![]() |
I just realized that kernel 3.0.0 is available since friday. So I just compiled it with my config from 2.6.39 and... trara: it's working! Here you can find my new config-3.0.0 for the fit-pc2.
Compared to the last kernel upgrade, this one was quite easy. The psb_gfx driver comes still from the psb-dkms package.
What's new in kernel 3.0.0? First: the version numbering. After 2.6.39 follows now 3.0.0. Ok, if Linus like so, why not. We are working to long with 2.6. But I hope, the linux kernel is not following now the version numbering inflation like chrome and firefox...
There is a quite good site where the changes are explaned: http://kernelnewbies.org/LinuxChanges
GMA still in staging?
Nice one, Alex!
The GMA500 drivers are still in the staging tree, yes?
I completely agree with your comments about the version numbering inflation.
Yes, still in staging :-(
Hi Matthew,
yes, gma500 is still in the staging tree :-(
But there have been several patches on 15th july by Alan Cox. Maybe he is merging somehow...?
But I have still a problem with heavy swapping. Either I really use too much memory (there is only 1 GB) or something has changed in kernel since 2.6.38....?
Best,
Alexander
Yes, noticed Alan pushing a lot of stuff so that's a good sign!
Swapping problems related to the GMA500? I wasn't aware there were any issues with this; I thought it was just functionality that was lacking. Or are you talking about swapping issues on 3.0 in general?
gma500 in staging is working!
Hi Matthew,
first: I'm using now the gma500/psb_gfx driver from the staging tree - not from the psb-dkms-package. So it really works now the code in the kernel tree.
My swapping problem is not related to gma500. I use a lot chrome with several tabs and opera/firefox in parallel. A lot for only 1 GB memory. And there is continuos music playback on a console with mplayer (to save some memory and CPU performance).
From time to time every access to any chrome-tab or what ever gets very slow and the system swaps a lot and sometimes even the music stops and nothing works for several seconds.
I had the impression that this behavior startet with 2.6.38 and I cannot find the fault. I could imagine that there is some kernel missconfiguration (because I have a very minimalist kernel config). Or maybe, I do too much in parallel...?
Best
Alexander