Hello i can not run Lunatics on my second machine if i load Lunatics i keep getting error on opencl Astropulse unit everything else works it is only happening to the opencl units

---

Stock Boinc working and doing opencl Astroplulse as stock set up on latest release

---

Sigh. I see three machines, and no way of telling which is the second. * Please use Host ID numbers * (that's for everyone, not just you)

I'm going to assume, in the absence of any real evidence, that you're talking about host 7326757 - the one with the NVIDIA GeForce GT 220.

You are getting the same error -

Error : Building Program (source, clBuildProgram):main kernels: not OK code -42
ptxas : error : Entry function 'GPU_fetch_array_kernel_twin_1D_cl' uses too much shared data (0x4034 bytes, 0x4000 max)

- as Ivailo Bonev reported in message 1557766 five days ago. (The tasks have been purged now, but I copied the error message to pass on to the developer).

We are putting most effort into eliminating the problem for these older cards ready for the release of the next-generation AP v7 application. It is probable that this problem, and one or two others like it, will delay the release of AP v7: in that case, we may be able to back-port the fix into a new AP v6 application to be delivered via a refreshed v0.42a installer, but that decision hasn't been confirmed yet.

In the meantime, please be aware that the problem isn't with the delivery system - the Lunatics installer - but with just one of the 22 different science applications it can deliver at your choice.

The quickest solution in your situation is to re-run the installer, but don't check the option for Astropulse (OpenCL) to run on your NVidia GPU. The multibeam (CUDA) application uses a different technology, and should run fine on the GT 220.

<core_client_version>7.2.42</core_client_version>
<![CDATA[
<message>
(unknown error) - exit code -42 (0xffffffd6)
</message>
<stderr_txt>
Running on device number: 0
DATA_CHUNK_UNROLL at default:2
DATA_CHUNK_UNROLL at default:2
Priority of worker thread raised successfully
Priority of process adjusted successfully, below normal priority class used
OpenCL platform detected: NVIDIA Corporation
BOINC assigns device 0
Info: BOINC provided device ID used
Used GPU device parameters are:
Number of compute units: 6
Single buffer allocation size: 128MB
max WG size: 512
FERMI path used: no

Build features: Non-graphics OpenCL USE_OPENCL_NV COMBINED_DECHIRP_KERNEL FFTW USE_INCREASED_PRECISION USE_SSE2 x86
CPUID: Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz

Cache: L1=64K L2=2048K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
AstroPulse v.6
Non-graphics FFTW USE_CONVERSION_OPT
Windows x86 rev 1316, V6 match, by Raistmer with support of Lunatics.kwsn.net team. SSE2

OpenCL version by Raistmer

ffa threshold mods by Joe Segur
SSE3 dechirping by JDWhale
Combined dechirp kernel by Frizz
Number of OpenCL platforms: 1


OpenCL Platform Name: NVIDIA CUDA
Number of devices: 1
Max compute units: 6
Max work group size: 512
Max clock frequency: 1480Mhz
Max memory allocation: 134217728
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 9
Local memory type: Scratchpad
Local memory size: 16384
Queue properties:
Out-of-Order: Yes
Name: GeForce GT 220
Vendor: NVIDIA Corporation
Driver version: 337.88
Version: OpenCL 1.0 CUDA
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 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics


state.fold_buf_size_short=65536; state.fold_buf_size_long=262144
INFO: can't open binary kernel file: C:\Bionic Data/projects/setiathome.berkeley.edu\AstroPulse_Kernels_r1316.cl_GeForceGT220.bin_V6, continue with recompile...
Info : Building Program (source, clBuildProgram):main kernels: OK code 0


INFO: binary kernel file created
WARNING: can't open binary kernel file for oclFFT plan: C:\Bionic Data/projects/setiathome.berkeley.edu\oclFFTplan_GeForceGT220_32768_r1316.bin, continue with recompile...
Termination request detected. GPU device synched, awaiting termination...
Not using ap_cmdline.txt-file, using commandline options.
Running on device number: 0
Priority of worker thread raised successfully
Priority of process adjusted successfully, below normal priority class used
OpenCL platform detected: NVIDIA Corporation
BOINC assigns device 0
Info: BOINC provided OpenCL device ID used
Used GPU device parameters are:
Number of compute units: 6
Single buffer allocation size: 128MB
max WG size: 512
FERMI path used: no

Build features: Non-graphics OpenCL USE_OPENCL_NV TWIN_FFA OCL_ZERO_COPY COMBINED_DECHIRP_KERNEL FFTW USE_INCREASED_PRECISION USE_SSE2 x86
CPUID: Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz

