[LU-12621] o2iblnd QP sizing -EINVAL retries

• Type: Improvement
• Resolution: Fixed
• Priority: Major
• Fix Version/s: Lustre 2.13.0, Lustre 2.12.3
• Assignee: Amir Shehata (Inactive)
• Reporter: Mike Marciniszyn

Description

The logic currently in Lustre that loops reducing the queue sizes is silly.

The ib_query_device() in kiblnd_hdev_get_attr() is done to save off the max_mr_size in the hdev.

The code could save off max_qp_wr and use that cap the send queue calculations to avoid creating a QP above that limit.

 

 

 

 

Comments

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/36253/
Subject: LU-12621 o2iblnd: cache max_qp_wr
Project: fs/lustre-release
Branch: b2_12
Current Patch Set:
Commit: 41acf6bdd51c7a015d61f746a82a9155c01d8596

[Gerrit Updater]

Minh Diep ([e-mail]) uploaded a new patch: https://review.whamcloud.com/36253
Subject: LU-12621 o2iblnd: cache max_qp_wr
Project: fs/lustre-release
Branch: b2_12
Current Patch Set: 1
Commit: e13ff349e6a1a59df377245f27f660de5771e709

[Peter Jones]

Landed for 2.13

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/36073/
Subject: LU-12621 o2iblnd: cache max_qp_wr
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7ee319ed7f9dfa365a66b20b03f2141c54fb0293

[Gerrit Updater]

Amir Shehata ([e-mail]) uploaded a new patch: https://review.whamcloud.com/36073
Subject: LU-12621 o2iblnd: cache max_qp_wr
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: e9f6d3c6cddb5a47a1af25b294634c17e3b66214