On Sat, 2014-08-23 at 20:14 -0700, Andres Cimmarusti wrote: > > As this is apparently specific to a non-free driver, you should not > > expect that the kernel team will ever do any work on it. > > I understand > > > However, if you can find a specific commit that introduced it, or that > > fixes it, we might be able to make some progress. > > At the moment this is difficult for me because this issue is very > erratic and I think it has to do with some configuration option in the > kernel. I would have tried different kernels, but currently everytime > I try to compile a vanilla kernel either the deb-pkg way or the > kernel-package way I get a blocking error... > > > On Sun, 2014-04-27 at 11:13 -0400, Andres Cimmarusti wrote: > >> found 720528 linux/3.13.10-1 linux/3.11.10-1 > >> notfound 720528 linux/3.12.9-1 linux/3.9.6-1 > >> thanks > > [...] > > > > So this worked in 3.9, broke in 3.11, worked in 3.12, broke again in > > 3.13? That seems unlikely. > > Let me re-state: worked with debian 3.9, broke with debian 3.11 (but > works with Ubuntu vanilla LTS 3.11.10.x series), worked with debian > 3.12, broke again with debian 3.13 (but worked again with Ubuntu > vanilla LTS 3.13.11.x series) and finally doesn't work with debian > 3.14 and also not with 3.15, I have yet to try 3.16. > My guess is that a configuration option I'm using (processor timer > frequency 1000 Hz) when building the Ubuntu vanilla LTS kernels is > doing the trick, but then again I don't know why debian 3.12 did work. > Could this faster 1000 Hz vs 250 Hz (default in debian) have anything > to do with this ACPI issue Nvidia reported...
It doesn't seem very likely. I also can't see any obviously related patches to ACPI support in Ubuntu's kernel. Ben. -- Ben Hutchings One of the nice things about standards is that there are so many of them.
signature.asc
Description: This is a digitally signed message part
