Commit Graph

811 Commits

Author SHA1 Message Date
Stoiko Ivanov
dd539d4154 d/control.in: Provide versioned linux-image/linux-headers
pve-headers-$(uname -r) is equivalent to
linux-headers-$(uname -r)-amd64

pve-kernel-$(uname -r) is equivalent to
linux-image-$(uname -r)-amd64

By adding a provides this should help users running
`apt install linux-headers-$(uname -r)-amd64` which is commonly
suggested in install instructions for third-party kernel-drivers on
plain debian.

Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2021-12-14 13:31:17 +01:00
Fabian Grünbichler
bc1d191389 KConfig: disable module BTF debug info
enabling this effectively makes the debug info of the kernel image part
of the kernel <-> module ABI[0], potentially breaking module loading for
otherwise compatible kernel/module combinations.

there were several user reports[1,2,3,..], and although a workaround
exists (re-installing the currently booted version, loading the module,
then upgrading again) disabling this until it's clear how to proceed
seems like the way to go.

disabling via patch is needed, since leaving it unset via `-d` or
explicitly disabled via `--set-val` doesn't work - it's auto-enabled
based on build-env pahole supporting split BTF generation.

0: https://lore.kernel.org/all/1637926692.uyvrkty41j.astroid@nora.none/
1: https://forum.proxmox.com/threads/realtek-8156-2-5gbe-usb-nic.69903/#post-433793
2: https://forum.proxmox.com/threads/usb-storage-can-not-be-mounted.100480/
3: https://forum.proxmox.com/threads/neuer-server-und-probleme.99098/#post-427862

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Acked-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Reviewed-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-12-14 13:31:05 +01:00
Fabian Grünbichler
7c4bae5665 d/control: add zstd to B-D
used for compressing the kernel image, build fails if not installed.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Acked-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Reviewed-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-12-14 13:31:00 +01:00
Thomas Lamprecht
892a5d3219 update ABI file for 5.15.5-1-pve
(generated with debian/scripts/abi-generate)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-30 16:42:03 +01:00
Thomas Lamprecht
f6163ca84c add fwlist from 5.15.5
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-30 16:37:47 +01:00
Thomas Lamprecht
4be20d430b bump version to 5.15.5-1
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-30 15:56:37 +01:00
Thomas Lamprecht
74962917c2 switch from impish to jelly as base ubuntu kernel
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-30 15:55:43 +01:00
Thomas Lamprecht
6c95b2e31c update ABI file for 5.13.19-2-pve
(generated with debian/scripts/abi-generate)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-29 18:35:46 +01:00
Thomas Lamprecht
899fd8d0aa bump version to 5.13.19-4
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-29 18:09:38 +01:00
Thomas Lamprecht
3ea6393f63 rebase patches on top of Ubuntu-5.13.0-23.23
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-29 18:09:38 +01:00
Thomas Lamprecht
dbe708a919 update sources to Ubuntu-5.13.0-23.23
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-29 18:09:38 +01:00
Thomas Lamprecht
fda70d3865 drop oboslete ocfs2 patch
now upstreamed in a slightly different way

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-29 18:09:38 +01:00
Thomas Lamprecht
92e753f360 bump version to 5.13.19-3
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-23 13:31:41 +01:00
Fabian Ebner
25aba572ca Backport two io-wq fixes relevant for io_uring
There were quite a few reports in the community forum about Windows
VMs with SATA disks not working after upgrading to kernel 5.13.
Issue was reproducible during the installation of Win2019 (suggested
by Thomas), and it's already fixed in 5.15. Bisecting led to
    io-wq: split bounded and unbounded work into separate lists
as the commit fixing the issue.

Indeed, the commit states
    Fixes: ecc53c48c13d ("io-wq: check max_worker limits if a worker transitions bound state")
which is present as a backport in ubuntu-impish:
    f9eb79f840052285408ae9082dc4419dc1397954

The first backport
    io-wq: fix queue stalling race
also sounds nice to have and additionally served as a preparation for
the second one to apply more cleanly.

Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
2021-11-23 13:19:53 +01:00
Thomas Lamprecht
43414bd219 bump version to 5.13.19-2
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-09 13:27:05 +01:00
Thomas Lamprecht
38c898a1b6 Revert "block: remove the -ERESTARTSYS handling in blkdev_get_by_dev"
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2021-11-09 12:56:54 +01:00
Thomas Lamprecht
db84324aab update ABI file for 5.13.19-1-pve
(generated with debian/scripts/abi-generate)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-06 14:28:58 +01:00
Thomas Lamprecht
1f9eca1c7a update firmware list for 5.13.19-1-pve
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-06 14:06:00 +01:00
Thomas Lamprecht
4e5a919c13 bump version to 5.13.19-1
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-06 13:08:45 +01:00
Thomas Lamprecht
46951972bc rebase patches on top of Ubuntu-5.13.0-22.22
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-06 13:07:52 +01:00
Thomas Lamprecht
23caf6a578 update sources to Ubuntu-5.13.0-22.22
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-11-06 13:07:52 +01:00
Thomas Lamprecht
7956d7d5b2 update ABI file for 5.13.18-1-pve
(generated with debian/scripts/abi-generate)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-10-19 14:13:11 +02:00
Thomas Lamprecht
7d04290406 bump version to 5.13.18-1
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-10-19 13:43:23 +02:00
Thomas Lamprecht
4fce12053e rebase patches on top of Ubuntu-5.13.0-21.21
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-10-19 13:40:56 +02:00
Thomas Lamprecht
7ceb3b1224 update sources to Ubuntu-5.13.0-21.21
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-10-19 13:40:56 +02:00
Thomas Lamprecht
68145b3f84 backport "ocfs2: mount fails with buffer overflow in strlen"
we've got good feedback from customers and the patch got two R-b's
and no NAK or the like upstream:

https://lore.kernel.org/all/20210929180654.32460-1-vvidic@valentin-vidic.from.hr/t/

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-10-18 14:04:36 +02:00
Fabian Ebner
fb9edd57ac remove outdated io_uring patch
The patch is not needed anymore, because the fix is already in
ubuntu-impish (commit d0b69849e40b2c3582f1cd6573f8e0d3a033d078).
Unfortunately, the patch still applied (in the wrong place), making it
hard to notice.

Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
2021-09-30 17:41:47 +02:00
Thomas Lamprecht
68481eb746 delete hirsute submodule directory entry
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-28 07:12:11 +02:00
Thomas Lamprecht
5d60271071 readme: general updates
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-28 07:08:46 +02:00
Thomas Lamprecht
5e8e351552 readme: refer to impish kernel
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-28 07:08:37 +02:00
Thomas Lamprecht
aaf2b2c31b update README
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-28 07:08:07 +02:00
Thomas Lamprecht
c8bbfe0d9f bump version to 5.13.14-1
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-28 06:17:00 +02:00
Thomas Lamprecht
58d428de99 update ZFS submodule to 2.1.1
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-28 06:16:40 +02:00
Thomas Lamprecht
9ac8d8ec39 update ABI file for 5.13.14-1-pve
(generated with debian/scripts/abi-generate)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-27 13:31:29 +02:00
Thomas Lamprecht
3edc172ef3 update fwlist-previous
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-27 13:31:12 +02:00
Thomas Lamprecht
df6168ebe5 buildsys: switch base over to 5.13 branched Ubuntu Impish kernel
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-27 13:30:16 +02:00
Thomas Lamprecht
2445fd0873 update patches for Ubuntu-5.13.0-17.17
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-27 12:43:06 +02:00
Thomas Lamprecht
4f243ed063 add Ubuntu Impish submodule
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-27 09:59:32 +02:00
Thomas Lamprecht
112453be49 bump version to 5.11.22-9
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-22 10:57:10 +02:00
Thomas Lamprecht
b77e82cce8 d/rules: kconfig: enable SQUASHFS_DECOMP_MULTI_PERCPU
Debian did so since 5.10~rc7-1~exp1 and ubuntu only disabled it due
some concerns about "high" memory usage on many-core systems[0], high
is to be seen relative here as its 26 MiB on 208 cores[1] and only
matters for ubuntu as due to their snaps they may have a lot of
active squashfs mounts.

Proxmox projects do not use snaps, or other things that uses squashfs
instances a tall besides the installer. While some users may use a
few it is unlikely to cause much problems (a few 100 MiB should not
be a big problem on a server with hundreds of online cores.

Any how, to speed up decompression in our installer and use a similar
setting as Debian, the distro we're most similar too, enable this
Kconfig knob.

[0]: https://bugs.launchpad.net/snappy/+bug/1636847
[1]: https://bugs.launchpad.net/snappy/+bug/1636847/comments/21

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-22 10:03:32 +02:00
Thomas Lamprecht
d6a491197c backport "blk-mq: fix kernel panic during iterating over flush request"
see https://forum.proxmox.com/threads/96598/#post-418247

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-09-21 07:15:18 +02:00
Thomas Lamprecht
84a5f7dcd9 update ABI file for 5.11.22-4-pve
(generated with debian/scripts/abi-generate)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-27 15:47:08 +02:00
Thomas Lamprecht
6d33e79b29 bump version to 5.11.22-8
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-27 11:52:31 +02:00
Thomas Lamprecht
88ca3a12ab update fwlist-previous
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-27 11:52:31 +02:00
Thomas Lamprecht
85fafb9d66 d/rules: kconfig: disable revocation certificates for now
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-27 10:15:18 +02:00
Thomas Lamprecht
c058ed8e33 rebase patches on top of Ubuntu-5.11.0-34.36
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-27 09:20:57 +02:00
Thomas Lamprecht
92518c5228 update sources to Ubuntu-5.11.0-34.36
(generated with debian/scripts/import-upstream-tag)

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-27 09:20:57 +02:00
Fabian Grünbichler
a7de27ff7d fix #3552: cherry-pick PCI probe fixes
breaking some NVME setups. these should be picked up by one of the next
Ubuntu kernel releases, since both the breaking change and the fix are
authored by Canonical devs.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2021-08-26 14:51:27 +02:00
Thomas Lamprecht
f4ddb1493e make nvem a module again
Requested in the forum, as some (misbehaving) mellanox module relies
on loading nvme-core and apparently cannot cope with built-in
modules... meh, but we're probably easier to get to make the change
than the mellanox people ;-)
https://forum.proxmox.com/threads/kernel-5-11.86225/post-412645

Originally set when this was another repo:
https://git.proxmox.com/?p=pve-kernel-jessie.git;a=commitdiff;h=0817f603d0e5bc3165f512eec8860543430cfd2d;hp=60c76fe4f5ab178af78be9ca2ea8baad3b4a4bc9

The Debian bug is resolved since quite a bit, bullseye shouldn't be
affected by this at all anymore.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2021-08-24 14:02:51 +02:00
Stoiko Ivanov
3ae5a3f4e0 bump to 5.11.22-7
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
2021-08-19 08:49:30 +02:00