Received-SPF: pass (google.com: domain of linux-wireless+bounces-6295-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Sender: Larry Finger <larry.finger@gmail.com>
Message-ID: <4d36d201-bb39-469c-9861-5e08c9003c73@lwfinger.net>
Date: Sun, 14 Apr 2024 20:23:22 -0500
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [RFC] rtw88: Fix startup problems for SDIO wifi plus UART
 Bluetooth
To: Ping-Ke Shih <pkshih@realtek.com>, Kalle Valo <kvalo@kernel.org>
Cc: Larry Finger <Larry.Finger@gmail.com>,
 Johannes Berg <johannes@sipsolutions.net>,
 "linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>
References: <4umro86wvv84.MjDfYvt4P5uZryt8boBK8Q2@1EHFQ.trk.elasticemail.com>
 <5af71338f3904aac9d2c237637c851e7@realtek.com>
 <287e9d4e-316a-4579-961e-58e75abea534@lwfinger.net>
 <87o7agjs00.fsf@kernel.org>
 <46745fe7-a43d-4f7c-b247-f0fde01ee63e@lwfinger.net>
 <d3141ad7a7a9455f8a6dbcdd3f46eecf@realtek.com>
Content-Language: en-US
From: Larry Finger <Larry.Finger@lwfinger.net>
In-Reply-To: <d3141ad7a7a9455f8a6dbcdd3f46eecf@realtek.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

On 4/14/24 7:45 PM, Ping-Ke Shih wrote:
> Larry Finger <larry.finger@gmail.com> wrote:
>>
>> I originally wondered if there was a power problem when both were simultaneously
>> started, but I see that is not the case. To reiterate, the BT must start first
>> and complete UART initialization before the wifi starts initializing. It seems
>> to me that a suitable msleep() at the start of the SDIO probe routine seems to
>> be a viable workaround even though it is not aesthetically pleasing. A sleep of
>> 150 ms is too short, but 500 seems to work reliably.
> 
> Suggest to apply this workaround but only if UART-BT + SDIO-WiFi, because SDIO
> experts didn't remember they have met this problem. They need real hardware to
> measure signals to know what it is wrong, but unfortunately they don't have
> bandwidth to process this because of limitation of human resources. Sorry
> for that.

The OP at GitHub reported today that there was a DTB error for that chip [1]. I 
am going to drop the patch for the stable kernel, but I will still apply it to 
my rtw88 repo at GitHub.com. That way, users with older kernels will get the 
benefit of the msleep() even though their DTB may not be fixed.

Thanks to you and the SDIO team for your efforts toward a problem that turned 
out to be in the kernel. Sorry for some noise.

Larry


[1] 
https://github.com/ROCKNIX/distribution/pull/63/files#diff-e2de6222b1794f89311bdc1597c1597c76f34503c575ea3f7e7ec9c5376218d6