AMD RDNA 2 GPU not detected on Linux with ROCm hip libraries #102158

New Issue

Joseph McElroy · 2022-10-30T02:34:37+02:00

Joseph McElroy commented

2022-10-30 02:34:37 +02:00

System Information
Operating system: Opensuse Tumbleweed 20221028
Graphics card: AMD 6700S [Radeon RX 6650 XT] dimgrey_cavefish (link )

Blender Version
Broken: 3.3 from Opensuse repo as well as 3.3 flatpak
Worked: n/a

Short description of error

When going to Edit > Preferences > System > HIP to enable my GPU for rendering it says: No compatible GPUs found for Cycles, requires Vega/RDNA architecture and driver version 22.10 or newer.

Exact steps for others to reproduce the error
Based on the default startup or an attached .blend file (as simple as possible).

I just opened a blank new file.

Details

I have Mesa 22.2.2 and ROCm 5.3.0 installed, along with hip-devel, hip-runtime-amd, hipblas, etc., all v5.3.0.

I have the following environmental variables:

DRI_PRIME=1
LD_LIBRARY_PATH=/opt/rocm-5.3.0/opencl/lib
HSA_PATH=/opt/rocm-5.3.0/hsa/
HIP_PATH=/opt/rocm-5.3.0/hip/
HIP_CLANG_PATH=/bin
DEVICE_LIB_PATH=/opt/rocm-5.3.0/lib/
HIPCC_VERBOSE=7

These set my GPU as default for graphics, enable OpenCL to work, specify the locations for hip/hsa/clang, and allow more verbose logs to be output when running hipcc.

I can verify that Mesa and ROCm work in other applications, so given that the GPU architecture is RDNA 2 and I have all the proper libraries installed, I'm unsure why Blender isn't able to detect things.

The output of hipconfig is:

HIP version  : 5.3.22061-e8e78f1a

== hipconfig
HIP_PATH     : /opt/rocm-5.3.0/hip/
ROCM_PATH    : /opt/rocm-5.3.0
HIP_COMPILER : clang
HIP_PLATFORM : amd
HIP_RUNTIME  : rocclr
CPP_CONFIG   :  -D__HIP_PLATFORM_HCC__= -D__HIP_PLATFORM_AMD__= -I/opt/rocm-5.3.0/hip*include -I/bin/../lib/clang/15.0.2 -I/opt/rocm-5.3.0/hsa*include

== hip-clang
HSA_PATH         : /opt/rocm-5.3.0/hsa/
HIP_CLANG_PATH   : /bin
clang version 15.0.2
Target: x86_64-suse-linux
Thread model: posix
InstalledDir: /bin
LLVM (http://llvm.org/):
  LLVM version 7.0.1
  Optimized build.
  Default target: x86_64-unknown-linux-gnu
  Host CPU: (unknown)

