I've Built a Couple OSX CUDA Apps...

Message boards : Number crunching : I've Built a Couple OSX CUDA Apps...
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 58 · Next

AuthorMessage
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1747299 - Posted: 6 Dec 2015, 20:53:08 UTC - in response to Message 1747294.  

...It reads the first file that it can find out of ~/.bash_profile, ~/.bash_login, and ~/.profile and does not read any further files.

Hmmm, ALL of those have the same effect. If I have one present the Compiler can Not find pkg-config and terminates immediately. Same thing in Mavericks and Mountain Lion. Wasted about 30 minutes trying to reinstall/fix pkg-config without any success, remove the file and it works the way it use to.

Now with CUDA Toolkit 6.5 I'm seeing about the same as with CUDA 6.0. In Mavericks I still get;
clang: error: no such file or directory: 'seti_cuda-analyzeFuncs_sse2.o'
clang: error: no such file or directory: 'seti_cuda-analyzeFuncs_sse3.o'
clang: error: no such file or directory: 'seti_cuda-analyzeFuncs.o'
clang: error: no such file or directory: 'cudaAcceleration.o'
clang: error: no such file or directory: 'cudaAcc_CalcChirpData.o'
clang: error: no such file or directory: 'cudaAcc_fft.o'
clang: error: no such file or directory: 'cudaAcc_gaussfit.o'
clang: error: no such file or directory: 'cudaAcc_PowerSpectrum.o'
clang: error: no such file or directory: 'cudaAcc_pulsefind.o'
clang: error: no such file or directory: 'cudaAcc_summax.o'
clang: error: no such file or directory: 'cudaAcc_transpose.o'
clang: error: no such file or directory: 'cudaAcc_utilities.o'
clang: error: no such file or directory: 'cudaAcc_autocorr.o'
make[2]: [seti_cuda] Error 1 (ignored)
...
/bin/cp seti_cuda setiathome_x41zc_x86_64-apple-darwin_cuda65
cp: seti_cuda: No such file or directory
make[2]: [setiathome_x41zc_x86_64-apple-darwin_cuda65] Error 1 (ignored)
strip setiathome_x41zc_x86_64-apple-darwin_cuda65
error: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/strip: can't open file: setiathome_x41zc_x86_64-apple-darwin_cuda65 (No such file or directory)
make[2]: [setiathome_x41zc_x86_64-apple-darwin_cuda65] Error 1 (ignored)

In Mountain Lion the SSE errors are not there.
Also, with the Modified code it is just complaining about 'compute_10' now, but it has the same complaint with the Stock code;
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_pulsefind.cu -o cudaAcc_pulsefind.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'

So, is it possible to have it ignore 'compute_10'?
Also in Mavericks, I'm still getting the same Error with both codes;
In file included from analyzeFuncs.cpp:73:
./cuda/cudaAcc_utilities.h:4:37: error: unknown type name 'cudaError_t'
extern int cudaAcc_SafeCall_No_Sync(cudaError_t err, char* errMessage, char* file, int line);
                                    ^
./cuda/cudaAcc_utilities.h:5:34: error: unknown type name 'cudaError_t'
extern int cudaAcc_SafeCall_Sync(cudaError_t err, char* errMessage, char* file, int line);
                                 ^
./cuda/cudaAcc_utilities.h:6:43: error: unknown type name 'cudaError_t'
extern int cudaAcc_SafeCall_No_Sync_wExit(cudaError_t err, char* errMessage, char* file, int line);

make[2]: [cudaAcc_pulsefind.o] Error 1 (ignored)


You can try add to the cudaAcc_utilities.h in the first line

#include <cuda_runtime_api.h>
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1747299 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747307 - Posted: 6 Dec 2015, 21:10:12 UTC - in response to Message 1747299.  
Last modified: 6 Dec 2015, 21:42:08 UTC

You can try add to the cudaAcc_utilities.h in the first line

#include <cuda_runtime_api.h>

But, I get the same Error when using your code that already has it;
// analyzeFuncs.C
// $Id: analyzeFuncs.cpp,v 1.34.2.44 2007/08/16 10:13:55 charlief Exp $
//

