diff options
author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2022-09-03 13:12:15 +0100 |
---|---|---|
committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2022-09-03 13:12:22 +0100 |
commit | e2d140e9c33739032db40b18a019ece4fb3f687a (patch) | |
tree | 3b0b20c211e5f2e7334fe3c1e9bcdbec3930f526 /clang/lib/CodeGen/CodeGenFunction.cpp | |
parent | 30dadaa2eb499395f5bef44b9e1a18db6360ad1f (diff) | |
download | llvm-e2d140e9c33739032db40b18a019ece4fb3f687a.zip llvm-e2d140e9c33739032db40b18a019ece4fb3f687a.tar.gz llvm-e2d140e9c33739032db40b18a019ece4fb3f687a.tar.bz2 |
[TTI] Add isExpensiveToSpeculativelyExecute wrapper
CGP uses a raw `getInstructionCost(I, TargetTransformInfo::TCK_SizeAndLatency) >= TCC_Expensive` check to see if its better to move an expensive instruction used in a select behind a branch instead.
This is causing issues with upcoming improvements to TCK_SizeAndLatency costs on X86 as we need to use TCK_SizeAndLatency as an uop count (so its compatible with various target-specific buffer sizes - see D132288), but we can have instructions that have a low TCK_SizeAndLatency value but should still be treated as 'expensive' (FDIV for example) - by adding a isExpensiveToSpeculativelyExecute wrapper we can keep the current behaviour but still add an x86 override in a future patch when the cost tables are updated to compensate.
Diffstat (limited to 'clang/lib/CodeGen/CodeGenFunction.cpp')
0 files changed, 0 insertions, 0 deletions