Message boards :
Number crunching :
Half of my AMD-GPU work units end in an error. Help?
Message board moderation
Author | Message |
---|---|
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
Hi Seti@Home folks, Just got SETI running on my new XFX R9 390 graphics card. Everything seemed fine until I looked at the results page and saw 50% of my tasks are ending with the "Error while computing" status. http://setiathome.berkeley.edu/results.php?userid=9584707 Perhaps the problem is my system is Win7 64-bit, yet the task log says x86? Or perhaps its something with my graphics card? Here is the output for a clinfo.exe: C:\Users\PC\Downloads\clinfo>clinfo.exe Number of platforms: 1 Platform Profile: FULL_PROFILE Platform Version: OpenCL 2.0 AMD-APP (2079.5) Platform Name: AMD Accelerated Parallel Proces sing Platform Vendor: Advanced Micro Devices, Inc. Platform Extensions: cl_khr_icd cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_amd_event_callback cl_amd_offl ine_devices Platform Name: AMD Accelerated Parallel Proces sing Number of devices: 2 Device Type: CL_DEVICE_TYPE_GPU Device ID: 4098 Max compute units: 40 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 Max clock frequency: 1015Mhz Address bits: 32 Max memory allocation: 3221225472 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 Error correction support: 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: 616CE698 Name: Hawaii Vendor: Advanced Micro Devices, Inc. Driver version: 2079.5 (VM) Profile: FULL_PROFILE Version: OpenCL 1.2 AMD-APP (2079.5) Extensions: cl_khr_fp64 cl_amd_fp64 cl_khr_ global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int3 2_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_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 cl_khr_d3d11_sha ring cl_khr_dx9_media_sharing cl_khr_image2d_from_buffer cl_khr_spir cl_khr_gl_e vent Device Type: CL_DEVICE_TYPE_CPU Device ID: 4098 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: 8 Preferred vector width double: 4 Max clock frequency: 3499Mhz Address bits: 32 Max memory allocation: 1073741824 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: 32768 Global memory size: 2147483648 Constant buffer size: 65536 Max number of constant args: 8 Local memory type: Global Local memory size: 32768 Error correction support: 0 Profiling timer resolution: 292 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: 616CE698 Name: Intel(R) Core(TM) i5-4690K CPU @ 3.50GHz Vendor: GenuineIntel Driver version: 2079.5 (sse2,avx) Profile: FULL_PROFILE Version: OpenCL 1.2 AMD-APP (2079.5) Extensions: cl_khr_fp64 cl_amd_fp64 cl_khr_ global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int3 2_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_at tribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_am d_popcnt cl_khr_d3d10_sharing cl_khr_spir cl_khr_gl_event C:\Users\PC\Downloads\clinfo>pause Press any key to continue . . . Any help? Thank you in advance Edit: Here's a verbose output of the SETI@HOME log: http://pastebin.com/wW8kPux6 I noticed there's a lot of "Access is denied" errors. This may be my firewall asking me to give permission for the program to contact the servers. I tried whitelisting the executable, but it seems to keep asking. |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
ok, in the prereferences you can set that, that BOINC wont ask again to connect. Use the freeware CPU-Z to look about your mainboards name and bios version. Which power supply do you use ? please post the infos btw you got a haswell cpu, no wonder to me, that this ultra new gpu wont work properly ;) ok ,no answers...here some tips, if you won´t upgrade your mainboard bios, you can set you PCIe bus speed to GEN2 and test again. |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
ok, in the prereferences you can set that, that BOINC wont ask again to connect. I thought I did that. I just respecified it, so hopefully that helps. CPU-Z: http://pastebin.com/ZTYx9TTM PSU: 850W |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
there is a solution in my post before, read it, the only chance, bios update not needed it´s allready up to date. |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
I'm also interested in hearing if there's any optimizations I can run for the R9 390. |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
there is a solution in my post before, read it, the only chance, bios update not needed it´s allready up to date. Motherboard BIOS is up to date. No problem there. Any other ideas? |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
before you download new workunits, you could install "Lunarics 0.44" 64-bit version. And choose optimized apps for the BOINC client to be installed. BEWARE your old workunits will be moved to another folder, can´t promise you get it back. you can set you PCIe bus speed from AUTO to GEN2 and test again. |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
before you download new workunits, you could install "Lunarics 0.44" 64-bit version. And choose optimized apps for the BOINC client to be installed. I really don't think the Haswell CPU is the problem. It's only a few years old. The GPU is only a year old. They should work together, right? As for Lunatics 0.44, I really don't want to install third party programs yet. I don't know this guy. I'd rather stay with the official BOINC client. |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
They should work together, right? no, there u wrong. Some people told me too ,Intel plus AMD gpu isn´t a good combination, but can work. But that´s not the reason , put the PCIe bus speed on GEN2. And watch if ur happy with it. |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
any results on that "Gen2" configuration of the mainboard bios entries ? https://boinc.berkeley.edu/trac/wiki/ClientAppConfig This maybe interesting for u, if the gpu does it u can run 2-3 workunits in the same time. |
betreger Send message Joined: 29 Jun 99 Posts: 11361 Credit: 29,581,041 RAC: 66 |
I don't know this guy AFAIK Lunatics do much of the code for the stock apps |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
any results on that "Gen2" configuration of the mainboard bios entries ? I changed the setting but it will take 2-3 days before new WU pass/fail validation. I will let you know. |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
cheers, see ya, happy crunshing |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
saw ur results this morning, looking very very good, in order to raise your income in points, u´ll config the app_config.xml file in your "ProgramDATA/BOINC/projects/setiathome.berkeley.edu" directory. <app_config> <app> <name>setiathome_v8</name> <gpu_versions> <gpu_usage>.33</gpu_usage> <cpu_usage>.25</cpu_usage> </gpu_versions> </app> </app_config> This example will work with 3 workunits in a time. |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
Thank you for your help! Do I need additional code for Astropulse? Also, I don't need a "max_concurrant" code? |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
AP are very very RARE to get, possibly its already over to crunsh them. The max_concurrant will be need only when less than expected workunits are running. For example : theoreticly running 24 workunits (including cpu) but there running only 12 at once. ok you got 4 for cpu and 3 for gpu, means max_concurrant are at minimum 7, to got save you can use 10. Edit:Sorry, thought u got a i7 cpu. |
rob smith Send message Joined: 7 Mar 03 Posts: 22190 Credit: 416,307,556 RAC: 380 |
AstroPulse are not only very rare, but are processed (currently) using a separate application. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
saw ur results this morning, looking very very good, I switched from .33 per GPU to .5 because I am trying to run a different program on my CPU, and each WU was eating -10% CPU, so with 3 SETI running it was -30% hit to my CPU. At 2 SETI running (gpu usage .50 - cpu usage .25) its only -20% to CPU, which is ok. We'll see how the results look over the next 24 hours. Thank you for your help. |
The_Matrix Send message Joined: 17 Nov 03 Posts: 414 Credit: 5,827,850 RAC: 0 |
ok, hope you will have good start then. Ask if you need, this forum got plenty of space to write. |
S\ Send message Joined: 18 Oct 11 Posts: 28 Credit: 9,093,005 RAC: 0 |
ok, hope you will have good start then. Ask if you need, this forum got plenty of space to write. After adding the multiple GPU xml, I'm getting "Triple Data Corrupt" messages in the GUI. http://pastebin.com/xa4gnSwt |
©2024 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.