Re: [PATCH] coresight: etm4x: Skip setting LPOVERRIDE bit for qcom,skip-power-up

From: Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
To: Suzuki Poulose <suzuki.poulose@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>,
	Mike Leach <mike.leach@linaro.org>,
	coresight@lists.linaro.org, Stephen Boyd <swboyd@chromium.org>,
	denik@chromium.org, linux-arm-msm@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH] coresight: etm4x: Skip setting LPOVERRIDE bit for qcom,skip-power-up
Date: Fri, 16 Oct 2020 19:10:07 +0530	[thread overview]
Message-ID: <85ade254494144efc20c8c7512828654@codeaurora.org> (raw)
In-Reply-To: <9e19d312-9de4-2ed8-75ca-c774b93bfe11@arm.com>

Hi Suzuki,

On 2020-10-16 18:45, Suzuki Poulose wrote:
> On 10/16/20 12:47 PM, Sai Prakash Ranjan wrote:
>> Hi Suzuki,
>> 
>> On 2020-10-16 16:51, Suzuki Poulose wrote:
>>> Hi Sai,
>>> 
>>> On 10/16/20 11:10 AM, Sai Prakash Ranjan wrote:
>>>> There is a bug on the systems supporting to skip power up
>>>> (qcom,skip-power-up) where setting LPOVERRIDE bit(low-power
>>>> state override behaviour) will result in CPU hangs/lockups
>>>> even on the implementations which supports it. So skip
>>>> setting the LPOVERRIDE bit for such platforms.
>>>> 
>>>> Fixes: 02510a5aa78d ("coresight: etm4x: Add support to skip trace 
>>>> unit power up")
>>>> Signed-off-by: Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
>>> 
>>> The fix is fine by me. Btw, is there a hardware Erratum assigned for
>>> this ? It would be good to have the Erratum documented somewhere,
>>> preferrably ( Documentation/arm64/silicon-errata.rst )
>>> 
>> 
>> No, afaik we don't have any erratum assigned to this bug.
> 
> Ok. Please double check, if there are any.
> 

Sure I will check again.

>> It was already present in downstream kernel and since we
>> support these targets with the previous HW bug
>> (qcom,skip-power-up) now in upstream, we would need this
>> fix in upstream kernel as well.
> 
> I understand the need for the fix and we must fix it. I was
> looking to document this in the central place for errata's
> handled in the kernel. And I missed asking this question
> when the original patch was posted. So, thought of asking
> the question now anyways. Better late than never ;-)
> 
> Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>

Thanks. One more thing, does the internal erratum
number (if it exists) is good enough to be documented in
the Documentation/arm64/silicon-errata.rst ? I ask this
because outside qualcomm, it won't mean much right.

Thanks,
Sai

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a 
member
of Code Aurora Forum, hosted by The Linux Foundation