-
Notifications
You must be signed in to change notification settings - Fork 33
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
New tunable --shm-metadata-msg-size
#435
Conversation
2613938
to
b1c8d8d
Compare
b1c8d8d
to
60fc7f1
Compare
5c9c611
to
6b508e1
Compare
6b508e1
to
6908616
Compare
476e4f8
to
2592e2c
Compare
fb01977
to
4cf36a0
Compare
2316af0
to
726873c
Compare
Thank you! |
a184677
to
732b932
Compare
732b932
to
a3057a9
Compare
@knopers8 I haven't been able to push a release until now. However, I think, with the latest update, this is close to become mergeable. Also, @rbx will be back soon and let's see what he is thinking. Since I am on a trip from tomorrow till sunday, I can only continue next Monday. If you have the chance to test this branch here, it would be appreciated. @rbx I intentionally removed some of the already merged api removal commits in |
@knopers8 Currently, there is no way to tune this per channel, only per device and applies to all sending shmem channels in that device. Is this granular enough? |
Yes, that's good enough for the cases I'd like to apply this. |
a3057a9
to
e903652
Compare
a364a10
to
4a73b31
Compare
Implement move semantics.
4a73b31
to
afcb0d4
Compare
The shm metadata msg will be right-padded to the given size. This tunable may be used to saturate the kernel msg buffers more quickly with the effect that the ZeroMQ message queue size - on which the FairMQ shmem transport relies upon - behaves more accurately for very small queue sizes. This introduces a change for the meta msg format in the multipart case: old: | MetaHeader 1 | ... | MetaHeader n | new: | n | MetaHeader 1 | ... | MetaHeader n | padded to fMetadataMsgSize | where `n` is a `size_t` and contains the number of following meta headers. Previously, this number was infered from the msg buffer size itself which is no longer possible due to the potential padding. Implements FairRootGroup#432
afcb0d4
to
dee0bc5
Compare
@rbx thx for the test coverage. I shortened the code even more and added the changes from what we discussed earlier today. |
Looks good! |
tests/docs and multipart send coming tomorrow, but should already be enough for a first round of discussion.
Implements #432