[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20131014024126.GA24871@thunk.org>
Date: Sun, 13 Oct 2013 22:41:26 -0400
From: Theodore Ts'o <tytso@....edu>
To: Robert Yang <liezhi.yang@...driver.com>
Cc: linux-ext4@...r.kernel.org, dvhart@...ux.intel.com
Subject: Re: [RFC 01/10] mke2fs.c: add an option: -d root-directory
On Wed, Aug 28, 2013 at 01:25:51PM +0800, Robert Yang wrote:
> @@ -2773,7 +2776,6 @@ no_journal:
> "filesystem accounting information: "));
> checkinterval = fs->super->s_checkinterval;
> max_mnt_count = fs->super->s_max_mnt_count;
> - retval = ext2fs_close(fs);
> if (retval) {
> fprintf(stderr,
> _("\nWarning, had trouble writing out superblocks."));
You can't just move the call to ext2fs_close(). You also need to move
if (!quiet)
printf(_("Writing superblocks and "
"filesystem accounting information: "));
before the call to ext2fs_close() since this is used to print the
message for the progress information that will be emitted by
ext2fs_close(), and you also have to move the error checking:
if (retval) {
fprintf(stderr,
_("\nWarning, had trouble writing out superblocks."));
...
after the call to ext2fs_close().
This would have been ***painfully*** obvious if you had run the
regression test suite. ("make -j8 ; make -j8 check"), since the
inconsistent move of ext2fs_close() without the preceeding printf
would cause all of the mke2fs tests (the m_* tests) to fail.
This is why regression test suites are so important. :-)
- Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists