Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <d9e3009b-1fd0-e81b-715e-d93c9343b55c@collabora.com>
Date:   Wed, 14 Jun 2023 13:18:48 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.11.2
Subject: Re: [PATCH] wifi: mt76: mt7921e: Perform FLR to recovery the device
Content-Language: en-US
To:     Kai-Heng Feng <kai.heng.feng@canonical.com>, nbd@nbd.name,
        lorenzo@kernel.org, ryder.lee@mediatek.com
Cc:     Shayne Chen <shayne.chen@mediatek.com>,
        Sean Wang <sean.wang@mediatek.com>,
        Kalle Valo <kvalo@kernel.org>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        Deren Wu <deren.wu@mediatek.com>,
        Leon Yen <Leon.Yen@mediatek.com>,
        linux-wireless@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org,
        linux-mediatek@lists.infradead.org
References: <20230614063252.1650824-1-kai.heng.feng@canonical.com>
From:   AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@collabora.com>
In-Reply-To: <20230614063252.1650824-1-kai.heng.feng@canonical.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

Il 14/06/23 08:32, Kai-Heng Feng ha scritto:
> When "Pre-boot WiFi" is enabled in BIOS, the mt7921e device may not
> work:
> mt7921e 0000:02:00.0: ASIC revision: 79220010
> mt7921e 0000:02:00.0: Message 00000010 (seq 1) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 2) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 3) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 4) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 5) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 6) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 7) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 8) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 9) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: Message 00000010 (seq 10) timeout
> mt7921e 0000:02:00.0: Failed to get patch semaphore
> mt7921e 0000:02:00.0: hardware init failed
> 
> After some trials and errors, it shows that PCI function level reset can
> recover the device back to a functional state.
> 
> So perform FLR at probe routine to workaround the issue.

Is there any way to check if the device was enabled before booting Linux?

I'm thinking of something like

if (device_is_already_enabled)
	ret = pci_reset_function_locked(....)

Regards,
Angelo

> 
> Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
> ---
>   drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
> index ddb1fa4ee01d..9671fbe35a8e 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
> +++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
> @@ -262,6 +262,10 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
>   	int ret;
>   	u16 cmd;
>   
> +	ret = pci_reset_function_locked(pdev);
> +	if (ret)
> +		pci_info(pdev, "Unable to perform FLR\n");
> +
>   	ret = pcim_enable_device(pdev);
>   	if (ret)
>   		return ret;