Re: [PATCH v1] mm/gup: remove (VM_)BUG_ONs

[Lorenzo Stoakes <lorenzo.stoakes@oracle.com>]

On Wed, Jun 04, 2025 at 08:42:30AM -0700, Linus Torvalds wrote:
[snip]

> I really think that the *ONLY* situation where BUG() is valid is when
> you absolutely *know* that corruption will happen, and you cannot
> continue.
>
> Very much *not* some kind of "this is problematic, and who knows what
> corruption it might cause".  But "I *know* I can't continue without
> major system because the hardware is broken sh*t".
>
> In other words, don't use it. Ever. Unless you can explain exactly why
> without any handwaving.

Thanks, this aligns with my understanding.

This does make VM_BUG_ON_xxx() look even more silly :) so I think we definitely
need to get rid of that...

'Absolutely definitely corruption but we only when CONFIG_DEBUG_VM is set' is
you know, insane.

>
> Cloud providers or others can do "panic-on-warn" if they want to stop
> the machine at the first sign of trouble.
>
>                   Linus

Yeah, I have seen people object to WARN_ON()'s because of this though 'hey
some people might panic here!!'. My view on that is - right, they can,
that's fine, they asked for it :)

Cheers, Lorenzo