-
Notifications
You must be signed in to change notification settings - Fork 801
[SYCL] Enable FTZ operations for CUDA/PTX backend via -fcuda-flush-denormals-to-zero #6411
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
| case options::OPT_fgpu_flush_denormals_to_zero: | ||
| DenormalFPMath = llvm::DenormalMode::getPositiveZero(); | ||
| DenormalFP32Math = llvm::DenormalMode::getPositiveZero(); | ||
| break; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you add a driver test verifying the expect values when the options are set?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@mdtoguchi Do you prefer a separate test for this, or should I add it to the clang/test/Driver/fp-model.c?
|
Please add a CodeGen test for this. |
|
@andykaylor, FYI. |
mdtoguchi
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK for driver
|
@premanandrao Please review this PR. |
@premanandrao is OOO next week . I see he has reviewed the patch earlier and requested a CodeGen test which hasn't been added yet |
|
@elizabethandrews @againull This is my bad. I'll add a test case next week. |
|
@elizabethandrews @againull I just added a CodeGen test case. |
premanandrao
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
FE changes look good to me.
|
@intel/llvm-gatekeepers, this looks ready to go. |
Gives a small performance benefit for Gromacs on NVidia targets