The Highest Ranked SETI AMD Host is a MAC: Time for a STOCK MAC APP?
Message boards :
Number crunching :
The Highest Ranked SETI AMD Host is a MAC: Time for a STOCK MAC APP?
Message board moderation
Previous · 1 . . . 14 · 15 · 16 · 17 · 18 · 19 · 20 . . . 21 · Next
| Author | Message |
|---|---|
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 
 |
I think there may be some other problem. I tested Urs app r2839 in my new 10.8.5 and it works without the Error I've been receiving. I tested the last Joe Fox App and it failed with the known 'Symbol not found: ___sincos_stret' error. All of mine have been crashing with; ERROR: OpenCL kernel/call 'Enqueueing kernel: (strip) PC_find_spike32_kernel_cl' call failed (-54) in file analyzeFuncs.cpp near line 3171. Mine have also been printing the lines; period_iterations_num=23 end FindSpikes_ev==NULL end FindSpikes_ev==NULL Sometimes a build will not print the FindSpikes lines but most of them do. I had the same Error on the one build from Lion, and receive the same error in the old Mountain Lion. The builds don't have the error in Yosemite. I'm just wondering if something has been changed since around 2839 in analyzeFuncs.cpp that can be causing the recent builds to crash and print the FindSpikes lines. As for the old Mountain Lion system, it appears to have fftw 3.2.2 misplaced in the system somewhere and will probably need a system restore to clear it. Obviously reinstalling 3.3.4 again and again is not fixing it. The system restore will remove all the third party items from usr. I can even stop short of installing the last security update on that system as well. Right now I keep getting builds on the new Mountain Lion that either work with the shorties and fail with the longer ones or visa versa. The old Mountain Lion didn't have that problem but the builds were only 2.1 mb in size whereas they are 3.6 mb on the new system. ID: 1647398 ·  |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
Hmmmm, the last build seems to work on both tasks, although it is a little slower than previous builds. Seems the more configure options you add the slower the app runs. I have seen this task finish in 19 mins with a CPU time of 400 secs, but it had a few too many spikes and ACs. Notice this build doesn't have the 'end FindSpikes_ev==NULL' lines. This task is from here, http://setiathome.berkeley.edu/result.php?resultid=3999222858 15:10:02 (592): Can't open init data file - running in standalone mode
Not using mb_cmdline.txt-file, using commandline options.
15:10:02 (592): Can't open init data file - running in standalone mode
WARNING: init_data.xml missing
OpenCL platform detected: Apple
WARNING: BOINC supplied wrong platform!
Number of OpenCL devices found : 3
BOINC assigns slot on device #2.
WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities
Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx OCL_CHIRP3 ASYNC_SPIKE FFTW SSSE3 64bit
System: Darwin x86_64 Kernel: 14.1.0
CPU : Intel(R) Xeon(R) CPU E5462 @ 2.80GHz
GenuineIntel x86, Family 6 Model 23 Stepping 6
Features : FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 DTES64 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1
OpenCL-kernels filename : MultiBeam_Kernels_r2855.cl
ar=0.421468 NumCfft=197331 NumGauss=1121461076 NumPulse=226461018600 NumTriplet=452853385294
Currently allocated 145 MB for GPU buffers
In v_BaseLineSmooth: NumDataPoints=1048576, BoxCarLength=8192, NumPointsInChunk=32768
OS X optimized S@H Enhanced application by Alex Kan
OpenCL version by Raistmer, r2855
AMD HD5 version by Raistmer
Number of OpenCL platforms: 1
OpenCL Platform Name: Apple
Number of devices: 3
Max compute units: 14
Max work group size: 256
Max clock frequency: 775Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts PRO Prototype
Vendor: AMD
Driver version: 1.2 (Dec 23 2014 00:17:18)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Max compute units: 14
Max work group size: 256
Max clock frequency: 775Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts PRO Prototype
Vendor: AMD
Driver version: 1.2 (Dec 23 2014 00:17:18)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Max compute units: 14
Max work group size: 256
Max clock frequency: 900Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts XT Prototype
Vendor: AMD
Driver version: 1.2 (Dec 23 2014 00:17:18)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Work Unit Info:
...............
