From: Konrad Dybcio <konrad.dybcio@linaro.org>
To: Bjorn Andersson <andersson@kernel.org>
Cc: Andy Gross <agross@kernel.org>, Rob Herring <robh+dt@kernel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
Marijn Suijten <marijn.suijten@somainline.org>,
linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] arm64: dts: qcom: sm6115: Set up CPU cooling maps
Date: Tue, 20 Jun 2023 15:19:26 +0200 [thread overview]
Message-ID: <29e4d965-787c-b832-3916-722bad496fae@linaro.org> (raw)
In-Reply-To: <965bc277-b31b-e278-a793-90248615c9d6@linaro.org>
On 31.05.2023 16:55, Konrad Dybcio wrote:
>
>
> On 18.04.2023 17:51, Bjorn Andersson wrote:
>> On Tue, Apr 18, 2023 at 03:01:47PM +0200, Konrad Dybcio wrote:
>>>
>>>
>>> On 18.04.2023 15:02, Bjorn Andersson wrote:
>>>> On Tue, Apr 18, 2023 at 01:56:56PM +0200, Konrad Dybcio wrote:
>>>>> Set up CPU cooling maps to ensure the thermal framework is aware of
>>>>> the connection between the CPUs and the TSENS sensors.
>>>>>
>>>>> All of the maps refer to all 4 CPUs within a given cluster at a time,
>>>>> as that's what can be considered the smallest DVFS target unit - they
>>>>> all share the same voltage line and clock source.
>>>>>
>>>>
>>>> Generally software based CPU cooling is considered too slow to cope with
>>>> CPU core temperature changes, and the limits hardware working together
>>>> with OSM/EPSS will do a better job maintaining appropriate core
>>>> temperature levels.
>>>>
>>>> Is there a reason why this can't be used/relied upon on this platform?
>>> I haven't set up LMH yet and the default limits, at least with my dubious
>>> meta build, seem to let the board go to 75degC with just 4 cores working..
>>>
>>> Not sure if there's a voltage droop when I let it go full throttle or
>>> something, but pushing it to the limit definitely causes the board to be
>>> (even) less stable..
>>>
>>>>
>>>>
>>>> PS. Amending this mechanism with software based cooling to keep the
>>>> system at a reasonable/lower temperature is a good idea.
>>> We don't like burned chips around here!
>>>
>>
>> Limits hardware will help you with that, software based cooling will
>> help avoid burning the user.
> So, are there any reasons not to take this patch?
Bump
Konrad
>
> Konrad
>>
>> Regards,
>> Bjorn
>>
>>> Konrad
>>>>
>>>> Regards,
>>>> Bjorn
>>>>
>>>>> Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
>>>>> ---
>>>>> arch/arm64/boot/dts/qcom/sm6115.dtsi | 137 +++++++++++++++++++++++++++++++++++
>>>>> 1 file changed, 137 insertions(+)
>>>>>
>>>>> diff --git a/arch/arm64/boot/dts/qcom/sm6115.dtsi b/arch/arm64/boot/dts/qcom/sm6115.dtsi
>>>>> index 631ca327e064..36ff913c1a60 100644
>>>>> --- a/arch/arm64/boot/dts/qcom/sm6115.dtsi
>>>>> +++ b/arch/arm64/boot/dts/qcom/sm6115.dtsi
>>>>> @@ -12,6 +12,7 @@
>>>>> #include <dt-bindings/gpio/gpio.h>
>>>>> #include <dt-bindings/interrupt-controller/arm-gic.h>
>>>>> #include <dt-bindings/power/qcom-rpmpd.h>
>>>>> +#include <dt-bindings/thermal/thermal.h>
>>>>>
>>>>> / {
>>>>> interrupt-parent = <&intc>;
>>>>> @@ -47,6 +48,8 @@ CPU0: cpu@0 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_0>;
>>>>> qcom,freq-domain = <&cpufreq_hw 0>;
>>>>> + #cooling-cells = <2>;
>>>>> +
>>>>> L2_0: l2-cache {
>>>>> compatible = "cache";
>>>>> cache-level = <2>;
>>>>> @@ -63,6 +66,7 @@ CPU1: cpu@1 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_0>;
>>>>> qcom,freq-domain = <&cpufreq_hw 0>;
>>>>> + #cooling-cells = <2>;
>>>>> };
>>>>>
>>>>> CPU2: cpu@2 {
>>>>> @@ -75,6 +79,7 @@ CPU2: cpu@2 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_0>;
>>>>> qcom,freq-domain = <&cpufreq_hw 0>;
>>>>> + #cooling-cells = <2>;
>>>>> };
>>>>>
>>>>> CPU3: cpu@3 {
>>>>> @@ -87,6 +92,7 @@ CPU3: cpu@3 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_0>;
>>>>> qcom,freq-domain = <&cpufreq_hw 0>;
>>>>> + #cooling-cells = <2>;
>>>>> };
>>>>>
>>>>> CPU4: cpu@100 {
>>>>> @@ -99,6 +105,8 @@ CPU4: cpu@100 {
>>>>> dynamic-power-coefficient = <282>;
>>>>> next-level-cache = <&L2_1>;
>>>>> qcom,freq-domain = <&cpufreq_hw 1>;
>>>>> + #cooling-cells = <2>;
>>>>> +
>>>>> L2_1: l2-cache {
>>>>> compatible = "cache";
>>>>> cache-level = <2>;
>>>>> @@ -115,6 +123,7 @@ CPU5: cpu@101 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_1>;
>>>>> qcom,freq-domain = <&cpufreq_hw 1>;
>>>>> + #cooling-cells = <2>;
>>>>> };
>>>>>
>>>>> CPU6: cpu@102 {
