[Attributor][PM] Introduce `-attributor-enable={none,cgscc,module,all}`

[jdoerfert]

Summary

The old command line option -attributor-disable was too coarse grained
as we want to measure the effects of the module or cgscc pass without
the other as well.

Since none is the default there is no real functional change.

[lebedev.ri]

Looks reasonable to me

llvm/include/llvm/Transforms/IPO/Attributor.h

2868–2869

1 << 0
1 << 1

llvm/include/llvm/Transforms/IPO/Attributor.h

if (RangeV.isEmptySet())
       return llvm::None;
     return nullptr;
   }
 
   /// Unique ID (due to the unique address)
   static const char ID;
 };
 
+/// Run options, used by the pass manager.
+enum AttributorRunOption {
+  NONE = 0,
+  MODULE = 1,
+  CGSCC = 2,
+  ALL = MODULE | CGSCC
+};
+
 } // end namespace llvm
 
 #endif // LLVM_TRANSFORMS_IPO_FUNCTIONATTRS_H

llvm/lib/Passes/PassBuilder.cpp

PipelineTuningOptions::PipelineTuningOptions() {
   CallGraphProfile = EnableCallGraphProfile;
 }
 
 extern cl::opt<bool> EnableHotColdSplit;
 extern cl::opt<bool> EnableOrderFileInstrumentation;
 
 extern cl::opt<bool> FlattenedProfileUsed;
 
-extern cl::opt<bool> DisableAttributor;
+extern cl::opt<AttributorRunOption> AttributorRun;
 
 const PassBuilder::OptimizationLevel PassBuilder::OptimizationLevel::O0 = {
     /*SpeedLevel*/ 0,
     /*SizeLevel*/ 0};
 const PassBuilder::OptimizationLevel PassBuilder::OptimizationLevel::O1 = {
     /*SpeedLevel*/ 1,
     /*SizeLevel*/ 0};
 const PassBuilder::OptimizationLevel PassBuilder::OptimizationLevel::O2 = {
if (Phase != ThinLTOPhase::PreLink)
       // We perform early indirect call promotion here, before globalopt.
       // This is important for the ThinLTO backend phase because otherwise
       // imported available_externally functions look unreferenced and are
       // removed.
       MPM.addPass(PGOIndirectCallPromotion(Phase == ThinLTOPhase::PostLink,
                                            true /* SamplePGO */));
   }
 
-  if (!DisableAttributor)
+  if (AttributorRun & AttributorRunOption::MODULE)
     MPM.addPass(AttributorPass());
 
   // Interprocedural constant propagation now that basic cleanup has occurred
   // and prior to optimizing globals.
   // FIXME: This position in the pipeline hasn't been carefully considered in
   // years, it should be re-analyzed.
   MPM.addPass(IPSCCPPass());
 
PassBuilder::buildModuleSimplificationPipeline(OptimizationLevel Level,
   // For PreLinkThinLTO pass, we disable hot-caller heuristic for sample PGO
   // because it makes profile annotation in the backend inaccurate.
   InlineParams IP = getInlineParamsFromOptLevel(Level);
   if (Phase == ThinLTOPhase::PreLink && PGOOpt &&
       PGOOpt->Action == PGOOptions::SampleUse)
     IP.HotCallSiteThreshold = 0;
   MainCGPipeline.addPass(InlinerPass(IP));
 
-  if (!DisableAttributor)
+  if (AttributorRun & AttributorRunOption::CGSCC)
     MainCGPipeline.addPass(AttributorCGSCCPass());
 
   if (PTO.Coroutines)
     MainCGPipeline.addPass(CoroSplitPass());
 
   // Now deduce any function attributes based in the current code.
   MainCGPipeline.addPass(PostOrderFunctionAttrsPass());
 

llvm/lib/Transforms/IPO/PassManagerBuilder.cpp

 cl::opt<bool> EnableOrderFileInstrumentation(
     "enable-order-file-instrumentation", cl::init(false), cl::Hidden,
     cl::desc("Enable order file instrumentation (default = off)"));
 
 static cl::opt<bool>
     EnableMatrix("enable-matrix", cl::init(false), cl::Hidden,
                  cl::desc("Enable lowering of the matrix intrinsics"));
 
