Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v5] brcmfmac: firmware: Fix firmware loading
To:     Arend van Spriel <arend.vanspriel@broadcom.com>,
        Arend van Spriel <aspriel@gmail.com>,
        Franky Lin <franky.lin@broadcom.com>,
        Hante Meuleman <hante.meuleman@broadcom.com>,
        Chi-hsien Lin <chi-hsien.lin@infineon.com>,
        Wright Feng <wright.feng@infineon.com>,
        Chung-hsien Hsu <chung-hsien.hsu@infineon.com>,
        Kalle Valo <kvalo@codeaurora.org>,
        Linus Walleij <linus.walleij@linaro.org>
Cc:     linux-wireless@vger.kernel.org,
        Stefan Hansson <newbyte@disroot.org>
References: <20210808180510.8753-1-digetx@gmail.com>
 <902984be-2c27-e6b0-247b-8d9a94c08cc4@broadcom.com>
From:   Dmitry Osipenko <digetx@gmail.com>
Message-ID: <36277b51-b647-898c-c1b4-a0e363c52b0d@gmail.com>
Date:   Mon, 9 Aug 2021 17:56:43 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.11.0
MIME-Version: 1.0
In-Reply-To: <902984be-2c27-e6b0-247b-8d9a94c08cc4@broadcom.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk

09.08.2021 11:23, Arend van Spriel пишет:
> On 8/8/2021 8:05 PM, Dmitry Osipenko wrote:
>> From: Linus Walleij <linus.walleij@linaro.org>
>>
>> The patch that would first try the board-specific firmware
>> had a bug because the fallback would not be called: the
>> asynchronous interface is used meaning request_firmware_nowait()
>> returns 0 immediately.
>>
>> Harden the firmware loading like this:
>>
>> - If we cannot build an alt_path (like if no board_type is
>>    specified) just request the first firmware without any
>>    suffix, like in the past.
>>
>> - If the lookup of a board specific firmware fails, we get
>>    a NULL fw in the async callback, so just try again without
>>    the alt_path from a dedicated brcm_fw_request_done_alt_path
>>    callback.
>>
>> - Drop the unnecessary prototype of brcm_fw_request_done.
>>
>> - Added MODULE_FIRMWARE match for per-board SDIO bins, making
>>    userspace tools to pull all the relevant firmware files.
> The original idea was to setup the path names in
> brcmf_fw_alloc_request() function, but with the introduction of the
> board_type for NVRAM files that was abandoned and we cook up alternative
> paths. Now similar is done for the firmware files. So I would want to
> rework the code, but for now I am going with Linus's/Your fix for the
> sake of having the regression more or less quickly resolved.

Thanks, indeed it could be improved further later on.

> You reported an issue earlier where the firmware callback was called
> from the probe context causing it to hang and it is not clear to me
> whether that is fixed with this version of the patch.

It's independent minor problem that can't be easily reproduced in
practice and seems it existed for a long time. It's not fixed by this patch.