GPU target temperature reduced in 595 for Turing cards

NVIDIA Open GPU Kernel Modules Version

595.58.03

Please confirm this issue does not happen with the proprietary driver (of the same version). This issue tracker is only for bugs specific to the open kernel driver.

• I confirm that this does not happen with the proprietary driver package.

Operating System and Version

Arch Linux

Kernel Release

6.19.9-arch1-1

Please confirm you are running a stable release kernel (e.g. not a -rc). We do not accept bug reports for unreleased kernels.

• I am running on a stable kernel release.

Hardware: GPU

RTX2070 Max-Q and GTX 1660Ti Mobile

Describe the bug

Since updating to nvidia-open 595, the GPU target temperature (gtt) is set to 75 degrees Celsius. Before (nvidia-open 580 and 590) it was 87 degrees Celsius. Confirmed with:

nvidia-smi -q | grep -i target

The gtt also cannot be changed with nvidia-smi -gtt 87 (GPU Temperature Target Threshold not supported).

This results in throttling and performance reduction once the temperature reaches 74 degrees. Before, the GPU was operating normally at stable 85 degrees under full load.

To Reproduce

With nvidia-open 595 installed, run:

nvidia-smi -q | grep -i target

Bug Incidence

Always

nvidia-bug-report.log.gz

nvidia-bug-report.log.gz

More Info

No response

Start with the nvidia-smi target-temperature output and the attached nvidia-bug-report.log.gz, then compare the behavior between nvidia-open 580/590 and 595 on the listed Turing GPUs. Reproduce with nvidia-smi -q | grep -i target and verify that the target returns to the expected value without throttling under load.