From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from wolverine02.qualcomm.com ([199.106.114.251]:34767 "EHLO wolverine02.qualcomm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754111AbbJPMwH (ORCPT ); Fri, 16 Oct 2015 08:52:07 -0400 From: Kalle Valo To: Rajkumar Manoharan CC: , Subject: Re: [PATCH v2 0/7] ath10k: add copy engine fast path support References: <1444654626-3290-1-git-send-email-rmanohar@qti.qualcomm.com> Date: Fri, 16 Oct 2015 15:51:44 +0300 In-Reply-To: <1444654626-3290-1-git-send-email-rmanohar@qti.qualcomm.com> (Rajkumar Manoharan's message of "Mon, 12 Oct 2015 18:26:59 +0530") Message-ID: <87pp0fgen3.fsf@kamboji.qca.qualcomm.com> (sfid-20151016_145220_665490_1DCA8F8A) MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: linux-wireless-owner@vger.kernel.org List-ID: Rajkumar Manoharan writes: > Below patchset adds fast path support for uplink traffic by bypassing > HTC layer processing. This is enabled by making use of unused copy > engine 5 to receive HTT messages directly from HIF layer. From initial > validation in VHT80/5G mode TCP UL is improved to 900Mbps from ~840Mbps > in conducted test. > > -Rajkumar > > v2: > * fix invalid dma memory access (ATH10K_SKB_RXCB is used instead of > ATH10K_SKB_CB in htt_tx_cb) > * register either send_cb or recv_cb callback not both for > appropriate CE. > * Process CE 4 send completion first before processing rx > > Rajkumar Manoharan (7): > ath10k: export htc tx rx handlers > ath10k: register per copy engine send completion callbacks > ath10k: register per copy engine receive callbacks > ath10k: export htt tx rx handlers > ath10k: Configure copy engine 5 for HTT messages > ath10k: remove unused dl_is_polled > ath10k: remove htc polling for tx completion All 7 applied, thanks. I wasn't able to reproduce the warning anymore, tested with both qca988x and qca99x0. -- Kalle Valo