[llvm] ea9276d - [AMDGPU] GlobalISel for f8 conversions (#80503)

[via llvm-commits]

Author: Stanislav Mekhanoshin
Date: 2024-02-05T09:41:37-08:00
New Revision: ea9276d47efb22e26483bd5ad31c2e249ed9846f

URL: https://github.com/llvm/llvm-project/commit/ea9276d47efb22e26483bd5ad31c2e249ed9846f
DIFF: https://github.com/llvm/llvm-project/commit/ea9276d47efb22e26483bd5ad31c2e249ed9846f.diff

LOG: [AMDGPU] GlobalISel for f8 conversions (#80503)

Added: 
    

Modified: 
    llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
    llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
index 09fac963d222d..5323e4fc58de8 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
@@ -4497,6 +4497,14 @@ AMDGPURegisterBankInfo::getInstrMapping(const MachineInstr &MI) const {
     case Intrinsic::amdgcn_dot4_f32_bf8_fp8:
     case Intrinsic::amdgcn_dot4_f32_fp8_fp8:
     case Intrinsic::amdgcn_dot4_f32_bf8_bf8:
+    case Intrinsic::amdgcn_cvt_f32_fp8:
+    case Intrinsic::amdgcn_cvt_f32_bf8:
+    case Intrinsic::amdgcn_cvt_pk_f32_fp8:
+    case Intrinsic::amdgcn_cvt_pk_f32_bf8:
+    case Intrinsic::amdgcn_cvt_pk_fp8_f32:
+    case Intrinsic::amdgcn_cvt_pk_bf8_f32:
+    case Intrinsic::amdgcn_cvt_sr_fp8_f32:
+    case Intrinsic::amdgcn_cvt_sr_bf8_f32:
     case Intrinsic::amdgcn_wmma_bf16_16x16x16_bf16:
     case Intrinsic::amdgcn_wmma_f16_16x16x16_f16:
     case Intrinsic::amdgcn_wmma_bf16_16x16x16_bf16_tied:

diff  --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll
index f915fa8e6cd1c..fc4b663b85a61 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll
@@ -1,6 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn -mcpu=gfx940 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX940 %s
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX12 %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx940 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX940 %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx940 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX940 %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX12 %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX12 %s
 
 declare float @llvm.amdgcn.cvt.f32.bf8(i32, i32)
 declare float @llvm.amdgcn.cvt.f32.fp8(i32, i32)