Re: [PATCH] docs: ABI: sysfs-module: list all taint flags

[Petr Pavlu]

On 10/16/25 12:13 AM, Randy Dunlap wrote:
> The list of module taint flags has not been updated lately as the
> taint flags list grows. Instead of trying to keep multiple lists
> updated, just refer to the list of kernel taint flags since they are
> the same.
> 
> Signed-off-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
> ---
> Cc: Jonathan Corbet <corbet@xxxxxxx>
> Cc: linux-doc@xxxxxxxxxxxxxxx
> Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> Cc: Luis Chamberlain <mcgrof@xxxxxxxxxx>
> Cc: Petr Pavlu <petr.pavlu@xxxxxxxx>
> Cc: Daniel Gomez <da.gomez@xxxxxxxxxx>
> Cc: Sami Tolvanen <samitolvanen@xxxxxxxxxx>
> Cc: linux-modules@xxxxxxxxxxxxxxx
> ---
>  Documentation/ABI/testing/sysfs-module        |   10 ++--------
>  Documentation/admin-guide/tainted-kernels.rst |    2 ++
>  2 files changed, 4 insertions(+), 8 deletions(-)
> 
> --- linux-next-20251014.orig/Documentation/ABI/testing/sysfs-module
> +++ linux-next-20251014/Documentation/ABI/testing/sysfs-module
> @@ -52,14 +52,8 @@ What:		/sys/module/*/taint
>  Date:		Jan 2012
>  KernelVersion:	3.3
>  Contact:	Kay Sievers <kay.sievers@xxxxxxxx>
> -Description:	Module taint flags:
> -			==  =====================
> -			P   proprietary module
> -			O   out-of-tree module
> -			F   force-loaded module
> -			C   staging driver module
> -			E   unsigned module
> -			==  =====================
> +Description:	Module taint flags: same as the kernel taint flags.
> +		See: :ref:`taint_flags` in Documentation/admin-guide/tainted-kernels.rst

The module taint flags that can appear in /sys/module/*/taint are
a subset of the kernel taint flags. By looking at the calls to
add_taint_module(), they are as follows:

Present:
TAINT_PROPRIETARY_MODULE
TAINT_OOT_MODULE
TAINT_FORCED_MODULE
TAINT_CRAP
TAINT_UNSIGNED_MODULE

Missing:
TAINT_LIVEPATCH
TAINT_TEST

+ potentially TEST_AUX.

Since this text specifically documents what can appear in
/sys/module/*/taint, I think we should still maintain a list of these
flags for accuracy.

Additionally, Documentation/admin-guide/tainted-kernels.rst provides
taint descriptions for the kernel as a whole, which can be misleading
for individual modules. For instance, for TAINT_LIVEPATCH, the document
says "kernel has been live patched", but in the context of
/sys/module/*/taint, it means "this is a livepatch module".

-- 
Thanks,
Petr