  Registered Targets:
    aarch64    - AArch64 (little endian)
    aarch64_be - AArch64 (big endian)
    amdgcn     - AMD GCN GPUs
    arm        - ARM
    arm64      - ARM64 (little endian)
    armeb      - ARM (big endian)
    bpf        - BPF (host endian)
    bpfeb      - BPF (big endian)
    bpfel      - BPF (little endian)
    hexagon    - Hexagon
    lanai      - Lanai
    mips       - Mips
    mips64     - Mips64 [experimental]
    mips64el   - Mips64el [experimental]
    mipsel     - Mipsel
    msp430     - MSP430 [experimental]
    nvptx      - NVIDIA PTX 32-bit
    nvptx64    - NVIDIA PTX 64-bit
    ppc32      - PowerPC 32
    ppc64      - PowerPC 64
    ppc64le    - PowerPC 64 LE
    r600       - AMD GPUs HD2XXX-HD6XXX
    sparc      - Sparc
    sparcel    - Sparc LE
    sparcv9    - Sparc V9
    systemz    - SystemZ
    thumb      - Thumb
    thumbeb    - Thumb (big endian)
    x86        - 32-bit X86: Pentium-Pro and above
    x86-64     - 64-bit X86: EM64T and AMD64
    xcore      - XCore
hip-clang-cxxflags : HIP_PATH=/opt/rocm-5.3.0/hip/
HIP_PLATFORM=amd
HIP_COMPILER=clang
HIP_RUNTIME=rocclr
ROCM_PATH=/opt/rocm-5.3.0
HIP_ROCCLR_HOME=/opt/rocm-5.3.0/hip/
HIP_CLANG_PATH=/bin
Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 183.
HIP_CLANG_INCLUDE_PATH=
HIP_INCLUDE_PATH=/opt/rocm-5.3.0/hip//include
HIP_LIB_PATH=/opt/rocm-5.3.0/hip//lib
DEVICE_LIB_PATH=/opt/rocm-5.3.0/lib/
HIP_CLANG_TARGET=1
Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 195.
Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 196.
hipcc-args: --cxxflags
hipcc-cmd: /bin/clang++  -L"/opt/rocm-5.3.0/hip*lib" -O3 -lgcc_s -lgcc -lpthread -lm -lrt  --cxxflags -Wl,--enable-new-dtags -Wl,-rpath=/opt/rocm-5.3.0/hip*lib:/opt/rocm-5.3.0/lib -lamdhip64  -L/bin/../lib/clang/15.0.2/lib/linux -lclang_rt.builtins-x86_64 
 -std=c++11 -isystem "/.." -isystem /opt/rocm-5.3.0/hsa*include -isystem "/opt/rocm-5.3.0/hip*include" -O3
hip-clang-ldflags  : HIP_PATH=/opt/rocm-5.3.0/hip/
HIP_PLATFORM=amd
HIP_COMPILER=clang
HIP_RUNTIME=rocclr
ROCM_PATH=/opt/rocm-5.3.0
HIP_ROCCLR_HOME=/opt/rocm-5.3.0/hip/
HIP_CLANG_PATH=/bin
Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 183.
HIP_CLANG_INCLUDE_PATH=
HIP_INCLUDE_PATH=/opt/rocm-5.3.0/hip//include
HIP_LIB_PATH=/opt/rocm-5.3.0/hip//lib
DEVICE_LIB_PATH=/opt/rocm-5.3.0/lib/
HIP_CLANG_TARGET=1
Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 195.
Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 196.
hipcc-args: --ldflags
hipcc-cmd: /bin/clang++  -L"/opt/rocm-5.3.0/hip*lib" -O3 -lgcc_s -lgcc -lpthread -lm -lrt  --ldflags -Wl,--enable-new-dtags -Wl,-rpath=/opt/rocm-5.3.0/hip*lib:/opt/rocm-5.3.0/lib -lamdhip64  -L/bin/../lib/clang/15.0.2/lib/linux -lclang_rt.builtins-x86_64 
 -L"/opt/rocm-5.3.0/hip//lib" -O3 -lgcc_s -lgcc -lpthread -lm -lrt

=== Environment Variables
PATH=/home/username/.local/bin:/home/username/bin:/usr/local/bin:/usr/bin:/bin:/opt/rocm/bin:/opt/rocm/profiler/bin:/opt/rocm/opencl/bin:/opt/rocm/bin:/opt/rocm/profiler/bin:/opt/rocm/opencl/bin/x86_64
HIPCC_VERBOSE=7
HSA_PATH=/opt/rocm-5.3.0/hsa/
HIP_PATH=/opt/rocm-5.3.0/hip/
HIP_CLANG_PATH=/bin
LD_LIBRARY_PATH=/opt/rocm-5.3.0/opencl/lib

== Linux Kernel
Hostname     :  Hostname
Linux Hostname 6.0.5-1-default #1 SMP PREEMPT_DYNAMIC Thu Oct 27 05:14:41 UTC 2022 (7359656) x86_64 x86_64 x86_64 GNU/Linux
LSB Version:    n/a
Distributor ID: openSUSE
Description:    openSUSE Tumbleweed
Release:        20221028
Codename:       n/a

And the output of rocminfo is:

