Re: [PATCH v2 4/4] fstests: add generic test for file handles

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Apr 20, 2017 at 10:21 AM, Eryu Guan <eguan@xxxxxxxxxx> wrote:
> On Thu, Apr 20, 2017 at 09:34:48AM +0300, Amir Goldstein wrote:
> ..
>> >> +# Check non-stale handles to linked files
>> >> +rm -f $testdir/*
>> >> +src/open_by_handle -c $testdir $numfiles
>> >> +src/open_by_handle    $testdir $numfiles
>> >> +
>> >> +# Check non-stale handles to files that were hardlinked and original deleted
>> >> +src/open_by_handle -l $testdir $numfiles
>> >> +src/open_by_handle -u $testdir $numfiles
>> >
>> > This last test still depends on test files created in the second test
>> > implicitly. Forgot to update this part?
>>
>> Intentional.
>> It's the intended usage to create a test set with -c
>> (imagine this is going to done on lower layer)
>> then link the test set with -l (imagine this on either lower or overlay)
>> then unlink the test set with -u (imagine this on overlay).
>
> Yeah, that makes more sense in an overlay-specific test & operating on
> multiple layers. But I don't see the point in this generic test by doing
> so, as all these operations are clearly done on the same layer. So I
> think adding "rm -f" and "-c" run in this test won't change test
> behavior, but could avoid the implicit dependency.
>

You are right. It wont change the test.

> Anyway, at least this is worth some comments on the implicit dependency.
>

I'll take the easier choice ;-)

>>
>> So as a rule of thumb, rm -f is only added as extra safety
>> before a src/open_by_handle -c line
>>
>> >
>> > # Check non-stale handles to files that were hardlinked and original deleted
>> > rm -f $testdir/*
>> > src/open_by_handle -c $testdir $numfiles
>> > src/open_by_handle -l $testdir $numfiles
>> > src/open_by_handle -u $testdir $numfiles
>> >
>> >> +
>> >> +echo "Silence is golden"
>> >> +status=$?
>> >
>> > $? is always 0 after echo. And any test failure could break golden
>> > image, I think status can be set to 0 unconditionally.
>> >
>>
>> Indeed. a combination of copy&paste from xfs/238 and me moving
>> Silence is golden to end. Let me know if you want me to re-post for this
>
> I can fix this status issue, but seems you need to re-post anyway to add
> more comments or empty $testdir before each test :)
>
> Thanks,
> Eryu
--
To unsubscribe from this list: send the line "unsubscribe linux-unionfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Filesystems Devel]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux