[llvm] 8f9dd5e - [AMDGPU] Vector register spill test cleanup (NFC)

Christudasan Devadasan via llvm-commits llvm-commits at lists.llvm.org
Tue Apr 26 00:48:52 PDT 2022


Author: Christudasan Devadasan
Date: 2022-04-26T13:17:16+05:30
New Revision: 8f9dd5e608c0ac201ab682ccc89ac3be2dfd0d29

URL: https://github.com/llvm/llvm-project/commit/8f9dd5e608c0ac201ab682ccc89ac3be2dfd0d29
DIFF: https://github.com/llvm/llvm-project/commit/8f9dd5e608c0ac201ab682ccc89ac3be2dfd0d29.diff

LOG: [AMDGPU] Vector register spill test cleanup (NFC)

The vector register spills have no dependency with
SILowerSGPRSpills pass anymore. The entire handling
has been moved to PrologEpilogInserter with D55301.

Added: 
    

Modified: 
    llvm/test/CodeGen/AMDGPU/pei-build-spill-partial-agpr.mir
    llvm/test/CodeGen/AMDGPU/spill-agpr.mir
    llvm/test/CodeGen/AMDGPU/vgpr-spill-dead-frame-in-dbg-value.mir

Removed: 
    


################################################################################
diff  --git a/llvm/test/CodeGen/AMDGPU/pei-build-spill-partial-agpr.mir b/llvm/test/CodeGen/AMDGPU/pei-build-spill-partial-agpr.mir
index 3eaa44691ab9c..a1d9c17d9c393 100644
--- a/llvm/test/CodeGen/AMDGPU/pei-build-spill-partial-agpr.mir
+++ b/llvm/test/CodeGen/AMDGPU/pei-build-spill-partial-agpr.mir
@@ -1,6 +1,6 @@
 # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
-# RUN: llc -march=amdgcn -mcpu=gfx908 -verify-machineinstrs -run-pass=si-lower-sgpr-spills,prologepilog -o - %s | FileCheck -check-prefix=MUBUF-V2A %s
-# RUN: llc -march=amdgcn -mcpu=gfx908 -verify-machineinstrs -mattr=+enable-flat-scratch -run-pass=si-lower-sgpr-spills,prologepilog -o - %s | FileCheck -check-prefix=FLATSCR-V2A %s
+# RUN: llc -march=amdgcn -mcpu=gfx908 -verify-machineinstrs -run-pass=prologepilog -o - %s | FileCheck -check-prefix=MUBUF-V2A %s
+# RUN: llc -march=amdgcn -mcpu=gfx908 -verify-machineinstrs -mattr=+enable-flat-scratch -run-pass=prologepilog -o - %s | FileCheck -check-prefix=FLATSCR-V2A %s
 
 --- |
 

diff  --git a/llvm/test/CodeGen/AMDGPU/spill-agpr.mir b/llvm/test/CodeGen/AMDGPU/spill-agpr.mir
index 9798999068b5a..db883b986e49e 100644
--- a/llvm/test/CodeGen/AMDGPU/spill-agpr.mir
+++ b/llvm/test/CodeGen/AMDGPU/spill-agpr.mir
@@ -1,8 +1,8 @@
 # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
 # RUN: llc -march=amdgcn -mcpu=gfx908 -run-pass=regallocfast -o - %s | FileCheck -check-prefix=GFX908-SPILLED %s
-# RUN: llc -march=amdgcn -mcpu=gfx908 -run-pass=regallocfast,si-lower-sgpr-spills,prologepilog -o - %s | FileCheck -check-prefix=GFX908-EXPANDED %s
+# RUN: llc -march=amdgcn -mcpu=gfx908 -run-pass=regallocfast,prologepilog -o - %s | FileCheck -check-prefix=GFX908-EXPANDED %s
 # RUN: llc -march=amdgcn -mcpu=gfx90a -run-pass=regallocfast -o - %s | FileCheck -check-prefix=GFX90A-SPILLED %s
-# RUN: llc -march=amdgcn -mcpu=gfx90a -run-pass=regallocfast,si-lower-sgpr-spills,prologepilog -o - %s | FileCheck -check-prefix=GFX90A-EXPANDED %s
+# RUN: llc -march=amdgcn -mcpu=gfx90a -run-pass=regallocfast,prologepilog -o - %s | FileCheck -check-prefix=GFX90A-EXPANDED %s
 
 ---
 name: spill_restore_agpr32

diff  --git a/llvm/test/CodeGen/AMDGPU/vgpr-spill-dead-frame-in-dbg-value.mir b/llvm/test/CodeGen/AMDGPU/vgpr-spill-dead-frame-in-dbg-value.mir
index 76a6baf139868..7c559f74f949a 100644
--- a/llvm/test/CodeGen/AMDGPU/vgpr-spill-dead-frame-in-dbg-value.mir
+++ b/llvm/test/CodeGen/AMDGPU/vgpr-spill-dead-frame-in-dbg-value.mir
@@ -1,7 +1,7 @@
-# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx908 -amdgpu-spill-vgpr-to-agpr=true -verify-machineinstrs -run-pass=si-lower-sgpr-spills,prologepilog -o - %s | FileCheck %s
+# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx908 -amdgpu-spill-vgpr-to-agpr=true -verify-machineinstrs -run-pass=prologepilog -o - %s | FileCheck %s
 
-# After handling the VGPR spill to AGPR copy in SILowerSGPRSpills pass, replace the dead frame index in the DBG_VALUE instruction with reg 0.
-# Otherwise, the test would crash during PEI while trying to replace the dead frame index.
+# After handling the VGPR spill to AGPR copy, replace the dead frame index in the DBG_VALUE instruction with reg 0.
+# Otherwise, the test would crash while trying to replace the dead frame index.
 --- |
   define amdgpu_kernel void @test() { ret void }
 


        


More information about the llvm-commits mailing list