We are seeing deadlock in cifs code while updating volume in
cifs_reconnect. There are few fixes available in stable trees
already. This series backports some patches back to 5.4 stable.
__schedule+0x268/0x6e0
schedule+0x2f/0xa0
schedule_preempt_disabled+0xa/0x10
__mutex_lock.isra.7+0x20b/0x470
? dfs_cache_update_vol+0x45/0x2a0 [cifs]
dfs_cache_update_vol+0x45/0x2a0 [cifs]
cifs_reconnect+0x6f2/0xef0 [cifs]
cifs_handle_standard+0x18d/0x1b0 [cifs]
cifs_demultiplex_thread+0xa5c/0xc90 [cifs]
? cifs_handle_standard+0x1b0/0x1b0 [cifs]
Paulo Alcantara (SUSE) (5):
cifs: Clean up DFS referral cache
cifs: Get rid of kstrdup_const()'d paths
cifs: Introduce helpers for finding TCP connection
cifs: Merge is_path_valid() into get_normalized_path()
cifs: Fix potential deadlock when updating vol in cifs_reconnect()
fs/cifs/dfs_cache.c | 701 +++++++++++++++++++++++---------------------
1 file changed, 372 insertions(+), 329 deletions(-)
--
2.40.1
Hi Pablo,
While checking netfilter backports to the stable series, I noticed
that 6e1acfa387b9 ("netfilter: nf_tables: validate registers coming
from userspace.") was backported in various series for stable, and
included in 4.14.316, 4.19.284, 5.4.244, 5.15.32, 5.16.18, 5.17.1,
where the original fix was in 5.18-rc1.
While the commit has
Fixes: 49499c3e6e18 ("netfilter: nf_tables: switch registers to 32 bit addressing")
the 6e1acfa387b9 change got not backported to the 5.10.y series.
The backports to the other series are
https://lore.kernel.org/stable/20230516151606.4892-1-pablo@netfilter.org/https://lore.kernel.org/stable/20230516150613.4566-1-pablo@netfilter.org/https://lore.kernel.org/stable/20230516144435.4010-1-pablo@netfilter.org/
Pablo, was this an oversight and can the change as well be applied to
5.10.y?
From looking at the 5.4.y series, from the stable dependencies,
08a01c11a5bb ("netfilter: nftables: statify nft_parse_register()") is
missing in 5.10.y, then 6e1acfa387b9 ("netfilter: nf_tables: validate
registers coming from userspace.") can be applied (almost, the comment
needs to be dropped, as done in the backports).
I'm right now not understanding what I'm missing that it was for 5.4.y
but not 5.10.y after the report of the failed apply by Greg.
At least the two attached bring 5.10.y inline with 5.4.y up to 4) from
https://lore.kernel.org/stable/20230516144435.4010-1-pablo@netfilter.org/
but I'm unsure if you want/need as well the remaining 5), 6), 7), 8)
and 9).
Regards,
Salvatore
Linux 6.2 and newer are (mostly) unbootable on my old HP 6730b laptop, the 6.1.30 works still fine.
The weirdest thing is that newer kernels (like 6.3.4 and 6.4-rc3) may boot ok on the first try, but when rebooting, the very same version doesn't boot.
Some times, when trying to boot, I get this message repeated forever:
ACPI Error: No handler or method for GPE [XX], disabling event (20221020/evgpe-839)
On newer kernels, the date is 20230331 instead of 20221020. There is also some other error, but I can't read it as it gets overwritten by the other ACPI error, see image linked at the end.
And some times, the screen will just stay completely blank.
I tried booting with acpi=off, but it does not help.
I bisected and this is the first bad commit 7e68dd7d07a2
"Merge tag 'net-next-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next"
As the later kernels had the seemingly random booting behaviour (mentioned above), I retested the last good one 7c4a6309e27f by booting it several times and it boots every time.
I tried getting some boot logs, but the boot process does not go far enough to make any logs.
Kernel .config file: https://0x0.st/Hqt1.txt
Environment (outputs of a working Linux 6.1 build):
Software (output of the ver_linux script): https://0x0.st/Hqte.txt
Processor information (from /proc/cpuinfo): https://0x0.st/Hqt2.txt
Module information (from /proc/modules): https://0x0.st/HqtL.txt
/proc/ioports: https://0x0.st/Hqt9.txt
/proc/iomem: https://0x0.st/Hqtf.txt
PCI information ('lspci -vvv' as root): https://0x0.st/HqtO.txt
SCSI information (from /proc/scsi/scsi)
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: ATA Model: KINGSTON SVP200S Rev: C4
Type: Direct-Access ANSI SCSI revision: 05
Host: scsi1 Channel: 00 Id: 00 Lun: 00
Vendor: hp Model: CDDVDW TS-L633M Rev: 0301
Type: CD-ROM ANSI SCSI revision: 05
Distribution: Arch Linux
Boot manager: systemd-boot (UEFI)
git bisect log: https://0x0.st/Hqgx.txt
ACPI Error (sorry for the dusty screen): https://0x0.st/HqEk.jpeg
#regzbot ^introduced 7e68dd7d07a2
Best regards
Sami Korkalainen