GPUs recognized, but no work is being assigned to them?

Message boards : GPUs : GPUs recognized, but no work is being assigned to them?
Message board moderation

To post messages, you must log in.

AuthorMessage
Alexander Falk

Send message
Joined: 22 Apr 16
Posts: 3
United States
Message 69100 - Posted: 22 Apr 2016, 15:12:23 UTC

I have an interesting issue with one of my older machines that I'm trying to use for the PrimeGrid project on BOINC: http://www.primegrid.com/show_host_detail.php?hostid=514865

It is a dedicated box with 3 AMD Radeon HD 6990 graphics cards that I previously had been using for Bitcoin mining and want to repurpose for PrimeGrid operation. The graphics cards have 2 GPUs per card, so a total of 6 GPUs.

I am running Ubuntu Linux 14.04 LTS but in a headless configuration, as the machine is in the basement, so I'm doing all work from within a SSH connection from my desktop. I've installed the latest AMD Catalyst drivers, and I've installed the AMD APP SDK.

However, even though the 6 GPUs are being recognized, I get no OpenCL tasks assigned to that machine from BOINC.

Here is the recognition of the GPUs upon start of boinc:

8: 18-Apr-2016 21:43:41 (low) [] OpenCL: AMD/ATI GPU 0: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
9: 18-Apr-2016 21:43:41 (low) [] OpenCL: AMD/ATI GPU 1: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
10: 18-Apr-2016 21:43:41 (low) [] OpenCL: AMD/ATI GPU 2: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
11: 18-Apr-2016 21:43:41 (low) [] OpenCL: AMD/ATI GPU 3: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
12: 18-Apr-2016 21:43:41 (low) [] OpenCL: AMD/ATI GPU 4: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
13: 18-Apr-2016 21:43:41 (low) [] OpenCL: AMD/ATI GPU 5: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)

But no work is assigned to those GPUs:

47: 18-Apr-2016 21:47:43 (low) [PrimeGrid] Requesting new tasks for ATI
48: 18-Apr-2016 21:47:46 (low) [PrimeGrid] Scheduler request completed: got 0 new tasks

I have double-checked all settings and I am requesting OpenCL based tasks from PrimeGrid (and it works on all my other machines).

I have already read all forums, scoured the Internet, and can't seem to find any explanation. The only possible hint I've found is that other people who are using this old Radeon HD 69xx card have reported that boinc recognizes their cards differently:

OpenCL: AMD/ATI GPU 0: AMD Radeon HD 6900 series (Cayman) (driver version 1124.2, device version OpenCL 1.2 AMD-APP (1124.2), 2048MB, 1812MB available, 50 GFLOPS peak)

The difference is that in their boinc startup messages, they see the GPU identified as "AMD Radeon HD 6900 series (Cayman)" whereas in mined it just says "Cayman".

Does anybody have any idea on (a) what causes that difference; (b) if there is any way I can get boinc on my machine to report it with the full names; and (c) if that is indeed the source of the problem, i.e. does that cause the lack of work being assigned?

For debugging and background info purposes, here are the cards as reported by lspci:

07:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Antilles [Radeon HD 6990] [1002:671d] (prog-if 00 [VGA controller])
Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Device [1002:0b2a]
Flags: bus master, fast devsel, latency 0, IRQ 88
Memory at 80000000 (64-bit, prefetchable) [size=256M]
Memory at fdfc0000 (64-bit, non-prefetchable) [size=128K]
I/O ports at 7000 [size=256]
Expansion ROM at fdfa0000 [disabled] [size=128K]
Capabilities: [50] Power Management version 3
Capabilities: [58] Express Legacy Endpoint, MSI 00
Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
Capabilities: [150] Advanced Error Reporting
Kernel driver in use: fglrx_pci
--
0b:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Antilles [Radeon HD 6990] [1002:671d] (prog-if 00 [VGA controller])
Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Device [1002:0b2a]
Flags: bus master, fast devsel, latency 0, IRQ 86
Memory at a0000000 (64-bit, prefetchable) [size=256M]
Memory at fe2c0000 (64-bit, non-prefetchable) [size=128K]
I/O ports at 9000 [size=256]
Expansion ROM at fe2a0000 [disabled] [size=128K]
Capabilities: [50] Power Management version 3
Capabilities: [58] Express Legacy Endpoint, MSI 00
Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
Capabilities: [150] Advanced Error Reporting
Kernel driver in use: fglrx_pci
--
13:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Antilles [Radeon HD 6990] [1002:671d] (prog-if 00 [VGA controller])
Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Device [1002:0b2a]
Flags: bus master, fast devsel, latency 0, IRQ 84
Memory at d0000000 (64-bit, prefetchable) [size=256M]
Memory at fe9c0000 (64-bit, non-prefetchable) [size=128K]
I/O ports at e000 [size=256]
Expansion ROM at fe9a0000 [disabled] [size=128K]
Capabilities: [50] Power Management version 3
Capabilities: [58] Express Legacy Endpoint, MSI 00
Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
Capabilities: [150] Advanced Error Reporting
Kernel driver in use: fglrx_pci


