[llvm] [AMDGPU] Implement codegen for GFX11+ V_CVT_PK_[IU]16_F32 (PR #168719)
Jay Foad via llvm-commits
llvm-commits at lists.llvm.org
Wed Nov 19 06:56:28 PST 2025
https://github.com/jayfoad created https://github.com/llvm/llvm-project/pull/168719
None
>From a871e4cfaa4224966bc8e2dc46f89835c568e3a0 Mon Sep 17 00:00:00 2001
From: Jay Foad <jay.foad at amd.com>
Date: Wed, 19 Nov 2025 14:17:19 +0000
Subject: [PATCH] [AMDGPU] Implement codegen for GFX11+ V_CVT_PK_[IU]16_F32
---
llvm/lib/Target/AMDGPU/SIISelLowering.cpp | 5 +
llvm/lib/Target/AMDGPU/SOPInstructions.td | 8 +
llvm/lib/Target/AMDGPU/VOP3Instructions.td | 22 ++
llvm/test/CodeGen/AMDGPU/bf16.ll | 231 ++++--------
llvm/test/CodeGen/AMDGPU/fp_to_sint.ll | 329 ++++++++++++++++++
llvm/test/CodeGen/AMDGPU/fp_to_uint.ll | 273 +++++++++++++++
llvm/test/CodeGen/AMDGPU/scalar-float-sop1.ll | 26 +-
7 files changed, 718 insertions(+), 176 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index e37d739fc25df..0a9ae52d5dabf 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -6806,6 +6806,11 @@ SDValue SITargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const {
return splitTernaryVectorOp(Op, DAG);
case ISD::FP_TO_SINT:
case ISD::FP_TO_UINT:
+ if (AMDGPU::isGFX11Plus(*Subtarget) && Op.getValueType() == MVT::i16 &&
+ Op.getOperand(0).getValueType() == MVT::f32) {
+ // Make f32->i16 legal so we can select V_CVT_PK_[IU]16_F32.
+ return Op;
+ }
return LowerFP_TO_INT(Op, DAG);
case ISD::SHL:
case ISD::SRA:
diff --git a/llvm/lib/Target/AMDGPU/SOPInstructions.td b/llvm/lib/Target/AMDGPU/SOPInstructions.td
index 1931e0be15152..936eb545563ae 100644
--- a/llvm/lib/Target/AMDGPU/SOPInstructions.td
+++ b/llvm/lib/Target/AMDGPU/SOPInstructions.td
@@ -469,6 +469,14 @@ let SubtargetPredicate = HasSALUFloatInsts, Uses = [MODE],
} // End SubtargetPredicate = HasSALUFloatInsts, Uses = [MODE]
// SchedRW = [WriteSFPU], isReMaterializable = 1
+let SubtargetPredicate = HasSALUFloatInsts, AddedComplexity = 9 in {
+ // Fallback patterns for f32->i16 conversion.
+ def : GCNPat<(i16 (UniformUnaryFrag<fp_to_sint> f32:$src0)),
+ (S_CVT_I32_F32 $src0)>;
+ def : GCNPat<(i16 (UniformUnaryFrag<fp_to_uint> f32:$src0)),
+ (S_CVT_U32_F32 $src0)>;
+}
+
let hasSideEffects = 1 in {
let has_sdst = 0 in {
let Uses = [M0] in {
diff --git a/llvm/lib/Target/AMDGPU/VOP3Instructions.td b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
index 872bde501cd2d..e1b22c6804544 100644
--- a/llvm/lib/Target/AMDGPU/VOP3Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
@@ -1721,6 +1721,28 @@ let SubtargetPredicate = isGFX11Plus in {
defm V_MINMAX_I32 : VOP3Inst<"v_minmax_i32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
defm V_CVT_PK_I16_F32 : VOP3Inst<"v_cvt_pk_i16_f32", VOP3_Profile<VOP_V2I16_F32_F32>>;
defm V_CVT_PK_U16_F32 : VOP3Inst<"v_cvt_pk_u16_f32", VOP3_Profile<VOP_V2I16_F32_F32>>;
+
+ def : GCNPat<(v2i16 (build_vector (i16 (fp_to_sint (f32 (VOP3Mods f32:$src0, i32:$src0_modifiers)))),
+ (i16 (fp_to_sint (f32 (VOP3Mods f32:$src1, i32:$src1_modifiers)))))),
+ (V_CVT_PK_I16_F32_e64 $src0_modifiers, $src0, $src1_modifiers, $src1)>;
+ def : GCNPat<(v2i16 (build_vector (i16 (fp_to_uint (f32 (VOP3Mods f32:$src0, i32:$src0_modifiers)))),
+ (i16 (fp_to_uint (f32 (VOP3Mods f32:$src1, i32:$src1_modifiers)))))),
+ (V_CVT_PK_U16_F32_e64 $src0_modifiers, $src0, $src1_modifiers, $src1)>;
+
+ // Fallback patterns for f32->i16 conversion. These are only required because
+ // f32->i16 has to be legal so that we can select V_CVT_PK_[IU]16_F32 above.
