[Patch v7] ACPI: processor: reduce CPUFREQ thermal reduction pctg for Tegra241

Sumit Gupta sumitg at nvidia.com
Thu Nov 23 06:17:41 PST 2023



On 23/11/23 19:30, Sudeep Holla wrote:
> External email: Use caution opening links or attachments
> 
> 
> On Thu, Nov 23, 2023 at 05:44:33PM +0530, Sumit Gupta wrote:
>> From: Srikar Srimath Tirumala <srikars at nvidia.com>
>>
>> Current implementation of processor_thermal performs software throttling
>> in fixed steps of "20%" which can be too coarse for some platforms.
>> We observed some performance gain after reducing the throttle percentage.
>> Change the CPUFREQ thermal reduction percentage and maximum thermal steps
>> to be configurable. Also, update the default values of both for Nvidia
>> Tegra241 (Grace) SoC. The thermal reduction percentage is reduced to "5%"
>> and accordingly the maximum number of thermal steps are increased as they
>> are derived from the reduction percentage.
>>
>> Signed-off-by: Srikar Srimath Tirumala <srikars at nvidia.com>
>> Co-developed-by: Sumit Gupta <sumitg at nvidia.com>
>> Signed-off-by: Sumit Gupta <sumitg at nvidia.com>
>> ---
>>
>> Sending this patch separately as the other patch in the series is
>> applied by Rafael in v6[1]. Revision history before this version is
>> in the cover letter of v6[1].
>>
>> Please review and provide ACK if looks fine.
>>
> 
> For arm64 specific changes(a minor nit below though),
> 
> Acked-by: Sudeep Holla <sudeep.holla at arm.com>
> 
> 
> [...]
> 

Thank you.

>> diff --git a/drivers/acpi/arm64/thermal_cpufreq.c b/drivers/acpi/arm64/thermal_cpufreq.c
>> new file mode 100644
>> index 000000000000..d524f2cd6044
>> --- /dev/null
>> +++ b/drivers/acpi/arm64/thermal_cpufreq.c
>> @@ -0,0 +1,20 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +#include <linux/acpi.h>
>> +
>> +#include "../internal.h"
>> +
>> +#define SMCCC_SOC_ID_T241      0x036b0241
>> +
> 
> [nit] We really need to find better place to define this globally and not
> locally at each usage site like this. We already have it in GICv3 driver.
> But that can come as a cleanup later if it causes issue for merging this
> change.
> 
> --
> Regards,
> Sudeep

Sure, will check and send a separate patch later on top.

Best Regards,
Sumit Gupta



More information about the linux-arm-kernel mailing list