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 6900HS with Radeon Graphics
  Uuid:                    CPU-XX                             
  Marketing Name:          AMD Ryzen 9 6900HS 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)                             
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   3300                               
  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:                    40277236(0x26694f4) KB             
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 2                   
      Segment:                 GLOBAL; FLAGS: KERNARG, FINE GRAINED
      Size:                    40277236(0x26694f4) KB             
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
    Pool 3                   
      Segment:                 GLOBAL; FLAGS: COARSE GRAINED      
      Size:                    40277236(0x26694f4) KB             
      Allocatable:             TRUE                               
      Alloc Granule:           4KB                                
      Alloc Alignment:         4KB                                
      Accessible by all:       TRUE                               
  ISA Info:                
*******Agent 2*******                  
  Name:                    gfx1032                            
  Uuid:                    GPU-XX                             
  Marketing Name:          AMD Radeon RX 6700S                
  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:                      2048(0x800) KB                     
    L3:                      32768(0x8000) KB                   
  Chip ID:                 29679(0x73ef)                      
  ASIC Revision:           0(0x0)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   2435                               
  BDFID:                   768                                
  Internal Node ID:        1                                  
  Compute Unit:            28                                 
  SIMDs per CU:            2                                  
  Shader Engines:          4                                  
  Shader Arrs. per Eng.:   2                                  
  WatchPts on Addr. Ranges:4                                  
  Features:                KERNEL_DISPATCH 
  Fast F16 Operation:      TRUE                               
  Wavefront Size:          32(0x20)                           
  Workgroup Max Size:      1024(0x400)                        
  Workgroup Max Size per Dimension:
    x                        1024(0x400)                        
    y                        1024(0x400)                        
    z                        1024(0x400)                        
  Max Waves Per CU:        32(0x20)                           
  Max Work-item Per CU:    1024(0x400)                        
  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:                    8372224(0x7fc000) 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--gfx1032         
      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                                 
*******Agent 3*******                  
  Name:                    gfx1035                            
  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:                    2                                  
  Device Type:             GPU                                
  Cache Info:              
    L1:                      16(0x10) KB                        
    L2:                      2048(0x800) KB                     
  Chip ID:                 5761(0x1681)                       
  ASIC Revision:           2(0x2)                             
  Cacheline Size:          64(0x40)                           
  Max Clock Freq. (MHz):   2400                               
  BDFID:                   1792                               
  Internal Node ID:        2                                  
  Compute Unit:            12                                 
  SIMDs per CU:            2                                  
  Shader Engines:          2                                  
  Shader Arrs. per Eng.:   2                                  
  WatchPts on Addr. Ranges:4                                  
  Features:                KERNEL_DISPATCH 
  Fast F16 Operation:      TRUE                               
  Wavefront Size:          32(0x20)                           
  Workgroup Max Size:      1024(0x400)                        
  Workgroup Max Size per Dimension:
    x                        1024(0x400)                        
    y                        1024(0x400)                        
    z                        1024(0x400)                        
  Max Waves Per CU:        32(0x20)                           
  Max Work-item Per CU:    1024(0x400)                        
  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--gfx1035         
      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***

I tried setting HSA_OVERRIDE_GFX_VERSION=10.3.0 before running Blender from the terminal, as that has enabled me to use Tensorflow with this GPU, but it didn't help either.

I'm also using KDE Wayland, not X11, in case that could be an issue. Let me know if there's any more info I can provide or debugging steps I should attempt.

**System Information** Operating system: Opensuse Tumbleweed 20221028 Graphics card: AMD 6700S [Radeon RX 6650 XT] dimgrey_cavefish ([link ](https://www.techpowerup.com/gpu-specs/radeon-rx-6700s.c3868)) **Blender Version** Broken: 3.3 from Opensuse repo as well as 3.3 flatpak Worked: n/a **Short description of error** When going to Edit > Preferences > System > HIP to enable my GPU for rendering it says: No compatible GPUs found for Cycles, requires Vega/RDNA architecture and driver version 22.10 or newer. **Exact steps for others to reproduce the error** Based on the default startup or an attached .blend file (as simple as possible). I just opened a blank new file. **Details** I have Mesa 22.2.2 and ROCm 5.3.0 installed, along with hip-devel, hip-runtime-amd, hipblas, etc., all v5.3.0. I have the following environmental variables: ``` DRI_PRIME=1 LD_LIBRARY_PATH=/opt/rocm-5.3.0/opencl/lib HSA_PATH=/opt/rocm-5.3.0/hsa/ HIP_PATH=/opt/rocm-5.3.0/hip/ HIP_CLANG_PATH=/bin DEVICE_LIB_PATH=/opt/rocm-5.3.0/lib/ HIPCC_VERBOSE=7 ``` These set my GPU as default for graphics, enable OpenCL to work, specify the locations for hip/hsa/clang, and allow more verbose logs to be output when running hipcc. I can verify that Mesa and ROCm work in other applications, so given that the GPU architecture is RDNA 2 and I have all the proper libraries installed, I'm unsure why Blender isn't able to detect things. The output of `hipconfig` is: ``` HIP version : 5.3.22061-e8e78f1a == hipconfig HIP_PATH : /opt/rocm-5.3.0/hip/ ROCM_PATH : /opt/rocm-5.3.0 HIP_COMPILER : clang HIP_PLATFORM : amd HIP_RUNTIME : rocclr CPP_CONFIG : -D__HIP_PLATFORM_HCC__= -D__HIP_PLATFORM_AMD__= -I/opt/rocm-5.3.0/hip*include -I/bin/../lib/clang/15.0.2 -I/opt/rocm-5.3.0/hsa*include == hip-clang HSA_PATH : /opt/rocm-5.3.0/hsa/ HIP_CLANG_PATH : /bin clang version 15.0.2 Target: x86_64-suse-linux Thread model: posix InstalledDir: /bin LLVM (http://llvm.org/): LLVM version 7.0.1 Optimized build. Default target: x86_64-unknown-linux-gnu Host CPU: (unknown) Registered Targets: aarch64 - AArch64 (little endian) aarch64_be - AArch64 (big endian) amdgcn - AMD GCN GPUs arm - ARM arm64 - ARM64 (little endian) armeb - ARM (big endian) bpf - BPF (host endian) bpfeb - BPF (big endian) bpfel - BPF (little endian) hexagon - Hexagon lanai - Lanai mips - Mips mips64 - Mips64 [experimental] mips64el - Mips64el [experimental] mipsel - Mipsel msp430 - MSP430 [experimental] nvptx - NVIDIA PTX 32-bit nvptx64 - NVIDIA PTX 64-bit ppc32 - PowerPC 32 ppc64 - PowerPC 64 ppc64le - PowerPC 64 LE r600 - AMD GPUs HD2XXX-HD6XXX sparc - Sparc sparcel - Sparc LE sparcv9 - Sparc V9 systemz - SystemZ thumb - Thumb thumbeb - Thumb (big endian) x86 - 32-bit X86: Pentium-Pro and above x86-64 - 64-bit X86: EM64T and AMD64 xcore - XCore hip-clang-cxxflags : HIP_PATH=/opt/rocm-5.3.0/hip/ HIP_PLATFORM=amd HIP_COMPILER=clang HIP_RUNTIME=rocclr ROCM_PATH=/opt/rocm-5.3.0 HIP_ROCCLR_HOME=/opt/rocm-5.3.0/hip/ HIP_CLANG_PATH=/bin Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 183. HIP_CLANG_INCLUDE_PATH= HIP_INCLUDE_PATH=/opt/rocm-5.3.0/hip//include HIP_LIB_PATH=/opt/rocm-5.3.0/hip//lib DEVICE_LIB_PATH=/opt/rocm-5.3.0/lib/ HIP_CLANG_TARGET=1 Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 195. Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 196. hipcc-args: --cxxflags hipcc-cmd: /bin/clang++ -L"/opt/rocm-5.3.0/hip*lib" -O3 -lgcc_s -lgcc -lpthread -lm -lrt --cxxflags -Wl,--enable-new-dtags -Wl,-rpath=/opt/rocm-5.3.0/hip*lib:/opt/rocm-5.3.0/lib -lamdhip64 -L/bin/../lib/clang/15.0.2/lib/linux -lclang_rt.builtins-x86_64 -std=c++11 -isystem "/.." -isystem /opt/rocm-5.3.0/hsa*include -isystem "/opt/rocm-5.3.0/hip*include" -O3 hip-clang-ldflags : HIP_PATH=/opt/rocm-5.3.0/hip/ HIP_PLATFORM=amd HIP_COMPILER=clang HIP_RUNTIME=rocclr ROCM_PATH=/opt/rocm-5.3.0 HIP_ROCCLR_HOME=/opt/rocm-5.3.0/hip/ HIP_CLANG_PATH=/bin Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 183. HIP_CLANG_INCLUDE_PATH= HIP_INCLUDE_PATH=/opt/rocm-5.3.0/hip//include