[PATCH] D119469: [AArch64] Turn truncating buildvectors into truncates

Tue Feb 15 02:51:12 PST 2022

david-arm added inline comments.

================
Comment at: llvm/lib/Target/AArch64/AArch64ISelLowering.cpp:9269
+// concats and truncate nodes.
+static SDValue ReconstructTruncateFromBuildVector(SDValue V, SelectionDAG &DAG) {
+  if (V.getValueType() != MVT::v16i8)
----------------
dmgreen wrote:
> david-arm wrote:
> > Hi @dmgreen, this feels a bit *too* specific to a particular optimisation. For example, there is a test in CodeGen/AArch64/neon-extracttruncate.ll called `@extract_2_v4i32` that could also benefit from something like this. Also, some of the code in here looks very similar to what ReconstructShuffle attempts to do, which also looks at BUILD_VECTORS that are constructed from extractelement operations. Would it be better to simply extend ReconstructShuffle to cover this case?
> ReconstructShuffle is for reconstructing buildvectors into Legal Shuffles, and if you look at the code only handles 2 sources. This method is for reconstructing a truncates, not shuffles, and needs to handle 4 sources. As this isn't a legal shuffle, it doesn't really fit there.
> The tests in neon-extracttruncate.ll I just added as a quick test. They are not expected to come up in practice. https://godbolt.org/z/eEser71Gq. It is the lowering of large fptosi.sat that we are targeting.
OK, but is it worth making this more generic then than it currently is? i.e. dealing with more cases than just this one very specific issue? @extract_2_v4i32 looks like it could benefit too.

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D119469/new/

https://reviews.llvm.org/D119469