Credit multiplier is : 2.85
WU true angle range is : 0.421468
Used GPU device parameters are:
Number of compute units: 14
Single buffer allocation size: 64MB
Total device global memory: 1024MB
max WG size: 256
local mem type: Real
period_iterations_num=23
Autocorr: peak=18.11777, time=33.55, delay=5.9239, d_freq=1420869629.27, chirp=14.563, fft_len=128k
Spike: peak=24.11937, time=46.98, d_freq=1420870010.01, chirp=15.511, fft_len=128k
Triplet: peak=10.37436, time=33.14, period=1.979, d_freq=1420871016.67, chirp=-27.424, fft_len=256
Triplet: peak=11.31312, time=33.14, period=1.979, d_freq=1420871018.6, chirp=-29.668, fft_len=256
Triplet: peak=11.20124, time=33.14, period=1.979, d_freq=1420871015.44, chirp=-30.915, fft_len=256
Triplet: peak=11.01234, time=33.14, period=1.979, d_freq=1420871020.54, chirp=-31.912, fft_len=256
Triplet: peak=10.54942, time=33.14, period=1.979, d_freq=1420871017.4, chirp=-33.158, fft_len=256
Best spike: peak=24.11937, time=46.98, d_freq=1420870010.01, chirp=15.511, fft_len=128k
Best autocorr: peak=18.11777, time=33.55, delay=5.9239, d_freq=1420869629.27, chirp=14.563, fft_len=128k
Best gaussian: peak=3.175113, mean=0.5315582, ChiSq=1.363123, time=88.08, d_freq=1420873307.08,
score=-0.5016232, null_hyp=2.196321, chirp=-82.483, fft_len=16k
Best pulse: peak=2.437596, time=19.12, period=0.7034, d_freq=1420867328.53, score=0.9697, chirp=36.897, fft_len=256
Best triplet: peak=11.31312, time=33.14, period=1.979, d_freq=1420871018.6, chirp=-29.668, fft_len=256
Flopcounter: 20389834125122.226562
Spike count: 1
Autocorr count: 1
Pulse count: 0
Triplet count: 5
Gaussian count: 0
Time cpu in use since last restart: 494.3 seconds
GPU device sync requested... ...GPU device synched
15:34:17 (592): called boinc_finish(0)The shortie seems to have been removed already, but it matches the results from 2839. I suppose I'll need to grab a newer shortie. Anyone have a fix for the KWSN-OSX-bench-MB_v2.1.07 benchmark tool failing in Yosemite? I keep getting; KWSN-Darwin-MBbench v2.1.07 Running on TomsMacPro.local at Fri Feb 27 20:52:15 2015 --------------------------------------------------- Starting benchmark run... --------------------------------------------------- Listing wu-file(s) in /testWUs : PG0009_v7.wu Listing executable(s) in /APPS : MBv7_7.05r2855_ssse3_clGPU_x86_64-apple-darwin Listing executable in /REF_APPs : MBv7_7.05r2839_sse3_clATI_HD5_OSX64 --------------------------------------------------- Current WU: PG0009_v7.wu --------------------------------------------------- Skipping default app MBv7_7.05r2839_sse3_clATI_HD5_OSX64, displaying saved result(s) Elapsed Time: ………………………………… 0 seconds --------------------------------------------------- Running app with command : MBv7_7.05r2855_ssse3_clGPU_x86_64-apple-darwin 0.59 real 0.46 user 0.06 sys ./benchmark: line 145: 0 / 0: division by 0 (error token is "0") 2015-02-27 15:52:16.892 system_profiler[1015:14113] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:16.894 system_profiler[1015:14113] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.132 system_profiler[1019:14130] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.134 system_profiler[1019:14130] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.372 system_profiler[1023:14150] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.373 system_profiler[1023:14150] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.613 system_profiler[1028:14170] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.615 system_profiler[1028:14170] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.855 system_profiler[1032:14195] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:17.857 system_profiler[1032:14195] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.099 system_profiler[1036:14214] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.101 system_profiler[1036:14214] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.336 system_profiler[1040:14233] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.338 system_profiler[1040:14233] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.581 system_profiler[1044:14255] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.582 system_profiler[1044:14255] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.825 system_profiler[1048:14278] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 2015-02-27 15:52:18.827 system_profiler[1048:14278] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0 Done with Benchmark run! Removing temporary files! TomsMacPro:KWSN-OSX-bench-MB_v2.1.07 Tom$ Here's another long one that seems to have passed, http://setiathome.berkeley.edu/workunit.php?wuid=1718091289 16:28:05 (1241): Can't open init data file - running in standalone mode Work Unit Info: ............... Credit multiplier is : 2.85 WU true angle range is : 0.361064 Used GPU device parameters are: Number of compute units: 14 Single buffer allocation size: 64MB Total device global memory: 1024MB max WG size: 256 local mem type: Real period_iterations_num=23 Spike: peak=24.63067, time=97.31, d_freq=1419485482.51, chirp=-17.809, fft_len=64k Spike: peak=24.07767, time=2.517, d_freq=1419485122.7, chirp=-22.209, fft_len=16k Spike: peak=24.15903, time=2.517, d_freq=1419485122.69, chirp=-22.212, fft_len=16k Spike: peak=24.22926, time=2.517, d_freq=1419485122.68, chirp=-22.216, fft_len=16k Spike: peak=24.25545, time=2.517, d_freq=1419485122.67, chirp=-22.219, fft_len=16k Spike: peak=24.24602, time=2.517, d_freq=1419485122.67, chirp=-22.222, fft_len=16k Spike: peak=24.19466, time=2.517, d_freq=1419485122.66, chirp=-22.226, fft_len=16k Spike: peak=24.1271, time=2.517, d_freq=1419485122.65, chirp=-22.228, fft_len=16k Spike: peak=24.0833, time=2.517, d_freq=1419485122.72, chirp=-22.439, fft_len=16k Spike: peak=24.2542, time=2.517, d_freq=1419485122.71, chirp=-22.443, fft_len=16k Spike: peak=24.35423, time=2.517, d_freq=1419485122.7, chirp=-22.446, fft_len=16k Spike: peak=24.44955, time=2.517, d_freq=1419485122.69, chirp=-22.449, fft_len=16k Spike: peak=24.49229, time=2.517, d_freq=1419485122.68, chirp=-22.452, fft_len=16k Spike: peak=24.51072, time=2.517, d_freq=1419485122.67, chirp=-22.456, fft_len=16k Spike: peak=24.48502, time=2.517, d_freq=1419485122.66, chirp=-22.46, fft_len=16k Spike: peak=24.43682, time=2.517, d_freq=1419485122.66, chirp=-22.462, fft_len=16k Spike: peak=24.33426, time=2.517, d_freq=1419485122.65, chirp=-22.466, fft_len=16k Spike: peak=24.22879, time=2.517, d_freq=1419485122.64, chirp=-22.469, fft_len=16k Spike: peak=24.05069, time=2.517, d_freq=1419485122.63, chirp=-22.472, fft_len=16k Pulse: peak=6.115143, time=14.94, period=2.884, d_freq=1419478408.08, score=1.03, chirp=-61.831, fft_len=1024 Best spike: peak=24.63067, time=97.31, d_freq=1419485482.51, chirp=-17.809, fft_len=64k Best autocorr: peak=17.13059, time=73.82, delay=5.5257, d_freq=1419482485.94, chirp=0.86788, fft_len=128k Best gaussian: peak=2.988612, mean=0.5315296, ChiSq=1.396593, time=86.4, d_freq=1419481914.98, score=-0.7719021, null_hyp=2.210215, chirp=-35.758, fft_len=16k Best pulse: peak=6.115143, time=14.94, period=2.884, d_freq=1419478408.08, score=1.03, chirp=-61.831, fft_len=1024 Best triplet: peak=0, time=-2.122e+11, period=0, d_freq=0, chirp=0, fft_len=0 Flopcounter: 21619962827225.500000 Spike count: 19 Autocorr count: 0 Pulse count: 1 Triplet count: 0 Gaussian count: 0 Time cpu in use since last restart: 490.8 seconds GPU device sync requested... ...GPU device synched 16:55:10 (1241): called boinc_finish(0) ID: 1647454 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
It appears the "end FindSpikes_ev==NULL" comes and goes. Anyone know where they can be coming from? Look at them all. At least the numbers match, http://setiathome.berkeley.edu/result.php?resultid=4001778056 This was the line used on that build from the config log, as you see the quotes are missing from the log; ./configure --disable-server --disable-graphics --disable-shared --enable-bitness=64 --enable-client --enable-static-client --enable-static --enable-dependency-tracking --enable-intrinsics --enable-ssse3 --build=x86_64-apple-darwin --host=x86_64-apple-darwin --target=x86_64-apple-darwin --with-boinc-platform=x86_64-apple-darwin CC=/usr/bin/clang CPPFLAGS= -DUSE_I386_OPTIMIZATIONS -DUSE_I386_XEON -DUSE_OPENCL_HD5xxx -DUSE_OPENCL -DUSE_FFTW -DASYNC_SPIKE -DSETI7 -DOCL_CHIRP3 -DUSE_SSSE3 -O3 -D_THREAD_SAFE -mtune=generic -I/Users/TBar/sah_v7_opt/AKv8/client -I/Users/TBar/sah_v7_opt/src -I/Users/TBar/sah_v7_opt/src/boinc -I/Users/TBar/sah_v7_opt/src/OpenCL_FFT LDFLAGS= -mmacosx-version-min=10.7 -ldl -lz -lpthread -framework Carbon BOINCDIR=/Users/TBar/boinc 17:25:54 (1546): Can't open init data file - running in standalone mode Not using mb_cmdline.txt-file, using commandline options. 17:25:54 (1546): Can't open init data file - running in standalone mode WARNING: init_data.xml missing OpenCL platform detected: Apple WARNING: BOINC supplied wrong platform! Number of OpenCL devices found : 3 BOINC assigns slot on device #2. WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx OCL_CHIRP3 ASYNC_SPIKE FFTW SSSE3 64bit System: Darwin x86_64 Kernel: 14.1.0 CPU : Intel(R) Xeon(R) CPU E5462 @ 2.80GHz GenuineIntel x86, Family 6 Model 23 Stepping 6 Features : FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 DTES64 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 OpenCL-kernels filename : MultiBeam_Kernels_r2855.cl ar=2.720411 NumCfft=99655 NumGauss=0 NumPulse=27784126371 NumTriplet=27784126371 Currently allocated 121 MB for GPU buffers In v_BaseLineSmooth: NumDataPoints=1048576, BoxCarLength=8192, NumPointsInChunk=32768 OS X optimized S@H Enhanced application by Alex Kan OpenCL version by Raistmer, r2855 AMD HD5 version by Raistmer Number of OpenCL platforms: 1 Work Unit Info: ............... Credit multiplier is : 2.85 WU true angle range is : 2.720411 Used GPU device parameters are: Number of compute units: 14 Single buffer allocation size: 64MB Total device global memory: 1024MB max WG size: 256 local mem type: Real period_iterations_num=23 end FindSpikes_ev==NULL end FindSpikes_ev==NULL Spike: peak=24.64636, time=57.88, d_freq=1419086387.8, chirp=-18.751, fft_len=16k Spike: peak=24.17659, time=57.88, d_freq=1419086387.81, chirp=-18.988, fft_len=16k Autocorr: peak=17.938, time=60.4, delay=1.0785, d_freq=1419080755.18, chirp=-21.128, fft_len=128k Autocorr: peak=18.43378, time=60.4, delay=1.0785, d_freq=1419080754.29, chirp=-21.142, fft_len=128k end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL end FindSpikes_ev==NULL Best spike: peak=24.64636, time=57.88, d_freq=1419086387.8, chirp=-18.751, fft_len=16k Best autocorr: peak=18.43378, time=60.4, delay=1.0785, d_freq=1419080754.29, chirp=-21.142, fft_len=128k Best gaussian: peak=0, mean=0, ChiSq=0, time=-2.122e+11, d_freq=0, score=-12, null_hyp=0, chirp=0, fft_len=0 Best pulse: peak=0.741393, time=16.79, period=0.02094, d_freq=1419077148.44, score=0.9013, chirp=0, fft_len=16 Best triplet: peak=0, time=-2.122e+11, period=0, d_freq=0, chirp=0, fft_len=0 Flopcounter: 11879411877221.558594 Spike count: 2 Autocorr count: 2 Pulse count: 0 Triplet count: 0 Gaussian count: 0 Time cpu in use since last restart: 111.1 seconds GPU device sync requested... ...GPU device synched 17:33:58 (1546): called boinc_finish(0) ID: 1647491 · |
Richard Haselgrove  Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 
|
It appears the "end FindSpikes_ev==NULL" comes and goes. Anyone know where they can be coming from? Look at them all. I think you'll need to actually read the source code you're tying to compile, for that one. ID: 1647492 · |
Urs Echternacht  Send message Joined: 15 May 99 Posts: 692 Credit: 135,197,781 RAC: 211 
|
That is normal behaviour for config.log : no more quotes, not on OS X or Linux or FreeBSD... its a log-file not a script-file. _\|/_ U r s ID: 1647624 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
I think I've progressed as much as possible under the circumstances. I've got and app that consistently matches results while running as standalone in Yosemite. The problem now is with all the NULL errors. I've constructed a list; The app will print NULL statements while running. Usually with shorties; end FindSpikes_ev==NULL When I try to run the app as standalone in Lion and 2 Mountain Lions I get; end FindSpikes_ev==NULL ERROR: OpenCL kernel/call 'Enqueueing kernel: (strip) PC_find_spike32_kernel_cl' call failed (-54) in file analyzeFuncs.cpp near line 3171. Waiting 30 sec before restart... The older r2839 runs fine on these systems. When I try to run the app in BOINC I get an immediate; MultiBeam_Kernels_r2855.clHD5_ATIRadeonBartsPROPrototype.bin_V7_14.1.0_12Dec23201400, continue with recompile... libc++abi.dylib: terminating with uncaught exception of type std::logic_error: basic_string::_S_construct NULL not valid SIGABRT: abort called It won't even compile a binary before it crashes. It runs as standalone fine, except for all the NULLs. I decided to see what it would do with the older MultiBeam_Kernels.cl in BOINC. I duplicated the .cl from r2839 and renamed it. It then created All the Binaries before crashing with; ERROR: OpenCL kernel/call 'Creating PC_find_triplets_kernel_HD5_cl from program' call failed (-46) in file analyzeFuncs.cpp near line 1510. Waiting 30 sec before restart... Which is similar to what it gives in the older systems as standalone. At least with the older .cl it created the binaries in BOINC before crashing... ID: 1647672 · |
 Raistmer  Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 
 |
It appears the "end FindSpikes_ev==NULL" comes and goes. Anyone know where they can be coming from? Try to disable ASYNC_SPIKE define. This path not fully mainained now cause I don't use it in last Windows builds. "should" work though... EDIT: Also, better to experiment with prev revs, w/o PC_find_triplets_*_HD5_cl kernels. They still under development and known to work incorrectly in last build. ID: 1647735 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
I suppose I can try it without the ASYNC_SPIKE. However, if you look at Urs r2839 you can see his App is using that option without any problems; http://setiathome.berkeley.edu/result.php?resultid=4001702650 Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx OCL_CHIRP3 ASYNC_SPIKE FFTW SSE3 64bit System: Darwin x86_64 Kernel: 14.1.0 CPU : Intel(R) Xeon(R) CPU: E5462 @ 2.80GHz OpenCL-kernels filename : MultiBeam_Kernels_r2839.cl Can you point me to the prev revs to use. I still have the repository copy I downloaded on 5 Feb but all the Executables have been stripped by Xcode. I did try using the MultiBeam_Kernels.cl from that download in a new build, but still received the 'ERROR: OpenCL kernel/call 'Creating PC_find_triplets_kernel_HD5_cl from program' call failed (-46) in file analyzeFuncs.cpp near line 1510.' error. ID: 1647822 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
New build without the ASYNC_SPIKE. I exchanged the MultiBeam_Kernels.cl & MultiBeam_Kernels_additional.cl with the ones from 5 Feb...before the compile. When run as standalone in the new Mountain Lion it shows; 12:41:41 (5968): Can't open init data file - running in standalone mode 12:41:41 (5968): Can't open init data file - running in standalone mode Not using mb_cmdline.txt-file, using commandline options. 12:41:41 (5968): Can't open init data file - running in standalone mode WARNING: init_data.xml missing OpenCL platform detected: Apple WARNING: BOINC supplied wrong platform! Number of OpenCL devices found : 3 BOINC assigns slot on device #2. WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx OCL_CHIRP3 FFTW SSSE3 64bit System: Darwin x86_64 Kernel: 12.5.0 CPU : Intel(R) Xeon(R) CPU E5462 @ 2.80GHz GenuineIntel x86, Family 6 Model 23 Stepping 6 Features : FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 DTES64 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 OpenCL-kernels filename : MultiBeam_Kernels_r2855.cl INFO: can't open binary kernel file: .//MultiBeam_Kernels_r2855.clHD5_ATIRadeonBartsPROPrototype.bin_V7_12.5.0_10, continue with recompile... Info : Building Program (binary, clBuildProgram):main kernels: OK code 0 INFO: binary kernel file created WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_524288_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_8_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_16_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_32_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_64_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_128_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_256_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_512_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_1024_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_2048_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_4096_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_8192_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_16384_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_32768_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_65536_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsPROPrototype_131072_gr64_lr16_wg256_tw0_r2855.bin_12.5.0_10, continue with recompile... ar=4.045585 NumCfft=99339 NumGauss=0 NumPulse=13750382726 NumTriplet=27806644708 Currently allocated 121 MB for GPU buffers ERROR: OpenCL kernel/call 'Creating PC_find_triplets_kernel_HD5_cl from program' call failed (-46) in file analyzeFuncs.cpp near line 1510. Waiting 30 sec before restart... It has the same Error when run as standalone in Yosemite, ERROR: OpenCL kernel/call 'Creating PC_find_triplets_kernel_HD5_cl from program' call failed (-46) in file analyzeFuncs.cpp near line 1510. Just for the heck of it I renamed the MultiBeam_Kernels.cl from Joe's build r2760 and tried that as standalone with the same build in Yosemite, basically the same results except line 1508. Then I tried the new build, without the ASYNC_SPIKE, with the current MultiBeam_Kernels.cl in Yosemite standalone with this task, http://setiathome.berkeley.edu/workunit.php?wuid=1719190641, and received; 13:05:48 (572): Can't open init data file - running in standalone mode
13:05:48 (572): Can't open init data file - running in standalone mode
Not using mb_cmdline.txt-file, using commandline options.
