test/recvsend_bundle: enlarge recv buf-ring to 2 MiB to avoid ENOBUFS… #1453
+1
−1
Conversation
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
… on ppc64 Sub-test (d) "bundling with full socket" pre-fills the peer until EAGAIN. On some architectures (like ppc64), once the tcp_w/rmem min is set to higher value, like 64k and with autotuning on, the sender can queue more than the 1MiB before the receiver arms multishot recv. With a 1MiB provided-buffer pool (MSG_SIZE * RECV_BIDS, so 8192 * 128), the first recv runs out of provided buffers and fails with -105 (ENOBUFS). Bump RECV_BIDS from 8192 to 16384, so the pool no longer underflows during the prefill on ppc64. This is a minimal, test-only tweak to stabilize results across different arches without changing test logic. Signed-off-by: Sebastian Chlad <sebastian.chlad@suse.com>
Owner
|
I think that work-around is likely just fine. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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 is a minimal, test-only tweak to stabilize results across different arches without changing test logic.
IMHO, ideally we would find a way to allocate the buffer based on the system under test runtime socket SO_RCVBUF, so we would have a test without the hardcoded RECV_BIDS and thus the buffer size.
So while I do not have a ready patch/solution, I wanted to see what do you think about it.