,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_1d_v4f32_f32_f32(float %zc
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d.cl
@@ -4576,12 +4576,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_1d_v4f32_f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.l
@@ -4592,12 +4592,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_l_1d_v4f32_f32(float %zcompa
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b
@@ -4608,12 +4608,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_1d_v4f32_f32_f32(float %bi
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b.cl
@@ -4624,12 +4624,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_1d_v4f32_f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.lz
@@ -4640,12 +4640,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_lz_1d_v4f32_f32(float %zcomp
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd
@@ -4656,12 +4656,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_1d_v4f32_f32_f32(float %z
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd.cl
@@ -4672,12 +4672,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_1d_v4f32_f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.o
@@ -4688,12 +4688,12 @@ define amdgpu_ps float @extract_elt0_image_sample_o_1d_v4f32_f32(i32 %offset, fl
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cl.o
@@ -4704,12 +4704,12 @@ define amdgpu_ps float @extract_elt0_image_sample_cl_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d.o
@@ -4720,12 +4720,12 @@ define amdgpu_ps float @extract_elt0_image_sample_d_o_1d_v4f32_f32_f32(i32 %offs
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d.cl.o
@@ -4736,12 +4736,12 @@ define amdgpu_ps float @extract_elt0_image_sample_d_cl_o_1d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.l.o
@@ -4752,12 +4752,12 @@ define amdgpu_ps float @extract_elt0_image_sample_l_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b.o
@@ -4768,12 +4768,12 @@ define amdgpu_ps float @extract_elt0_image_sample_b_o_1d_v4f32_f32_f32(i32 %offs
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b.cl.o
@@ -4784,12 +4784,12 @@ define amdgpu_ps float @extract_elt0_image_sample_b_cl_o_1d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.lz.o
@@ -4800,12 +4800,12 @@ define amdgpu_ps float @extract_elt0_image_sample_lz_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd.o
@@ -4816,12 +4816,12 @@ define amdgpu_ps float @extract_elt0_image_sample_cd_o_1d_v4f32_f32_f32(i32 %off
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd.cl.o
@@ -4832,12 +4832,12 @@ define amdgpu_ps float @extract_elt0_image_sample_cd_cl_o_1d_v4f32_f32_f32(i32 %
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.o
@@ -4848,12 +4848,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cl.o
@@ -4864,12 +4864,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cl_o_1d_v4f32_f32(i32 %offse
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d.o
@@ -4880,12 +4880,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_o_1d_v4f32_f32_f32(i32 %of
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d.cl.o
@@ -4896,12 +4896,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_o_1d_v4f32_f32_f32(i32
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.l.o
@@ -4912,12 +4912,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_l_o_1d_v4f32_f32(i32 %offset
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b.o
@@ -4928,12 +4928,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_o_1d_v4f32_f32_f32(i32 %of
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b.cl.o
@@ -4944,12 +4944,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_o_1d_v4f32_f32_f32(i32
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.lz.o
@@ -4960,12 +4960,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_lz_o_1d_v4f32_f32(i32 %offse
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd.o
@@ -4976,12 +4976,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_o_1d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd.cl.o
@@ -4992,12 +4992,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_o_1d_v4f32_f32_f32(i32
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4
@@ -5011,12 +5011,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_2d_v4f32_f32(float %s, float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32 1, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32(i32 1, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.cl
@@ -5028,12 +5028,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_cl_2d_v4f32_f32(float %s, flo
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32 2, float %s, float %t, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32.v8i32(i32 2, float %s, float %t, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32.v8i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.l
@@ -5045,12 +5045,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_l_2d_v4f32_f32(float %s, floa
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32 4, float %s, float %t, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32(i32 4, float %s, float %t, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b
@@ -5062,12 +5062,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_2darray_v4f32_f32_f32(float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32 8, float %bias, float %s, float %t, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32.v8i32(i32 8, float %bias, float %s, float %t, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32.v8i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b.cl
@@ -5079,12 +5079,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_cl_cube_v4f32_f32_f32(float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32 1, float %bias, float %s, float %t, float %face, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32.v8i32(i32 1, float %bias, float %s, float %t, float %face, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32.v8i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.lz
@@ -5096,12 +5096,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_lz_2d_v4f32_f16(half %s, half
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32 1, half %s, half %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16.v8i32(i32 1, half %s, half %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32, half, half, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16.v8i32(i32, half, half, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.o
@@ -5113,12 +5113,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_o_2d_v4f32_f32(i32 %offset, f
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.cl.o
@@ -5130,12 +5130,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_cl_o_2d_v4f32_f32(i32 %offset
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.l.o
@@ -5147,12 +5147,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_l_o_2d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b.o
@@ -5164,12 +5164,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_o_2d_v4f32_f32_f32(i32 %off
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b.cl.o
@@ -5181,12 +5181,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_cl_o_2d_v4f32_f32_f32(i32 %
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.lz.o
@@ -5198,12 +5198,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_lz_o_2d_v4f32_f32(i32 %offset
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.o
@@ -5215,12 +5215,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_o_2d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.cl.o
@@ -5232,12 +5232,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_cl_o_2d_v4f32_f32(i32 %offs
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.l.o
@@ -5249,12 +5249,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_l_o_2d_v4f32_f32(i32 %offse
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.b.o
@@ -5266,12 +5266,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_b_o_2d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.b.cl.o
@@ -5283,12 +5283,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_b_cl_o_2d_v4f32_f32_f32(i32
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.lz.o
@@ -5300,12 +5300,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_lz_o_2d_v4f32_f32(i32 %offs
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.getlod
@@ -5316,12 +5316,12 @@ define amdgpu_ps float @extract_elt0_image_getlod_1d_v4f32_f32(float %s, <8 x i3
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getlod.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32.v8i32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32.v8i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.load
@@ -5332,12 +5332,12 @@ define amdgpu_ps float @extract_elt0_image_load_2dmsaa_v4f32_i32(i32 %s, i32 %t,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SAMPLE:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %sample, <8 x i32> %sampler, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 %sample, <8 x i32> %sampler, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.load.mip
@@ -5348,12 +5348,12 @@ define amdgpu_ps float @extract_elt0_image_load_mip_1d_v4f32_i32(i32 %s, i32 %mi
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.mip.1d.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32 15, i32 %s, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32, i32, i32, <8 x i32>, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.getresinfo
@@ -5364,12 +5364,12 @@ define amdgpu_ps float @extract_elt0_image_getresinfo_1d_v4f32_i32(i32 %mip, <8
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getresinfo.1d.f32.i32.v8i32(i32 1, i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32 15, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32.v8i32(i32 15, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #1
; --------------------------------------------------------------------
; TFE / LWE
@@ -5382,25 +5382,25 @@ define amdgpu_ps float @extract_elt0_tfe_image_load_1d_v4f32i32_i32(i32 %s, <8 x
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[RGBA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 1)
+ %data = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 1)
%rgba = extractvalue { <4 x float>, i32 } %data, 0
%elt0 = extractelement <4 x float> %rgba, i32 0
ret float %elt0
}
-declare {<4 x float>, i32} @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32, i32, <8 x i32>, i32, i32) #1
+declare {<4 x float>, i32} @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #1
define amdgpu_hs float @tfe_check_assert() #0 {
; CHECK-LABEL: @tfe_check_assert(
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2d.f32.i32.v8i32(i32 1, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call nsz <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 15, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1) #2
+ %data = call nsz <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32.v8i32(i32 15, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1) #2
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 immarg, i32, i32, <8 x i32>, i32 immarg, i32 immarg) #1
+declare <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32.v8i32(i32 immarg, i32, i32, <8 x i32>, i32 immarg, i32 immarg) #1
attributes #0 = { nounwind }
attributes #1 = { nounwind readonly }
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll
index 28403ffd33b99c..66184405f30922 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts.ll
@@ -4087,7 +4087,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32(float %vaddr, <8
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[VADDR:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4100,7 +4100,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_tfe(float %vaddr,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 1, i32 0)
+ %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 1, i32 0)
%data.vec = extractvalue {<4 x float>,i32} %data, 0
%elt0 = extractelement <4 x float> %data.vec, i32 0
ret float %elt0
@@ -4114,7 +4114,7 @@ define amdgpu_ps float @extract_elt0_image_sample_1d_v4f32_f32_lwe(float %vaddr,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA_VEC]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 2, i32 0)
+ %data = call {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 15, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 2, i32 0)
%data.vec = extractvalue {<4 x float>,i32} %data, 0
%elt0 = extractelement <4 x float> %data.vec, i32 0
ret float %elt0
@@ -4125,7 +4125,7 @@ define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.2d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[T:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4136,7 +4136,7 @@ define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float %s, float %t, float %r, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 0, float %s, float %t, float %r, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4146,7 +4146,7 @@ define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_1darray_v4f32_f32(f
