[00:57:38] --- Russ has left: Disconnected [03:31:35] --- haba has become available [05:01:47] --- brantgurga has become available [05:26:31] --- SecureEndpoints has left: Replaced by new connection [06:02:01] --- SecureEndpoints has become available [06:33:45] --- mmeffie has become available [08:13:31] --- reuteras has left [09:40:27] Hm, how shold I say this? Maybe: AFS version: OpenAFS 1.2.11 built 2009-04-24 (with patch on alpha_dux50) - UGH! [09:42:21] --- haba has left [09:44:43] --- Russ has become available [10:35:05] about to go over RT, but, who has outstanding issues for 1.5.60? [10:44:46] we need to resolve the issues surrounding the rename of the ResidencyCmd pioctl (since rename-residency-from-mrafs-to-osd-20090119 was never undone in 1.4.x, we have a divergence; since, as before, it changes nothing on the wire and the memory format of no structure, i am still of the position that we should do it) [10:45:30] The only changes are to identifiers in local API's? [10:47:05] Presumably, it means that odd things may happen if you use a MR-AFS fs against a modern OpenAFS kernel module. [10:47:11] I don't think I care about that. [10:47:36] Uh, why? Or is the issue that the pioctl was _renumbered_? [10:48:09] also, the transaction retry issue should be addressed, or dropped. if we address it, it should be in the client, not the server; until async rx exists, and given the low number of threads the volserver is capable of, tying those up is just not a good idea [10:48:20] > The only changes are to identifiers in local API's? correct [10:48:33] the pioctl was not renumbered [10:48:44] Ah. okay. [10:49:25] There's stuff in RT that should go in. [10:49:29] If the only change was to the name of a pioctl, then I suggest going ahead with the change, but also continuing to define the old preprocessor macro so that code that calls pioctl() with the old macro will still work. [10:49:47] Felix's mmap stuff, if we're happy that its correct. The rename of dprintf to afs_dprintf [10:49:52] the existing number was renamed. for an unregistered use (as far as i am concerned: it's not on the web pages, and besides, mr-afs is dead), with the wireless and data formats all being the same. don't care. old fs will work with new kernel, and vice versa. don't. care. [10:49:59] The patch I gave you last night. [10:49:59] If the structure was also renamed, you may need to define two identical structures to get that effect. [10:50:08] there's plenty of stuff in RT which should go in. most is uncontroversial [10:50:19] > Felix's mmap stuff, if we're happy that its correct. [10:50:30] i get the idea marc and he don't think it's correct, just better. i am still reading [10:51:07] Lemme see if there's anything else in my tree that I should give you. [10:51:23] Don't assume it's unregistered just because it's not on the web pages. Unfortunately there are a number of cases where an approved registation is recorded only in the state of the ticket that requested it. But... I don't think anything was ever registered for mr-afs per se. I believe there is a set of requests related to osd [10:51:37] > Don't assume it's unregistered just because it's not on the web pages. [10:51:42] set up kula *cough* [10:51:50] I get the idea the mmap stuff is not correct yet. I sort of wish Chaskiel would weighin [10:52:09] Yes, I'll set up kula. _after_ the marathon [10:52:19] Hopefully as soon after the marathon as possible. [10:52:30] (but, I'll sleep first :-) [10:52:46] > I get the idea the mmap stuff is not correct yet. [10:52:46] How do we feel about the cfileopen stuff? It would be good to get that in. [10:52:50] yeah. that's the problem [10:53:03] cfileopen being the union struct for agnostic cache types? [10:53:08] Yes. [10:53:50] it looks fine here. [10:54:20] I'm pretty sure its good. There _may_ be architectures that I've missed, but 1.5 would seem to be a perfect place to find that out. [10:55:43] I've got a couple of prototyping patches that are probably ready, too. [10:56:12] And, toby's Mac OS X installer widget is ready, too. [10:58:07] yay! [11:04:15] --- dev-zero@jabber.org has become available [11:04:18] --- dev-zero@jabber.org has left: offline [11:38:22] --- mmeffie has left [12:27:08] --- haba has become available [12:44:08] --- Simon Wilkinson has left: Lost connection [13:22:42] --- dlc has become available [14:27:18] --- haba has left [14:27:35] --- Simon Wilkinson has become available [14:36:43] --- Simon Wilkinson has left [17:37:50] --- Russ has left: Disconnected [17:57:06] --- Russ has become available [18:16:51] --- dwbotsch has left [18:17:17] --- dwbotsch has become available [20:25:37] --- dev-zero@jabber.org has become available [22:25:43] --- Russ has left: Disconnected [22:45:59] --- brantgurga has left [22:55:33] --- dev-zero@jabber.org has left: offline