Return-path: Received: from nbd.name ([46.4.11.11]:40618 "EHLO nbd.name" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753134Ab1ISPDv (ORCPT ); Mon, 19 Sep 2011 11:03:51 -0400 Message-ID: <4E7759CA.4010100@openwrt.org> (sfid-20110919_170354_393187_373262D2) Date: Mon, 19 Sep 2011 17:03:38 +0200 From: Felix Fietkau MIME-Version: 1.0 To: Rajkumar Manoharan CC: linux-wireless@vger.kernel.org, linville@tuxdriver.com, rodrigue@qca.qualcomm.com Subject: Re: [PATCH 12/12] ath9k: do not insert padding into tx buffers on AR9380+ References: <1316028267-34187-3-git-send-email-nbd@openwrt.org> <1316028267-34187-4-git-send-email-nbd@openwrt.org> <1316028267-34187-5-git-send-email-nbd@openwrt.org> <1316028267-34187-6-git-send-email-nbd@openwrt.org> <1316028267-34187-7-git-send-email-nbd@openwrt.org> <1316028267-34187-8-git-send-email-nbd@openwrt.org> <1316028267-34187-9-git-send-email-nbd@openwrt.org> <1316028267-34187-10-git-send-email-nbd@openwrt.org> <1316028267-34187-11-git-send-email-nbd@openwrt.org> <1316028267-34187-12-git-send-email-nbd@openwrt.org> <20110919144002.GA22040@vmraj-lnx.users.atheros.com> In-Reply-To: <20110919144002.GA22040@vmraj-lnx.users.atheros.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: On 2011-09-19 4:40 PM, Rajkumar Manoharan wrote: > On Wed, Sep 14, 2011 at 09:24:27PM +0200, Felix Fietkau wrote: >> With the new EDMA descriptor format, a single descriptor can contain up >> to four buffer pointers. By splitting the buffer into two parts, we can >> let the hardware add the padding internally instead of using memmove on >> the skb data. >> >> Signed-off-by: Felix Fietkau >> --- >> drivers/net/wireless/ath/ath9k/xmit.c | 56 +++++++++++++++++++++----------- >> 1 files changed, 37 insertions(+), 19 deletions(-) >> > Felix, > > This patch again introduced "ath: Failed to stop TX DMA!" for AR9380+ chips. > After reverting this, the errors are completely gone away. Please look into this ASAP. How frequent are these errors in your tests? I ran several tests on embedded hardware with AR9380 and didn't notice any issues. - Felix