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1darray.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], float [[SLICE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 1, float %s, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32 1, float %s, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4156,7 +4156,7 @@ define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 2, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 2, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4166,7 +4166,7 @@ define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 4, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 4, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4176,7 +4176,7 @@ define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 8, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 8, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 8, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4186,7 +4186,7 @@ define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 9, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 9, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4196,7 +4196,7 @@ define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4206,7 +4206,7 @@ define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_1d_v4f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
@@ -4217,7 +4217,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_1d_v4f32
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <2 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <2 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4227,7 +4227,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_1d_v4f32
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4237,7 +4237,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_1d_v4f32
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 3, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4247,7 +4247,7 @@ define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_1d_v4f32
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32.v8i32.v4i32(i32 5, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 0, i32 1>
ret <2 x float> %shuf
}
@@ -4258,7 +4258,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_1d_
; CHECK-NEXT: [[SHUF:%.*]] = insertelement <3 x float> poison, float [[DATA]], i64 0
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 1, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4269,7 +4269,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_1d_
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 3, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4280,7 +4280,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_1d_
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x float> [[DATA]], <2 x float> poison, <3 x i32> <i32 0, i32 1, i32 poison>
; CHECK-NEXT: ret <3 x float> [[SHUF]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 5, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4290,7 +4290,7 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_1d_
; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 7, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
@@ -4300,16 +4300,16 @@ define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_1111_image_sample_1d_
; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.1d.v3f32.f32.v8i32.v4i32(i32 7, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 0, i32 1, i32 2>
ret <3 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare {<4 x float>,i32} @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cl
@@ -4320,12 +4320,12 @@ define amdgpu_ps float @extract_elt1_image_sample_cl_2darray_v4f32_f32(float %s,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.2darray.f32.f32.v8i32.v4i32(i32 2, float [[S:%.*]], float [[T:%.*]], float [[SLICE:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32 15, float %s, float %t, float %slice, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32.v8i32.v4i32(i32 15, float %s, float %t, float %slice, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 1
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.2darray.v4f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d
@@ -4336,12 +4336,12 @@ define amdgpu_ps float @extract_elt2_image_sample_d_cube_v4f32_f32_f32(float %ds
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cube.f32.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DTDH:%.*]], float [[DSDV:%.*]], float [[DTDV:%.*]], float [[S:%.*]], float [[T:%.*]], float [[FACE:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32 15, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %face, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 2
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cube.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d.cl
@@ -4352,12 +4352,12 @@ define amdgpu_ps float @extract_elt3_image_sample_d_cl_1d_v4f32_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 3
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.l
@@ -4368,12 +4368,12 @@ define amdgpu_ps float @extract_elt1_dmask_0110_image_sample_l_1d_v2f32_f32(floa
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.1d.f32.f32.v8i32.v4i32(i32 4, float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32 6, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32.v8i32.v4i32(i32 6, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <2 x float> %data, i32 1
ret float %elt0
}
-declare <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <2 x float> @llvm.amdgcn.image.sample.l.1d.v2f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b
@@ -4384,12 +4384,12 @@ define amdgpu_ps float @extract_elt1_dmask_1001_image_sample_b_1d_v4f32_f32_f32(
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.1d.f32.f32.f32.v8i32.v4i32(i32 8, float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 9, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 9, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 1
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b.cl
@@ -4400,12 +4400,12 @@ define amdgpu_ps <2 x float> @extract_elt1_elt2_dmask_1101_image_sample_b_cl_1d_
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.b.cl.1d.v2f32.f32.f32.v8i32.v4i32(i32 12, float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 13, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 13, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 1, i32 2>
ret <2 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.lz
@@ -4416,12 +4416,12 @@ define amdgpu_ps <2 x float> @extract_elt1_elt3_image_sample_lz_1d_v4f32_f32(flo
; CHECK-NEXT: [[DATA:%.*]] = call <2 x float> @llvm.amdgcn.image.sample.lz.1d.v2f32.f32.v8i32.v4i32(i32 10, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <2 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <2 x i32> <i32 1, i32 3>
ret <2 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd
@@ -4432,12 +4432,12 @@ define amdgpu_ps <3 x float> @extract_elt1_elt2_elt3_image_sample_cd_1d_v4f32_f3
; CHECK-NEXT: [[DATA:%.*]] = call <3 x float> @llvm.amdgcn.image.sample.cd.1d.v3f32.f32.f32.v8i32.v4i32(i32 14, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret <3 x float> [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%shuf = shufflevector <4 x float> %data, <4 x float> poison, <3 x i32> <i32 1, i32 2, i32 3>
ret <3 x float> %shuf
}
-declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd.cl
@@ -4448,7 +4448,7 @@ define amdgpu_ps half @extract_elt3_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 8, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 3
ret half %elt0
}
@@ -4458,7 +4458,7 @@ define amdgpu_ps half @extract_elt2_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 4, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 2
ret half %elt0
}
@@ -4468,7 +4468,7 @@ define amdgpu_ps half @extract_elt1_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 2, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 1
ret half %elt0
}
@@ -4479,7 +4479,7 @@ define amdgpu_ps <4 x half> @extract_elt_to3_image_sample_cd_cl_1d_v4f16_f32_f32
; CHECK-NEXT: [[RES:%.*]] = shufflevector <3 x half> [[DATA]], <3 x half> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%res = shufflevector <4 x half> %data, <4 x half> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
ret <4 x half> %res
}
@@ -4490,7 +4490,7 @@ define amdgpu_ps <4 x half> @extract_elt_to2_image_sample_cd_cl_1d_v4f16_f32_f32
; CHECK-NEXT: [[RES:%.*]] = shufflevector <2 x half> [[DATA]], <2 x half> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; CHECK-NEXT: ret <4 x half> [[RES]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%res = shufflevector <4 x half> %data, <4 x half> poison, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
ret <4 x half> %res
}
@@ -4501,7 +4501,7 @@ define amdgpu_ps <4 x half> @extract_elt_to1_image_sample_cd_cl_1d_v4f16_f32_f32
; CHECK-NEXT: [[RES:%.*]] = insertelement <4 x half> poison, half [[DATA]], i64 0
; CHECK-NEXT: ret <4 x half> [[RES]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%res = shufflevector <4 x half> %data, <4 x half> poison, <4 x i32> <i32 0, i32 4, i32 5, i32 6>
ret <4 x half> %res
}
@@ -4511,12 +4511,12 @@ define amdgpu_ps half @extract_elt0_image_sample_cd_cl_1d_v4f16_f32_f32(float %d
; CHECK-NEXT: [[DATA:%.*]] = call half @llvm.amdgcn.image.sample.cd.cl.1d.f16.f32.f32.v8i32.v4i32(i32 1, float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret half [[DATA]]
;
- %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32 15, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x half> %data, i32 0
ret half %elt0
}
-declare <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x half> @llvm.amdgcn.image.sample.cd.cl.1d.v4f16.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c
@@ -4527,12 +4527,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_1d_v4f32_f32(float %zcompare
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cl
@@ -4543,12 +4543,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cl_1d_v4f32_f32(float %zcomp
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d
@@ -4559,12 +4559,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_1d_v4f32_f32_f32(float %zc
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d.cl
@@ -4575,12 +4575,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_1d_v4f32_f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.l
@@ -4591,12 +4591,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_l_1d_v4f32_f32(float %zcompa
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b
@@ -4607,12 +4607,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_1d_v4f32_f32_f32(float %bi
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b.cl
@@ -4623,12 +4623,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_1d_v4f32_f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.lz
@@ -4639,12 +4639,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_lz_1d_v4f32_f32(float %zcomp
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.1d.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd
@@ -4655,12 +4655,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_1d_v4f32_f32_f32(float %z
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd.cl
@@ -4671,12 +4671,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_1d_v4f32_f32_f32(float
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.1d.f32.f32.f32.v8i32.v4i32(i32 1, float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.o
@@ -4687,12 +4687,12 @@ define amdgpu_ps float @extract_elt0_image_sample_o_1d_v4f32_f32(i32 %offset, fl
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cl.o
@@ -4703,12 +4703,12 @@ define amdgpu_ps float @extract_elt0_image_sample_cl_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d.o
@@ -4719,12 +4719,12 @@ define amdgpu_ps float @extract_elt0_image_sample_d_o_1d_v4f32_f32_f32(i32 %offs
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.d.cl.o
@@ -4735,12 +4735,12 @@ define amdgpu_ps float @extract_elt0_image_sample_d_cl_o_1d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.l.o
@@ -4751,12 +4751,12 @@ define amdgpu_ps float @extract_elt0_image_sample_l_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b.o
@@ -4767,12 +4767,12 @@ define amdgpu_ps float @extract_elt0_image_sample_b_o_1d_v4f32_f32_f32(i32 %offs
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.b.cl.o
@@ -4783,12 +4783,12 @@ define amdgpu_ps float @extract_elt0_image_sample_b_cl_o_1d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.lz.o
@@ -4799,12 +4799,12 @@ define amdgpu_ps float @extract_elt0_image_sample_lz_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd.o
@@ -4815,12 +4815,12 @@ define amdgpu_ps float @extract_elt0_image_sample_cd_o_1d_v4f32_f32_f32(i32 %off
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.cd.cl.o
@@ -4831,12 +4831,12 @@ define amdgpu_ps float @extract_elt0_image_sample_cd_cl_o_1d_v4f32_f32_f32(i32 %
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.o
@@ -4847,12 +4847,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_o_1d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cl.o
@@ -4863,12 +4863,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cl_o_1d_v4f32_f32(i32 %offse
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cl.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d.o
@@ -4879,12 +4879,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_o_1d_v4f32_f32_f32(i32 %of
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.d.cl.o
@@ -4895,12 +4895,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_o_1d_v4f32_f32_f32(i32
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.d.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.l.o
@@ -4911,12 +4911,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_l_o_1d_v4f32_f32(i32 %offset
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.l.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[LOD:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b.o
@@ -4927,12 +4927,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_o_1d_v4f32_f32_f32(i32 %of
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.b.cl.o
@@ -4943,12 +4943,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_o_1d_v4f32_f32_f32(i32
