On 31 October 2014 23:41, Evgeniy Dushistov <[email protected]> wrote: > On Fri, Oct 31, 2014 at 09:12:39PM +0100, Rafał Miłecki wrote: >> > So it is 3.18 Mbit/s vs 71.57 MBit/s under the same conditions, >> > and the only difference is OS. >> > >> > Any hints, how to speedup connection under linux? >> >> We have some improvements for HT-PHY (BCM4331) in 3.18-rc1 (and >> newer). Could you try it? >> >> b43 is still limited to 802.11g speeds, but I hope it will be better >> than 3.18 Mb/s. >> > > I got: > % Total % Received % Xferd Average Speed Time Time Time > Current > Dload Upload Total Spent Left Speed > 100 2147M 100 2147M 0 0 2246k 0 0:16:18 0:16:18 --:--:-- 1842k > > real 16m18.774s > user 0m27.340s > sys 0m36.350s > > This is ~ 17.55Mbit/s for file transfer. > What is much, much better. Thank you. > > Are these changes(3.16 [slow b43] vs 3.18 [fast b43]) located > in drivers/net/wireless/b43? > I would like to port them(changes) to my 3.16 kernel.
You may also look at drivers/ssb. Anyway, there is already a project of backporting drivers, see: https://backports.wiki.kernel.org/index.php/Main_Page > By the way, during test connection was hang, > and I can not ping any machine outside of my notebook. > > I rmmod/modprobe b43, restart wpa_supplicant/dhcpcd > and all back to normal. > And after restart I was unable to reproduce it(bug). > dmesg also show nothing. Any hints how to debug such > kind of issues, if I see it again? I'm afraid not. You would need to hack driver to display info about interrupts. Then see if there are any packets appearing or not, etc. -- Rafał _______________________________________________ b43-dev mailing list [email protected] http://lists.infradead.org/mailman/listinfo/b43-dev
