1

How is it possible to get the size of the memory on a GPU, which is available for programs with use OpenCL for computations, such as darktable?

I know of lspci which gives some general informations, but not what I am looking for.

$ sudo lspci -v -s 01:00.0
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Curacao XT [Radeon R9 270X] (prog-if 00 [VGA controller])
    Subsystem: Gigabyte Technology Co., Ltd Device 227d
    Flags: bus master, fast devsel, latency 0, IRQ 49
    Memory at d0000000 (64-bit, prefetchable) [size=256M]
    Memory at fe780000 (64-bit, non-prefetchable) [size=256K]
    I/O ports at c000 [size=256]
    Expansion ROM at fe7c0000 [disabled] [size=128K]
    Capabilities: [48] Vendor Specific Information: Len=08 <?>
    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
    Capabilities: [270] #19
    Capabilities: [2b0] Address Translation Service (ATS)
    Capabilities: [2c0] #13
    Capabilities: [2d0] #1b
    Kernel driver in use: fglrx_pci

It show 256MB which is unrealistic and too little (the GPU has 4GB total memory), because darktable works fine with OpenCL and requires 768MB at minimum.

Then there is clinfo (package clinfo), which gives the following:

Number of platforms:                 1
  Platform Profile:              FULL_PROFILE
  Platform Version:              OpenCL 1.2 AMD-APP (1411.4)
  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 cl_amd_hsa 


  Platform Name:                 AMD Accelerated Parallel Processing
Number of devices:               2
  Device Type:                   CL_DEVICE_TYPE_GPU
  Device ID:                     4098
  Board name:                    AMD Radeon R9 200 Series
  Device Topology:               PCI[ B#1, D#0, F#0 ]
  Max compute units:                 20
  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:           4
  Preferred vector width short:          2
  Preferred vector width int:            1
  Preferred vector width long:           1
  Preferred vector width float:          1
  Preferred vector width double:         1
  Native vector width char:          4
  Native vector width short:             2
  Native vector width int:           1
  Native vector width long:          1
  Native vector width float:             1
  Native vector width double:            1
  Max clock frequency:               1100Mhz
  Address bits:                  32
  Max memory allocation:             1073741824
  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:                    Read/Write
  Cache line size:               64
  Cache size:                    16384
  Global memory size:                3221225472
  Constant buffer size:              65536
  Max number of constant args:           8
  Local memory type:                 Scratchpad
  Local memory size:                 32768
  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 properties:              
    Out-of-Order:                No
    Profiling :                  Yes
  Platform ID:                   0x00007fce5d932500
  Name:                      Pitcairn
  Vendor:                    Advanced Micro Devices, Inc.
  Device OpenCL C version:           OpenCL C 1.2 
  Driver version:                1411.4 (VM)
  Profile:                   FULL_PROFILE
  Version:                   OpenCL 1.2 AMD-APP (1411.4)
  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_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_khr_image2d_from_buffer cl_khr_spir 


  Device Type:                   CL_DEVICE_TYPE_CPU
  Device ID:                     4098
  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:               2664Mhz
  Address bits:                  64
  Max memory allocation:             2147483648
  Image support:                 Yes
  Max number of images read arguments:       128
  Max number of images write arguments:      8
  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:                    32768
  Global memory size:                6258630656
  Constant buffer size:              65536
  Max number of constant args:           8
  Local memory type:                 Global
  Local memory size:                 32768
  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 properties:              
    Out-of-Order:                No
    Profiling :                  Yes
  Platform ID:                   0x00007fce5d932500
  Name:                      Intel(R) Core(TM)2 Duo CPU     E6750  @ 2.66GHz
  Vendor:                    GenuineIntel
  Device OpenCL C version:           OpenCL C 1.2 
  Driver version:                1411.4 (sse2)
  Profile:                   FULL_PROFILE
  Version:                   OpenCL 1.2 AMD-APP (1411.4)
  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_amd_svm 

There are some values with memory in their names, but which of them is the total available memory? In which unit? Global memory size is 512MB in bits and Max memory allocation is 256MB in bits. Local memory size could be 4GB in MB. clinfo has no manpage or builtin help with -h.

How to correctly interpret all these values to get the amount of available GPU memory? Are there any other programs that I can use?

Also: why is there no tag for OpenCL yet?

2

You probably have the answer by now, but the output from clinfo is in bytes and not bits. The global memory size is therefore around 3 GB, not 512 MB.

1

So you need a general purpose linux-world-wide utility for scripts etc to get this information? I'm afraid getting such specific information won't be easy. I'm not familiar with clinfo package, I presume you have to sudo apt-get install it.

Because if it doesn't have to be general, you could write a OpenCL application which would obtain this info. I believe OpenCL should have some method to give you such information, it's just a matter of writing simple application which initializes OpenCL context and printfs GPU_MEMORY (or sth similiar) to console.

As for OpenCL tag, I think you'd have more luck on https://stackoverflow.com/questions/tagged/opencl

  • It does not have to be a package, can also be a program I need to compile. You mention clinfo (I posted the output in the question above), so how to interpret it's output? – sebix Feb 27 '15 at 17:43
  • @sebix clinfo has source on github ie. github.com/Oblomov/clinfo/blob/master/src/clinfo.c , you can look for things like CL_DEVICE_GLOBAL_MEM_SIZE and see with which output string it is associated with. Then if you know that which OCL flag corresponds to your interest (size of GPU memory available for OCL) you could look for that, ie. clinfo | grep "Global memory size" . – Koshmaar Feb 27 '15 at 17:51
  • CL_DEVICE_GLOBAL_MEM_SIZE is - as also posted above in the question - 512MB, but this is not what I am searching for, see the explanation in my question. – sebix Feb 27 '15 at 18:17
  • Do you know if there is even this kind of date somewhere inside OpenCL? If that is not available under CL_DEVICE_* flags etc or any other way, then obviously clinfo won't give that to you. Also, if I understand correctly, you need the currently available (not theoretically possible) amount of memory on the GPU for CL? – Koshmaar Feb 27 '15 at 23:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.