Ext4 data corruption in stable kernels

Posted Dec 10, 2023 22:35 UTC (Sun) by ballombe (subscriber, #9523)
In reply to: Ext4 data corruption in stable kernels by rolexhamster
Parent article: Ext4 data corruption in stable kernels

The problem is that the time to run any meaningful QA on the kernel is far longer than the time between to two stable kernel releases.

Ext4 data corruption in stable kernels: automated QA

Posted Dec 11, 2023 0:06 UTC (Mon) by geofft (subscriber, #59789) [Link]

Is there an element of inherent wall-clock time here, or is it parallelizable to the point where it can run quickly enough if someone were to throw a couple racks of physical machines or a large public cloud budget at the problem?

(Someone like the big companies that the stable kernels are in large part made for and by?)

Ext4 data corruption in stable kernels

Posted Dec 11, 2023 6:52 UTC (Mon) by vegard (subscriber, #52330) [Link]

In this case the problem was actually picked up by QA on November 24 and reported about 12 hours after the patches were posted on the mailing list, and that is also why it got dropped from 5.15 and older kernels: https://lore.kernel.org/stable/81a11ebe-ea47-4e21-b5eb-53...