Comment 23 for bug 1913395
Revision history for this message
| bugproxy (bugproxy) wrote Comment bridged from LTC Bugzilla | #23 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
------- Comment From <email address hidden> 2021-02-11 09:37 EDT-------
No problem, I suspected this was the case but wanted to be sure.
So, I've run verification on both focal and groovy using the qemu available in -proposed. With the caveat that for focal I used Frank's kernel build provided in 1907421 as we're still waiting for the backport from upstream stable on that for focal.
On both focal and groovy, I created a guest with a vfio-pci passthrough mlx device on s390x and forced data-transfer workloads that could, prior to this fix, quickly and reliably overrun the DMA space and result in transfer stalls / QEMU error messages. No more errors -- looks good!