>>>>> @@ -127,6 +136,7 @@ CPU6: cpu@102 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_1>;
>>>>> qcom,freq-domain = <&cpufreq_hw 1>;
>>>>> + #cooling-cells = <2>;
>>>>> };
>>>>>
>>>>> CPU7: cpu@103 {
>>>>> @@ -139,6 +149,7 @@ CPU7: cpu@103 {
>>>>> enable-method = "psci";
>>>>> next-level-cache = <&L2_1>;
>>>>> qcom,freq-domain = <&cpufreq_hw 1>;
>>>>> + #cooling-cells = <2>;
>>>>> };
>>>>>
>>>>> cpu-map {
>>>>> @@ -2471,6 +2482,24 @@ cpu4-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 6>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu4_alert0>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu4_alert1>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu4_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>> @@ -2497,6 +2526,24 @@ cpu5-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 7>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu5_alert0>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu5_alert1>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu5_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>> @@ -2523,6 +2570,24 @@ cpu6-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 8>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu6_alert0>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu6_alert1>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu6_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>> @@ -2549,6 +2614,24 @@ cpu7-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 9>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu7_alert0>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu7_alert1>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu7_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>> @@ -2575,6 +2658,24 @@ cpu45-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 10>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu45_alert0>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu45_alert1>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu45_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>> @@ -2601,6 +2702,24 @@ cpu67-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 11>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu67_alert0>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu67_alert1>;
>>>>> + cooling-device = <&CPU4 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU5 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU6 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU7 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu67_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>> @@ -2627,6 +2746,24 @@ cpu0123-thermal {
>>>>> polling-delay = <0>;
>>>>> thermal-sensors = <&tsens0 12>;
>>>>>
>>>>> + cooling-maps {
>>>>> + map0 {
>>>>> + trip = <&cpu0123_alert0>;
>>>>> + cooling-device = <&CPU0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU1 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU2 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU3 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> +
>>>>> + map1 {
>>>>> + trip = <&cpu0123_alert1>;
>>>>> + cooling-device = <&CPU0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU1 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU2 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>>>>> + <&CPU3 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>>>>> + };
>>>>> + };
>>>>> +
>>>>> trips {
>>>>> cpu0123_alert0: trip-point0 {
>>>>> temperature = <90000>;
>>>>>
>>>>> ---
>>>>> base-commit: 4aa1da8d99724f6c0b762b58a71cee7c5e2e109b
>>>>> change-id: 20230418-topic-cool_bengal-2f5f3f47269c
>>>>>
>>>>> Best regards,
>>>>> --
>>>>> Konrad Dybcio <konrad.dybcio@linaro.org>
>>>>>
prev parent reply other threads:[~2023-06-20 13:21 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-18 11:56 [PATCH] arm64: dts: qcom: sm6115: Set up CPU cooling maps Konrad Dybcio
2023-04-18 13:02 ` Bjorn Andersson
2023-04-18 13:01 ` Konrad Dybcio
2023-04-18 15:51 ` Bjorn Andersson
2023-05-31 14:55 ` Konrad Dybcio
2023-06-20 13:19 ` Konrad Dybcio [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=29e4d965-787c-b832-3916-722bad496fae@linaro.org \
--to=konrad.dybcio@linaro.org \
--cc=agross@kernel.org \
--cc=andersson@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=krzysztof.kozlowski+dt@linaro.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=marijn.suijten@somainline.org \
--cc=robh+dt@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).