[llvm] [RISCV][VLOPT] Add support for vrgather (PR #148249)
Mikhail R. Gadelha via llvm-commits
llvm-commits at lists.llvm.org
Tue Jul 15 07:48:52 PDT 2025
https://github.com/mikhailramalho updated https://github.com/llvm/llvm-project/pull/148249
>From 32f3dfa376afef71980934116449f51ea079635d Mon Sep 17 00:00:00 2001
From: "Mikhail R. Gadelha" <mikhail at igalia.com>
Date: Thu, 10 Jul 2025 17:15:01 -0300
Subject: [PATCH 1/4] [RISCV][VLOPT] Add support for vrgather
Signed-off-by: Mikhail R. Gadelha <mikhail at igalia.com>
---
llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
index 2d9f38221d424..79d7a21f19de2 100644
--- a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
+++ b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
@@ -747,6 +747,15 @@ getOperandLog2EEW(const MachineOperand &MO, const MachineRegisterInfo *MRI) {
return TwoTimes ? MILog2SEW + 1 : MILog2SEW;
}
+ // Vector Register Gather with 16-bit Index Elements Instruction
+ // vrgatherei16.vv vd, vs2, vs1
+ // Dest and source data EEW=SEW. Index vector EEW=16.
+ case RISCV::VRGATHEREI16_VV: {
+ if (MO.getOperandNo() == 0 || MO.getOperandNo() == 1)
+ return MILog2SEW;
+ return 4; // vs1 index vector has EEW=16, log2(16)=4
+ }
+
default:
return std::nullopt;
}
@@ -1051,6 +1060,11 @@ static bool isSupportedInstr(const MachineInstr &MI) {
case RISCV::VSLIDEDOWN_VI:
case RISCV::VSLIDE1UP_VX:
case RISCV::VFSLIDE1UP_VF:
+ // Vector Register Gather Instructions
+ case RISCV::VRGATHER_VI:
+ case RISCV::VRGATHER_VV:
+ case RISCV::VRGATHER_VX:
+ case RISCV::VRGATHEREI16_VV:
// Vector Single-Width Floating-Point Add/Subtract Instructions
case RISCV::VFADD_VF:
case RISCV::VFADD_VV:
>From ea20a7882cd5cb1ae770e7c583315fc903fdc9a5 Mon Sep 17 00:00:00 2001
From: "Mikhail R. Gadelha" <mikhail at igalia.com>
Date: Thu, 10 Jul 2025 17:15:11 -0300
Subject: [PATCH 2/4] Update test
Signed-off-by: Mikhail R. Gadelha <mikhail at igalia.com>
---
llvm/test/CodeGen/RISCV/rvv/vl-opt-instrs.ll | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vl-opt-instrs.ll b/llvm/test/CodeGen/RISCV/rvv/vl-opt-instrs.ll
index 317ad0c124e73..a8eba6d3db256 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vl-opt-instrs.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vl-opt-instrs.ll
@@ -5476,9 +5476,8 @@ define <vscale x 4 x i32> @vrgather_vi(<vscale x 4 x i32> %a, <vscale x 4 x i32>
;
; VLOPT-LABEL: vrgather_vi:
; VLOPT: # %bb.0:
-; VLOPT-NEXT: vsetvli a1, zero, e32, m2, ta, ma
-; VLOPT-NEXT: vrgather.vi v12, v8, 5
; VLOPT-NEXT: vsetvli zero, a0, e32, m2, ta, ma
+; VLOPT-NEXT: vrgather.vi v12, v8, 5
; VLOPT-NEXT: vadd.vv v8, v12, v10
; VLOPT-NEXT: ret
%1 = call <vscale x 4 x i32> @llvm.riscv.vrgather.vx.nxv4i32.iXLen(<vscale x 4 x i32> poison, <vscale x 4 x i32> %a, iXLen 5, iXLen -1)
@@ -5497,9 +5496,8 @@ define <vscale x 4 x i32> @vrgather_vv(<vscale x 4 x i32> %a, <vscale x 4 x i32>
;
; VLOPT-LABEL: vrgather_vv:
; VLOPT: # %bb.0:
-; VLOPT-NEXT: vsetvli a1, zero, e32, m2, ta, ma
-; VLOPT-NEXT: vrgather.vv v12, v8, v10
; VLOPT-NEXT: vsetvli zero, a0, e32, m2, ta, ma
