Re: [PATCH v1 4/4] x86/vmware: Run panic diagnostics before kdump by default

["Guilherme G. Piccoli" <gpiccoli@igalia.com>]

On 28/09/2026 14:35, Stephen Brennan wrote:
> [...]
> Hi Guilherme, thanks for CCing me! You're right, I'm interested too :D
> Your email arrived on the first day of my vacation, so I'm sorry for the
> delayed response.

Hi Stephen! Nice talking to you =)
No worries at all, this is also conference season heh

> [...] 
> I do think we're going to have trouble articulating exactly what are the
> criteria for a "pre_kdump" notifier vs a "panic" notifier? Anyone
> submitting a patch believes their notifier is important, so of course
> they believe it belongs on the pre_kdump list! How will we draw the line
> in the future?
> 
> - Is it safety? As in "the notifier doesn't take any locks"? I feel
>   this should be part of the criteria. (Though I have never been
>   successful in predicting what code would cause a kdump/panic failure)
> - Is it urgency? As in "without this, the expected behavior of the
>   system during a panic will not occur".
> - Is it the fact that the notifier is communicating with the hypervisor?
> 
> From my perspective: informing the hypervisor of a panic is both safe
> and urgent. If the hypervisor is configured to immediately halt and do
> its own panic handling, great.
> 

This is a very good point. What is the criteria? I honestly don't
know...I think hypervisor-related things, that are safe (as you put)
makes sense. Anything that might also make kdump work more reliably, and
is safe as well...but everyone will judge their notifier important, in
the end heh


> Providing the last 4KB of logs seems less safe and less urgent (though
> obviously still immensely helpful). The kernel provides a huge variety
> of consoles through which the hypervisor could have already received
> these logs. If the system is configured not to use them... it seems the
> tradeoff to not be debuggable has already been made :P
> To me, that logic belongs in a standard panic notifier.
> 
> Plenty of people will likely disagree with me on that, but that's why it
> would be helpful to have clarity on what separates pre_kdump from
> standard panic notifiers.
> 

But at the same time ... there's something else, the "self-contained"
nature of hypervisor-related notifiers. This one is restricted for
vmware only, so personally, I think it's not a super biggie to run it
early. It is, in the end, a decision of vmware stakeholders. If that
would be an arch notifier, something that all arm64 users would
necessarily run, for example, I'd be way more skeptic on putting it early.

Anyway, not sure if I helped or if I contributed nothing to the criteria
discussion heheh

But I hope to see this discussion happening, maybe should also gets
documented as part of future Zach's patch iteration as well.
Cheers,


Guilherme