Crash using GPU Compute in Cycles on AMD in 2.92 and 2.93 but not 2.91 #86248

New Issue

Ian Andolina · 2021-03-04T01:48:05+01:00

Ian Andolina commented

2021-03-04 01:48:05 +01:00

System Information
Operating system: Ubuntu 20.10
Graphics card: Radeon Pro WX5100

Blender Version
Broken: 2.92, 2.93
Worked: 2.91

Enabling GPU compute for cycles crashes Blender in 2.92 but not 2.91

Exact steps for others to reproduce the error

I am using latest MESA (21.1) OpenGL drivers, but the AMD-Pro OpenCL driver (installed using this script: https://gist.github.com/tuxutku/79daa2edca131c1525a136b650cdbe0a). This works fine in clinfo and LuxCore benchmarks all run without issue. I am using the vanilla downloads for Blender from the website.

Using the default cube blend and clean preferences, in 2.91.2 I can render the cube using either CPU or GPU:

➜ Apps/blender-2.91.2/blender -d --debug-gpu                     
Switching to fully guarded memory allocator.
Blender 2.91.2
Build: 2021-01-19 16:16:34 Linux Release
argv[0] = Apps/blender-2.91.2/blender
argv[1] = -d
argv[2] = --debug-gpu
Read prefs: /home/cog5/.config/blender/2.91/config/userpref.blend
read file 
  Version 280 sub 39 date unknown hash unknown
found bundled python: /home/cog5/Apps/blender-2.91.2/2.91/python
LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction
renamable $vgpr1 = V_CNDMASK_B32_e32 32768, killed $vgpr2, implicit killed $vcc, implicit $exec
LLVM failed to compile shader
radeonsi: can't compile a main shader part
LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction
renamable $vgpr3 = V_CNDMASK_B32_e32 32768, killed $vgpr3, implicit killed $vcc, implicit $exec
LLVM failed to compile shader
radeonsi: can't compile a main shader part
read file 
  Version 280 sub 39 date unknown hash unknown
/opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory
Saved session recovery to '/tmp/quit.blend'

Blender quit

If I do exactly the same steps in 2.92.0 (or 2.93alpha), I see this:

➜ Apps/blender-2.92.0/blender -d --debug-gpu                   
Switching to fully guarded memory allocator.
Blender 2.92.0
Build: 2021-02-25 09:31:37 Linux Release
argv[0] = Apps/blender-2.92.0/blender
argv[1] = -d
argv[2] = --debug-gpu
Read prefs: /home/cog5/.config/blender/2.92/config/userpref.blend
read file 
  Version 280 sub 39 date unknown hash unknown
LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction
renamable $vgpr1 = V_CNDMASK_B32_e32 32768, killed $vgpr2, implicit killed $vcc, implicit $exec
LLVM failed to compile shader
radeonsi: can't compile a main shader part
LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction
renamable $vgpr3 = V_CNDMASK_B32_e32 32768, killed $vgpr3, implicit killed $vcc, implicit $exec
LLVM failed to compile shader
radeonsi: can't compile a main shader part
read file 
  Version 280 sub 39 date unknown hash unknown
/opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory
Cycles: compiling OpenCL program background...
/opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory
Segmentation fault (core dumped)
Cycles: compiling OpenCL program background...
Writing: /tmp/blender.crash.txt
zsh: segmentation fault (core dumped)  Apps/blender-2.92.0/blender -d --debug-gpu

# Blender 2.92.0, Commit date: 2021-02-24 16:25, Hash 02948a2cab44
bpy.context.space_data.context = 'OUTPUT'  # Property
bpy.context.space_data.context = 'RENDER'  # Property
bpy.context.scene.render.engine = 'CYCLES'  # Property
bpy.context.space_data.shading.type = 'RENDERED'  # Property
bpy.context.scene.cycles.device = 'GPU'  # Property

