[all-commits] [llvm/llvm-project] 9e77f5: [LV] Account for vp_merge in out of loop EVL reduc...
Luke Lau via All-commits
all-commits at lists.llvm.org
Wed Nov 13 23:55:39 PST 2024
Branch: refs/heads/main
Home: https://github.com/llvm/llvm-project
Commit: 9e77f59005917e32f09136fa43018f471267f5bd
https://github.com/llvm/llvm-project/commit/9e77f59005917e32f09136fa43018f471267f5bd
Author: Luke Lau <luke at igalia.com>
Date: 2024-11-14 (Thu, 14 Nov 2024)
Changed paths:
M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
A llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reduction-cost.ll
Log Message:
-----------
[LV] Account for vp_merge in out of loop EVL reductions in legacy cost model (#115903)
In #101641, support for out of loop reductions with EVL tail folding was
added by transforming selects to vp_merges in
transformRecipestoEVLRecipes.
Whilst the select was previously free, the vp_merge wasn't and incurs a
cost on RISC-V with the VPlan cost model. But this diverged from the
legacy cost model and caused the "VPlan cost model and legacy cost model
disagreed" assertion to trigger when building 502.gcc_r from SPEC CPU
2017.
Neither the select nor vp_merge recipes from the VPlan exist in the
underlying instructions, so I thought it would make the most sense to
fix this by adding the cost to the underlying phi instruction in
getInstructionCost.
It's worth noting that on RISC-V this vp_merge won't actually generate
any instructions because the mask is all true, and will be folded away.
So we should update the cost model at some point to reflect that.
To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications
More information about the All-commits
mailing list