AMD Vega card w/ ROCm only getting Astropulse jobs. CPU and nVidia GPU getting tasks galore

Message boards : Number crunching : AMD Vega card w/ ROCm only getting Astropulse jobs. CPU and nVidia GPU getting tasks galore
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Justin Turner Arthur Project Donor

Send message
Joined: 20 Oct 03
Posts: 3
Credit: 934,594
RAC: 0
United States
Message 1982408 - Posted: 27 Feb 2019, 5:54:39 UTC
Last modified: 27 Feb 2019, 5:55:19 UTC

I'm in a dual-vendor card setup in Linux (computer 8678180). BOINC recognizes all my devices, I've enabled all the GPU+CPU types in my prefs, and have set use_all_gpus to 1 in cc_config.xml

I've yet to see a regular Seti@home task application use my AMD cards, only Astropulse. The nVidia cards get a mix of CUDA and OpenCL work and the CPU tends to be maxed out and happy.

Startup stdout:
26-Feb-2019 21:57:05 [---] Starting BOINC client version 7.14.2 for x86_64-pc-linux-gnu
26-Feb-2019 21:57:05 [---] log flags: file_xfer, sched_ops, task
26-Feb-2019 21:57:05 [---] Libraries: libcurl/7.64.0 OpenSSL/1.0.2q zlib/1.2.11
26-Feb-2019 21:57:05 [---] Data directory: /var/lib/boinc
26-Feb-2019 21:57:06 [---] CUDA: NVIDIA GPU 0: GeForce GTX 1080 Ti (driver version 415.27, CUDA version 10.0, compute capability 6.1, 4096MB, 3976MB available, 11702 GFLOPS peak)
26-Feb-2019 21:57:06 [---] CUDA: NVIDIA GPU 1: GeForce GTX 1080 Ti (driver version 415.27, CUDA version 10.0, compute capability 6.1, 4096MB, 3976MB available, 11702 GFLOPS peak)
26-Feb-2019 21:57:06 [---] OpenCL: NVIDIA GPU 0: GeForce GTX 1080 Ti (driver version 415.27, device version OpenCL 1.2 CUDA, 11178MB, 3976MB available, 11702 GFLOPS peak)
26-Feb-2019 21:57:06 [---] OpenCL: NVIDIA GPU 1: GeForce GTX 1080 Ti (driver version 415.27, device version OpenCL 1.2 CUDA, 11178MB, 3976MB available, 11702 GFLOPS peak)
26-Feb-2019 21:57:06 [---] OpenCL: AMD/ATI GPU 0: Vega 10 XTX [Radeon Vega Frontier Edition] (driver version 2783.0 (HSA1.1,LC), device version OpenCL 1.2, 16368MB, 16368MB available, 13107 GFLOPS peak)
26-Feb-2019 21:57:06 [---] OpenCL: AMD/ATI GPU 1: Vega 10 XTX [Radeon Vega Frontier Edition] (driver version 2783.0 (HSA1.1,LC), device version OpenCL 1.2, 16368MB, 16368MB available, 13107 GFLOPS peak)
26-Feb-2019 21:57:06 [---] [libc detection] gathered: 2.27, Gentoo 2.27-r6 p3
26-Feb-2019 21:57:06 [---] Host name: allspice
26-Feb-2019 21:57:06 [---] Processor: 8 AuthenticAMD AMD Ryzen 7 1800X Eight-Core Processor [Family 23 Model 1 Stepping 1]
26-Feb-2019 21:57:06 [---] Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb hw_pstate sme ssbd sev vmmcall fsgsbase bmi1 avx2 smep bmi2 rdseed adx smap clflushopt sha_ni xsaveopt xsavec xgetbv1 xsaves clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif overflow_recov succor smca
26-Feb-2019 21:57:06 [---] OS: Linux Gentoo: Gentoo/Linux [4.19.23-gentoo|libc 2.27 (Gentoo 2.27-r6 p3)]
26-Feb-2019 21:57:06 [---] Memory: 15.66 GB physical, 20.75 GB virtual
26-Feb-2019 21:57:06 [---] Disk: 56.84 GB total, 17.14 GB free
26-Feb-2019 21:57:06 [---] Local time is UTC +0 hours
26-Feb-2019 21:57:06 [---] Config: use all coprocessors
26-Feb-2019 21:57:06 [SETI@home] URL http://setiathome.berkeley.edu/; Computer ID 8678180; resource share 100
26-Feb-2019 21:57:06 [SETI@home] General prefs: from SETI@home (last modified 26-Feb-2019 07:26:26)
26-Feb-2019 21:57:06 [SETI@home] Computer location: home
26-Feb-2019 21:57:06 [---] General prefs: using separate prefs for home
26-Feb-2019 21:57:06 [---] Preferences:
26-Feb-2019 21:57:06 [---]    max memory usage when active: 8016.07 MB
26-Feb-2019 21:57:06 [---]    max memory usage when idle: 14428.93 MB
26-Feb-2019 21:57:06 [---]    max disk usage: 17.30 GB
26-Feb-2019 21:57:06 [---]    max upload rate: 20480 bytes/sec
26-Feb-2019 21:57:06 [---]    (to change preferences, visit a project web site or select Preferences in the Manager)
26-Feb-2019 21:57:06 [---] Setting up project and slot directories
26-Feb-2019 21:57:06 [---] Checking active tasks
26-Feb-2019 21:57:06 [---] Setting up GUI RPC socket
26-Feb-2019 21:57:06 [---] Checking presence of 646 project files
26-Feb-2019 21:57:06 Initialization completed
26-Feb-2019 21:57:06 [SETI@home] Sending scheduler request: To report completed tasks.
26-Feb-2019 21:57:06 [SETI@home] Reporting 82 completed tasks
26-Feb-2019 21:57:06 [SETI@home] Requesting new tasks for CPU and NVIDIA GPU and AMD/ATI GPU


