Return-path: Received: from mail-wg0-f41.google.com ([74.125.82.41]:58295 "EHLO mail-wg0-f41.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751108AbbBPGyT convert rfc822-to-8bit (ORCPT ); Mon, 16 Feb 2015 01:54:19 -0500 Received: by mail-wg0-f41.google.com with SMTP id b13so27591091wgh.0 for ; Sun, 15 Feb 2015 22:54:17 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <87mw4fgolp.fsf@kamboji.qca.qualcomm.com> References: <1423749198-11346-1-git-send-email-michal.kazior@tieto.com> <87mw4fgolp.fsf@kamboji.qca.qualcomm.com> Date: Mon, 16 Feb 2015 07:54:17 +0100 Message-ID: (sfid-20150216_075423_264246_6FAD624B) Subject: Re: [PATCH] ath10k: fix AP/IBSS CSA with template based fw From: Michal Kazior To: Kalle Valo Cc: "ath10k@lists.infradead.org" , linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 15 February 2015 at 16:51, Kalle Valo wrote: > Michal Kazior writes: > >> qca6174 with wmi-tlv firmware uses offloaded >> beaconing scheme (i.e. templates). This requires a >> little different approach when implementing CSA. >> >> Add missing code to update CS count and report CSA >> completion to mac80211. Without it channel switch >> was never finished. >> >> To avoid races during interface teardown data_lock >> has been used to protect is_up and is_started so >> they can be compared against before scheduling >> count down work. >> >> Signed-off-by: Michal Kazior >> --- >> >> @Kalle: I'm deferring the chanctx patchset v4 >> until this gets applied. I'd like to avoid posting >> patches without a valid parent id and leaving you >> with some weird conflicts and implicit depndencies >> in the code to deal with. > > Ok. > > I actually had strange conflicts already with this patch (in mac.c). As > I'm travelling, please check my resolution from the pending branch very > carefully. I might have done something stupid. There's a mistake. Only `arvif->is_up` should be protected by `ar->data_lock` in ath10k_bss_assoc(): @@ -1852,7 +1901,9 @@ static void ath10k_bss_assoc(struct ieee80211_hw *hw, return; } + spin_lock_bh(&arvif->ar->data_lock); arvif->is_up = true; + spin_unlock_bh(&arvif->ar->data_lock); /* Workaround: Some firmware revisions (tested with qca6174 (...) } Michał