[llvm] [SLP] Bail out on store-to-load forwarding hazards (PR #199606)
Alexey Bataev via llvm-commits
llvm-commits at lists.llvm.org
Thu Jul 23 05:18:21 PDT 2026
================
@@ -27384,6 +27418,142 @@ bool SLPVectorizerPass::runImpl(Function &F, ScalarEvolution *SE_,
return Changed;
}
+bool BoUpSLP::findStoreLoadForwardingConflict(StoreInst *BaseStore,
----------------
alexey-bataev wrote:
Yes, but scanning through the graph is more correct and ribust, not all trees include store-loads pairs, better to do the analysis only for actually used loads/stores to save compile time
https://github.com/llvm/llvm-project/pull/199606
More information about the llvm-commits
mailing list