[PATCH] D54855: [AMDGPU] An exp must be branched over if exec=0

[Tim Renouf via Phabricator via llvm-commits]

tpr created this revision.
Herald added subscribers: llvm-commits, t-tye, dstuttard, yaxunl, nhaehnle, wdng, jvesely, kzhuravl, arsenm.

Change-Id: I1f57e70da08c08fe65917d762d82ba28251a5b5b


Repository:
  rL LLVM

https://reviews.llvm.org/D54855

Files:
  lib/Target/AMDGPU/SIInsertSkips.cpp
  test/CodeGen/AMDGPU/branch-over-exp.mir


Index: test/CodeGen/AMDGPU/branch-over-exp.mir
===================================================================
--- /dev/null
+++ test/CodeGen/AMDGPU/branch-over-exp.mir
@@ -0,0 +1,28 @@
+# RUN: llc -march=amdgcn -mcpu=gfx900 -run-pass si-insert-skips -verify-machineinstrs -o - %s | FileCheck --enable-var-scope %s
+#
+# Check that we get a S_CBRANCH_EXECZ over an export.
+#
+# CHECK: S_CBRANCH_EXECZ
+
+---
+name:            _func
+body:             |
+  bb.0:
+    successors: %bb.1(0x40000000), %bb.2(0x40000000)
+    liveins: $vgpr0, $vgpr1, $vgpr2
+  
+    V_CMP_LT_U32_e32 killed $vgpr0, killed $vgpr1, implicit-def $vcc, implicit $exec
+    $sgpr0_sgpr1 = S_AND_SAVEEXEC_B64 killed $vcc, implicit-def $exec, implicit-def $scc, implicit $exec
+    SI_MASK_BRANCH %bb.2, implicit $exec
+    S_BRANCH %bb.1
+  
+  bb.1:
+    successors: %bb.2(0x80000000)
+    liveins: $vgpr2
+  
+    EXP_DONE 32, killed renamable $vgpr2, undef renamable $vgpr0, undef renamable $vgpr0, undef renamable $vgpr0, 0, 0, 1, implicit $exec
+  
+  bb.2:
+    S_ENDPGM
+
+...
Index: lib/Target/AMDGPU/SIInsertSkips.cpp
===================================================================
--- lib/Target/AMDGPU/SIInsertSkips.cpp
+++ lib/Target/AMDGPU/SIInsertSkips.cpp
@@ -145,6 +145,11 @@
         return true;
       }
 
+      // We must branch over export. Export with no lanes enabled is not the same as
+      // not exporting at all, at least in some cases.
+      if (I->getOpcode() == AMDGPU::EXP || I->getOpcode() == AMDGPU::EXP_DONE)
+        return true;
+
       ++NumInstr;
 
       if (NumInstr >= SkipThreshold)


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D54855.175110.patch
Type: text/x-patch
Size: 1612 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20181123/2fb77ddf/attachment.bin>