+; VLOPT-NEXT: vrgather.vv v12, v8, v10
; VLOPT-NEXT: vadd.vv v8, v12, v8
; VLOPT-NEXT: ret
%1 = call <vscale x 4 x i32> @llvm.riscv.vrgather.vv.nxv4i32(<vscale x 4 x i32> poison, <vscale x 4 x i32> %a, <vscale x 4 x i32> %idx, iXLen -1)
@@ -5518,9 +5516,8 @@ define <vscale x 4 x i32> @vrgather_vx(<vscale x 4 x i32> %a, iXLen %idx, <vscal
;
; VLOPT-LABEL: vrgather_vx:
; VLOPT: # %bb.0:
-; VLOPT-NEXT: vsetvli a2, zero, e32, m2, ta, ma
-; VLOPT-NEXT: vrgather.vx v12, v8, a0
; VLOPT-NEXT: vsetvli zero, a1, e32, m2, ta, ma
+; VLOPT-NEXT: vrgather.vx v12, v8, a0
; VLOPT-NEXT: vadd.vv v8, v12, v10
; VLOPT-NEXT: ret
%1 = call <vscale x 4 x i32> @llvm.riscv.vrgather.vx.nxv4i32.iXLen(<vscale x 4 x i32> poison, <vscale x 4 x i32> %a, iXLen %idx, iXLen -1)
@@ -5539,9 +5536,8 @@ define <vscale x 4 x i32> @vrgatherei16_vv(<vscale x 4 x i32> %a, <vscale x 4 x
;
; VLOPT-LABEL: vrgatherei16_vv:
; VLOPT: # %bb.0:
-; VLOPT-NEXT: vsetvli a1, zero, e32, m2, ta, ma
-; VLOPT-NEXT: vrgatherei16.vv v12, v8, v10
; VLOPT-NEXT: vsetvli zero, a0, e32, m2, ta, ma
+; VLOPT-NEXT: vrgatherei16.vv v12, v8, v10
; VLOPT-NEXT: vadd.vv v8, v12, v8
; VLOPT-NEXT: ret
%1 = call <vscale x 4 x i32> @llvm.riscv.vrgatherei16.vv.nxv4i32(<vscale x 4 x i32> poison, <vscale x 4 x i32> %a, <vscale x 4 x i16> %idx, iXLen -1)
>From 768189d11dd0252a4024e2a86a25325f26a48fb9 Mon Sep 17 00:00:00 2001
From: "Mikhail R. Gadelha" <mikhail at igalia.com>
Date: Fri, 11 Jul 2025 11:28:24 -0300
Subject: [PATCH 3/4] Swap condition
Signed-off-by: Mikhail R. Gadelha <mikhail at igalia.com>
---
llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
index 79d7a21f19de2..e940ab6c62637 100644
--- a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
+++ b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
@@ -748,12 +748,11 @@ getOperandLog2EEW(const MachineOperand &MO, const MachineRegisterInfo *MRI) {
}
// Vector Register Gather with 16-bit Index Elements Instruction
- // vrgatherei16.vv vd, vs2, vs1
// Dest and source data EEW=SEW. Index vector EEW=16.
case RISCV::VRGATHEREI16_VV: {
- if (MO.getOperandNo() == 0 || MO.getOperandNo() == 1)
- return MILog2SEW;
- return 4; // vs1 index vector has EEW=16, log2(16)=4
+ if (MO.getOperandNo() == 2)
+ return 4;
+ return MILog2SEW;
}
default:
>From eed2358008d8dff6d0e5cff7c0957c6c4fbfff42 Mon Sep 17 00:00:00 2001
From: "Mikhail R. Gadelha" <mikhail at igalia.com>
Date: Tue, 15 Jul 2025 11:48:25 -0300
Subject: [PATCH 4/4] Added tests
Signed-off-by: Mikhail R. Gadelha <mikhail at igalia.com>
---
.../test/CodeGen/RISCV/rvv/vl-opt-op-info.mir | 59 +++++++++++++++++++
1 file changed, 59 insertions(+)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir b/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir
index b39ba422bd349..2b32066ebc01a 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir
+++ b/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir
@@ -1801,4 +1801,63 @@ body: |
; CHECK-NEXT: %y:vr = PseudoVMAND_MM_B16 $noreg, %x, 1, 0 /* e8 */
%x:vr = PseudoVMSET_M_B8 -1, 0
%y:vr = PseudoVMAND_MM_B16 $noreg, %x, 1, 0
+...
