[Codegen][GPU] Let integer range optimization narrow GPU computations to i32 #19473
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
benvanik
reviewed
Dec 12, 2024
krzysz00
force-pushed
the
index-narrowing
branch
3 times, most recently
from
be116ef to
ffa5fc4
Compare
krzysz00
requested review from
antiagainst,
MaheshRavishankar,
kuhar,
qedawkins and
Groverkss
as code owners
krzysz00
force-pushed
the
index-narrowing
branch
2 times, most recently
from
01736ba to
a367857
Compare
qedawkins
reviewed
Jan 7, 2025
| ArrayRef<ArrayAttr> castAssumptions = ArrayRef( | ||
| static_cast<const ArrayAttr *>(assumptions.data()), assumptions.size()); | ||
| auto newOp = rewriter.create<Util::AssumeIntOp>(op.getLoc(), newArgs, | ||
| castAssumptions); |
There was a problem hiding this comment.
Can you not just do rewriter.create<Util::AssumeIntOp>(op.getLoc(), newArgs, op.getAssumptions());
If not we can either just clone with new operands or add a new builder.
There was a problem hiding this comment.
There's a builder missing - and no really room to insert one
There was a problem hiding this comment.
wdym there is no room to add a new builder? Can't we add a builder that take ArrayRef<Value> and ArrayAttr as operands?
There was a problem hiding this comment.
... Yeah, might not be a bad idea, could add one.
compiler/src/iree/compiler/Dialect/Util/Transforms/OptimizeIntArithmetic.cpp
Outdated
Show resolved
Hide resolved
compiler/src/iree/compiler/Dialect/Util/Transforms/OptimizeIntArithmetic.cpp
Outdated
Show resolved
Hide resolved
krzysz00
force-pushed
the
index-narrowing
branch
from
a367857 to
09d7c2d
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Outdated suggestions cannot be applied.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.
Note: This PR is stacked on top of #19372, and so looks bigger than it is. The relevant changes are in the last commit.
Add an option to -iree-util-optimize-int-arithmetic to have it perform computations in i32 where possible, which is enabled when optimizing arithmetic for GPU codegen. This allows LLVM co correctly conclude that various computations don't need to be done at full 64-bit precision, thus saving registers and instructions. (LLVM has some rewrites for this, but they're, for example, gated on only having one use of the potentially-truncated value, which means that shared math stays in an over-wide data type).