[LU-12254] lnet: correct discovery event queue release

• Status: Resolved
• Project: Lustre
• Fix Version/s: Lustre 2.13.0, Lustre 2.12.3
• Type: Technical task
• Priority: Minor
• Reporter: Amir Shehata (Inactive)
• Assignee: Amir Shehata (Inactive)
• Resolution: Fixed

Description

It is possibly for events to have a ref count on the event queue when it's released in the discovery thread. When shutting down we can see an error output:

LNetError: 2770:0:(api-ni.c:1004:lnet_res_container_cleanup()) 1 active elements on exit of EQ container
LNetError: 2807:0:(module.c:655:libcfs_exit()) Portals memory leaked: 304 bytes

The eq needs to be freed after all events have been processed.

Comments

[Gerrit Updater]

Amir Shehata ([e-mail]) uploaded a new patch: https://review.whamcloud.com/34796
Subject: LU-12254 lnet: correct discovery LNetEQFree()
Project: fs/lustre-release
Branch: multi-rail
Current Patch Set: 1
Commit: ea1ec208b908863bc2fa803df4837aee04a33e13

[Gerrit Updater]

Amir Shehata ([e-mail]) merged in patch https://review.whamcloud.com/34796/
Subject: LU-12254 lnet: correct discovery LNetEQFree()
Project: fs/lustre-release
Branch: multi-rail
Current Patch Set:
Commit: a0879b5985b41f92dede96e7f27623eb72102b15

[Joseph Gmitter (Inactive)]

Work has landed as part of the MR Routing merge commit: https://review.whamcloud.com/#/c/34983/

[Gerrit Updater]

Minh Diep ([e-mail]) uploaded a new patch: https://review.whamcloud.com/36036
Subject: LU-12254 lnet: correct discovery LNetEQFree()
Project: fs/lustre-release
Branch: b2_12
Current Patch Set: 1
Commit: f7de523811c1259fb5faa7b6d47a2033cafd3124

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/36036/
Subject: LU-12254 lnet: correct discovery LNetEQFree()
Project: fs/lustre-release
Branch: b2_12
Current Patch Set:
Commit: 9c2d72073319ccffe39bdb72f22d79e4a43499ab