支持多核吗？怎么支持

[sumo2017]

如题，如果要编译支持多核（多core）的算子，应该怎么做？

[lyuchuny3]

可以通过parallel()调度原语来调用： https://autokernel-docs-en.readthedocs.io/en/latest/tutorials/halide/halide_schedule.html#schedules-within-stages-domain-order

parallel (x,size) | Splits the dimension by size and parallelizes the outer dimension.

如果要控制具体的线程数，可以通后环境变量HL_NUM_THREADS:

HL_NUM_THREADS=... specifies the number of threads to create for the thread pool. When the async scheduling directive is used, more threads than this number may be required and thus allocated. A maximum of 256 threads is allowed. (By default, the number of cores on the host is used.)