coproc_info.xml:
    <coprocs>
    <have_cuda>1</have_cuda>
    <cuda_version>10000</cuda_version>
<coproc_cuda>
   <count>1</count>
   <name>GeForce GTX 1080 Ti</name>
   <available_ram>4169138176.000000</available_ram>
   <have_cuda>1</have_cuda>
   <have_opencl>0</have_opencl>
   <peak_flops>11701760000000.000000</peak_flops>
   <cudaVersion>10000</cudaVersion>
   <drvVersion>41527</drvVersion>
   <totalGlobalMem>4294967295.000000</totalGlobalMem>
   <sharedMemPerBlock>49152.000000</sharedMemPerBlock>
   <regsPerBlock>65536</regsPerBlock>
   <warpSize>32</warpSize>
   <memPitch>2147483647.000000</memPitch>
   <maxThreadsPerBlock>1024</maxThreadsPerBlock>
   <maxThreadsDim>1024 1024 64</maxThreadsDim>
   <maxGridSize>2147483647 65535 65535</maxGridSize>
   <clockRate>1632500</clockRate>
   <totalConstMem>65536.000000</totalConstMem>
   <major>6</major>
   <minor>1</minor>
   <textureAlignment>512.000000</textureAlignment>
   <deviceOverlap>1</deviceOverlap>
   <multiProcessorCount>28</multiProcessorCount>
<pci_info>
   <bus_id>6</bus_id>
   <device_id>0</device_id>
   <domain_id>0</domain_id>
</pci_info>
</coproc_cuda>
<coproc_cuda>
   <count>1</count>
   <name>GeForce GTX 1080 Ti</name>
   <available_ram>4169138176.000000</available_ram>
   <have_cuda>1</have_cuda>
   <have_opencl>0</have_opencl>
   <peak_flops>11701760000000.000000</peak_flops>
   <cudaVersion>10000</cudaVersion>
   <drvVersion>41527</drvVersion>
   <totalGlobalMem>4294967295.000000</totalGlobalMem>
   <sharedMemPerBlock>49152.000000</sharedMemPerBlock>
   <regsPerBlock>65536</regsPerBlock>
   <warpSize>32</warpSize>
   <memPitch>2147483647.000000</memPitch>
   <maxThreadsPerBlock>1024</maxThreadsPerBlock>
   <maxThreadsDim>1024 1024 64</maxThreadsDim>
   <maxGridSize>2147483647 65535 65535</maxGridSize>
   <clockRate>1632500</clockRate>
   <totalConstMem>65536.000000</totalConstMem>
   <major>6</major>
   <minor>1</minor>
   <textureAlignment>512.000000</textureAlignment>
   <deviceOverlap>1</deviceOverlap>
   <multiProcessorCount>28</multiProcessorCount>
<pci_info>
   <bus_id>9</bus_id>
   <device_id>0</device_id>
   <domain_id>0</domain_id>
</pci_info>
</coproc_cuda>
   <ati_opencl>
      <name>Vega 10 XTX [Radeon Vega Frontier Edition]</name>
      <vendor>Advanced Micro Devices, Inc.</vendor>
      <vendor_id>4098</vendor_id>
      <available>1</available>
      <half_fp_config>0</half_fp_config>
      <single_fp_config>191</single_fp_config>
      <double_fp_config>63</double_fp_config>
      <endian_little>1</endian_little>
      <execution_capabilities>1</execution_capabilities>
      <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_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program </extensions>
      <global_mem_size>17163091968</global_mem_size>
      <local_mem_size>65536</local_mem_size>
      <max_clock_frequency>1600</max_clock_frequency>
      <max_compute_units>64</max_compute_units>
      <nv_compute_capability_major>0</nv_compute_capability_major>
      <nv_compute_capability_minor>0</nv_compute_capability_minor>
      <amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
      <amd_simd_width>16</amd_simd_width>
      <amd_simd_instruction_width>1</amd_simd_instruction_width>
      <opencl_platform_version>OpenCL 2.0 AMD-APP.internal (2783.0)</opencl_platform_version>
      <opencl_device_version>OpenCL 1.2 </opencl_device_version>
      <opencl_driver_version>2783.0 (HSA1.1,LC)</opencl_driver_version>
      <device_num>0</device_num>
      <peak_flops>13107200000000.000000</peak_flops>
      <opencl_available_ram>17163091968.000000</opencl_available_ram>
      <opencl_device_index>0</opencl_device_index>
      <warn_bad_cuda>0</warn_bad_cuda>
   </ati_opencl>
   <ati_opencl>
      <name>Vega 10 XTX [Radeon Vega Frontier Edition]</name>
      <vendor>Advanced Micro Devices, Inc.</vendor>
      <vendor_id>4098</vendor_id>
      <available>1</available>
      <half_fp_config>0</half_fp_config>
      <single_fp_config>191</single_fp_config>
      <double_fp_config>63</double_fp_config>
      <endian_little>1</endian_little>
      <execution_capabilities>1</execution_capabilities>
      <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_subgroups cl_khr_depth_images cl_amd_copy_buffer_p2p cl_amd_assembly_program </extensions>
