2

I've checked out GPU usage monitoring (CUDA), but are there similar tools for AMD/ATI cards? Or kind of universal tools? I want to check out if my applications use the 256 MB of RAM of the video card at all, since I've seen applications that uses lots system memory, while they should rather use the video cards.

glxinfo does not provide the information I'm looking for, but maybe you will ask if I have HW Acceleration:

$ glxinfo | grep render
direct rendering: Yes
OpenGL renderer string: Gallium 0.4 on ATI RV515

The info about the card:

03:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV515 [Radeon X1300/X1550] (prog-if 00 [VGA controller])
    Subsystem: VISIONTEK Device 2352
    Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
    Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
    Latency: 0, Cache Line Size: 16 bytes
    Interrupt: pin A routed to IRQ 19
    Region 0: Memory at d0000000 (64-bit, prefetchable) [size=256M]
    Region 2: Memory at bffe0000 (64-bit, non-prefetchable) [size=64K]
    Region 4: I/O ports at e000 [size=256]
    Expansion ROM at bffc0000 [disabled] [size=128K]
    Capabilities: [50] Power Management version 2
        Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
        Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
    Capabilities: [58] Express (v1) Endpoint, MSI 00
        DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <4us, L1 unlimited
            ExtTag+ AttnBtn- AttnInd- PwrInd- RBE- FLReset-
        DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
            RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop+
            MaxPayload 128 bytes, MaxReadReq 128 bytes
        DevSta: CorrErr- UncorrErr- FatalErr- UnsuppReq- AuxPwr- TransPend-
        LnkCap: Port #0, Speed 2.5GT/s, Width x16, ASPM L0s L1, Latency L0 <64ns, L1 <1us
            ClockPM- Surprise- LLActRep- BwNot-
        LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- Retrain- CommClk-
            ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
        LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
    Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit+
        Address: 0000000000000000  Data: 0000
    Kernel driver in use: radeon

I know there is Process Explorer in Windows, and it works with my card so the thing is Linux Kernel - driver/module - tool.

Braiam
  • 35,991

1 Answers1

1

There's a program called radeontop which should provide some or all of the information you're after.

I've installed and run it on my debian laptop (which has a Radeon HD 6320 GPU) and it seems to work as advertised.

If you need data for further processing rather than a top-like display, it has a -d or --dump option for dumping the data to a file (unfortunately, only as percentages rather than as raw numbers). Examining the source code will tell you how to get at the raw data yourself.

The debian packaged version has the following description.

Package: radeontop
Description-en: Utility to show Radeon GPU utilization
 radeontop is a small utility which allows one to monitor the utilization of
 Radeon GPUs starting from the R600 series and newer using undocumented
 performance counters in the hardware. The utility works with the free
 drivers.
 .
 It displays the utilization of the graphics pipe, event engine, vertex cache,
 vertex group and tesselator, texture addresser and cache, the shader units
 and more, both with a relative percent value as well as a colorful bar diagram.
Homepage: https://github.com/clbr/radeontop
cas
  • 78,579