device-libs: Replace MATH_SQRT f32 macro with function #880
Conversation
This is equivalent to the correctly rounded sqrt function, which is already available in __ocml_sqrt_f32. We can't just directly use the sqrt builtin because of the OpenCL default has lower precision, and there's currently no fine grained control for a specific operation. There's only the flag -cl-fp32-correctly-rounded-divide-sqrt applied to the translation unit where __ocml_sqrt_f32 is implemented. This also results in a 2 instruction savings in cacosh as one sample user.
|
Thank you for submitting a Pull Request (PR) to the LLVM Project! This PR will be automatically labeled and the relevant teams will be notified. If you wish to, you can add reviewers by using the "Reviewers" section on this page. If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers. If you have further questions, they may be answered by the LLVM GitHub User Guide. You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums. |
This is equivalent to the correctly rounded sqrt function, which is already available in __ocml_sqrt_f32. We can't just directly use the sqrt builtin because the OpenCL default has lower precision, and there's currently no fine grained control for a specific operation. There's only the flag -cl-fp32-correctly-rounded-divide-sqrt applied to the translation unit where __ocml_sqrt_f32 is implemented.
This also results in a 2 instruction savings in cacosh as one sample user.