Skip to content
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/UD: Fix issue N544 (CREQ assert) #953

Merged
merged 2 commits into from
Sep 11, 2016

Conversation

brminich
Copy link
Contributor

@brminich brminich commented Sep 8, 2016

Fixed:

  • If CREQ and CREP to peer are lost, peer may stay unconnected
  • type == UCT_UD_PACKET_CREQ assertion

fixes #544

Fixed:
 - If CREQ and CREP to peer are lost, peer may stay unconnected
 - type == UCT_UD_PACKET_CREQ assertion
@mellanox-github
Copy link
Contributor

Test PASSed.
See http://bgate.mellanox.com/jenkins/job/gh-ucx-pr/2337/ for details.

@mellanox-github
Copy link
Contributor

Test FAILed.
See http://hpc-master.lab.mtl.com:8080/job/hpc-ucx-pr/1667/ for details (Mellanox internal link).

@yosefe yosefe added the Bugfix label Sep 8, 2016
@brminich
Copy link
Contributor Author

brminich commented Sep 8, 2016

@alex-mikheev, could you please look thru?

@mellanox-github
Copy link
Contributor

Test FAILed.
See http://hpc-master.lab.mtl.com:8080/job/hpc-ucx-pr/1668/ for details (Mellanox internal link).

@@ -578,6 +579,9 @@ void uct_ud_ep_process_rx(uct_ud_iface_t *iface, uct_ud_neth_t *neth, unsigned b
goto out;
}
if (neth->packet_type & UCT_UD_PACKET_FLAG_CTL) {
/* No need to track duplications, CREP always goes
* with ACK_REQ flag */
ep->rx.ooo_pkts.head_sn = neth->psn;
uct_ud_ep_rx_ctl(iface, ep, (uct_ud_ctl_hdr_t *)(neth + 1));
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it think it is better to add this line in uct_ud_ep_rx_ctl()

@mellanox-github
Copy link
Contributor

Test FAILed.
See http://hpc-master.lab.mtl.com:8080/job/hpc-ucx-pr/1672/ for details (Mellanox internal link).

@mellanox-github
Copy link
Contributor

Test PASSed.
See http://bgate.mellanox.com/jenkins/job/gh-ucx-pr/2341/ for details.

@mellanox-github
Copy link
Contributor

Test PASSed.
See http://hpc-master.lab.mtl.com:8080/job/hpc-ucx-pr/1674/ for details (Mellanox internal link).

@alex-mikheev
Copy link
Contributor

👍

@yosefe yosefe merged commit eda2c58 into openucx:master Sep 11, 2016
yosefe added a commit to yosefe/ucx that referenced this pull request Sep 11, 2016
 Fix issue introduced by openucx#953 - when getting duplicate CREP with lower
psn, the current expected psn should not be affected. Otherwise packets
already acknowledged would be considered missing.
yosefe added a commit to yosefe/ucx that referenced this pull request Sep 11, 2016
 Fix issue introduced by openucx#953 - when getting duplicate CREP with lower
psn, the current expected psn should not be affected. Otherwise packets
already acknowledged would be considered missing.
@brminich brminich deleted the topic/uct_ud_fix_conns branch January 31, 2017 17:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

uct_ud_ep_rx_creq error at np 1280
4 participants