[Bug 675050] Review Request: cloudfs - Cloud Filesystem

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

 



Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.


https://bugzilla.redhat.com/show_bug.cgi?id=675050

Kaleb KEITHLEY <kkeithle@xxxxxxxxxx> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kkeithle@xxxxxxxxxx

--- Comment #9 from Kaleb KEITHLEY <kkeithle@xxxxxxxxxx> 2011-06-27 12:16:29 EDT ---
IMO it's sort of a spurious argument to claim that some arbitrary package built
-I/usr/include/glusterfs can't get /usr/include/uuid.h without resorting to
unusual measures. In all likelihood the only package that will use it is
CloudFS, and CloudFS doesn't need /usr/include/uuid.h at all, and so doesn't
need to resort to any measures whatsoever. Is there a concrete example of
another package besides CloudFS that would use glusterfs-devel?

NB. The glusterfs uuid.h is, apart from an extra #include "config.h", identical
to Ted T'so's libuuid-devel uuid.h installed in /usr/include/uuid/uuid.h.
(These are _not_ the same as uuid-devel /usr/include/uuid.h.)

Note too that it doesn't work to just change the '#include "uuid.h"' in iatt.h
to "glusterfs/uuid.h". The glusterfs headers come from different subdirs in the
glusterfs source changing their iatt.h or their source tree upstream is, as
jdarcy said, highly unlikely. The only way this would work would be to patch
all the gluster headers while building glusterfs-devel to change all the
#include "foo.h" to #include <glusterfs/foo.h>.

Finally note that the extraordinary lengths suggested above (-I/usr/include or
#include_next<uuid.h>) don't work anyway. At least not with F14's gcc-4.5.1 or
F15's gcc-4.6.0, which appear to be overly clever in determining that they have
already included uuid.h from somewhere and won't include it again. (And no,
it's not a function of #ifndef _UUID_H_/#define _UUID_H_ type protection.)

It might make more sense to remove the uuid.h from the glusterfs-devel package
and fix cloudfs to use the libuuid-devel version. However doing this means we
add a dependency to libuuid-devel.

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
_______________________________________________
package-review mailing list
package-review@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/package-review


[Index of Archives]     [Fedora Legacy]     [Fedora Desktop]     [Fedora SELinux]     [Yosemite News]     [KDE Users]     [Fedora Tools]