13:05:48 (572): Can't open init data file - running in standalone mode
WARNING: init_data.xml missing
OpenCL platform detected: Apple
WARNING: BOINC supplied wrong platform!
Number of OpenCL devices found : 3
BOINC assigns slot on device #2.
WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities
Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx OCL_CHIRP3 FFTW SSSE3 64bit
System: Darwin x86_64 Kernel: 14.1.0
CPU : Intel(R) Xeon(R) CPU E5462 @ 2.80GHz
GenuineIntel x86, Family 6 Model 23 Stepping 6
Features : FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 DTES64 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1
OpenCL-kernels filename : MultiBeam_Kernels_r2855.cl
INFO: can't open binary kernel file: .//MultiBeam_Kernels_r2855.clHD5_ATIRadeonBartsXTPrototype.bin_V7_14.1.0_12Dec23201400, continue with recompile...
Info : Building Program (binary, clBuildProgram):main kernels: OK code 0
INFO: binary kernel file created
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_524288_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_8_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_16_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_32_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_64_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_128_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_256_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_512_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_1024_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_2048_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_4096_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_8192_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_16384_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_32768_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_65536_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
WARNING: can't open binary kernel file for oclFFT plan: .//MB_clFFTplan_ATIRadeonBartsXTPrototype_131072_gr64_lr16_wg256_tw0_r2855.bin_14.1.0_12Dec23201400, continue with recompile...
ar=2.719270 NumCfft=99655 NumGauss=0 NumPulse=27772502708 NumTriplet=27772502708
Currently allocated 121 MB for GPU buffers
In v_BaseLineSmooth: NumDataPoints=1048576, BoxCarLength=8192, NumPointsInChunk=32768
OS X optimized S@H Enhanced application by Alex Kan
OpenCL version by Raistmer, r2855
AMD HD5 version by Raistmer
Number of OpenCL platforms: 1
OpenCL Platform Name: Apple
Number of devices: 3
Max compute units: 14
Max work group size: 256
Max clock frequency: 775Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts PRO Prototype
Vendor: AMD
Driver version: 1.2 (Dec 23 2014 00:17:18)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Max compute units: 14
Max work group size: 256
Max clock frequency: 775Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts PRO Prototype
Vendor: AMD
Driver version: 1.2 (Dec 23 2014 00:17:18)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Max compute units: 14
Max work group size: 256
Max clock frequency: 900Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts XT Prototype
Vendor: AMD
Driver version: 1.2 (Dec 23 2014 00:17:18)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Work Unit Info:
...............
Credit multiplier is : 2.85
WU true angle range is : 2.719270
Used GPU device parameters are:
Number of compute units: 14
Single buffer allocation size: 64MB
Total device global memory: 1024MB
max WG size: 256
local mem type: Real
period_iterations_num=23
Spike: peak=24.17966, time=6.711, d_freq=1419585066.63, chirp=-9.2639, fft_len=128k
Spike: peak=24.90196, time=6.711, d_freq=1419585066.62, chirp=-9.2648, fft_len=128k
Spike: peak=24.60994, time=6.711, d_freq=1419585066.62, chirp=-9.2657, fft_len=128k
Spike: peak=24.34005, time=3.355, d_freq=1419586787.61, chirp=-28.708, fft_len=64k
Spike: peak=24.16706, time=3.355, d_freq=1419586787.59, chirp=-28.711, fft_len=64k
Best spike: peak=24.90196, time=6.711, d_freq=1419585066.62, chirp=-9.2648, fft_len=128k
Best autocorr: peak=17.23334, time=87.24, delay=3.3741, d_freq=1419589292.86, chirp=-6.3145, fft_len=128k
Best gaussian: peak=0, mean=0, ChiSq=0, time=-2.122e+11, d_freq=0,
score=-12, null_hyp=0, chirp=0, fft_len=0
Best pulse: peak=4.237789, time=77.66, period=0.249, d_freq=1419586288.84, score=0.876, chirp=-77.209, fft_len=64
Best triplet: peak=0, time=-2.122e+11, period=0, d_freq=0, chirp=0, fft_len=0
Flopcounter: 14695275975907.673828
Spike count: 5
Autocorr count: 0
Pulse count: 0
Triplet count: 0
Gaussian count: 0
Time cpu in use since last restart: 120.6 seconds
GPU device sync requested... ...GPU device synched