HIP_LIB_PATH=/opt/rocm-5.3.0/hip//lib DEVICE_LIB_PATH=/opt/rocm-5.3.0/lib/ HIP_CLANG_TARGET=1 Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 195. Use of uninitialized value $HIP_CLANG_INCLUDE_PATH in concatenation (.) or string at /opt/rocm-5.3.0/hip/bin//hipcc.pl line 196. hipcc-args: --ldflags hipcc-cmd: /bin/clang++ -L"/opt/rocm-5.3.0/hip*lib" -O3 -lgcc_s -lgcc -lpthread -lm -lrt --ldflags -Wl,--enable-new-dtags -Wl,-rpath=/opt/rocm-5.3.0/hip*lib:/opt/rocm-5.3.0/lib -lamdhip64 -L/bin/../lib/clang/15.0.2/lib/linux -lclang_rt.builtins-x86_64 -L"/opt/rocm-5.3.0/hip//lib" -O3 -lgcc_s -lgcc -lpthread -lm -lrt === Environment Variables PATH=/home/username/.local/bin:/home/username/bin:/usr/local/bin:/usr/bin:/bin:/opt/rocm/bin:/opt/rocm/profiler/bin:/opt/rocm/opencl/bin:/opt/rocm/bin:/opt/rocm/profiler/bin:/opt/rocm/opencl/bin/x86_64 HIPCC_VERBOSE=7 HSA_PATH=/opt/rocm-5.3.0/hsa/ HIP_PATH=/opt/rocm-5.3.0/hip/ HIP_CLANG_PATH=/bin LD_LIBRARY_PATH=/opt/rocm-5.3.0/opencl/lib == Linux Kernel Hostname : Hostname Linux Hostname 6.0.5-1-default #1 SMP PREEMPT_DYNAMIC Thu Oct 27 05:14:41 UTC 2022 (7359656) x86_64 x86_64 x86_64 GNU/Linux LSB Version: n/a Distributor ID: openSUSE Description: openSUSE Tumbleweed Release: 20221028 Codename: n/a ``` And the output of `rocminfo` is: ``` 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 6900HS with Radeon Graphics Uuid: CPU-XX Marketing Name: AMD Ryzen 9 6900HS 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) ASIC Revision: 0(0x0) Cacheline Size: 64(0x40) Max Clock Freq. (MHz): 3300 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: 40277236(0x26694f4) KB Allocatable: TRUE Alloc Granule: 4KB Alloc Alignment: 4KB Accessible by all: TRUE Pool 2 Segment: GLOBAL; FLAGS: KERNARG, FINE GRAINED Size: 40277236(0x26694f4) KB Allocatable: TRUE Alloc Granule: 4KB Alloc Alignment: 4KB Accessible by all: TRUE Pool 3 Segment: GLOBAL; FLAGS: COARSE GRAINED Size: 40277236(0x26694f4) KB Allocatable: TRUE Alloc Granule: 4KB Alloc Alignment: 4KB Accessible by all: TRUE ISA Info: *******Agent 2******* Name: gfx1032 Uuid: GPU-XX Marketing Name: AMD Radeon RX 6700S 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: 2048(0x800) KB L3: 32768(0x8000) KB Chip ID: 29679(0x73ef) ASIC Revision: 0(0x0) Cacheline Size: 64(0x40) Max Clock Freq. (MHz): 2435 BDFID: 768 Internal Node ID: 1 Compute Unit: 28 SIMDs per CU: 2 Shader Engines: 4 Shader Arrs. per Eng.: 2 WatchPts on Addr. Ranges:4 Features: KERNEL_DISPATCH Fast F16 Operation: TRUE Wavefront Size: 32(0x20) Workgroup Max Size: 1024(0x400) Workgroup Max Size per Dimension: x 1024(0x400) y 1024(0x400) z 1024(0x400) Max Waves Per CU: 32(0x20) Max Work-item Per CU: 1024(0x400) 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: 8372224(0x7fc000) 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--gfx1032 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 *******Agent 3******* Name: gfx1035 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: 2 Device Type: GPU Cache Info: L1: 16(0x10) KB L2: 2048(0x800) KB Chip ID: 5761(0x1681) ASIC Revision: 2(0x2) Cacheline Size: 64(0x40) Max Clock Freq. (MHz): 2400 BDFID: 1792 Internal Node ID: 2 Compute Unit: 12 SIMDs per CU: 2 Shader Engines: 2 Shader Arrs. per Eng.: 2 WatchPts on Addr. Ranges:4 Features: KERNEL_DISPATCH Fast F16 Operation: TRUE Wavefront Size: 32(0x20) Workgroup Max Size: 1024(0x400) Workgroup Max Size per Dimension: x 1024(0x400) y 1024(0x400) z 1024(0x400) Max Waves Per CU: 32(0x20) Max Work-item Per CU: 1024(0x400) 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--gfx1035 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*** ``` I tried setting `HSA_OVERRIDE_GFX_VERSION=10.3.0` before running Blender from the terminal, as that has enabled me to use Tensorflow with this GPU, but it didn't help either. I'm also using KDE Wayland, not X11, in case that could be an issue. Let me know if there's any more info I can provide or debugging steps I should attempt.

