linux-kernel - Re: [linux-next:master] [VFS/nfsd/cachefiles/ovl] 7ab96df840: WARNING:at_fs/dcache.c:#umount

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <aSe6cWAoUpEWL44E@xsang-OptiPlex-9020>
Date: Thu, 27 Nov 2025 10:41:53 +0800
From: Oliver Sang <oliver.sang@...el.com>
To: Amir Goldstein <amir73il@...il.com>
CC: Christian Brauner <brauner@...nel.org>, NeilBrown <neil@...wn.name>, "Jeff
 Layton" <jlayton@...nel.org>, <oe-lkp@...ts.linux.dev>, <lkp@...el.com>,
	<netfs@...ts.linux.dev>, <linux-fsdevel@...r.kernel.org>,
	<linux-nfs@...r.kernel.org>, <linux-unionfs@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>, <oliver.sang@...el.com>
Subject: Re: [linux-next:master] [VFS/nfsd/cachefiles/ovl] 7ab96df840:
 WARNING:at_fs/dcache.c:#umount_check

hi, Amir,

On Wed, Nov 26, 2025 at 01:29:57PM +0100, Amir Goldstein wrote:
> On Wed, Nov 26, 2025 at 11:42 AM Christian Brauner <brauner@...nel.org> wrote:
> >
> > On Tue, Nov 25, 2025 at 09:48:18PM +0800, kernel test robot wrote:
> > >
> > > Hello,
> > >
> > > kernel test robot noticed "WARNING:at_fs/dcache.c:#umount_check" on:
> > >
> > > commit: 7ab96df840e60eb933abfe65fc5fe44e72f16dc0 ("VFS/nfsd/cachefiles/ovl: add start_creating() and end_creating()")
> > > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
> > >
> > > [test failed on linux-next/master d724c6f85e80a23ed46b7ebc6e38b527c09d64f5]
> >
> > Neil, can you please take a look at this soon?
> > I plan on sending the batch of PRs for this cycle on Friday.
> >
> > >
> > > in testcase: filebench
> > > version: filebench-x86_64-22620e6-1_20251009
> > > with following parameters:
> > >
> > >       disk: 1SSD
> > >       fs: ext4
> > >       fs2: nfsv4
> > >       test: ratelimcopyfiles.f
> > >       cpufreq_governor: performance
> > >
> 
> Test is copying to nfsv4 so that's the immediate suspect.
> WARN_ON is in unmount of ext4, but I suspect that nfs
> was loop mounted for the test.
> 
> FWIW, nfsd_proc_create() looks very suspicious.
> 
> nfsd_create_locked() does end_creating() internally (internal API change)
> but nfsd_create_locked() still does end_creating() regardless.
> 
> Oliver,
> 
> Can you test this handwritten change or need a patch/branch for testing:

thanks for the patch! but it cannot solved the issues we reported.
since we are now testing refined patch from NeilBrown, I won't supply more
detail such like dmesg here. if you still want it, please let us know. thanks!

> 
> diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c
> index 28f03a6a3cc38..35618122705db 100644
> --- a/fs/nfsd/nfsproc.c
> +++ b/fs/nfsd/nfsproc.c
> @@ -407,6 +407,7 @@ nfsd_proc_create(struct svc_rqst *rqstp)
>                 /* File doesn't exist. Create it and set attrs */
>                 resp->status = nfsd_create_locked(rqstp, dirfhp, &attrs, type,
>                                                   rdev, newfhp);
> +               goto out_write;
>         } else if (type == S_IFREG) {
>                 dprintk("nfsd:   existing %s, valid=%x, size=%ld\n",
>                         argp->name, attr->ia_valid, (long) attr->ia_size);
> 
> 
> Thanks,
> Amir.
>