#define DO_SMOOTH

#if defined(__linux__)
#include <cuda_runtime_api.h>
#include <unistd.h>
#endif

#if defined(__APPLE__) || defined(__linux__) // Duplicate definitions in windows
  #include "version.h"
  const char *BOINC_PACKAGE_STRING="libboinc: "BOINC_VERSION_STRING;
//#else
#elif !defined(_WIN32) && !defined(__linux__)
   #include "config.h"
   const char *BOINC_PACKAGE_STRING="libboinc: "PACKAGE_STRING;
#endif


On the Compute_10 Error, Every Missing file ends with the same Error;
nvcc fatal : Unsupported gpu architecture 'compute_10'

Surely there's some way to Kill 'Compute_10'.
This is the latest run in Mavericks with the Modified Code;
mv -f .deps/seti_cuda-xml_util.Tpo .deps/seti_cuda-xml_util.Po
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcceleration.cu -o cudaAcceleration.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcceleration.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_CalcChirpData.cu -o cudaAcc_CalcChirpData.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_CalcChirpData.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_fft.cu -o cudaAcc_fft.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_fft.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_gaussfit.cu -o cudaAcc_gaussfit.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_gaussfit.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_PowerSpectrum.cu -o cudaAcc_PowerSpectrum.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_PowerSpectrum.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_pulsefind.cu -o cudaAcc_pulsefind.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_pulsefind.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_summax.cu -o cudaAcc_summax.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_summax.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_transpose.cu -o cudaAcc_transpose.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_transpose.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_utilities.cu -o cudaAcc_utilities.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_utilities.o] Error 1 (ignored)
/Developer/NVIDIA/CUDA-6.5/bin/nvcc -c cuda/cudaAcc_autocorr.cu -o cudaAcc_autocorr.o -Icuda -DHAVE_CONFIG_H -I. -I..  -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db --maxrregcount=32 -O3 --use_fast_math --ptxas-options="-v" --compiler-options "-g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H  -fno-strict-aliasing" -m64  -gencode arch=compute_10,code=sm_10 -gencode arch=compute_11,code=sm_11 -gencode arch=compute_12,code=sm_12 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -I/Developer/NVIDIA/CUDA-6.5/include
nvcc fatal   : Unsupported gpu architecture 'compute_10'
make[2]: [cudaAcc_autocorr.o] Error 1 (ignored)
g++ -DHAVE_CONFIG_H -I. -I..  -g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H -DUSE_I386_OPTIMIZATIONS -DUSE_I386_XEON -DHALF_STRIDE -DSETI7 -DUSE_SSE41 -O3 -I/Users/Tom/sah_v7_opt/Xbranch/client/vector -I/usr/local/cuda/lib -I/usr/local/include -g -O2 -g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math -DHAVE_CONFIG_H -DTEXT_UI -DNDEBUG -DCLIENT -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db -D_THREAD_SAFE   -msse2 -mfpmath=sse -DUSE_SSE -DUSE_SSE2   -c cuda/cudaAcc_analyzeReport.cpp -o cudaAcc_analyzeReport.o -Icuda
In file included from cuda/cudaAcc_analyzeReport.cpp:32:
In file included from ./analyzeReport.h:33:
In file included from ./seti.h:43:
In file included from ./seti_header.h:43:
In file included from ../db/db_table.h:27:
In file included from ../db/sqlrow.h:28:
ID: 1747307 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747320 - Posted: 6 Dec 2015, 22:19:29 UTC - in response to Message 1747307.  
Last modified: 6 Dec 2015, 22:26:14 UTC

Those gencode entries are in the makefile template in the client directory, so when you run configure populate the generated Makefile. So the thing to do is open the template ( IIRC .am or .in, not looking at it right now), search for gencode, and remove any entries below compute_20, then rerun configure from the root.

