Skip to content

Update drivers/base to v5.18

Mark Langsdorf requested to merge mlangsdorf/centos-stream-9:bz2067284 into main

Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2067284
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2067252
Omitted-fix: 68af28426b3ca1bf9ba21c7d8bdd0ff639e5134c
Requires substantial amd-pmc backport commits that are not in this MR or the centos-9 codebase yet.
Omitted-fix: b1f66033cd4e9ce8cbe2c74c98d4e04c0b2d7b40
Requires substantial amd-pmc backport commits that are not in this MR or the centos-9 codebase yet.
Omitted-fix: 753ee989f7cf0c0a76a7f56956827a8863a60f97
Fix for the previous omitted fix.
Omitted-fix: 32370191c0851da069d242f581cbe2fdb80040cb
Same fix as "platform/x86: amd-pmc: Set QOS during suspend on CZN w/ timer wakeup" but with a different commit hash

Update drivers/base subsystem to match Linux v5.18, including v5.17, v5.16, and v5.15. In general, all the commits applied cleanly, as would be expected with applying those commits to a v5.14 base.

A bunch of memory hotplug and memblock commits tangentially touched drivers/base in from v5.17 or v5.18, but David Hildebrand wrote them and can backport them better than I can, so I don't consider them to really be drivers/base commits. I pulled in a few commits that don't strictly touch drivers/base to finish out Luis Chamberlain's series on reworking the firmware loader.

A half-dozen MSI related commits that touched drivers/base in v5.17 were not backported, as they were part of a roughly 30 commit series that would be better handled by the PCI subsystem team.

Signed-off-by: Mark Langsdorf mlangsdo@redhat.com

Edited by Mark Langsdorf

Merge request reports