Re: [linux-next:master] [fs/mnt_idmapping.c] b4291c7fd9: xfstests.generic.645.fail

Re: [linux-next:master] [fs/mnt_idmapping.c] b4291c7fd9: xfstests.generic.645.fail

[Darrick J. Wong]

On Tue, Feb 20, 2024 at 09:57:30AM +0100, Christian Brauner wrote:
> On Mon, Feb 19, 2024 at 02:55:42PM +0800, kernel test robot wrote:
> > 
> > 
> > Hello,
> > 
> > kernel test robot noticed "xfstests.generic.645.fail" on:
> > 
> > commit: b4291c7fd9e550b91b10c3d7787b9bf5be38de67 ("fs/mnt_idmapping.c: Return -EINVAL when no map is written")
> > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
> 
> The test needs to be updated. We now explicitly fail when no map is
> written.

Has there been any progress on updating generic/645?  6.9-rc1 is out,
and Dave and I have both noticed this regressing.

--D

> > 
> > [test failed on linux-next/master d37e1e4c52bc60578969f391fb81f947c3e83118]
> > 
> > in testcase: xfstests
> > version: xfstests-x86_64-c46ca4d1-1_20240205
> > with following parameters:
> > 
> > 	disk: 4HDD
> > 	fs: f2fs
> > 	test: generic-645
> > 
> > 
> > 
> > compiler: gcc-12
> > test machine: 4 threads Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz (Skylake) with 32G memory
> > 
> > (please refer to attached dmesg/kmsg for entire log/backtrace)
> > 
> > 
> > 
> > 
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > | Reported-by: kernel test robot <oliver.sang@intel.com>
> > | Closes: https://lore.kernel.org/oe-lkp/202402191416.17ec9160-oliver.sang@intel.com
> > 
> > 2024-02-17 15:04:19 export TEST_DIR=/fs/sda1
> > 2024-02-17 15:04:19 export TEST_DEV=/dev/sda1
> > 2024-02-17 15:04:19 export FSTYP=f2fs
> > 2024-02-17 15:04:19 export SCRATCH_MNT=/fs/scratch
> > 2024-02-17 15:04:19 mkdir /fs/scratch -p
> > 2024-02-17 15:04:19 export SCRATCH_DEV=/dev/sda4
> > 2024-02-17 15:04:19 export MKFS_OPTIONS=-f
> > 2024-02-17 15:04:19 echo generic/645
> > 2024-02-17 15:04:19 ./check generic/645
> > FSTYP         -- f2fs
> > PLATFORM      -- Linux/x86_64 lkp-skl-d03 6.8.0-rc1-00033-gb4291c7fd9e5 #1 SMP PREEMPT_DYNAMIC Sat Feb 17 22:11:35 CST 2024
> > MKFS_OPTIONS  -- -f /dev/sda4
> > MOUNT_OPTIONS -- -o acl,user_xattr /dev/sda4 /fs/scratch
> > 
> > generic/645       [failed, exit status 1]- output mismatch (see /lkp/benchmarks/xfstests/results//generic/645.out.bad)
> >     --- tests/generic/645.out	2024-02-05 17:37:40.000000000 +0000
> >     +++ /lkp/benchmarks/xfstests/results//generic/645.out.bad	2024-02-17 15:07:42.613312168 +0000
> >     @@ -1,2 +1,4 @@
> >      QA output created by 645
> >      Silence is golden
> >     +idmapped-mounts.c: 6671: nested_userns - Invalid argument - failure: sys_mount_setattr
> >     +vfstest.c: 2418: run_test - Invalid argument - failure: test that nested user namespaces behave correctly when attached to idmapped mounts
> >     ...
> >     (Run 'diff -u /lkp/benchmarks/xfstests/tests/generic/645.out /lkp/benchmarks/xfstests/results//generic/645.out.bad'  to see the entire diff)
> > Ran: generic/645
> > Failures: generic/645
> > Failed 1 of 1 tests
> > 
> > 
> > 
> > 
> > The kernel config and materials to reproduce are available at:
> > https://download.01.org/0day-ci/archive/20240219/202402191416.17ec9160-oliver.sang@intel.com
> > 
> > 
> > 
> > -- 
> > 0-DAY CI Kernel Test Service
> > https://github.com/intel/lkp-tests/wiki
> > 
> 

Re: [linux-next:master] [fs/mnt_idmapping.c] b4291c7fd9: xfstests.generic.645.fail

[Christian Brauner]

On Mon, Mar 25, 2024 at 09:58:09AM -0700, Darrick J. Wong wrote:
> On Tue, Feb 20, 2024 at 09:57:30AM +0100, Christian Brauner wrote:
> > On Mon, Feb 19, 2024 at 02:55:42PM +0800, kernel test robot wrote:
> > > 
> > > 
> > > Hello,
> > > 
> > > kernel test robot noticed "xfstests.generic.645.fail" on:
> > > 
> > > commit: b4291c7fd9e550b91b10c3d7787b9bf5be38de67 ("fs/mnt_idmapping.c: Return -EINVAL when no map is written")
> > > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
> > 
> > The test needs to be updated. We now explicitly fail when no map is
> > written.
> 
> Has there been any progress on updating generic/645?  6.9-rc1 is out,
> and Dave and I have both noticed this regressing.

