On Wednesday 27 January 2010, Rafał Miłecki wrote: > W dniu 27 stycznia 2010 17:05 użytkownik Larry Finger > > <larry.fin...@lwfinger.net> napisał: > > On 01/27/2010 09:38 AM, Rafał Miłecki wrote: > >> 2010/1/27 Yuval Hager<yha...@yhager.com>: > >>> On Sunday 27 December 2009, Yuval Hager wrote: > >>>> On Saturday 26 December 2009, Gábor Stefanik wrote: > >>>>> Update your pci.ids file - what you have is a BCM4311/02 ABG. > >>>> > >>>> I ran update-pciids successfully, now dmesg says: > >>>> b43-phy0: Broadcom 4311 WLAN found (core revision 13) > >>>> > >>>> and lspci: > >>>> 02:00.0 Network controller [0280]: Broadcom Corporation BCM4312 > >>>> 802.11a/b/g [14e4:4312] (rev 02) > >>>> > >>>>> Also, could you check a few more kernels in-between .29 and .32? > >>>> > >>>> I have checked the following kernels, and found they all work > >>>> correctly: 2.6.29 > >>>> 2.6.29.6 > >>>> 2.6.30.4 > >>>> 2.6.31.9 > >>>> > >>>> Kernels 2.6.32 and 2.6.32.2 experience the disconnection I mentioned > >>>> in the original post. Before checking 2.6.32, I have removed > >>>> /lib/firmware/b43 completely and installed the firmware based on > >>>> the detailed instructions on the web site. I also enabled > >>>> CONFIG_B43_PHY_LP=y for 2.6.32. > >>>> > >>>> all kernels checked are vanilla kernels, from kernel.org. > >>> > >>> Just a quick update on this - this still happes on latest - 2.6.32.6. > >> > >> Could you try to (git) bisect this problem? > >> > >> This will unfortunately take hours to do, but should give us commit > >> that broke support for your wifi. > >> > >> For howto you can check http://landley.net/writing/git-quick.html > >> > >> Generally I think you should do something like: > >> > >> git clone > >> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git > >> cd linux-2.6 > >> git bisect bad 22763c5cf3690a681551162c15d34d935308c8d7 > >> git bisect good 74fca6a42863ffacaf7ba6f1936a9f228950f657 > >> > >> You can eventually bisect on drivers/net/wireless but I'm not sure how > >> to do this properly. > > > > git start drivers/net/wireless > > git good blahblah > > git bad blahblah > > Thanks Larry. > > Yuval: this way it should be really fast to find first-bad-commit. > Would be great if you could do this bisect. >
Yes, I know about 'git bisect', I used it in the past, and it's an awesome tool. I will run the test, and reply here when I have the result. It might take me some time though.. --y
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Bcm43xx-dev mailing list Bcm43xx-dev@lists.berlios.de https://lists.berlios.de/mailman/listinfo/bcm43xx-dev
