All ps_separation_17_test2 wu's seem to be working about the right length of time, but at the end fail.

This is across 3 systems, from 4830,50,70 up to the 6950.

All ps_separation_17_test2 wu's seem to be working about the right length of time, but at the end fail.

Same here. They all fail on every machine.

Dito here,all test2 crashed.

still bad.

Same here. All the ps_separation_17_test2 workunits crunch to completion and come up with a computation error. Std output is here:

Task 60957635

Name ps_separation_17_test2_7565_0
Workunit 41546458
Created 28 Jun 2011 | 23:21:30 UTC
Sent 28 Jun 2011 | 23:48:17 UTC
Received 30 Jun 2011 | 0:56:15 UTC
Server state Over
Outcome Computation error
Client state Compute error
Exit status -1073740940 (0xffffffffc0000374)
Computer ID 257518
Report deadline 10 Jul 2011 | 23:48:17 UTC
Run time 994.17
CPU time 932.81
Validate state Invalid
Credit 0.00
Application version MilkyWay@Home
Anonymous platform (NVIDIA GPU)
Stderr output

<core_client_version>6.12.28</core_client_version>
<![CDATA[
<message>
- exit code -1073740940 (0xc0000374)
</message>
<stderr_txt>
Using SSE3 path
Found 1 platform(s)
Platform 0 information:
Name: NVIDIA CUDA
Version: OpenCL 1.0 CUDA 4.0.1
Vendor: NVIDIA Corporation
Extensions: cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll
Profile: FULL_PROFILE
Using device 1 on platform 0
Found 2 CL device(s)
Device GeForce GTX 460 (NVIDIA Corporation:0x10de) (CL_DEVICE_TYPE_GPU)
Driver version: 275.33
Version: OpenCL 1.0 CUDA
Compute capability: 2.1
Image support: CL_TRUE
Max compute units: 7
Clock frequency: 1520 Mhz
Global mem size: 1008402432
Local mem size: 49152
Max const buf size: 65536
Double extension: MW_CL_KHR_FP64

Compiler flags:
-cl-mad-enable -cl-no-signed-zeros -cl-strict-aliasing -cl-finite-math-only -cl-nv-verbose -DDOUBLEPREC=1 -DFAST_H_PROB=1 -DAUX_BG_PROFILE=0 -DNSTREAM=3 -DCONVOLVE=120 -DR0=5.019892539665562 -DSUN_R0=8.500000000000000 -DQ_INV_SQR=2.914444641769131 -DUSE_IMAGES=0

Build status: CL_BUILD_SUCCESS
Build log:

ptxas info : Compiling entry function 'mu_sum_kernel' for 'sm_21'
ptxas info : Function properties for mu_sum_kernel
384 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info : Used 52 registers, 8+0 bytes lmem, 80 bytes cmem[0], 22 bytes cmem[2], 20 bytes cmem[16]
Kernel work group info:
Work group size = 576
Kernel local mem size = 0
Compile work group size = { 0, 0, 0 }
Keeping chunk boundaries as multiples of 448
Range: { nu_steps = 640, mu_steps = 1600, r_steps = 1400 }
Iteration area: 2240000
Chunk estimate: 79
Num chunks: 79
Added area: 0
Effective area: 2240000
Integration time: 651.681885 s. Average time per iteration = 1018.252946 ms
Integral 0 time = 653.193827 s
Kernel work group info:
Work group size = 576
Kernel local mem size = 0
Compile work group size = { 0, 0, 0 }
Keeping chunk boundaries as multiples of 448
Range: { nu_steps = 640, mu_steps = 400, r_steps = 1400 }
Iteration area: 560000
Chunk estimate: 20
Num chunks: 20
Added area: 0
Effective area: 560000
Integration time: 169.322670 s. Average time per iteration = 264.566672 ms
Integral 1 time = 169.626810 s
Kernel work group info:
Work group size = 576
Kernel local mem size = 0
Compile work group size = { 0, 0, 0 }
Keeping chunk boundaries as multiples of 448
Range: { nu_steps = 640, mu_steps = 400, r_steps = 1400 }
Iteration area: 560000
Chunk estimate: 20
Num chunks: 20
Added area: 0
Effective area: 560000
Integration time: 164.092331 s. Average time per iteration = 256.394268 ms
21:39:16 (2424): No heartbeat from core client for 30 sec - exiting
Using SSE3 path
Found 1 platform(s)
Platform 0 information:
Name: NVIDIA CUDA
Version: OpenCL 1.0 CUDA 4.0.1
Vendor: NVIDIA Corporation
Extensions: cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll
Profile: FULL_PROFILE
Using device 1 on platform 0
Found 2 CL device(s)
Device GeForce GTX 460 (NVIDIA Corporation:0x10de) (CL_DEVICE_TYPE_GPU)
Driver version: 275.33
Version: OpenCL 1.0 CUDA
Compute capability: 2.1
Image support: CL_TRUE
Max compute units: 7
Clock frequency: 1520 Mhz
Global mem size: 1008402432
Local mem size: 49152
Max const buf size: 65536
Double extension: MW_CL_KHR_FP64

Compiler flags:
-cl-mad-enable -cl-no-signed-zeros -cl-strict-aliasing -cl-finite-math-only -cl-nv-verbose -DDOUBLEPREC=1 -DFAST_H_PROB=1 -DAUX_BG_PROFILE=0 -DNSTREAM=3 -DCONVOLVE=120 -DR0=5.019892539665562 -DSUN_R0=8.500000000000000 -DQ_INV_SQR=2.914444641769131 -DUSE_IMAGES=0

Build status: CL_BUILD_SUCCESS
Build log:


Kernel work group info:
Work group size = 576
Kernel local mem size = 0
Compile work group size = { 0, 0, 0 }
Keeping chunk boundaries as multiples of 448
Range: { nu_steps = 640, mu_steps = 400, r_steps = 1400 }
Iteration area: 560000
Chunk estimate: 20
Num chunks: 20
Added area: 0
Effective area: 560000
Integration time: 35.698868 s. Average time per iteration = 55.779482 ms
Integral 2 time = 35.983819 s
Likelihood time = 1.612008 s
<background_integral> 0.000755275827362 </background_integral>
<stream_integral> 57.795035441076493 283.231368773365200 1804.407722704481600 </stream_integral>
<background_likelihood> -3.376735412375381 </background_likelihood>
<stream_only_likelihood> -45.623959289950136 -3.684359922150092 -4.243788383397004 </stream_only_likelihood>
<search_likelihood> -2.981189150220247 </search_likelihood>

</stderr_txt>
]]>

---

I just abort all ps_separation_17_test... , if I happen to spot them.

It looks as they\we need an other application to make them work.