I have been experimenting with the stream SDK on a machine running Ubuntu 10.04 desktop and with 2 H5970 cards installed. I am unable to get FindNumDevices to return more than 1 or CLInfo to return more than 2 devices (one CPU and 1 GPU). I am beginning to wonder if the combination of OS + ATI Driver + SDK does not support 2 GPU. Here are the specific details:
OS - Ubuntu Desktop 10.04 64-bit
SDK Package - ati-stream-sdk-v2.1-lnx64.tgz
Catalyst Driver Package - ati-driver-installer-10-6-x86.x86_64.run
Monitor(s) - 1
I believe I am following the Catalyst driver installation properly mostly due to the fact that the Catalyst Control Center appears in the X menu after installation. Building the SDK at this point is simple, just cd-ing into the SDK directory and typing 'make' will build without error. The next step is where I think I might be fouling up. I know that I have to use the aticonfig command after installation of the driver. I see conflicting reports on this forum on the exact syntax to use:
aticonfig --initial
aticonfig --adapter=all initial
aticonfig --adapter=all initial -f
The first version of aticonfig doesn't seem to help me much. When I move over to the SDK samples, this is when FindNumDevices reports a single device and CLInfo reports 2 devices, the CPU and a single GPU device. Although technically this is a working state, it doesn't take full advantage of the hardware. This is why I researched finding the other syntax.
The second and third options seem to actually corrupt X.org because when I reboot, the display is screwy and I am unable to do anything (save for SSH login where the samples report no devices). I even use the environment variable change EXPORT DISPLAY=:0 but this doesn't help either. I can restore the display to the last know working state by rebooting the box with the Ubuntu installer disc and copying a previous version of the xorg.conf back over top of the modified one.
Here is some output snippets of the sample programs, first FindNumDevices:
user@ubuntu:/usr/src/ati-stream-sdk-v2.1-lnx64/samples/cal/bin/x86_64$ sudo ./FindNumDevices [sudo] password for user: Supported CAL Runtime Version: 1.3.185 Found CAL Runtime Version: 1.4.636 Use -? for help CAL initialized. Finding out number of devices :- Device Count = 1 CAL shutdown successful. Press enter to exit...
and now CLInfo:
root@ubuntu:/usr/src/ati-stream-sdk-v2.1-lnx64/samples/opencl/bin/x86_64# ./CLInfo Number of platforms: 1 Platform Profile: FULL_PROFILE Platform Version: OpenCL 1.0 ATI-Stream-v2.1 (145) Platform Name: ATI Stream Platform Vendor: Advanced Micro Devices, Inc. Platform Extensions: cl_khr_icd Platform Name: ATI Stream Number of devices: 2 Device Type: CL_DEVICE_TYPE_CPU Device ID: 4098 Max compute units: 12 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: 0 Max clock frequency: 1600Mhz Address bits: 64 Max memory allocation: 1073741824 Image support: No 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: No Round to +ve and infinity: No IEEE754-2008 fused multiply-add: No Cache type: Read/Write Cache line size: 0 Cache size: 0 Global memory size: 3221225472 Constant buffer size: 65536 Max number of constant args: 8 Local memory type: Global Local memory size: 32768 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: 0x7f8e76df8228 Name: Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz Vendor: GenuineIntel Driver version: 1.1 Profile: FULL_PROFILE Version: OpenCL 1.0 ATI-Stream-v2.1 (145) Extensions: cl_khr_icd 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_byte_addressable_store cl_khr_gl_sharing cl_ext_device_fission cl_amd_device_attribute_query cl_amd_printf Device Type: CL_DEVICE_TYPE_GPU Device ID: 4098 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: 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: 0 Max clock frequency: 725Mhz Address bits: 32 Max memory allocation: 268435456 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: 32768 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: No Round to +ve and infinity: No IEEE754-2008 fused multiply-add: No Cache type: None Cache line size: 0 Cache size: 0 Global memory size: 268435456 Constant buffer size: 65536 Max number of constant args: 8 Local memory type: Scratchpad Local memory size: 32768 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: 0x7f8e76df8228 Name: Cypress Vendor: Advanced Micro Devices, Inc. Driver version: CAL 1.4.696 Profile: FULL_PROFILE Version: OpenCL 1.0 ATI-Stream-v2.1 (145) Extensions: cl_khr_icd 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_byte_addressable_store cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_media_ops
and finally aticonfig --lsa
root@ubuntu:/usr/src/ati-stream-sdk-v2.1-lnx64/samples/opencl/bin/x86_64# aticonfig --lsa * 0. 05:00.0 ATI Radeon HD 5900 Series 1. 06:00.0 ATI Radeon HD 5900 Series 2. 09:00.0 ATI Radeon HD 5900 Series 3. 0a:00.0 ATI Radeon HD 5900 Series
* - Default adapter
Does anyone else out there run a similar setup? I would imagine this setup is possible but perhaps I am wrong. Any suggestions? Downgrade Ubuntu?
- MM