      <global_mem_size>17163091968</global_mem_size>
      <local_mem_size>65536</local_mem_size>
      <max_clock_frequency>1600</max_clock_frequency>
      <max_compute_units>64</max_compute_units>
      <nv_compute_capability_major>0</nv_compute_capability_major>
      <nv_compute_capability_minor>0</nv_compute_capability_minor>
      <amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
      <amd_simd_width>16</amd_simd_width>
      <amd_simd_instruction_width>1</amd_simd_instruction_width>
      <opencl_platform_version>OpenCL 2.0 AMD-APP.internal (2783.0)</opencl_platform_version>
      <opencl_device_version>OpenCL 1.2 </opencl_device_version>
      <opencl_driver_version>2783.0 (HSA1.1,LC)</opencl_driver_version>
      <device_num>1</device_num>
      <peak_flops>13107200000000.000000</peak_flops>
      <opencl_available_ram>17163091968.000000</opencl_available_ram>
      <opencl_device_index>1</opencl_device_index>
      <warn_bad_cuda>0</warn_bad_cuda>
   </ati_opencl>
   <nvidia_opencl>
      <name>GeForce GTX 1080 Ti</name>
      <vendor>NVIDIA Corporation</vendor>
      <vendor_id>4318</vendor_id>
      <available>1</available>
      <half_fp_config>0</half_fp_config>
      <single_fp_config>191</single_fp_config>
      <double_fp_config>63</double_fp_config>
      <endian_little>1</endian_little>
      <execution_capabilities>1</execution_capabilities>
      <extensions>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_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer</extensions>
      <global_mem_size>11721506816</global_mem_size>
      <local_mem_size>49152</local_mem_size>
      <max_clock_frequency>1632</max_clock_frequency>
      <max_compute_units>28</max_compute_units>
      <nv_compute_capability_major>6</nv_compute_capability_major>
      <nv_compute_capability_minor>1</nv_compute_capability_minor>
      <amd_simd_per_compute_unit>0</amd_simd_per_compute_unit>
      <amd_simd_width>0</amd_simd_width>
      <amd_simd_instruction_width>0</amd_simd_instruction_width>
      <opencl_platform_version>OpenCL 1.2 CUDA 10.0.132</opencl_platform_version>
      <opencl_device_version>OpenCL 1.2 CUDA</opencl_device_version>
      <opencl_driver_version>415.27</opencl_driver_version>
      <device_num>0</device_num>
      <peak_flops>11701760000000.000000</peak_flops>
      <opencl_available_ram>4169138176.000000</opencl_available_ram>
      <opencl_device_index>0</opencl_device_index>
      <warn_bad_cuda>0</warn_bad_cuda>
   </nvidia_opencl>
   <nvidia_opencl>
      <name>GeForce GTX 1080 Ti</name>
      <vendor>NVIDIA Corporation</vendor>
      <vendor_id>4318</vendor_id>
      <available>1</available>
      <half_fp_config>0</half_fp_config>
      <single_fp_config>191</single_fp_config>
      <double_fp_config>63</double_fp_config>
      <endian_little>1</endian_little>
      <execution_capabilities>1</execution_capabilities>
      <extensions>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_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer</extensions>
      <global_mem_size>11721506816</global_mem_size>
      <local_mem_size>49152</local_mem_size>
      <max_clock_frequency>1632</max_clock_frequency>
      <max_compute_units>28</max_compute_units>
      <nv_compute_capability_major>6</nv_compute_capability_major>
      <nv_compute_capability_minor>1</nv_compute_capability_minor>
      <amd_simd_per_compute_unit>0</amd_simd_per_compute_unit>
      <amd_simd_width>0</amd_simd_width>
      <amd_simd_instruction_width>0</amd_simd_instruction_width>
      <opencl_platform_version>OpenCL 1.2 CUDA 10.0.132</opencl_platform_version>
      <opencl_device_version>OpenCL 1.2 CUDA</opencl_device_version>
      <opencl_driver_version>415.27</opencl_driver_version>
      <device_num>1</device_num>
      <peak_flops>11701760000000.000000</peak_flops>
      <opencl_available_ram>4169138176.000000</opencl_available_ram>
      <opencl_device_index>1</opencl_device_index>
      <warn_bad_cuda>0</warn_bad_cuda>
   </nvidia_opencl>
<warning>NVIDIA library reports 2 GPUs</warning>
<warning>ATI: libaticalrt.so: cannot open shared object file: No such file or directory</warning>
    </coprocs>

