From mboxrd@z Thu Jan 1 00:00:00 1970 From: chansun.xjs Date: Tue, 01 May 2012 10:54:37 +0800 Subject: [ath9k-devel] =?utf-8?b?5Zue5aSNOiAgUmU6IFJlOiBSZTogRnJhbWUgQWdn?= =?utf-8?q?regation_and_Block_ACK_pros?= In-Reply-To: References: <4f9e321a.c8d8440a.6880.529e@mx.google.com> <4f9e450c.a886440a.7395.ffffd8fb@mx.google.com> <4f9f29f8.4114440a.2806.ffffb61e@mx.google.com> Message-ID: <4f9f507e.0686440a.3e6d.ffffc23f@mx.google.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ath9k-devel@lists.ath9k.org Thanks a lot, I will try it 2012-05-01 chansun.xjs ????Adrian Chadd ?????2012-05-01 10:48 ???Re: Re: Re: Frame Aggregation and Block ACK pros ????"chansun.xjs" ???"Felix Fietkau","Mohammed Shafi","fredmm","Daniel Smith","Luis R","ath9k-devel" On 30 April 2012 17:10, chansun.xjs wrote: > Yes,appreciate for your tips, but what I am paying attention to is enabling > normal A-MPDU without Block ACK. Is there any possibility to do that? Any > guides? I personally have no idea, sorry. I haven't tried doing it so I don't know if it's possible. You could test for yourself though - you just need to hack up ath9k to override: * when queuing a set of aggregate frames to the hardware, override the TX descriptor ACK setting (ie, forcibly disable ACK in the TX descriptor when it's an aggregate); * always treat sub-frames as "ACKed" when handling an aggregate TX completion (as if it did work, the BA bitmap is likely to be all 0's..) Adrian -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20120501/e7c0544c/attachment-0001.htm