-cl::opt<bool> DisableAttributor(
-    "attributor-disable", cl::Hidden,
-    cl::desc("Disable the attributor inter-procedural deduction pass."),
-    cl::init(true));
+cl::opt<AttributorRunOption> AttributorRun(
+    "attributor-enable", cl::Hidden, cl::init(AttributorRunOption::NONE),
+    cl::desc("Enable the attributor inter-procedural deduction pass."),
+    cl::values(clEnumValN(AttributorRunOption::ALL, "all",
+                          "enable all attributor runs"),
+               clEnumValN(AttributorRunOption::MODULE, "module",
+                          "enable module-wide attributor runs"),
+               clEnumValN(AttributorRunOption::CGSCC, "cgscc",
+                          "enable call graph SCC attributor runs"),
+               clEnumValN(AttributorRunOption::NONE, "none",
+                          "disable attributor runs")));
 
 PassManagerBuilder::PassManagerBuilder() {
     OptLevel = 2;
     SizeLevel = 0;
     LibraryInfo = nullptr;
     Inliner = nullptr;
     DisableUnrollLoops = false;
     SLPVectorize = false;
bool PrepareForThinLTOUsingPGOSampleProfile =
       PrepareForThinLTO && !PGOSampleUse.empty();
   if (PrepareForThinLTOUsingPGOSampleProfile)
     DisableUnrollLoops = true;
 
   // Infer attributes about declarations if possible.
   MPM.add(createInferFunctionAttrsLegacyPass());
 
   // Infer attributes on declarations, call sites, arguments, etc.
-  if (!DisableAttributor)
+  if (AttributorRun & AttributorRunOption::MODULE)
     MPM.add(createAttributorLegacyPass());
 
   addExtensionsToPM(EP_ModuleOptimizerEarly, MPM);
 
   if (OptLevel > 2)
     MPM.add(createCallSiteSplittingPass());
 
   MPM.add(createIPSCCPPass());          // IP SCCP
void PassManagerBuilder::populateModulePassManager(
   bool RunInliner = false;
   if (Inliner) {
     MPM.add(Inliner);
     Inliner = nullptr;
     RunInliner = true;
   }
 
   // Infer attributes on declarations, call sites, arguments, etc. for an SCC.
-  if (!DisableAttributor)
+  if (AttributorRun & AttributorRunOption::CGSCC)
     MPM.add(createAttributorCGSCCLegacyPass());
 
   // Try to perform OpenMP specific optimizations. This is a (quick!) no-op if
   // there are no OpenMP runtime calls present in the module.
   if (OptLevel > 1)
     MPM.add(createOpenMPOptLegacyPass());
 
   MPM.add(createPostOrderFunctionAttrsLegacyPass());
if (OptLevel > 1) {
     // pointers passed as arguments to direct uses of functions.
     PM.add(createIPSCCPPass());
 
     // Attach metadata to indirect call sites indicating the set of functions
     // they may target at run-time. This should follow IPSCCP.
     PM.add(createCalledValuePropagationPass());
 
     // Infer attributes on declarations, call sites, arguments, etc.
-    if (!DisableAttributor)
+    if (AttributorRun & AttributorRunOption::MODULE)
       PM.add(createAttributorLegacyPass());
   }
 
   // Infer attributes about definitions. The readnone attribute in particular is
   // required for virtual constant propagation.
   PM.add(createPostOrderFunctionAttrsLegacyPass());
   PM.add(createReversePostOrderFunctionAttrsPass());
 
void PassManagerBuilder::addLTOOptimizationPasses(legacy::PassManagerBase &PM) {
   }
 
   PM.add(createPruneEHPass());   // Remove dead EH info.
 
