[0/3] sysfs: Group visibility fixups

Message

Dan Williams Feb. 22, 2024, 8:40 p.m. UTC

Hey Greg,

Marc was able to get me a backtrace for that bootup hang scenario that
Pierre noted here [1]. A Tested-by is still pending, but I am certain
this is the issue, and it may impact more than just the one platform if
that "empty_attrs" pattern has been repeated anywhere else in the
kernel.

I also took some time to document how to use the helpers a bit better,
and introduce DEFINE_SIMPLE_SYSFS_GROUP_VISIBLE() for cases where all
that matters is group visibility and not per attribute visibility.

[1]: http://lore.kernel.org/r/[e-mail]

---

Dan Williams (3):
      sysfs: Fix crash on empty group attributes array
      sysfs: Document new "group visible" helpers
      sysfs: Introduce DEFINE_SIMPLE_SYSFS_GROUP_VISIBLE()


 fs/sysfs/group.c      |    4 +-
 include/linux/sysfs.h |   85 +++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 84 insertions(+), 5 deletions(-)

base-commit: 70317fd24b419091aa0a6dc3ea3ec7bb50c37c32

Comments

Greg KH Feb. 23, 2024, 6:33 a.m. UTC | #1

On Thu, Feb 22, 2024 at 12:40:48PM -0800, Dan Williams wrote:
> Hey Greg,
> 
> Marc was able to get me a backtrace for that bootup hang scenario that
> Pierre noted here [1]. A Tested-by is still pending, but I am certain
> this is the issue, and it may impact more than just the one platform if
> that "empty_attrs" pattern has been repeated anywhere else in the
> kernel.
> 
> I also took some time to document how to use the helpers a bit better,
> and introduce DEFINE_SIMPLE_SYSFS_GROUP_VISIBLE() for cases where all
> that matters is group visibility and not per attribute visibility.
> 
> [1]: http://lore.kernel.org/r/[e-mail]

I'll just queue these up now on my normal driver-core-next branch, and
not worry about a stable tag as I doubt anyone wants that now.  But if
they do, please let me know and I can provide one.

thanks for the quick fixes!

greg k-h