[llvm] r282875 - [AMDGPU] Ask subtarget if waitcnt instruction is needed before barrier instruction

[Konstantin Zhuravlyov via llvm-commits]

Author: kzhuravl
Date: Fri Sep 30 11:50:36 2016
New Revision: 282875

URL: http://llvm.org/viewvc/llvm-project?rev=282875&view=rev
Log:
[AMDGPU] Ask subtarget if waitcnt instruction is needed before barrier instruction

Differential Revision: https://reviews.llvm.org/D24985

Modified:
    llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
    llvm/trunk/lib/Target/AMDGPU/SIInsertWaits.cpp

Modified: llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h?rev=282875&r1=282874&r2=282875&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h (original)
+++ llvm/trunk/lib/Target/AMDGPU/AMDGPUSubtarget.h Fri Sep 30 11:50:36 2016
@@ -540,6 +540,12 @@ public:
 
   /// Return the maximum number of waves per SIMD for kernels using \p VGPRs VGPRs
   unsigned getOccupancyWithNumVGPRs(unsigned VGPRs) const;
+
+  /// \returns True if waitcnt instruction is needed before barrier instruction,
+  /// false otherwise.
+  bool needWaitcntBeforeBarrier() const {
+    return true;
+  }
 };
 
 } // End namespace llvm

Modified: llvm/trunk/lib/Target/AMDGPU/SIInsertWaits.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/SIInsertWaits.cpp?rev=282875&r1=282874&r2=282875&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/SIInsertWaits.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/SIInsertWaits.cpp Fri Sep 30 11:50:36 2016
@@ -590,8 +590,9 @@ bool SIInsertWaits::runOnMachineFunction
       // S_SENDMSG implicitly waits for all outstanding LGKM transfers to finish,
       // but we also want to wait for any other outstanding transfers before
       // signalling other hardware blocks
-      if (I->getOpcode() == AMDGPU::S_BARRIER ||
-          I->getOpcode() == AMDGPU::S_SENDMSG)
+      if ((I->getOpcode() == AMDGPU::S_BARRIER &&
+               ST->needWaitcntBeforeBarrier()) ||
+           I->getOpcode() == AMDGPU::S_SENDMSG)
         Required = LastIssued;
       else
         Required = handleOperands(*I);