   // CSFDO instrumentation and use pass.
   addPGOInstrPasses(PM, /* IsCS */ true);
 
   // Infer attributes on declarations, call sites, arguments, etc. for an SCC.
-  if (!DisableAttributor)
+  if (AttributorRun & AttributorRunOption::CGSCC)
     PM.add(createAttributorCGSCCLegacyPass());
 
   // Try to perform OpenMP specific optimizations. This is a (quick!) no-op if
   // there are no OpenMP runtime calls present in the module.
   if (OptLevel > 1)
     PM.add(createOpenMPOptLegacyPass());
 
   // Optimize globals again if we ran the inliner.

llvm/test/Transforms/IPConstantProp/multiple_callbacks.ll

 ; RUN: opt -ipconstprop -S < %s | FileCheck %s
 ; FIXME: There seems to be another instance of nondeterminism which causes the number of iterations to be either 1 or 3, depending on the system. 
 ;        This needs to be investigated and resolved. In the meantime we do not verify the number of iterations.
-; opt -S -passes=attributor -aa-pipeline='basic-aa' -attributor-disable=false -attributor-max-iterations-verify -attributor-max-iterations=1 < %s | FileCheck %s
-; RUN: opt -S -passes=attributor -aa-pipeline='basic-aa' -attributor-disable=false -attributor-max-iterations-verify=false -attributor-max-iterations=3 < %s | FileCheck %s
+; opt -S -passes=attributor -aa-pipeline='basic-aa' -attributor-enable=all -attributor-max-iterations-verify -attributor-max-iterations=1 < %s | FileCheck %s
+; RUN: opt -S -passes=attributor -aa-pipeline='basic-aa' -attributor-enable=all -attributor-max-iterations-verify=false -attributor-max-iterations=3 < %s | FileCheck %s
 ;
 ;
 ;                            /---------------------------------------|
 ;                            |                /----------------------|----|
 ;                            |                |                /-----|    |
 ;                            V                V                V     |    |
 ;    void broker(int (*cb0)(int), int (*cb1)(int), int (*cb2)(int), int, int);
 ;

llvm/test/Transforms/OpenMP/parallel_deletion.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature
-; RUN: opt -S -attributor -openmpopt -attributor-disable=false < %s | FileCheck %s
-; RUN: opt -S -passes='attributor,cgscc(openmpopt)' -attributor-disable=false < %s | FileCheck %s
+; RUN: opt -S -attributor -openmpopt -attributor-enable=all < %s | FileCheck %s
+; RUN: opt -S -passes='attributor,cgscc(openmpopt)' -attributor-enable=all < %s | FileCheck %s
 ;
 target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
 
 %struct.ident_t = type { i32, i32, i32, i32, i8* }
 
 @.str = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00", align 1
 @0 = private unnamed_addr global %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @.str, i32 0, i32 0) }, align 8
 @1 = private unnamed_addr global %struct.ident_t { i32 0, i32 322, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @.str, i32 0, i32 0) }, align 8

llvm/test/Transforms/OpenMP/rtf_type_checking.ll

 ; RUN: opt -S -openmpopt -stats < %s 2>&1 | FileCheck %s
-; RUN: opt -S -attributor -attributor-disable=false -openmpopt -stats < %s 2>&1 | FileCheck %s
+; RUN: opt -S -attributor -attributor-enable=all -openmpopt -stats < %s 2>&1 | FileCheck %s
 ; REQUIRES: asserts
 
 target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
 
 %struct.ident_t = type { i32, i32, i32, i32, i8* }
 
 @.str = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00", align 1
 @0 = private unnamed_addr global %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* @.str, i32 0, i32 0) }, align 8

 ; ===-------------------------------------------------------------------------===
 ;                         ... Statistics Collected ...
 ; ===-------------------------------------------------------------------------===
 ;
 ; CHECK: 1 cgscc-passmgr - Maximum CGSCCPassMgr iterations on one SCC
 ; CHECK: 2 openmp-opt{{.*}}Number of OpenMP runtime functions identified
 ;
 ; There are two matches since the pass is run once per function.
 No newline at end of file