UCT/UGNI: Spinlocks around all of the GNI_* functions #1546
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
|
Test FAILed. |
|
Test PASSed. |
|
bot:bgate:retest |
|
Test PASSed. |
|
how well it was tested ? |
|
@shamisp There are no dead locks so far. There are a few places that still aren't thread safe (mpools and unaligned zcopy code). I'm building up more threading work in OpenSHMEM so it will be well exercised. |
|
👍 |
shamisp
approved these changes
May 30, 2017
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Outdated suggestions cannot be applied.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 completes most of the work to make the UGNI UCT layer thread safe. All GNI calls are protected with a spin lock attached to the device and remove the pthread mutex from being visible outside of ugni_md. It's still in ugni_md to protect md creation before the spin lock has been initialized.
As part of this effort the cq code has been factored out into it's own code path with error reporting and locks. This fixes #1529.
SMSG also does the right thing down, locking mbox access and bouncing threads that try and process an iface mbox while another thread is processing it.