Cache: L1=64K L2=2048K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3 SSSE3 SSE4.1
### Restart at 0.00 percent.
state.fold_buf_size_short=65536; state.fold_buf_size_long=262144
INFO: can't open binary kernel file: C:\Bionic Data/projects/setiathome.berkeley.edu\AstroPulse_Kernels_r2399.cl_GeForceGT220.bin_V6_TWIN_FFA_33788, continue with recompile...
Error : Building Program (source, clBuildProgram):main kernels: not OK code -42
ptxas : error : Entry function 'GPU_fetch_array_kernel_twin_1D_cl' uses too much shared data (0x4034 bytes, 0x4000 max)


</stderr_txt>
]]>

this is the sterrr file for 1 of the units that have failed

please help

---

core_client_version>7.2.42</core_client_version>
<![CDATA[
<message>
(unknown error) - exit code -42 (0xffffffd6)
</message>
<stderr_txt>
Running on device number: 0
Priority of worker thread raised successfully
Priority of process adjusted successfully, below normal priority class used
OpenCL platform detected: NVIDIA Corporation
BOINC assigns device 0
Info: BOINC provided OpenCL device ID used
Used GPU device parameters are:
Number of compute units: 6
Single buffer allocation size: 128MB
max WG size: 512
FERMI path used: no

Build features: Non-graphics OpenCL USE_OPENCL_NV TWIN_FFA OCL_ZERO_COPY COMBINED_DECHIRP_KERNEL FFTW USE_INCREASED_PRECISION USE_SSE2 x86
CPUID: Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz

Cache: L1=64K L2=2048K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3 SSSE3 SSE4.1
AstroPulse v6 Windows x86 rev 2399, V6 match, by Raistmer with support of Lunatics.kwsn.net team. SSE2

OpenCL version by Raistmer

ffa threshold mods by Joe Segur
SSE3 dechirping by JDWhale
Combined dechirp kernel by Frizz
Number of OpenCL platforms: 1


OpenCL Platform Name: NVIDIA CUDA
Number of devices: 1
Max compute units: 6
Max work group size: 512
Max clock frequency: 1480Mhz
Max memory allocation: 134217728
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 9
Local memory type: Scratchpad
Local memory size: 16384
Queue properties:
Out-of-Order: Yes
Name: GeForce GT 220
Vendor: NVIDIA Corporation
Driver version: 337.88
Version: OpenCL 1.0 CUDA
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 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics


state.fold_buf_size_short=65536; state.fold_buf_size_long=262144
INFO: can't open binary kernel file: C:\Bionic Data/projects/setiathome.berkeley.edu\AstroPulse_Kernels_r2399.cl_GeForceGT220.bin_V6_TWIN_FFA_33788, continue with recompile...
Error : Building Program (source, clBuildProgram):main kernels: not OK code -42
ptxas : error : Entry function 'GPU_fetch_array_kernel_twin_1D_cl' uses too much shared data (0x4034 bytes, 0x4000 max)


</stderr_txt>
]]>

this is a second unit also computaion error

---

<core_client_version>7.2.42</core_client_version>
<![CDATA[
<message>
(unknown error) - exit code -42 (0xffffffd6)
</message>
<stderr_txt>
Running on device number: 0
DATA_CHUNK_UNROLL at default:2
DATA_CHUNK_UNROLL at default:2
Priority of worker thread raised successfully
Priority of process adjusted successfully, below normal priority class used
OpenCL platform detected: NVIDIA Corporation
BOINC assigns device 0
Info: BOINC provided device ID used
Used GPU device parameters are:
Number of compute units: 6
Single buffer allocation size: 128MB
max WG size: 512
FERMI path used: no

Build features: Non-graphics OpenCL USE_OPENCL_NV COMBINED_DECHIRP_KERNEL FFTW USE_INCREASED_PRECISION USE_SSE2 x86
CPUID: Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz

Cache: L1=64K L2=2048K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
AstroPulse v.6
Non-graphics FFTW USE_CONVERSION_OPT
Windows x86 rev 1316, V6 match, by Raistmer with support of Lunatics.kwsn.net team. SSE2

OpenCL version by Raistmer

ffa threshold mods by Joe Segur
SSE3 dechirping by JDWhale
Combined dechirp kernel by Frizz
Number of OpenCL platforms: 1


