Other Parts Discussed in Thread: SYSBIOS
Tool/software: TI C/C++ Compiler
Hi， I'm running the example project that does matrix computation below omp directory supplied by TI on C6678. Here I met two problems as below:
1. In the original example source code, the dimension of the matrix is 10. As I increase the size of the matrix, for example, 128, an error happens when the program runs to enter the parallel region, that corresponds to the OpenMP instruction: #pragama omp parallel. The error displayed in the console shows:
"ti.sdo.ipc.heaps.HeapMemMP: line 628 assertion failure: A_internal: An internal error has occurred xdc.runtime.Error.raise: terminating execution"
In my project, I use the original .cfg file without any change. Could you tell me how to solve this problem? Do I need to add some pieces of script?
2. In the above condition, sometimes another error also happens which seems to be out of memory boundary in the local address space in the slave cores( accept core 0) . The prompt in the console returns:
"ti.sysbios.heaps.HeapMem: line x out of memory" I need to operate on the matrix of 256*256 integers. Could you show me how to do some alterations to make the example routine run with such amount of data on 8 cores?
Best regards
