On Mon, Aug 08, 2022 at 12:14:30PM -0300, Guilherme G. Piccoli wrote:

On 08/08/2022 02:07, Evan Green wrote:

quoted

On Tue, Jul 19, 2022 at 12:55 PM Guilherme G. Piccoli
[off-list ref] wrote:

quoted

Currently the gsmi driver registers a panic notifier as well as
reboot and die notifiers. The callbacks registered are called in
atomic and very limited context - for instance, panic disables
preemption and local IRQs, also all secondary CPUs (not executing
the panic path) are shutdown.

With that said, taking a spinlock in this scenario is a dangerous
invitation for lockup scenarios. So, fix that by checking if the
spinlock is free to acquire in the panic notifier callback - if not,
bail-out and avoid a potential hang.

Fixes: 74c5b31c6618 ("driver: Google EFI SMI")
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: David Gow <redacted>
Cc: Evan Green <redacted>
Cc: Julius Werner <jwerner@chromium.org>
Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com>

Reviewed-by: Evan Green <redacted>

Thanks a bunch Evan!

Ard / Greg, do you think you could get this patch through your -next (or
-fixes) trees? Not sure which tree is the most common for picking GSMI
stuff.

Picking out an individual patch from a series with as many responses and
threads like this one is quite difficult.

Just resend this as a stand-alone patch if you want it applied
stand-alone as our tools want to apply a whole patch series at once.

I'm trying to get these fixes merged individually in their trees to not
stall the whole series and increase the burden of re-submitting.

The burden is on the submitter, not the maintainer as we have more
submitters than reviewers/maintainers.

thanks,

greg k-h