# backtrace
Apps/blender-2.92.0/blender(BLI_system_backtrace+0x20) [0x865aed0]
Apps/blender-2.92.0/blender() [0xeb124a]
/lib/x86_64-linux-gnu/libc.so.6(+0x41950) [0x7fcf01ba0950]
/lib/libamdocl-orca64.so(+0x10364cc) [0x7fce990f94cc]
/lib/libamdocl-orca64.so(+0x1044e4f) [0x7fce99107e4f]
/lib/libamdocl-orca64.so(+0x14ea205) [0x7fce995ad205]
/lib/libamdocl-orca64.so(+0xf2e029) [0x7fce98ff1029]
/lib/libamdocl-orca64.so(+0xf2ef7f) [0x7fce98ff1f7f]
/lib/libamdocl-orca64.so(+0xf1bdff) [0x7fce98fdedff]
/lib/libamdocl-orca64.so(+0x17d62d9) [0x7fce998992d9]
/lib/libamdocl-orca64.so(+0x1795251) [0x7fce99858251]
/lib/libamdocl-orca64.so(+0x1799505) [0x7fce9985c505]
/lib/libamdocl-orca64.so(aclCompile+0x1b8) [0x7fce9982c698]
/lib/libamdocl-orca64.so(+0xee7a12) [0x7fce98faaa12]
/lib/libamdocl-orca64.so(+0xccb837) [0x7fce98d8e837]
/lib/libamdocl-orca64.so(+0xcb1865) [0x7fce98d74865]
/lib/libamdocl-orca64.so(clBuildProgram+0x224) [0x7fce98d60334]
Apps/blender-2.92.0/blender(_ZN3ccl12OpenCLDevice13OpenCLProgram12build_kernelEPKSs+0x11f) [0x25bd0af]
Apps/blender-2.92.0/blender(_ZN3ccl12OpenCLDevice13OpenCLProgram14compile_kernelEPKSs+0x1cb) [0x25bd95b]
Apps/blender-2.92.0/blender(_ZN3ccl12OpenCLDevice13OpenCLProgram7compileEv+0x533) [0x25c0533]
Apps/blender-2.92.0/blender() [0x63c1405]
Apps/blender-2.92.0/blender() [0x1233735]
Apps/blender-2.92.0/blender() [0x12339eb]
Apps/blender-2.92.0/blender() [0x1225d47]
Apps/blender-2.92.0/blender() [0x122d4b0]
Apps/blender-2.92.0/blender() [0x122f4bc]
Apps/blender-2.92.0/blender() [0x122f6b9]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x9590) [0x7fcf0229a590]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x43) [0x7fcf01c78223]

# Python backtrace

Here is clinfo status:

➜ clinfo                     
/opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory
Number of platforms                               1
  Platform Name                                   AMD Accelerated Parallel Processing
  Platform Vendor                                 Advanced Micro Devices, Inc.
  Platform Version                                OpenCL 2.1 AMD-APP (3188.4)
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd cl_amd_event_callback cl_amd_offline_devices 
  Platform Host timer resolution                  1ns
  Platform Extensions function suffix             AMD

