Score:1

Blender 3.3 does not recognize my GPU for HIP (on Ubuntu 20.04)

ml flag

My GPU (RX 5600 XT) does not appear as an option in Blender Render Devices menu. It reads:

No compatible GPUs found for Cycles. Requires AMD GPU with Vega or RDNA architecture and AMD driver version 22.10 or newer

My card does meet those requirements and I have Radeon™ Software for Linux® installer version 22.20 for Ubuntu 20.04.5 HWE installed. I have a similar issue with DaVinci Resolve.

The interesting thing is that when I run blender as root, Both my GPU and CPU appear as options.

Additionally, using clinfo will result in (this is obviously not everything):

Platform Name AMD Accelerated Parallel Processing

Number of devices 0

However, using clinfo as root will result in:

Platform Name AMD Accelerated Parallel Processing

Number of devices 1

Edit: Blender will output HIP hipInit: Invalid device to the terminal if preferences is opened. Edit 2: I've added myself to the video and render groups, to no avail.

Here is some information I get from lspci -vnn

    Subsystem: XFX Pine Group Inc. Navi 10 [Radeon RX 5600 OEM/5600 XT / 5700/5700 XT] [1682:5710]
    Flags: bus master, fast devsel, latency 0, IRQ 77
    Memory at d0000000 (64-bit, prefetchable) [size=256M]
    Memory at e0000000 (64-bit, prefetchable) [size=2M]
    I/O ports at f000 [size=256]
    Memory at fcd00000 (32-bit, non-prefetchable) [size=512K]
    Expansion ROM at fcd80000 [disabled] [size=128K]
    Capabilities: <access denied>
    Kernel driver in use: amdgpu
    Kernel modules: amdgpu```

Score:1
ml flag

I fixed my issue. After doing all of the things above, what finally worked was logging out, then restarting my device.

It probably also helped that I ran this beforehand: sudo usermod -a -G video username sudo usermod -a -G render username

I sit in a Tesla and translated this thread with Ai:

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.