From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail-lf0-f66.google.com ([209.85.215.66]:33852 "EHLO mail-lf0-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750903AbdJBJGp (ORCPT ); Mon, 2 Oct 2017 05:06:45 -0400 Received: by mail-lf0-f66.google.com with SMTP id h80so1969805lfe.1 for ; Mon, 02 Oct 2017 02:06:44 -0700 (PDT) Subject: Re: [PATCH 02/11] ath10k_sdio: wb396 reference card fix To: Alagu Sankar , Steve deRosier Cc: silexcommon@gmail.com, linux-wireless , ath10k@lists.infradead.org References: <1506793068-27445-1-git-send-email-alagusankar@silex-india.com> <1506793068-27445-3-git-send-email-alagusankar@silex-india.com> From: Erik Stromdahl Message-ID: <427bfce1-5adb-9cad-b674-e42f8005bb02@gmail.com> (sfid-20171002_110459_463768_722ACABC) Date: Mon, 2 Oct 2017 11:06:42 +0200 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: Hi Alagu, On 2017-10-02 09:02, Alagu Sankar wrote: > Hi Steve, > > On 2017-10-02 04:17, Steve deRosier wrote: >> Hi Alagu, >> >> >> On Sat, Sep 30, 2017 at 10:37 AM, wrote: >>> >>> From: Alagu Sankar >>> >>> The QCA9377-3 WB396 sdio reference card does not get initialized >>> due to the conflict in uart gpio pins. This fix is not required >>> for other QCA9377 sdio cards. >>> >>> Signed-off-by: Alagu Sankar >>> --- >>>  drivers/net/wireless/ath/ath10k/core.c | 9 ++++++++- >>>  1 file changed, 8 insertions(+), 1 deletion(-) >>> >>> diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c >>> index b4f66cd..86247c8 100644 >>> --- a/drivers/net/wireless/ath/ath10k/core.c >>> +++ b/drivers/net/wireless/ath/ath10k/core.c >>> @@ -1708,8 +1708,15 @@ static int ath10k_init_uart(struct ath10k *ar) >>>                 return ret; >>>         } >>> >>> -       if (!uart_print) >>> +       if (!uart_print) { >>> +               /* Hack: override dbg TX pin to avoid side effects of default >>> +                * GPIO_6 in QCA9377 WB396 reference card >>> +                */ >>> +               if (ar->hif.bus == ATH10K_BUS_SDIO) >>> +                       ath10k_bmi_write32(ar, hi_dbg_uart_txpin, >>> +                                          ar->hw_params.uart_pin); >> >> If it is indeed a "hack", then I don't think the maintainer should >> accept this upstream. If you want it upstream you need a clean enough >> implementation that doesn't need to be labeled a "hack". > > It is a hack as per the qcacld reference driver. > >> Your commit message states that this is only needed for a very >> specific card and not for other QCA9377 sdio cards. Yet, you're doing >> this for all ATH10K_BUS_SDIO devices. Not good. I think that it's a >> quirk and it's limited to a particular implementation of the device. >> My suggestion: if it can be automatically determined, then do so >> explicitly. If not, then it needs to be a DT setting or a module >> parameter or something like that so the platform maker can decide to >> do it. Having it affect all users of a SDIO QCA9377 when it doesn't >> apply doesn't seem like a good idea to me. >> >> >> - Steve > > Got it. The qcacld reference driver had it for all the QCA9377 sdio cards. > But we found it to be a problem only for the WB396 reference card. Will > have this checked again and release a v2 patch accordingly. > While you are at it, you might as well change the commit comments to: "ath10k: sdio: " or perhaps just: "ath10k: " > Best Regards, > Alagu Sankar > > _______________________________________________ > ath10k mailing list > ath10k@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/ath10k