Re: [PATCH] arm64: dts: qcom: sm8350: fix BAM DMA crash and reboot

From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
To: Luca Weiss <luca.weiss@fairphone.com>,
	Andy Gross <agross@kernel.org>,
	Bjorn Andersson <andersson@kernel.org>,
	Konrad Dybcio <konrad.dybcio@linaro.org>,
	Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Bhupesh Sharma <bhupesh.sharma@linaro.org>,
	linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Subject: Re: [PATCH] arm64: dts: qcom: sm8350: fix BAM DMA crash and reboot
Date: Mon, 26 Jun 2023 15:24:04 +0200	[thread overview]
Message-ID: <593b3b3e-7613-734d-91a9-9a48bbe5a9da@linaro.org> (raw)
In-Reply-To: <7b909a56-1b3d-79fb-0286-b10117982dc3@linaro.org>

On 26/06/2023 14:46, Krzysztof Kozlowski wrote:
> On 26/06/2023 13:32, Luca Weiss wrote:
>> Hi Krzysztof,
>>
>> On Wed Jun 21, 2023 at 4:36 PM CEST, Krzysztof Kozlowski wrote:
>>> SM8350 HDK and MTP boards were silently dying and rebooting during BAM
>>> DMA probe:
>>>
>>>   [    1.574304] vreg_bob: Setting 3008000-3960000uV
>>>   [    1.576918] bam-dFormat: Log Type - Time(microsec) - Message -
>>>   Optional Info
>>>   Log Type: B - Since Boot(Power On Reset),  D - Delta,  S - Statistic
>>>   S - QC_IMAGE_VERSION_STRING=BOOT.MXF.1.0-00637.1-LAHAINA-1
>>>   S - IMAGE_VARIANT_STRING=SocLahainaLAA
>>>   S - OEM_IMAGE_VERSION_STRING=crm-ubuntu77
>>>   S - Boot Interface: UFS
>>>
>>> It seems that BAM DMA is locally controller (not by firmware) and
>>> requires proper initialization by the driver prior to use, at least on
>>> HDK8350 and MTP8350, but probably on all boards.
>>
>> Are you sure that the bam (and subsequent the qce) actually probes with
>> this change? From reading the code I don't see how the bam should probe
>> without either qcom,controlled-remotely or qcom,powered-remotely but no
> 
> Why the binding does not require either this or that? Eh, buggy stuff...
> 
> 
>> clocks supplied. I think the probe just fails with this change, right?
> 
> I will need to double check. I was happy enough to be able to boot my
> device instead of having crashes, but indeed it would be nice to fix it
> fully.

You were right and my patch is not correct. I checked downstream and
there are no clocks there. Therefore I assume something else is wrong here.

I will send v2 disabling this.

Best regards,
Krzysztof