Once through that, I'd like to see if you get the same Framework problem that's sending me to XCode land. If you don't, then I'll be scratching my head, but yay! if you do run into the same wall, well we'd be at pretty much the same point... annoyed with Cuda+autotools on Mac. There seems to be something in one of the includes that triggers this Framework inclusion, but haven't had a chance to isolate it yet. probably after work if it isn't 50 billion degrees again (working on a Mac pro in Aussie summer is toasty, lol)
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747320 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747329 - Posted: 6 Dec 2015, 23:02:52 UTC - in response to Message 1747320.  
Last modified: 6 Dec 2015, 23:05:29 UTC

OK, they appear to be in both Makefiles that are generated from configure. I ran the configure line;
./configure --disable-graphics --disable-shared --enable-bitness=64 --enable-client --enable-static-client --enable-static --enable-dependency-tracking --enable-intrinsics --build=x86_64-apple-darwin --host=x86_64-apple-darwin --target=x86_64-apple-darwin --with-boinc-platform=x86_64-apple-darwin --enable-fast-math CC="/usr/bin/clang" CPPFLAGS=" -DUSE_I386_OPTIMIZATIONS -DUSE_I386_XEON -DHALF_STRIDE -DSETI7 -DUSE_SSE41 -O3 -I/Users/Tom/sah_v7_opt/Xbranch/client/vector -I/usr/local/cuda/lib" LDFLAGS=" -mmacosx-version-min=10.8 -ldl /usr/lib/libz.1.dylib -lpthread -mtune=core2 -march=core2 -fstrict-aliasing -framework Carbon" BOINCDIR="/Users/Tom/boinc"

Then I opened the makefiles and removed everything less than -gencode arch=compute_20,code=sm_20. Afterwards I ran make -i -k. I received a few of these;
cuda/cudaAcceleration.cu(116): warning: conversion from a string literal to "char *" is deprecated
cuda/cudaAcceleration.cu(153): error: identifier "nvGetDriverVersion" is undefined
cuda/cudaAcceleration.cu(154): error: identifier "ddverstring" is undefined
2 errors detected in the compilation of "/var/folders/6c/hy0ffg_90sz3wbxfk7gzw3800000gn/T//tmpxft_0000ef6d_00000000-6_cudaAcceleration.cpp1.ii".
make[2]: [cudaAcceleration.o] Error 2 (ignored)
cuda/cudaAcc_fft.cu(21): warning: conversion from a string literal to "char *" is deprecated
cuda/cudaAcc_gaussfit.cu(490): warning: expression has no effect
cuda/cudaAcc_gaussfit.cu(503): warning: conversion from a string literal to "char *" is deprecated
cuda/cudaAcc_pulsefind.cu(245): warning: conversion from a string literal to "char *" is deprecated
cuda/cudaAcc_summax.cu(252): warning: conversion from a string literal to "char *" is deprecated
cuda/cudaAcc_transpose.cu(72): warning: conversion from a string literal to "char *" is deprecated

