Cherry-pick four PRs for SWDEV-517168 (llvm#766)

Author: jharryma

llvm/docs/AMDGPUUsage.rst

@@ -1697,6 +1697,24 @@ The AMDGPU backend supports the following LLVM IR attributes.
                                              function which requires AGPRs is reached through any function marked
                                              with this attribute.
 
+     "amdgpu-hidden-argument"                This attribute is used internally by the backend to mark function arguments
+                                             as hidden. Hidden arguments are managed by the compiler and are not part of
+                                             the explicit arguments supplied by the user.
+
+     "amdgpu-sgpr-hazard-wait"               Disabled SGPR hazard wait insertion if set to 0.
+                                             Exists for testing performance impact of SGPR hazard waits only.
+
+     "amdgpu-sgpr-hazard-boundary-cull"      Enable insertion of SGPR hazard cull sequences at function call boundaries.
+                                             Cull sequence reduces future hazard waits, but has a performance cost.
+
+     "amdgpu-sgpr-hazard-mem-wait-cull"      Enable insertion of SGPR hazard cull sequences before memory waits.
+                                             Cull sequence reduces future hazard waits, but has a performance cost.
+                                             Attempt to amortize cost by overlapping with memory accesses.
+
+     "amdgpu-sgpr-hazard-mem-wait-cull-threshold"
+                                             Sets the number of active SGPR hazards that must be present before
+                                             inserting a cull sequence at a memory wait.
+
      ======================================= ==========================================================
 
 Calling Conventions

llvm/lib/Target/AMDGPU/AMDGPU.h

@@ -421,6 +421,9 @@ void initializeAMDGPUSetWavePriorityPass(PassRegistry &);
 void initializeGCNRewritePartialRegUsesPass(llvm::PassRegistry &);
 extern char &GCNRewritePartialRegUsesID;
 
+void initializeAMDGPUWaitSGPRHazardsLegacyPass(PassRegistry &);
+extern char &AMDGPUWaitSGPRHazardsLegacyID;
+
 namespace AMDGPU {
 enum TargetIndex {
   TI_CONSTDATA_START,

llvm/lib/Target/AMDGPU/AMDGPU.td

@@ -1769,7 +1769,8 @@ def FeatureISAVersion11_Generic: FeatureSet<
      FeatureUserSGPRInit16Bug,
      FeatureMADIntraFwdBug,
      FeaturePrivEnabledTrap2NopBug,
-     FeatureRequiresCOV6])>;
+     FeatureRequiresCOV6,
+     FeatureRequiredExportPriority])>;
 
 def FeatureISAVersion11_0_Common : FeatureSet<
   !listconcat(FeatureISAVersion11_Common.Features,
@@ -1799,20 +1800,23 @@ def FeatureISAVersion11_5_0 : FeatureSet<
   !listconcat(FeatureISAVersion11_Common.Features,
     [FeatureSALUFloatInsts,
      FeatureDPPSrc1SGPR,
-     FeatureVGPRSingleUseHintInsts])>;
+     FeatureVGPRSingleUseHintInsts,
+     FeatureRequiredExportPriority])>;
 
 def FeatureISAVersion11_5_1 : FeatureSet<
   !listconcat(FeatureISAVersion11_Common.Features,
     [FeatureSALUFloatInsts,
      FeatureDPPSrc1SGPR,
      FeatureVGPRSingleUseHintInsts,
-     Feature1_5xVGPRs])>;
+     Feature1_5xVGPRs,
+     FeatureRequiredExportPriority])>;
 
 def FeatureISAVersion11_5_2 : FeatureSet<
   !listconcat(FeatureISAVersion11_Common.Features,
     [FeatureSALUFloatInsts,
      FeatureDPPSrc1SGPR,
-     FeatureVGPRSingleUseHintInsts])>;
+     FeatureVGPRSingleUseHintInsts,
+     FeatureRequiredExportPriority])>;
 
 def FeatureISAVersion11_5_3 : FeatureSet<
   !listconcat(FeatureISAVersion11_Common.Features,

llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp

@@ -26,6 +26,7 @@
 #include "AMDGPUTargetObjectFile.h"
 #include "AMDGPUTargetTransformInfo.h"
 #include "AMDGPUUnifyDivergentExitNodes.h"
+#include "AMDGPUWaitSGPRHazards.h"
 #include "GCNIterativeScheduler.h"
 #include "GCNSchedStrategy.h"
 #include "GCNVOPDUtils.h"
@@ -521,6 +522,7 @@ extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeAMDGPUTarget() {
   initializeGCNPreRALongBranchRegPass(*PR);
   initializeGCNRewritePartialRegUsesPass(*PR);
   initializeGCNRegPressurePrinterPass(*PR);
+  initializeAMDGPUWaitSGPRHazardsLegacyPass(*PR);
 }
 
 static std::unique_ptr<TargetLoweringObjectFile> createTLOF(const Triple &TT) {
@@ -1623,6 +1625,8 @@ void GCNPassConfig::addPreEmitPass() {
   if (isPassEnabled(EnableInsertSingleUseVDST, CodeGenOptLevel::Less))
     addPass(&AMDGPUInsertSingleUseVDSTID);
 
+  addPass(&AMDGPUWaitSGPRHazardsLegacyID);
+
   if (isPassEnabled(EnableInsertDelayAlu, CodeGenOptLevel::Less))
     addPass(&AMDGPUInsertDelayAluID);