Iirc, Taylor wanted to fix this but it seems that hasn't happened yet.
I'll ping again and if nothing's happened until tomorrow I'll send a
patch.

Re: [linux-next:master] [fs/mnt_idmapping.c] b4291c7fd9: xfstests.generic.645.fail

[Darrick J. Wong]

On Tue, Mar 26, 2024 at 12:43:27PM +0100, Christian Brauner wrote:
> On Mon, Mar 25, 2024 at 09:58:09AM -0700, Darrick J. Wong wrote:
> > On Tue, Feb 20, 2024 at 09:57:30AM +0100, Christian Brauner wrote:
> > > On Mon, Feb 19, 2024 at 02:55:42PM +0800, kernel test robot wrote:
> > > > 
> > > > 
> > > > Hello,
> > > > 
> > > > kernel test robot noticed "xfstests.generic.645.fail" on:
> > > > 
> > > > commit: b4291c7fd9e550b91b10c3d7787b9bf5be38de67 ("fs/mnt_idmapping.c: Return -EINVAL when no map is written")
> > > > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
> > > 
> > > The test needs to be updated. We now explicitly fail when no map is
> > > written.
> > 
> > Has there been any progress on updating generic/645?  6.9-rc1 is out,
> > and Dave and I have both noticed this regressing.
> 
> Iirc, Taylor wanted to fix this but it seems that hasn't happened yet.
> I'll ping again and if nothing's happened until tomorrow I'll send a
> patch.

Ok, glad to hear that this is still on your radar.  Thank you for
following up!

--D

Re: [PATCH 2/2] vfs/idmapped_mounts.c: Change mount_setattr expected output

[Christian Brauner]

On Tue, Mar 26, 2024 at 08:33:52PM +0000, Taylor Jackson via B4 Relay wrote:
> From: Taylor Jackson <tjackson9431@gmail.com>
> 
> In kernel commit dacfd001eaf2 (“fs/mnt_idmapping.c: Return -EINVAL
> when no map is written”), the behavior of mount_setattr changed to
> return EINVAL when attempting to create an idmapped mount when using
> a user namespace with no mappings. The following commit updates the test
> to expect no mount to be created in that case. And since no mount is created,
> this commit also removes the check for overflow IDs because it does not make
> sense to check for overflow IDs for a mount that was not created.
> 
> Signed-off-by: Taylor Jackson <tjackson9431@gmail.com>
> ---

Thanks for fixing this!
Reviewed-by: Christian Brauner <brauner@kernel.org>

>  src/vfs/idmapped-mounts.c | 7 +------
>  1 file changed, 1 insertion(+), 6 deletions(-)
> 
> diff --git a/src/vfs/idmapped-mounts.c b/src/vfs/idmapped-mounts.c
> index 34052ca3..f4dfc3f3 100644
> --- a/src/vfs/idmapped-mounts.c
> +++ b/src/vfs/idmapped-mounts.c
> @@ -6667,7 +6667,7 @@ static int nested_userns(const struct vfstest_info *info)
>  	}
>  
>  	if (sys_mount_setattr(fd_open_tree_level4, "", AT_EMPTY_PATH,
> -			      &attr_level4, sizeof(attr_level4))) {
> +			      &attr_level4, sizeof(attr_level4)) != -1 || errno != EINVAL) {
>  		log_stderr("failure: sys_mount_setattr");
>  		goto out;
>  	}
> @@ -6706,11 +6706,6 @@ static int nested_userns(const struct vfstest_info *info)
>  			log_stderr("failure: check ownership %s", file);
>  			goto out;
>  		}
> -
> -		if (!expected_uid_gid(fd_open_tree_level4, file, 0, info->t_overflowuid, info->t_overflowgid)) {
> -			log_stderr("failure: check ownership %s", file);
> -			goto out;
> -		}
>  	}
>  
>  	/* Verify that ownership looks correct for callers in the first userns. */
> 
> -- 
> 2.34.1
> 
> 

On Tue, Mar 26, 2024 at 08:22:28AM -0700, Darrick J. Wong wrote:
> On Tue, Mar 26, 2024 at 12:43:27PM +0100, Christian Brauner wrote:
> > On Mon, Mar 25, 2024 at 09:58:09AM -0700, Darrick J. Wong wrote:
> > > On Tue, Feb 20, 2024 at 09:57:30AM +0100, Christian Brauner wrote:
> > > > On Mon, Feb 19, 2024 at 02:55:42PM +0800, kernel test robot wrote:
> > > > > 
> > > > > 
> > > > > Hello,
> > > > > 
> > > > > kernel test robot noticed "xfstests.generic.645.fail" on:
> > > > > 
> > > > > commit: b4291c7fd9e550b91b10c3d7787b9bf5be38de67 ("fs/mnt_idmapping.c: Return -EINVAL when no map is written")
> > > > > https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
> > > > 
> > > > The test needs to be updated. We now explicitly fail when no map is
> > > > written.
> > > 
> > > Has there been any progress on updating generic/645?  6.9-rc1 is out,
> > > and Dave and I have both noticed this regressing.
> > 
> > Iirc, Taylor wanted to fix this but it seems that hasn't happened yet.
> > I'll ping again and if nothing's happened until tomorrow I'll send a
> > patch.
> 
> Ok, glad to hear that this is still on your radar.  Thank you for
> following up!

@Darrick, Taylor sent fixes for this now (I've took the liberty to
respond to both mails combined.).