Re: [PATCH v5 2/2] x86/MCE/AMD: Fix memory leak when `threshold_create_bank()` fails

[Borislav Petkov]

On Mon, Mar 28, 2022 at 11:12:53AM +0700, Ammar Faizi wrote:
> Although, I am not sure if that 100% guarantees mce_threshold_remove_device()
> will not mess up with the interrupt (e.g. freeing the data while the interrupt
> reading it), unless we're using RCU stuff.
> 
> What do you think?

I would've said it doesn't matter but that thresholding device creation
is part of hotplug and it can happen multiple times even *after* the
interrupt vector has been set during setup so a potential teardown and
concurrent thresholding interrupt firing might really hit in a not fully
initialized/cleaned up state so yeah, let's do Yazen's thing.

The alternative would be the temporarily re-assign mce_threshold_vector
to default_threshold_interrupt while setup is being done but that's not
really necessary atm.

But call that helper function __threshold_remove_device().

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette