[llvm] [RISCV][GISel] Add initial support for rvv intrinsics (PR #156415)
via llvm-commits
llvm-commits at lists.llvm.org
Sun Sep 7 23:33:19 PDT 2025
llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-llvm-globalisel
Author: Jianjian Guan (jacquesguan)
<details>
<summary>Changes</summary>
This pr removes the falling back to SDISel of rvv intrinsics and marks them legalized in the legalize pass. Another pr would be created for regbankselect pass to make vf form intriniscs have the right scalar register bank.
---
Patch is 29.90 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/156415.diff
4 Files Affected:
- (modified) llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp (+4)
- (modified) llvm/lib/Target/RISCV/RISCVISelLowering.cpp (+7)
- (modified) llvm/test/CodeGen/RISCV/GlobalISel/irtranslator/fallback.ll (-33)
- (added) llvm/test/CodeGen/RISCV/GlobalISel/rvv/vfadd.ll (+764)
``````````diff
diff --git a/llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp b/llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp
index ab5c9e17b9a37..64e71e2a968a7 100644
--- a/llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp
+++ b/llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp
@@ -714,6 +714,10 @@ RISCVLegalizerInfo::RISCVLegalizerInfo(const RISCVSubtarget &ST)
bool RISCVLegalizerInfo::legalizeIntrinsic(LegalizerHelper &Helper,
MachineInstr &MI) const {
Intrinsic::ID IntrinsicID = cast<GIntrinsic>(MI).getIntrinsicID();
+
+ if (RISCVVIntrinsicsTable::getRISCVVIntrinsicInfo(IntrinsicID))
+ return true;
+
switch (IntrinsicID) {
default:
return false;
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index 3ab08f990c289..2aea09891d51e 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -38,6 +38,7 @@
#include "llvm/IR/DiagnosticPrinter.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/IR/Instructions.h"
+#include "llvm/IR/IntrinsicInst.h"
#include "llvm/IR/IntrinsicsRISCV.h"
#include "llvm/MC/MCCodeEmitter.h"
#include "llvm/MC/MCInstBuilder.h"
@@ -24880,6 +24881,12 @@ bool RISCVTargetLowering::fallBackToDAGISel(const Instruction &Inst) const {
Op == Instruction::Freeze || Op == Instruction::Store)
return false;
+ if (auto *II = dyn_cast<IntrinsicInst>(&Inst)) {
+ // Mark RVV intrinsic as supported.
+ if (RISCVVIntrinsicsTable::getRISCVVIntrinsicInfo(II->getIntrinsicID()))
+ return false;
+ }
+
if (Inst.getType()->isScalableTy())
return true;
diff --git a/llvm/test/CodeGen/RISCV/GlobalISel/irtranslator/fallback.ll b/llvm/test/CodeGen/RISCV/GlobalISel/irtranslator/fallback.ll
index 49276c9416234..8e43e044b7ee5 100644
--- a/llvm/test/CodeGen/RISCV/GlobalISel/irtranslator/fallback.ll
+++ b/llvm/test/CodeGen/RISCV/GlobalISel/irtranslator/fallback.ll
@@ -2,39 +2,6 @@
; RUN: FileCheck %s --check-prefix=FALLBACK-WITH-REPORT-OUT < %t.out
; RUN: FileCheck %s --check-prefix=FALLBACK-WITH-REPORT-ERR < %t.err
-
-declare <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
- <vscale x 1 x i8>,
- <vscale x 1 x i8>,
- <vscale x 1 x i8>,
- i64)
-
-; FALLBACK_WITH_REPORT_ERR: <unknown>:0:0: unable to translate instruction: call
-; FALLBACK-WITH-REPORT-OUT-LABEL: scalable_arg
-define <vscale x 1 x i8> @scalable_arg(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i64 %2) nounwind {
-entry:
- %a = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
- <vscale x 1 x i8> undef,
- <vscale x 1 x i8> %0,
- <vscale x 1 x i8> %1,
- i64 %2)
-
- ret <vscale x 1 x i8> %a
-}
-
-; FALLBACK-WITH-REPORT-ERR: remark: <unknown>:0:0: unable to translate instruction: call
-; FALLBACK-WITH-REPORT-OUT-LABEL: scalable_inst
-define <vscale x 1 x i8> @scalable_inst(i64 %0) nounwind {
-entry:
- %a = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
- <vscale x 1 x i8> undef,
- <vscale x 1 x i8> undef,
- <vscale x 1 x i8> undef,
- i64 %0)
-
- ret <vscale x 1 x i8> %a
-}
-
; FALLBACK-WITH-REPORT-ERR: remark: <unknown>:0:0: unable to translate instruction: alloca:
; FALLBACK-WITH-REPORT-OUT-LABEL: scalable_alloca
define void @scalable_alloca() #1 {
diff --git a/llvm/test/CodeGen/RISCV/GlobalISel/rvv/vfadd.ll b/llvm/test/CodeGen/RISCV/GlobalISel/rvv/vfadd.ll
new file mode 100644
index 0000000000000..f4b46b6b5857a
--- /dev/null
+++ b/llvm/test/CodeGen/RISCV/GlobalISel/rvv/vfadd.ll
