Re: Bug related with a 6.6.24 platform/x86 commit signed by you - Enormous memory leak

[Thomas Weißschuh]

On 2024-07-19 17:40:59+0000, Max Dubois wrote:
> I’m sorry, yes that file i called ‘working’ contain multiple kernels instances becouae I had to do a path to find out the crash happened at 6.6.24 level and that the last working kernel was 6.6.23. So I did a try to a lot of kernels over and under 6.6.24 to isolate the problem to 6.6.24 and everything after.
> 
> 
> If needed I can try to isolate the instance of the working 6.6.23 and
> I can’t confirm it does NO ERRORS at all like the kernels before it.
> 
> Buggy 6.6.24 log it is corrrext, the real 6.6.24 log with buga on vmalloc.
> 
> If needed I can confirm I can test any possible fix and/or patch against 6.6.24.

Can you try to reproduce the issue with
1) 6.6.41
2) 6.10
3) https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/

If it is fixed in one of them it should also be fixed in the ones later
in the list.

It would be very unfortunate to investigate and fix a bug that has
already been fixed before.

> Thank you
> 
> MD
> Inviato da iPhone
> 
> > Il giorno 19 lug 2024, alle ore 17:12, Dan Carpenter <dan.carpenter@xxxxxxxxxx> ha scritto:
> > 
> > On Fri, Jul 19, 2024 at 10:01:14AM -0500, Dan Carpenter wrote:
> >> 
> >> The interesting thing about that is the working kernel had tons of these
> >> allocation failures as well.
> >> https://bugzilla.kernel.org/show_bug.cgi?id=219061
> >> See the attachment which called "This is a session with the last WORKING
> >> KERNEL 6.6.23, NO ERRORS, everything fine".
> > 
> > Never mind.  There are a bunch of different reboots in that file with
> > different kernels.
> > 
> > regards,
> > dan carpenter
> > 
>