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.b.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[BIAS:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.lz.o
@@ -4959,12 +4959,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_lz_o_1d_v4f32_f32(i32 %offse
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.lz.o.1d.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd.o
@@ -4975,12 +4975,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_o_1d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample.c.cd.cl.o
@@ -4991,12 +4991,12 @@ define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_o_1d_v4f32_f32_f32(i32
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.sample.c.cd.cl.o.1d.f32.f32.f32.v8i32.v4i32(i32 1, i32 [[OFFSET:%.*]], float [[ZCOMPARE:%.*]], float [[DSDH:%.*]], float [[DSDV:%.*]], float [[S:%.*]], float [[CLAMP:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4
@@ -5010,12 +5010,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_2d_v4f32_f32(float %s, float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32 1, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32(i32 1, float %s, float %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f32.v8i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.cl
@@ -5027,12 +5027,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_cl_2d_v4f32_f32(float %s, flo
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32 2, float %s, float %t, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32.v8i32(i32 2, float %s, float %t, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.cl.2d.v4f32.f32.v8i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.l
@@ -5044,12 +5044,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_l_2d_v4f32_f32(float %s, floa
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32 4, float %s, float %t, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32(i32 4, float %s, float %t, float %lod, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b
@@ -5061,12 +5061,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_2darray_v4f32_f32_f32(float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32 8, float %bias, float %s, float %t, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32.v8i32(i32 8, float %bias, float %s, float %t, float %slice, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.2darray.v4f32.f32.f32.v8i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b.cl
@@ -5078,12 +5078,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_cl_cube_v4f32_f32_f32(float
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32 1, float %bias, float %s, float %t, float %face, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32.v8i32(i32 1, float %bias, float %s, float %t, float %face, float %clamp, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.cube.v4f32.f32.f32.v8i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.lz
@@ -5095,12 +5095,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_lz_2d_v4f32_f16(half %s, half
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32 1, half %s, half %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16.v8i32(i32 1, half %s, half %t, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16(i32, half, half, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.lz.2d.v4f32.f16.v8i32(i32, half, half, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.o
@@ -5112,12 +5112,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_o_2d_v4f32_f32(i32 %offset, f
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.o.2d.v4f32.f32.v8i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.cl.o
@@ -5129,12 +5129,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_cl_o_2d_v4f32_f32(i32 %offset
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.l.o
@@ -5146,12 +5146,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_l_o_2d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b.o
@@ -5163,12 +5163,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_o_2d_v4f32_f32_f32(i32 %off
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.b.cl.o
@@ -5180,12 +5180,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_b_cl_o_2d_v4f32_f32_f32(i32 %
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.lz.o
@@ -5197,12 +5197,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_lz_o_2d_v4f32_f32(i32 %offset
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.2d.v4f32.f32.v8i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.o
@@ -5214,12 +5214,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_o_2d_v4f32_f32(i32 %offset,
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.cl.o
@@ -5231,12 +5231,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_cl_o_2d_v4f32_f32(i32 %offs
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.l.o
@@ -5248,12 +5248,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_l_o_2d_v4f32_f32(i32 %offse
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, float %lod, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.b.o
@@ -5265,12 +5265,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_b_o_2d_v4f32_f32_f32(i32 %o
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.b.cl.o
@@ -5282,12 +5282,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_b_cl_o_2d_v4f32_f32_f32(i32
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32.v8i32(i32 1, i32 %offset, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.2d.v4f32.f32.f32.v8i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.gather4.c.lz.o
@@ -5299,12 +5299,12 @@ define amdgpu_ps float @extract_elt0_image_gather4_c_lz_o_2d_v4f32_f32(i32 %offs
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32(i32 1, i32 %offset, float %zcompare, float %s, float %t, <8 x i32> %gather4r, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.getlod
@@ -5315,12 +5315,12 @@ define amdgpu_ps float @extract_elt0_image_getlod_1d_v4f32_f32(float %s, <8 x i3
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getlod.1d.f32.f32.v8i32.v4i32(i32 1, float [[S:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32.v8i32(i32 15, float %s, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32.v8i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.load
@@ -5331,12 +5331,12 @@ define amdgpu_ps float @extract_elt0_image_load_2dmsaa_v4f32_i32(i32 %s, i32 %t,
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2dmsaa.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[T:%.*]], i32 [[SAMPLE:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %sample, <8 x i32> %sampler, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 %sample, <8 x i32> %sampler, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.2dmsaa.v4f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.load.mip
@@ -5347,12 +5347,12 @@ define amdgpu_ps float @extract_elt0_image_load_mip_1d_v4f32_i32(i32 %s, i32 %mi
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.mip.1d.f32.i32.v8i32(i32 1, i32 [[S:%.*]], i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32 15, i32 %s, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32, i32, i32, <8 x i32>, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.image.getresinfo
@@ -5363,12 +5363,12 @@ define amdgpu_ps float @extract_elt0_image_getresinfo_1d_v4f32_i32(i32 %mip, <8
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.getresinfo.1d.f32.i32.v8i32(i32 1, i32 [[MIP:%.*]], <8 x i32> [[SAMPLER:%.*]], i32 0, i32 0)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32 15, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
+ %data = call <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32.v8i32(i32 15, i32 %mip, <8 x i32> %sampler, i32 0, i32 0)
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32(i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.getresinfo.1d.v4f32.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #1
; --------------------------------------------------------------------
; TFE / LWE
@@ -5381,25 +5381,25 @@ define amdgpu_ps float @extract_elt0_tfe_image_load_1d_v4f32i32_i32(i32 %s, <8 x
; CHECK-NEXT: [[ELT0:%.*]] = extractelement <4 x float> [[RGBA]], i64 0
; CHECK-NEXT: ret float [[ELT0]]
;
- %data = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 1)
+ %data = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 1)
%rgba = extractvalue { <4 x float>, i32 } %data, 0
%elt0 = extractelement <4 x float> %rgba, i32 0
ret float %elt0
}
-declare {<4 x float>, i32} @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32(i32, i32, <8 x i32>, i32, i32) #1
+declare {<4 x float>, i32} @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32, i32, <8 x i32>, i32, i32) #1
define amdgpu_hs float @tfe_check_assert() #0 {
; CHECK-LABEL: @tfe_check_assert(
; CHECK-NEXT: [[DATA:%.*]] = call float @llvm.amdgcn.image.load.2d.f32.i32.v8i32(i32 1, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1)
; CHECK-NEXT: ret float [[DATA]]
;
- %data = call nsz <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 15, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1) #2
+ %data = call nsz <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32.v8i32(i32 15, i32 undef, i32 undef, <8 x i32> undef, i32 0, i32 1) #2
%elt0 = extractelement <4 x float> %data, i32 0
ret float %elt0
}
-declare <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32(i32 immarg, i32, i32, <8 x i32>, i32 immarg, i32 immarg) #1
+declare <4 x float> @llvm.amdgcn.image.load.2d.v4f32.i32.v8i32(i32 immarg, i32, i32, <8 x i32>, i32 immarg, i32 immarg) #1
attributes #0 = { nounwind }
attributes #1 = { nounwind readonly }
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll
index fc9dd99c4803a8..4df170dc8f85ca 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics-gfx8.ll
@@ -5,12 +5,12 @@
; llvm.amdgcn.image.sample a16 is disabled on pre-gfx9
; --------------------------------------------------------------------
-declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d(
@@ -20,7 +20,7 @@ define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -35,7 +35,7 @@ define amdgpu_kernel void @image_sample_a16_2d(ptr addrspace(1) %out, <8 x i32>
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -52,7 +52,7 @@ define amdgpu_kernel void @image_sample_a16_3d(ptr addrspace(1) %out, <8 x i32>
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%r32 = fpext half %r to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 15, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32(i32 15, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -70,7 +70,7 @@ define amdgpu_kernel void @image_sample_a16_cube(ptr addrspace(1) %out, <8 x i32
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%face32 = fpext half %face to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32(i32 15, float %s32, float %t32, float %face32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32.v8i32(i32 15, float %s32, float %t32, float %face32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -85,7 +85,7 @@ define amdgpu_kernel void @image_sample_a16_1darray(ptr addrspace(1) %out, <8 x
;
%s32 = fpext half %s to float
%slice32 = fpext half %slice to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 15, float %s32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32(i32 15, float %s32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -102,7 +102,7 @@ define amdgpu_kernel void @image_sample_a16_2darray(ptr addrspace(1) %out, <8 x
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%slice32 = fpext half %slice to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32 15, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32.v8i32(i32 15, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
index e886dbc509e1d5..e4f8bb133e6ce6 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll
@@ -3159,60 +3159,60 @@ define amdgpu_kernel void @permlane64_uniform(ptr addrspace(1) %out, i32 %src0)
; llvm.amdgcn.image.sample a16
; --------------------------------------------------------------------
-declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32(i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f32.f32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, half %s) {
; CHECK-LABEL: @image_sample_a16_1d(
@@ -3221,7 +3221,7 @@ define amdgpu_kernel void @image_sample_a16_1d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3234,7 +3234,7 @@ define amdgpu_kernel void @image_sample_a16_2d(ptr addrspace(1) %out, <8 x i32>
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3248,7 +3248,7 @@ define amdgpu_kernel void @image_sample_a16_3d(ptr addrspace(1) %out, <8 x i32>
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%r32 = fpext half %r to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 15, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3263,7 +3263,7 @@ define amdgpu_kernel void @image_sample_a16_cube(ptr addrspace(1) %out, <8 x i32
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%face32 = fpext half %face to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32(i32 15, float %s32, float %t32, float %face32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %face32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3276,7 +3276,7 @@ define amdgpu_kernel void @image_sample_a16_1darray(ptr addrspace(1) %out, <8 x
;
%s32 = fpext half %s to float
%slice32 = fpext half %slice to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 15, float %s32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3290,7 +3290,7 @@ define amdgpu_kernel void @image_sample_a16_2darray(ptr addrspace(1) %out, <8 x