It ended with this;
7 warnings generated.
g++ -mmacosx-version-min=10.8 -ldl /usr/lib/libz.1.dylib -lpthread -mtune=core2 -march=core2 -fstrict-aliasing -framework Carbon -Wl,-rpath,\$ORIGIN -o seti_cuda seti_cuda-main.o seti_cuda-confsettings.o seti_cuda-analyzeFuncs_vector.o seti_cuda-analyzeFuncs_fpu.o seti_cuda-analyzeFuncs_sse.o seti_cuda-analyzeFuncs_sse2.o seti_cuda-analyzeFuncs_sse3.o seti_cuda-analyzeFuncs_x86_64.o seti_cuda-analyzeFuncs_altivec.o seti_cuda-x86_float4.o seti_cuda-hires_timer.o seti_cuda-analyzeFuncs.o seti_cuda-analyzeReport.o seti_cuda-analyzePoT.o seti_cuda-pulsefind.o seti_cuda-gaussfit.o seti_cuda-lcgamm.o seti_cuda-malloc_a.o seti_cuda-seti.o seti_cuda-seti_header.o seti_cuda-timecvt.o seti_cuda-s_util.o seti_cuda-sah_version.o seti_cuda-worker.o seti_cuda-chirpfft.o seti_cuda-spike.o seti_cuda-progress.o seti_cuda-fft8g.o seti_cuda-gdata.o seti_cuda-autocorr.o seti_cuda-schema_master.o seti_cuda-sqlrow.o seti_cuda-sqlblob.o seti_cuda-xml_util.o   -L/Users/Tom/boinc/api -lboinc_api -L/Users/Tom/boinc/lib -lboinc  -lm      cudaAcceleration.o cudaAcc_CalcChirpData.o cudaAcc_fft.o cudaAcc_gaussfit.o cudaAcc_PowerSpectrum.o cudaAcc_pulsefind.o cudaAcc_summax.o cudaAcc_transpose.o cudaAcc_utilities.o cudaAcc_autocorr.o cudaAcc_analyzeReport.o -lpthread -L/Developer/NVIDIA/CUDA-6.5/lib64 -lcudart -lcufft -I/Developer/NVIDIA/CUDA-6.5/include
clang: error: no such file or directory: 'seti_cuda-analyzeFuncs_sse2.o'
clang: error: no such file or directory: 'seti_cuda-analyzeFuncs_sse3.o'
clang: error: no such file or directory: 'seti_cuda-analyzeFuncs.o'
clang: error: no such file or directory: 'cudaAcceleration.o'
make[2]: [seti_cuda] Error 1 (ignored)
...
/bin/cp seti_cuda setiathome_x41zc_x86_64-apple-darwin_cuda65
cp: seti_cuda: No such file or directory
make[2]: [setiathome_x41zc_x86_64-apple-darwin_cuda65] Error 1 (ignored)
strip setiathome_x41zc_x86_64-apple-darwin_cuda65
error: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/strip: can't open file: setiathome_x41zc_x86_64-apple-darwin_cuda65 (No such file or directory)
make[2]: [setiathome_x41zc_x86_64-apple-darwin_cuda65] Error 1 (ignored)
/bin/cp setiathome_x41zc_x86_64-apple-darwin_cuda65 ../
cp: setiathome_x41zc_x86_64-apple-darwin_cuda65: No such file or directory
make[2]: [setiathome_x41zc_x86_64-apple-darwin_cuda65] Error 1 (ignored)
make[2]: Nothing to be done for `all-am'.

Much closer to an App, and No framework land...as far as I can see.
ID: 1747329 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747330 - Posted: 6 Dec 2015, 23:13:35 UTC - in response to Message 1747329.  
Last modified: 6 Dec 2015, 23:17:35 UTC

Much closer to an App, and No framework land...as far as I can see.


Good! yeah I was managing to reach a different brick wall, so will check it out before going the XCode route.

[Edit:] the call to nvapi to get the driver version, and the associated include, are Windows calls, and so should be wrapped if #ifdef _WIN32 ... #endif , that's my hamfist at work.
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747330 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747332 - Posted: 6 Dec 2015, 23:42:31 UTC - in response to Message 1747330.  

If I use Mountain Lion the two SSE errors will be gone. That leaves;
cuda/cudaAcceleration.cu(153): error: identifier "nvGetDriverVersion" is undefined
cuda/cudaAcceleration.cu(154): error: identifier "ddverstring" is undefined
and the;
In file included from analyzeFuncs.cpp:73:
./cuda/cudaAcc_utilities.h:4:37: error: unknown type name 'cudaError_t'
extern int cudaAcc_SafeCall_No_Sync(cudaError_t err, char* errMessage, char* file, int line);

Close, very close...
ID: 1747332 · Report as offensive
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1747338 - Posted: 7 Dec 2015, 0:45:49 UTC - in response to Message 1747332.  
Last modified: 7 Dec 2015, 0:47:41 UTC

If I use Mountain Lion the two SSE errors will be gone. That leaves;
cuda/cudaAcceleration.cu(153): error: identifier "nvGetDriverVersion" is undefined
cuda/cudaAcceleration.cu(154): error: identifier "ddverstring" is undefined
and the;
In file included from analyzeFuncs.cpp:73:
./cuda/cudaAcc_utilities.h:4:37: error: unknown type name 'cudaError_t'
extern int cudaAcc_SafeCall_No_Sync(cudaError_t err, char* errMessage, char* file, int line);

Close, very close...


