Return-path: Received: from mga11.intel.com ([192.55.52.93]:16241 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754083Ab0CZD5M (ORCPT ); Thu, 25 Mar 2010 23:57:12 -0400 Subject: Re: [PATCH 1/2] mac80211: support paged rx SKBs From: Zhu Yi To: Kalle Valo Cc: "linville@tuxdriver.com" , "linux-wireless@vger.kernel.org" In-Reply-To: <87eij8c3ld.fsf@purkki.valot.fi> References: <1269399463-5244-1-git-send-email-yi.zhu@intel.com> <87eij8c3ld.fsf@purkki.valot.fi> Content-Type: text/plain; charset="UTF-8" Date: Fri, 26 Mar 2010 11:57:57 +0800 Message-ID: <1269575877.4043.203.camel@debian> Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Thu, 2010-03-25 at 14:33 +0800, Kalle Valo wrote: > Zhu Yi writes: > > > Mac80211 drivers can now pass paged SKBs to mac80211 via > > ieee80211_rx{_irqsafe}. The implementation currently use > > skb_linearize() in a few places i.e. management frame handling, > > software decryption, defragmentation and A-MSDU process. We can > > optimize them one by one later. > > I think ieee80211_rx{_irqsafe} documentation should now mention that > paged skbs are supported. Better to make this suppot explicit for the > driver authors. Good idea it should be documented somewhere (after the patch is accepted). Maybe the developer wiki page? For the code itself, I think Linux drivers can always assume the network stacks support paged buffers. In case some stack doesn't support paged buffer, it should do skb_linearize() itself in the very beginning of its rcv handling (e.g. sctp_rcv, tipc_recv_msg, etc). Thanks, -yi