From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from xc.sipsolutions.net ([83.246.72.84]:38276 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751207AbYJUGnd (ORCPT ); Tue, 21 Oct 2008 02:43:33 -0400 Subject: Re: [PATCH v3] mac80211/drivers: rewrite the rate control API From: Johannes Berg To: Zhu Yi Cc: John Linville , linux-wireless , Vasanthakumar Thiagarajan , Felix Fietkau In-Reply-To: <1224570194.24677.301.camel@debian.sh.intel.com> References: <1223996147.10113.33.camel@johannes.berg> (sfid-20081014_165703_092655_BE6BA58D) <1224052607.3027.20.camel@johannes.berg> <1224484940.24677.169.camel@debian.sh.intel.com> <1224485319.18024.9.camel@johannes.berg> <1224486368.24677.184.camel@debian.sh.intel.com> <1224486587.18024.25.camel@johannes.berg> <1224562072.24677.277.camel@debian.sh.intel.com> <1224570194.24677.301.camel@debian.sh.intel.com> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-0HK3SWfaktI8LoqWmlld" Date: Tue, 21 Oct 2008 08:43:19 +0200 Message-Id: <1224571399.27899.67.camel@johannes.berg> (sfid-20081021_084338_263182_68255A45) Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: --=-0HK3SWfaktI8LoqWmlld Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Tue, 2008-10-21 at 14:23 +0800, Zhu Yi wrote: > On Mon, 2008-10-20 at 22:07 -0600, Zhu Yi wrote: > > The max retry rates number for iwl4965 is 16. See iwl-commands.h for > > more details: > >=20 > > a) Use the initial rate for only one entry. > > b) For each following entry, reduce the rate to next lower available > > rate, until reaching the lowest available rate. > > c) When reducing rate, also switch antenna selection. > > d) Once lowest available rate is reached, repeat this rate until > > rate table is filled (16 entries), switching antenna each entry. Thanks for the info. > Now I even wonder if abstract the common MRR info is possible. Yes, the > algorithm itself is hardware independent. But the one we use best > utilizes the hardware capability and it communicates with the hardware > frequently. I don't know about other hardware. But I highly doubt the > interface is so easy to be defined within IEEE80211_SKB_CB. Yeah, we definitely don't have enough space for 16 rates. If you want to properly report all that to userspace you'd probably have to add some radiotap information in the driver, but that's not currently possible... For other HW, all I know is Atheros (4 different rates with up to 16 tries each iirc), Broadcom (2 different rates, but we can rewrite ucode) and p54 (8 different rates, but no firmware influence) I'll clean up the three things you pointed out in this patch, I don't think we can achieve perfect reporting of what your hardware does at this time. johannes --=-0HK3SWfaktI8LoqWmlld Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Comment: Johannes Berg (powerbook) iQIcBAABAgAGBQJI/XoDAAoJEKVg1VMiehFYWY8P/0anpEEW7oYPzgpO9j3iZ32r 9lKgTNutq61SHpq3G8Ns0Xy2Ck0hQNc2MAZN6BQgy6dNLDTPnrmwqT+5xsindqnZ OhPeDH2Y8gWSnMA8GGOcPfaiLub+a8fE4hdWwRODVD8RSpv7wY5Tmc2DrOUXZJtT qb/6XGOIGpI/gKF0NhAuZvhlkCk2qs1fWS4A65SuNkwmTutouvXr5Obn+v6ZuCLB QSMhbtqPxaw5VbOBuRKkjbfbyPWhfrBoRp4l1xsJAhO2VyIxl9YpsHTtrzK7+6QD ZYC0+jKKYwRdM/QW/lm5DozTsKlcsNpG2TkV7fyBOBBjRLc2TaTlro3MhCrMhwGg AHCumEn9hGVypZrEXhILMRM7thr0+uGD0TJMccnGJuNetwL0CQU1SajVZtlZn00s MJnc7bwRw0XXfhvF2YTMuxvapU5PikE2EJzo5pwUtEElkqcvlj0PY27F1kSfHO4I VEyEY4mmps2njw4H+3Hd0DNmGtpRmBeBaOLbu3j1MD8XKd0VCyaQBk5pedtmQc4/ NJ/6w8aCUbmGlEsmyHMUkL+GT819NRaAoF6anf/Xk2ijxOagRQhJjhDFCyinmDx7 DlEnMocNsiQs0/4dZFCkZ5Ap5T/Tf1dF6xF77ReYywjwhoWWYK+tj8x1+fJKzWcG KtzfnZjSgS7bF4SVQKZR =hRVP -----END PGP SIGNATURE----- --=-0HK3SWfaktI8LoqWmlld--