1) can you comment out the lines 153 and 154 like this
//nvGetDriverVersion();
//fprintf(stderr,"nVidia Driver Version %s\n",ddverstring);

2) change cudaError_t to int (a temporary fix).
2b) My #include has an #if defined that checks for __linux__. It should check for __APPLE__ too. like this
#if defined(__APPLE__) || defined(__linux__)
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1747338 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747357 - Posted: 7 Dec 2015, 3:21:56 UTC - in response to Message 1747338.  
Last modified: 7 Dec 2015, 3:34:47 UTC

Now in Mountain Lion with the modified code. I can't seem to find either of these anywhere;
cuda/cudaAcceleration.cu(153): error: identifier "nvGetDriverVersion" is undefined
cuda/cudaAcceleration.cu(154): error: identifier "ddverstring" is undefined

Right now it doesn't appear to be a problem.
I made this change; #if defined(__APPLE__) || defined(__linux__), and that appears to work.

Right now the problem is Many of these errors;
ptxas /var/folders/6c/hy0ffg_90sz3wbxfk7gzw3800000gn/T//tmpxft_00005108_00000000-17_cudaAcc_CalcChirpData.compute_20.ptx, line 47; error : Instruction 'ld.nc' requires .target sm_32 or higher
I added a few sm(s) to the makefile but it still gives the error, the makefiles now have;
NVCCFLAGS = -O3 --use_fast_math --ptxas-options="-v" --compiler-options "$(AM_CXXFLAGS) $(CXXFLAGS) -fno-strict-aliasing" -m64 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -gencode arch=compute_32,code=compute_32 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50

I had to remove sm_52 as it was unsupported. I need a list of the sm(s) to include.

Changed the line to;
NVCCFLAGS = -O3 --use_fast_math --ptxas-options="-v" --compiler-options "$(AM_CXXFLAGS) $(CXXFLAGS) -fno-strict-aliasing" -m64 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -gencode arch=compute_32,code=sm_32 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50
But still getting Hundreds of lines similar to;
ptxas /var/folders/6c/hy0ffg_90sz3wbxfk7gzw3800000gn/T//tmpxft_00005c34_00000000-17_cudaAcc_gaussfit.compute_20.ptx, line 70; error : Instruction 'ld.nc' requires .target sm_32 or higher


The end result is;
clang: error: no such file or directory: 'cudaAcc_CalcChirpData.o'
clang: error: no such file or directory: 'cudaAcc_gaussfit.o'
clang: error: no such file or directory: 'cudaAcc_PowerSpectrum.o'
clang: error: no such file or directory: 'cudaAcc_pulsefind.o'
clang: error: no such file or directory: 'cudaAcc_summax.o'
clang: error: no such file or directory: 'cudaAcc_autocorr.o'
make[2]: [seti_cuda] Error 1 (ignored)
ID: 1747357 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747370 - Posted: 7 Dec 2015, 3:55:37 UTC
Last modified: 7 Dec 2015, 4:21:51 UTC

OK, I was reading it wrong. What it wanted to do was Remove everything Lower than sm_32...whatever card sm_32 is. I can do that.