  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 1
  Device Name                                     Ellesmere
  Device Vendor                                   Advanced Micro Devices, Inc.
  Device Vendor ID                                0x1002
  Device Version                                  OpenCL 1.2 AMD-APP (3188.4)
  Driver Version                                  3188.4
  Device OpenCL C Version                         OpenCL C 1.2 
  Device Type                                     GPU
  Device Board Name (AMD)                         AMD Radeon Graphics
  Device Topology (AMD)                           PCI-E, 65:00.0
  Device Profile                                  FULL_PROFILE
  Device Available                                Yes
  Compiler Available                              Yes
  Linker Available                                Yes
  Max compute units                               28
  SIMD per compute unit (AMD)                     4
  SIMD width (AMD)                                16
  SIMD instruction width (AMD)                    1
  Max clock frequency                             1086MHz
  Graphics IP (AMD)                               8.0
  Device Partition                                (core)
    Max number of sub-devices                     28
    Supported partition types                     None
    Supported affinity domains                    (n/a)
  Max work item dimensions                        3
  Max work item sizes                             1024x1024x1024
  Max work group size                             256
  Preferred work group size (AMD)                 256
  Max work group size (AMD)                       1024
  Preferred work group size multiple              64
  Wavefront width (AMD)                           64
  Preferred / native vector sizes                 
    char                                                 4 / 4       
    short                                                2 / 2       
    int                                                  1 / 1       
    long                                                 1 / 1       
    half                                                 1 / 1        (cl_khr_fp16)
    float                                                1 / 1       
    double                                               1 / 1        (cl_khr_fp64)
  Half-precision Floating-point support           (cl_khr_fp16)
    Denormals                                     No
    Infinity and NANs                             No
    Round to nearest                              No
    Round to zero                                 No
    Round to infinity                             No
    IEEE754-2008 fused multiply-add               No
    Support is emulated in software               No
  Single-precision Floating-point support         (core)
    Denormals                                     No
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  Yes
  Double-precision Floating-point support         (cl_khr_fp64)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Address bits                                    64, Little-Endian
  Global memory size                              8076738560 (7.522GiB)
  Global free memory (AMD)                        7867320 (7.503GiB)
  Global memory channels (AMD)                    8
  Global memory banks per channel (AMD)           16
  Global memory bank width (AMD)                  256 bytes
  Error Correction support                        No
  Max memory allocation                           6700353126 (6.24GiB)
  Unified memory for Host and Device              No
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       2048 bits (256 bytes)
  Global Memory cache type                        Read/Write
  Global Memory cache size                        16384 (16KiB)
  Global Memory cache line size                   64 bytes
  Image support                                   Yes
    Max number of samplers per kernel             16
    Max size for 1D images from buffer            134217728 pixels
    Max 1D or 2D image array size                 2048 images
    Base address alignment for 2D image buffers   256 bytes
    Pitch alignment for 2D image buffers          256 pixels
    Max 2D image size                             16384x16384 pixels
    Max 3D image size                             2048x2048x2048 pixels
    Max number of read image args                 128
    Max number of write image args                8
  Local memory type                               Local
  Local memory size                               32768 (32KiB)
  Local memory syze per CU (AMD)                  65536 (64KiB)
  Local memory banks (AMD)                        32
  Max number of constant args                     8
  Max constant buffer size                        6700353126 (6.24GiB)
  Preferred constant buffer size (AMD)            16384 (16KiB)
  Max size of kernel argument                     1024
  Queue properties                                
    Out-of-order execution                        No
    Profiling                                     Yes
  Prefer user sync for interop                    Yes
  Profiling timer resolution                      1ns
  Profiling timer offset since Epoch (AMD)        1614769070051965685ns (Wed Mar  3 18:57:50 2021)
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
    Thread trace supported (AMD)                  Yes
    Number of async queues (AMD)                  2
    Max real-time compute queues (AMD)            0
    Max real-time compute units (AMD)             0
    SPIR versions                                 1.2
  printf() buffer size                            4194304 (4MiB)
  Built-in kernels                                (n/a)
  Device Extensions                               cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event 

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  AMD Accelerated Parallel Processing
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   Success [AMD]
  clCreateContext(NULL, ...) [default]            Success [AMD]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   Ellesmere
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   Ellesmere
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  Success (1)
    Platform Name                                 AMD Accelerated Parallel Processing
    Device Name                                   Ellesmere

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.2.12
  ICD loader Profile                              OpenCL 2.2
(base)

