How can I parallelize the execution of this benchmark? (https://github.com/google/XNNPACK/blob/master/bench/spmm-benchmark.h)

google / XNNPACK

High-efficiency floating-point neural network inference operators for mobile, server, and Web

Other

1.81k stars 342 forks source link

How can I parallelize the execution of this benchmark? (https://github.com/google/XNNPACK/blob/master/bench/spmm-benchmark.h) #6268

Open AnonymousYWL opened 4 months ago

fbarchard commented 3 months ago

The end2end_bench shows spmm on arm using threads.

FabianSchuetze commented 1 month ago

The end2end benchmarks benchmark a full model, or? Can I just benchmark individual gemms? Do you have an example in mind which shows how to benchmark individual gemms with different number of threads?