[clang] [llvm] [AMDGPU] Add support for `v_permlane16_swap_b32` on gfx1250 (PR #149518)
Shilei Tian via cfe-commits
cfe-commits at lists.llvm.org
Fri Jul 18 09:51:19 PDT 2025
https://github.com/shiltian updated https://github.com/llvm/llvm-project/pull/149518
>From 1d045407b88bc8efae91410223e8ba980cdec6d1 Mon Sep 17 00:00:00 2001
From: Shilei Tian <i at tianshilei.me>
Date: Fri, 18 Jul 2025 10:02:30 -0400
Subject: [PATCH] [AMDGPU] Add support for `v_permlane16_swap_b32` on gfx1250
Co-authored-by: Mekhanoshin, Stanislav <Stanislav.Mekhanoshin at amd.com>
---
.../CodeGenOpenCL/builtins-amdgcn-gfx1250.cl | 47 ++++++
llvm/lib/Target/AMDGPU/VOP1Instructions.td | 8 +
llvm/lib/Target/AMDGPU/VOPInstructions.td | 11 +-
.../AMDGPU/llvm.amdgcn.permlane16.swap.ll | 153 +++++++++++++++++-
llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s | 6 +
llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s | 6 +
.../gfx1250_asm_vop3_from_vop1-fake16.s | 21 +++
.../MC/AMDGPU/gfx1250_asm_vop3_from_vop1.s | 21 +++
8 files changed, 270 insertions(+), 3 deletions(-)
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
index d42e51d04ab9d..4c3f308a6cf75 100644
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
+++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx1250.cl
@@ -5,6 +5,7 @@
#pragma OPENCL EXTENSION cl_khr_fp16 : enable
typedef unsigned int uint;
+typedef unsigned int __attribute__((ext_vector_type(2))) uint2;
typedef half __attribute__((ext_vector_type(2))) half2;
// CHECK-LABEL: @test_setprio_inc_wg(
@@ -368,6 +369,52 @@ void test_cvt_pk_f16_bf8(global half2* out, short a)
out[0] = __builtin_amdgcn_cvt_pk_f16_bf8(a);
}
+// CHECK-LABEL: @test_permlane16_swap(
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
+// CHECK-NEXT: [[OLD_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// CHECK-NEXT: [[SRC_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
+// CHECK-NEXT: [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[OUT_ADDR]] to ptr
+// CHECK-NEXT: [[OLD_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[OLD_ADDR]] to ptr
+// CHECK-NEXT: [[SRC_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SRC_ADDR]] to ptr
+// CHECK-NEXT: store ptr addrspace(1) [[OUT:%.*]], ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT: store i32 [[OLD:%.*]], ptr [[OLD_ADDR_ASCAST]], align 4
+// CHECK-NEXT: store i32 [[SRC:%.*]], ptr [[SRC_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP0:%.*]] = load i32, ptr [[OLD_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP1:%.*]] = load i32, ptr [[SRC_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP2:%.*]] = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 [[TMP0]], i32 [[TMP1]], i1 false, i1 false)
+// CHECK-NEXT: [[TMP3:%.*]] = extractvalue { i32, i32 } [[TMP2]], 0
+// CHECK-NEXT: [[TMP4:%.*]] = extractvalue { i32, i32 } [[TMP2]], 1
+// CHECK-NEXT: [[TMP5:%.*]] = insertelement <2 x i32> poison, i32 [[TMP3]], i64 0
+// CHECK-NEXT: [[TMP6:%.*]] = insertelement <2 x i32> [[TMP5]], i32 [[TMP4]], i64 1
+// CHECK-NEXT: [[TMP7:%.*]] = load ptr addrspace(1), ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT: store <2 x i32> [[TMP6]], ptr addrspace(1) [[TMP7]], align 8
+// CHECK-NEXT: [[TMP8:%.*]] = load i32, ptr [[OLD_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[SRC_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP10:%.*]] = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 [[TMP8]], i32 [[TMP9]], i1 true, i1 false)
+// CHECK-NEXT: [[TMP11:%.*]] = extractvalue { i32, i32 } [[TMP10]], 0
+// CHECK-NEXT: [[TMP12:%.*]] = extractvalue { i32, i32 } [[TMP10]], 1
