[PATCH] D131700: [AArch64] Add pattern for SQDML*Lv1i32_indexed
Dave Green via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Aug 15 09:04:54 PDT 2022
dmgreen added a comment.
Can you add a testcase where the extractelement is not the 0 lane? It would be good to have tests to make sure that the new pattern with non-zero offset works OK.
The moved pattern looks OK from what I can see. The input hasn't changed, just been reformatted? Just the output of it has changed? I agree that I'm not sure what the FIXME meant. Tablegen can be finicky at times.
================
Comment at: llvm/lib/Target/AArch64/AArch64InstrFormats.td:8908
+ (v4i16 V64:$Rm))),
+ VectorIndexH:$idx)))),
+ (!cast<Instruction>(NAME # v1i32_indexed)
----------------
Does this need to be 0, if we use the lowest lane in the output pattern `(EXTRACT_SUBREG V64:$Rn, hsub)`?
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D131700/new/
https://reviews.llvm.org/D131700
More information about the llvm-commits
mailing list