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%slice32 = fpext half %slice to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32 15, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3302,7 +3302,7 @@ define amdgpu_kernel void @image_sample_a16_c_1d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32(i32 15, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3315,7 +3315,7 @@ define amdgpu_kernel void @image_sample_a16_c_2d(ptr addrspace(1) %out, <8 x i32
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32(i32 15, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.2d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3328,7 +3328,7 @@ define amdgpu_kernel void @image_sample_a16_cl_1d(ptr addrspace(1) %out, <8 x i3
;
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f32(i32 15, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cl.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3342,7 +3342,7 @@ define amdgpu_kernel void @image_sample_a16_cl_2d(ptr addrspace(1) %out, <8 x i3
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f32(i32 15, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cl.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3355,7 +3355,7 @@ define amdgpu_kernel void @image_sample_a16_c_cl_1d(ptr addrspace(1) %out, <8 x
;
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32(i32 15, float %zcompare, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cl.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3369,7 +3369,7 @@ define amdgpu_kernel void @image_sample_a16_c_cl_2d(ptr addrspace(1) %out, <8 x
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f32(i32 15, float %zcompare, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cl.2d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3382,7 +3382,7 @@ define amdgpu_kernel void @image_sample_a16_b16_1d(ptr addrspace(1) %out, <8 x i
;
%bias32 = fpext half %bias to float
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 15, float %bias32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3395,7 +3395,7 @@ define amdgpu_kernel void @image_sample_a16_b32_1d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32(i32 15, float %bias, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3409,7 +3409,7 @@ define amdgpu_kernel void @image_sample_a16_b16_2d(ptr addrspace(1) %out, <8 x i
%bias32 = fpext half %bias to float
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32(i32 15, float %bias32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3424,7 +3424,7 @@ define amdgpu_kernel void @image_sample_a16_b32_2d(ptr addrspace(1) %out, <8 x i
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32(i32 15, float %bias, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3437,7 +3437,7 @@ define amdgpu_kernel void @image_sample_a16_c_b16_1d(ptr addrspace(1) %out, <8 x
;
%bias32 = fpext half %bias to float
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float %bias32, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3450,7 +3450,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_1d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3464,7 +3464,7 @@ define amdgpu_kernel void @image_sample_a16_c_b16_2d(ptr addrspace(1) %out, <8 x
%bias32 = fpext half %bias to float
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32(i32 15, float %bias32, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3479,7 +3479,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_2d(ptr addrspace(1) %out, <8 x
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3493,7 +3493,7 @@ define amdgpu_kernel void @image_sample_a16_b16_cl_1d(ptr addrspace(1) %out, <8
%bias32 = fpext half %bias to float
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 15, float %bias32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3508,7 +3508,7 @@ define amdgpu_kernel void @image_sample_a16_b32_cl_1d(ptr addrspace(1) %out, <8
;
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32(i32 15, float %bias, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3523,7 +3523,7 @@ define amdgpu_kernel void @image_sample_a16_b16_cl_2d(ptr addrspace(1) %out, <8
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32(i32 15, float %bias32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3540,7 +3540,7 @@ define amdgpu_kernel void @image_sample_a16_b32_cl_2d(ptr addrspace(1) %out, <8
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32(i32 15, float %bias, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3554,7 +3554,7 @@ define amdgpu_kernel void @image_sample_a16_c_b16_cl_1d(ptr addrspace(1) %out, <
%bias32 = fpext half %bias to float
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float %bias32, float %zcompare, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %zcompare, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3569,7 +3569,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_cl_1d(ptr addrspace(1) %out, <
;
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3584,7 +3584,7 @@ define amdgpu_kernel void @image_sample_a16_c_b16_cl_2d(ptr addrspace(1) %out, <
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32(i32 15, float %bias32, float %zcompare, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias32, float %zcompare, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3601,7 +3601,7 @@ define amdgpu_kernel void @image_sample_a16_c_b32_cl_2d(ptr addrspace(1) %out, <
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32(i32 15, float %bias, float %zcompare, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %bias, float %zcompare, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3615,7 +3615,7 @@ define amdgpu_kernel void @image_sample_a16_d_1d(ptr addrspace(1) %out, <8 x i32
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3632,7 +3632,7 @@ define amdgpu_kernel void @image_sample_a16_d_2d(ptr addrspace(1) %out, <8 x i32
%dtdv32 = fpext half %dtdv to float
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3652,7 +3652,7 @@ define amdgpu_kernel void @image_sample_a16_d_3d(ptr addrspace(1) %out, <8 x i32
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%r32 = fpext half %r to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %drdh32, float %dsdv32, float %dtdv32, float %drdv32, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %drdh32, float %dsdv32, float %dtdv32, float %drdv32, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3666,7 +3666,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_1d(ptr addrspace(1) %out, <8 x i
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3683,7 +3683,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_2d(ptr addrspace(1) %out, <8 x i
%dtdv32 = fpext half %dtdv to float
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3698,7 +3698,7 @@ define amdgpu_kernel void @image_sample_a16_d_cl_1d(ptr addrspace(1) %out, <8 x
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3716,7 +3716,7 @@ define amdgpu_kernel void @image_sample_a16_d_cl_2d(ptr addrspace(1) %out, <8 x
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3731,7 +3731,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_cl_1d(ptr addrspace(1) %out, <8
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3749,7 +3749,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_cl_2d(ptr addrspace(1) %out, <8
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3763,7 +3763,7 @@ define amdgpu_kernel void @image_sample_a16_cd_1d(ptr addrspace(1) %out, <8 x i3
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3780,7 +3780,7 @@ define amdgpu_kernel void @image_sample_a16_cd_2d(ptr addrspace(1) %out, <8 x i3
%dtdv32 = fpext half %dtdv to float
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3794,7 +3794,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_1d(ptr addrspace(1) %out, <8 x
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3811,7 +3811,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_2d(ptr addrspace(1) %out, <8 x
%dtdv32 = fpext half %dtdv to float
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3826,7 +3826,7 @@ define amdgpu_kernel void @image_sample_a16_cd_cl_1d(ptr addrspace(1) %out, <8 x
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3844,7 +3844,7 @@ define amdgpu_kernel void @image_sample_a16_cd_cl_2d(ptr addrspace(1) %out, <8 x
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3859,7 +3859,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_cl_1d(ptr addrspace(1) %out, <8
%dsdv32 = fpext half %dsdv to float
%s32 = fpext half %s to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3877,7 +3877,7 @@ define amdgpu_kernel void @image_sample_a16_c_cd_cl_2d(ptr addrspace(1) %out, <8
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%clamp32 = fpext half %clamp to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %clamp32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3890,7 +3890,7 @@ define amdgpu_kernel void @image_sample_a16_l_1d(ptr addrspace(1) %out, <8 x i32
;
%s32 = fpext half %s to float
%lod32 = fpext half %lod to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32(i32 15, float %s32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3904,7 +3904,7 @@ define amdgpu_kernel void @image_sample_a16_l_2d(ptr addrspace(1) %out, <8 x i32
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%lod32 = fpext half %lod to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32(i32 15, float %s32, float %t32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3917,7 +3917,7 @@ define amdgpu_kernel void @image_sample_a16_c_l_1d(ptr addrspace(1) %out, <8 x i
;
%s32 = fpext half %s to float
%lod32 = fpext half %lod to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float %zcompare, float %s32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3931,7 +3931,7 @@ define amdgpu_kernel void @image_sample_a16_c_l_2d(ptr addrspace(1) %out, <8 x i
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%lod32 = fpext half %lod to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32(i32 15, float %zcompare, float %s32, float %t32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, float %t32, float %lod32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3943,7 +3943,7 @@ define amdgpu_kernel void @image_sample_a16_lz_1d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3956,7 +3956,7 @@ define amdgpu_kernel void @image_sample_a16_lz_2d(ptr addrspace(1) %out, <8 x i3
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3968,7 +3968,7 @@ define amdgpu_kernel void @image_sample_a16_c_lz_1d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32(i32 15, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.lz.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3981,7 +3981,7 @@ define amdgpu_kernel void @image_sample_a16_c_lz_2d(ptr addrspace(1) %out, <8 x
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32(i32 15, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -3999,7 +3999,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_V1(ptr addrspace(1) %o
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%slice32 = fpext half %slice to float
- %res = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f32.f32(i32 4, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f32.f32.v8i32.v4i32(i32 4, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store float %res, ptr addrspace(1) %out
ret void
}
@@ -4017,7 +4017,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_V2(ptr addrspace(1) %o
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%slice32 = fpext half %slice to float
- %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32.v8i32.v4i32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <2 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4034,7 +4034,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_const(ptr addrspace(1)
%dtdv32 = fpext half %dtdv to float
%s32 = fpext half %s to float
%slice32 = fpext half %slice to float
- %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float 0.25, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32.v8i32.v4i32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float 0.25, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <2 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4053,7 +4053,7 @@ define amdgpu_kernel void @image_sample_a16_c_d_o_2darray_const_noopt(ptr addrsp
%dtdv32 = fpext half %dtdv to float
%s32 = fpext half %s to float
%slice32 = fpext half %slice to float
- %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float 1.0e+10, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32.v8i32.v4i32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s32, float 1.0e+10, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <2 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4065,7 +4065,7 @@ define amdgpu_kernel void @image_load_a16_mip_1d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
%s32 = zext i16 %s to i32
- %res = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s32, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32 15, i32 %s32, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4078,7 +4078,7 @@ define amdgpu_kernel void @image_load_a16_mip_1d_noopt(ptr addrspace(1) %out, <8
; CHECK-NEXT: ret void
;
%s32 = sext i16 %s to i32
- %res = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s32, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32 15, i32 %s32, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4091,7 +4091,7 @@ define amdgpu_kernel void @image_load_a16_mip_2d(ptr addrspace(1) %out, <8 x i32
;
%s32 = zext i16 %s to i32
%t32 = zext i16 %t to i32
