Hi Luis, Christophe,

In addition to the previous iteration, since this particular series does
indeed modify last_unloaded_module, I decided to use strscpy() as a
replacement for the now deprecated strlcpy().

Changes since v1 [1][2]:

  - Replaced the deprecated strlcpy() for strscpy()
  - Replaced last_unloaded_module[] with an anonymous structure
    i.e. last_unloaded_module.name and last_unloaded_module.taints
  - Ensured we modify last_unloaded_module.taints only when required

[1]: https://lore.kernel.org/lkml/20220627164052.2416485-1-atomlin@redhat.com/
[2]: https://lore.kernel.org/lkml/20220627164052.2416485-2-atomlin@redhat.com/


Aaron Tomlin (3):
  module: Modify module_flags() to accept show_state argument
  module: Use strscpy() for last_unloaded_module
  module: Show the last unloaded module's taint flag(s)

 kernel/module/internal.h |  2 +-
 kernel/module/main.c     | 27 ++++++++++++++++++---------
 kernel/module/procfs.c   |  2 +-
 3 files changed, 20 insertions(+), 11 deletions(-)

-- 
2.34.3

On Thu, Jul 14, 2022 at 04:39:30PM +0100, Aaron Tomlin wrote:
> Hi Luis, Christophe,
> 
> In addition to the previous iteration, since this particular series does
> indeed modify last_unloaded_module, I decided to use strscpy() as a
> replacement for the now deprecated strlcpy().

Nice, looks super clean now, applied and pushed to modules-next, thanks!
BTW since you and Christophe seem to contribute a lot to modules lately,
any chance for future stuff you can ask 0day folks to add your trees
and branches to get tested prior to posting patches? Then you can
suggest in your cover letter they've been blessed by 0-day or whatever.

Typically I push to modules-testing, wait and if no complaints come back
then I push to modules-next but in this case I just went to
modules-next directly as I couldn't see how this could break.

  Luis