[PATCH] tests/file_handle.c: skip if fs can't do handles

[Dmitry V. Levin]

On Mon, May 03, 2021 at 10:42:07PM +0000, Alyssa Ross wrote:
> On Mon, May 03, 2021 at 05:06:52PM +0300, Dmitry V. Levin wrote:
> > On Mon, May 03, 2021 at 01:12:13PM +0000, Alyssa Ross wrote:
> > > Quoting name_to_handle_at(2):
> > >
> > > > Some care is needed here as EOVERFLOW can also indicate that no file
> > > > handle is available for this particular name in a filesystem which
> > > > does normally support file-handle lookup.  This case can be detected
> > > > when the EOVERFLOW error is returned without handle_bytes being
> > > > increased.
> > >
> > > This patch handles that case, resolving 20 failures for me on ZFS, and
> > > making `make check' able to run successfully.
> >
> > I must admit I'm a bit surprised that a filesystem supporting file-handle
> > lookups can fail to return a file handle for ".".
> 
> Good point.
> 
> I looked into it further, and it's a bug in ZFS.  I sent this fix:
> https://github.com/openzfs/zfs/pull/11995

Thanks.

> I feel like if that fix is accepted, it's probably not worth fixing in
> strace (even though the buggy behaviour is probably technically valid).
> What do you think?  If you still think it's worth fixing, I'm happy to
> send a v2 with your suggestion applied.

Since it's a strace test, its main purpose is to test strace rather than
the kernel, so I don't mind applying a workaround like this.


-- 
ldv