Return-path: Received: from nbd.name ([88.198.39.176]:49877 "EHLO ds10.nbd.name" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754310Ab0JPNKP (ORCPT ); Sat, 16 Oct 2010 09:10:15 -0400 Message-ID: <4CB9A433.5050007@openwrt.org> Date: Sat, 16 Oct 2010 15:10:11 +0200 From: Felix Fietkau MIME-Version: 1.0 To: Rajkumar Manoharan CC: "linux-wireless@vger.kernel.org" , "linville@tuxdriver.com" , Luis Rodriguez Subject: Re: [PATCH 1/6] ath9k_hw: optimize interrupt mask changes References: <1287165813-94331-1-git-send-email-nbd@openwrt.org> <20101016115041.GA4028@vmraj-lnx.users.atheros.com> In-Reply-To: <20101016115041.GA4028@vmraj-lnx.users.atheros.com> Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 2010-10-16 1:51 PM, Rajkumar Manoharan wrote: > On Fri, Oct 15, 2010 at 11:33:28PM +0530, Felix Fietkau wrote: >> @@ -751,8 +751,8 @@ chip_reset: >> ath_debug_stat_interrupt(sc, status); >> >> if (sched) { >> - /* turn off every interrupt except SWBA */ >> - ath9k_hw_set_interrupts(ah, (ah->imask & ATH9K_INT_SWBA)); >> + /* turn off every interrupt */ >> + ath9k_hw_disable_interrupts(ah); > > won't it affect beacon generation? No, because the original code did not leave SWBA interrupts enabled either, as it was masking out ATH9K_INT_GLOBAL. So far I haven't seen issues with beacons under high load, even on slow embedded hardware. - Felix