And here is the full clinfo result, so the cards are all correctly identified and available to OpenCL:

Number of platforms: 1
Platform Profile: FULL_PROFILE
Platform Version: OpenCL 2.0 AMD-APP (1800.11)
Platform Name: AMD Accelerated Parallel Processing
Platform Vendor: Advanced Micro Devices, Inc.
Platform Extensions: cl_khr_icd cl_amd_event_callback cl_amd_offline_devices


Platform Name: AMD Accelerated Parallel Processing
Number of devices: 7
Device Type: CL_DEVICE_TYPE_GPU
Vendor ID: 1002h
Board name:
Device Topology: PCI[ B#19, D#0, F#0 ]
Max compute units: 24
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 830Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 16384
Max image 2D height: 16384
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 2048
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 2117074944
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Max pipe arguments: 0
Max pipe active reservations: 0
Max pipe packet size: 0
Max global variable size: 0
Max global variable preferred total size: 0
Max read/write image args: 0
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: Cayman
Vendor: Advanced Micro Devices, Inc.
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event


Device Type: CL_DEVICE_TYPE_GPU
Vendor ID: 1002h
Board name:
Device Topology: PCI[ B#18, D#0, F#0 ]
Max compute units: 24
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 830Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 16384
Max image 2D height: 16384
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 2048
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 2117074944
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Max pipe arguments: 0
Max pipe active reservations: 0
Max pipe packet size: 0
Max global variable size: 0
Max global variable preferred total size: 0
Max read/write image args: 0
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: Cayman
Vendor: Advanced Micro Devices, Inc.
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event


Device Type: CL_DEVICE_TYPE_GPU
Vendor ID: 1002h
Board name:
Device Topology: PCI[ B#11, D#0, F#0 ]
Max compute units: 24
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 830Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 16384
Max image 2D height: 16384
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 2048
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 2117074944
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Max pipe arguments: 0
Max pipe active reservations: 0
Max pipe packet size: 0
Max global variable size: 0
Max global variable preferred total size: 0
Max read/write image args: 0
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: Cayman
Vendor: Advanced Micro Devices, Inc.
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event


Device Type: CL_DEVICE_TYPE_GPU
Vendor ID: 1002h
Board name:
Device Topology: PCI[ B#10, D#0, F#0 ]
Max compute units: 24
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 830Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 16384
Max image 2D height: 16384
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 2048
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 2117074944
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Max pipe arguments: 0
Max pipe active reservations: 0
Max pipe packet size: 0
Max global variable size: 0
Max global variable preferred total size: 0
Max read/write image args: 0
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: Cayman
Vendor: Advanced Micro Devices, Inc.
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event


Device Type: CL_DEVICE_TYPE_GPU
Vendor ID: 1002h
Board name:
Device Topology: PCI[ B#7, D#0, F#0 ]
Max compute units: 24
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 830Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 16384
Max image 2D height: 16384
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 2048
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 2117074944
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Max pipe arguments: 0
Max pipe active reservations: 0
Max pipe packet size: 0
Max global variable size: 0
Max global variable preferred total size: 0
Max read/write image args: 0
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: Cayman
Vendor: Advanced Micro Devices, Inc.
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event


Device Type: CL_DEVICE_TYPE_GPU
Vendor ID: 1002h
Board name:
Device Topology: PCI[ B#6, D#0, F#0 ]
Max compute units: 24
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 830Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 16384
Max image 2D height: 16384
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 2048
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 2117074944
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Max pipe arguments: 0
Max pipe active reservations: 0
Max pipe packet size: 0
Max global variable size: 0
Max global variable preferred total size: 0
Max read/write image args: 0
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: Cayman
Vendor: Advanced Micro Devices, Inc.
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_amd_image2d_from_buffer_read_only cl_khr_spir cl_khr_gl_event


