Re: Possible hungtask issue will be introduced with device_lock() in uevent_show()

[Greg KH]

On Sat, Jan 04, 2025 at 02:02:54PM +0800, zhangzekun (A) wrote:
> 
> 
> 在 2024/12/31 16:26, Greg KH 写道:
> > On Tue, Dec 31, 2024 at 03:56:08PM +0800, Zhang Zekun wrote:
> > > Hi, Dan, Greg,
> > > 
> > > We have found a potential tungtask issue has been introduce by commit 9a71892cbcdb ("Revert "driver core: Fix uevent_show() vs driver detach race""), which revert the rcu in device_uevent but reintroduce the device_lock() in uevent_show(). The reproduce procedure is quite simple:
> > 
> > The revert just puts the original logic back in place, so this is not
> > anything new that has been introduced, right?  It's just that the
> > attempted fix didn't work, so a different fix needs to happen.
> Hi, Greg,
> 
> Yes, there is nothing new introduced here. We have been testing the rcu fix
> (commit 15fffc6a5624 ("driver core: Fix uevent_show() vs driver detach
> race")) for monthes but has not obersved problems.

That's great, but other people did report problems, which can't be
ignored.

> Noticing that, brmails+k write in bugzilla "Also, I can't say why the issue
> appeared in the past even without this commit being present, as I haven't
> bisected any kernel version before v6.6.45.". I doubt that there could still
> have problem described in bugzilla [1] even if commit c0a40097f0bc
> ("drivers: core: synchronize really_probe() and dev_uevent()") has been
> reverted, and the problem is not directly introduced by rcu in dev_uevent.
> 
> [1] https://bugzilla.kernel.org/show_bug.cgi?id=219244

So you think that the bug report is wrong?  Ok, then you might need to
take the time to prove it :)

If you feel this should be added back, please feel free to resubmit it
with the new information as to why it should be ok now to apply it.

thanks,

greg k-h