Unable to use blender with any tested version of rocm installed

[rexy712]

I'm not even trying to render with GPU yet. When I try to render with the cpu while having amd-rocm-meta installed, blender closes with this message:

mesa: CommandLine Error: Option 'help-list' registered more than once!
LLVM ERROR: inconsistency in registered CommandLine options

Seems like llvm-roc and my system llvm-9.0.1_rc3 are having conflicts? Tested with 2.7.0, 2.9.0, 2.10.0, and 3.0.0. Have not tried installing directly from rocm git yet. Uninstalling rocm fixes the issue. clinfo doesn't work without performing mv /etc/OpenCL/vendors/mesa.icd{,.bak}. After this workaround, I get this output:

Number of platforms                               1
  Platform Name                                   AMD Accelerated Parallel Processing
  Platform Vendor                                 Advanced Micro Devices, Inc.
  Platform Version                                OpenCL 2.0 AMD-APP.internal (3019.0)
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd cl_amd_object_metadata cl_amd_event_callback 
  Platform Max metadata object keys (AMD)         8
  Platform Extensions function suffix             AMD

  Platform Name                                   AMD Accelerated Parallel Processing
Number of devices                                 1
  Device Name                                     gfx900
  Device Vendor                                   Advanced Micro Devices, Inc.
  Device Vendor ID                                0x1002
  Device Version                                  OpenCL 2.0 
  Driver Version                                  3019.0 (HSA1.1,LC)
  Device OpenCL C Version                         OpenCL C 2.0 
  Device Type                                     GPU
  Device Board Name (AMD)                         Vega 10 XL/XT [Radeon RX Vega 56/64]
  Device Topology (AMD)                           PCI-E, 2f:00.0
  Device Profile                                  FULL_PROFILE
  Device Available                                Yes
  Compiler Available                              Yes
  Linker Available                                Yes
  Max compute units                               64
  SIMD per compute unit (AMD)                     4
  SIMD width (AMD)                                16
  SIMD instruction width (AMD)                    1
  Max clock frequency                             1630MHz
  Graphics IP (AMD)                               9.0
  Device Partition                                (core)
    Max number of sub-devices                     64
    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                                     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
    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                              8573157376 (7.984GiB)
  Global free memory (AMD)                        8372224 (7.984GiB)
  Global memory channels (AMD)                    64
  Global memory banks per channel (AMD)           4
  Global memory bank width (AMD)                  256 bytes
  Error Correction support                        No
  Max memory allocation                           7287183769 (6.787GiB)
  Unified memory for Host and Device              No
  Shared Virtual Memory (SVM) capabilities        (core)
    Coarse-grained buffer sharing                 Yes
    Fine-grained buffer sharing                   Yes
    Fine-grained system sharing                   No
    Atomics                                       No
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       1024 bits (128 bytes)
  Preferred alignment for atomics                 
    SVM                                           0 bytes
    Global                                        0 bytes
    Local                                         0 bytes
  Max size for global variable                    7287183769 (6.787GiB)
  Preferred total size of global vars             8573157376 (7.984GiB)
  Global Memory cache type                        Read/Write
  Global Memory cache size                        16384 (16KiB)
  Global Memory cache line size                   64 bytes
  Image support                                   No
    Base address alignment for 2D image buffers   0 bytes
    Pitch alignment for 2D image buffers          0 pixels
  Max number of pipe args                         16
  Max active pipe reservations                    16
  Max pipe packet size                            2992216473 (2.787GiB)
  Local memory type                               Local
  Local memory size                               65536 (64KiB)
  Local memory syze per CU (AMD)                  65536 (64KiB)
  Local memory banks (AMD)                        32
  Max number of constant args                     8
  Max constant buffer size                        7287183769 (6.787GiB)
  Preferred constant buffer size (AMD)            16384 (16KiB)
  Max size of kernel argument                     1024
  Queue properties (on host)                      
    Out-of-order execution                        No
    Profiling                                     Yes
  Queue properties (on device)                    
    Out-of-order execution                        Yes
    Profiling                                     Yes
    Preferred size                                262144 (256KiB)
    Max size                                      8388608 (8MiB)
  Max queues on device                            1
  Max events on device                            1024
  Prefer user sync for interop                    Yes
  Number of P2P devices (AMD)                     0
  P2P devices (AMD)                               <printDeviceInfo:147: get number of CL_DEVICE_P2P_DEVICES_AMD : error -30>
  Profiling timer resolution                      1ns
  Profiling timer offset since Epoch (AMD)        0ns (Wed Dec 31 16:00:00 1969)
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
    Thread trace supported (AMD)                  No
    Number of async queues (AMD)                  8
    Max real-time compute queues (AMD)            8
    Max real-time compute units (AMD)             64
  printf() buffer size                            4194304 (4MiB)
  Built-in kernels                                (n/a)
  Device Extensions                               cl_khr_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_media_ops cl_amd_media_ops2 cl_khr_image2d_from_buffer cl_khr_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program 

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                                   gfx900
  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                                   gfx900
  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                                   gfx900

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

[justxi]

Yes I have this problem too. I have to run "OLC_ICD_VENDORS=amdocl64.icd clinfo", or uninstall mesa. Seems to be this problem: https://github.com/justxi/rocm/issues/8#issuecomment-551017447

[justxi]

Due to the fact this is related to ebuilds that are in Gentoo portage now, please submit a bugreport at https://bugs.gentoo.org.

[rexy712]

Alright I'll go report it there and close this issue

[Technohacker]

Funnily enough, using amd-rocm-meta-bin instead of source compiling it prevents blender from crashing on the System preferences (though it eventually crashes on rendering). Maybe when I built ROCm manually it used different LLVM versions?

[justxi]

If you use the ebuilds, I will use llvm-rocm. This is intended because not all sources from AMD are up streamed yet.

Have you tried to render a picture using ROCM-OpenCL-Runtiime?

[Technohacker]

Yup, but in my case what happens is (Blender 2.82): 1) Using rocm-opencl-runtime (or even amd-rocm-meta), Blender would crash on opening the System section of its preferences, with the stacktrace showing a SEGV at getting OpenCL info (but clinfo works), so I can't render an image 2) Using amd-rocm-meta-bin, Blender gets all the way to compiling the render kernels and then crashes with a SEGV on one of them. I believe this is also an upstream bug, but either way it doesn't allow me to render an image :sweat_smile:

I'm assuming I made some error in building ROCm, but for now I'll use the prebuilts :)

[justxi]

So with the prebuilt packages all works fine?

[Technohacker]

I do have to add, for things like LuxRender (from what I remember doing a while back), that worked with my compiled version of ROCm, even rendering. But yes the prebuilts got me farther than my compiled version on Blender :)

(Of course, Blender doesn't work with ROCm very consistently atm, but the prebuilt got me a little further than when I compiled myself)

The prebuilt EBuild required some changes though, the python patch wasn't needed because that program didn't exist, and ROCm's RPM prebuilt uses versioned folders so the ICD file would need the versioned path or a symlink at /opt/rocm

[justxi]

If you made changes to the ebuilds, feel free to create a PR.

[Technohacker]

Surely! :)

[justxi]

Thanks.

[Technohacker]

I've made the PR: #149 :)