+// CHECK-NEXT: [[TMP13:%.*]] = insertelement <2 x i32> poison, i32 [[TMP11]], i64 0
+// CHECK-NEXT: [[TMP14:%.*]] = insertelement <2 x i32> [[TMP13]], i32 [[TMP12]], i64 1
+// CHECK-NEXT: [[TMP15:%.*]] = load ptr addrspace(1), ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT: store <2 x i32> [[TMP14]], ptr addrspace(1) [[TMP15]], align 8
+// CHECK-NEXT: [[TMP16:%.*]] = load i32, ptr [[OLD_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP17:%.*]] = load i32, ptr [[SRC_ADDR_ASCAST]], align 4
+// CHECK-NEXT: [[TMP18:%.*]] = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 [[TMP16]], i32 [[TMP17]], i1 false, i1 true)
+// CHECK-NEXT: [[TMP19:%.*]] = extractvalue { i32, i32 } [[TMP18]], 0
+// CHECK-NEXT: [[TMP20:%.*]] = extractvalue { i32, i32 } [[TMP18]], 1
+// CHECK-NEXT: [[TMP21:%.*]] = insertelement <2 x i32> poison, i32 [[TMP19]], i64 0
+// CHECK-NEXT: [[TMP22:%.*]] = insertelement <2 x i32> [[TMP21]], i32 [[TMP20]], i64 1
+// CHECK-NEXT: [[TMP23:%.*]] = load ptr addrspace(1), ptr [[OUT_ADDR_ASCAST]], align 8
+// CHECK-NEXT: store <2 x i32> [[TMP22]], ptr addrspace(1) [[TMP23]], align 8
+// CHECK-NEXT: ret void
+//
+void test_permlane16_swap(global uint2* out, uint old, uint src) {
+ *out = __builtin_amdgcn_permlane16_swap(old, src, false, false);
+ *out = __builtin_amdgcn_permlane16_swap(old, src, true, false);
+ *out = __builtin_amdgcn_permlane16_swap(old, src, false, true);
+}
+
// CHECK-LABEL: @test_cvt_f32_fp8_e5m3(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(1), align 8, addrspace(5)
diff --git a/llvm/lib/Target/AMDGPU/VOP1Instructions.td b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
index 3ee90857b34b8..80eb5d8b7d571 100644
--- a/llvm/lib/Target/AMDGPU/VOP1Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
@@ -1080,6 +1080,13 @@ multiclass VOP1_Real_FULL_t16_and_fake16_gfx1250<
VOP1_Real_FULL_with_name<GFX1250Gen, op, opName#"_fake16", asmName>;
}
+multiclass VOP1_Real_OpSelIsDPP_gfx1250<bits<9> op> : VOP1_Real_e32<GFX1250Gen, op> {
+ defvar ps = !cast<VOP_Pseudo>(NAME#"_e64");
+ def _e64_gfx1250 :
+ VOP3_Real_Gen<ps, GFX1250Gen>,
+ VOP3OpSelIsDPP_gfx12<{0, 1, 1, op{6-0}}, ps.Pfl>;
+}
+
defm V_CVT_F32_FP8 : VOP1_Real_FULL_with_name<GFX12Not12_50Gen, 0x06c, "V_CVT_F32_FP8_OP_SEL", "v_cvt_f32_fp8">;
defm V_CVT_F32_FP8 : VOP1_Real_FULL_with_name<GFX1250Gen, 0x06c, "V_CVT_F32_FP8_gfx1250", "v_cvt_f32_fp8">;
@@ -1147,6 +1154,7 @@ defm V_MOV_B64 : VOP1_Real_FULL <GFX1250Gen, 0x1d>;
defm V_TANH_F32 : VOP1_Real_FULL<GFX1250Gen, 0x01e>;
defm V_TANH_F16 : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x01f>;
+defm V_PERMLANE16_SWAP_B32 : VOP1_Real_OpSelIsDPP_gfx1250<0x049>;
defm V_TANH_BF16 : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x04a>;
defm V_PRNG_B32 : VOP1_Real_FULL<GFX1250Gen, 0x04b>;
defm V_CVT_F32_BF16 : VOP1_Real_FULL_t16_and_fake16_gfx1250<0x072, "v_cvt_f32_bf16", "V_CVT_F32_BF16_gfx1250">;
diff --git a/llvm/lib/Target/AMDGPU/VOPInstructions.td b/llvm/lib/Target/AMDGPU/VOPInstructions.td
index 2b91ea7386be4..a25ebdf3e5f6d 100644
--- a/llvm/lib/Target/AMDGPU/VOPInstructions.td
+++ b/llvm/lib/Target/AMDGPU/VOPInstructions.td
@@ -331,10 +331,19 @@ class VOP3OpSel_gfx9 <bits<10> op, VOPProfile P> : VOP3e_vi <op, P> {
// Special case for v_permlane16_swap_b32/v_permlane32_swap_b32
// op_sel[0]/op_sel[1] are treated as bound_ctrl and fi dpp operands.
-class VOP3OpSelIsDPP_gfx9 <bits<10> op, VOPProfile P> : VOP3e_vi <op, P> {
+class VOP3OpSelIsDPP_base {
bits<1> fi;
bits<1> bound_ctrl;
+}
+
+class VOP3OpSelIsDPP_gfx9 <bits<10> op, VOPProfile P> : VOP3OpSelIsDPP_base, VOP3e_vi <op, P> {
+ // OPSEL[0] specifies FI
+ let Inst{11} = fi;
+ // OPSEL[1] specifies BOUND_CTRL
