[00:50:40] --- Simon Wilkinson has left [00:50:40] --- Simon Wilkinson has become available [00:50:40] --- Simon Wilkinson has left [00:56:44] --- Simon Wilkinson has become available [01:00:25] --- Simon Wilkinson has left [01:03:25] --- Simon Wilkinson has become available [01:03:34] --- Simon Wilkinson has left [02:21:41] --- dev-zero@jabber.org has become available [02:21:44] --- dev-zero@jabber.org has left [03:23:00] --- Simon Wilkinson has become available [03:24:02] --- kula has left: Lost connection [03:51:31] --- Simon Wilkinson has left [03:51:41] --- Simon Wilkinson has become available [03:59:11] --- Simon Wilkinson has left [03:59:36] --- Simon Wilkinson has become available [05:09:06] --- Simon Wilkinson has left [05:10:15] --- Simon Wilkinson has become available [05:23:11] --- meffie has become available [05:23:37] --- Simon Wilkinson has left [05:23:46] --- Simon Wilkinson has become available [06:00:30] --- Simon Wilkinson has left [06:02:19] --- mvita has become available [06:09:31] --- meffie has left [06:09:47] --- Simon Wilkinson has become available [06:15:24] --- meffie has become available [06:23:38] --- Simon Wilkinson has left [06:24:14] --- Simon Wilkinson has become available [06:28:32] --- Simon Wilkinson has left [06:51:40] --- wiesand has become available [07:02:23] --- wiesand has left [07:02:27] --- wiesand has become available [07:58:24] --- Simon Wilkinson has become available [08:10:53] --- wiesand has left [08:17:34] --- deason has become available [08:17:38] --- deason has left [08:46:59] --- Simon Wilkinson has left [08:59:58] --- Simon Wilkinson has become available [09:04:34] It looks like the headers in src/opr are declaring prototypes as extern, is there a reason to do so? [09:07:51] That was the original convention [09:08:04] Original to ... opr? [09:08:26] The whole tree [09:08:58] Okay, then. [09:09:02] When I came on the scene, all prototypes went in a blah_prototypes.h file, and were all declared extern [09:09:20] I see there is no opr_prototypes.h ;) [09:09:23] When I did all of the prototyping work, I didn't realise that, so things became somewhat inconsistent. [09:10:50] At this stage the only advantage to separate header files is that we could possibly autogenerate the prototypes in the same way as Heimdal does [09:11:20] Heimdal's scheme seems to end up frustrating/confusing me fairly often, since I don't usually have a build tree sitting around for it. [09:12:01] ... which reminds me about that _krb5_internal_hmac() thing. [09:12:28] Its nice when you're actually developing for it, as you don't need to update prototypes in multiple locations. And it knows from symbol naming whether stuff goes into a public or private header [09:12:51] Also, since you're here, I'll ask if you'd prefer the procsize routine to be in opr/proc.c or opr/sys.c or some other name. [09:19:57] Either sound fine [09:46:24] Hmm, I wonder whether GetNewFID wants to be zeroing the new vnode when it (re)allocs to make one, instead of just setting it to be of type vNull. [09:56:00] --- Simon Wilkinson has left [11:10:25] --- meffie has left [11:50:17] --- meffie has become available [15:17:38] --- mvita has left [17:05:13] --- meffie has left [18:56:00] --- mvita has become available [21:14:57] Hmm, maybe I don't need to manually mark as verified if it's already been approved (but is not yet merged). [21:19:03] patches cannot be merged without both verified and approved [21:19:52] I'm not sure what is causing that suse buildslave to error out with an ~empty build log. [21:20:32] Or maybe I'm just looking too late and this is normal. [21:21:09] I thought that looking too late resulted in just a dead link, though; I guess there's some sort of intermediate state where the summary information is available but not the actual log. [21:58:14] --- mvita has left [22:46:16] --- ktdreyer has left