On 27/09/2021 18:45, Guilherme G. Piccoli wrote:

Hi Yassine, thanks for the patch! I was doing some experiments with a
dragonboard 820-based board, and without your patch, a CPU benchmark can
quickly cause overheating - throttling mechanism doesn't start and we
get a FW reset to a bad state (likely a HW protection mechanism).

I noticed that a similar patch including cooling maps is present in
Linaro's qcom tree [0], and it was submitted upstream [1], but there was
a re-submission [2] by Konrad that was merged and dropped the thermals
part. Based on some threads I read, it seems a FW lockup was detected
with the complete patch?

I'm not sure, so I'm looping Konrad / Loic / Bjorn here, to understand
better what made this portion of the patch to be dropped.
Anyway, I think worth to mention this in your commit message Yassine,
including perhaps a fixes tag like:

Fixes: 90173a954a22 ("arm64: dts: qcom: msm8996: Add CPU opps").

Also, I'm not sure why there was no response here or why it wasn't
merged, but if it helps, please have my:

Tested-by: Guilherme G. Piccoli <gpiccoli@igalia.com>

Thanks,


Guilherme


[0]
https://git.linaro.org/landing-teams/working/qualcomm/kernel.git/commit/?h=release/qcomlt-4.14&id=2274c48c671

[1]
https://patchwork.kernel.org/project/linux-arm-msm/patch/1595253740-29466-6-git-send-email-loic.poulain@linaro.org/

[2]
https://lore.kernel.org/linux-arm-msm/20210527194455.782108-2-konrad.dybcio@somainline.org/ (local)

Unfortunately, (likely) due to a typo in the Subject field, this wasn't
threaded correctly. So, the original message I was responding is:

https://lore.kernel.org/linux-arm-msm/jmayJcXoExAK2G7UBIXMz5CDN0BYgYkFZguHlPNRFOU@cp4-web-038.plabs.ch/ (local)

I'm sorry about the confusion with the mail thread!
Cheers,


Guilherme