AMD VLIW4 GPUs and fp64

Discussion in 'GPGPU Technology & Programming' started by codedivine, Jul 26, 2013.

  1. Dade

    Newcomer

    Joined:
    Dec 20, 2009
    Messages:
    206
    Likes Received:
    20
    HD5800 chips my have the support full double precision but they dont' have the support OpenCL compliant double precision. For instance, some operation as different rounding mode, etc. (this is quite deadly for most scientific applications and it makes cl_amd_fp64 quite useless in that context).

    From cl_amd_fp64 extensions overview:

     
  2. codedivine

    Regular

    Joined:
    Jan 22, 2009
    Messages:
    271
    Likes Received:
    0
    They do. I have a 5850 and I can confirm that the device supports cl_khr_fp64 under OpenCL.
     
  3. OpenGL guy

    Veteran

    Joined:
    Feb 6, 2002
    Messages:
    2,357
    Likes Received:
    28
    You're right, we still do expose the cl_khr_fp64 extension. The difference is that you don't have to enable the extension before using doubles in the kernel.

    AFAIK, we haven't changed our double precision support for HD6900 boards in OpenCL. Unfortunately, I don't have a board to check.
     
  4. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    14,815
    Likes Received:
    2,231
    I have a 6900 board I can check
     
  5. codedivine

    Regular

    Joined:
    Jan 22, 2009
    Messages:
    271
    Likes Received:
    0
    Can you run the tool "clinfo" on the command line? It comes bundled with the AMD drivers so you it should already be installed. It will give a long output including a list of OpenCL extensions supported by each device. It will be great if you can simply paste the output.

    If you want, you can do something like: clinfo > results.txt and copy/paste from results.txt.
     
  6. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    14,815
    Likes Received:
    2,231
    will do
    meanwhile i found fah opencl benchmark
    [​IMG]
    ps: gtx 680 scores
    4.99 explicit
    6.37 explicit

    ps: what folder do I find clinfo (edit: syswow64)
     
  7. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    14,815
    Likes Received:
    2,231
    Number of platforms: 1
    Platform Profile: FULL_PROFILE
    Platform Version: OpenCL 1.2 AMD-APP (1124.2)
    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_khr_d3d10_sharing cl_khr_d3d11_sharing


    Platform Name: AMD Accelerated Parallel Processing
    Number of devices: 2
    Device Type: CL_DEVICE_TYPE_GPU
    Device ID: 4098
    Board name: AMD Radeon HD 6900 Series
    Max compute units: 22
    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: 800Mhz
    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: 2147483648
    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: 666FB864
    Name: Cayman
    Vendor: Advanced Micro Devices, Inc.
    Device OpenCL C version: OpenCL C 1.2
    Driver version: 1124.2 (VM)
    Profile: FULL_PROFILE
    Version: OpenCL 1.2 AMD-APP (1124.2)
    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_khr_d3d10_sharing


    Device Type: CL_DEVICE_TYPE_CPU
    Device ID: 4098
    Board name:
    Max compute units: 4
    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: 2404Mhz
    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: 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: 2147483648
    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: 425
    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: 666FB864
    Name: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz
    Vendor: GenuineIntel
    Device OpenCL C version: OpenCL C 1.2
    Driver version: 1124.2 (sse2)
    Profile: FULL_PROFILE
    Version: OpenCL 1.2 AMD-APP (1124.2)
    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_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_d3d10_sharing
     
  8. codedivine

    Regular

    Joined:
    Jan 22, 2009
    Messages:
    271
    Likes Received:
    0
    clinfo should be in C:\windows\system32. On most systems, if you just do "clinfo" on cmd (from any folder) it works as clinfo should be on the system PATH.
     
  9. codedivine

    Regular

    Joined:
    Jan 22, 2009
    Messages:
    271
    Likes Received:
    0
    Great. Thanks. Confirms that cl_khr_fp64 is supported.
     
  10. Davros

    Legend

    Joined:
    Jun 7, 2004
    Messages:
    14,815
    Likes Received:
    2,231
    question
    cpu Address bits: 32 ???
    its a 64bit cpu
     
  11. darkblu

    Veteran

    Joined:
    Feb 7, 2002
    Messages:
    2,642
    Likes Received:
    22
    Apropos, Devastator (Trinity) seems to be doing khr_fp64 just fine nowadays:

    edit: Bah. Seems that output is form the early drivers, and khr_fp64 was subsequently removed. My bad.
    Code:
    Number of platforms:				 1
      Platform Profile:				 FULL_PROFILE
      Platform Version:				 OpenCL 1.1 AMD-APP (873.1)
      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_khr_d3d10_sharing
    
    
      Platform Name:				 AMD Accelerated Parallel Processing
    Number of devices:				 2
      Device Type:					 CL_DEVICE_TYPE_GPU
      Device ID:					 4098
      Board name:					 AMD Radeon HD 7660G
      Max compute units:				 6
      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:				 685Mhz
      Address bits:					 32
      Max memory allocation:			 200540160
      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:			 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:				 536870912
      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:		 1
      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:					 000007FEE7CB7AA0
      Name:						 Devastator
      Vendor:					 Advanced Micro Devices, Inc.
      Device OpenCL C version:			 OpenCL C 1.1 
      Driver version:				 CAL 1.4.1683 (VM)
      Profile:					 FULL_PROFILE
      Version:					 OpenCL 1.1 AMD-APP (873.1)
      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_popcnt cl_khr_d3d10_sharing cl_amd_meminfo
    
    In case somebody feel adventurous, here's a CLI tool which will report the double caps as well (AMD's clinfo currently does not). Supplied build script is for linux and osx, but building it for windows should be trivial. To build it you'll need CL_VERSION_1_2 headers.
     
    #31 darkblu, Aug 21, 2013
    Last edited by a moderator: Aug 21, 2013
Loading...

Share This Page

  • About Us

    Beyond3D has been around for over a decade and prides itself on being the best place on the web for in-depth, technically-driven discussion and analysis of 3D graphics hardware. If you love pixels and transistors, you've come to the right place!

    Beyond3D is proudly published by GPU Tools Ltd.
Loading...