From mboxrd@z Thu Jan  1 00:00:00 1970
Return-path: <linux-wireless-owner@vger.kernel.org>
Received: from mail-pz0-f46.google.com ([209.85.210.46]:57290 "EHLO
	mail-pz0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752658Ab0GZQ4T convert rfc822-to-8bit (ORCPT
	<rfc822;linux-wireless@vger.kernel.org>);
	Mon, 26 Jul 2010 12:56:19 -0400
MIME-Version: 1.0
In-Reply-To: <AANLkTikrmM0pwBLLMWTCO-stRnoCxwQf0M3ZByc7=8_b@mail.gmail.com>
References: <4C3D5F3F.7070901@odi.ch> <AANLkTimn7Z3LqhLOt4_6ggFgjdzIZQPd_CznOIW-6D1N@mail.gmail.com>
	<4C499E6F.8030507@odi.ch> <AANLkTikrmM0pwBLLMWTCO-stRnoCxwQf0M3ZByc7=8_b@mail.gmail.com>
From: "Luis R. Rodriguez" <lrodriguez@atheros.com>
Date: Mon, 26 Jul 2010 09:55:58 -0700
Message-ID: <AANLkTi=Ga2Sy60=wXKQT3nv_PRYw-og7279bD9LTWENE@mail.gmail.com>
Subject: Re: BUG at arch/x86/mm/physaddr.c:5
To: Jike Song <albcamus@gmail.com>
Cc: =?UTF-8?Q?Ortwin_Gl=C3=BCck?= <odi@odi.ch>,
	linux-kernel@vger.kernel.org, Felix Fietkau <nbd@openwrt.org>,
	Vasanthakumar Thiagarajan <vasanth@atheros.com>,
	jmalinen@atheros.com, Sujith.Manoharan@atheros.com,
	senthilkumar@atheros.com, linux-wireless@vger.kernel.org
Content-Type: text/plain; charset=UTF-8
Sender: linux-wireless-owner@vger.kernel.org
List-ID: <linux-wireless.vger.kernel.org>

2010/7/25 Jike Song <albcamus@gmail.com>:
> 2010/7/23 Ortwin Glück <odi@odi.ch>:
>> On 23.07.2010 10:04, Jike Song wrote:
>>> Seems to be a logical error?  Does the following patch remove your Oops?
>>
>> Looking at the init code I agree. Yes, the patch fixes the OOPS. I am sure
>> because the OOPS occurred at every boot and now it's gone.
>>
>> Thanks!
>>
>> Ortwin
>
> Hi Maintainers,
>
>  Since Ortwin has confirmed this patch , would you give me an ACK/NAK please?
>
>
> --
> Thanks,
> Jike
>
>
> From 1414829dcc5b53ef4802f5de2b9be58e2ca0fb23 Mon Sep 17 00:00:00 2001
> From: Jike Song <albcamus@gmail.com>
> Date: Mon, 26 Jul 2010 13:42:41 +0800
> Subject: [PATCH] ath9k: fix wrong DMA direction in RX tasklet
> MIME-Version: 1.0
> Content-Type: text/plain; charset=utf-8
> Content-Transfer-Encoding: 8bit
>
> Commit b5c80475abaad015699384ca64ef8229fdd88758 introduced edma
> support for RX, hence bidirectional DMA support. But it
> specified the DMA type as DMA_FROM_DEVICE by mistake.
>
> Reported-and-tested-by: Ortwin Glück <odi@odi.ch>
> Signed-off-by: Jike Song <albcamus@gmail.com>
> ---
>  drivers/net/wireless/ath/ath9k/recv.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/wireless/ath/ath9k/recv.c
> b/drivers/net/wireless/ath/ath9k/recv.c
> index ca6065b..e3e5291 100644
> --- a/drivers/net/wireless/ath/ath9k/recv.c
> +++ b/drivers/net/wireless/ath/ath9k/recv.c
> @@ -844,9 +844,9 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush, bool hp)
>        int dma_type;
>
>        if (edma)
> -               dma_type = DMA_FROM_DEVICE;
> -       else
>                dma_type = DMA_BIDIRECTIONAL;
> +       else
> +               dma_type = DMA_FROM_DEVICE;
>
>        qtype = hp ? ATH9K_RX_QUEUE_HP : ATH9K_RX_QUEUE_LP;
>        spin_lock_bh(&sc->rx.rxbuflock);
> --

I'll let Felix chime in on this one as he last reviewed this path.
There were some serious issues with MIPs and DMA that at this point I
think only he grocked.

  Luis