Now WTH is this?
7 warnings generated.
g++ -mmacosx-version-min=10.8 -ldl /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -framework Carbon -Wl,-rpath,\$ORIGIN -o seti_cuda seti_cuda-main.o seti_cuda-confsettings.o seti_cuda-analyzeFuncs_vector.o seti_cuda-analyzeFuncs_fpu.o seti_cuda-analyzeFuncs_sse.o seti_cuda-analyzeFuncs_sse2.o seti_cuda-analyzeFuncs_sse3.o seti_cuda-analyzeFuncs_x86_64.o seti_cuda-analyzeFuncs_altivec.o seti_cuda-x86_float4.o seti_cuda-hires_timer.o seti_cuda-analyzeFuncs.o seti_cuda-analyzeReport.o seti_cuda-analyzePoT.o seti_cuda-pulsefind.o seti_cuda-gaussfit.o seti_cuda-lcgamm.o seti_cuda-malloc_a.o seti_cuda-seti.o seti_cuda-seti_header.o seti_cuda-timecvt.o seti_cuda-s_util.o seti_cuda-sah_version.o seti_cuda-worker.o seti_cuda-chirpfft.o seti_cuda-spike.o seti_cuda-progress.o seti_cuda-fft8g.o seti_cuda-gdata.o seti_cuda-autocorr.o seti_cuda-schema_master.o seti_cuda-sqlrow.o seti_cuda-sqlblob.o seti_cuda-xml_util.o   -L/Users/Tom/boinc/api -lboinc_api -L/Users/Tom/boinc/lib -lboinc  -lm      cudaAcceleration.o cudaAcc_CalcChirpData.o cudaAcc_fft.o cudaAcc_gaussfit.o cudaAcc_PowerSpectrum.o cudaAcc_pulsefind.o cudaAcc_summax.o cudaAcc_transpose.o cudaAcc_utilities.o cudaAcc_autocorr.o cudaAcc_analyzeReport.o -lpthread -L/Developer/NVIDIA/CUDA-6.5/lib64 -lcudart -lcufft -I/Developer/NVIDIA/CUDA-6.5/include
ld: warning: directory not found for option '-L/Developer/NVIDIA/CUDA-6.5/lib64'
ld: library not found for -lcudart
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: [seti_cuda] Error 1 (ignored)
g++ -DHAVE_CONFIG_H -I. -I..  -g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math  -fno-common -DPIC -DHAVE_CONFIG_H -DUSE_I386_OPTIMIZATIONS -DUSE_I386_XEON -DHALF_STRIDE -DSETI7 -DUSE_SSE41 -O3 -I/Users/Tom/sah_v7_opt/Xbranch/client/vector -I/usr/local/cuda/lib -DTEST_TIMER  -I/usr/local/include -g -O2 -g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math -DHAVE_CONFIG_H -DTEXT_UI -DNDEBUG -DCLIENT -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db -D_THREAD_SAFE   -msse2 -mfpmath=sse -DUSE_SSE -DUSE_SSE2   -MT hires_timer_test-hires_timer.o -MD -MP -MF .deps/hires_timer_test-hires_timer.Tpo -c -o hires_timer_test-hires_timer.o `test -f 'vector/hires_timer.cpp' || echo './'`vector/hires_timer.cpp
mv -f .deps/hires_timer_test-hires_timer.Tpo .deps/hires_timer_test-hires_timer.Po
/bin/sh ../libtool  --tag=CXX   --mode=link g++ -DTEST_TIMER  -I/usr/local/include -g -O2 -g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math -DHAVE_CONFIG_H -DTEXT_UI -DNDEBUG -DCLIENT -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db -D_THREAD_SAFE   -msse2 -mfpmath=sse -DUSE_SSE -DUSE_SSE2   -D_THREAD_SAFE  -mmacosx-version-min=10.8 -ldl /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -framework Carbon -Wl,-rpath,\$ORIGIN -L/usr/local/lib  -mmacosx-version-min=10.8 -ldl /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -framework Carbon  -mmacosx-version-min=10.8 -ldl /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -framework Carbon -Wl,-rpath,\$ORIGIN -o hires_timer_test hires_timer_test-hires_timer.o  -L/Users/Tom/boinc/api -lboinc_api -L/Users/Tom/boinc/lib -lboinc  -lm    
mkdir .libs
g++ -DTEST_TIMER -I/usr/local/include -g -O2 -g -O2 -I/Developer/NVIDIA/CUDA-6.5/include -ffast-math -DHAVE_CONFIG_H -DTEXT_UI -DNDEBUG -DCLIENT -I../db -I/Users/Tom/boinc -I/Users/Tom/boinc/api -I/Users/Tom/boinc/lib -I/Users/Tom/boinc/sched -I/Users/Tom/boinc/db -D_THREAD_SAFE -msse2 -mfpmath=sse -DUSE_SSE -DUSE_SSE2 -D_THREAD_SAFE -mmacosx-version-min=10.8 /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -Wl,-rpath -Wl,\$ORIGIN -mmacosx-version-min=10.8 /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -mmacosx-version-min=10.8 /usr/lib/libz.1.dylib -mtune=core2 -march=core2 -fstrict-aliasing -Wl,-rpath -Wl,\$ORIGIN -o hires_timer_test hires_timer_test-hires_timer.o -Wl,-bind_at_load  -L/usr/local/lib -ldl -L/Users/Tom/boinc/api /Users/Tom/boinc/api/.libs/libboinc_api.dylib -L/Users/Tom/boinc/lib /Users/Tom/boinc/lib/.libs/libboinc.a -lm
/bin/cp seti_cuda setiathome_x41zc_x86_64-apple-darwin_cuda65
cp: seti_cuda: No such file or directory
make[2]: [setiathome_x41zc_x86_64-apple-darwin_cuda65] Error 1 (ignored)

ld: warning: directory not found for option '-L/Developer/NVIDIA/CUDA-6.5/lib64'
ld: library not found for -lcudart
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: [seti_cuda] Error 1 (ignored)

I looked, there is Not a /Developer/NVIDIA/CUDA-6.5/lib64
???