+ let True16Predicate = UseRealTrue16Insts in {
+ def : GCNPat<(i16 (fp_to_sint (f32 (VOP3Mods f32:$src0, i32:$src0_modifiers)))),
+ (EXTRACT_SUBREG (V_CVT_I32_F32_e64 $src0_modifiers, $src0), lo16)>;
+ def : GCNPat<(i16 (fp_to_uint (f32 (VOP3Mods f32:$src0, i32:$src0_modifiers)))),
+ (EXTRACT_SUBREG (V_CVT_U32_F32_e64 $src0_modifiers, $src0), lo16)>;
+ }
+ let True16Predicate = NotUseRealTrue16Insts in {
+ def : GCNPat<(i16 (fp_to_sint (f32 (VOP3Mods f32:$src0, i32:$src0_modifiers)))),
+ (V_CVT_I32_F32_e64 $src0_modifiers, $src0)>;
+ def : GCNPat<(i16 (fp_to_uint (f32 (VOP3Mods f32:$src0, i32:$src0_modifiers)))),
+ (V_CVT_U32_F32_e64 $src0_modifiers, $src0)>;
+ }
} // End SubtargetPredicate = isGFX11Plus
class VOP3_CVT_SR_FP16_TiedInput_Profile<VOPProfile P> : VOP3_CVT_SCALE_F1632_FP8BF8_TiedInput_Profile<P> {
diff --git a/llvm/test/CodeGen/AMDGPU/bf16.ll b/llvm/test/CodeGen/AMDGPU/bf16.ll
index 28d7e6916e519..c9b2c8c08b41c 100644
--- a/llvm/test/CodeGen/AMDGPU/bf16.ll
+++ b/llvm/test/CodeGen/AMDGPU/bf16.ll
@@ -35513,55 +35513,24 @@ define <2 x i16> @v_fptosi_v2bf16_to_v2i16(<2 x bfloat> %x) {
; GFX10-NEXT: v_perm_b32 v0, v0, v1, 0x5040100
; GFX10-NEXT: s_setpc_b64 s[30:31]
;
-; GFX11TRUE16-LABEL: v_fptosi_v2bf16_to_v2i16:
-; GFX11TRUE16: ; %bb.0:
-; GFX11TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX11TRUE16-NEXT: v_and_b32_e32 v1, 0xffff0000, v0
-; GFX11TRUE16-NEXT: v_lshlrev_b32_e32 v0, 16, v0
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2)
-; GFX11TRUE16-NEXT: v_mov_b16_e32 v0.h, v1.l
-; GFX11TRUE16-NEXT: s_setpc_b64 s[30:31]
-;
-; GFX11FAKE16-LABEL: v_fptosi_v2bf16_to_v2i16:
-; GFX11FAKE16: ; %bb.0:
-; GFX11FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX11FAKE16-NEXT: v_lshlrev_b32_e32 v1, 16, v0
-; GFX11FAKE16-NEXT: v_and_b32_e32 v0, 0xffff0000, v0
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
-; GFX11FAKE16-NEXT: v_perm_b32 v0, v0, v1, 0x5040100
-; GFX11FAKE16-NEXT: s_setpc_b64 s[30:31]
-;
-; GFX1250TRUE16-LABEL: v_fptosi_v2bf16_to_v2i16:
-; GFX1250TRUE16: ; %bb.0:
-; GFX1250TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
-; GFX1250TRUE16-NEXT: s_wait_kmcnt 0x0
-; GFX1250TRUE16-NEXT: v_and_b32_e32 v1, 0xffff0000, v0
-; GFX1250TRUE16-NEXT: v_lshlrev_b32_e32 v0, 16, v0
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2)
-; GFX1250TRUE16-NEXT: v_mov_b16_e32 v0.h, v1.l
-; GFX1250TRUE16-NEXT: s_set_pc_i64 s[30:31]
+; GFX11-LABEL: v_fptosi_v2bf16_to_v2i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_and_b32_e32 v1, 0xffff0000, v0
+; GFX11-NEXT: v_lshlrev_b32_e32 v0, 16, v0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cvt_pk_i16_f32 v0, v0, v1
+; GFX11-NEXT: s_setpc_b64 s[30:31]
;
-; GFX1250FAKE16-LABEL: v_fptosi_v2bf16_to_v2i16:
-; GFX1250FAKE16: ; %bb.0:
-; GFX1250FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
-; GFX1250FAKE16-NEXT: s_wait_kmcnt 0x0
-; GFX1250FAKE16-NEXT: v_lshlrev_b32_e32 v1, 16, v0
-; GFX1250FAKE16-NEXT: v_and_b32_e32 v0, 0xffff0000, v0
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
-; GFX1250FAKE16-NEXT: v_perm_b32 v0, v0, v1, 0x5040100
-; GFX1250FAKE16-NEXT: s_set_pc_i64 s[30:31]
+; GFX1250-LABEL: v_fptosi_v2bf16_to_v2i16:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_and_b32_e32 v1, 0xffff0000, v0
+; GFX1250-NEXT: v_lshlrev_b32_e32 v0, 16, v0
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX1250-NEXT: v_cvt_pk_i16_f32 v0, v0, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%op = fptosi <2 x bfloat> %x to <2 x i16>
ret <2 x i16> %op
}
@@ -35655,61 +35624,27 @@ define <3 x i16> @v_fptosi_v3bf16_to_v3i16(<3 x bfloat> %x) {
; GFX10-NEXT: v_perm_b32 v0, v0, v2, 0x5040100
; GFX10-NEXT: s_setpc_b64 s[30:31]
;
-; GFX11TRUE16-LABEL: v_fptosi_v3bf16_to_v3i16:
-; GFX11TRUE16: ; %bb.0:
-; GFX11TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX11TRUE16-NEXT: v_and_b32_e32 v2, 0xffff0000, v0
-; GFX11TRUE16-NEXT: v_lshlrev_b32_e32 v0, 16, v0
-; GFX11TRUE16-NEXT: v_lshlrev_b32_e32 v1, 16, v1
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX11TRUE16-NEXT: v_mov_b16_e32 v0.h, v2.l
-; GFX11TRUE16-NEXT: s_setpc_b64 s[30:31]
-;
-; GFX11FAKE16-LABEL: v_fptosi_v3bf16_to_v3i16:
-; GFX11FAKE16: ; %bb.0:
-; GFX11FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX11FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v0
-; GFX11FAKE16-NEXT: v_and_b32_e32 v0, 0xffff0000, v0
-; GFX11FAKE16-NEXT: v_lshlrev_b32_e32 v1, 16, v1
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX11FAKE16-NEXT: v_perm_b32 v0, v0, v2, 0x5040100
-; GFX11FAKE16-NEXT: s_setpc_b64 s[30:31]
-;
-; GFX1250TRUE16-LABEL: v_fptosi_v3bf16_to_v3i16:
-; GFX1250TRUE16: ; %bb.0:
-; GFX1250TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
-; GFX1250TRUE16-NEXT: s_wait_kmcnt 0x0
-; GFX1250TRUE16-NEXT: v_and_b32_e32 v2, 0xffff0000, v0
-; GFX1250TRUE16-NEXT: v_dual_lshlrev_b32 v0, 16, v0 :: v_dual_lshlrev_b32 v1, 16, v1
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX1250TRUE16-NEXT: v_mov_b16_e32 v0.h, v2.l
-; GFX1250TRUE16-NEXT: s_set_pc_i64 s[30:31]
+; GFX11-LABEL: v_fptosi_v3bf16_to_v3i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_and_b32_e32 v2, 0xffff0000, v0
+; GFX11-NEXT: v_lshlrev_b32_e32 v0, 16, v0
+; GFX11-NEXT: v_lshlrev_b32_e32 v1, 16, v1
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_cvt_pk_i16_f32 v0, v0, v2
+; GFX11-NEXT: v_cvt_i32_f32_e32 v1, v1
+; GFX11-NEXT: s_setpc_b64 s[30:31]
;
-; GFX1250FAKE16-LABEL: v_fptosi_v3bf16_to_v3i16:
-; GFX1250FAKE16: ; %bb.0:
