Date: Wed, 20 Mar 2013 21:43:20 -0500 From: Joshua Isom <jrisom@gmail.com> To: freebsd-wireless@freebsd.org Subject: Re: update - better EDMA RX support, testing AR9580 Message-ID: <514A73C8.7050102@gmail.com> In-Reply-To: <CAJ-Vmo=MZXU3qdOHmv99Hq-XRZCCphw-O_M8Tv=T7srE=apq%2Bg@mail.gmail.com> References: <CAJ-Vmo=UZDCb%2BeSRLO5fiJQNJkzVYqjcy4zWNc70CBePQyKGjw@mail.gmail.com> <CAJ-Vmo=MZXU3qdOHmv99Hq-XRZCCphw-O_M8Tv=T7srE=apq%2Bg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 3/19/2013 10:31 PM, Adrian Chadd wrote: > On 19 March 2013 12:39, Adrian Chadd <adrian@freebsd.org> wrote: >> Hi all, >> >> I've just committed some new RX handling code. Since the RX FIFO on >> the AR9380 and later chips is very shallow (128 entries), you have to >> refill the FIFO much quicker than you can handle frames. > > .. and an AR9390 STA -> AR9380 AP at MCS24 is giving me around 230mbit TCP. > > My goal is to hit 200mbit TCP on 2x2 and 300mbit TCP on 3x3. > > But, I won't complain about getting to 230mbit this early in the game. > > > > Adrian Any tips for backporting the code to -STABLE? Currently I had to comment out the scsi cd driver because it was causing kernel panics. Now I'm getting kernel panics from geom_io.c due to a failed assert.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?514A73C8.7050102>