diff options
| author | Hongtao Yu <hoy@fb.com> | 2021-06-17 11:09:13 -0700 |
|---|---|---|
| committer | Hongtao Yu <hoy@fb.com> | 2021-06-18 15:14:11 -0700 |
| commit | bd52495518808bdbf24f4d8e9e20774d6d2e3333 (patch) | |
| tree | 584ed0830c3b5ad0a0e9e5007b07972392df3d57 /llvm/lib/CodeGen/BranchFolding.cpp | |
| parent | 36f93fc594ea168d541bb2fde95014a5502bef88 (diff) | |
| download | llvm-bd52495518808bdbf24f4d8e9e20774d6d2e3333.zip llvm-bd52495518808bdbf24f4d8e9e20774d6d2e3333.tar.gz llvm-bd52495518808bdbf24f4d8e9e20774d6d2e3333.tar.bz2 | |
[CSSPGO] Undoing the concept of dangling pseudo probe
As a follow-up to https://reviews.llvm.org/D104129, I'm cleaning up the danling probe related code in both the compiler and llvm-profgen.
I'm seeing a 5% size win for the pseudo_probe section for SPEC2017 and 10% for Ciner. Certain benchmark such as 602.gcc has a 20% size win. No obvious difference seen on build time for SPEC2017 and Cinder.
Reviewed By: wenlei
Differential Revision: https://reviews.llvm.org/D104477
Diffstat (limited to 'llvm/lib/CodeGen/BranchFolding.cpp')
| -rw-r--r-- | llvm/lib/CodeGen/BranchFolding.cpp | 10 |
1 files changed, 0 insertions, 10 deletions
diff --git a/llvm/lib/CodeGen/BranchFolding.cpp b/llvm/lib/CodeGen/BranchFolding.cpp index 6deb2dc0..65e7e92 100644 --- a/llvm/lib/CodeGen/BranchFolding.cpp +++ b/llvm/lib/CodeGen/BranchFolding.cpp @@ -1307,16 +1307,6 @@ static void salvageDebugInfoFromEmptyBlock(const TargetInstrInfo *TII, for (MachineBasicBlock *PredBB : MBB.predecessors()) if (PredBB->succ_size() == 1) copyDebugInfoToPredecessor(TII, MBB, *PredBB); - - // For AutoFDO, if the block is removed, we won't be able to sample it. To - // avoid assigning a zero weight for BB, move all its pseudo probes into once - // of its predecessors or successors and mark them dangling. This should allow - // the counts inference a chance to get a more reasonable weight for the - // block. - if (!MBB.pred_empty()) - MBB.moveAndDanglePseudoProbes(*MBB.pred_begin()); - else if (!MBB.succ_empty()) - MBB.moveAndDanglePseudoProbes(*MBB.succ_begin()); } bool BranchFolder::OptimizeBlock(MachineBasicBlock *MBB) { |