+ let Inst{12} = bound_ctrl;
+}
+class VOP3OpSelIsDPP_gfx12 <bits<10> op, VOPProfile P> : VOP3OpSelIsDPP_base, VOP3e_gfx11_gfx12 <op, P> {
// OPSEL[0] specifies FI
let Inst{11} = fi;
// OPSEL[1] specifies BOUND_CTRL
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.permlane16.swap.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.permlane16.swap.ll
index 814086685880d..ed6a02b62ae9a 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.permlane16.swap.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.permlane16.swap.ll
@@ -1,6 +1,8 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx950 < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx950 < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx950 < %s | FileCheck -check-prefix=GFX950 %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx950 < %s | FileCheck -check-prefix=GFX950 %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1250 < %s | FileCheck -check-prefix=GFX1250 %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1250 < %s | FileCheck -check-prefix=GFX1250 %s
; RUN: not --crash llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 -filetype=null %s 2>&1 | FileCheck -check-prefix=ERR-SDAG %s
; RUN: not llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 -filetype=null %s 2>&1 | FileCheck -check-prefix=ERR-GISEL %s
@@ -17,6 +19,18 @@ define { i32, i32 } @v_permlane16_swap_b32_vv(i32 %vdst_old, i32 %src0_old) {
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vv:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vv:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -29,6 +43,22 @@ define { i32, i32 } @v_permlane16_swap_b32_vi(i32 %vdst_old) {
; GCN-NEXT: s_nop 1
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vi:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_mov_b32_e32 v1, 1
+; GFX950-NEXT: s_nop 1
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vi:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_mov_b32_e32 v1, 1
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 1, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -41,6 +71,22 @@ define { i32, i32 } @v_permlane16_swap_b32_vl(i32 %vdst_old) {
; GCN-NEXT: s_nop 1
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vl:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_mov_b32_e32 v1, 0xc1d1
+; GFX950-NEXT: s_nop 1
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vl:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_mov_b32_e32 v1, 0xc1d1
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 49617, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -54,6 +100,23 @@ define { i32, i32 } @v_permlane16_swap_b32_iv(i32 %src0_old) {
; GCN-NEXT: s_nop 1
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_iv:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_mov_b32_e32 v1, v0
+; GFX950-NEXT: v_mov_b32_e32 v0, 1
+; GFX950-NEXT: s_nop 1
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_iv:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_dual_mov_b32 v1, v0 :: v_dual_mov_b32 v0, 1
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 1, i32 %src0_old, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -67,6 +130,23 @@ define { i32, i32 } @v_permlane16_swap_b32_ss(i32 inreg %vdst_old, i32 inreg %sr
; GCN-NEXT: s_nop 1
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_ss:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_mov_b32_e32 v0, s0
+; GFX950-NEXT: v_mov_b32_e32 v1, s1
+; GFX950-NEXT: s_nop 1
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_ss:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_dual_mov_b32 v0, s0 :: v_dual_mov_b32 v1, s1
