-
Notifications
You must be signed in to change notification settings - Fork 423
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
UCT/IB/MLX5: fixed compilation issue #2353
Conversation
hoopoepg
commented
Feb 27, 2018
- fixed compilation on ARM platforms
- fixed compilation on ARM platforms
Test PASSed. |
src/uct/ib/mlx5/ib_mlx5.inl
Outdated
n = wq->qend - dest; | ||
memcpy(dest, src, n); | ||
memcpy(wq->qstart, src + n, length - n); | ||
if (length) { |
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.
looks like extra branch on fast-path? can't we silence the compiler somehow or make it only for ARM?
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.
below there are conditions where length used, new condition should not affect to performance.
and original code is not valid: memcpy has unpredictable behavior on NULL ptr in zero-length.
so, look for ARM specific suppression?
Test PASSed. |
looks like gcc has hard time to propagate constants, this happens from uct_rc/dc_mlx5_ep_tag_eager_zcopy() only. where the inline copy function is not called. |
Test PASSed. |
Test FAILed. |
Test PASSed. |
bot:mlx:retest |
Test FAILed. |
Test PASSed. |
@yosefe tests are passed |