Device Type: CL_DEVICE_TYPE_CPU
Vendor ID: 1002h
Board name:
Max compute units: 2
Max work items dimensions: 3
Max work items[0]: 1024
Max work items[1]: 1024
Max work items[2]: 1024
Max work group size: 1024
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 2
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 2
Max clock frequency: 3000Mhz
Address bits: 64
Max memory allocation: 2029027328
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 64
Max image 2D width: 8192
Max image 2D height: 8192
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 4096
Alignment (bits) of base address: 1024
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: Yes
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: Read/Write
Cache line size: 64
Cache size: 65536
Global memory size: 2029027328
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Global
Local memory size: 32768
Max pipe arguments: 16
Max pipe active reservations: 16
Max pipe packet size: 2029027328
Max global variable size: 1879048192
Max global variable preferred total size: 1879048192
Max read/write image args: 64
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: No
Fine grain buffer: No
Fine grain system: No
Atomics: No
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 1
Error correction support: 0
Unified memory for Host and Device: 1
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: Yes
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f883f30a430
Name: AMD Athlon(tm) II X2 250 Processor
Vendor: AuthenticAMD
Device OpenCL C version: OpenCL C 1.2
Driver version: 1800.11 (sse2)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1800.11)
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_gl_sharing cl_ext_device_fission cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_spir cl_khr_gl_event
ID: 69100 · Report as offensive
Profile Jord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15493
Netherlands
Message 69101 - Posted: 22 Apr 2016, 15:33:38 UTC

Having your GPU(s) detected by BOINC is only one part of the story, to be able to use them in a project the project needs to have an actual science application that's capable of running on this GPU under the OS of choice.

Not all projects have science applications that can run on an ATI GPU.
Those that do don't all have a science application for the GPU capable of running under Linux. And even here it may be required to install a certain version of drivers, that not all drivers are supported.
Those are questions best asked at the project.

The projects that do allow work to be done on ATI GPUs may run multiple application sorts that not all run on the ATI GPU. In the case of Primegrid, we can see in http://www.primegrid.com/apps.php that they have applications for ATI GPUs for under Linux, but only for PPS (Sieve), and all Genefer sub-projects.

If you do not have the project set to ask for work from those sub-projects, you don't get work for them. Also, if these sub-projects are out of work, it's difficult for them to give out some. You can see the status of available work per sub-project at the top of the right-side column on Primegrid's homepage http://www.primegrid.com/.

And when in doubt, ask on their forums. It's their work, their applications, they'll ought to know why your GPUs don't get any at this time.
ID: 69101 · Report as offensive
Richard Haselgrove
Volunteer tester
Help desk expert

Send message
Joined: 5 Oct 06
Posts: 5082
United Kingdom
Message 69102 - Posted: 22 Apr 2016, 15:36:55 UTC - in response to Message 69100.  

"BOINC version 7.2.42" - is that standard across your other machines too?
ID: 69102 · Report as offensive
Alexander Falk

Send message
Joined: 22 Apr 16
Posts: 3
United States
Message 69103 - Posted: 22 Apr 2016, 16:51:59 UTC - in response to Message 69101.  

Having your GPU(s) detected by BOINC is only one part of the story, to be able to use them in a project the project needs to have an actual science application that's capable of running on this GPU under the OS of choice.

Not all projects have science applications that can run on an ATI GPU.
Those that do don't all have a science application for the GPU capable of running under Linux. And even here it may be required to install a certain version of drivers, that not all drivers are supported.
Those are questions best asked at the project.

I did already do that as a first inclination that it was an issue related to the particular project, rather than BOINC. However, I received no help there and upon further investigation it looks to me like the GPUs are not identified correctly by BOINC, which is why I posted here.

The projects that do allow work to be done on ATI GPUs may run multiple application sorts that not all run on the ATI GPU. In the case of Primegrid, we can see in http://www.primegrid.com/apps.php that they have applications for ATI GPUs for under Linux, but only for PPS (Sieve), and all Genefer sub-projects.

That's correct, and I'm trying to do exactly those calculations where they have applications for ATI GPUs under Linux.

If you do not have the project set to ask for work from those sub-projects, you don't get work for them. Also, if these sub-projects are out of work, it's difficult for them to give out some. You can see the status of available work per sub-project at the top of the right-side column on Primegrid's homepage http://www.primegrid.com/.

And when in doubt, ask on their forums. It's their work, their applications, they'll ought to know why your GPUs don't get any at this time.