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -80,6 +160,23 @@ define { i32, i32 } @v_permlane16_swap_b32_sv(i32 inreg %vdst_old, i32 %src0_old
; GCN-NEXT: s_nop 1
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_sv:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_mov_b32_e32 v1, v0
+; GFX950-NEXT: v_mov_b32_e32 v0, s0
+; GFX950-NEXT: s_nop 1
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_sv:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_dual_mov_b32 v1, v0 :: v_dual_mov_b32 v0, s0
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -92,6 +189,22 @@ define { i32, i32 } @v_permlane16_swap_b32_vs(i32 %vdst_old, i32 inreg %src0_old
; GCN-NEXT: s_nop 1
; GCN-NEXT: v_permlane16_swap_b32_e32 v0, v1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vs:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_mov_b32_e32 v1, s0
+; GFX950-NEXT: s_nop 1
+; GFX950-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vs:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_mov_b32_e32 v1, s0
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_permlane16_swap_b32_e32 v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 false, i1 false)
ret { i32, i32 } %v
}
@@ -102,6 +215,18 @@ define { i32, i32 } @v_permlane16_swap_b32_vv_fi(i32 %vdst_old, i32 %src0_old) {
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_permlane16_swap_b32_e64 v0, v1 fi:1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vv_fi:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_permlane16_swap_b32_e64 v0, v1 fi:1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vv_fi:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_permlane16_swap_b32_e64 v0, v1 fi:1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 true, i1 false)
ret { i32, i32 } %v
}
@@ -112,6 +237,18 @@ define { i32, i32 } @v_permlane16_swap_b32_vv_bc(i32 %vdst_old, i32 %src0_old) {
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_permlane16_swap_b32_e64 v0, v1 bound_ctrl:1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vv_bc:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_permlane16_swap_b32_e64 v0, v1 bound_ctrl:1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vv_bc:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_permlane16_swap_b32_e64 v0, v1 bound_ctrl:1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 false, i1 true)
ret { i32, i32 } %v
}
@@ -122,6 +259,18 @@ define { i32, i32 } @v_permlane16_swap_b32_vv_fi_bc(i32 %vdst_old, i32 %src0_old
; GCN-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GCN-NEXT: v_permlane16_swap_b32_e64 v0, v1 bound_ctrl:1 fi:1
; GCN-NEXT: s_setpc_b64 s[30:31]
+; GFX950-LABEL: v_permlane16_swap_b32_vv_fi_bc:
+; GFX950: ; %bb.0:
+; GFX950-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX950-NEXT: v_permlane16_swap_b32_e64 v0, v1 bound_ctrl:1 fi:1
+; GFX950-NEXT: s_setpc_b64 s[30:31]
+;
+; GFX1250-LABEL: v_permlane16_swap_b32_vv_fi_bc:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_permlane16_swap_b32_e64 v0, v1 bound_ctrl:1 fi:1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%v = call { i32, i32 } @llvm.amdgcn.permlane16.swap(i32 %vdst_old, i32 %src0_old, i1 true, i1 true)
ret { i32, i32 } %v
}
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s
index 5f310a9954ad0..f2cf3d58fb0cf 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1-fake16.s
@@ -627,3 +627,9 @@ v_cvt_f32_fp8_e32 v1, 3
v_cvt_f32_fp8_e32 v1, v3
// GFX1250: v_cvt_f32_fp8_e32 v1, v3 ; encoding: [0x03,0xd9,0x02,0x7e]
+
+v_permlane16_swap_b32 v1, v2
+// GFX1250: v_permlane16_swap_b32_e32 v1, v2 ; encoding: [0x02,0x93,0x02,0x7e]
+
+v_permlane16_swap_b32_e32 v1, v2