- %res = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32(i32 15, i32 %s32, i32 %t32, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32.v8i32(i32 15, i32 %s32, i32 %t32, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4103,7 +4103,7 @@ define amdgpu_kernel void @image_load_a16_mip_2d_const(ptr addrspace(1) %out, <8
; CHECK-NEXT: ret void
;
%s32 = zext i16 %s to i32
- %res = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32(i32 15, i32 %s32, i32 65535, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32.v8i32(i32 15, i32 %s32, i32 65535, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4116,7 +4116,7 @@ define amdgpu_kernel void @image_load_a16_mip_2d_const_noopt(ptr addrspace(1) %o
; CHECK-NEXT: ret void
;
%s32 = zext i16 %s to i32
- %res = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32(i32 15, i32 %s32, i32 65536, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32.v8i32(i32 15, i32 %s32, i32 65536, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4133,7 +4133,7 @@ define amdgpu_kernel void @image_sample_g16_d_1d(ptr addrspace(1) %out, <8 x i32
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4148,7 +4148,7 @@ define amdgpu_kernel void @image_sample_g16_d_2d(ptr addrspace(1) %out, <8 x i32
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4165,7 +4165,7 @@ define amdgpu_kernel void @image_sample_g16_d_3d(ptr addrspace(1) %out, <8 x i32
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
%drdv32 = fpext half %drdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %drdh32, float %dsdv32, float %dtdv32, float %drdv32, float %s, float %t, float %r, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.3d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %drdh32, float %dsdv32, float %dtdv32, float %drdv32, float %s, float %t, float %r, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4178,7 +4178,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_1d(ptr addrspace(1) %out, <8 x i
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4193,7 +4193,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_2d(ptr addrspace(1) %out, <8 x i
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4206,7 +4206,7 @@ define amdgpu_kernel void @image_sample_g16_d_cl_1d(ptr addrspace(1) %out, <8 x
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4221,7 +4221,7 @@ define amdgpu_kernel void @image_sample_g16_d_cl_2d(ptr addrspace(1) %out, <8 x
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4234,7 +4234,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_cl_1d(ptr addrspace(1) %out, <8
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4249,7 +4249,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_cl_2d(ptr addrspace(1) %out, <8
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4262,7 +4262,7 @@ define amdgpu_kernel void @image_sample_g16_cd_1d(ptr addrspace(1) %out, <8 x i3
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4277,7 +4277,7 @@ define amdgpu_kernel void @image_sample_g16_cd_2d(ptr addrspace(1) %out, <8 x i3
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4290,7 +4290,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_1d(ptr addrspace(1) %out, <8 x
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4305,7 +4305,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_2d(ptr addrspace(1) %out, <8 x
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4318,7 +4318,7 @@ define amdgpu_kernel void @image_sample_g16_cd_cl_1d(ptr addrspace(1) %out, <8 x
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32(i32 15, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4333,7 +4333,7 @@ define amdgpu_kernel void @image_sample_g16_cd_cl_2d(ptr addrspace(1) %out, <8 x
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4346,7 +4346,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_cl_1d(ptr addrspace(1) %out, <8
;
%dsdh32 = fpext half %dsdh to float
%dsdv32 = fpext half %dsdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dsdv32, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4361,7 +4361,7 @@ define amdgpu_kernel void @image_sample_g16_c_cd_cl_2d(ptr addrspace(1) %out, <8
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4376,7 +4376,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_o_2darray_V1(ptr addrspace(1) %o
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f32.f32(i32 4, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %slice, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call float @llvm.amdgcn.image.sample.c.d.o.2darray.f32.f32.f32.v8i32.v4i32(i32 4, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %slice, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store float %res, ptr addrspace(1) %out
ret void
}
@@ -4391,7 +4391,7 @@ define amdgpu_kernel void @image_sample_g16_c_d_o_2darray_V2(ptr addrspace(1) %o
%dtdh32 = fpext half %dtdh to float
%dsdv32 = fpext half %dsdv to float
%dtdv32 = fpext half %dtdv to float
- %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %slice, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call <2 x float> @llvm.amdgcn.image.sample.c.d.o.2darray.v2f32.f32.f32.v8i32.v4i32(i32 6, i32 %offset, float %zcompare, float %dsdh32, float %dtdh32, float %dsdv32, float %dtdv32, float %s, float %t, float %slice, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <2 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4407,7 +4407,7 @@ define amdgpu_kernel void @image_sample_a16_1d_nnan(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call nnan <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4419,7 +4419,7 @@ define amdgpu_kernel void @image_sample_a16_1d_nnan_ninf_nsz(ptr addrspace(1) %o
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call nnan ninf nsz <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan ninf nsz <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4431,7 +4431,7 @@ define amdgpu_kernel void @image_sample_a16_1d_fast(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
%s32 = fpext half %s to float
- %res = call fast <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call fast <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4444,7 +4444,7 @@ define amdgpu_kernel void @image_sample_a16_2d_nnan(ptr addrspace(1) %out, <8 x
;
%s32 = fpext half %s to float
%t32 = fpext half %t to float
- %res = call nnan <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4458,7 +4458,7 @@ define amdgpu_kernel void @image_sample_a16_3d_nnan(ptr addrspace(1) %out, <8 x
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%r32 = fpext half %r to float
- %res = call nnan <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 15, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %r32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4473,7 +4473,7 @@ define amdgpu_kernel void @image_sample_a16_cube_nnan(ptr addrspace(1) %out, <8
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%face32 = fpext half %face to float
- %res = call nnan <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32(i32 15, float %s32, float %t32, float %face32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan <4 x float> @llvm.amdgcn.image.sample.cube.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %face32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4486,7 +4486,7 @@ define amdgpu_kernel void @image_sample_a16_1darray_nnan(ptr addrspace(1) %out,
;
%s32 = fpext half %s to float
%slice32 = fpext half %slice to float
- %res = call nnan <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32(i32 15, float %s32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan <4 x float> @llvm.amdgcn.image.sample.1darray.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4500,7 +4500,7 @@ define amdgpu_kernel void @image_sample_a16_2darray_nnan(ptr addrspace(1) %out,
%s32 = fpext half %s to float
%t32 = fpext half %t to float
%slice32 = fpext half %slice to float
- %res = call nnan <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32(i32 15, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %res = call nnan <4 x float> @llvm.amdgcn.image.sample.2darray.v4f32.f32.v8i32.v4i32(i32 15, float %s32, float %t32, float %slice32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %res, ptr addrspace(1) %out
ret void
}
@@ -4509,16 +4509,16 @@ define amdgpu_kernel void @image_sample_a16_2darray_nnan(ptr addrspace(1) %out,
; llvm.amdgcn.image.sample l to lz
; --------------------------------------------------------------------
-declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.l.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.l.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.c.l.2d.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2darray.v4f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.l.2d.v4f32.f32.v8i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2darray.v4f32.f32.v8i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
define amdgpu_kernel void @sample_l_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %lod) {
; CHECK-LABEL: @sample_l_1d(
@@ -4528,7 +4528,7 @@ define amdgpu_kernel void @sample_l_1d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32(i32 15, float %s, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float %s, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4541,7 +4541,7 @@ define amdgpu_kernel void @sample_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32(i32 15, float %s, float %t, float -0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.l.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s, float %t, float -0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4554,7 +4554,7 @@ define amdgpu_kernel void @sample_c_l_1d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32(i32 15, float %zcompare, float %s, float -2.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.1d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, float -2.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4567,7 +4567,7 @@ define amdgpu_kernel void @sample_c_l_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32(i32 15, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.2d.v4f32.f32.v8i32.v4i32(i32 15, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4580,7 +4580,7 @@ define amdgpu_kernel void @sample_l_o_1d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %s, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4593,7 +4593,7 @@ define amdgpu_kernel void @sample_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.2d.v4f32.f32(i32 15, i32 %offset, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4606,7 +4606,7 @@ define amdgpu_kernel void @sample_c_l_o_1d(ptr addrspace(1) %out, <8 x i32> inre
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4619,7 +4619,7 @@ define amdgpu_kernel void @sample_c_l_o_2d(ptr addrspace(1) %out, <8 x i32> inre
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.2d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4632,7 +4632,7 @@ define amdgpu_kernel void @gather4_l_2d(ptr addrspace(1) %out, <8 x i32> inreg %
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32(i32 15, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.l.2d.v4f32.f32.v8i32(i32 15, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4645,7 +4645,7 @@ define amdgpu_kernel void @gather4_c_l_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.c.l.2d.v4f32.f32(i32 15, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.c.l.2d.v4f32.f32.v8i32(i32 15, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4658,7 +4658,7 @@ define amdgpu_kernel void @gather4_l_o_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32(i32 15, i32 %offset, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.l.o.2d.v4f32.f32.v8i32(i32 15, i32 %offset, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4671,7 +4671,7 @@ define amdgpu_kernel void @gather4_c_l_o_2d(ptr addrspace(1) %out, <8 x i32> inr
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2d.v4f32.f32.v8i32(i32 15, i32 %offset, float %zcompare, float %s, float %t, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4684,7 +4684,7 @@ define amdgpu_kernel void @gather4_c_l_o_2darray(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2darray.v4f32.f32(i32 15, i32 %offset, float %zcompare, float %s, float %t, float %slice, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.2darray.v4f32.f32.v8i32(i32 15, i32 %offset, float %zcompare, float %s, float %t, float %slice, float 0.0, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4701,7 +4701,7 @@ define amdgpu_kernel void @load_mip_1d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32 15, i32 %s, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32 15, i32 %s, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4714,7 +4714,7 @@ define amdgpu_kernel void @load_mip_2d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32(i32 15, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4727,7 +4727,7 @@ define amdgpu_kernel void @load_mip_3d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.load.mip.3d.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %u, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.load.mip.3d.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 %u, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4740,7 +4740,7 @@ define amdgpu_kernel void @load_mip_1darray(ptr addrspace(1) %out, <8 x i32> inr
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.load.mip.1darray.v4f32.i32(i32 15, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.load.mip.1darray.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4753,7 +4753,7 @@ define amdgpu_kernel void @load_mip_2darray(ptr addrspace(1) %out, <8 x i32> inr