-; GFX1250FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
-; GFX1250FAKE16-NEXT: s_wait_kmcnt 0x0
-; GFX1250FAKE16-NEXT: v_dual_lshlrev_b32 v2, 16, v0 :: v_dual_lshlrev_b32 v1, 16, v1
-; GFX1250FAKE16-NEXT: v_and_b32_e32 v0, 0xffff0000, v0
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3)
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1)
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX1250FAKE16-NEXT: v_perm_b32 v0, v0, v2, 0x5040100
-; GFX1250FAKE16-NEXT: s_set_pc_i64 s[30:31]
+; GFX1250-LABEL: v_fptosi_v3bf16_to_v3i16:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_and_b32_e32 v2, 0xffff0000, v0
+; GFX1250-NEXT: v_dual_lshlrev_b32 v0, 16, v0 :: v_dual_lshlrev_b32 v1, 16, v1
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX1250-NEXT: v_cvt_pk_i16_f32 v0, v0, v2
+; GFX1250-NEXT: v_cvt_i32_f32_e32 v1, v1
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%op = fptosi <3 x bfloat> %x to <3 x i16>
ret <3 x i16> %op
}
@@ -35827,77 +35762,29 @@ define <4 x i16> @v_fptosi_v4bf16_to_v4i16(<4 x bfloat> %x) {
; GFX10-NEXT: v_perm_b32 v1, v1, v2, 0x5040100
; GFX10-NEXT: s_setpc_b64 s[30:31]
;
-; GFX11TRUE16-LABEL: v_fptosi_v4bf16_to_v4i16:
-; GFX11TRUE16: ; %bb.0:
-; GFX11TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX11TRUE16-NEXT: v_and_b32_e32 v2, 0xffff0000, v0
-; GFX11TRUE16-NEXT: v_and_b32_e32 v3, 0xffff0000, v1
-; GFX11TRUE16-NEXT: v_lshlrev_b32_e32 v1, 16, v1
-; GFX11TRUE16-NEXT: v_lshlrev_b32_e32 v0, 16, v0
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v3, v3
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX11TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX11TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX11TRUE16-NEXT: v_mov_b16_e32 v0.h, v2.l
-; GFX11TRUE16-NEXT: v_mov_b16_e32 v1.h, v3.l
-; GFX11TRUE16-NEXT: s_setpc_b64 s[30:31]
-;
-; GFX11FAKE16-LABEL: v_fptosi_v4bf16_to_v4i16:
-; GFX11FAKE16: ; %bb.0:
-; GFX11FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX11FAKE16-NEXT: v_lshlrev_b32_e32 v2, 16, v1
-; GFX11FAKE16-NEXT: v_lshlrev_b32_e32 v3, 16, v0
-; GFX11FAKE16-NEXT: v_and_b32_e32 v0, 0xffff0000, v0
-; GFX11FAKE16-NEXT: v_and_b32_e32 v1, 0xffff0000, v1
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v3, v3
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX11FAKE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX11FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11FAKE16-NEXT: v_perm_b32 v0, v0, v3, 0x5040100
-; GFX11FAKE16-NEXT: v_perm_b32 v1, v1, v2, 0x5040100
-; GFX11FAKE16-NEXT: s_setpc_b64 s[30:31]
-;
-; GFX1250TRUE16-LABEL: v_fptosi_v4bf16_to_v4i16:
-; GFX1250TRUE16: ; %bb.0:
-; GFX1250TRUE16-NEXT: s_wait_loadcnt_dscnt 0x0
-; GFX1250TRUE16-NEXT: s_wait_kmcnt 0x0
-; GFX1250TRUE16-NEXT: v_and_b32_e32 v2, 0xffff0000, v0
-; GFX1250TRUE16-NEXT: v_and_b32_e32 v3, 0xffff0000, v1
-; GFX1250TRUE16-NEXT: v_dual_lshlrev_b32 v1, 16, v1 :: v_dual_lshlrev_b32 v0, 16, v0
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3)
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v3, v3
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX1250TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX1250TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX1250TRUE16-NEXT: v_mov_b16_e32 v0.h, v2.l
-; GFX1250TRUE16-NEXT: v_mov_b16_e32 v1.h, v3.l
-; GFX1250TRUE16-NEXT: s_set_pc_i64 s[30:31]
+; GFX11-LABEL: v_fptosi_v4bf16_to_v4i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT: v_and_b32_e32 v2, 0xffff0000, v1
+; GFX11-NEXT: v_and_b32_e32 v3, 0xffff0000, v0
+; GFX11-NEXT: v_lshlrev_b32_e32 v0, 16, v0
+; GFX11-NEXT: v_lshlrev_b32_e32 v1, 16, v1
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_cvt_pk_i16_f32 v0, v0, v3
+; GFX11-NEXT: v_cvt_pk_i16_f32 v1, v1, v2
+; GFX11-NEXT: s_setpc_b64 s[30:31]
;
-; GFX1250FAKE16-LABEL: v_fptosi_v4bf16_to_v4i16:
-; GFX1250FAKE16: ; %bb.0:
-; GFX1250FAKE16-NEXT: s_wait_loadcnt_dscnt 0x0
-; GFX1250FAKE16-NEXT: s_wait_kmcnt 0x0
-; GFX1250FAKE16-NEXT: v_dual_lshlrev_b32 v2, 16, v1 :: v_dual_lshlrev_b32 v3, 16, v0
-; GFX1250FAKE16-NEXT: v_and_b32_e32 v0, 0xffff0000, v0
-; GFX1250FAKE16-NEXT: v_and_b32_e32 v1, 0xffff0000, v1
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v2, v2
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v3, v3
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
-; GFX1250FAKE16-NEXT: v_cvt_i32_f32_e32 v1, v1
-; GFX1250FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX1250FAKE16-NEXT: v_perm_b32 v0, v0, v3, 0x5040100
-; GFX1250FAKE16-NEXT: v_perm_b32 v1, v1, v2, 0x5040100
-; GFX1250FAKE16-NEXT: s_set_pc_i64 s[30:31]
+; GFX1250-LABEL: v_fptosi_v4bf16_to_v4i16:
+; GFX1250: ; %bb.0:
+; GFX1250-NEXT: s_wait_loadcnt_dscnt 0x0
+; GFX1250-NEXT: s_wait_kmcnt 0x0
+; GFX1250-NEXT: v_and_b32_e32 v2, 0xffff0000, v1
+; GFX1250-NEXT: v_and_b32_e32 v3, 0xffff0000, v0
+; GFX1250-NEXT: v_dual_lshlrev_b32 v0, 16, v0 :: v_dual_lshlrev_b32 v1, 16, v1
+; GFX1250-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX1250-NEXT: v_cvt_pk_i16_f32 v0, v0, v3
+; GFX1250-NEXT: v_cvt_pk_i16_f32 v1, v1, v2
+; GFX1250-NEXT: s_set_pc_i64 s[30:31]
%op = fptosi <4 x bfloat> %x to <4 x i16>
ret <4 x i16> %op
}
diff --git a/llvm/test/CodeGen/AMDGPU/fp_to_sint.ll b/llvm/test/CodeGen/AMDGPU/fp_to_sint.ll
index 0c5ed00b58d90..dea2dbe18a05d 100644
--- a/llvm/test/CodeGen/AMDGPU/fp_to_sint.ll
+++ b/llvm/test/CodeGen/AMDGPU/fp_to_sint.ll
@@ -1,6 +1,7 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=amdgcn < %s | FileCheck %s --check-prefixes=SI