**System Information** Operating system: Ubuntu 20.10 Graphics card: Radeon Pro WX5100 **Blender Version** Broken: 2.92, 2.93 Worked: 2.91 **Enabling GPU compute for cycles crashes Blender in 2.92 but not 2.91** **Exact steps for others to reproduce the error** I am using latest MESA (21.1) OpenGL drivers, but the AMD-Pro OpenCL driver (installed using this script: https://gist.github.com/tuxutku/79daa2edca131c1525a136b650cdbe0a). This works fine in clinfo and LuxCore benchmarks all run without issue. I am using the vanilla downloads for Blender from the website. Using the default cube blend and clean preferences, in 2.91.2 I can render the cube using either CPU or GPU: ``` ➜ Apps/blender-2.91.2/blender -d --debug-gpu Switching to fully guarded memory allocator. Blender 2.91.2 Build: 2021-01-19 16:16:34 Linux Release argv[0] = Apps/blender-2.91.2/blender argv[1] = -d argv[2] = --debug-gpu Read prefs: /home/cog5/.config/blender/2.91/config/userpref.blend read file Version 280 sub 39 date unknown hash unknown found bundled python: /home/cog5/Apps/blender-2.91.2/2.91/python LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction renamable $vgpr1 = V_CNDMASK_B32_e32 32768, killed $vgpr2, implicit killed $vcc, implicit $exec LLVM failed to compile shader radeonsi: can't compile a main shader part LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction renamable $vgpr3 = V_CNDMASK_B32_e32 32768, killed $vgpr3, implicit killed $vcc, implicit $exec LLVM failed to compile shader radeonsi: can't compile a main shader part read file Version 280 sub 39 date unknown hash unknown /opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory Saved session recovery to '/tmp/quit.blend' Blender quit ``` If I do exactly the same steps in 2.92.0 (or 2.93alpha), I see this: ``` ➜ Apps/blender-2.92.0/blender -d --debug-gpu Switching to fully guarded memory allocator. Blender 2.92.0 Build: 2021-02-25 09:31:37 Linux Release argv[0] = Apps/blender-2.92.0/blender argv[1] = -d argv[2] = --debug-gpu Read prefs: /home/cog5/.config/blender/2.92/config/userpref.blend read file Version 280 sub 39 date unknown hash unknown LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction renamable $vgpr1 = V_CNDMASK_B32_e32 32768, killed $vgpr2, implicit killed $vcc, implicit $exec LLVM failed to compile shader radeonsi: can't compile a main shader part LLVM triggered Diagnostic Handler: Illegal instruction detected: VOP* instruction violates constant bus restriction renamable $vgpr3 = V_CNDMASK_B32_e32 32768, killed $vgpr3, implicit killed $vcc, implicit $exec LLVM failed to compile shader radeonsi: can't compile a main shader part read file Version 280 sub 39 date unknown hash unknown /opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory Cycles: compiling OpenCL program background... /opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory Segmentation fault (core dumped) Cycles: compiling OpenCL program background... Writing: /tmp/blender.crash.txt zsh: segmentation fault (core dumped) Apps/blender-2.92.0/blender -d --debug-gpu ``` ``` # Blender 2.92.0, Commit date: 2021-02-24 16:25, Hash 02948a2cab44 bpy.context.space_data.context = 'OUTPUT' # Property bpy.context.space_data.context = 'RENDER' # Property bpy.context.scene.render.engine = 'CYCLES' # Property bpy.context.space_data.shading.type = 'RENDERED' # Property bpy.context.scene.cycles.device = 'GPU' # Property # backtrace Apps/blender-2.92.0/blender(BLI_system_backtrace+0x20) [0x865aed0] Apps/blender-2.92.0/blender() [0xeb124a] /lib/x86_64-linux-gnu/libc.so.6(+0x41950) [0x7fcf01ba0950] /lib/libamdocl-orca64.so(+0x10364cc) [0x7fce990f94cc] /lib/libamdocl-orca64.so(+0x1044e4f) [0x7fce99107e4f] /lib/libamdocl-orca64.so(+0x14ea205) [0x7fce995ad205] /lib/libamdocl-orca64.so(+0xf2e029) [0x7fce98ff1029] /lib/libamdocl-orca64.so(+0xf2ef7f) [0x7fce98ff1f7f] /lib/libamdocl-orca64.so(+0xf1bdff) [0x7fce98fdedff] /lib/libamdocl-orca64.so(+0x17d62d9) [0x7fce998992d9] /lib/libamdocl-orca64.so(+0x1795251) [0x7fce99858251] /lib/libamdocl-orca64.so(+0x1799505) [0x7fce9985c505] /lib/libamdocl-orca64.so(aclCompile+0x1b8) [0x7fce9982c698] /lib/libamdocl-orca64.so(+0xee7a12) [0x7fce98faaa12] /lib/libamdocl-orca64.so(+0xccb837) [0x7fce98d8e837] /lib/libamdocl-orca64.so(+0xcb1865) [0x7fce98d74865] /lib/libamdocl-orca64.so(clBuildProgram+0x224) [0x7fce98d60334] Apps/blender-2.92.0/blender(_ZN3ccl12OpenCLDevice13OpenCLProgram12build_kernelEPKSs+0x11f) [0x25bd0af] Apps/blender-2.92.0/blender(_ZN3ccl12OpenCLDevice13OpenCLProgram14compile_kernelEPKSs+0x1cb) [0x25bd95b] Apps/blender-2.92.0/blender(_ZN3ccl12OpenCLDevice13OpenCLProgram7compileEv+0x533) [0x25c0533] Apps/blender-2.92.0/blender() [0x63c1405] Apps/blender-2.92.0/blender() [0x1233735] Apps/blender-2.92.0/blender() [0x12339eb] Apps/blender-2.92.0/blender() [0x1225d47] Apps/blender-2.92.0/blender() [0x122d4b0] Apps/blender-2.92.0/blender() [0x122f4bc] Apps/blender-2.92.0/blender() [0x122f6b9] /lib/x86_64-linux-gnu/libpthread.so.0(+0x9590) [0x7fcf0229a590] /lib/x86_64-linux-gnu/libc.so.6(clone+0x43) [0x7fcf01c78223] # Python