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.load.mip.2darray.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %u, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.load.mip.2darray.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 %u, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4766,7 +4766,7 @@ define amdgpu_kernel void @load_mip_cube(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.load.mip.cube.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %u, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.load.mip.cube.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 %u, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4838,39 +4838,39 @@ main_body:
ret void
}
-declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32(i32, i32, i32, <8 x i32>, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.load.mip.3d.v4f32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.load.mip.1darray.v4f32.i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.load.mip.2darray.v4f32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.load.mip.cube.v4f32.i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.1d.v4f32.i32.v8i32(i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.2d.v4f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.3d.v4f32.i32.v8i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.1darray.v4f32.i32.v8i32(i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.2darray.v4f32.i32.v8i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.load.mip.cube.v4f32.i32.v8i32(i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #1
-declare void @llvm.amdgcn.image.store.mip.1d.v4f32.i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.2d.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.3d.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.cube.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.1darray.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.2darray.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.1d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.2d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.3d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.cube.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.1darray.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.2darray.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
; --------------------------------------------------------------------
; llvm.amdgcn.image.sample bias zero
; --------------------------------------------------------------------
-declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f16(i32, i32, half, float, half, half, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.gather4.b.2d.v4f32.f32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.c.b.2d.v4f32.f32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.v8i32.v4i32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.v8i32.v4i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.v8i32.v4i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f16.v8i32.v4i32(i32, i32, half, float, half, half, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.gather4.b.2d.v4f32.f32.v8i32(i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.2d.v4f32.f32.v8i32(i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.v8i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
define amdgpu_kernel void @sample_b_1d(ptr addrspace(1) %out, <8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s) {
; CHECK-LABEL: @sample_b_1d(
@@ -4880,7 +4880,7 @@ define amdgpu_kernel void @sample_b_1d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32(i32 15, float 0.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.v8i32.v4i32(i32 15, float 0.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4893,7 +4893,7 @@ define amdgpu_kernel void @sample_b_2d(ptr addrspace(1) %out, <8 x i32> inreg %r
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32(i32 15, float -0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.2d.v4f32.f32.v8i32.v4i32(i32 15, float -0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4906,7 +4906,7 @@ define amdgpu_kernel void @sample_c_b_1d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32(i32 15, float -0.0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.1d.v4f32.f32.v8i32.v4i32(i32 15, float -0.0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4919,7 +4919,7 @@ define amdgpu_kernel void @sample_c_b_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32(i32 15, float 0.0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.2d.v4f32.f32.v8i32.v4i32(i32 15, float 0.0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4932,7 +4932,7 @@ define amdgpu_kernel void @sample_b_o_1d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32(i32 15, i32 %offset, float 0.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float 0.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4945,7 +4945,7 @@ define amdgpu_kernel void @sample_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32(i32 15, i32 %offset, float 0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float 0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4958,7 +4958,7 @@ define amdgpu_kernel void @sample_c_b_o_1d(ptr addrspace(1) %out, <8 x i32> inre
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32(i32 15, i32 %offset, float 0.0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float 0.0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4971,7 +4971,7 @@ define amdgpu_kernel void @sample_c_b_o_2d(ptr addrspace(1) %out, <8 x i32> inre
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32(i32 15, i32 %offset, float 0.0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 %offset, float 0.0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4984,7 +4984,7 @@ define amdgpu_kernel void @gather4_b_2d(ptr addrspace(1) %out, <8 x i32> inreg %
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.b.2d.v4f32.f32(i32 15, float 0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.b.2d.v4f32.f32.v8i32(i32 15, float 0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -4997,7 +4997,7 @@ define amdgpu_kernel void @gather4_c_b_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.c.b.2d.v4f32.f32(i32 15, float 0.0, float %zcompare,float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.c.b.2d.v4f32.f32.v8i32(i32 15, float 0.0, float %zcompare,float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5010,7 +5010,7 @@ define amdgpu_kernel void @gather4_b_o_2d(ptr addrspace(1) %out, <8 x i32> inreg
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32(i32 15, i32 %offset, float 0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.b.o.2d.v4f32.f32.v8i32(i32 15, i32 %offset, float 0.0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5023,7 +5023,7 @@ define amdgpu_kernel void @gather4_c_b_o_2d(ptr addrspace(1) %out, <8 x i32> inr
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32(i32 15, i32 %offset, float 0.0, float %zcompare,float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.2d.v4f32.f32.v8i32(i32 15, i32 %offset, float 0.0, float %zcompare,float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5036,7 +5036,7 @@ define amdgpu_kernel void @sample_c_b_o_a16_2d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f16(i32 15, i32 %offset, half 0.0, float %zcompare, half %s, half %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f16.v8i32.v4i32(i32 15, i32 %offset, half 0.0, float %zcompare, half %s, half %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5050,7 +5050,7 @@ define amdgpu_kernel void @sample_b_1d_pos(ptr addrspace(1) %out, <8 x i32> inre
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32(i32 15, float 1.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.v8i32.v4i32(i32 15, float 1.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5064,7 +5064,7 @@ define amdgpu_kernel void @sample_b_1d_neg(ptr addrspace(1) %out, <8 x i32> inre
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32(i32 15, float -1.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.v8i32.v4i32(i32 15, float -1.0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5079,7 +5079,7 @@ define amdgpu_kernel void @sample_b_1d_a16(ptr addrspace(1) %out, <8 x i32> inre
;
main_body:
%s32 = fpext half %s to float
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32(i32 15, float -0.0, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.1d.v4f32.f32.v8i32.v4i32(i32 15, float -0.0, float %s32, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5096,7 +5096,7 @@ define amdgpu_kernel void @offset_sample_o_1d(ptr addrspace(1) %out, <8 x i32> i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32 15, i32 0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5109,7 +5109,7 @@ define amdgpu_kernel void @offset_sample_o_2d(ptr addrspace(1) %out, <8 x i32> i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.o.2d.v4f32.f32(i32 15, i32 0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5122,7 +5122,7 @@ define amdgpu_kernel void @offset_sample_c_o_1d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5135,7 +5135,7 @@ define amdgpu_kernel void @offset_sample_c_o_2d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5148,7 +5148,7 @@ define amdgpu_kernel void @offset_sample_cl_o_1d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32 15, i32 0, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5161,7 +5161,7 @@ define amdgpu_kernel void @offset_sample_cl_o_2d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.cl.o.2d.v4f32.f32(i32 15, i32 0, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.cl.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5174,7 +5174,7 @@ define amdgpu_kernel void @offset_sample_c_cl_o_1d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5187,7 +5187,7 @@ define amdgpu_kernel void @offset_sample_c_cl_o_2d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.2d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5200,7 +5200,7 @@ define amdgpu_kernel void @offset_sample_b_o_1d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5213,7 +5213,7 @@ define amdgpu_kernel void @offset_sample_b_o_2d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5226,7 +5226,7 @@ define amdgpu_kernel void @offset_sample_c_b_o_1d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5239,7 +5239,7 @@ define amdgpu_kernel void @offset_sample_c_b_o_2d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5252,7 +5252,7 @@ define amdgpu_kernel void @offset_sample_b_cl_o_1d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5265,7 +5265,7 @@ define amdgpu_kernel void @offset_sample_b_cl_o_2d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.2d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5278,7 +5278,7 @@ define amdgpu_kernel void @offset_sample_c_b_cl_o_1d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %zcompare, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5291,7 +5291,7 @@ define amdgpu_kernel void @offset_sample_c_b_cl_o_2d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.2d.v4f32.f32.f32(i32 15, i32 0, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %bias, float %zcompare, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5304,7 +5304,7 @@ define amdgpu_kernel void @offset_sample_d_o_1d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5317,7 +5317,7 @@ define amdgpu_kernel void @offset_sample_d_o_2d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.d.o.2d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.d.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5330,7 +5330,7 @@ define amdgpu_kernel void @offset_sample_c_d_o_1d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5343,7 +5343,7 @@ define amdgpu_kernel void @offset_sample_c_d_o_2d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.2d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5356,7 +5356,7 @@ define amdgpu_kernel void @offset_sample_d_cl_o_1d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5369,7 +5369,7 @@ define amdgpu_kernel void @offset_sample_d_cl_o_2d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.2d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5382,7 +5382,7 @@ define amdgpu_kernel void @offset_sample_c_d_cl_o_1d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5395,7 +5395,7 @@ define amdgpu_kernel void @offset_sample_c_d_cl_o_2d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.2d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5408,7 +5408,7 @@ define amdgpu_kernel void @offset_sample_cd_o_1d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5421,7 +5421,7 @@ define amdgpu_kernel void @offset_sample_cd_o_2d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.cd.o.2d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.cd.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5434,7 +5434,7 @@ define amdgpu_kernel void @offset_sample_c_cd_o_1d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5447,7 +5447,7 @@ define amdgpu_kernel void @offset_sample_c_cd_o_2d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.2d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5460,7 +5460,7 @@ define amdgpu_kernel void @offset_sample_cd_cl_o_1d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5473,7 +5473,7 @@ define amdgpu_kernel void @offset_sample_cd_cl_o_2d(ptr addrspace(1) %out, <8 x
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.2d.v4f32.f32.f32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5486,7 +5486,7 @@ define amdgpu_kernel void @offset_sample_c_cd_cl_o_1d(ptr addrspace(1) %out, <8
