[PATCH] D42756: AMDGPU: Remove the s_buffer workaround for GFX9 chips

[Marek Olšák via Phabricator via llvm-commits]

This revision was not accepted when it landed; it landed in state "Needs Review".
This revision was automatically updated to reflect the committed changes.
Closed by commit rL324486: AMDGPU: Remove the s_buffer workaround for GFX9 chips (authored by mareko, committed by ).

Repository:
  rL LLVM

https://reviews.llvm.org/D42756

Files:
  llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
  llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
  llvm/trunk/test/CodeGen/AMDGPU/smrd.ll


Index: llvm/trunk/test/CodeGen/AMDGPU/smrd.ll
===================================================================
--- llvm/trunk/test/CodeGen/AMDGPU/smrd.ll
+++ llvm/trunk/test/CodeGen/AMDGPU/smrd.ll
@@ -217,14 +217,8 @@
 ; 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)
Index: llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
===================================================================
--- llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
+++ llvm/trunk/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
@@ -853,9 +853,8 @@
 
       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);
Index: llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
===================================================================
--- llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
+++ llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
@@ -333,14 +333,6 @@
     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);
   }


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D42756.133215.patch
Type: text/x-patch
Size: 2532 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180207/edaf711f/attachment.bin>