Joseph McElroy commented

2022-10-30 02:34:37 +02:00

Added subscriber: @jmcelroy

Jesse Yurkovich commented

2022-10-30 05:36:33 +01:00

Added subscriber: @deadpin

Jesse Yurkovich commented

2022-10-30 05:36:33 +01:00

Changed status from 'Needs Triage' to: 'Needs User Info'

Jesse Yurkovich commented

2022-10-30 05:36:33 +01:00

There is a possibility that this is the same issue as #102018 (Cycles HIP issues on Debian)

@jmcelroy Can you do a search for the following file on your device and let us know its location: libamdhip64.so

There is a possibility that this is the same issue as #102018 (Cycles HIP issues on Debian) @jmcelroy Can you do a search for the following file on your device and let us know its location: `libamdhip64.so`

Joseph McElroy commented

2022-10-30 12:05:11 +01:00

I found two copies of that file in my ROCm folders, one in /opt/rocm-5.3.0/hip/lib/, and one in /opt/rocm-5.3.0/lib/.

I found two copies of that file in my ROCm folders, one in `/opt/rocm-5.3.0/hip/lib/`, and one in `/opt/rocm-5.3.0/lib/`.

Jesse Yurkovich commented

2022-10-30 21:11:33 +01:00

Thanks for getting back. Yeah, this is probably the same issue as the one mentioned above. I'll close this as a duplicate and carry over your information to the original bug.

Jesse Yurkovich commented

2022-10-30 21:12:02 +01:00

Closed as duplicate of #102018

Jesse Yurkovich closed this issue

2022-10-30 21:12:02 +01:00

Brecht Van Lommel commented

2022-11-01 18:41:37 +01:00

Added subscriber: @brecht

Brecht Van Lommel commented

2022-11-01 18:41:37 +01:00

With the fix I committed for the other report, I think you'd need to add the hip library folder to the path as well:

LD_LIBRARY_PATH=/opt/rocm-5.3.0/opencl/lib:/opt/rocm-5.3.0/hip/lib

With the fix I committed for the other report, I think you'd need to add the hip library folder to the path as well: ``` LD_LIBRARY_PATH=/opt/rocm-5.3.0/opencl/lib:/opt/rocm-5.3.0/hip/lib ```

Joseph McElroy commented

2023-01-27 20:39:28 +01:00

Thank you for your help. I found since the recent updates that simply adding export PATH=/opt/rocm/hip/bin/:$PATHto /etc/profile.local allowed Blender to work with the GPU and I could eliminate the other environmental variables from before (except for DRI_PRIME=1).

Thank you for your help. I found since the recent updates that simply adding `export PATH=/opt/rocm/hip/bin/:$PATH`to /etc/profile.local allowed Blender to work with the GPU and I could eliminate the other environmental variables from before (except for `DRI_PRIME=1`).

Luwef referenced this issue

2023-04-28 16:28:02 +02:00

Blender crashing when enabling HIP. #107442

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

AMD RDNA 2 GPU not detected on Linux with ROCm hip libraries #102158