fix(vpto): add SIMT fastmath control - #1138
Merged
Merged
Conversation
A5 板测失败
失败用例
|
A3 板测失败
失败用例
|
A3 板测失败详情:PR #1138ci
gather_legacy
|
This file contains hidden or 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
--simt-fastmath, disabled by default.--cce-simt-fpmath-combine=falseto Bisheng by default and enable it only when explicitly requested.Validation
ptoas --simt-fastmath --helpandptoas --simt-fastmath=false --helpsucceed.--simt-fastmathmodes.fmulandfaddoperations.Scope
This PR addresses the implicit SIMT
mulf + addfcontraction described in issue #1117. The scalar high-precisionfdivquestion remains open: CANN 9.0.0 exposes high-precision division only through vectorbisheng::cce::simd::hp::vdiv/ PTODiv754-style helpers; the candidate scalar LLVM intrinsics are not selectable on the current A5 Bisheng target.Refs #1117