cuda-gdb

cuda-gdb exits with “[1] stopped” when it hits a kernel call

由 若如初见. 提交于 2020-01-17 04:33:07
问题 I'm pretty new to CUDA and flying a bit by the seat of my pants here... I'm trying to debug my CUDA program on a remote machine I don't have admin rights on. I compile my program with nvcc -g -G and then try to debug it with cuda-gdb. However, as soon as gdb hits a call to a kernel (doesn't even have to enter it, and it doesn't happen in host code), I get: (cuda-gdb) run Starting program: /path/to/my/binary/cuda_clustered_tree [Thread debugging using libthread_db enabled] [1]+ Stopped cuda

Cuda-gdb not stopping at breakpoints inside kernel

由 元气小坏坏 提交于 2020-01-03 04:39:19
问题 Cuda-gdb was obeying all the breakpoints I would set, before adding '-arch sm_20' flag while compiling. I had to add this to avoid error being thrown : 'atomicAdd is undefined' (as pointed here). Here is my current statement to compile the code: nvcc -g -G --maxrregcount=32 Main.cu -o SW_exe (..including header files...) -arch sm_20 and when I set a breakpoint inside kernel, cuda-gdb stops once at the last line of the kernel, and then the program continues. (cuda-gdb) b SW_kernel_1.cu:49

CUDA cross-compiling error! cuda-gdb error message RSEG1058

由 混江龙づ霸主 提交于 2019-12-24 12:22:58
问题 I'm new in this CUDA and parallel computing staff and I have a recent problem. I have an Ubuntu 12.04 system which is host and Jetson TK1 as the target. I'm using Nsight Eclipse to write, edit and compile the algorithms. I'm using SSH protocol to connect TK1 from the host computer. When I try some file transfer operations it's okay always. However, sometimes the cross-compiling process sucks and errors "Connect to ubuntu@192.168.0.94 was cancelled" having title "RSEG1058". I don't understand

cuda-gdb not working in nsight on linux

由 亡梦爱人 提交于 2019-12-24 08:31:30
问题 I installed cuda 5 in my Ubuntu 12.10 and it is working well, I can compile and debug through cuda-gdb in terminal as well. I'm trying to use nsight, it compiles and executes my code with no issues, but when I try to debug I got the following error Error in final launch sequence Failed to execute MI command: -gdb-set cuda api_failures ignore Error message from debugger back end: Undefined set cuda command: "api_failures ignore". Try "help set cuda". Undefined set cuda command: "api_failures

nsight eclipse remote debugging timed out error

由 瘦欲@ 提交于 2019-12-20 06:24:55
问题 I have a Server running CentOS 6.0 and I'm trying to use it as a remote host for cuda debugging. In order to do this, I installed cuda-toolkit 5.5 both on the server and my notebook, which is running ubuntu 12.10 OS. I configured the two machines as the NVIDIA-cuda-instruction told me, yet when I started the Nsight eclipse edition and tried to remote debug my cuda applications, I ran into error, whcih says: Failed to execute MI command: -target-select remote 192.168.2.105:2345 Error message

cuda-gdb Error message

由 坚强是说给别人听的谎言 提交于 2019-12-20 06:01:17
问题 I tried to debug my CUDA application with cuda-gdb but got some weird error. I set option -g -G -O0 to build my application. I could run my program without cuda-gdb, but didn't get correct result. Hence I decided to use cuda-gdb, however, I got following error message while running program with cuda-gdb Error: Failed to read the valid warps mask (dev=1, sm=0, error=16). What does it means? Why sm=0 and what's the meaning of error=16 ? Update 1 : I tried to use cuda-gdb to CUDA samples, but it

How to do remote debugging with Nsight 5.5 on Linux?

由 為{幸葍}努か 提交于 2019-12-13 03:39:19
问题 Disclaimer: I know that this question has been asked numerous times, but before Nsight 5.5 the answer simply was "You can't." I'm trying to debug a CUDA program with Nsight 5.5 on Ubuntu. The remote machine is also an Ubuntu with CUDA Toolkit 5.5. I setup debugging in NSight using the "Debug an application on a remote system". Whenever I try to run the application I get a "Could not start gdbserver on the remote host" error. This is what I get on the console: Last login: Thu Aug 1 16:09:26

nsight eclipse for mac lion debug error

由 坚强是说给别人听的谎言 提交于 2019-12-12 03:39:14
问题 I am using nsight eclipse edition for mac to debug my cuda code. I was able to run cuda-gdb all fine from a terminal but was not able to use the debug function in Nsight. Whenever I started to debug, an error window pops up saying "Error in final launch sequence Connection is shut down". My OS is Mac Lion (10.7.4). I understand the big thing in Cuda 5.0 is the interactive debugging with eclipse but sadly I won't be able to use it unless I switch to Windows or Linux. If anybody has seen this

CUDA Constant Memory Error

由 隐身守侯 提交于 2019-12-11 10:55:39
问题 I am trying to do a sample code with constant memory with CUDA 5.5. I have 2 constant arrays of size 3000 each. I have another global array X of size N. I want to compute Y[tid] = X[tid]*A[tid%3000] + B[tid%3000] Here is the code. #include <iostream> #include <stdio.h> using namespace std; #include <cuda.h> __device__ __constant__ int A[3000]; __device__ __constant__ int B[3000]; __global__ void kernel( int *dc_A, int *dc_B, int *X, int *out, int N) { int tid = threadIdx.x + blockIdx.x

nsight eclipse remote debugging timed out error

由 删除回忆录丶 提交于 2019-12-02 12:29:27
I have a Server running CentOS 6.0 and I'm trying to use it as a remote host for cuda debugging. In order to do this, I installed cuda-toolkit 5.5 both on the server and my notebook, which is running ubuntu 12.10 OS. I configured the two machines as the NVIDIA-cuda-instruction told me, yet when I started the Nsight eclipse edition and tried to remote debug my cuda applications, I ran into error, whcih says: Failed to execute MI command: -target-select remote 192.168.2.105:2345 Error message from debugger back end: 192.168.2.105:2345: Connection timed out I googled this error, someones says it
订阅 cuda-gdb