[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1b7d4c2f-98f7-4b80-4e32-6e0fe5c906ce@arm.com>
Date: Fri, 18 Jan 2019 12:04:20 +0000
From: Suzuki K Poulose <suzuki.poulose@....com>
To: saiprakash.ranjan@...eaurora.org, robh+dt@...nel.org,
mathieu.poirier@...aro.org, leo.yan@...aro.org,
alexander.shishkin@...ux.intel.com, andy.gross@...aro.org,
david.brown@...aro.org, vivek.gautam@...eaurora.org,
dianders@...omium.org, sboyd@...nel.org,
bjorn.andersson@...aro.org, devicetree@...r.kernel.org,
mark.rutland@....com
Cc: rnayak@...eaurora.org, sibis@...eaurora.org,
linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
linux-arm-msm@...r.kernel.org
Subject: Re: [PATCHv2 2/4] arm64: dts: qcom: msm8996: Add Coresight support
Hi,
On 18/01/2019 11:48, Sai Prakash Ranjan wrote:
> Hi Suzuki,
>
> On 1/18/2019 4:27 PM, Suzuki K Poulose wrote:
>>
>> If you want to use the scatter-gather mode, you must add
>> "arm,scatter-gather"
>> property to the node here.
>>
>> Have you tested the scatter-gather mode on the chip ? Does that work at
>> all ?
>> Note: You could try this by setting a buffer_size > 1M for the ETR and use
>> it as a sink from the sysfs mode.
>>
>
> Yes the chip supports scatter-gather mode[1]. But is it required in
> dt node since it's an optional property?
If you would like to use the scatter-gather mode, you must specify it
explicitly. We have had some other boards with broken scatter-gather
mode integration which causes mysterious issues when used. So we have
a white-list approach for enabling the scatter-gather mode in the driver.
So, if you have tested to make sure that the mode works fine with a recent
kernel version (v4.19 onwards), you may want to add the property.
Suzuki
Powered by blists - more mailing lists