@@ -0,0 +1,764 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zvfh \
+; RUN: -verify-machineinstrs -target-abi=ilp32d -global-isel | FileCheck %s
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zvfh \
+; RUN: -verify-machineinstrs -target-abi=ilp32d -global-isel | FileCheck %s
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+v,+zvfh \
+; RUN: -verify-machineinstrs -target-abi=lp64d -global-isel | FileCheck %s
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zfhmin,+zvfh \
+; RUN: -verify-machineinstrs -target-abi=ilp32d -global-isel | FileCheck %s
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+v,+zfhmin,+zvfh \
+; RUN: -verify-machineinstrs -target-abi=lp64d -global-isel | FileCheck %s
+
+declare <vscale x 1 x half> @llvm.riscv.vfadd.nxv1f16.nxv1f16(
+ <vscale x 1 x half>,
+ <vscale x 1 x half>,
+ <vscale x 1 x half>,
+ iXLen, iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfadd_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv1f16_nxv1f16_nxv1f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, mf4, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v9
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 1 x half> @llvm.riscv.vfadd.nxv1f16.nxv1f16(
+ <vscale x 1 x half> undef,
+ <vscale x 1 x half> %0,
+ <vscale x 1 x half> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+ <vscale x 1 x half>,
+ <vscale x 1 x half>,
+ <vscale x 1 x half>,
+ <vscale x 1 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+ <vscale x 1 x half> %0,
+ <vscale x 1 x half> %1,
+ <vscale x 1 x half> %2,
+ <vscale x 1 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 2 x half> @llvm.riscv.vfadd.nxv2f16.nxv2f16(
+ <vscale x 2 x half>,
+ <vscale x 2 x half>,
+ <vscale x 2 x half>,
+ iXLen, iXLen);
+
+define <vscale x 2 x half> @intrinsic_vfadd_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv2f16_nxv2f16_nxv2f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, mf2, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v9
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 2 x half> @llvm.riscv.vfadd.nxv2f16.nxv2f16(
+ <vscale x 2 x half> undef,
+ <vscale x 2 x half> %0,
+ <vscale x 2 x half> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 2 x half> %a
+}
+
+declare <vscale x 2 x half> @llvm.riscv.vfadd.mask.nxv2f16.nxv2f16(
+ <vscale x 2 x half>,
+ <vscale x 2 x half>,
+ <vscale x 2 x half>,
+ <vscale x 2 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 2 x half> @intrinsic_vfadd_mask_vv_nxv2f16_nxv2f16_nxv2f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv2f16_nxv2f16_nxv2f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 2 x half> @llvm.riscv.vfadd.mask.nxv2f16.nxv2f16(
+ <vscale x 2 x half> %0,
+ <vscale x 2 x half> %1,
+ <vscale x 2 x half> %2,
+ <vscale x 2 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 2 x half> %a
+}
+
+declare <vscale x 4 x half> @llvm.riscv.vfadd.nxv4f16.nxv4f16(
+ <vscale x 4 x half>,
+ <vscale x 4 x half>,
+ <vscale x 4 x half>,
+ iXLen, iXLen);
+
+define <vscale x 4 x half> @intrinsic_vfadd_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv4f16_nxv4f16_nxv4f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v9
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 4 x half> @llvm.riscv.vfadd.nxv4f16.nxv4f16(
+ <vscale x 4 x half> undef,
+ <vscale x 4 x half> %0,
+ <vscale x 4 x half> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 4 x half> %a
+}
+
+declare <vscale x 4 x half> @llvm.riscv.vfadd.mask.nxv4f16.nxv4f16(
+ <vscale x 4 x half>,
+ <vscale x 4 x half>,
+ <vscale x 4 x half>,
+ <vscale x 4 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 4 x half> @intrinsic_vfadd_mask_vv_nxv4f16_nxv4f16_nxv4f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv4f16_nxv4f16_nxv4f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 4 x half> @llvm.riscv.vfadd.mask.nxv4f16.nxv4f16(
+ <vscale x 4 x half> %0,
+ <vscale x 4 x half> %1,
+ <vscale x 4 x half> %2,
+ <vscale x 4 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 4 x half> %a
+}
+
+declare <vscale x 8 x half> @llvm.riscv.vfadd.nxv8f16.nxv8f16(
+ <vscale x 8 x half>,
+ <vscale x 8 x half>,
+ <vscale x 8 x half>,
+ iXLen, iXLen);
+
