[llvm] 35255cc - [RISCV][SLEEF] Fix incorrect vector function mapping for llvm.exp.f64 (#121641)

[via llvm-commits]

Author: Rux124
Date: 2025-05-12T13:34:46+08:00
New Revision: 35255cce20ceea0b9895633d6976bc7e79a6e2d2

URL: https://github.com/llvm/llvm-project/commit/35255cce20ceea0b9895633d6976bc7e79a6e2d2
DIFF: https://github.com/llvm/llvm-project/commit/35255cce20ceea0b9895633d6976bc7e79a6e2d2.diff

LOG: [RISCV][SLEEF] Fix incorrect vector function mapping for llvm.exp.f64 (#121641)

The element count needs to be SCALABLE(4) to generate operations with
lmul == 2 for the f64 type.

Added: 
    

Modified: 
    llvm/include/llvm/Analysis/VecFuncs.def
    llvm/test/CodeGen/RISCV/replace-with-veclib-sleef-scalable.ll

Removed: 
    


################################################################################
diff  --git a/llvm/include/llvm/Analysis/VecFuncs.def b/llvm/include/llvm/Analysis/VecFuncs.def
index b9f7a38943b05..68753a2497db2 100644
--- a/llvm/include/llvm/Analysis/VecFuncs.def
+++ b/llvm/include/llvm/Analysis/VecFuncs.def
@@ -954,7 +954,7 @@ TLI_DEFINE_VECFUNC("erfc", "Sleef_erfcdx_u15rvvm2",  SCALABLE(2), NOMASK, "_ZGVr
 TLI_DEFINE_VECFUNC("erfcf", "Sleef_erfcfx_u15rvvm2", SCALABLE(4), NOMASK, "_ZGVrNxv")
 
 TLI_DEFINE_VECFUNC("exp", "Sleef_expdx_u10rvvm2", SCALABLE(2), NOMASK, "_ZGVrNxv")
-TLI_DEFINE_VECFUNC("llvm.exp.f64", "Sleef_expdx_u10rvvm2", SCALABLE(4), NOMASK, "_ZGVrNxv")
+TLI_DEFINE_VECFUNC("llvm.exp.f64", "Sleef_expdx_u10rvvm2", SCALABLE(2), NOMASK, "_ZGVrNxv")
 TLI_DEFINE_VECFUNC("expf", "Sleef_expfx_u10rvvm2", SCALABLE(4), NOMASK, "_ZGVrNxv")
 TLI_DEFINE_VECFUNC("llvm.exp.f32", "Sleef_expfx_u10rvvm2", SCALABLE(4), NOMASK, "_ZGVrNxv")
 

diff  --git a/llvm/test/CodeGen/RISCV/replace-with-veclib-sleef-scalable.ll b/llvm/test/CodeGen/RISCV/replace-with-veclib-sleef-scalable.ll
index 134c2b8932003..c489bc3681876 100644
--- a/llvm/test/CodeGen/RISCV/replace-with-veclib-sleef-scalable.ll
+++ b/llvm/test/CodeGen/RISCV/replace-with-veclib-sleef-scalable.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -mtriple=riscv64-unknown-linux-gnu -mattr=+v -vector-library=sleefgnuabi -replace-with-veclib -S < %s | FileCheck %s
 
 ;.
-; CHECK: @llvm.compiler.used = appending global [21 x ptr] [ptr @Sleef_cosdx_u10rvvm2, ptr @Sleef_cosfx_u10rvvm2, ptr @Sleef_expfx_u10rvvm2, ptr @Sleef_exp10dx_u10rvvm2, ptr @Sleef_exp10fx_u10rvvm2, ptr @Sleef_exp2dx_u10rvvm2, ptr @Sleef_exp2fx_u10rvvm2, ptr @Sleef_fmadx_rvvm2, ptr @Sleef_fmafx_rvvm2, ptr @Sleef_logdx_u10rvvm2, ptr @Sleef_logfx_u10rvvm2, ptr @Sleef_log10dx_u10rvvm2, ptr @Sleef_log10fx_u10rvvm2, ptr @Sleef_log2dx_u10rvvm2, ptr @Sleef_log2fx_u10rvvm2, ptr @Sleef_powdx_u10rvvm2, ptr @Sleef_powfx_u10rvvm2, ptr @Sleef_sindx_u10rvvm2, ptr @Sleef_sinfx_u10rvvm2, ptr @Sleef_sqrtdx_u05rvvm2, ptr @Sleef_sqrtfx_u05rvvm2], section "llvm.metadata"
+; CHECK: @llvm.compiler.used = appending global [22 x ptr] [ptr @Sleef_cosdx_u10rvvm2, ptr @Sleef_cosfx_u10rvvm2, ptr @Sleef_expdx_u10rvvm2, ptr @Sleef_expfx_u10rvvm2, ptr @Sleef_exp10dx_u10rvvm2, ptr @Sleef_exp10fx_u10rvvm2, ptr @Sleef_exp2dx_u10rvvm2, ptr @Sleef_exp2fx_u10rvvm2, ptr @Sleef_fmadx_rvvm2, ptr @Sleef_fmafx_rvvm2, ptr @Sleef_logdx_u10rvvm2, ptr @Sleef_logfx_u10rvvm2, ptr @Sleef_log10dx_u10rvvm2, ptr @Sleef_log10fx_u10rvvm2, ptr @Sleef_log2dx_u10rvvm2, ptr @Sleef_log2fx_u10rvvm2, ptr @Sleef_powdx_u10rvvm2, ptr @Sleef_powfx_u10rvvm2, ptr @Sleef_sindx_u10rvvm2, ptr @Sleef_sinfx_u10rvvm2, ptr @Sleef_sqrtdx_u05rvvm2, ptr @Sleef_sqrtfx_u05rvvm2], section "llvm.metadata"
 ;.
 define <vscale x 2 x double> @llvm_ceil_vscale_f64(<vscale x 2 x double> %in) {
 ; CHECK-LABEL: define <vscale x 2 x double> @llvm_ceil_vscale_f64(
@@ -67,7 +67,7 @@ define <vscale x 4 x float> @llvm_cos_vscale_f32(<vscale x 4 x float> %in) {
 define <vscale x 2 x double> @llvm_exp_vscale_f64(<vscale x 2 x double> %in) {
 ; CHECK-LABEL: define <vscale x 2 x double> @llvm_exp_vscale_f64(
 ; CHECK-SAME: <vscale x 2 x double> [[IN:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 2 x double> @llvm.exp.nxv2f64(<vscale x 2 x double> [[IN]])
+; CHECK-NEXT:    [[TMP1:%.*]] = call fast <vscale x 2 x double> @Sleef_expdx_u10rvvm2(<vscale x 2 x double> [[IN]])
 ; CHECK-NEXT:    ret <vscale x 2 x double> [[TMP1]]
 ;
   %1 = call fast <vscale x 2 x double> @llvm.exp.nxv2f64(<vscale x 2 x double> %in)