Performance loss on SLES 12 SP5 after updating to kernel 4.12.14-122.150 or higher
This document (000021233) is provided subject to the disclaimer at the end of this document.
Environment
Situation
This issue is not expected to affect SLES 15 SPx.
Resolution
For SLES 12 SP5 kernel-default 4.12.14-122.179
For kernel-azure: 4.12.14-16.152
For kernel-rt: 4.12.14-10.144
Cause
commit da5bd87e56640166a5a4d3053687c02b135bd5bd
Date: Tue Feb 7 08:32:03 2023 -0800
Revert "scsi: core: run queue if SCSI device queue isn't ready
and queue is idle" (git-fixes).
However, the corresponding upstream commit b4fd63f4264 mentions this revert in scsi core should be done in tandem with 2 other commits in blk-mq, which were not present at that time:
ab3cee376 blk-mq: In blk_mq_dispatch_rq_list() "no budget" is a reason to kick
a0823421a blk-mq: Rerun dispatching in the case of budget contention
These have now been added and performance is back to normal.
Status
Disclaimer
This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.
- Document ID:000021233
- Creation Date: 10-Oct-2023
- Modified Date:10-Oct-2023
-
- SUSE Linux Enterprise Server
For questions or concerns with the SUSE Knowledgebase please contact: tidfeedback[at]suse.com