+---
+name: vrgatherei16_vv
+body: |
+ bb.0:
+ ; CHECK-LABEL: name: vrgatherei16_vv
+ ; CHECK: early-clobber %x:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, $noreg, 1, 5 /* e32 */, 0 /* tu, mu */
+ ; CHECK-NEXT: %y:vr = PseudoVADD_VV_M1 $noreg, %x, $noreg, 1, 5 /* e32 */, 0 /* tu, mu */
+ %x:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, $noreg, -1, 5 /* e32 */, 0
+ %y:vr = PseudoVADD_VV_M1 $noreg, %x, $noreg, 1, 5 /* e32 */, 0
+...
+---
+name: vrgatherei16_vv_incompatible_data_eew
+body: |
+ bb.0:
+ ; CHECK-LABEL: name: vrgatherei16_vv_incompatible_data_eew
+ ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0 /* tu, mu */
+ ; CHECK-NEXT: early-clobber %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, %x, $noreg, 1, 5 /* e32 */, 0 /* tu, mu */
+ %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0
+ %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, %x, $noreg, 1, 5 /* e32 */, 0
+...
---
+name: vrgatherei16_vv_compatible_index_eew
+body: |
+ bb.0:
+ ; CHECK-LABEL: name: vrgatherei16_vv_compatible_index_eew
+ ; CHECK: %x:vr = PseudoVADD_VV_MF2 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0 /* tu, mu */
+ ; CHECK-NEXT: early-clobber %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, %x, 1, 5 /* e32 */, 0 /* tu, mu */
+ %x:vr = PseudoVADD_VV_MF2 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0
+ %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, %x, 1, 5 /* e32 */, 0
+...
+---
+name: vrgatherei16_vv_incompatible_dest_emul
+body: |
+ bb.0:
+ ; CHECK-LABEL: name: vrgatherei16_vv_incompatible_dest_emul
+ ; CHECK: early-clobber %x:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, $noreg, -1, 5 /* e32 */, 0 /* tu, mu */
+ ; CHECK-NEXT: %y:vr = PseudoVADD_VV_MF2 $noreg, %x, $noreg, 1, 5 /* e32 */, 0 /* tu, mu */
+ %x:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, $noreg, -1, 5 /* e32 */, 0
+ %y:vr = PseudoVADD_VV_MF2 $noreg, %x, $noreg, 1, 5 /* e32 */, 0
+...
+---
+name: vrgatherei16_vv_incompatible_source_emul
+body: |
+ bb.0:
+ ; CHECK-LABEL: name: vrgatherei16_vv_incompatible_source_emul
+ ; CHECK: %x:vr = PseudoVADD_VV_MF2 $noreg, $noreg, $noreg, -1, 5 /* e32 */, 0 /* tu, mu */
+ ; CHECK-NEXT: early-clobber %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, %x, $noreg, 1, 5 /* e32 */, 0 /* tu, mu */
+ %x:vr = PseudoVADD_VV_MF2 $noreg, $noreg, $noreg, -1, 5 /* e32 */, 0
+ %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, %x, $noreg, 1, 5 /* e32 */, 0
+...
+---
+name: vrgatherei16_vv_incompatible_index_emul
+body: |
+ bb.0:
+ ; CHECK-LABEL: name: vrgatherei16_vv_incompatible_index_emul
+ ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0 /* tu, mu */
+ ; CHECK-NEXT: early-clobber %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, %x, 1, 5 /* e32 */, 0 /* tu, mu */
+ %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0
+ %y:vr = PseudoVRGATHEREI16_VV_M1_E32_MF2 $noreg, $noreg, %x, 1, 5 /* e32 */, 0
More information about the llvm-commits
mailing list