; RUN: llc -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global < %s | FileCheck %s --check-prefixes=VI
+; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 < %s | FileCheck %s -check-prefixes=GFX11
; RUN: llc -mtriple=r600 -mcpu=redwood < %s | FileCheck %s --check-prefixes=EG
declare float @llvm.fabs.f32(float) #1
@@ -28,6 +29,17 @@ define amdgpu_kernel void @fp_to_sint_i32(ptr addrspace(1) %out, float %in) {
; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_i32:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_i32_f32_e32 v1, s2
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_i32:
; EG: ; %bb.0:
; EG-NEXT: ALU 3, @4, KC0[CB0:0-32], KC1[]
@@ -67,6 +79,17 @@ define amdgpu_kernel void @fp_to_sint_i32_fabs(ptr addrspace(1) %out, float %in)
; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_i32_fabs:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_i32_f32_e64 v1, |s2|
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_i32_fabs:
; EG: ; %bb.0:
; EG-NEXT: ALU 3, @4, KC0[CB0:0-32], KC1[]
@@ -108,6 +131,16 @@ define amdgpu_kernel void @fp_to_sint_v2i32(ptr addrspace(1) %out, <2 x float> %
; VI-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_v2i32:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v2, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_i32_f32_e32 v1, s3
+; GFX11-NEXT: v_cvt_i32_f32_e32 v0, s2
+; GFX11-NEXT: global_store_b64 v2, v[0:1], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_v2i32:
; EG: ; %bb.0:
; EG-NEXT: ALU 5, @4, KC0[CB0:0-32], KC1[]
@@ -157,6 +190,20 @@ define amdgpu_kernel void @fp_to_sint_v4i32(ptr addrspace(1) %out, ptr addrspace
; VI-NEXT: buffer_store_dwordx4 v[0:3], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_v4i32:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v4, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: s_load_b128 s[4:7], s[2:3], 0x0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_i32_f32_e32 v3, s7
+; GFX11-NEXT: v_cvt_i32_f32_e32 v2, s6
+; GFX11-NEXT: v_cvt_i32_f32_e32 v1, s5
+; GFX11-NEXT: v_cvt_i32_f32_e32 v0, s4
+; GFX11-NEXT: global_store_b128 v4, v[0:3], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_v4i32:
; EG: ; %bb.0:
; EG-NEXT: ALU 0, @8, KC0[CB0:0-32], KC1[]
@@ -234,6 +281,31 @@ define amdgpu_kernel void @fp_to_sint_i64 (ptr addrspace(1) %out, float %in) {
; VI-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_i64:
+; GFX11: ; %bb.0: ; %entry
+; GFX11-NEXT: s_load_b32 s0, s[4:5], 0x2c
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_trunc_f32_e32 v0, s0
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
+; GFX11-NEXT: v_ashrrev_i32_e32 v3, 31, v0
+; GFX11-NEXT: v_floor_f32_e32 v1, v1
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, v2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3)
+; GFX11-NEXT: v_xor_b32_e32 v1, v1, v3
+; GFX11-NEXT: v_mov_b32_e32 v2, 0
+; GFX11-NEXT: v_xor_b32_e32 v0, v0, v3
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v1, null, v1, v3, vcc_lo
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: global_store_b64 v2, v[0:1], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_i64:
; EG: ; %bb.0: ; %entry
; EG-NEXT: ALU 40, @4, KC0[CB0:0-32], KC1[]
@@ -357,6 +429,43 @@ define amdgpu_kernel void @fp_to_sint_v2i64(ptr addrspace(1) %out, <2 x float> %
; VI-NEXT: buffer_store_dwordx4 v[0:3], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_v2i64:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v6, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_trunc_f32_e32 v0, s3
+; GFX11-NEXT: v_trunc_f32_e32 v1, s2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v0|
+; GFX11-NEXT: v_mul_f32_e64 v3, 0x2f800000, |v1|
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_floor_f32_e32 v2, v2
+; GFX11-NEXT: v_floor_f32_e32 v3, v3
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_fma_f32 v4, 0xcf800000, v2, |v0|
+; GFX11-NEXT: v_fma_f32 v5, 0xcf800000, v3, |v1|
+; GFX11-NEXT: v_ashrrev_i32_e32 v0, 31, v0
+; GFX11-NEXT: v_ashrrev_i32_e32 v1, 31, v1
+; GFX11-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX11-NEXT: v_cvt_u32_f32_e32 v4, v4
+; GFX11-NEXT: v_cvt_u32_f32_e32 v5, v5
+; GFX11-NEXT: v_cvt_u32_f32_e32 v3, v3
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_xor_b32_e32 v7, v2, v0
+; GFX11-NEXT: v_xor_b32_e32 v4, v4, v0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_xor_b32_e32 v5, v5, v1
+; GFX11-NEXT: v_xor_b32_e32 v8, v3, v1
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-NEXT: v_sub_co_u32 v2, vcc_lo, v4, v0
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v3, null, v7, v0, vcc_lo
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-NEXT: v_sub_co_u32 v0, vcc_lo, v5, v1
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v1, null, v8, v1, vcc_lo
+; GFX11-NEXT: global_store_b128 v6, v[0:3], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_v2i64:
; EG: ; %bb.0:
; EG-NEXT: ALU 74, @4, KC0[CB0:0-32], KC1[]
@@ -559,6 +668,66 @@ define amdgpu_kernel void @fp_to_sint_v4i64(ptr addrspace(1) %out, <4 x float> %
; VI-NEXT: buffer_store_dwordx4 v[0:3], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_v4i64:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x34
+; GFX11-NEXT: s_load_b64 s[4:5], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v8, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_trunc_f32_e32 v0, s1
+; GFX11-NEXT: v_trunc_f32_e32 v2, s3
+; GFX11-NEXT: v_trunc_f32_e32 v3, s2
+; GFX11-NEXT: v_trunc_f32_e32 v1, s0
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_mul_f32_e64 v4, 0x2f800000, |v0|
