Blender 3.2 Debuts With AMD GPU Linux Rendering Support

[dpospisil]

I have asked AMD today, when there will be Ubuntu 22.04 AMDGPU-PRO drivers available. I have received very bizarre answer:

We cannot comment on availability of the new drivers for Ubuntu 22.04 as we are from AMD customer care.

What does that mean?

[reba]

Tried it just for fun on my Debian machine (I possibly selected too many packages to achieve base functionality) :

$ sudo apt-get install rocm-device-libs rocm-smi rocminfo

Rebooted for good measure (probably not necessary)

$ rocminfo  
ROCk module is loaded
=====================    
HSA System Attributes    
=====================    
Runtime Version:         1.1
System Timestamp Freq.:  1000.000000MHz
Sig. Max Wait Duration:  18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model:           LARGE                              
System Endianness:       LITTLE                              

==========                
HSA Agents                
==========                
*******                  
Agent 1                  
*******                  
 Name:                    AMD Ryzen 9 5900HX with Radeon Graphics
 Uuid:                    CPU-XX                              
 Marketing Name:          AMD Ryzen 9 5900HX with Radeon Graphics
 Vendor Name:             CPU                                
 Feature:                 None specified                      
 Profile:                 FULL_PROFILE                        
 Float Round Mode:        NEAR                                
 Max Queue Number:        0(0x0)                              
 Queue Min Size:          0(0x0)                              
 Queue Max Size:          0(0x0)                              
 Queue Type:              MULTI                              
 Node:                    0                                  
 Device Type:             CPU                                
 Cache Info:              
   L1:                      32768(0x8000) KB                    
 Chip ID:                 0(0x0)                              
 Cacheline Size:          64(0x40)                            
 Max Clock Freq. (MHz):   4680                                
 BDFID:                   0                                  
 Internal Node ID:        0                                  
 Compute Unit:            16                                  
 SIMDs per CU:            0                                  
 Shader Engines:          0                                  
 Shader Arrs. per Eng.:   0                                  
 WatchPts on Addr. Ranges:1                                  
 Features:                None
 Pool Info:                
   Pool 1                    
     Segment:                 GLOBAL; FLAGS: FINE GRAINED        
     Size:                    32225208(0x1ebb7b8) KB              
     Allocatable:             TRUE                                
     Alloc Granule:           4KB                                
     Alloc Alignment:         4KB                                
     Accessible by all:       TRUE                                
   Pool 2                    
     Segment:                 GLOBAL; FLAGS: KERNARG, FINE GRAINED
     Size:                    32225208(0x1ebb7b8) KB              
     Allocatable:             TRUE                                
     Alloc Granule:           4KB                                
     Alloc Alignment:         4KB                                
     Accessible by all:       TRUE                                
   Pool 3                    
     Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
     Size:                    32225208(0x1ebb7b8) KB              
     Allocatable:             TRUE                                
     Alloc Granule:           4KB                                
     Alloc Alignment:         4KB                                
     Accessible by all:       TRUE                                
 ISA Info:                
*******                  
Agent 2                  
*******                  
 Name:                    gfx90c                              
 Uuid:                    GPU-XX                              
 Marketing Name:                                              
 Vendor Name:             AMD                                
 Feature:                 KERNEL_DISPATCH                    
 Profile:                 BASE_PROFILE                        
 Float Round Mode:        NEAR                                
 Max Queue Number:        128(0x80)                          
 Queue Min Size:          64(0x40)                            
 Queue Max Size:          131072(0x20000)                    
 Queue Type:              MULTI                              
 Node:                    1                                  
 Device Type:             GPU                                
 Cache Info:              
   L1:                      16(0x10) KB                        
   L2:                      1024(0x400) KB                      
 Chip ID:                 5688(0x1638)                        
 Cacheline Size:          64(0x40)                            
 Max Clock Freq. (MHz):   2100                                
 BDFID:                   1280                                
 Internal Node ID:        1                                  
 Compute Unit:            8                                  
 SIMDs per CU:            4                                  
 Shader Engines:          1                                  
 Shader Arrs. per Eng.:   1                                  
 WatchPts on Addr. Ranges:4                                  
 Features:                KERNEL_DISPATCH  
 Fast F16 Operation:      TRUE                                
 Wavefront Size:          64(0x40)                            
 Workgroup Max Size:      1024(0x400)                        
 Workgroup Max Size per Dimension:
   x                        1024(0x400)                        
   y                        1024(0x400)                        
   z                        1024(0x400)                        
 Max Waves Per CU:        40(0x28)                            
 Max Work-item Per CU:    2560(0xa00)                        
 Grid Max Size:           4294967295(0xffffffff)              
 Grid Max Size per Dimension:
   x                        4294967295(0xffffffff)              
   y                        4294967295(0xffffffff)              
   z                        4294967295(0xffffffff)              
 Max fbarriers/Workgrp:   32                                  
 Pool Info:                
   Pool 1                    
     Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
     Size:                    524288(0x80000) KB                  
     Allocatable:             TRUE                                
     Alloc Granule:           4KB                                
     Alloc Alignment:         4KB                                
     Accessible by all:       FALSE                              
   Pool 2                    
     Segment:                 GROUP                              
     Size:                    64(0x40) KB                        
     Allocatable:             FALSE                              
     Alloc Granule:           0KB                                
     Alloc Alignment:         0KB                                
     Accessible by all:       FALSE                              
 ISA Info:                
   ISA 1                    
     Name:                    amdgcn-amd-amdhsa--gfx90c:xnack-    
     Machine Models:          HSA_MACHINE_MODEL_LARGE            
     Profiles:                HSA_PROFILE_BASE                    
     Default Rounding Mode:   NEAR                                
     Default Rounding Mode:   NEAR                                
     Fast f16:                TRUE                                
     Workgroup Max Size:      1024(0x400)                        
     Workgroup Max Size per Dimension:
       x                        1024(0x400)                        
       y                        1024(0x400)                        
       z                        1024(0x400)                        
     Grid Max Size:           4294967295(0xffffffff)              
     Grid Max Size per Dimension:
       x                        4294967295(0xffffffff)              
       y                        4294967295(0xffffffff)              
       z                        4294967295(0xffffffff)              
     FBarrier Max Size:       32                                  
*** Done ***

