[RISCV] Lower constant build_vectors with few non-sign bits via vsext #65648
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
I'm not sure what ZSEXT in the title is supposed to mean if we only consider SEXT? |
I'd meant that to be VSEXT. As in, the instruction. Will fix the typo. |
preames
changed the title
topperc
reviewed
Sep 7, 2023
| @@ -3456,6 +3456,27 @@ static SDValue lowerBuildVectorOfConstants(SDValue Op, SelectionDAG &DAG, | |||
| if (SDValue Res = lowerBuildVectorViaDominantValues(Op, DAG, Subtarget)) | |||
| return Res; | |||
|
|
|||
| // IF the number of signbits allows, see if we can lower as a <N x i8>. | |||
There was a problem hiding this comment.
If the capitalization of 'F' in 'If' intentional?
If we have a build_vector such as [i64 0, i64 3, i64 1, i64 2], we instead lower this as vsext([i8 0, i8 3, i8 1, i8 2]). For vectors with 4 or fewer elements, the resulting narrow vector can be generated via scalar materialization. For shuffles which get lowered to vrgathers, constant build_vectors of small constants are idiomatic. As such, this change covers all shuffles with an output type of 4 or less. I deliberately started narrow here. I think it makes sense to expand this to longer vectors, but we need a more robust profit model on the recursive expansion. It's questionable if we want to do the zsext if we're going to generate a constant pool load for the narrower type anyways. One possibility for future exploration is to allow the narrower VT to be less than 8 bits. We can't use vsext for that, but we could use something analogous to our widening interleave lowering with some extra shifts and ands.
preames
force-pushed
the
pr-20230907-riscv-constant-build-vector-4x-sext
branch
from
1b54578
to
105311b
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
If we have a build_vector such as [i64 0, i64 3, i64 1, i64 2], we instead lower this as vsext([i8 0, i8 3, i8 1, i8 2]). For vectors with 4 or fewer elements, the resulting narrow vector can be generated via scalar materialization.
For shuffles which get lowered to vrgathers, constant build_vectors of small constants are idiomatic. As such, this change covers all shuffles with an output type of 4 or less.
I deliberately started narrow here. I think it makes sense to expand this to longer vectors, but we need a more robust profit model on the recursive expansion. It's questionable if we want to do the zsext if we're going to generate a constant pool load for the narrower type anyways.
One possibility for future exploration is to allow the narrower VT to be less than 8 bits. We can't use vsext for that, but we could use something analogous to our widening interleave lowering with some extra shifts and ands.