13:14:00 (572): called boinc_finish(0)Hmmmmm.... It appears the Spike prints are gone, http://setiathome.berkeley.edu/workunit.php?wuid=1719324138 Credit multiplier is : 2.85 WU true angle range is : 2.720411 Used GPU device parameters are: Number of compute units: 14 Single buffer allocation size: 64MB Total device global memory: 1024MB max WG size: 256 local mem type: Real period_iterations_num=23 Spike: peak=24.64635, time=57.88, d_freq=1419086387.8, chirp=-18.751, fft_len=16k Spike: peak=24.17659, time=57.88, d_freq=1419086387.81, chirp=-18.988, fft_len=16k Autocorr: peak=17.938, time=60.4, delay=1.0785, d_freq=1419080755.18, chirp=-21.128, fft_len=128k Autocorr: peak=18.43379, time=60.4, delay=1.0785, d_freq=1419080754.29, chirp=-21.142, fft_len=128k Best spike: peak=24.64635, time=57.88, d_freq=1419086387.8, chirp=-18.751, fft_len=16k Best autocorr: peak=18.43379, time=60.4, delay=1.0785, d_freq=1419080754.29, chirp=-21.142, fft_len=128k Best gaussian: peak=0, mean=0, ChiSq=0, time=-2.122e+11, d_freq=0, score=-12, null_hyp=0, chirp=0, fft_len=0 Best pulse: peak=0.7413931, time=16.79, period=0.02094, d_freq=1419077148.44, score=0.9013, chirp=0, fft_len=16 Best triplet: peak=0, time=-2.122e+11, period=0, d_freq=0, chirp=0, fft_len=0 Flopcounter: 14695186564453.558594 Spike count: 2 Autocorr count: 2 Pulse count: 0 Triplet count: 0 Gaussian count: 0 Time cpu in use since last restart: 116.9 seconds GPU device sync requested... ...GPU device synched 13:53:10 (816): called boinc_finish(0) ID: 1647908 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
Still no progress with running in standalone Mountain Lion or Yosemite BOINC. New build with everything current, without ASYNC_SPIKE. In Mountain Lion it's the same error with the current .cl; ERROR: OpenCL kernel/call 'Enqueueing kernel: (strip) PC_find_spike32_kernel_cl' call failed (-54) in file analyzeFuncs.cpp near line 3171. With the older .cl it's; ERROR: OpenCL kernel/call 'Creating PC_find_triplets_kernel_HD5_cl from program' call failed (-46) in file analyzeFuncs.cpp near line 1510. It works fine in standalone Yosemite with the results matching results from main. When run in BOINC it immediately crashes with; OpenCL-kernels filename : MultiBeam_Kernels_r2855.cl INFO: can't open binary kernel file: /Volumes/Mov1/BOINC/Maverick/BOINC Data/projects/setiathome.berkeley.edu/MultiBeam_Kernels_r2855.clHD5_ATIRadeonBartsPROPrototype.bin_V7_14.1.0_12Dec23201400, continue with recompile... libc++abi.dylib: terminating with uncaught exception of type std::logic_error: basic_string::_S_construct NULL not valid SIGABRT: abort called Crashed executable name: MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin Machine type Intel 80486 (64-bit executable) System version: Macintosh OS 10.10.2 build 14C109 Sat Feb 28 16:18:00 2015 atos cannot load symbols for the file MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin for architecture x86_64. 0 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x00000001029fe42b 1 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x00000001029ef6f6 2 libsystem_platform.dylib 0x00007fff92b2cf1a 3 ??? 0x00007fff5d4bb5c8 4 libsystem_c.dylib 0x00007fff8ab88b53 5 libc++abi.dylib 0x00007fff8d996a21 6 libc++abi.dylib 0x00007fff8d9be9b9 7 libobjc.A.dylib 0x00007fff8af2f7eb 8 libc++abi.dylib 0x00007fff8d9bc0a1 9 libc++abi.dylib 0x00007fff8d9bbb30 10 libstdc++.6.dylib 0x00007fff8e03948b 11 libstdc++.6.dylib 0x00007fff8e063883 12 libstdc++.6.dylib 0x00007fff8e061c8e 13 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x0000000102998a87 14 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x000000010299929a 15 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x000000010286c577 16 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x0000000102884132 17 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x0000000102889d16 18 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x0000000102868d70 19 MBv7_7.05r2855_ati5_ssse3_x86_64-apple-darwin 0x0000000102742054 20 ??? 0x0000000000000005 No writing Binaries, no passing go...just crash. :-( ID: 1647977 · |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
when I said prev rev I meant whole rev sources, not just single CL file. HD5 triplet was added in rev 2799 so use 2763 to avoid them. r2850 and r2849 had some changes in OS X build path though, maybe worth to merge that changes too. ID: 1648131 · |
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874
|
when I said prev rev I meant whole rev sources, not just single CL file. I think he's working off a single tarball. You'll have to teach him the full subversion management if you want him to go down that route. ID: 1648132 · |
Brent Norman    Send message Joined: 1 Dec 99 Posts: 2786 Credit: 685,657,289 RAC: 835 
|
No offence TBar, but wouldn't enabling VNC let some of the programmers to see what your actually doing, and provide better input? Edit: They may not have a Mac in front of them, but they could. 
