Return-path: Received: from sabertooth01.qualcomm.com ([65.197.215.72]:47441 "EHLO sabertooth01.qualcomm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751709AbaBQPBy (ORCPT ); Mon, 17 Feb 2014 10:01:54 -0500 From: Kalle Valo To: Michal Kazior CC: , Subject: Re: [RFC/RFT 0/7] ath10k: performance improvements References: <1392629563-31046-1-git-send-email-michal.kazior@tieto.com> Date: Mon, 17 Feb 2014 17:01:49 +0200 In-Reply-To: <1392629563-31046-1-git-send-email-michal.kazior@tieto.com> (Michal Kazior's message of "Mon, 17 Feb 2014 10:32:36 +0100") Message-ID: <87y5193hr6.fsf@kamboji.qca.qualcomm.com> (sfid-20140217_160220_168825_1012C0D6) MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: linux-wireless-owner@vger.kernel.org List-ID: Michal Kazior writes: > These patches aim at reducing host CPU load and > thus improve performance on low-end systems. > > In my setup I get a relative improvement of > 100mbps for both UDP Tx and Rx. Nice! > Tx -----> > laptop:eth---eth:AP135:ath10k---ath10k:laptop > <----- Rx > > > Michal Kazior (7): > ath10k: remove DMA mapping wrappers > ath10k: remove is_aborted from skb_cb > ath10k: replace send_head() with tx_sg() > ath10k: bypass htc for htt tx path > ath10k: batch htt tx/rx completions > ath10k: remove pci completion list > ath10k: minimize coherent dma accesses I know this is RFC still, but I did some compilation tests and saw these: drivers/net/wireless/ath/ath10k/htt_rx.c:274:17: warning: context imbalance in 'ath10k_htt_rx_amsdu_pop' - unexpected unlock drivers/net/wireless/ath/ath10k/htt_tx.c:436: WARNING: line over 80 characters drivers/net/wireless/ath/ath10k/htt_tx.c:493: WARNING: line over 80 characters drivers/net/wireless/ath/ath10k/htt_rx.c:1214: WARNING: line over 80 characters -- Kalle Valo