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dsdv, float %s, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5499,7 +5499,7 @@ define amdgpu_kernel void @offset_sample_c_cd_cl_o_2d(ptr addrspace(1) %out, <8
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.2d.v4f32.f32.f32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %dsdh, float %dtdh, float %dsdv, float %dtdv, float %s, float %t, float %clamp, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5512,7 +5512,7 @@ define amdgpu_kernel void @offset_sample_l_o_1d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32(i32 15, i32 0, float %s, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5525,7 +5525,7 @@ define amdgpu_kernel void @offset_sample_l_o_2d(ptr addrspace(1) %out, <8 x i32>
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.2d.v4f32.f32(i32 15, i32 0, float %s, float %t, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.l.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, float %t, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5538,7 +5538,7 @@ define amdgpu_kernel void @offset_sample_c_l_o_1d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5551,7 +5551,7 @@ define amdgpu_kernel void @offset_sample_c_l_o_2d(ptr addrspace(1) %out, <8 x i3
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.2d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, float %t, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, float %t, float %lod, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5564,7 +5564,7 @@ define amdgpu_kernel void @offset_sample_lz_o_1d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32 15, i32 0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5577,7 +5577,7 @@ define amdgpu_kernel void @offset_sample_lz_o_2d(ptr addrspace(1) %out, <8 x i32
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.lz.o.2d.v4f32.f32(i32 15, i32 0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5590,7 +5590,7 @@ define amdgpu_kernel void @offset_sample_c_lz_o_1d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
@@ -5603,51 +5603,51 @@ define amdgpu_kernel void @offset_sample_c_lz_o_2d(ptr addrspace(1) %out, <8 x i
; CHECK-NEXT: ret void
;
main_body:
- %v = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.2d.v4f32.f32(i32 15, i32 0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
+ %v = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.2d.v4f32.f32.v8i32.v4i32(i32 15, i32 0, float %zcompare, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
store <4 x float> %v, ptr addrspace(1) %out
ret void
}
-declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.o.2d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cl.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.2d.v4f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cd.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.2d.v4f32.f32.f32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-
-declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.lz.o.2d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
-declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.2d.v4f32.f32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cl.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.1d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.2d.v4f32.f32.f32.v8i32.v4i32(i32, i32, float, float, float, float, float, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+
+declare <4 x float> @llvm.amdgcn.image.sample.lz.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.lz.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.1d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.2d.v4f32.f32.v8i32.v4i32(i32, i32, float, float, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
; --------------------------------------------------------------------
; llvm.amdgcn.is.shared
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll
index 92d11a2bbd7aca..e7a36e805a012f 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-simplify-image-buffer-stores.ll
@@ -18,7 +18,7 @@ define amdgpu_ps void @image_store_1d_store_all_zeros(<8 x i32> inreg %rsrc, i32
; GFXUNKNOWN-NEXT: call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> zeroinitializer, i32 15, i32 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFXUNKNOWN-NEXT: ret void
;
- call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> zeroinitializer, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> zeroinitializer, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
ret void
}
@@ -41,7 +41,7 @@ define amdgpu_ps void @image_store_1d_store_insert_zeros_at_end(<8 x i32> inreg
%newvdata2 = insertelement <4 x float> %newvdata1, float 0.0, i32 1
%newvdata3 = insertelement <4 x float> %newvdata2, float 0.0, i32 2
%newvdata4 = insertelement <4 x float> %newvdata3, float 0.0, i32 3
- call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> %newvdata4, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> %newvdata4, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
ret void
}
@@ -68,7 +68,7 @@ define amdgpu_ps void @image_store_mip_1d_store_insert_zeros_at_end(<8 x i32> in
%newvdata2 = insertelement <4 x float> %newvdata1, float %vdata1, i32 1
%newvdata3 = insertelement <4 x float> %newvdata2, float %vdata2, i32 2
%newvdata4 = insertelement <4 x float> %newvdata3, float 0.0, i32 3
- call void @llvm.amdgcn.image.store.mip.1d.v4f32.i32(<4 x float> %newvdata4, i32 7, i32 %s, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
+ call void @llvm.amdgcn.image.store.mip.1d.v4f32.i32.v8i32(<4 x float> %newvdata4, i32 7, i32 %s, i32 0, <8 x i32> %rsrc, i32 0, i32 0)
ret void
}
@@ -162,7 +162,7 @@ define amdgpu_ps void @image_store_1d_store_shufflevector_same(<8 x i32> inreg %
; GFXUNKNOWN-NEXT: ret void
;
%data = shufflevector <4 x float> %vdata1, <4 x float> poison, <4 x i32> <i32 0, i32 0, i32 0, i32 0>
- call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> %data, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> %data, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
ret void
}
@@ -180,7 +180,7 @@ define amdgpu_ps void @image_store_1d_store_shufflevector(<8 x i32> inreg %rsrc,
; GFXUNKNOWN-NEXT: ret void
;
%data = shufflevector <4 x float> <float 2.0, float 1.0, float 2.0, float 5.0>, <4 x float> poison, <4 x i32> <i32 0, i32 0, i32 3, i32 2>
- call void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float> %data, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ call void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float> %data, i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
ret void
}
@@ -280,20 +280,20 @@ declare void @llvm.amdgcn.raw.buffer.store.format.v4f32(<4 x float>, <4 x i32>,
declare void @llvm.amdgcn.struct.buffer.store.format.v4f32(<4 x float>, <4 x i32>, i32, i32, i32, i32) #2
declare void @llvm.amdgcn.struct.tbuffer.store.v4f32(<4 x float>, <4 x i32>, i32, i32, i32, i32, i32) #0
declare void @llvm.amdgcn.raw.tbuffer.store.v4f32(<4 x float>, <4 x i32>, i32, i32, i32, i32) #0
-declare void @llvm.amdgcn.image.store.1d.v4f32.i32(<4 x float>, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.2d.v4f32.i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.3d.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.cube.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.1darray.v4f32.i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.2darray.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.2dmsaa.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.2darraymsaa.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.1d.v4f32.i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.2d.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.3d.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.cube.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.1darray.v4f32.i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
-declare void @llvm.amdgcn.image.store.mip.2darray.v4f32.i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.1d.v4f32.i32.v8i32(<4 x float>, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.2d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.3d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.cube.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.1darray.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.2darray.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.2dmsaa.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.2darraymsaa.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.1d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.2d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.3d.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.cube.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.1darray.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
+declare void @llvm.amdgcn.image.store.mip.2darray.v4f32.i32.v8i32(<4 x float>, i32, i32, i32, i32, i32, <8 x i32>, i32, i32) #0
attributes #0 = { nounwind }
attributes #1 = { nounwind writeonly }
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll b/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll
index 6d40062a3fda7e..30431ad724843b 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/image-d16.ll
@@ -17,7 +17,7 @@ define amdgpu_ps half @image_sample_2d_fptrunc_to_d16(<8 x i32> inreg %rsrc, <4
; GFX81PLUS-NEXT: ret half [[TEX]]
;
main_body:
- %tex = call float @llvm.amdgcn.image.sample.lz.2d.f32.f32(i32 1, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
+ %tex = call float @llvm.amdgcn.image.sample.lz.2d.f32.f32.v8i32.v4i32(i32 1, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
%tex_half = fptrunc float %tex to half
ret half %tex_half
}
@@ -41,7 +41,7 @@ define amdgpu_ps half @image_sample_2d_v2f32(<8 x i32> inreg %rsrc, <4 x i32> in
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_0]]
;
main_body:
- %tex = call <2 x float> @llvm.amdgcn.image.sample.lz.2d.v2f32.f32(i32 3, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
+ %tex = call <2 x float> @llvm.amdgcn.image.sample.lz.2d.v2f32.f32.v8i32.v4i32(i32 3, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
%tex_2_half = fptrunc <2 x float> %tex to <2 x half>
%tex_half_0 = extractelement <2 x half> %tex_2_half, i64 0
%tex_half_1 = extractelement <2 x half> %tex_2_half, i64 1
@@ -72,7 +72,7 @@ define amdgpu_ps half @image_sample_2d_v3f32(<8 x i32> inreg %rsrc, <4 x i32> in
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_1]]
;
main_body:
- %tex = call <3 x float> @llvm.amdgcn.image.sample.lz.2d.v3f32.f32(i32 7, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
+ %tex = call <3 x float> @llvm.amdgcn.image.sample.lz.2d.v3f32.f32.v8i32.v4i32(i32 7, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
%tex_3_half = fptrunc <3 x float> %tex to <3 x half>
%tex_half_0 = extractelement <3 x half> %tex_3_half, i64 0
%tex_half_1 = extractelement <3 x half> %tex_3_half, i64 1
@@ -109,7 +109,7 @@ define amdgpu_ps half @image_sample_2d_v4f32(<8 x i32> inreg %rsrc, <4 x i32> in
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_2]]
;
main_body:
- %tex = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32(i32 15, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
+ %tex = call <4 x float> @llvm.amdgcn.image.sample.lz.2d.v4f32.f32.v8i32.v4i32(i32 15, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
%tex_4_half = fptrunc <4 x float> %tex to <4 x half>
%tex_half_0 = extractelement <4 x half> %tex_4_half, i64 0
%tex_half_1 = extractelement <4 x half> %tex_4_half, i64 1
@@ -148,7 +148,7 @@ define amdgpu_ps half @image_gather4_2d_v4f32(<8 x i32> inreg %rsrc, <4 x i32> i
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_2]]
;
main_body:
- %tex = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f16(i32 1, half %s, half %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
+ %tex = call <4 x float> @llvm.amdgcn.image.gather4.2d.v4f32.f16.v8i32.v4i32(i32 1, half %s, half %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 0, i32 0)
%tex_4_half = fptrunc <4 x float> %tex to <4 x half>
%tex_half_0 = extractelement <4 x half> %tex_4_half, i64 0
%tex_half_1 = extractelement <4 x half> %tex_4_half, i64 1
@@ -170,7 +170,7 @@ define amdgpu_ps half @load_1d(i16 %s, <8 x i32> inreg %rsrc) {
; GFX81PLUS-NEXT: [[S_FLOAT:%.*]] = call half @llvm.amdgcn.image.load.1d.f16.i16.v8i32(i32 1, i16 [[S:%.*]], <8 x i32> [[RSRC:%.*]], i32 0, i32 0)
; GFX81PLUS-NEXT: ret half [[S_FLOAT]]
;
- %s_float = call float @llvm.amdgcn.image.load.1d.f32.i16(i32 1, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ %s_float = call float @llvm.amdgcn.image.load.1d.f32.i16.v8i32(i32 1, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
%s_half = fptrunc float %s_float to half
ret half %s_half
}
@@ -191,7 +191,7 @@ define amdgpu_ps half @load_1d_v2(i16 %s, <8 x i32> inreg %rsrc) {
; GFX81PLUS-NEXT: [[ADDF_SUM_0:%.*]] = fadd half [[S0]], [[S1]]
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_0]]
;
- %v2_float = call <2 x float> @llvm.amdgcn.image.load.1d.v2f32.i16(i32 3, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ %v2_float = call <2 x float> @llvm.amdgcn.image.load.1d.v2f32.i16.v8i32(i32 3, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
%v2_half = fptrunc <2 x float> %v2_float to <2 x half>
%s0 = extractelement <2 x half> %v2_half, i64 0
%s1 = extractelement <2 x half> %v2_half, i64 1
@@ -219,7 +219,7 @@ define amdgpu_ps half @load_1d_v3(i16 %s, <8 x i32> inreg %rsrc) {
; GFX81PLUS-NEXT: [[ADDF_SUM_1:%.*]] = fadd half [[S2]], [[ADDF_SUM_0]]