That was my initial thought, too, but it appears to be an issue with how BOINC detects and reports the GPUs.

One more data point worth mentioning: I have not been able to identify yet why calInit() returns 1 (instead of 0) during the GPU-detection code - and what the significance of that (if any) is: calInit() returned 1

See full --get_messages output here:

1: 22-Apr-2016 12:36:07 (low) [] Starting BOINC client version 7.2.42 for x86_64-pc-linux-gnu
2: 22-Apr-2016 12:36:07 (low) [] log flags: file_xfer, sched_ops, task, coproc_debug
3: 22-Apr-2016 12:36:07 (low) [] Libraries: libcurl/7.35.0 OpenSSL/1.0.1f zlib/1.2.8 libidn/1.28 librtmp/2.3
4: 22-Apr-2016 12:36:07 (low) [] Data directory: /var/lib/boinc-client
5: 22-Apr-2016 12:36:07 (low) [] [coproc] launching child process at /usr/bin/boinc
6: 22-Apr-2016 12:36:07 (low) [] [coproc] relative to directory /var/lib/boinc-client
7: 22-Apr-2016 12:36:07 (low) [] [coproc] with data directory /var/lib/boinc-client
8: 22-Apr-2016 12:36:07 (low) [] OpenCL: AMD/ATI GPU 0: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
9: 22-Apr-2016 12:36:07 (low) [] OpenCL: AMD/ATI GPU 1: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
10: 22-Apr-2016 12:36:07 (low) [] OpenCL: AMD/ATI GPU 2: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
11: 22-Apr-2016 12:36:07 (low) [] OpenCL: AMD/ATI GPU 3: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
12: 22-Apr-2016 12:36:07 (low) [] OpenCL: AMD/ATI GPU 4: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
13: 22-Apr-2016 12:36:07 (low) [] OpenCL: AMD/ATI GPU 5: Cayman (driver version 1800.11 (VM), device version OpenCL 1.2 AMD-APP (1800.11), 2019MB, 2019MB available, 1594 GFLOPS peak)
14: 22-Apr-2016 12:36:07 (low) [] OpenCL CPU: AMD Athlon(tm) II X2 250 Processor (OpenCL driver vendor: Advanced Micro Devices, Inc., driver version 1800.11 (sse2), device version OpenCL 1.2 AMD-APP (1800.11))
15: 22-Apr-2016 12:36:07 (low) [] No NVIDIA library found
16: 22-Apr-2016 12:36:07 (low) [] calInit() returned 1

That is not present in the get_messages output on my other machines. So it seems something is going wrong with the coprocessor detection in BOINC...
ID: 69103 · Report as offensive
Alexander Falk

Send message
Joined: 22 Apr 16
Posts: 3
United States
Message 69104 - Posted: 22 Apr 2016, 16:52:22 UTC - in response to Message 69102.  

"BOINC version 7.2.42" - is that standard across your other machines too?

Yes, that's the standard for all my Linux machines.
ID: 69104 · Report as offensive
Juha
Volunteer developer
Volunteer tester
Help desk expert

Send message
Joined: 20 Nov 12
Posts: 801
Finland
Message 69105 - Posted: 22 Apr 2016, 19:12:36 UTC - in response to Message 69103.  

16: 22-Apr-2016 12:36:07 (low) [] calInit() returned 1


Either the GPU or the driver doesn't support CAL. Judging by the r9 390 not working thread PrimeGrid's server requires CAL support even for OpenCL apps.

If the GPU supports CAL you could try an older driver or you could try some of the workarounds suggested in the PrimeGrid thread.
ID: 69105 · Report as offensive
Claggy

Send message
Joined: 23 Apr 07
Posts: 1112
United Kingdom
Message 69107 - Posted: 22 Apr 2016, 19:58:21 UTC - in response to Message 69105.  

16: 22-Apr-2016 12:36:07 (low) [] calInit() returned 1


Either the GPU or the driver doesn't support CAL. Judging by the r9 390 not working thread PrimeGrid's server requires CAL support even for OpenCL apps.

If the GPU supports CAL you could try an older driver or you could try some of the workarounds suggested in the PrimeGrid thread.

Or PrimeGrid fixes their apps to not require CAL support, or fix their planclasses to not require CAL support.

Claggy
ID: 69107 · Report as offensive

Message boards : GPUs : GPUs recognized, but no work is being assigned to them?

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.