OpenCL Platform Name: NVIDIA CUDA
Number of devices: 1
Max compute units: 6
Max work group size: 512
Max clock frequency: 1480Mhz
Max memory allocation: 134217728
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 9
Local memory type: Scratchpad
Local memory size: 16384
Queue properties:
Out-of-Order: Yes
Name: GeForce GT 220
Vendor: NVIDIA Corporation
Driver version: 337.88
Version: OpenCL 1.0 CUDA
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 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics


state.fold_buf_size_short=65536; state.fold_buf_size_long=262144
INFO: can't open binary kernel file: C:\Bionic Data/projects/setiathome.berkeley.edu\AstroPulse_Kernels_r1316.cl_GeForceGT220.bin_V6, continue with recompile...
Info : Building Program (source, clBuildProgram):main kernels: OK code 0


INFO: binary kernel file created
WARNING: can't open binary kernel file for oclFFT plan: C:\Bionic Data/projects/setiathome.berkeley.edu\oclFFTplan_GeForceGT220_32768_r1316.bin, continue with recompile...
Termination request detected. GPU device synched, awaiting termination...
Not using ap_cmdline.txt-file, using commandline options.
Running on device number: 0
Priority of worker thread raised successfully
Priority of process adjusted successfully, below normal priority class used
OpenCL platform detected: NVIDIA Corporation
BOINC assigns device 0
Info: BOINC provided OpenCL device ID used
Used GPU device parameters are:
Number of compute units: 6
Single buffer allocation size: 128MB
max WG size: 512
FERMI path used: no

Build features: Non-graphics OpenCL USE_OPENCL_NV TWIN_FFA OCL_ZERO_COPY COMBINED_DECHIRP_KERNEL FFTW USE_INCREASED_PRECISION USE_SSE2 x86
CPUID: Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz

Cache: L1=64K L2=2048K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3 SSSE3 SSE4.1
### Restart at 0.00 percent.
state.fold_buf_size_short=65536; state.fold_buf_size_long=262144
INFO: can't open binary kernel file: C:\Bionic Data/projects/setiathome.berkeley.edu\AstroPulse_Kernels_r2399.cl_GeForceGT220.bin_V6_TWIN_FFA_33788, continue with recompile...
Error : Building Program (source, clBuildProgram):main kernels: not OK code -42
ptxas : error : Entry function 'GPU_fetch_array_kernel_twin_1D_cl' uses too much shared data (0x4034 bytes, 0x4000 max)


</stderr_txt>
]]>

this is the sterrr file for 1 of the units that have failed

please help

"Windows x86 rev 1316" wasn't delivered by the Lunatics v0.42 installer - we're up to r2399 now - but you have confirmed my guess at the GeForce GT 220, so my previous answer applies.

OK Thank you Richard so the new version 0.42 has a bug or 2 with older cards .ok

I spose it mite be better to just rollback to the older version of both Boinc and Lunatics until i can get a better card in a few months . Thanks again Richard

---

OK Thank you Richard so the new version 0.42 has a bug or 2 with older cards .ok

I spose it mite be better to just rollback to the older version of both Boinc and Lunatics until i can get a better card in a few months . Thanks again Richard

Neither of the two known issues with the Lunatics v0.42 installer has anything to do with the version of BOINC in use - no point in switching that back and forth.

Since you've given this thread a generic title implying the whole of the installer has a problem, it'll probably attract quite a wide readership. Let's use it constructively to itemise the two specific known problems that might catch people out.

1) The Astropulse/OpenCL application (r2399) for NVIDIA cards can fail - we think, specifically on older cards - with an error message like

Error : Building Program (source, clBuildProgram):main kernels: not OK code -42
ptxas : error : Entry function 'GPU_fetch_array_kernel_twin_1D_cl' uses too much shared data (0x4034 bytes, 0x4000 max)

at the very end of stderr_txt.

2) The Astropulse/OpenCL application (r2399) for ATI cards can fail on older cards like the HD5470/HD6370m with 'Max work group size: 128'

"C:\Users\Temp\OCL2536T5.cl", line 1370: error: work group size exceeds the
          maximum default value for the selected device

Again, the solution for this problem is being targeted at the AP v7 (installer v0.43) release, but if we need to deploy a v0.42a interim release installer, this problem will be addressed too.

Thanks again Richard i was a bit too quick . I've rolled back to the previous version of Boinc to ver 7.0.64 and Lunatics back to ver 0.41 and all is now working fine opencl for Astropulse is now working .

That's what i get for being to quick Richard oh well all sweet now but thanks again

---