Return-path: Received: from mail-wi0-f170.google.com ([209.85.212.170]:44487 "EHLO mail-wi0-f170.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751832AbbAGKOD convert rfc822-to-8bit (ORCPT ); Wed, 7 Jan 2015 05:14:03 -0500 Received: by mail-wi0-f170.google.com with SMTP id bs8so7729715wib.1 for ; Wed, 07 Jan 2015 02:14:01 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <1420432796-10765-1-git-send-email-sujith@msujith.org> References: <1420432796-10765-1-git-send-email-sujith@msujith.org> Date: Wed, 7 Jan 2015 11:14:01 +0100 Message-ID: (sfid-20150107_111408_880910_0BCA8C7E) Subject: Re: [PATCH] ath10k: Make HTT fill size configurable From: Michal Kazior To: Sujith Manoharan Cc: "ath10k@lists.infradead.org" , linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 5 January 2015 at 05:39, Sujith Manoharan wrote: > From: Sujith Manoharan > > The HTT RX ring is replenished with a maximum of 16 buffers, > but this might be insufficient when RX traffic is high. > Not having enough RX buffers throttles the FW, resulting > in low throughput. > > This patch adds a module parameter to adjust the fill size > based on the platform/usage. I don't think this should be adjustable. I assume you tested this in conjunction with rx batching patch - what fill level values did you use to see throughput improvement? How (un)responsive was serial/ssh connection to the machine? I have a patch internally which removes the replenishment tasklet and I wonder how will it behave with rx batching then, hmm.. Michał