[llvm] a17f8fe - [InstCombine] Use more inline elements in a SmallVector (#100942)

[via llvm-commits]

Author: Kazu Hirata
Date: 2024-07-28T13:19:44-07:00
New Revision: a17f8fe7d46e1708331df2dec733203de356b0cf

URL: https://github.com/llvm/llvm-project/commit/a17f8fe7d46e1708331df2dec733203de356b0cf
DIFF: https://github.com/llvm/llvm-project/commit/a17f8fe7d46e1708331df2dec733203de356b0cf.diff

LOG: [InstCombine] Use more inline elements in a SmallVector (#100942)

The 4 inline elements only cover 58% of cases encountered here during
the compilation of X86ISelLowering.cpp.ll, a .ll version of
X86ISelLowering.cpp.

The 8 inline elements cover 96% and save 0.27% of heap allocations.

Added: 
    

Modified: 
    llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
index f6c4b6e180937..e61480d7542d7 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -1500,7 +1500,7 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst &CI) {
   // Don't try to simplify calls without uses. It will not do anything useful,
   // but will result in the following folds being skipped.
   if (!CI.use_empty()) {
-    SmallVector<Value *, 4> Args;
+    SmallVector<Value *, 8> Args;
     Args.reserve(CI.arg_size());
     for (Value *Op : CI.args())
       Args.push_back(Op);