* Missing patches from kallsyms buildid series in 6.18.y and 6.19.y
@ 2026-04-08 11:38 Marek Kroemeke
2026-04-08 12:55 ` Greg KH
0 siblings, 1 reply; 2+ messages in thread
From: Marek Kroemeke @ 2026-04-08 11:38 UTC (permalink / raw)
To: stable; +Cc: gregkh, sashal, pmladek, kernel-team
Hi,
Petr Mladek's series "kallsyms: Prevent invalid access when showing
module buildid" [1] was partially applied to the 6.18.y longterm and
6.19.y stable trees. Patches 3, 5, and 6 from the 7-patch series
landed:
acfdbb4ab291 ("module: add helper function for reading module_buildid()")
cd6735896d03 ("kallsyms/bpf: rename __bpf_address_lookup() to bpf_address_lookup()")
e8a1e7eaa19d ("kallsyms/ftrace: set module buildid in ftrace_mod_address_lookup()")
But patches 1, 2, 4, and 7 did not:
426295ef18c5 ("kallsyms: clean up @namebuf initialization in kallsyms_lookup_buildid()")
fda024fb6476 ("kallsyms: clean up modname and modbuildid initialization in kallsyms_lookup_buildid()")
8e81dac4cd54 ("kallsyms: cleanup code for appending the module buildid")
3b07086444f8 ("kallsyms: prevent module removal when printing module name and buildid")
Without the missing patches, __sprint_symbol() can use an
uninitialized or dangling mod->build_id pointer during backtrace
printing. We hit KASAN errors and stack protector failures due to this partial application.
Could these four commits please be queued for both trees?
[1] https://lore.kernel.org/all/20251128135920.217303-1-pmladek@suse.com/
Thanks,
Marek Kroemeke
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: Missing patches from kallsyms buildid series in 6.18.y and 6.19.y
2026-04-08 11:38 Missing patches from kallsyms buildid series in 6.18.y and 6.19.y Marek Kroemeke
@ 2026-04-08 12:55 ` Greg KH
0 siblings, 0 replies; 2+ messages in thread
From: Greg KH @ 2026-04-08 12:55 UTC (permalink / raw)
To: Marek Kroemeke; +Cc: stable, sashal, pmladek, kernel-team
On Wed, Apr 08, 2026 at 11:38:51AM +0000, Marek Kroemeke wrote:
> Hi,
>
> Petr Mladek's series "kallsyms: Prevent invalid access when showing
> module buildid" [1] was partially applied to the 6.18.y longterm and
> 6.19.y stable trees. Patches 3, 5, and 6 from the 7-patch series
> landed:
>
> acfdbb4ab291 ("module: add helper function for reading module_buildid()")
> cd6735896d03 ("kallsyms/bpf: rename __bpf_address_lookup() to bpf_address_lookup()")
> e8a1e7eaa19d ("kallsyms/ftrace: set module buildid in ftrace_mod_address_lookup()")
>
> But patches 1, 2, 4, and 7 did not:
> 426295ef18c5 ("kallsyms: clean up @namebuf initialization in kallsyms_lookup_buildid()")
> fda024fb6476 ("kallsyms: clean up modname and modbuildid initialization in kallsyms_lookup_buildid()")
> 8e81dac4cd54 ("kallsyms: cleanup code for appending the module buildid")
> 3b07086444f8 ("kallsyms: prevent module removal when printing module name and buildid")
>
> Without the missing patches, __sprint_symbol() can use an
> uninitialized or dangling mod->build_id pointer during backtrace
> printing. We hit KASAN errors and stack protector failures due to this partial application.
>
>
> Could these four commits please be queued for both trees?
> [1] https://lore.kernel.org/all/20251128135920.217303-1-pmladek@suse.com/
All now queued up thanks.
greg k-h
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-04-08 12:55 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-08 11:38 Missing patches from kallsyms buildid series in 6.18.y and 6.19.y Marek Kroemeke
2026-04-08 12:55 ` Greg KH
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox