Re: fs/vfs/renamerace* tests

[Paul Goyette <paul%whooppee.com@localhost>]

On Sat, 12 Jul 2014, Andreas Gustafsson wrote:

> Paul Goyette wrote:
> > The test code still indicates that these tests are expected to fail for
> > udf, however the tests are actually passing successfully (at least in my
> > amd64 test-bed). [1]
>
> I asked Reinoud about this about a year ago, and he said they were
> failing for him, "with EINVAL and ENOSPC".

Hmmm.  My test bed runs with a fairly large "disk" (to accomodate the 
fs/cd9660/t_high_ino_big_file test) so it could be avoiding the ENOSPC
condition.

A quick check through historical data shows udf_renamerace failing with 
EINVAL until approximately May 29th, which is around the time that I 
increased the disk size.  However, I see no failures related to the 
udf_renameracedirs test, for as long as the expect_fail has been set.

> I think the current behavior where the tests runs are complaining to
> those who cannot reproduce the bug, but not complaining to those who
> can, is not helpful.  My suggestion is to remove the atf_tc_expect_fail()
> calls in case, and if this causes errors to be reported for someone,
> then that person should file a PR.

I'd be happy to remove the atf_tc_expect_fail() if there are no 
objections.


-------------------------------------------------------------------------
| Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:       |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com    |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net |
| Kernel Developer |                          | pgoyette at netbsd.org  |
-------------------------------------------------------------------------