+define <vscale x 8 x half> @intrinsic_vfadd_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv8f16_nxv8f16_nxv8f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m2, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v10
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 8 x half> @llvm.riscv.vfadd.nxv8f16.nxv8f16(
+ <vscale x 8 x half> undef,
+ <vscale x 8 x half> %0,
+ <vscale x 8 x half> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 8 x half> %a
+}
+
+declare <vscale x 8 x half> @llvm.riscv.vfadd.mask.nxv8f16.nxv8f16(
+ <vscale x 8 x half>,
+ <vscale x 8 x half>,
+ <vscale x 8 x half>,
+ <vscale x 8 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 8 x half> @intrinsic_vfadd_mask_vv_nxv8f16_nxv8f16_nxv8f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv8f16_nxv8f16_nxv8f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v10, v12, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 8 x half> @llvm.riscv.vfadd.mask.nxv8f16.nxv8f16(
+ <vscale x 8 x half> %0,
+ <vscale x 8 x half> %1,
+ <vscale x 8 x half> %2,
+ <vscale x 8 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 8 x half> %a
+}
+
+declare <vscale x 16 x half> @llvm.riscv.vfadd.nxv16f16.nxv16f16(
+ <vscale x 16 x half>,
+ <vscale x 16 x half>,
+ <vscale x 16 x half>,
+ iXLen, iXLen);
+
+define <vscale x 16 x half> @intrinsic_vfadd_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv16f16_nxv16f16_nxv16f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m4, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v12
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 16 x half> @llvm.riscv.vfadd.nxv16f16.nxv16f16(
+ <vscale x 16 x half> undef,
+ <vscale x 16 x half> %0,
+ <vscale x 16 x half> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 16 x half> %a
+}
+
+declare <vscale x 16 x half> @llvm.riscv.vfadd.mask.nxv16f16.nxv16f16(
+ <vscale x 16 x half>,
+ <vscale x 16 x half>,
+ <vscale x 16 x half>,
+ <vscale x 16 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 16 x half> @intrinsic_vfadd_mask_vv_nxv16f16_nxv16f16_nxv16f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv16f16_nxv16f16_nxv16f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v12, v16, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 16 x half> @llvm.riscv.vfadd.mask.nxv16f16.nxv16f16(
+ <vscale x 16 x half> %0,
+ <vscale x 16 x half> %1,
+ <vscale x 16 x half> %2,
+ <vscale x 16 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 16 x half> %a
+}
+
+declare <vscale x 32 x half> @llvm.riscv.vfadd.nxv32f16.nxv32f16(
+ <vscale x 32 x half>,
+ <vscale x 32 x half>,
+ <vscale x 32 x half>,
+ iXLen, iXLen);
+
+define <vscale x 32 x half> @intrinsic_vfadd_vv_nxv32f16_nxv32f16_nxv32f16(<vscale x 32 x half> %0, <vscale x 32 x half> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv32f16_nxv32f16_nxv32f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e16, m8, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v16
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 32 x half> @llvm.riscv.vfadd.nxv32f16.nxv32f16(
+ <vscale x 32 x half> undef,
+ <vscale x 32 x half> %0,
+ <vscale x 32 x half> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 32 x half> %a
+}
+
+declare <vscale x 32 x half> @llvm.riscv.vfadd.mask.nxv32f16.nxv32f16(
+ <vscale x 32 x half>,
+ <vscale x 32 x half>,
+ <vscale x 32 x half>,
+ <vscale x 32 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 32 x half> @intrinsic_vfadd_mask_vv_nxv32f16_nxv32f16_nxv32f16(<vscale x 32 x half> %0, <vscale x 32 x half> %1, <vscale x 32 x half> %2, <vscale x 32 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv32f16_nxv32f16_nxv32f16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vl8re16.v v24, (a0)
+; CHECK-NEXT: fsrmi a0, 0
+; CHECK-NEXT: vsetvli zero, a1, e16, m8, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v16, v24, v0.t
+; CHECK-NEXT: fsrm a0
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 32 x half> @llvm.riscv.vfadd.mask.nxv32f16.nxv32f16(
+ <vscale x 32 x half> %0,
+ <vscale x 32 x half> %1,
+ <vscale x 32 x half> %2,
+ <vscale x 32 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 32 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfadd.nxv1f32.nxv1f32(
+ <vscale x 1 x float>,
+ <vscale x 1 x float>,
+ <vscale x 1 x float>,
+ iXLen, iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfadd_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv1f32_nxv1f32_nxv1f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v9
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 1 x float> @llvm.riscv.vfadd.nxv1f32.nxv1f32(
+ <vscale x 1 x float> undef,
+ <vscale x 1 x float> %0,
+ <vscale x 1 x float> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfadd.mask.nxv1f32.nxv1f32(
+ <vscale x 1 x float>,
+ <vscale x 1 x float>,
+ <vscale x 1 x float>,
+ <vscale x 1 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfadd_mask_vv_nxv1f32_nxv1f32_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv1f32_nxv1f32_nxv1f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 1 x float> @llvm.riscv.vfadd.mask.nxv1f32.nxv1f32(
+ <vscale x 1 x float> %0,
+ <vscale x 1 x float> %1,
+ <vscale x 1 x float> %2,
+ <vscale x 1 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 2 x float> @llvm.riscv.vfadd.nxv2f32.nxv2f32(
+ <vscale x 2 x float>,
+ <vscale x 2 x float>,
+ <vscale x 2 x float>,
+ iXLen, iXLen);
+
+define <vscale x 2 x float> @intrinsic_vfadd_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv2f32_nxv2f32_nxv2f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, m1, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v9
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 2 x float> @llvm.riscv.vfadd.nxv2f32.nxv2f32(
+ <vscale x 2 x float> undef,
+ <vscale x 2 x float> %0,
+ <vscale x 2 x float> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 2 x float> %a
+}
+
+declare <vscale x 2 x float> @llvm.riscv.vfadd.mask.nxv2f32.nxv2f32(
+ <vscale x 2 x float>,
+ <vscale x 2 x float>,
+ <vscale x 2 x float>,
+ <vscale x 2 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 2 x float> @intrinsic_vfadd_mask_vv_nxv2f32_nxv2f32_nxv2f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv2f32_nxv2f32_nxv2f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 2 x float> @llvm.riscv.vfadd.mask.nxv2f32.nxv2f32(
+ <vscale x 2 x float> %0,
+ <vscale x 2 x float> %1,
+ <vscale x 2 x float> %2,
+ <vscale x 2 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 2 x float> %a
+}
+
+declare <vscale x 4 x float> @llvm.riscv.vfadd.nxv4f32.nxv4f32(
+ <vscale x 4 x float>,
+ <vscale x 4 x float>,
+ <vscale x 4 x float>,
+ iXLen, iXLen);
+
+define <vscale x 4 x float> @intrinsic_vfadd_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv4f32_nxv4f32_nxv4f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, m2, ta, ma
+; CHECK-NEXT: vfadd.vv v8, v8, v10
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 4 x float> @llvm.riscv.vfadd.nxv4f32.nxv4f32(
+ <vscale x 4 x float> undef,
+ <vscale x 4 x float> %0,
+ <vscale x 4 x float> %1,
+ iXLen 0, iXLen %2)
+
+ ret <vscale x 4 x float> %a
+}
+
+declare <vscale x 4 x float> @llvm.riscv.vfadd.mask.nxv4f32.nxv4f32(
+ <vscale x 4 x float>,
+ <vscale x 4 x float>,
+ <vscale x 4 x float>,
+ <vscale x 4 x i1>,
+ iXLen, iXLen, iXLen);
+
+define <vscale x 4 x float> @intrinsic_vfadd_mask_vv_nxv4f32_nxv4f32_nxv4f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x i1> %3, iXLen %4) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_mask_vv_nxv4f32_nxv4f32_nxv4f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT: vfadd.vv v8, v10, v12, v0.t
+; CHECK-NEXT: fsrm a1
+; CHECK-NEXT: ret
+entry:
+ %a = call <vscale x 4 x float> @llvm.riscv.vfadd.mask.nxv4f32.nxv4f32(
+ <vscale x 4 x float> %0,
+ <vscale x 4 x float> %1,
+ <vscale x 4 x float> %2,
+ <vscale x 4 x i1> %3,
+ iXLen 0, iXLen %4, iXLen 1)
+
+ ret <vscale x 4 x float> %a
+}
+
+declare <vscale x 8 x float> @llvm.riscv.vfadd.nxv8f32.nxv8f32(
+ <vscale x 8 x float>,
+ <vscale x 8 x float>,
+ <vscale x 8 x float>,
+ iXLen, iXLen);
+
+define <vscale x 8 x float> @intrinsic_vfadd_vv_nxv8f32_nxv8f32_nxv8f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1, iXLen %2) nounwind {
+; CHECK-LABEL: intrinsic_vfadd_vv_nxv8f32_nxv8f32_nxv8f32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: fsrmi a1, 0
+; CHECK-NEXT: vsetvli zero, a0, e32, m4, ta, ma
+; CHEC...
[truncated]
``````````
</details>
https://github.com/llvm/llvm-project/pull/156415
More information about the llvm-commits
mailing list