[01:51] is there something wrong with fglrx + u 37-12 kernel? [01:51] all i get is a crash [01:52] with 37 mainline it works [02:25] Kano: which 37 mainline? [02:26] ie. how recent? What is the head? [02:26] yes 37 mainline kernel compared to 37-12 [02:27] Kano: right, but how recent is the mainline? [02:27] well i did not test ubuntu, i tested with debian squeeze [02:27] 2.6.37 final [02:27] the u precompiled one [02:27] will check nvidia tomorrow, but fglrx never worked [02:28] http://kanotix.com/files/hellfire/ubuntu/ [02:28] thats that i used [02:29] test system was i7-880 with hd5670 64 bit [02:29] nothing changed expect kernel, .37 mainline works... [02:30] Kano: hrmmm okay, I am not aware of any issues but I haven't been following the video drivers closely [02:30] I'll poke apw and see if he knows anything [02:31] maybe i install u tomorrow too [02:31] but thats really curious [02:31] Kano: yeah it is, what behavior did you see? [02:31] crash on load of kdm [02:31] black screen on boot, purple ubuntu screen [02:32] i tested it with debian squeeze currently [02:32] but u kernel [02:32] + firmware+wireless-crda [02:32] right, I recalled after the puple ubuntu screen Q [02:33] so none of the crazy plymouth boot issues [02:33] well i have got plymouth installed,but p works with mainline too [02:33] it does not matter if i disable it or not [02:34] using oss ati driver x works [02:36] basically i thought i could use u kernel now without recompile, i always disabled your hpa hack [02:36] but that was reverted for -12 kernel... so i wanted to try it [02:37] ah, well I'm not sure why it isn't working, especially with mainline .37 working [02:37] well the gcc is definitely different which is used [02:37] i can install mainline kernel even with lenny [02:37] Kano: which mainline did you use? [02:37] ie who built it [02:38] http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.37-natty/ [02:38] okay, that should be the same version of gcc [02:38] the linux headers install with lenny libc6 [02:39] but not the natty ones [02:39] so gcc is different [02:40] hrmmm /me is puzzled, our mainline kernels should be using the same build env [02:40] they dont use pbuilder [02:40] i really like the mainline kernels because of the backwards compat [02:41] you have to disable modeset then you can use em with lenny too [02:43] ah nice, I wasn't aware they were that backwards compatable [02:43] only when you know what you do ;) [02:49] the hardy gcc is used, hardy is mainly compatible with lenny [02:49] Linux version 2.6.37-020637-generic (root@zinc) (gcc version 4.2.3 (Ubuntu 4.2.3-2ubuntu7)) #201101050908 SMP Wed Jan 5 10:17:13 UTC 2011 [02:49] from cat /proc/version [03:00] I'm going to guess that seeing something on 3 different machines with three (or is it 2?) different types of hardware makes it likely to be in the kernel, not the individual hardware... [03:01] when I hit the disk drive really hard, the swraid devices seem to decide that one of the drives has failed (usually complaining about a timeout in SCSI).. rebooting the machine magically restores everything to hunky-dory status [03:01] lucid and maverick kernels, fwiw [03:02] and pure mainline too? [03:02] haven't looked [03:02] lacking a machine to test on [03:14] I may have to eat my words... the most recent failure may well have been a controller falling over === JanC_ is now known as JanC === ghostcube_ is now known as ghostcube === yofel_ is now known as yofel === Quintasan_ is now known as Quintasan === jjohansen is now known as jj-afk === jj-afk is now known as jjohansen === gnomefreak76 is now known as gnomefreak