port updated version of llvm float div #531
Merged
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
knickish
force-pushed
the
float_div_subnormal_rounding
branch
2 times, most recently
from
96bf09e
to
5563410
Compare
|
Thanks! Regarding ARM, I know the the ARM SIMD instructions always flush subnormals to zero, so that might be what is happening here. The normal scalar FP instructions properly support subnormals though. I think it's fine to leave it under a |
knickish
force-pushed
the
float_div_subnormal_rounding
branch
from
5563410
to
71b5f58
Compare
|
Okay, added a comment next the |
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Outdated suggestions cannot be applied.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.
This pr updates the float division code to match more recent versions of LLVM's implementations from compiler-rt, adding proper handling of subnormal values. I tested this by fuzzing all f32 values and an equivalent number of f64s against hard float operations on my computer (AMD zen3 cpu) in addition to updating the existing tests in the repo.
There seem to be very slight differences against the
armtarget, not sure if there's a better approach than just ignoring subnormals on that arch.