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_1]]
;
- %v3_float = call <3 x float> @llvm.amdgcn.image.load.1d.v3f32.i16(i32 7, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ %v3_float = call <3 x float> @llvm.amdgcn.image.load.1d.v3f32.i16.v8i32(i32 7, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
%v3_half = fptrunc <3 x float> %v3_float to <3 x half>
%s0 = extractelement <3 x half> %v3_half, i64 0
%s1 = extractelement <3 x half> %v3_half, i64 1
@@ -253,7 +253,7 @@ define amdgpu_ps half @load_1d_v4(i16 %s, <8 x i32> inreg %rsrc) {
; GFX81PLUS-NEXT: [[ADDF_SUM_2:%.*]] = fadd half [[ADDF_SUM_0]], [[ADDF_SUM_1]]
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_2]]
;
- %v4_float = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ %v4_float = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
%v4_half = fptrunc <4 x float> %v4_float to <4 x half>
%s0 = extractelement <4 x half> %v4_half, i64 0
%s1 = extractelement <4 x half> %v4_half, i64 1
@@ -278,7 +278,7 @@ define amdgpu_ps half @load_2dmsaa(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32 %f
; GFX81PLUS-NEXT: ret half [[S_FLOAT]]
;
main_body:
- %s_float = call float @llvm.amdgcn.image.msaa.load.x.2dmsaa.f32.i32(i32 1, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
+ %s_float = call float @llvm.amdgcn.image.msaa.load.x.2dmsaa.f32.i32.v8i32(i32 1, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
%s_half = fptrunc float %s_float to half
ret half %s_half
}
@@ -302,7 +302,7 @@ define amdgpu_ps half @load_2dmsaa_v2(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_0]]
;
main_body:
- %v2_float = call <2 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v2f32.i32(i32 3, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
+ %v2_float = call <2 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v2f32.i32.v8i32(i32 3, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
%v2_half = fptrunc <2 x float> %v2_float to <2 x half>
%s0 = extractelement <2 x half> %v2_half, i64 0
%s1 = extractelement <2 x half> %v2_half, i64 1
@@ -333,7 +333,7 @@ define amdgpu_ps half @load_2dmsaa_v3(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_1]]
;
main_body:
- %v3_float = call <3 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v3f32.i32(i32 7, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
+ %v3_float = call <3 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v3f32.i32.v8i32(i32 7, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
%v3_half = fptrunc <3 x float> %v3_float to <3 x half>
%s0 = extractelement <3 x half> %v3_half, i64 0
%s1 = extractelement <3 x half> %v3_half, i64 1
@@ -370,7 +370,7 @@ define amdgpu_ps half @load_2dmsaa_v4(<8 x i32> inreg %rsrc, i32 %s, i32 %t, i32
; GFX81PLUS-NEXT: ret half [[ADDF_SUM_2]]
;
main_body:
- %v4_float = call <4 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v4f32.i32(i32 15, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
+ %v4_float = call <4 x float> @llvm.amdgcn.image.msaa.load.x.2dmsaa.v4f32.i32.v8i32(i32 15, i32 %s, i32 %t, i32 %fragid, <8 x i32> %rsrc, i32 0, i32 0)
%v4_half = fptrunc <4 x float> %v4_float to <4 x half>
%s0 = extractelement <4 x half> %v4_half, i64 0
%s1 = extractelement <4 x half> %v4_half, i64 1
diff --git a/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll b/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll
index 3de0ac17d840fc..cb99632c287b38 100644
--- a/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll
+++ b/llvm/test/Verifier/AMDGPU/intrinsic-immarg.ll
@@ -36,27 +36,27 @@ define void @struct_buffer_load_format_f32(<4 x i32> inreg %rsrc, i32 %idx, i32
ret void
}
-declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32)
+declare <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32)
define void @invalid_image_sample_1d_v4f32_f32(float %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc, i32 %dmask, i1 %bool, i32 %arg) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %dmask
; CHECK-NEXT: %data0 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 %dmask, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
- %data0 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 %dmask, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
+ %data0 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 %dmask, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i1 %bool
; CHECK-NEXT: %data1 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 %bool, i32 0, i32 0)
- %data1 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 %bool, i32 0, i32 0)
+ %data1 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 %bool, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %arg
; CHECK-NEXT: %data2 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 %arg, i32 0)
- %data2 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 %arg, i32 0)
+ %data2 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 %arg, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %arg
; CHECK-NEXT: %data3 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 %arg)
- %data3 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 %arg)
+ %data3 = call <4 x float> @llvm.amdgcn.image.sample.1d.v4f32.f32.v8i32.v4i32(i32 0, float %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 %arg)
ret void
}
@@ -336,84 +336,84 @@ define amdgpu_kernel void @update_dpp_test(ptr addrspace(1) %out, i32 %in1, i32
ret void
}
-declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32, i32, <8 x i32>, i32, i32)
+declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32, i32, <8 x i32>, i32, i32)
define amdgpu_ps void @load_1d(<8 x i32> inreg %rsrc, i32 %s, i32 %var) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 %var, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
- %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 %var, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 %var, i32 %s, <8 x i32> %rsrc, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %var, i32 0)
- %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %var, i32 0)
+ %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %var, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 %var)
- %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 %var)
+ %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 0, i32 %var)
ret void
}
-declare {<4 x float>,i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32, i32, <8 x i32>, i32, i32)
+declare {<4 x float>,i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32)
define amdgpu_ps void @load_1d_tfe(<8 x i32> inreg %rsrc, ptr addrspace(1) inreg %out, i32 %s, i32 %val) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val0 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 %val, i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
- %val0 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 %val, i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
+ %val0 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32.v8i32(i32 %val, i32 %s, <8 x i32> %rsrc, i32 1, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val1 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
- %val1 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
+ %val1 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val2 = call { <4 x float>, i32 } @llvm.amdgcn.image.load.1d.sl_v4f32i32s.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 1, i32 %val)
- %val2 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 1, i32 %val)
+ %val2 = call {<4 x float>, i32} @llvm.amdgcn.image.load.1d.v4f32i32.i32.v8i32(i32 15, i32 %s, <8 x i32> %rsrc, i32 1, i32 %val)
ret void
}
-declare {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32)
+declare {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32.v8i32.v4i32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32)
define amdgpu_ps void @sample_1d_tfe(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, ptr addrspace(1) inreg %out, float %s, i32 %var, i1 %bool) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val0 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 0)
- %val0 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 0)
+ %val0 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32.v8i32.v4i32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i1 %bool
; CHECK-NEXT: %val1 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 %bool, i32 1, i32 0)
- %val1 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 %bool, i32 1, i32 0)
+ %val1 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32.v8i32.v4i32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 %bool, i32 1, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val2 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 %var, i32 0)
- %val2 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 %var, i32 0)
+ %val2 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32.v8i32.v4i32(i32 16, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 %var, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val3 = call { <4 x float>, i32 } @llvm.amdgcn.image.sample.1d.sl_v4f32i32s.f32.v8i32.v4i32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 %var)
- %val3 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 %var)
+ %val3 = call {<4 x float>, i32} @llvm.amdgcn.image.sample.1d.v4f32i32.f32.v8i32.v4i32(i32 %var, float %s, <8 x i32> %rsrc, <4 x i32> %samp, i1 false, i32 1, i32 %var)
ret void
}
-declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32, i16, <8 x i32>, i32, i32)
+declare <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32, i16, <8 x i32>, i32, i32)
define amdgpu_ps void @load_1d_a16(<8 x i32> inreg %rsrc, <2 x i16> %coords, i16 %s, i32 %var) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 %var, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
- %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 %var, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
+ %val0 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 %var, i16 %s, <8 x i32> %rsrc, i32 0, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 %var, i32 0)
- %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 %var, i32 0)
+ %val1 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 %var, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %var
; CHECK-NEXT: %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 %var)
- %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 %var)
+ %val2 = call <4 x float> @llvm.amdgcn.image.load.1d.v4f32.i16.v8i32(i32 15, i16 %s, <8 x i32> %rsrc, i32 0, i32 %var)
ret void
}
@@ -426,31 +426,31 @@ define amdgpu_ps void @raw_buffer_atomic_swap(<4 x i32> inreg %rsrc, i32 %data,
ret void
}
-declare i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32, i32, <8 x i32>, i32, i32)
+declare i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32, i32, <8 x i32>, i32, i32)
define amdgpu_ps void @atomic_swap_1d(<8 x i32> inreg %rsrc, i32 %data, i32 %s, i32 %val) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val0 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
- %val0 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
+ %val0 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val1 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
- %val1 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
+ %val1 = call i32 @llvm.amdgcn.image.atomic.swap.1d.i32.i32.v8i32(i32 %data, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
ret void
}
-declare i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32, i32, i32, <8 x i32>, i32, i32) #0
+declare i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32, i32, i32, <8 x i32>, i32, i32) #0
define amdgpu_ps void @atomic_cmpswap_1d(<8 x i32> inreg %rsrc, i32 %cmp, i32 %swap, i32 %s, i32 %val) {
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val0 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
- %val0 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
+ %val0 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 %val, i32 0)
; CHECK: immarg operand has non-immediate parameter
; CHECK-NEXT: i32 %val
; CHECK-NEXT: %val1 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
- %val1 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
+ %val1 = call i32 @llvm.amdgcn.image.atomic.cmpswap.1d.i32.i32.v8i32(i32 %cmp, i32 %swap, i32 %s, <8 x i32> %rsrc, i32 0, i32 %val)
ret void
}
>From b7099df2af18bd6cffdeb2fedff96abe9f4ff398 Mon Sep 17 00:00:00 2001
From: Stefan Stipanovic <Stefan.Stipanovic at amd.com>
Date: Thu, 26 Sep 2024 16:52:30 +0200
Subject: [PATCH 3/3] Fail with cannot select in case rsrc is neither v8i32 nor
v4i32
---
llvm/lib/Target/AMDGPU/SIISelLowering.cpp | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index 885ecab891b1f5..79ec9c0db44889 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -8209,9 +8209,17 @@ SDValue SITargetLowering::lowerImage(SDValue Op,
append_range(Ops, VAddrs);
else
Ops.push_back(VAddr);
- Ops.push_back(Op.getOperand(ArgOffset + Intr->RsrcIndex));
- if (BaseOpcode->Sampler)
- Ops.push_back(Op.getOperand(ArgOffset + Intr->SampIndex));
+ SDValue Rsrc = Op.getOperand(ArgOffset + Intr->RsrcIndex);
+ EVT VT = Rsrc.getValueType();
+ if (VT != MVT::v4i32 && VT != MVT::v8i32)
+ return Op;
+ Ops.push_back(Rsrc);
+ if (BaseOpcode->Sampler) {
+ SDValue Samp = Op.getOperand(ArgOffset + Intr->SampIndex);
+ if (Samp.getValueType() != MVT::v4i32)
+ return Op;
+ Ops.push_back(Samp);
+ }
Ops.push_back(DAG.getTargetConstant(DMask, DL, MVT::i32));
if (IsGFX10Plus)
Ops.push_back(DAG.getTargetConstant(DimInfo->Encoding, DL, MVT::i32));
More information about the llvm-commits
mailing list