backtrace ``` Here is `clinfo` status: ``` ➜ clinfo /opt/amdgpu/share/libdrm/amdgpu.ids: No such file or directory Number of platforms 1 Platform Name AMD Accelerated Parallel Processing Platform Vendor Advanced Micro Devices, Inc. Platform Version OpenCL 2.1 AMD-APP (3188.4) Platform Profile FULL_PROFILE Platform Extensions cl_khr_icd cl_amd_event_callback cl_amd_offline_devices Platform Host timer resolution 1ns Platform Extensions function suffix AMD Platform Name AMD Accelerated Parallel Processing Number of devices 1 Device Name Ellesmere Device Vendor Advanced Micro Devices, Inc. Device Vendor ID 0x1002 Device Version OpenCL 1.2 AMD-APP (3188.4) Driver Version 3188.4 Device OpenCL C Version OpenCL C 1.2 Device Type GPU Device Board Name (AMD) AMD Radeon Graphics Device Topology (AMD) PCI-E, 65:00.0 Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Linker Available Yes Max compute units 28 SIMD per compute unit (AMD) 4 SIMD width (AMD) 16 SIMD instruction width (AMD) 1 Max clock frequency 1086MHz Graphics IP (AMD) 8.0 Device Partition (core) Max number of sub-devices 28 Supported partition types None Supported affinity domains (n/a) Max work item dimensions 3 Max work item sizes 1024x1024x1024 Max work group size 256 Preferred work group size (AMD) 256 Max work group size (AMD) 1024 Preferred work group size multiple 64 Wavefront width (AMD) 64 Preferred / native vector sizes char 4 / 4 short 2 / 2 int 1 / 1 long 1 / 1 half 1 / 1 (cl_khr_fp16) float 1 / 1 double 1 / 1 (cl_khr_fp64) Half-precision Floating-point support (cl_khr_fp16) Denormals No Infinity and NANs No Round to nearest No Round to zero No Round to infinity No IEEE754-2008 fused multiply-add No Support is emulated in software No Single-precision Floating-point support (core) Denormals No Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Correctly-rounded divide and sqrt operations Yes Double-precision Floating-point support (cl_khr_fp64) Denormals Yes Infinity and NANs Yes Round to nearest Yes Round to zero Yes Round to infinity Yes IEEE754-2008 fused multiply-add Yes Support is emulated in software No Address bits 64, Little-Endian Global memory size 8076738560 (7.522GiB) Global free memory (AMD) 7867320 (7.503GiB) Global memory channels (AMD) 8 Global memory banks per channel (AMD) 16 Global memory bank width (AMD) 256 bytes Error Correction support No Max memory allocation 6700353126 (6.24GiB) Unified memory for Host and Device No Minimum alignment for any data type 128 bytes Alignment of base address 2048 bits (256 bytes) Global Memory cache type Read/Write Global Memory cache size 16384 (16KiB) Global Memory cache line size 64 bytes Image support Yes Max number of samplers per kernel 16 Max size for 1D images from buffer 134217728 pixels Max 1D or 2D image array size 2048 images Base address alignment for 2D image buffers 256 bytes Pitch alignment for 2D image buffers 256 pixels Max 2D image size 16384x16384 pixels Max 3D image size 2048x2048x2048 pixels Max number of read image args 128 Max number of write image args 8 Local memory type Local Local memory size 32768 (32KiB) Local memory syze per CU (AMD) 65536 (64KiB) Local memory banks (AMD) 32 Max number of constant args 8 Max constant buffer size 6700353126 (6.24GiB) Preferred constant buffer size (AMD) 16384 (16KiB) Max size of kernel argument 1024 Queue properties Out-of-order execution No Profiling Yes Prefer user sync for interop Yes Profiling timer resolution 1ns Profiling timer offset since Epoch (AMD) 1614769070051965685ns (Wed Mar 3 18:57:50 2021) Execution capabilities Run OpenCL kernels Yes Run native kernels No Thread trace supported (AMD) Yes Number of async queues (AMD) 2 Max real-time compute queues (AMD) 0 Max real-time compute units (AMD) 0 SPIR versions 1.2 printf() buffer size 4194304 (4MiB) Built-in kernels (n/a) Device Extensions cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_amd_bus_addressable_memory cl_khr_spir cl_khr_gl_event NULL platform behavior clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) AMD Accelerated Parallel Processing clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [AMD] clCreateContext(NULL, ...) [default] Success [AMD] clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1) Platform Name AMD Accelerated Parallel Processing Device Name Ellesmere clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1) Platform Name AMD Accelerated Parallel Processing Device Name Ellesmere clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1) Platform Name AMD Accelerated Parallel Processing Device Name Ellesmere ICD loader properties ICD loader Name OpenCL ICD Loader ICD loader Vendor OCL Icd free software ICD loader Version 2.2.12 ICD loader Profile OpenCL 2.2 (base) ```