The warning about libaticalrt.so makes sense as I'm using AMD's new open source drivers from the Linux kernel and compute loaders/runtimes from AMD's ROCm project.
ID: 1982408 · Report as offensive
Ben

Send message
Joined: 15 Jun 99
Posts: 54
Credit: 42,165,914
RAC: 130,668
United States
Message 1982418 - Posted: 27 Feb 2019, 6:28:47 UTC

I had the same problem. The only way I could get it to work was to compile the seti app myself. Unfortunately it also gave me too many errors. Switching to the amdgpu "pro" drivers fixed the problem.
ID: 1982418 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 9870
Credit: 930,286,616
RAC: 1,524,299
United States
Message 1982425 - Posted: 27 Feb 2019, 7:07:01 UTC

Unfortunately as Ben indicates, ROCm isn't ready for primetime yet for the majority of mainstream distributed computing projects. Only seems to run for specialized scientific applications so far from what I have been reading on phoronix.com.
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1982425 · Report as offensive
Profile Justin Turner Arthur Project Donor

Send message
Joined: 20 Oct 03
Posts: 3
Credit: 934,594
RAC: 0
United States
Message 1982428 - Posted: 27 Feb 2019, 7:28:34 UTC - in response to Message 1982425.  

Well it seems to be awesome at those few Astropulse OpenCL tasks it gets :D
ID: 1982428 · Report as offensive

Message boards : Number crunching : AMD Vega card w/ ROCm only getting Astropulse jobs. CPU and nVidia GPU getting tasks galore


 
©2019 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.