Closed Samueru-sama closed 10 months ago
IIRC this came up before. Printing the total each line would grow the logs unnecessarily. I recommend either calculating the total from the percentage, or grabbing the total dmesg. It may also be available in /sys, not sure though.
*from dmesg.
Alternatively, I'd be open to taking a patch for a new cli option like "--print-maximums" that would just output the vram/gtt/clock max amounts and exit.
IIRC this came up before. Printing the total each line would grow the logs unnecessarily. I recommend either calculating the total from the percentage, or grabbing the total dmesg. It may also be available in /sys, not sure though.
I see, well you know what is funny? amdgpu_top -d prints the used/total vram but doesn't print the gpu usage in percentage like radeontop does, looks like I will have to use it along with radeontop to get the info lol.
Edit: I ended up using amdgpu_top, if anyone wants it:
amdgpu_top -J -n 1 | jq -c -r '(.devices[] | ((.gpu_activity.GFX.value|tostring) + "% " + ((.VRAM."Total VRAM Usage".value/1024*10|round/10|tostring) + "/" + (.VRAM."Total VRAM".value/1024*10|round/10|tostring)) + "GiB"))'
Which nicely prints the gpu usage plus vram in GiB.
I'm trying to write a polybar script that takes the used/total vram of the GPU and prints the results.
The issue I have is that while radeontop with its GUI does show the total VRAM of the GPU, when printing the info witth radeontop -d - the value is not shown, only used vram is shown instead. Any help?