Ian Andolina commented

2021-03-04 01:48:05 +01:00

Added subscriber: @iandol

Ian Andolina commented

2021-03-04 01:54:04 +01:00

Note it does not seem to be related to https://developer.blender.org/T84461 as that also occurs on 2.91 and the error messages are different.

Ian Andolina commented

2021-03-04 01:54:51 +01:00

But it does seem similar to this one: https://developer.blender.org/T85099 -- possibly I am a duplicate of that one, although I didn't see this split-kernel error in 2.91

Ian Andolina commented

2021-03-04 01:56:42 +01:00

system-info.txt system-info.txt

[system-info.txt](https://archive.blender.org/developer/F9863922/system-info.txt) system-info.txt

Ian Andolina commented

2021-03-04 07:48:46 +01:00

Just on an off chance, I reverted to the older MESA drivers that come by default with Ubuntu, but no change to this crash:

OpenGL renderer string: AMD Radeon (TM) Pro WX 5100 Graphics (POLARIS10, DRM 3.40.0, 5.10.17-051017-lowlatency, LLVM 11.0.0)
OpenGL core profile version string: 4.6 (Core Profile) Mesa 20.2.6

Just on an off chance, I reverted to the older MESA drivers that come by default with Ubuntu, but no change to this crash: ``` OpenGL renderer string: AMD Radeon (TM) Pro WX 5100 Graphics (POLARIS10, DRM 3.40.0, 5.10.17-051017-lowlatency, LLVM 11.0.0) OpenGL core profile version string: 4.6 (Core Profile) Mesa 20.2.6 ```

Svend Sønderby commented

2021-03-04 20:32:59 +01:00

Added subscriber: @Svend

Germano Cavalcante commented

2021-04-12 15:08:32 +02:00

Added subscriber: @mano-wii

Germano Cavalcante commented

2021-04-12 15:08:32 +02:00

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

Germano Cavalcante commented

2021-04-12 15:08:32 +02:00

91c44fe885 may have solved the problem, you can try the latest daily build .
(If you can't, no problem, we can take it as resolved)

91c44fe885 may have solved the problem, you can try the [latest daily build ](https://builder.blender.org/download/). (If you can't, no problem, we can take it as resolved)

Germano Cavalcante commented

2021-04-19 16:24:20 +02:00

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

Germano Cavalcante closed this issue

2021-04-19 16:24:20 +02:00

Germano Cavalcante commented

2021-04-19 16:24:20 +02:00

No activity for more than a week. As per the tracker policy we assume the issue is gone and can be closed.

Thanks for the report. If the problem persists please open a new report with the required information.

No activity for more than a week. As per the tracker policy we assume the issue is gone and can be closed. Thanks for the report. If the problem persists please open a new report with the required information.

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

Crash using GPU Compute in Cycles on AMD in 2.92 and 2.93 but not 2.91 #86248