+; GFX11-NEXT: v_mul_f32_e64 v7, 0x2f800000, |v2|
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_mul_f32_e64 v11, 0x2f800000, |v3|
+; GFX11-NEXT: v_mul_f32_e64 v6, 0x2f800000, |v1|
+; GFX11-NEXT: v_ashrrev_i32_e32 v5, 31, v0
+; GFX11-NEXT: v_floor_f32_e32 v4, v4
+; GFX11-NEXT: v_floor_f32_e32 v7, v7
+; GFX11-NEXT: v_floor_f32_e32 v11, v11
+; GFX11-NEXT: v_floor_f32_e32 v6, v6
+; GFX11-NEXT: v_ashrrev_i32_e32 v10, 31, v2
+; GFX11-NEXT: v_fma_f32 v0, 0xcf800000, v4, |v0|
+; GFX11-NEXT: v_fma_f32 v2, 0xcf800000, v7, |v2|
+; GFX11-NEXT: v_ashrrev_i32_e32 v12, 31, v3
+; GFX11-NEXT: v_fma_f32 v3, 0xcf800000, v11, |v3|
+; GFX11-NEXT: v_ashrrev_i32_e32 v9, 31, v1
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, v0
+; GFX11-NEXT: v_fma_f32 v1, 0xcf800000, v6, |v1|
+; GFX11-NEXT: v_cvt_u32_f32_e32 v13, v4
+; GFX11-NEXT: v_cvt_u32_f32_e32 v4, v6
+; GFX11-NEXT: v_cvt_u32_f32_e32 v6, v7
+; GFX11-NEXT: v_cvt_u32_f32_e32 v2, v2
+; GFX11-NEXT: v_cvt_u32_f32_e32 v3, v3
+; GFX11-NEXT: v_xor_b32_e32 v0, v0, v5
+; GFX11-NEXT: v_cvt_u32_f32_e32 v7, v11
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX11-NEXT: v_xor_b32_e32 v11, v13, v5
+; GFX11-NEXT: v_xor_b32_e32 v13, v4, v9
+; GFX11-NEXT: v_xor_b32_e32 v4, v6, v10
+; GFX11-NEXT: v_xor_b32_e32 v6, v2, v10
+; GFX11-NEXT: v_xor_b32_e32 v15, v3, v12
+; GFX11-NEXT: v_sub_co_u32 v2, vcc_lo, v0, v5
+; GFX11-NEXT: v_xor_b32_e32 v14, v7, v12
+; GFX11-NEXT: v_xor_b32_e32 v1, v1, v9
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v3, null, v11, v5, vcc_lo
+; GFX11-NEXT: v_sub_co_u32 v6, vcc_lo, v6, v10
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1)
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v7, null, v4, v10, vcc_lo
+; GFX11-NEXT: v_sub_co_u32 v4, vcc_lo, v15, v12
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v5, null, v14, v12, vcc_lo
+; GFX11-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v9
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_sub_co_ci_u32_e64 v1, null, v13, v9, vcc_lo
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: global_store_b128 v8, v[4:7], s[4:5] offset:16
+; GFX11-NEXT: global_store_b128 v8, v[0:3], s[4:5]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_v4i64:
; EG: ; %bb.0:
; EG-NEXT: ALU 99, @6, KC0[CB0:0-32], KC1[]
@@ -754,6 +923,19 @@ define amdgpu_kernel void @fp_to_uint_f32_to_i1(ptr addrspace(1) %out, float %in
; VI-NEXT: buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_f32_to_i1:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cmp_eq_f32_e64 s2, -1.0, s2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, 1, s2
+; GFX11-NEXT: global_store_b8 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_f32_to_i1:
; EG: ; %bb.0:
; EG-NEXT: ALU 12, @4, KC0[CB0:0-32], KC1[]
@@ -804,6 +986,19 @@ define amdgpu_kernel void @fp_to_uint_fabs_f32_to_i1(ptr addrspace(1) %out, floa
; VI-NEXT: buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_fabs_f32_to_i1:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cmp_eq_f32_e64 s2, -1.0, |s2|
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, 1, s2
+; GFX11-NEXT: global_store_b8 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_fabs_f32_to_i1:
; EG: ; %bb.0:
; EG-NEXT: ALU 12, @4, KC0[CB0:0-32], KC1[]
@@ -853,6 +1048,17 @@ define amdgpu_kernel void @fp_to_sint_f32_i16(ptr addrspace(1) %out, float %in)
; VI-NEXT: buffer_store_short v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_sint_f32_i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v1, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_i32_f32_e32 v0, s2
+; GFX11-NEXT: global_store_b16 v1, v0, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_sint_f32_i16:
; EG: ; %bb.0:
; EG-NEXT: ALU 13, @4, KC0[CB0:0-32], KC1[]
@@ -879,5 +1085,128 @@ define amdgpu_kernel void @fp_to_sint_f32_i16(ptr addrspace(1) %out, float %in)
ret void
}
+define amdgpu_kernel void @fp_to_sint_v2f32_to_v2i16(ptr addrspace(1) %out, <2 x float> %in) {
+; SI-LABEL: fp_to_sint_v2f32_to_v2i16:
+; SI: ; %bb.0:
+; SI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x9
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: s_mov_b64 s[4:5], s[2:3]
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_cvt_i32_f32_e32 v0, s5
+; SI-NEXT: v_cvt_i32_f32_e32 v1, s4
+; SI-NEXT: v_lshlrev_b32_e32 v0, 16, v0
+; SI-NEXT: v_and_b32_e32 v1, 0xffff, v1
+; SI-NEXT: v_or_b32_e32 v0, v1, v0
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_endpgm
+;
+; VI-LABEL: fp_to_sint_v2f32_to_v2i16:
+; VI: ; %bb.0:
+; VI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x24
+; VI-NEXT: s_waitcnt lgkmcnt(0)
+; VI-NEXT: v_mov_b32_e32 v0, s3
+; VI-NEXT: v_cvt_i32_f32_sdwa v0, v0 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD
+; VI-NEXT: v_cvt_i32_f32_e32 v1, s2
+; VI-NEXT: s_mov_b32 s3, 0xf000
+; VI-NEXT: s_mov_b32 s2, -1
+; VI-NEXT: v_or_b32_sdwa v0, v1, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:DWORD
+; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; VI-NEXT: s_endpgm
+;
+; GFX11-LABEL: fp_to_sint_v2f32_to_v2i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_pk_i16_f32 v1, s2, s3
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
+; EG-LABEL: fp_to_sint_v2f32_to_v2i16:
+; EG: ; %bb.0:
+; EG-NEXT: ALU 9, @4, KC0[CB0:0-32], KC1[]
+; EG-NEXT: MEM_RAT_CACHELESS STORE_RAW T4.X, T5.X, 1
+; EG-NEXT: CF_END
+; EG-NEXT: PAD
+; EG-NEXT: ALU clause starting at 4:
+; EG-NEXT: TRUNC T0.W, KC0[2].W,
+; EG-NEXT: TRUNC * T1.W, KC0[3].X,
+; EG-NEXT: FLT_TO_INT T1.W, PS,
