Date: Fri, 28 Apr 2006 15:55:24 -0700 From: Ross Finlayson <finlayson@live555.com> To: freebsd-mobile@freebsd.org Subject: Re: kernel: ath0: device timeout Message-ID: <7.0.1.0.1.20060428154610.01d4a828@live555.com> In-Reply-To: <445290B9.5050807@errno.com> References: <7.0.1.0.1.20060428141609.01d4a828@live555.com> <445290B9.5050807@errno.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>ath_rate.ko is the target built by any of ath_rate_sample,
>ath_rate_onoe, and ath_rate_amrr so if you're using modules you
>build+install whichever is appropriate and then kldload if_ath.
OK, I see - "ath_rate_sample" turns out to be the default because it
appears last in the list of "ath_rate_*" in
"/usr/src/sys/modules/Makefile". Thanks.
FYI, the performance of my "ath0" device has improved noticeably
since I started using "ath_rate_onoe" instead of
"ath_rate_sample". However, I am still seeing the occasional "ath0:
device timeout" error.
>I have done extensive testing of all the rate control algorithms as
>well as a proprietary one and chose sample as the default. However
>none are anywhere near as effective as the proprietary one.
What is the problem with the 'proprietary' algorithm? Is it that
Atheros has not released the source code? Or are there patent
issues?? How does Linux deal with this - do they just used a closed
source binary kernel module, or something?
Also, out of curiosity, what is "ath_rate_amrr", and how does it
compare with the other two available algorithms (onoe and sample)?
Ross.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?7.0.1.0.1.20060428154610.01d4a828>