[llvm] [GlobalISel][IRTranslator] Add tiny hot caches for Value->VReg/Type and Alloca->FI (NFCI) (PR #158092)
Alexis Engelke via llvm-commits
llvm-commits at lists.llvm.org
Fri Sep 12 00:01:24 PDT 2025
- Previous message: [llvm] [GlobalISel][IRTranslator] Add tiny hot caches for Value->VReg/Type and Alloca->FI (NFCI) (PR #158092)
- Next message: [llvm] Exploit xxeval instruction for operations of the form ternary(A, X, NOR(B,C)), ternary(A, X, EQV(B,C)), ternary(A, X, NAND(B,C)), ternary(A, X, NOT(B)) and ternary(A, X, NOT(C)) (PR #158096)
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
aengelke wrote:
Repeating from #157964: If this really is a performance problem (I don't remember seeing this as substantial in my profiles when I looked at it last year), I think we should pursue adding an auxiliary integer field to Value or Instruction and use this instead.
When I did that change to SelectionDAG (ec8bcda2b024efedadde12c3eab1aa0ed291bb6d, unmerged), however, I didn't feel the improvement was big enough (-0.18%, [c-t-t](http://llvm-compile-time-tracker.com/compare.php?from=c59a59e3a5009b0fd924a3b871c2ae0fd0e71815&to=ec8bcda2b024efedadde12c3eab1aa0ed291bb6d&stat=instructions:u)) to push for that aux field. It might be worth re-considering, but it needs a more work on other parts to show that it's worth the hidden complexity.
https://github.com/llvm/llvm-project/pull/158092
- Previous message: [llvm] [GlobalISel][IRTranslator] Add tiny hot caches for Value->VReg/Type and Alloca->FI (NFCI) (PR #158092)
- Next message: [llvm] Exploit xxeval instruction for operations of the form ternary(A, X, NOR(B,C)), ternary(A, X, EQV(B,C)), ternary(A, X, NAND(B,C)), ternary(A, X, NOT(B)) and ternary(A, X, NOT(C)) (PR #158096)
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
More information about the llvm-commits
mailing list