ID: 1648141 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
What's there to see? I'm running a configure line in the terminal and it either spits out an App or doesn't. Right now the 'doesn't' is overwhelming. I think I know where Joe Fox went. Seems Mavericks doesn't compile an App either now, it may have earlier, but the updated copy you get from Apple at present will not. That leaves One OSX that will compile an App using the Terminal. Xcode is hopeless because all it takes is One Error and it will not compile either. Right now it doesn't look good, I'm going to have to fix my old copy of Mountain Lion before another build can be made. Here's an interesting error from the build on Mavericks, there are a few of these; "std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::__init(char const*, unsigned long)", referenced from:
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-main.o
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-analyzeFuncs.o
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-analyzeReport.o
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-analyzePoT.o
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-autocorr.o
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-pulsefind.o
seti_error::seti_error(int, char const*, int, char const*) in seti_boinc-gaussfit.oKinda reminds me of the BOINC Error I'm seeing with the Mountain Lion build. Yes, I'm using a downloaded copy of the branch. I never intended to make a career out of this. I just wanted to build a single App that wasn't such a pain to use. If an App makes your typing stutter by a second or two, it probably won't get used much. ID: 1648207 · |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
If an App makes your typing stutter by a second or two, it probably won't get used much. Did you try all builtin tuning capabilities before building own binary? ID: 1648254 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
If an App makes your typing stutter by a second or two, it probably won't get used much. I'm not aware of any prebuild tuning. The last line I used was; ./configure --disable-server --disable-graphics --disable-shared --enable-bitness=64 --enable-client --enable-static-client --enable-static --enable-dependency-tracking --enable-intrinsics --enable-ssse3 --build=x86_64-apple-darwin --host=x86_64-apple-darwin --target=x86_64-apple-darwin --with-boinc-platform=x86_64-apple-darwin CC=/usr/bin/clang CPPFLAGS= -DUSE_I386_OPTIMIZATIONS -DUSE_I386_XEON -DUSE_OPENCL_HD5xxx -DUSE_OPENCL -DUSE_FFTW -DSETI7 -DOCL_CHIRP3 -DUSE_SSSE3 -O3 -D_THREAD_SAFE -mtune=generic -I/Users/TBar/sah_v7_opt/AKv8/client -I/Users/TBar/sah_v7_opt/src -I/Users/TBar/sah_v7_opt/src/boinc -I/Users/TBar/sah_v7_opt/src/OpenCL_FFT LDFLAGS= -mmacosx-version-min=10.7 -ldl -lz -framework Carbon BOINCDIR=/Users/TBar/boinc If you can make some suggestions I'll give it a try, as soon as I restore Mountain Lion. The first attempt failed, all it did was revert the system back to 10.8.0 but left all the third party items in usr. If you try the built-in Apple restore partition it just says Mountain Lion is not available at the moment from the Apple store. I might have to burn a disk and try a complete reinstall. ID: 1648262 · |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
You have binary compiled by others. It should accept some params described in ReadMe. Did you try to tune that params (namely, period_iterations_num) to reduce lags? ID: 1648289 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
I went up to period_iterations 100, from what I understand higher than that won't matter. It didn't make any difference, just like it didn't make any difference in Windows XP or Linux on this machine. So I set it back to 32, as seen here, http://setiathome.berkeley.edu/results.php?hostid=7199204&offset=0&show_names=0&state=0&appid=11 I just installed Mavericks again last night, I was going to see how it worked so I set it at default for the first few hours. Unfortunately the server went belly up last night and I haven't been able to report the tasks it finished or receive new tasks. I suspect Mavericks will be the same as Lion, Mountain Lion, Yosemite, Windows XP, and Linux though. Namely, once you pass 32 it doesn't make much difference. There isn't any stutter on my ATI cards with AstroPulses, too bad they have been hard to find for the last 4 months. Running the Mountain Lion ESD installer from Snow Leopard didn't work either, it didn't replace the ML system, the files are still in usr, and my new build in ML came out at 2.1 mb. I expect if I booted to Yosemite to run it the fftw version will be 3.2.2 as before. Yes, I did reinstall 3.3.4 again, didn't make a difference. 3.2.2 is dug like a tick. I wish I knew where it was hiding. As for running the new build in ML, well... 13:15:15 (79299): Can't open init data file - running in standalone mode ERROR: OpenCL kernel/call 'Enqueueing kernel: (strip) PC_find_spike32_kernel_cl' call failed (-54) in file analyzeFuncs.cpp near line 3171. Same old same old. ID: 1648302 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
Finally fixed the old Mountain Lion. Now it's producing Apps 3.6 mb in size that use FFTW 3.3.4 with sse2. The Apps also work in Mavericks now, and Yosemite in standalone. Still have the same Error in BOINC though. It would be nice to run one in BOINC and see if it's still about a minute faster on the shorties. I ended up just deleting the entire usr folder while running from a different system and having the installer replace the folder. Bad juju in that one, deleting every file with FFTW in it's name wouldn't even work. So, I reinstalled all the Terminal Apps and now all is well, except it still won't run the Apps in standalone in Mountain Lion...same error. I was hoping I could build And test the App on one system, rebooting to test the App grew old days ago. I think the current build is about the best it's going to get though, anyone know what's going on with this error; INFO: can't open binary kernel file: /Volumes/Mov1/BOINC/Maverick/BOINC Data/projects/setiathome.berkeley.edu/MultiBeam_Kernels_r2855.clHD5_ATIRadeonBartsPROPrototype.bin_V7_14.1.0_12Dec23201400, continue with recompile... It would really be nice to run this App in BOINC. BTW, did you know you can take that ESD installer and make it into a bootable USB drive? This works, http://www.hongkiat.com/blog/clean-install-mountain-lion/ ID: 1648516 · |
|
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768
|
BTW, checkout this VLAR run last night. Please don't send any GPU VLARS to Macs, 1hr24mins to finish, and there was some screen stutter. There's a little stutter right now running 3 r2839 tasks with the 'Number of period iterations for PulseFind set to 48'. http://setiathome.berkeley.edu/result.php?resultid=4005071518 Spike count: 8 Autocorr count: 1 Pulse count: 12 Triplet count: 2 Gaussian count: 0 Time cpu in use since last restart: 8841.1 seconds 03:17:03 (1558): Can't open init data file - running in standalone mode
03:17:03 (1558): Can't open init data file - running in standalone mode
Not using mb_cmdline.txt-file, using commandline options.
03:17:03 (1558): Can't open init data file - running in standalone mode
WARNING: init_data.xml missing
OpenCL platform detected: Apple
WARNING: BOINC supplied wrong platform!
Number of OpenCL devices found : 3
BOINC assigns slot on device #2.
