Hi !
I have several memory leaks which are detected with the valgrind (version 3.2.3) memcheck tool with --error-limit=no --leak-check=full --show-reachable=yes options.
I am running a software under a industrial framework and, for the time being, because of memory leaks that were already reported in June 2008 by mmetz_nv (see http://forums.nvidia.com/index.php?showtopic=68958) and said to be closed under CUDA 2.0, I am unable to make it exploitable.
Similar results are obtained with Quadro NVS 290 and Tesla C870 GPUs.
CUDA Driver is x86_64-180.22 and CUDA SDK is 2.02.0807.1535 under Linux Fedora 8 kernel 2.6.26.8-57.
Please help me find an explanation and tell me if there is something to do about it.
Here are the valgrind output.
[codebox]==8455== malloc/free: in use at exit: 363,185 bytes in 184 blocks.
==8455== malloc/free: 70,565 allocs, 70,381 frees, 13,138,060 bytes allocated.
==8455== For counts of detected errors, rerun with: -v
==8455== searching for pointers to 184 not-freed blocks.
==8455== checked 919,032 bytes.
==8455==
==8455==
==8455== 4 bytes in 1 blocks are still reachable in loss record 1 of 49
==8455== at 0x4A06205: operator new(unsigned long) (vg_replace_malloc.c:167)
==8455== by 0x413A62: cutGetCmdLineArgumenti (in /home/fboum/NVIDIA_CUDA_SDK$
==8455== by 0x40F404: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release$
==8455==
==8455==
==8455== 24 bytes in 1 blocks are still reachable in loss record 2 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC7D0A: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB42D6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2699: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/b$
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release$
==8455==
==8455==
==8455== 24 bytes in 1 blocks are still reachable in loss record 3 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC6682: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455==
==8455==
==8455== 32 bytes in 1 blocks are still reachable in loss record 4 of 49
==8455== at 0x4A04D1F: calloc (vg_replace_malloc.c:279)
==8455== by 0x3CCE40158A: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C32FFE: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.$
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release$
==8455==
==8455==
==8455== 39 bytes in 2 blocks are still reachable in loss record 5 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x3CCC40A3C5: _dl_new_object (in /lib64/ld-2.7.so)
==8455== by 0x3CCC405C4B: _dl_map_object_from_fd (in /lib64/ld-2.7.so)
==8455== by 0x3CCC407F02: _dl_map_object (in /lib64/ld-2.7.so)
==8455== by 0x3CCC41118C: dl_open_worker (in /lib64/ld-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCC410B7B: _dl_open (in /lib64/ld-2.7.so)
==8455== by 0x3CCE400FB9: dlopen_doit (in /lib64/libdl-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCE40152C: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455==
==8455==
==8455== 39 bytes in 2 blocks are still reachable in loss record 6 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x3CCC408385: _dl_map_object (in /lib64/ld-2.7.so)
==8455== by 0x3CCC41118C: dl_open_worker (in /lib64/ld-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCC410B7B: _dl_open (in /lib64/ld-2.7.so)
==8455== by 0x3CCE400FB9: dlopen_doit (in /lib64/libdl-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCE40152C: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C32FFE: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.$
==8455==
==8455==
==8455== 40 bytes in 1 blocks are still reachable in loss record 7 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x3CCC40BC7D: _dl_map_object_deps (in /lib64/ld-2.7.so)
==8455== by 0x3CCC4111EC: dl_open_worker (in /lib64/ld-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCC410B7B: _dl_open (in /lib64/ld-2.7.so)
==8455== by 0x3CCE400FB9: dlopen_doit (in /lib64/libdl-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCE40152C: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C32FFE: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.$
==8455==
==8455==
==8455== 44 bytes in 2 blocks are still reachable in loss record 8 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC5530: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB168: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC5C5A: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC63A6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD924: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 48 bytes in 1 blocks are still reachable in loss record 9 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB7284: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2912: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 48 bytes in 1 blocks are still reachable in loss record 10 of 49
==8455== at 0x4A06205: operator new(unsigned long) (vg_replace_malloc.c:167)
==8455== by 0x41C54F: std::_Rb_tree<std::string, std::pair<std::string const, std::string>, std::_Select1st<std::pair<std::string const, std::string> >, $
==8455== by 0x41C7B0: std::_Rb_tree<std::string, std::pair<std::string const, std::string>, std::_Select1st<std::pair<std::string const, std::string> >, $
==8455== by 0x41CA66: std::_Rb_tree<std::string, std::pair<std::string const, std::string>, std::_Select1st<std::pair<std::string const, std::string> >, $
==8455== by 0x41BE1D: CmdArgReader::createArgsMaps(int, char const**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x41C4B3: CmdArgReader::init(int, char const**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x41389D: cutGetCmdLineArgumenti (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F404: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 56 bytes in 1 blocks are still reachable in loss record 11 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB7099: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB72CE: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2912: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 56 bytes in 1 blocks are still reachable in loss record 12 of 49
==8455== at 0x4A06205: operator new(unsigned long) (vg_replace_malloc.c:167)
==8455== by 0x4155CF: std::_Rb_tree<std::string, std::pair<std::string const, std::pair<std::type_info const*, void*> >, std::_Select1st<std::pair<std::s$
==8455== by 0x415800: std::_Rb_tree<std::string, std::pair<std::string const, std::pair<std::type_info const*, void*> >, std::_Select1st<std::pair<std::s$
==8455== by 0x415AB6: std::_Rb_tree<std::string, std::pair<std::string const, std::pair<std::type_info const*, void*> >, std::_Select1st<std::pair<std::s$
==8455== by 0x413F17: cutGetCmdLineArgumenti (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F404: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 72 bytes in 1 blocks are still reachable in loss record 13 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAD12E: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC5789: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD919: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 88 bytes in 3 blocks are possibly lost in loss record 14 of 49
