Re: [PATCH] fs/namespace: correctly handle errors returned by grab_requested_mnt_ns

[Andrei Vagin]

On Sun, Nov 23, 2025 at 4:51 AM Mark Brown <broonie@xxxxxxxxxx> wrote:
>
> On Tue, Nov 11, 2025 at 06:28:15AM +0000, Andrei Vagin wrote:
> > grab_requested_mnt_ns was changed to return error codes on failure, but
> > its callers were not updated to check for error pointers, still checking
> > only for a NULL return value.
>
> I'm seeing regressions in mainline in the LTP listmount04 test on some
> arm64 platforms:
>
> tst_test.c:1953: TINFO: LTP version: 20250530
> tst_test.c:1956: TINFO: Tested kernel: 6.18.0-rc6-00270-g89edd36fd801 #1 SMP PREEMPT @1763835814 aarch64
> tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
> tst_test.c:1774: TINFO: Overall timeout per run is 0h 05m 24s
> listmount04.c:128: TPASS: request points to unaccessible memory : EFAULT (14)
> listmount04.c:128: TPASS: mnt_ids points to unaccessible memory : EFAULT (14)
> listmount04.c:128: TPASS: invalid flags : EINVAL (22)
> listmount04.c:128: TPASS: insufficient mnt_id_req.size : EINVAL (22)
> listmount04.c:128: TFAIL: invalid mnt_id_req.spare expected EINVAL: EBADF (9)
> listmount04.c:128: TPASS: invalid mnt_id_req.param : EINVAL (22)
> listmount04.c:128: TPASS: invalid mnt_id_req.mnt_id : EINVAL (22)
> listmount04.c:128: TPASS: non-existant mnt_id : ENOENT (2)
>
> which bisect to this patch.  I'm not sure if the change in error code
> here is actually a real issue or not, this feels like an overly
> sensitive test, but perhaps there's some spec requirement or something
> so it seemed reasonable to report.

The merged patch is slightly different from what you can see on the
mailing list, so it's better to look at commit 78f0e33cd6c93
("fs/namespace: correctly handle errors returned by
grab_requested_mnt_ns") to understand what is going on here.

With this patch, the spare field can be used as the `mnt_ns_fd`. EINVAL
is returned if both mnt_ns_fd and mnt_ns_id are set. A non-zero
mnt_ns_fd (the old spare) is interpreted as a namespace file descriptor.

Thanks,
Andrei