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Success!

Well, there was one line in Both makefiles, I simply changed the line to, /Developer/NVIDIA/CUDA-6.5/lib
Now I have a file named setiathome_x41zc_x86_64-apple-darwin_cuda65 which is 9,276,944 bytes 9.3 MB in size...think it will work?
/bin/cp seti_cuda setiathome_x41zc_x86_64-apple-darwin_cuda65
strip setiathome_x41zc_x86_64-apple-darwin_cuda65
/bin/cp setiathome_x41zc_x86_64-apple-darwin_cuda65 ../
make[2]: Nothing to be done for `all-am'.
TomsMacPro:Xbranch Tom$
ID: 1747370 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747403 - Posted: 7 Dec 2015, 7:27:43 UTC - in response to Message 1747338.  

You can comment the applicable lines out for certain ( #ifdef preprocessor macro would be better, but not essential for a private build) . As soon as It cools down a bit here, and I recover from work a bit, can try again on the Mac pro and see where we're at on El Capitan
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747403 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747404 - Posted: 7 Dec 2015, 7:30:23 UTC - in response to Message 1747370.  
Last modified: 7 Dec 2015, 7:30:52 UTC

Cool, will cool down, see if I can replicate on El Capitan, then generalise the changes such they won't break Linux + Windows build systems, then we'll be on the way :D
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747404 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747412 - Posted: 7 Dec 2015, 8:42:30 UTC

Alrighty then. tracked down my Coreframework bizarre issues to a clear indication that current boinc api header file, HEAD revision, is broken somehow (at least for my Mac)

For me commenting out the boinc_api.h include in files that don't really need it allows them to build fine

What version of boinc sources are you building against ?
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747412 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1747419 - Posted: 7 Dec 2015, 9:34:31 UTC - in response to Message 1747412.  

Alrighty then. tracked down my Coreframework bizarre issues to a clear indication that current boinc api header file, HEAD revision, is broken somehow (at least for my Mac)

For me commenting out the boinc_api.h include in files that don't really need it allows them to build fine

What version of boinc sources are you building against ?

Judging by Raistmer's email to boinc_dev he ran into a similar api problem...

Should Christian pick up that ball I suggest you try liaise directly with him.
Or you reiterate the problems in a more diplomatic way and cite not working on Mac...
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1747419 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747424 - Posted: 7 Dec 2015, 10:01:27 UTC - in response to Message 1747419.  

Well I'd probably have to work out what broke first :P. Probably will just try and older one in a bit.
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747424 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747429 - Posted: 7 Dec 2015, 10:09:30 UTC - in response to Message 1747412.  
Last modified: 7 Dec 2015, 10:17:09 UTC

Alrighty then. tracked down my Coreframework bizarre issues to a clear indication that current boinc api header file, HEAD revision, is broken somehow (at least for my Mac)

For me commenting out the boinc_api.h include in files that don't really need it allows them to build fine

What version of boinc sources are you building against ?

I'm not sure how to check the version. It's the same one I installed back in Feb. At the time it was the latest version boinc-master.zip Sunday, February 22, 2015 at 7:56 PM 46.5 MB.

In other News...IT WORKS! Which is more than I can say about the BOINC Manager 7.4.36 in 10.11.1. The counters aren't working right.

So, the App passed the standalone tests and here is the first result, http://setiathome.berkeley.edu/result.php?resultid=4578205287
Here's a Shorty, http://setiathome.berkeley.edu/result.php?resultid=4578226919
Run time 2 min 57 sec
CPU time 2 min 44 sec
It's still getting 100% CPU use but....Who cares?
Pretty fast for a 750Ti.

Now how to get the counters working right...
ID: 1747429 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1747433 - Posted: 7 Dec 2015, 10:21:53 UTC - in response to Message 1747429.  
Last modified: 7 Dec 2015, 10:24:03 UTC

Looks like a mid anglerange, so most of the codepaths would have exercised. Fingers cross it holds up, and I can get el capitan to a similar stage with another boinc checkout.

