[llvm] r324486 - AMDGPU: Remove the s_buffer workaround for GFX9 chips
Marek Olsak via llvm-commits
llvm-commits at lists.llvm.org
Wed Feb 7 08:00:40 PST 2018
Author: mareko
Date: Wed Feb 7 08:00:40 2018
New Revision: 324486
URL: http://llvm.org/viewvc/llvm-project?rev=324486&view=rev
Log:
AMDGPU: Remove the s_buffer workaround for GFX9 chips
Summary:
I checked the AMD closed source compiler and the workaround is only
needed when x3 is emulated as x4, which we don't do in LLVM.
SMEM x3 opcodes don't exist, and instead there is a possibility to use x4
with the last component being unused. If the last component is out of
buffer bounds and falls on the next 4K page, the hw hangs.
Reviewers: arsenm, nhaehnle
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye
Differential Revision: https://reviews.llvm.org/D42756
Modified:
llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
llvm/trunk/test/CodeGen/AMDGPU/smrd.ll
Modified: llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h?rev=324486&r1=324485&r2=324486&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h (original)
+++ llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h Wed Feb 7 08:00:40 2018
@@ -333,14 +333,6 @@ public:
return HasMadMixInsts;
}
- bool hasSBufferLoadStoreAtomicDwordxN() const {
- // Only use the "x1" variants on GFX9 or don't use the buffer variants.
- // For x2 and higher variants, if the accessed region spans 2 VM pages and
- // the second page is unmapped, the hw hangs.
- // TODO: There is one future GFX9 chip that doesn't have this bug.
- return getGeneration() != GFX9;
- }
-
bool hasCARRY() const {
return (getGeneration() >= EVERGREEN);
}
Modified: llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp?rev=324486&r1=324485&r2=324486&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp Wed Feb 7 08:00:40 2018
@@ -853,9 +853,8 @@ bool SILoadStoreOptimizer::optimizeBlock
continue;
}
- if (STM->hasSBufferLoadStoreAtomicDwordxN() &&
- (Opc == AMDGPU::S_BUFFER_LOAD_DWORD_IMM ||
- Opc == AMDGPU::S_BUFFER_LOAD_DWORDX2_IMM)) {
+ if (Opc == AMDGPU::S_BUFFER_LOAD_DWORD_IMM ||
+ Opc == AMDGPU::S_BUFFER_LOAD_DWORDX2_IMM) {
// EltSize is in units of the offset encoding.
CI.InstClass = S_BUFFER_LOAD_IMM;
CI.EltSize = AMDGPU::getSMRDEncodedOffset(*STM, 4);
Modified: llvm/trunk/test/CodeGen/AMDGPU/smrd.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/smrd.ll?rev=324486&r1=324485&r2=324486&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/AMDGPU/smrd.ll (original)
+++ llvm/trunk/test/CodeGen/AMDGPU/smrd.ll Wed Feb 7 08:00:40 2018
@@ -217,14 +217,8 @@ main_body:
; GCN-NEXT: %bb.
; SICI-NEXT: s_buffer_load_dwordx4 s[{{[0-9]}}:{{[0-9]}}], s[0:3], 0x1
; SICI-NEXT: s_buffer_load_dwordx2 s[{{[0-9]}}:{{[0-9]}}], s[0:3], 0x7
-; VI-NEXT: s_buffer_load_dwordx4 s[{{[0-9]}}:{{[0-9]}}], s[0:3], 0x4
-; VI-NEXT: s_buffer_load_dwordx2 s[{{[0-9]}}:{{[0-9]}}], s[0:3], 0x1c
-; GFX9-NEXT: s_buffer_load_dword s{{[0-9]}}
-; GFX9-NEXT: s_buffer_load_dword s{{[0-9]}}
-; GFX9-NEXT: s_buffer_load_dword s{{[0-9]}}
-; GFX9-NEXT: s_buffer_load_dword s{{[0-9]}}
-; GFX9-NEXT: s_buffer_load_dword s{{[0-9]}}
-; GFX9-NEXT: s_buffer_load_dword s{{[0-9]}}
+; VIGFX9-NEXT: s_buffer_load_dwordx4 s[{{[0-9]}}:{{[0-9]}}], s[0:3], 0x4
+; VIGFX9-NEXT: s_buffer_load_dwordx2 s[{{[0-9]}}:{{[0-9]}}], s[0:3], 0x1c
define amdgpu_ps void @smrd_imm_merged(<4 x i32> inreg %desc) #0 {
main_body:
%r1 = call float @llvm.SI.load.const.v4i32(<4 x i32> %desc, i32 4)
More information about the llvm-commits
mailing list