On Wed, 6 Nov 2024, Roman Gushchin wrote:
On Tue, Nov 05, 2024 at 05:09:13PM -0800, Sean Christopherson wrote:
On Tue, Oct 22, 2024, Roman Gushchin wrote:
On Mon, Oct 21, 2024 at 12:49:28PM -0700, Hugh Dickins wrote:
On Mon, 21 Oct 2024, Roman Gushchin wrote: I don't think there's any need to change your text, but let me remind us that any "Bad page" report stops that page from being allocated again (because it's in an undefined, potentially dangerous state): so does amount to a small memory leak even if otherwise harmless.
It looks like I need to post v3 as soon as I get a publicly available syzkaller report, so I'll add this to the commit log.
Today is your lucky day :-)
I've been waiting for it for a long time :) Thanks for forwarding it my way!
I'm still not sure what the conclusion of our discussion was. My understanding is that my fix is not that pretty, but there are no better immediate ideas, only long-term improvement projects. Does it matches everybody else's understanding?
Yes, that matches my understanding, and my Acked-by stands: thanks a lot for keeping on this, Roman and Sean.
Hugh
If so, I'll prepare a v3 with an updated link. Otherwise, please, let me know.
Thanks!