Hi,
I'm trying to compare the cpu time when using the !$ open mp in nested loops compared to the one without it.
I'm using MVS2012 with intel composer xe.
Problem is when i disable the process openmp directives in Language the code runs fine but the !$ open mp loop takes the same cpu time compared to normal loop and slightly higher. When switching the process openMP directives to Generate parallel code i get the stack overflow. The code is attached to this post. I'll be thankful if you can guide me to how to use the open mp properly with mvs/intel composer xe.
Thanks in advance !
After enlarge the stack size to 20MB by setting "Project Properties -> Linker -> System -> Stack Reserve Size" to 20000000 the program can run without error. Detail info about stack size on Windows can be found at https://msdn.microsoft.com/en-us/library/8cxs58a6.aspx
Setting the number of bytes to allocate for each OpenMP* thread to use as the private stack for the thread can use environment variable "OMP_STACKSIZE". Detail about it can be found under the compiler user's guide and reference "Supported Environment Variables" topic.
Thanks,
Xiaoping Duan
Intel Customer Support
連結已複製
After enlarge the stack size to 20MB by setting "Project Properties -> Linker -> System -> Stack Reserve Size" to 20000000 the program can run without error. Detail info about stack size on Windows can be found at https://msdn.microsoft.com/en-us/library/8cxs58a6.aspx
Setting the number of bytes to allocate for each OpenMP* thread to use as the private stack for the thread can use environment variable "OMP_STACKSIZE". Detail about it can be found under the compiler user's guide and reference "Supported Environment Variables" topic.
Thanks,
Xiaoping Duan
Intel Customer Support