Return-path: Received: from he.sipsolutions.net ([78.46.109.217]:52664 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751495Ab1CPIUP (ORCPT ); Wed, 16 Mar 2011 04:20:15 -0400 Subject: Re: [PATCH] mac80211: fix aggregation frame release during timeout From: Johannes Berg To: Daniel Halperin Cc: linux-wireless In-Reply-To: <507B31BE-2066-4434-870D-108F7DF912A6@cs.washington.edu> References: <507B31BE-2066-4434-870D-108F7DF912A6@cs.washington.edu> Content-Type: text/plain; charset="UTF-8" Date: Wed, 16 Mar 2011 09:21:56 +0100 Message-ID: <1300263716.3761.1.camel@jlt3.sipsolutions.net> Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Tue, 2011-03-15 at 16:49 -0700, Daniel Halperin wrote: > x-T-R1-y-R2, > if (!time_after(jiffies, tid_agg_rx->reorder_time[j] + > - HT_RX_REORDER_BUF_TIMEOUT)) > + HT_RX_REORDER_BUF_TIMEOUT) && skipped) Wait, your previous example of xT---R worked fine, but if you say x-T- this patch won't work I think? Basically you're saying if we received frames 2 and 4 after 3, and 3 times out, we can release 2 through 4. I agree, but your code won't do that since skipped starts out at 1 due to the first x. Or am I misreading this? johannes