==8455== at 0x4A06205: operator new(unsigned long) (vg_replace_malloc.c:167)
==8455== by 0x325A49B8A0: std::string::_Rep::_S_create(unsigned long, unsigned long, std::allocator const&) (in /usr/lib64/libstdc++.so.6.0.8)
==8455== by 0x325A49C5F4: (within /usr/lib64/libstdc++.so.6.0.8)
==8455== by 0x325A49C727: std::string::string(std::string const&, unsigned long, unsigned long) (in /usr/lib64/libstdc++.so.6.0.8)
==8455== by 0x41BCB2: CmdArgReader::createArgsMaps(int, char const**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x41C4B3: CmdArgReader::init(int, char const**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x41389D: cutGetCmdLineArgumenti (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F404: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 112 bytes in 1 blocks are still reachable in loss record 15 of 49
==8455== at 0x4A06205: operator new(unsigned long) (vg_replace_malloc.c:167)
==8455== by 0x41C435: CmdArgReader::init(int, char const**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x41389D: cutGetCmdLineArgumenti (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F404: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 112 bytes in 2 blocks are still reachable in loss record 16 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC5988: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC63A6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD924: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 120 bytes in 1 blocks are still reachable in loss record 17 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x3CCC40BE0E: _dl_map_object_deps (in /lib64/ld-2.7.so)
==8455== by 0x3CCC4111EC: dl_open_worker (in /lib64/ld-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCC410B7B: _dl_open (in /lib64/ld-2.7.so)
==8455== by 0x3CCE400FB9: dlopen_doit (in /lib64/libdl-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCE40152C: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C32FFE: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455==
==8455==
==8455== 128 bytes in 2 blocks are still reachable in loss record 18 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAF00A: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC7D7D: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB42D6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2699: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 136 bytes in 1 blocks are still reachable in loss record 19 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB423B: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2699: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 144 bytes in 3 blocks are still reachable in loss record 20 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAD042: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC7724: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB307: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC67D5: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 154 bytes in 8 blocks are still reachable in loss record 21 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x3CCDC79301: strdup (in /lib64/libc-2.7.so)
==8455== by 0x4FA4288: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 168 bytes in 1 blocks are still reachable in loss record 22 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x52425A3: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x52426B4: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x52385B8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x5204548: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x51FE8D8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAACB1: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBA4EA: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA293D: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455==
==8455==
==8455== 192 bytes in 6 blocks are still reachable in loss record 23 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FADB2F: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA60C9: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5F66: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 208 bytes in 1 blocks are still reachable in loss record 24 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC6597: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 208 bytes in 1 blocks are still reachable in loss record 25 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FD1539: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC57B3: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD919: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 216 bytes in 3 blocks are still reachable in loss record 26 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAD04F: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB83F: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC67E8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 288 bytes in 4 blocks are still reachable in loss record 27 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB8EBD: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC7571: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB2E8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC67D5: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 288 bytes in 2 blocks are still reachable in loss record 28 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC74C9: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB2E8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC67D5: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 312 bytes in 2 blocks are still reachable in loss record 29 of 49
==8455== at 0x4A04D1F: calloc (vg_replace_malloc.c:279)
==8455== by 0x3CCC40ECE5: _dl_check_map_versions (in /lib64/ld-2.7.so)
==8455== by 0x3CCC411488: dl_open_worker (in /lib64/ld-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCC410B7B: _dl_open (in /lib64/ld-2.7.so)
==8455== by 0x3CCE400FB9: dlopen_doit (in /lib64/libdl-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCE40152C: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C32FFE: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455==
==8455==
