- Wednesday, February 11, 2015

openafs@conference.openafs.org

Wednesday, February 11, 2015< ^ >

Room Configuration

Room Occupants

GMT+0
[13:20:00] shadow@gmail.com/barnowlC91EEAD3 leaves the room
[13:20:09] shadow@gmail.com/barnowlC91EEAD3 joins the room
[13:30:57] meffie joins the room
[14:23:43] Simon Wilkinson joins the room
[14:25:34] wiesand joins the room
[14:49:40] Simon Wilkinson leaves the room
[14:49:46] Simon Wilkinson joins the room
[14:51:36] shadow@gmail.com/barnowlC91EEAD3 leaves the room
[14:51:45] shadow@gmail.com/barnowlC91EEAD3 joins the room
[14:51:52] Simon Wilkinson leaves the room
[14:52:11] Simon Wilkinson joins the room
[16:39:27] wiesand leaves the room
[18:01:55] Simon Wilkinson leaves the room
[18:55:11] shadow@gmail.com/barnowlC91EEAD3 joins the room
[19:00:52] shadow@gmail.com/barnowlC91EEAD3 leaves the room
[19:02:49] shadow@gmail.com/barnowlC91EEAD3 leaves the room
[19:02:57] shadow@gmail.com/barnowlC91EEAD3 joins the room
[19:20:45] shadow@gmail.com/barnowlC91EEAD3 leaves the room
[19:20:53] shadow@gmail.com/barnowlC91EEAD3 joins the room
[19:22:40] shadow@gmail.com/barnowlC91EEAD3 leaves the room
[19:23:18] shadow@gmail.com/barnowlC91EEAD3 joins the room
[20:37:48] meffie leaves the room
[20:59:41] kaduk joins the room
[21:04:22] meffie joins the room
[21:04:35] meffie leaves the room
[21:35:10] kaduk leaves the room
[21:47:47] kaduk joins the room
[21:55:36] <kaduk> Hmm, apparently clang 3.5 (new-ish in freebsd HEAD) will start whingeing about abs() calls on unsigned quantities and of the wrong width.
[21:59:56] <Jeffrey Altman> because abs(1) != 1 ?
[22:02:31] <kaduk> Well, calling abs() on a long just plain loses precision. abs() on an unsigned type has no effect, so clearly the programmer was wrong to do so.
[22:02:36] <kaduk> Or something.
[22:03:59] <kaduk> In other FreeBSD news, /bin/sh /path/to/our/libtool --quite --mode=link --tag=CC gcc -static [...] fails to actually produce a statically linked executable. Removing the libtool prefix to just invoke gcc -static does produce a statically linked executable.
[22:06:40] <kaduk> It is always hard to work up the motivation to go dig into libtool and figure out what it's doing wrong.
[22:12:15] <ballbery> I think you need to tell *libtool*, not gcc, to do --static
[22:12:25] <kaduk> That's probably true, yes.
[22:12:39] <ballbery> and it will arrange for gcc etc. to do the right thing and change its own processing to work with static libs
[22:54:37] <kaduk> Apparently freebsd current also fails the same way if I tell configure to use --with-roken=internal; otherwise it picks up the system roken and masks the bug
[23:32:46] <kaduk> Maybe -all-static will fix my Makefile...
[23:44:29] <kaduk> Well, it sort-of fixes things.
Now I discover that we are failing to sufficiently clean our rfc3961 namespace.
[23:57:51] <kaduk> And ... the system libkrb5.a has an undefined reference to heim_ipc_call, grump.