$ rocm-smi --showhw


======================= ROCm System Management Interface =======================
============================ Concise Hardware Info =============================
GPU DID GFX RAS SDMA RAS UMC RAS VBIOS BUS
0 1638 N/A N/A N/A 113-CEZANNE-017 0000:05:00.0
================================================== ==============================
============================= End of ROCm SMI Log ==============================
$ lspci | grep '05:00.0'
05:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Cezanne (rev c4)

a) What's the problem people seem to have?
b) Is there a way to verify functionality besides rocminfo and rocm-smi?

[bridgman]

Been trying with 22.10.3 on Radeon RX 6800 / 6800 XT cards on Ubuntu 20.04.4.... No go so far. Blender detects no supported GPUs for HIP / HIP device init failed.

Clean Ubuntu 20.04 install, tried as well the rocm and hip / hiplibsdk use-cases for the amdgpu-install, no go....

Any ideas bridgman or any other areas of documentation I am seeming overlooking? The driver stack appears fine unless there is some undocumented changes needed...

I'm not sure if we include enough of the ROCm stack in the packaged drivers to run Blender yet - will try to find out.

Does /opt/rocm/bin/rocminfo show anything interesting ?

EDIT - just curious, did someone recommend 22.10.3 ? AFAIK it is in the repos but our download pages are still recommending 22.10.2.

[bridgman]

We cannot comment on availability of the new drivers for Ubuntu 22.04 as we are from AMD customer care.

What does that mean?

"Gentlemen, you can't fight in here - this is the War Room"

Seriously, I think they were trying to say that their job is to support already-released drivers, and that they do not have visibility into unreleased drivers except to the extent that the engineering teams say "this is fixed in <release>, tell them it'll be out around <date>".

[photom]

b) Is there a way to verify functionality besides rocminfo and rocm-smi?

maybe run Blender? (ok, bad joke - sorry - Cezanne/Vega APU is disabled in the source code because of too many errors - they want to fix RDNA1 errors next -> https://developer.blender.org/T97591)

[Grinness]

Thank you, very useful!

The closest I could come re. documentation of currently "supported" hardware is this PR, which has AMD's attention:

https://github.com/RadeonOpenCompute/ROCm/pull/1738

If you compile from source you have support from Polaris to RDNA2, including Vega 56, 64 , CDNA ... at least this is my experience with Arch AUR, rx480, currently running ROCm 5.1.3.

$ rocminfo | grep gfx
Name: gfx803
Name: amdgcn-amd-amdhsa--gfx803

If you go for binary, you get what officially supported.
To get blender working with polaris we need to wait for their (blender) support or attempt compile from source and enable support there.
Similarly to get i.e. pytorch with polaris (gfx803), you need to compile it (pytorch) from source (after you compiled ROCm &HIP etc with Polaris support) ....

Edit: AMD ProRenderer plugin 3.4.0 works on Blender 3.2 and Polaris (rx480) fine, opensource amdgpu kernel module, ROCm 5.1.3 (tested a couple of hours ago)

[bridgman]

Quick update from bsavery - looks like they are installing just the ROCm stack from the following location:

https://github.com/RadeonOpenCompute/ROCm/releases

I believe they are using 5.1.3 but have asked for confirmation. Reading quickly through the install instructions, seems like it involves downloading the 21.50 amdgpu installer then (a) installing kernel driver from AMDGPU repo, (b) installing the rest of ROCm stack from the ROCm repo.

EDIT - OK, looks like our testing was done with both 22.10 AMDGPU drivers (probably 22.10.2) and the ROCm 5.1 release, both working.

[illwieckz]

A few days ago I've stumbled upon a Reddit post of someone who got it to work in Blender 3.1 with an RX 580.

This is another renderer (Radeon ProRender), this is not using default Blender renderer with Radeon.

[Keats]

Is there any technical/design reason this couldn't have been done with Vulkan? Most AMD GPU users don't have ROCm installed because it's otherwise useless. It's not too tough to get the runtime on Arch but...

I'm not an expert, but as I understand it, Vulcan is a graphics API, OpenCL is a compute API. They both use the GPU, yes, but they do very different things.

[JacekJagosz]

If you compile from source you have support from Polaris to RDNA2, including Vega 56, 64 , CDNA ... at least this is my experience with Arch AUR, rx480, currently running ROCm 5.1.3.

Just be aware that on AUR they patched in Polaris support, so probably AMD binaries don't contain, and on other distributions you need to use this patch as well.