WARNING: BOINC failed to provide OpenCL device, using own enumeration abilities
Build features: SETI7 Non-graphics OpenCL USE_OPENCL_HD5xxx OCL_CHIRP3 FFTW SSSE3 64bit
System: Darwin x86_64 Kernel: 13.4.0
CPU : Intel(R) Xeon(R) CPU E5462 @ 2.80GHz
GenuineIntel x86, Family 6 Model 23 Stepping 6
Features : FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 DTES64 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1
OpenCL-kernels filename : MultiBeam_Kernels_r2855.cl
ar=0.010129 NumCfft=146007 NumGauss=0 NumPulse=50183457664 NumTriplet=67977339808
Currently allocated 121 MB for GPU buffers
In v_BaseLineSmooth: NumDataPoints=1048576, BoxCarLength=8192, NumPointsInChunk=32768
OS X optimized S@H Enhanced application by Alex Kan
OpenCL version by Raistmer, r2855
AMD HD5 version by Raistmer
Number of OpenCL platforms: 1
OpenCL Platform Name: Apple
Number of devices: 3
Max compute units: 14
Max work group size: 256
Max clock frequency: 775Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts PRO Prototype
Vendor: AMD
Driver version: 1.2 (Aug 17 2014 20:28:00)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Max compute units: 14
Max work group size: 256
Max clock frequency: 775Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts PRO Prototype
Vendor: AMD
Driver version: 1.2 (Aug 17 2014 20:28:00)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Max compute units: 14
Max work group size: 256
Max clock frequency: 900Mhz
Max memory allocation: 268435456
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Queue properties:
Out-of-Order: No
Name: ATI Radeon Barts XT Prototype
Vendor: AMD
Driver version: 1.2 (Aug 17 2014 20:28:00)
Version: OpenCL 1.2
Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event 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_image2d_from_buffer cl_khr_depth_images
Work Unit Info:
...............
Credit multiplier is : 2.85
WU true angle range is : 0.010129
Used GPU device parameters are:
Number of compute units: 14
Single buffer allocation size: 64MB
Total device global memory: 1024MB
max WG size: 256
local mem type: Real
period_iterations_num=23
Pulse: peak=6.90952, time=53.7, period=19.37, d_freq=1420419025.2, score=1.019, chirp=-3.2007, fft_len=256
Spike: peak=24.52609, time=46.98, d_freq=1420423661.25, chirp=6.7582, fft_len=128k
Triplet: peak=10.62535, time=15.31, period=4.614, d_freq=1420421538.05, chirp=-9.9866, fft_len=4k
Spike: peak=24.55023, time=46.98, d_freq=1420420665.19, chirp=-13.055, fft_len=128k
Spike: peak=25.43507, time=46.98, d_freq=1420420665.2, chirp=-13.06, fft_len=128k
Spike: peak=24.5384, time=46.98, d_freq=1420420665.19, chirp=-13.062, fft_len=128k
Spike: peak=24.00377, time=46.98, d_freq=1420420665.19, chirp=-13.066, fft_len=128k
Autocorr: peak=18.25633, time=46.98, delay=2.8997, d_freq=1420419262.92, chirp=-14.028, fft_len=128k
Pulse: peak=6.551886, time=53.74, period=20.53, d_freq=1420420317.68, score=1.01, chirp=-20.674, fft_len=1024
Pulse: peak=0.5515752, time=53.7, period=0.4866, d_freq=1420421603.4, score=1.024, chirp=-22.675, fft_len=256
Spike: peak=24.3829, time=60.4, d_freq=1420418558.3, chirp=25.999, fft_len=128k
Spike: peak=25.2093, time=60.4, d_freq=1420418558.3, chirp=26.003, fft_len=128k
Spike: peak=24.96848, time=60.4, d_freq=1420418558.3, chirp=26.007, fft_len=128k
Pulse: peak=3.691362, time=53.79, period=10.35, d_freq=1420418497.69, score=1.008, chirp=45.149, fft_len=2k
Pulse: peak=3.926738, time=53.7, period=10.28, d_freq=1420418902.37, score=1.003, chirp=-50.952, fft_len=256
Pulse: peak=3.922193, time=53.7, period=10.28, d_freq=1420418888.08, score=1.002, chirp=-51.218, fft_len=256
Triplet: peak=13.01367, time=49.49, period=15.47, d_freq=1420417240.31, chirp=-61.888, fft_len=128
Pulse: peak=3.23059, time=53.79, period=8.677, d_freq=1420423392.47, score=1.013, chirp=70.192, fft_len=2k
Pulse: peak=0.5404494, time=53.7, period=0.5417, d_freq=1420422543.73, score=1.001, chirp=-73.359, fft_len=256
Pulse: peak=10.71628, time=53.7, period=27.84, d_freq=1420418829.13, score=1.033, chirp=-73.627, fft_len=256
Pulse: peak=3.208741, time=53.79, period=8.756, d_freq=1420421248.68, score=1.006, chirp=-78.694, fft_len=2k
Pulse: peak=2.775339, time=53.71, period=6.379, d_freq=1420424315.86, score=1.03, chirp=80.029, fft_len=512
Pulse: peak=3.356088, time=53.79, period=8.389, d_freq=1420416219.36, score=1.053, chirp=89.932, fft_len=2k
Best spike: peak=25.43507, time=46.98, d_freq=1420420665.2, chirp=-13.06, fft_len=128k
Best autocorr: peak=18.25633, time=46.98, delay=2.8997, d_freq=1420419262.92, chirp=-14.028, fft_len=128k
Best gaussian: peak=0, mean=0, ChiSq=0, time=-2.122e+11, d_freq=0,
score=-12, null_hyp=0, chirp=0, fft_len=0
Best pulse: peak=3.356088, time=53.79, period=8.389, d_freq=1420416219.36, score=1.053, chirp=89.932, fft_len=2k
Best triplet: peak=13.01367, time=49.49, period=15.47, d_freq=1420417240.31, chirp=-61.888, fft_len=128
Flopcounter: 26443466789612.894531
Spike count: 8
Autocorr count: 1
Pulse count: 12
Triplet count: 2
Gaussian count: 0
Time cpu in use since last restart: 1781.0 seconds
GPU device sync requested... ...GPU device synched
04:31:28 (1558): called boinc_finish(0)ID: 1648539 · |
©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.