how do I make cuda-gdb not print a line for each kernel executed

I am running a long calculation which invokes thousands of kernels. When I run inside the debugger, cuda-gdb prints a line for each kernel executed so this takes ages. Is there any way to make cuda-gdb not print these messages?


If you are talking about messages like this:

[Launch of CUDA Kernel 1 (kernel3) on Device 0]


set cuda kernel_events 0

should turn them off. Page 12 of the cuda-gdb 3.2 guide for future reference.