[00:10:24] --- Simon Wilkinson has become available [00:18:44] --- Simon Wilkinson has left [00:18:49] --- Simon Wilkinson has become available [00:18:56] --- Simon Wilkinson has left [00:56:48] --- Simon Wilkinson has become available [01:15:44] --- Simon Wilkinson has left [01:15:57] --- Simon Wilkinson has become available [01:18:35] --- Simon Wilkinson has left [01:19:44] --- Simon Wilkinson has become available [01:37:15] --- Simon Wilkinson has left [04:56:03] --- ballbery has become available [05:43:16] --- Simon Wilkinson has become available [05:59:16] --- meffie has become available [06:37:56] --- Simon Wilkinson has left [06:38:07] --- Simon Wilkinson has become available [06:49:30] --- Simon Wilkinson has left [06:49:54] --- Simon Wilkinson has become available [06:54:42] --- Simon Wilkinson has left [06:54:52] --- Simon Wilkinson has become available [08:03:31] --- deason has become available [08:15:13] --- Simon Wilkinson has left [08:15:35] --- Simon Wilkinson has become available [08:16:01] --- Simon Wilkinson has left [09:50:34] --- Simon Wilkinson has become available [10:02:51] --- Simon Wilkinson has left [10:03:10] --- Simon Wilkinson has become available [10:06:55] --- Simon Wilkinson has left [10:17:58] --- Simon Wilkinson has become available [10:28:23] --- Simon Wilkinson has left [10:28:39] --- Simon Wilkinson has become available [10:37:32] --- Simon Wilkinson has left [10:39:16] --- Simon Wilkinson has become available [10:49:14] --- Simon Wilkinson has left [10:49:29] --- Simon Wilkinson has become available [10:56:22] --- Simon Wilkinson has left [10:56:37] --- Simon Wilkinson has become available [11:16:12] --- Simon Wilkinson has left [11:16:21] --- Simon Wilkinson has become available [11:34:22] --- Simon Wilkinson has left [11:58:40] --- Marc has left: Disconnected [13:43:14] --- Simon Wilkinson has become available [13:51:08] --- Simon Wilkinson has left [13:51:25] --- Simon Wilkinson has become available [14:01:03] --- meffie has left [14:08:37] --- Simon Wilkinson has left [14:09:04] --- Simon Wilkinson has become available [14:09:44] --- Simon Wilkinson has left [14:09:56] --- Simon Wilkinson has become available [14:18:01] --- Simon Wilkinson has left [14:18:29] --- Simon Wilkinson has become available [14:20:36] --- Simon Wilkinson has left [14:52:33] --- Simon Wilkinson has become available [15:13:26] --- mvita has left [15:41:35] --- deason has left [15:46:50] --- mvita has become available [15:53:20] I guess I should have also asked which userspace afsd threads wake up periodically, since we probably should dump some more entropy into the kernel every now and then. [15:55:21] If it wasn't for AIX, I would say we just shouldn't care. [15:55:35] AIX has /dev/random. I wonder if there is any way to easily get to it from the kernel. [15:56:12] It is pretty hard for me to get excited about HPUX, IRIX, or AIX, really. But we do have an AIX buildlsave, so we're more forced to care there than we otherwise would be :) [15:56:45] I know that DFBSD should provide randomness to kernel modules; I assume that it's just that no one has written the glue for osi_readRandom yet. [15:57:49] I just wrote it for the platforms I could find documentation for. [15:58:28] (The AIX buildslave is actually really useful, since it enforces the libtool export symbol lists. Though the Mavericks buildslave should also do so, now.) [15:58:49] I guess I won't ask you if there's documentation for AIX kernel module developers freely available, then :) [15:59:50] While you're here, how much should we care in openafs about the lack of a prototype for _krb5_internal_hmac? [16:00:15] I don't have a good sense for where we would want to stick in a prototype for it (if we even needed to do so). [16:05:23] crypto/rfc3961/krb5_locl.h would seem like the logical place [19:10:20] --- mvita has left [21:56:29] Jeff: you want to backend osi_readRandom with a downcall to a userspace thread that calls back with the requested bits? [21:58:18] Also, for AIX, the docs make it seem like we could fp_open("/dev/urandom",...) and fp_read() from it. I am not sure how I would test whether such a scheme works, lacking an AIX box. [22:11:38] --- gendalia39652 has left [23:20:56] --- cudave has left: Replaced by new connection [23:20:57] --- cudave has become available [23:33:52] --- Simon Wilkinson has left