When I debug, i always get this problem. I have set CUDA debug focus. Anyone have any idea? Thanks a lot.
My environment: Windows server 2008, Parallel Nsight 1.51,CUDA SDK 3.2, toolkit 3.2
Summary of access violations:
Parallel Nsight Debug
Memory Checker detected 96 access violations!
error = access violation on load
blockIdx = {4,0,0}
threadIdx = {0,0,0}
address = 0x00002000
accessSize = 4
See Output view for details.
Parallel Nsight Debug
CUDA Memory Checker detected 32 threads caused an access violation:
Launch Parameters
CUcontext = 60c610a0
CUstream = 00000000
CUmodule = 61e51c50
CUfunction = 61e52c30
FunctionName = Z11MurmurHash2PKcPiS1_PjiS2
gridDim = {24,1,1}
blockDim = {512,1,1}
sharedSize = 2176
Parameters:
Parameters (raw):
0x00000000 0x05100000 0x05100200 0x05200000
0x00002ffd 0x05100400
GPU State:
Address Size Type Block Thread blockIdx threadIdx PC Source
00000000 4 adr ld 0 0 {0,0,0} {0,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000004 4 adr ld 0 1 {0,0,0} {1,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000008 4 adr ld 0 2 {0,0,0} {2,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000000c 4 adr ld 0 3 {0,0,0} {3,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000010 4 adr ld 0 4 {0,0,0} {4,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000014 4 adr ld 0 5 {0,0,0} {5,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000018 4 adr ld 0 6 {0,0,0} {6,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000001c 4 adr ld 0 7 {0,0,0} {7,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000020 4 adr ld 0 8 {0,0,0} {8,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000024 4 adr ld 0 9 {0,0,0} {9,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000028 4 adr ld 0 10 {0,0,0} {10,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000002c 4 adr ld 0 11 {0,0,0} {11,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000030 4 adr ld 0 12 {0,0,0} {12,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000034 4 adr ld 0 13 {0,0,0} {13,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000038 4 adr ld 0 14 {0,0,0} {14,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000003c 4 adr ld 0 15 {0,0,0} {15,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000040 4 adr ld 0 16 {0,0,0} {16,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000044 4 adr ld 0 17 {0,0,0} {17,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000048 4 adr ld 0 18 {0,0,0} {18,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000004c 4 adr ld 0 19 {0,0,0} {19,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000050 4 adr ld 0 20 {0,0,0} {20,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000054 4 adr ld 0 21 {0,0,0} {21,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000058 4 adr ld 0 22 {0,0,0} {22,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000005c 4 adr ld 0 23 {0,0,0} {23,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000060 4 adr ld 0 24 {0,0,0} {24,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000064 4 adr ld 0 25 {0,0,0} {25,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000068 4 adr ld 0 26 {0,0,0} {26,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000006c 4 adr ld 0 27 {0,0,0} {27,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000070 4 adr ld 0 28 {0,0,0} {28,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000074 4 adr ld 0 29 {0,0,0} {29,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
00000078 4 adr ld 0 30 {0,0,0} {30,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
0000007c 4 adr ld 0 31 {0,0,0} {31,0,0} 000180 g:\cppws\bytegpu\matrix multiply\murmur.cu:126
Summary of access violations:
Parallel Nsight Debug
Memory Checker detected 32 access violations!
error = access violation on load
blockIdx = {0,0,0}
threadIdx = {0,0,0}
address = 0x00000000
accessSize = 4
See Output view for details.
Parallel Nsight Debug
CUDA grid launch failed: CUcontext: 1623593120 CUmodule: 1642404944 Function: Z11MurmurHash2PKcPiS1_PjiS2