+; EG-NEXT: FLT_TO_INT * T0.W, PV.W,
+; EG-NEXT: AND_INT T0.W, PS, literal.x,
+; EG-NEXT: LSHL * T1.W, PV.W, literal.y,
+; EG-NEXT: 65535(9.183409e-41), 16(2.242078e-44)
+; EG-NEXT: OR_INT T4.X, PV.W, PS,
+; EG-NEXT: LSHR * T5.X, KC0[2].Y, literal.x,
+; EG-NEXT: 2(2.802597e-45), 0(0.000000e+00)
+ %sint = fptosi <2 x float> %in to <2 x i16>
+ store <2 x i16> %sint, ptr addrspace(1) %out
+ ret void
+}
+
+define amdgpu_kernel void @fp_to_sint_f32_to_v2i16(ptr addrspace(1) %out, float %in0, float %in1) {
+; SI-LABEL: fp_to_sint_f32_to_v2i16:
+; SI: ; %bb.0:
+; SI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x9
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: s_mov_b64 s[4:5], s[2:3]
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_cvt_i32_f32_e32 v0, s4
+; SI-NEXT: v_cvt_i32_f32_e32 v1, s5
+; SI-NEXT: v_lshlrev_b32_e32 v1, 16, v1
+; SI-NEXT: v_and_b32_e32 v0, 0xffff, v0
+; SI-NEXT: v_or_b32_e32 v0, v0, v1
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_endpgm
+;
+; VI-LABEL: fp_to_sint_f32_to_v2i16:
+; VI: ; %bb.0:
+; VI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x24
+; VI-NEXT: s_waitcnt lgkmcnt(0)
+; VI-NEXT: v_mov_b32_e32 v1, s3
+; VI-NEXT: v_cvt_i32_f32_e32 v0, s2
+; VI-NEXT: v_cvt_i32_f32_sdwa v1, v1 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD
+; VI-NEXT: s_mov_b32 s3, 0xf000
+; VI-NEXT: s_mov_b32 s2, -1
+; VI-NEXT: v_or_b32_sdwa v0, v0, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:DWORD
+; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; VI-NEXT: s_endpgm
+;
+; GFX11-LABEL: fp_to_sint_f32_to_v2i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_pk_i16_f32 v1, s2, s3
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
+; EG-LABEL: fp_to_sint_f32_to_v2i16:
+; EG: ; %bb.0:
+; EG-NEXT: ALU 9, @4, KC0[CB0:0-32], KC1[]
+; EG-NEXT: MEM_RAT_CACHELESS STORE_RAW T4.X, T5.X, 1
+; EG-NEXT: CF_END
+; EG-NEXT: PAD
+; EG-NEXT: ALU clause starting at 4:
+; EG-NEXT: TRUNC T0.W, KC0[2].W,
+; EG-NEXT: TRUNC * T1.W, KC0[2].Z,
+; EG-NEXT: FLT_TO_INT T1.W, PS,
+; EG-NEXT: FLT_TO_INT * T0.W, PV.W,
+; EG-NEXT: LSHL T0.W, PS, literal.x,
+; EG-NEXT: AND_INT * T1.W, PV.W, literal.y,
+; EG-NEXT: 16(2.242078e-44), 65535(9.183409e-41)
+; EG-NEXT: OR_INT T4.X, PV.W, PS,
+; EG-NEXT: LSHR * T5.X, KC0[2].Y, literal.x,
+; EG-NEXT: 2(2.802597e-45), 0(0.000000e+00)
+ %sint0 = fptosi float %in0 to i16
+ %sint1 = fptosi float %in1 to i16
+ %res0 = insertelement <2 x i16> poison, i16 %sint0, i32 0
+ %res1 = insertelement <2 x i16> %res0, i16 %sint1, i32 1
+ store <2 x i16> %res1, ptr addrspace(1) %out
+ ret void
+}
+
attributes #0 = { nounwind }
attributes #1 = { nounwind readnone }
diff --git a/llvm/test/CodeGen/AMDGPU/fp_to_uint.ll b/llvm/test/CodeGen/AMDGPU/fp_to_uint.ll
index c938475ab7675..c4fd321d027c7 100644
--- a/llvm/test/CodeGen/AMDGPU/fp_to_uint.ll
+++ b/llvm/test/CodeGen/AMDGPU/fp_to_uint.ll
@@ -1,6 +1,7 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=amdgcn < %s | FileCheck %s -check-prefixes=SI
; RUN: llc -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global < %s | FileCheck %s -check-prefixes=VI
+; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 < %s | FileCheck %s -check-prefixes=GFX11
; RUN: llc -mtriple=r600 -mcpu=redwood < %s | FileCheck %s -check-prefixes=EG
declare float @llvm.fabs.f32(float) #1
@@ -28,6 +29,17 @@ define amdgpu_kernel void @fp_to_uint_f32_to_i32 (ptr addrspace(1) %out, float %
; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_f32_to_i32:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, s2
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_f32_to_i32:
; EG: ; %bb.0:
; EG-NEXT: ALU 3, @4, KC0[CB0:0-32], KC1[]
@@ -68,6 +80,16 @@ define amdgpu_kernel void @fp_to_uint_v2f32_to_v2i32(ptr addrspace(1) %out, <2 x
; VI-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_v2f32_to_v2i32:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v2, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, s3
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, s2
+; GFX11-NEXT: global_store_b64 v2, v[0:1], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_v2f32_to_v2i32:
; EG: ; %bb.0:
; EG-NEXT: ALU 5, @4, KC0[CB0:0-32], KC1[]
@@ -117,6 +139,20 @@ define amdgpu_kernel void @fp_to_uint_v4f32_to_v4i32(ptr addrspace(1) %out, ptr
; VI-NEXT: buffer_store_dwordx4 v[0:3], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_v4f32_to_v4i32:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v4, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: s_load_b128 s[4:7], s[2:3], 0x0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_u32_f32_e32 v3, s7
+; GFX11-NEXT: v_cvt_u32_f32_e32 v2, s6
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, s5
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, s4
+; GFX11-NEXT: global_store_b128 v4, v[0:3], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_v4f32_to_v4i32:
; EG: ; %bb.0:
; EG-NEXT: ALU 0, @8, KC0[CB0:0-32], KC1[]
@@ -181,6 +217,24 @@ define amdgpu_kernel void @fp_to_uint_f32_to_i64(ptr addrspace(1) %out, float %x
; VI-NEXT: buffer_store_dwordx2 v[0:1], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_f32_to_i64:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b32 s0, s[4:5], 0x2c
+; GFX11-NEXT: v_mov_b32_e32 v2, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_trunc_f32_e32 v0, s0
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
+; GFX11-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0
+; GFX11-NEXT: v_floor_f32_e32 v1, v1
