Return-path: Received: from xc.sipsolutions.net ([83.246.72.84]:55246 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752476AbYJTKQE (ORCPT ); Mon, 20 Oct 2008 06:16:04 -0400 Subject: Re: [RFC] mac80211: Re-enable aggregation From: Johannes Berg To: Sujith Cc: "Luis R. Rodriguez" , "linville@tuxdriver.com" , "linux-wireless@vger.kernel.org" , Luis Rodriguez , "tomasw@gmail.com" In-Reply-To: <18684.22627.103501.578126@gargle.gargle.HOWL> References: <18684.16351.638713.791015@gargle.gargle.HOWL> <1224491480.18024.32.camel@johannes.berg> <18684.18492.94865.480736@gargle.gargle.HOWL> <43e72e890810200214g16f847caqa6cbf724b7b7b625@mail.gmail.com> <1224494669.18024.51.camel@johannes.berg> <18684.22627.103501.578126@gargle.gargle.HOWL> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-ufYpTJjXH8jR15oawBuB" Date: Mon, 20 Oct 2008 12:15:58 +0200 Message-Id: <1224497758.27899.3.camel@johannes.berg> (sfid-20081020_121614_789837_B531E9A7) Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: --=-ufYpTJjXH8jR15oawBuB Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Mon, 2008-10-20 at 15:37 +0530, Sujith wrote: > Johannes Berg wrote: > > On Mon, 2008-10-20 at 02:14 -0700, Luis R. Rodriguez wrote: > >=20 > > > Indeed I agree with this. I think we can do this in three steps: > > >=20 > > > 1. resolve aggregation for now for ath9k > >=20 > > That's what Sujith's patch does, basically. It just leaves the Intel > > hooks in place. > >=20 > > > 2. resolve aggregation for amdpu_queue case. My suggestion here is to > > > use skb_buf_head (after Jouni's suggestion for our driver in fact), > > > and do not make use of a qdisc at all for this > >=20 > > I agree on both, but this is more of an implementation detail than an > > API description, which we really want to have first. >=20 >=20 > Would be nice if an interim fix is accepted temporarily, > until the architectural details of proper mac80211 aggregation > support for the 2 classes of HW are defined, and implemented. Yeah I think that if you clean up this patch a bit wrt. that flag and double-check it won't cause problems on hardware that cannot do aggregation at all I'd be ok with it. Like I said, a later patch can remove all the if (ampdu_queues) stuff johannes --=-ufYpTJjXH8jR15oawBuB 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/FpaAAoJEKVg1VMiehFYKEQQAJURnqUYA/AQOyXQPDZoCaHp VZHXzrbNOlDVkNEqEOuzh73vWA1l5GpoXOhvjsmBvKjSjUo7XvnQaMXlMQfNUGSq fCPpReszlTKHVOAnZwXQ0RPyjgiMXJJJSU0p3BTLltbawmIMkdWlI7ASmiz1zCdT gv0kGpiTANANFDDqC4QC4PaBen1IK5FBR45pS9ZQtlbLI0/7almo/v3d9wGB2DOL WXNGVQHYC7p0iFjJSZV/2/+R9OACeUrPCJBjvSSbgSCj4oouUYz5MIfcyNVu6Zke JLmXScIJJ8OdnHrsw20qk25Ygp28PMVDoPwd1DNyyibp/NXRGVhal2McykJCPphx YcH0S3AMnWALFWQFBhwpJHMFvHUX9CaWHJ8QmMG+7Q9mtOoSaEXO2nv6+ZzRtKHQ NoX0Ac4Ff/t7/wQxObYNoxsIVHxWLfGd6wVMpgchIwx6Gt2Oh1DmroQ2ysJw7Wed lh4SO1m4YAngj/yxFbDwQ2k/nhXXiijJZv6JcKeKON63IWt6xyfpPmxu55kbQZmc SxJruoMTyRHA/dkVCq94Gk2x8eVe8j7tbiC3XRyJuTb6cNZW2ChgnAk/GKSCZX97 DRBRNA8uqg7ppwPHmaAOXIQcas9iigQnr3EOIfAOx575jN3MVpSnMS8Rg6qjkcuK IjcjtVSgymlbmfy0skoL =+l7w -----END PGP SIGNATURE----- --=-ufYpTJjXH8jR15oawBuB--