Can you ldd the exe for me please ? From what I can see earlier modifications may have already injected the origin into the library search pathg (I hope, making Cuda toolkit install unnecessary for users provided they have the right driver and get the dylibs)

[edit:] erm ... otool on mac, not ldd, will lookup the syntax a bit later
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1747433 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747434 - Posted: 7 Dec 2015, 10:30:28 UTC - in response to Message 1747433.  
Last modified: 7 Dec 2015, 10:35:49 UTC

So, ldd doesn't work on this Mac.
When running as standalone I had to enter the Path into the terminal, when launching BOINC I didn't do anything. The libraries were next to the App in both cases.
TomsMacPro:~ Tom$ otool -L -D /Volumes/Mov1/BOINC/Yosemite/BOINC\ Data/projects/setiathome.berkeley.edu/setiathome_x41zc_x86_64-apple-darwin_cuda65 
/Volumes/Mov1/BOINC/Yosemite/BOINC Data/projects/setiathome.berkeley.edu/setiathome_x41zc_x86_64-apple-darwin_cuda65:
	/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 169.3.0)
	/usr/lib/libz.1.dylib (compatibility version 1.0.0, current version 1.2.5)
	/System/Library/Frameworks/Carbon.framework/Versions/A/Carbon (compatibility version 2.0.0, current version 155.0.0)
	@rpath/libcudart.6.5.dylib (compatibility version 0.0.0, current version 6.5.14)
	@rpath/libcufft.6.5.dylib (compatibility version 0.0.0, current version 6.5.14)
	/usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current version 56.0.0)
ID: 1747434 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1747545 - Posted: 7 Dec 2015, 22:42:43 UTC - in response to Message 1747433.  

After a reboot the App still launched with BOINC without any problems. Just before I left Mountain Lion I was able to compile another App using the Stock code with devices;
-gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -gencode arch=compute_32,code=sm_32 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50
I have no idea which card is a sm_20 or sm_50, but I understand a sm_52 is a 980.

The App with the 'Special' code used;
gencode arch=compute_32,code=sm_32 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50
Again, I don't have a clue which card is a sm_32 or even what a 750Ti would be.

So far the App is working about the way it did in Linux, complete with the occasional flurry of errors. Strange how a certain angle range will trigger a number of consecutive errors. It's certainly making much shorter work of those MBs than my ATI 6850s ever did.
;-)
ID: 1747545 · Report as offensive
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1747561 - Posted: 7 Dec 2015, 23:42:01 UTC - in response to Message 1747545.  

After a reboot the App still launched with BOINC without any problems. Just before I left Mountain Lion I was able to compile another App using the Stock code with devices;
-gencode arch=compute_20,code=sm_20 -gencode arch=compute_20,code=sm_21 -gencode arch=compute_20,code=compute_20 -gencode arch=compute_32,code=sm_32 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50
I have no idea which card is a sm_20 or sm_50, but I understand a sm_52 is a 980.

The App with the 'Special' code used;
gencode arch=compute_32,code=sm_32 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50
Again, I don't have a clue which card is a sm_32 or even what a 750Ti would be.

So far the App is working about the way it did in Linux, complete with the occasional flurry of errors. Strange how a certain angle range will trigger a number of consecutive errors. It's certainly making much shorter work of those MBs than my ATI 6850s ever did.
;-)



sm35=780,sm50=750ti
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1747561 · Report as offensive
Profile petri33
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1668
Credit: 623,086,772
RAC: 156
Finland
Message 1747563 - Posted: 7 Dec 2015, 23:48:59 UTC - in response to Message 1747545.  
Last modified: 8 Dec 2015, 0:14:59 UTC


So far the App is working about the way it did in Linux, complete with the occasional flurry of errors. Strange how a certain angle range will trigger a number of consecutive errors.
;-)


I have sent you an email with updated source code for testing. Lot less errors for me. Hope it helps. You may need to do the #ifdefined __APPLE__ thing again. EDIT a new email fixed that.
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1747563 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 58 · Next

Message boards : Number crunching : I've Built a Couple OSX CUDA Apps...


 
©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.