[15:14] Hi all... I've some problem with 2.6.32-22. DVB does not work anymore. [15:14] I looked at the debdiff, there are a lot of lines quoting "dvb" [15:15] so.. I did this: [15:15] cat debdiff | grep dvb | grep ^- > dvb- [15:15] cat debdiff | grep dvb | grep ^+ > dvb+ [15:15] diff dvb- dvb+ [15:15] result is "null". [15:16] But... my DVB-S card works fine in 2.6.32-21 [15:43] hello. [15:45] Does anyone know what change from gutsy would require the use of noapic or nolapic to boot on a pentiumk 4 ? [15:46] in gutsy the system booted with apic no problems whatsoever.. [15:46] gutsy onwards the system requires adding no noapic or nolapic on the command line.. [15:48] l3on, what dvb card do you have ./ === JanC_ is now known as JanC [17:44] lapion: there are many, many possible candidates [17:44] lapion: I recommend you file a bug if you haven't. [18:06] the problem with the bug is that it's behaviour is very strange [18:07] if I boot with a newer kernel and with apic, at a certain moment during boot the system freezes, unless I press a key, any key.. such as ctrl or shift or alt or any other key, [18:09] during keypress the system will continue booting, but as soon as I release the key the startup procedure slowly grinds to a halt again.. untill key is pressed again.. [18:26] lapion: 04:07.0 Network controller: Techsan Electronics Co Ltd B2C2 FlexCopII DVB chip / Technisat SkyStar2 DVB card (rev 02) [18:31] this is all a bit ridiculous. We need to flip the default DMA update method for HDA, because the blacklist is exploding. I've pushed fifteen fixes in the past month. :( [18:33] lapion: what I see from my DVB client is that when I try to lock frequency I not receive the FE_HAS_LOCK [18:33] that was defined in frontend.h: [18:33] linux/dvb/frontend.h: FE_HAS_LOCK = 0x10, /* everything's working... */ [18:50] lapion: I discovered that there are no changes between: [18:50] diff -Nur linux_2.6.32-21.32/drivers/media/dvb/ linux_2.6.32-22.33/drivers/media/dvb/ [18:50] result is "null" [18:50] So, is it a build problem ? [20:05] hi, why there is no kernel ppa repository and there's only the site? It would be easier if small updates (meaning "the fourth number" updates) for newer kernels could be performed automatically... [20:11] m_gol: the way ppa's are structured and they way the kernel is packaged make it impossible with out changes to the LP infrastructure [20:13] pgraner: I don't understand... You provide debs for kernels - what is exactly the problem with putting them into a repository? [20:14] m_gol: its the nameing and abi, if we don't bump abi it will overwrite the previous kernel in ppa which is not desired, we need to keep each build [20:15] m_gol: we don't bump abi unless we have to as it breaks 3rd party modules (aka vmware et al.) [20:16] I'd say it's not that difficult in itself, but requires a LP change to special case the kernel packages? [20:16] m_gol: we use ppa's for scratch/test builds (ie. testing specific bug fixes) [20:16] JanC: correct LP needs to understand abi which it dosen't do now [20:17] JanC: since we have a workable solution and the change is only needed for the kernel its not high priority [20:17] OK, I see [20:17] anyway, if one day it was possible, I and many others would be grateful :) [20:18] so as long as the ABI doesn't change between 2.6.X & 2.6.X+1, LP can't handle that? [20:18] well, or maybe 2.6.X.Y & 2.6.X.Y+1 [20:19] there are a lot of problems that often can be solved by kernel upgrade (drivers mainly), e.g. I had to keep old version of wine because of some signal-related bug that causes HoMM3 to hang with newer wine on kernels < 2.6.34... [20:19] BTW: it was nice to have Steve Conklin(sp?) and his wife at LGM :) [20:21] JanC, the Conklin's are awesome! [20:21] JanC: yea steve is a great guy [20:22] akgraner: I didn't know you knew his wife or actually she was related to ubuntu somehow until a day or so after her talk [20:22] when you said so in #u-w or so [20:23] LGM was really cool BTW [20:23] yeah - she does some great design work on clothes [20:23] and she has some great ideas for re-purposing t-shirts and stuff [20:24] that's why I think LGM was so cool: a mix of artaists & developers & more "ordinary" users of "graphical" applications [20:24] and "design" etc. === dmarkey_ is now known as dmarkey [22:20] Is seeing this in kern.log during suspend any cause for concern? "ACPI handle has no context!" === MTecknology is now known as Guest00000001 === Guest00000001 is now known as MTecknology