How can I tell if my GPU is being utilized properly?
zamithal @ zamithal @programming.dev Posts 3Comments 35Joined 2 yr. ago
zamithal @ zamithal @programming.dev
Posts
3
Comments
35
Joined
2 yr. ago
Take a look in each game’s graphics settings. Not all of them show it, but some do. In Baldur’s Gate 3
KK will do.
I would expect System76 to be able to help with this more efficiently than we can, since they sold you the system for use with Linux.
I will certainly reach out to them, but it hadn't occurred to me until this post ¯(ツ)_/¯
uname -r
6.9.3-76060903-generic
I think this is the mesa version?
OpenGL version string: 4.5 (Compatibility Profile) Mesa 24.1.0-devel
cat /etc/os-release
NAME="Pop!_OS"
VERSION="22.04 LTS"
ID=pop
ID_LIKE="ubuntu debian"
PRETTY_NAME="Pop!_OS 22.04 LTS"
VERSION_ID="22.04"
HOME_URL="https://pop.system76.com/"
SUPPORT_URL="https://support.system76.com/"
BUG_REPORT_URL="https://github.com/pop-os/pop/issues"
PRIVACY_POLICY_URL="https://system76.com/privacy"
VERSION_CODENAME=jammy
UBUNTU_CODENAME=jammy
LOGO=distributor-logo-pop-os
dpkg -l|grep radeon
ii libdrm-amdgpu-radeon1:amd64 1:2.4.120.60103-1787201.22.04 amd64 Userspace interface to radeon-specific kernel DRM services -- runtime
ii libdrm-radeon1:amd64 2.4.120-1pop1~1706792268~22.04~bfb54ee amd64 Userspace interface to radeon-specific kernel DRM services -- runtime
ii libdrm-radeon1:i386 2.4.120-1pop1~1706792268~22.04~bfb54ee i386 Userspace interface to radeon-specific kernel DRM services -- runtime
ii radeontop 1.4-1 amd64 Utility to show Radeon GPU utilization
ii xserver-xorg-video-radeon 1:19.1.0-2ubuntu1 amd64 X.Org X server -- AMD/ATI Radeon display driver
I don't know exactly what i'm looking for in the xorg logs... cat /var/log/Xorg.*.log | grep "EE"
cat /var/log/Xorg.*.log | grep "WW"
[ 5068.047] (WW) Falling back to old probe method for modesetting
[ 5068.047] (WW) Falling back to old probe method for fbdev
cat /var/log/Xorg.*.log | grep "gpu"
[ 5067.696] (II) Applying OutputClass "AMDgpu" to /dev/dri/card1
[ 5067.696] loading driver: amdgpu
[ 5067.696] (==) Matched amdgpu as autoconfigured driver 0
[ 5067.696] (II) LoadModule: "amdgpu"
[ 5067.696] (II) Loading /usr/lib/xorg/modules/drivers/amdgpu_drv.so
[ 5067.696] (II) Module amdgpu: vendor="X.Org Foundation"
All GPUs supported by the amdgpu kernel driver
sudo journalctl -kb | grep gpu
Feb 15 08:59:51 pop-os kernel: [drm] amdgpu kernel modesetting enabled.
Feb 15 08:59:51 pop-os kernel: amdgpu: Virtual CRAT table created for CPU
Feb 15 08:59:51 pop-os kernel: amdgpu: Topology: Add CPU node
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: enabling device (0006 -> 0007)
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: Fetched VBIOS from VFCT
Feb 15 08:59:51 pop-os kernel: amdgpu: ATOM BIOS: 113-APM7199-002
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: CP RS64 enable
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: [drm:jpeg_v4_0_early_init [amdgpu]] JPEG decode is enabled in VM mode
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: vgaarb: deactivate vga console
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: Trusted Memory Zone (TMZ) feature not supported
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: MEM ECC is not presented.
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: SRAM ECC is not presented.
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: VRAM: 16368M 0x0000008000000000 - 0x00000083FEFFFFFF (16368M used)
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: GART: 512M 0x00007FFF00000000 - 0x00007FFF1FFFFFFF
Feb 15 08:59:51 pop-os kernel: [drm] amdgpu: 16368M of VRAM memory ready
Feb 15 08:59:51 pop-os kernel: [drm] amdgpu: 64286M of GTT memory ready.
Feb 15 08:59:51 pop-os kernel: [drm] GART: num cpu pages 131072, num gpu pages 131072
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: Will use PSP to load VCN firmware
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: reserve 0x1300000 from 0x83fc000000 for PSP TMR
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: RAP: optional rap ta ucode is not available
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: smu driver if version = 0x0000003d, smu fw if version = 0x00000040, smu fw program = 0, smu fw version = 0x004e7e00 (78.126.0)
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: SMU driver if version not matched
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: SMU is initialized successfully!
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: [drm:jpeg_v4_0_hw_init [amdgpu]] JPEG decode initialized successfully.
Feb 15 08:59:51 pop-os kernel: amdgpu: HMM registered 16368MB device memory
Feb 15 08:59:51 pop-os kernel: kfd kfd: amdgpu: Allocated 3969056 bytes on gart
Feb 15 08:59:51 pop-os kernel: kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
Feb 15 08:59:51 pop-os kernel: amdgpu: Virtual CRAT table created for GPU
Feb 15 08:59:51 pop-os kernel: amdgpu: Topology: Add dGPU node [0x744c:0x1002]
Feb 15 08:59:51 pop-os kernel: kfd kfd: amdgpu: added device 1002:744c
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: SE 6, SH per SE 2, CU per SH 8, active_cu_number 80
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_unified_0 uses VM inv eng 0 on hub 8
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_unified_1 uses VM inv eng 1 on hub 8
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring jpeg_dec uses VM inv eng 4 on hub 8
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: ring mes_kiq_3.1.0 uses VM inv eng 14 on hub 0
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: amdgpu: Using BACO for runtime pm
Feb 15 08:59:51 pop-os kernel: [drm] Initialized amdgpu 3.57.0 20150101 for 0000:03:00.0 on minor 2
Feb 15 08:59:51 pop-os kernel: fbcon: amdgpudrmfb (fb0) is primary device
Feb 15 08:59:51 pop-os kernel: amdgpu 0000:03:00.0: [drm] fb0: amdgpudrmfb frame buffer device
Feb 15 08:59:52 pop-os kernel: RAPL PMU: hw unit of domain pp1-gpu 2^-14 Joules
Feb 15 08:59:52 pop-os kernel: snd_hda_intel 0000:03:00.1: bound 0000:03:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])
...
Do any of your underperforming games have a screen that shows which GPU they are using?
I haven't found any that do but I can try launching a few and searching.
Is your monitor plugged into the motherboard’s video port, or one of the graphics card’s ports?
GPU port, 100%
Since you bought from System76, I would guess that they have support staff who can help make sure these things are set up properly. Have you called them?
I had not thought of doing that for whatever reason, but I will in the daytime tomorrow.
Does an overlay appear in the top left of the screen if you put DXVK_HUD=devinfo %command% in an underperforming game’s Steam Launch Options?
For the 3 I've checked so far no, I don't see that.
vulkan-tools | grep "GPU id":
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id = 0 (Radeon RX 7900 GRE (RADV NAVI31))
GPU id = 1 (Intel(R) Graphics (RPL-S))
GPU id = 2 (llvmpipe (LLVM 15.0.7, 256 bits))
GPU id : 0 (Radeon RX 7900 GRE (RADV NAVI31)):
GPU id : 1 (Intel(R) Graphics (RPL-S)):
GPU id : 2 (llvmpipe (LLVM 15.0.7, 256 bits)):
cat /var/log/Xorg.*.log | grep amd
[ 5067.696] (II) LoadModule: "amdgpu"
[ 5067.696] (II) Loading /usr/lib/xorg/modules/drivers/amdgpu_drv.so
[ 5067.696] (II) Module amdgpu: vendor="X.Org Foundation"
All GPUs supported by the amdgpu kernel driver
cat /var/log/Xorg.*.log | grep gpu
[ 5067.696] (II) Applying OutputClass "AMDgpu" to /dev/dri/card1
[ 5067.696] loading driver: amdgpu
[ 5067.696] (==) Matched amdgpu as autoconfigured driver 0
[ 5067.696] (II) LoadModule: "amdgpu"
[ 5067.696] (II) Loading /usr/lib/xorg/modules/drivers/amdgpu_drv.so
[ 5067.696] (II) Module amdgpu: vendor="X.Org Foundation"
All GPUs supported by the amdgpu kernel driver
Fair point, I should do that.
glxinfo | grep Vendor
Vendor: Mesa (0xffffffff)
glxinfo | grep Device
Device: llvmpipe (LLVM 17.0.6, 256 bits) (0xffffffff)
glxinfo | grep "OpenGL rend"
OpenGL renderer string: llvmpipe (LLVM 17.0.6, 256 bits)
Let me know if that's not right. glxinfo dumps a lot of text but those are the only hits for your comment.
When I launch radeontop it prints this before launching, and then the output suggests it isn't working:
Unknown Radeon card. <= R500 won't work, new cards might.
All stats sit at 0.00% except for Memory Clock @ 9%.
EDIT:
xorg, not wayland
Well for starters, my Wacom tablets drawing pen's eraser sometimes stops working until I restart my machine. How can I restart the service for this without restarting my machine? How can I identify what service that is? How can I debug the error to prevent it in the future?
This is a helpful tip, thanks
Way ahead of you but that only scratches the surface of Linux. I've got a docker compose stack with a bunch of services, DNS and reverse proxies... But that doesn't teach me about the internals on my workstation
Thanks for the tip on phoronix, I'll look into this.
I probably will :) but some baselineing would be nice too
While I'm usually a hands on type of learner, usually these days I find my time pretty limited and that's why I'm looking for a reading approach. I absolutely want to up a machine with arch, there's no doubt it will teach me many things. I've got a lot of items on my 'to learn' list though and was looking for a more passive approach I can fit between other tasks away from my machine
I cannot recommend Silver Bullet enough
This was the missing piece of the puzzle I needed to actually get organized and write things down. The graph based linking system allows me to link ideas together without any sort of hierarchy, which I adore.
I mostly use this to DM my groups Pathfinder campaign, but also use it for general note taking as well.
Yeah it appears my card isn't being used at all, which explains the poor performance. Honestly the fact the system runs as well as it does without it is impressive. I'm reaching out to system76 for their diagnostics on why this might be