Re: [PATCH 6.3 000/694] 6.3.2-rc2 review

[Greg Kroah-Hartman]

On Wed, May 10, 2023 at 09:58:06PM +1000, Rudi Heitbaum wrote:
> On Wed, 10 May 2023 at 19:09, Greg Kroah-Hartman
> <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> >
> > On Wed, May 10, 2023 at 06:29:23PM +1000, Rudi Heitbaum wrote:
> > > On Wed, 10 May 2023 at 17:25, Greg Kroah-Hartman
> > > <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > > >
> > > > On Tue, May 09, 2023 at 02:58:06PM +0000, Rudi Heitbaum wrote:
> > > > > On Tue, May 09, 2023 at 03:56:42PM +0200, Greg Kroah-Hartman wrote:
> > > > > > On Tue, May 09, 2023 at 01:10:32PM +0000, Rudi Heitbaum wrote:
> > > > > > > On Tue, May 09, 2023 at 08:06:58AM +0000, Rudi Heitbaum wrote:
> > > > > > > > On Tue, May 09, 2023 at 05:26:44AM +0200, Greg Kroah-Hartman wrote:
> > > > > > > > > This is the start of the stable review cycle for the 6.3.2 release.
> > > > > > > > > There are 694 patches in this series, all will be posted as a response
> > > > > > > > > to this one.  If anyone has any issues with these being applied, please
> > > > > > > > > let me know.
> > > > > > > > >
> > > > > > > > > Responses should be made by Thu, 11 May 2023 03:05:05 +0000.
> > > > > > > > > Anything received after that time might be too late.
> > > > > > > >
> > > > > > > > Hi Greg,
> > > > > > > >
> > > > > > > > 6.3.2-rc2 tested.
> > > > > > >
> > > > > > > Hi Greg,
> > > > > > >
> > > > > > > Further testing and have seen ntfs3: NULL pointer dereference with ntfs_lookup errors
> > > > > > > with 6.3.2-rc2 (I have not seen this error before.) No other errors in the logs.
> > > > > >
> > > > > > Can you reproduce this without the extern, gpl-violation module loaded?
> > > > > >
> > > > > > thanks,
> > > > > >
> > > > > > greg k-h
> > > > >
> > > > > Hi Greg,
> > > > >
> > > > > I dropped the bcm_sta and recompiled and commented out the i915.guc=3
> > > > > and was able to reproduce.
> > > > >
> > > > > [   84.745080] BUG: kernel NULL pointer dereference, address: 0000000000000020
> > > > > [   84.746239] #PF: supervisor read access in kernel mode
> > > > > [   84.747599] #PF: error_code(0x0000) - not-present page
> > > > > [   84.748929] PGD 0 P4D 0
> > > > > [   84.750240] Oops: 0000 [#1] SMP NOPTI
> > > > > [   84.751575] CPU: 2 PID: 3176 Comm: .NET ThreadPool Not tainted 6.3.2-rc2 #1
> > > > > [   84.752998] Hardware name: Intel(R) Client Systems NUC12WSKi7/NUC12WSBi7, BIOS WSADL357.0085.2022.0718.1739 07/18/2022
> > > > > [   84.754474] RIP: 0010:ntfs_lookup+0x76/0xe0 [ntfs3]
> > > >
> > > > And do you get this same crash on ntfs3 on 6.4-rc1?  Is this a new
> > > > regression, or does it also show up on 6.3.1?
> > >
> > > Tested with 6.3.1 during the day today. No errors, and had been
> > > running 6.3.1 with no issue. Retested with 6.3.2-rc2 and problem
> > > immediately evident. So yes - I believe a regression.
> > >
> > > I have built and am now testing 6.4.0-rc1 this evening - no errors so far.
> > >
> > > [    0.000000] Linux version 6.4.0-rc1 (docker@1ccd349e2545)
> > > (x86_64-libreelec-linux-gnu-gcc-13.1.0 (GCC) 13.1.0, GNU ld (GNU
> > > Binutils) 2.40) #1 SMP Wed May 10 07:51:37 UTC 2023
> > >
> > > > And ntfs, ick, why?  And .NET?  What a combination...
> > >
> > > Joys of media players. Test device gets to test exfat, ntfs3, .NET,
> > > and throw in a compile host/GHA runner to put it through paces.
> >
> > Yeah, this should work.  Thanks for verifying this works on other
> > releases.  Any chance you can do 'git bisect' to track down the
> > offending commit?  In looking things over, I don't see anything
> > obvious...
> 
> Hi Greg,
> 
> I can confirm the offending commit in 6.3.2-rc2 is
> 
> bf11fd528a97 fs/ntfs3: Fix null-ptr-deref on inode->i_op in ntfs_lookup()

Thanks!  Odd that this didn't show up for you on the other stable -rc
releases, as that commit is also in those trees.

I'll go revert this for now and ask the maintainer to send a fixed
version.

thanks,

greg k-h