llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT--> @llvm/pr-subscribers-clang Author: Yitzhak Mandelbaum (ymand) <details> <summary>Changes</summary> As per the FIXME, this cap never really served its purpose. This patch simplifies to a single, caller-specified, absolute cap. --- Full diff: https://github.com/llvm/llvm-project/pull/80033.diff 1 Files Affected: - (modified) clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp (-8) ``````````diff diff --git a/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp b/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp index 49e425bde66aa..4c88c46142d64 100644 --- a/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp +++ b/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp @@ -540,14 +540,6 @@ runTypeErasedDataflowAnalysis( Worklist.enqueueSuccessors(&Entry); AnalysisContext AC(CFCtx, Analysis, StartingEnv, BlockStates); - - // FIXME: remove relative cap. There isn't really any good setting for - // `MaxAverageVisitsPerBlock`, so it has no clear value over using - // `MaxBlockVisits` directly. - static constexpr std::int32_t MaxAverageVisitsPerBlock = 4; - const std::int32_t RelativeMaxBlockVisits = - MaxAverageVisitsPerBlock * BlockStates.size(); - MaxBlockVisits = std::min(RelativeMaxBlockVisits, MaxBlockVisits); std::int32_t BlockVisits = 0; while (const CFGBlock *Block = Worklist.dequeue()) { LLVM_DEBUG(llvm::dbgs() `````````` </details> https://github.com/llvm/llvm-project/pull/80033 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
