Hi folks,
I just stepped onto this:
#include <boost/date_time/posix_time/posix_time.hpp>
int main(int argc, char* argv[])
{
return EXIT_SUCCESS;
}
which produces:
nvcc -I . -I … -O2 -use_fast_math -DNODEBUG -arch=sm_20 -o bug1.o -c bug1.cu
nvcc error : ‘cudafe’ died due to signal 11 (Invalid memory reference)
nvcc error : ‘cudafe’ core dumped
I’m on CentOS 5.5 X86_64, gcc-4.1.2-48, boost-1.33.1-10, and CUDA 3.2rc:
nvcc -V
nvcc: NVIDIA ® Cuda compiler driver
Copyright © 2005-2010 NVIDIA Corporation
Built on Wed_Sep__8_17:12:45_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
just posted a bug report at https://partners.nvidia.com (NVOnline) External Media
Hi folks,
I just stepped onto this:
#include <boost/date_time/posix_time/posix_time.hpp>
int main(int argc, char* argv[])
{
return EXIT_SUCCESS;
}
which produces:
nvcc -I . -I … -O2 -use_fast_math -DNODEBUG -arch=sm_20 -o bug1.o -c bug1.cu
nvcc error : ‘cudafe’ died due to signal 11 (Invalid memory reference)
nvcc error : ‘cudafe’ core dumped
I’m on CentOS 5.5 X86_64, gcc-4.1.2-48, boost-1.33.1-10, and CUDA 3.2rc:
nvcc -V
nvcc: NVIDIA ® Cuda compiler driver
Copyright © 2005-2010 NVIDIA Corporation
Built on Wed_Sep__8_17:12:45_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
just posted a bug report at https://partners.nvidia.com (NVOnline) External Media
Hi folks,
I just stepped onto this:
#include <boost/date_time/posix_time/posix_time.hpp>
int main(int argc, char* argv[])
{
return EXIT_SUCCESS;
}
which produces:
I’m on CentOS 5.5 X86_64, gcc-4.1.2-48, boost-1.33.1-10, and CUDA 3.2rc:
just posted a bug report at https://partners.nvidia.com (NVOnline) External Image
The same happens with :
Linux janaka 2.6.34.7-61.fc13.x86_64 #1 SMP Tue Oct 19 04:06:30 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2010 NVIDIA Corporation
Built on Mon_Oct_18_11:54:53_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
Hi folks,
I just stepped onto this:
#include <boost/date_time/posix_time/posix_time.hpp>
int main(int argc, char* argv[])
{
return EXIT_SUCCESS;
}
which produces:
I’m on CentOS 5.5 X86_64, gcc-4.1.2-48, boost-1.33.1-10, and CUDA 3.2rc:
just posted a bug report at https://partners.nvidia.com (NVOnline) External Image
The same happens with :
Linux janaka 2.6.34.7-61.fc13.x86_64 #1 SMP Tue Oct 19 04:06:30 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2010 NVIDIA Corporation
Built on Mon_Oct_18_11:54:53_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
ceearem
November 7, 2010, 12:59pm
5
Hi
I actually had tons of problems with that combination (CentOS5.5+CUDA3.2RC)as well in my code. I am now waiting for the final release, because right now its hard to judge if the problems are because of bugs in CUDA or bugs in my code which just didnt result in any errors in the previous versions of CUDA.
Cheers
Ceearem
The same happens with :
Linux janaka 2.6.34.7-61.fc13.x86_64 #1 SMP Tue Oct 19 04:06:30 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2010 NVIDIA Corporation
Built on Mon_Oct_18_11:54:53_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
ceearem
November 7, 2010, 12:59pm
6
Hi
I actually had tons of problems with that combination (CentOS5.5+CUDA3.2RC)as well in my code. I am now waiting for the final release, because right now its hard to judge if the problems are because of bugs in CUDA or bugs in my code which just didnt result in any errors in the previous versions of CUDA.
Cheers
Ceearem
The same happens with :
Linux janaka 2.6.34.7-61.fc13.x86_64 #1 SMP Tue Oct 19 04:06:30 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2010 NVIDIA Corporation
Built on Mon_Oct_18_11:54:53_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
It seems like something is moving beyond… look at that ‘Bug - Fixed’
Please do not reply to this message
The following items have been modified for this Bug:
Bug Information
Requester: davide rossetti
Customer Bug ID:
NVIDIA Bug ID: 745732
Date: 10/13/2010 2:30:07 AM
Company/Division: GPU Computing
Severity: High
Priority: 3 - Low
Synopsis: CUDA 3.2RC 'cudafe' died due to signal 11 just including boost/date_time/posix_time/posix_time.hpp
Description: on CentOS 5.5, gcc-4.1.2-48, boost-1.33.1, simply including Boost posix_time.hpp makes cudafe to crash:
nvcc -I . -I … -O2 -use_fast_math -DNODEBUG -arch=sm_20 -o bug1.o -c bug1.cu
nvcc error : ‘cudafe’ died due to signal 11 (Invalid memory reference)
nvcc error : ‘cudafe’ core dumped
make: *** [bug1.o] Error 139
It seems like something is moving beyond… look at that ‘Bug - Fixed’
Please do not reply to this message
The following items have been modified for this Bug:
Bug Information
Requester: davide rossetti
Customer Bug ID:
NVIDIA Bug ID: 745732
Date: 10/13/2010 2:30:07 AM
Company/Division: GPU Computing
Severity: High
Priority: 3 - Low
Synopsis: CUDA 3.2RC 'cudafe' died due to signal 11 just including boost/date_time/posix_time/posix_time.hpp
Description: on CentOS 5.5, gcc-4.1.2-48, boost-1.33.1, simply including Boost posix_time.hpp makes cudafe to crash:
nvcc -I . -I … -O2 -use_fast_math -DNODEBUG -arch=sm_20 -o bug1.o -c bug1.cu
nvcc error : ‘cudafe’ died due to signal 11 (Invalid memory reference)
nvcc error : ‘cudafe’ core dumped
make: *** [bug1.o] Error 139
When will it be available for downloads?
When will it be available for downloads?
Hi folks,
I just stepped onto this:
#include <boost/date_time/posix_time/posix_time.hpp>
int main(int argc, char* argv[])
{
return EXIT_SUCCESS;
}
which produces:
I’m on CentOS 5.5 X86_64, gcc-4.1.2-48, boost-1.33.1-10, and CUDA 3.2rc:
just posted a bug report at https://partners.nvidia.com (NVOnline) External Image
I solved my crash problem by completely separating the CUDA code from c++ and compiling them separately with nvcc for cuda code and g++ for all my c++ code. Wrote a c function call to wrap the kernel calls. I do not see the crash anymore.
Hi folks,
I just stepped onto this:
#include <boost/date_time/posix_time/posix_time.hpp>
int main(int argc, char* argv[])
{
return EXIT_SUCCESS;
}
which produces:
I’m on CentOS 5.5 X86_64, gcc-4.1.2-48, boost-1.33.1-10, and CUDA 3.2rc:
just posted a bug report at https://partners.nvidia.com (NVOnline) External Image
I solved my crash problem by completely separating the CUDA code from c++ and compiling them separately with nvcc for cuda code and g++ for all my c++ code. Wrote a c function call to wrap the kernel calls. I do not see the crash anymore.