==8455== 372 bytes in 2 blocks are still reachable in loss record 30 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC5483: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB168: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC5C5A: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC63A6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD924: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 392 bytes in 7 blocks are still reachable in loss record 31 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FCE530: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCFA6E: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC7EC2: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB42D6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2699: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 408 bytes in 6 blocks are still reachable in loss record 32 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC463B: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4159: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 544 bytes in 4 blocks are still reachable in loss record 33 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAAFCB: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 600 bytes in 15 blocks are still reachable in loss record 34 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAD416: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAD788: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAC58D: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA3A05: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 624 bytes in 13 blocks are still reachable in loss record 35 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB8C0B: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC6F9B: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA6028: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9E36B: cuMemAlloc (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C18A7A: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D2B8: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 832 bytes in 4 blocks are still reachable in loss record 36 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FD2F34: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FD3710: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC5FE5: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD924: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 1,368 bytes in 19 blocks are still reachable in loss record 37 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FACFA1: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA3965: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4266: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 1,800 bytes in 9 blocks are still reachable in loss record 38 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC6E90: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FBB7C8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC67E8: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2567: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 2,309 bytes in 2 blocks are still reachable in loss record 39 of 49
==8455== at 0x4A04D1F: calloc (vg_replace_malloc.c:279)
==8455== by 0x3CCC40A15B: _dl_new_object (in /lib64/ld-2.7.so)
==8455== by 0x3CCC405C4B: _dl_map_object_from_fd (in /lib64/ld-2.7.so)
==8455== by 0x3CCC407F02: _dl_map_object (in /lib64/ld-2.7.so)
==8455== by 0x3CCC41118C: dl_open_worker (in /lib64/ld-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCC410B7B: _dl_open (in /lib64/ld-2.7.so)
==8455== by 0x3CCE400FB9: dlopen_doit (in /lib64/libdl-2.7.so)
==8455== by 0x3CCC40D305: _dl_catch_error (in /lib64/ld-2.7.so)
==8455== by 0x3CCE40152C: _dlerror_run (in /lib64/libdl-2.7.so)
==8455== by 0x3CCE400F30: dlopen@@GLIBC_2.2.5 (in /lib64/libdl-2.7.so)
==8455== by 0x4C1140D: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455==
==8455==
==8455== 2,728 bytes in 8 blocks are still reachable in loss record 40 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FA3D0F: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 2,728 bytes in 8 blocks are still reachable in loss record 41 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FA3CD5: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 6,208 bytes in 2 blocks are still reachable in loss record 42 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FC5957: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC63A6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD924: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 8,192 bytes in 1 blocks are still reachable in loss record 43 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAD14E: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC5789: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FCD919: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA4794: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9D7CF: cuInit (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C33014: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C14343: cudaGetDeviceCount (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40F3AE: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 16,456 bytes in 1 blocks are still reachable in loss record 44 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB49B7: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FC7CC0: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB42D6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2699: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 26,304 bytes in 8 blocks are still reachable in loss record 45 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FA39DE: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB5B1C: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAB017: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAA04A: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA294F: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 32,768 bytes in 1 blocks are still reachable in loss record 46 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB710B: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB72CE: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2912: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 65,536 bytes in 1 blocks are still reachable in loss record 47 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FB70C2: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FB72CE: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FA2912: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 66,736 bytes in 1 blocks are still reachable in loss record 48 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FA24A7: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9CED9: cuCtxCreate (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C36781: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36EEB: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455==
==8455== 122,880 bytes in 15 blocks are still reachable in loss record 49 of 49
==8455== at 0x4A059F6: malloc (vg_replace_malloc.c:149)
==8455== by 0x4FAD5BA: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAD788: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAC58D: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4FAAFF6: (within /usr/lib64/libcuda.so.180.22)
==8455== by 0x4F9FA2C: cuModuleLoadFatBinary (in /usr/lib64/libcuda.so.180.22)
==8455== by 0x4C3AC2E: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C36F12: (within /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x4C18933: cudaMalloc (in /usr/local/cuda/lib/libcudart.so.2.1)
==8455== by 0x40D1D7: runProg(int, char**) (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455== by 0x40F493: main (in /home/fboum/NVIDIA_CUDA_SDK/bin/linux/release/arrayScanGPU)
==8455==
==8455== LEAK SUMMARY:
==8455== definitely lost: 0 bytes in 0 blocks.
==8455== possibly lost: 88 bytes in 3 blocks.
==8455== still reachable: 363,097 bytes in 181 blocks.
==8455== suppressed: 0 bytes in 0 blocks.[/codebox]