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_fmamk_f32 v0, v1, 0xcf800000, v0
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, v1
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, v0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: global_store_b64 v2, v[0:1], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_f32_to_i64:
; EG: ; %bb.0:
; EG-NEXT: ALU 40, @4, KC0[CB0:0-32], KC1[]
@@ -280,6 +334,30 @@ define amdgpu_kernel void @fp_to_uint_v2f32_to_v2i64(ptr addrspace(1) %out, <2 x
; VI-NEXT: buffer_store_dwordx4 v[0:3], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_v2f32_to_v2i64:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v5, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_trunc_f32_e32 v0, s3
+; GFX11-NEXT: v_trunc_f32_e32 v4, s2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0
+; GFX11-NEXT: v_mul_f32_e32 v2, 0x2f800000, v4
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_floor_f32_e32 v1, v1
+; GFX11-NEXT: v_floor_f32_e32 v2, v2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT: v_fmac_f32_e32 v0, 0xcf800000, v1
+; GFX11-NEXT: v_fmac_f32_e32 v4, 0xcf800000, v2
+; GFX11-NEXT: v_cvt_u32_f32_e32 v3, v1
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_cvt_u32_f32_e32 v2, v0
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, v4
+; GFX11-NEXT: global_store_b128 v5, v[0:3], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_v2f32_to_v2i64:
; EG: ; %bb.0:
; EG-NEXT: ALU 74, @4, KC0[CB0:0-32], KC1[]
@@ -440,6 +518,45 @@ define amdgpu_kernel void @fp_to_uint_v4f32_to_v4i64(ptr addrspace(1) %out, <4 x
; VI-NEXT: buffer_store_dwordx4 v[0:3], off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_v4f32_to_v4i64:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x34
+; GFX11-NEXT: v_mov_b32_e32 v10, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_trunc_f32_e32 v0, s1
+; GFX11-NEXT: v_trunc_f32_e32 v8, s0
+; GFX11-NEXT: v_trunc_f32_e32 v4, s3
+; GFX11-NEXT: v_trunc_f32_e32 v9, s2
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0
+; GFX11-NEXT: v_mul_f32_e32 v2, 0x2f800000, v8
+; GFX11-NEXT: v_mul_f32_e32 v3, 0x2f800000, v4
+; GFX11-NEXT: v_mul_f32_e32 v5, 0x2f800000, v9
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_floor_f32_e32 v6, v1
+; GFX11-NEXT: v_floor_f32_e32 v2, v2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_floor_f32_e32 v7, v3
+; GFX11-NEXT: v_floor_f32_e32 v5, v5
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_fmac_f32_e32 v0, 0xcf800000, v6
+; GFX11-NEXT: v_fmac_f32_e32 v8, 0xcf800000, v2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_3) | instid1(VALU_DEP_4)
+; GFX11-NEXT: v_dual_fmac_f32 v4, 0xcf800000, v7 :: v_dual_fmac_f32 v9, 0xcf800000, v5
+; GFX11-NEXT: v_cvt_u32_f32_e32 v3, v6
+; GFX11-NEXT: v_cvt_u32_f32_e32 v7, v7
+; GFX11-NEXT: v_cvt_u32_f32_e32 v5, v5
+; GFX11-NEXT: v_cvt_u32_f32_e32 v6, v4
+; GFX11-NEXT: v_cvt_u32_f32_e32 v4, v9
+; GFX11-NEXT: v_cvt_u32_f32_e32 v1, v2
+; GFX11-NEXT: v_cvt_u32_f32_e32 v2, v0
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, v8
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: global_store_b128 v10, v[4:7], s[0:1] offset:16
+; GFX11-NEXT: global_store_b128 v10, v[0:3], s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_v4f32_to_v4i64:
; EG: ; %bb.0:
; EG-NEXT: ALU 99, @6, KC0[CB0:0-32], KC1[]
@@ -635,6 +752,19 @@ define amdgpu_kernel void @fp_to_uint_f32_to_i1(ptr addrspace(1) %out, float %in
; VI-NEXT: buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_f32_to_i1:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cmp_eq_f32_e64 s2, 1.0, s2
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, 1, s2
+; GFX11-NEXT: global_store_b8 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_f32_to_i1:
; EG: ; %bb.0:
; EG-NEXT: ALU 12, @4, KC0[CB0:0-32], KC1[]
@@ -685,6 +815,19 @@ define amdgpu_kernel void @fp_to_uint_fabs_f32_to_i1(ptr addrspace(1) %out, floa
; VI-NEXT: buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_fabs_f32_to_i1:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cmp_eq_f32_e64 s2, 1.0, |s2|
+; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
+; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, 1, s2
+; GFX11-NEXT: global_store_b8 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_fabs_f32_to_i1:
; EG: ; %bb.0:
; EG-NEXT: ALU 12, @4, KC0[CB0:0-32], KC1[]
@@ -734,6 +877,17 @@ define amdgpu_kernel void @fp_to_uint_f32_to_i16(ptr addrspace(1) %out, float %i
; VI-NEXT: buffer_store_short v0, off, s[0:3], 0
; VI-NEXT: s_endpgm
;
+; GFX11-LABEL: fp_to_uint_f32_to_i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_clause 0x1
+; GFX11-NEXT: s_load_b32 s2, s[4:5], 0x2c
+; GFX11-NEXT: s_load_b64 s[0:1], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v1, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_u32_f32_e32 v0, s2
+; GFX11-NEXT: global_store_b16 v1, v0, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
; EG-LABEL: fp_to_uint_f32_to_i16:
; EG: ; %bb.0:
; EG-NEXT: ALU 12, @4, KC0[CB0:0-32], KC1[]
@@ -759,5 +913,124 @@ define amdgpu_kernel void @fp_to_uint_f32_to_i16(ptr addrspace(1) %out, float %i
ret void
}
+define amdgpu_kernel void @fp_to_uint_v2f32_to_v2i16(ptr addrspace(1) %out, <2 x float> %in) {
+; SI-LABEL: fp_to_uint_v2f32_to_v2i16:
+; SI: ; %bb.0:
+; SI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x9
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: s_mov_b64 s[4:5], s[2:3]
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_cvt_u32_f32_e32 v0, s5
+; SI-NEXT: v_lshlrev_b32_e32 v0, 16, v0
