Return-path: <linux-wireless-owner@vger.kernel.org>
Received: from wolverine01.qualcomm.com ([199.106.114.254]:31296 "EHLO
	wolverine01.qualcomm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751963Ab2GIGqK (ORCPT
	<rfc822;linux-wireless@vger.kernel.org>);
	Mon, 9 Jul 2012 02:46:10 -0400
From: Sujith Manoharan <c_manoha@qca.qualcomm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Message-ID: <20474.32213.21743.932687@gargle.gargle.HOWL> (sfid-20120709_084615_164909_0D07C56C)
Date: Mon, 9 Jul 2012 12:14:37 +0530
To: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
CC: <linville@tuxdriver.com>, <linux-wireless@vger.kernel.org>
Subject: [PATCH] ath9k: fix reset work check properly
In-Reply-To: <1341815640-6299-1-git-send-email-rmanohar@qca.qualcomm.com>
References: <1341815640-6299-1-git-send-email-rmanohar@qca.qualcomm.com>
Sender: linux-wireless-owner@vger.kernel.org
List-ID: <linux-wireless.vger.kernel.org>

Rajkumar Manoharan wrote:
> Once the hw reset work is queued up and to bail out current
> execution properly, use HW_RESET bit ops instead of work_pending.
> As work_pending might return false when the queued work is in
> execution. So it is not correct to use the work utility for
> baining out.

I have one comment, though.

> diff --git a/drivers/net/wireless/ath/ath9k/beacon.c b/drivers/net/wireless/ath/ath9k/beacon.c
> index 40775da..567a655 100644
> --- a/drivers/net/wireless/ath/ath9k/beacon.c
> +++ b/drivers/net/wireless/ath/ath9k/beacon.c
> @@ -362,7 +362,7 @@ void ath_beacon_tasklet(unsigned long data)
>  	int slot;
>  	u32 bfaddr, bc = 0;
>  
> -	if (work_pending(&sc->hw_reset_work)) {
> +	if (test_bit(SC_OP_HW_RESET, &sc->sc_flags)) {

This is not needed, since we bail out in ath_isr().
(And interrupts are re-enabled only at the end of ath9k_tasklet()).

Sujith