+// GFX1250: v_permlane16_swap_b32_e32 v1, v2 ; encoding: [0x02,0x93,0x02,0x7e]
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s
index aa2e028f661e1..b1c4dc62edd6d 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop1.s
@@ -663,3 +663,9 @@ v_cvt_f32_fp8_e32 v1, 3
v_cvt_f32_fp8_e32 v1, v3
// GFX1250: v_cvt_f32_fp8_e32 v1, v3 ; encoding: [0x03,0xd9,0x02,0x7e]
+
+v_permlane16_swap_b32 v1, v2
+// GFX1250: v_permlane16_swap_b32_e32 v1, v2 ; encoding: [0x02,0x93,0x02,0x7e]
+
+v_permlane16_swap_b32_e32 v1, v2
+// GFX1250: v_permlane16_swap_b32_e32 v1, v2 ; encoding: [0x02,0x93,0x02,0x7e]
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1-fake16.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1-fake16.s
index 9c6a9127d82e4..6b45930a53d73 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1-fake16.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1-fake16.s
@@ -720,3 +720,24 @@ v_cvt_pk_f16_fp8 v1, v150 op_sel:[1]
v_cvt_pk_f16_fp8 v1, s2 op_sel:[1]
// GFX1250: v_cvt_pk_f16_fp8 v1, s2 op_sel:[1,0] ; encoding: [0x01,0x08,0xf5,0xd5,0x02,0x00,0x00,0x00]
+
+v_permlane16_swap_b32_e64 v1, v2
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 ; encoding: [0x01,0x00,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 bound_ctrl:0
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 ; encoding: [0x01,0x00,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 fi:0
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 ; encoding: [0x01,0x00,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 bound_ctrl:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 ; encoding: [0x01,0x10,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 fi:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 fi:1 ; encoding: [0x01,0x08,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 bound_ctrl:1 fi:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 fi:1 ; encoding: [0x01,0x18,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 fi:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 fi:1 ; encoding: [0x01,0x18,0xc9,0xd5,0x02,0x01,0x00,0x00]
diff --git a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1.s b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1.s
index 2f57d1c331c42..ad00832f7543d 100644
--- a/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1.s
+++ b/llvm/test/MC/AMDGPU/gfx1250_asm_vop3_from_vop1.s
@@ -750,3 +750,24 @@ v_cvt_pk_f16_fp8 v1, v150 op_sel:[1]
v_cvt_pk_f16_fp8 v1, s2 op_sel:[1]
// GFX1250: v_cvt_pk_f16_fp8 v1, s2 op_sel:[1,0] ; encoding: [0x01,0x08,0xf5,0xd5,0x02,0x00,0x00,0x00]
+
+v_permlane16_swap_b32_e64 v1, v2
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 ; encoding: [0x01,0x00,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 bound_ctrl:0
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 ; encoding: [0x01,0x00,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 fi:0
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 ; encoding: [0x01,0x00,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 bound_ctrl:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 ; encoding: [0x01,0x10,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 fi:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 fi:1 ; encoding: [0x01,0x08,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32 v1, v2 bound_ctrl:1 fi:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 fi:1 ; encoding: [0x01,0x18,0xc9,0xd5,0x02,0x01,0x00,0x00]
+
+v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 fi:1
+// GFX1250: v_permlane16_swap_b32_e64 v1, v2 bound_ctrl:1 fi:1 ; encoding: [0x01,0x18,0xc9,0xd5,0x02,0x01,0x00,0x00]
More information about the cfe-commits
mailing list