+; SI-NEXT: v_cvt_u32_f32_e32 v1, s4
+; SI-NEXT: v_or_b32_e32 v0, v1, v0
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_endpgm
+;
+; VI-LABEL: fp_to_uint_v2f32_to_v2i16:
+; VI: ; %bb.0:
+; VI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x24
+; VI-NEXT: s_waitcnt lgkmcnt(0)
+; VI-NEXT: v_mov_b32_e32 v0, s3
+; VI-NEXT: v_cvt_u32_f32_sdwa v0, v0 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD
+; VI-NEXT: v_cvt_u32_f32_e32 v1, s2
+; VI-NEXT: s_mov_b32 s3, 0xf000
+; VI-NEXT: s_mov_b32 s2, -1
+; VI-NEXT: v_or_b32_sdwa v0, v1, v0 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:DWORD
+; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; VI-NEXT: s_endpgm
+;
+; GFX11-LABEL: fp_to_uint_v2f32_to_v2i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_pk_u16_f32 v1, s2, s3
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
+; EG-LABEL: fp_to_uint_v2f32_to_v2i16:
+; EG: ; %bb.0:
+; EG-NEXT: ALU 8, @4, KC0[CB0:0-32], KC1[]
+; EG-NEXT: MEM_RAT_CACHELESS STORE_RAW T4.X, T5.X, 1
+; EG-NEXT: CF_END
+; EG-NEXT: PAD
+; EG-NEXT: ALU clause starting at 4:
+; EG-NEXT: TRUNC T0.W, KC0[3].X,
+; EG-NEXT: TRUNC * T1.W, KC0[2].W,
+; EG-NEXT: FLT_TO_UINT * T0.Y, PV.W,
+; EG-NEXT: LSHL T0.W, PS, literal.x,
+; EG-NEXT: FLT_TO_UINT * T0.Y, T1.W,
+; EG-NEXT: 16(2.242078e-44), 0(0.000000e+00)
+; EG-NEXT: OR_INT T4.X, PS, PV.W,
+; EG-NEXT: LSHR * T5.X, KC0[2].Y, literal.x,
+; EG-NEXT: 2(2.802597e-45), 0(0.000000e+00)
+ %uint = fptoui <2 x float> %in to <2 x i16>
+ store <2 x i16> %uint, ptr addrspace(1) %out
+ ret void
+}
+
+define amdgpu_kernel void @fp_to_uint_f32_to_v2i16(ptr addrspace(1) %out, float %in0, float %in1) {
+; SI-LABEL: fp_to_uint_f32_to_v2i16:
+; SI: ; %bb.0:
+; SI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x9
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: s_mov_b64 s[4:5], s[2:3]
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_cvt_u32_f32_e32 v0, s4
+; SI-NEXT: v_cvt_u32_f32_e32 v1, s5
+; SI-NEXT: v_lshlrev_b32_e32 v1, 16, v1
+; SI-NEXT: v_or_b32_e32 v0, v0, v1
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_endpgm
+;
+; VI-LABEL: fp_to_uint_f32_to_v2i16:
+; VI: ; %bb.0:
+; VI-NEXT: s_load_dwordx4 s[0:3], s[4:5], 0x24
+; VI-NEXT: s_waitcnt lgkmcnt(0)
+; VI-NEXT: v_mov_b32_e32 v1, s3
+; VI-NEXT: v_cvt_u32_f32_e32 v0, s2
+; VI-NEXT: v_cvt_u32_f32_sdwa v1, v1 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD
+; VI-NEXT: s_mov_b32 s3, 0xf000
+; VI-NEXT: s_mov_b32 s2, -1
+; VI-NEXT: v_or_b32_sdwa v0, v0, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:DWORD
+; VI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; VI-NEXT: s_endpgm
+;
+; GFX11-LABEL: fp_to_uint_f32_to_v2i16:
+; GFX11: ; %bb.0:
+; GFX11-NEXT: s_load_b128 s[0:3], s[4:5], 0x24
+; GFX11-NEXT: v_mov_b32_e32 v0, 0
+; GFX11-NEXT: s_waitcnt lgkmcnt(0)
+; GFX11-NEXT: v_cvt_pk_u16_f32 v1, s2, s3
+; GFX11-NEXT: global_store_b32 v0, v1, s[0:1]
+; GFX11-NEXT: s_endpgm
+;
+; EG-LABEL: fp_to_uint_f32_to_v2i16:
+; EG: ; %bb.0:
+; EG-NEXT: ALU 8, @4, KC0[CB0:0-32], KC1[]
+; EG-NEXT: MEM_RAT_CACHELESS STORE_RAW T4.X, T5.X, 1
+; EG-NEXT: CF_END
+; EG-NEXT: PAD
+; EG-NEXT: ALU clause starting at 4:
+; EG-NEXT: TRUNC T0.W, KC0[2].W,
+; EG-NEXT: TRUNC * T1.W, KC0[2].Z,
+; EG-NEXT: FLT_TO_UINT * T0.Y, PV.W,
+; EG-NEXT: LSHL T0.W, PS, literal.x,
+; EG-NEXT: FLT_TO_UINT * T0.Y, T1.W,
+; EG-NEXT: 16(2.242078e-44), 0(0.000000e+00)
+; EG-NEXT: OR_INT T4.X, PV.W, PS,
+; EG-NEXT: LSHR * T5.X, KC0[2].Y, literal.x,
+; EG-NEXT: 2(2.802597e-45), 0(0.000000e+00)
+ %uint0 = fptoui float %in0 to i16
+ %uint1 = fptoui float %in1 to i16
+ %res0 = insertelement <2 x i16> poison, i16 %uint0, i32 0
+ %res1 = insertelement <2 x i16> %res0, i16 %uint1, i32 1
+ store <2 x i16> %res1, ptr addrspace(1) %out
+ ret void
+}
+
attributes #0 = { nounwind }
attributes #1 = { nounwind readnone }
diff --git a/llvm/test/CodeGen/AMDGPU/scalar-float-sop1.ll b/llvm/test/CodeGen/AMDGPU/scalar-float-sop1.ll
index 689e918c61425..09686e2fa819a 100644
--- a/llvm/test/CodeGen/AMDGPU/scalar-float-sop1.ll
+++ b/llvm/test/CodeGen/AMDGPU/scalar-float-sop1.ll
@@ -1,8 +1,8 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1150 < %s | FileCheck %s
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1150 -global-isel < %s | FileCheck %s
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck %s
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -global-isel < %s | FileCheck %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1150 < %s | FileCheck %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1150 < %s | FileCheck %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck %s
define amdgpu_vs float @sitofp_i32_to_f32(i32 inreg %val) {
; CHECK-LABEL: sitofp_i32_to_f32:
@@ -44,6 +44,24 @@ define amdgpu_vs i32 @fptoui_f32_to_u32(float inreg %val) {
ret i32 %res
}
+define amdgpu_vs i16 @fptosi_f32_to_i16(float inreg %val) {
+; CHECK-LABEL: fptosi_f32_to_i16:
+; CHECK: ; %bb.0:
+; CHECK-NEXT: s_cvt_i32_f32 s0, s0
+; CHECK-NEXT: ; return to shader part epilog
+ %res = fptosi float %val to i16
+ ret i16 %res
+}
+
+define amdgpu_vs i16 @fptoui_f32_to_u16(float inreg %val) {
+; CHECK-LABEL: fptoui_f32_to_u16:
+; CHECK: ; %bb.0:
+; CHECK-NEXT: s_cvt_u32_f32 s0, s0
+; CHECK-NEXT: ; return to shader part epilog
+ %res = fptoui float %val to i16
+ ret i16 %res
+}
+
define amdgpu_vs float @fpext_f16_to_f32(half inreg %val) {
; CHECK-LABEL: fpext_f16_to_f32:
; CHECK: ; %bb.0:
More information about the llvm-commits
mailing list