Named pools allows storing multiple pools and referencing them later on. This would be useful for sparing certain domains for specific tasks.
This patch also introduces a global default task pool, which effectively makes pool an optional parameter in all the task functions. This aims to simplify parallel code by not having to pass around pool. It also aims to make introducing parallelism in existing codebases easier.
This is an initial cut, any opinions and ideas for improvement are welcome!
Named pools allows storing multiple pools and referencing them later on. This would be useful for sparing certain domains for specific tasks.
This patch also introduces a global default task pool, which effectively makes pool an optional parameter in all the task functions. This aims to simplify parallel code by not having to pass around pool. It also aims to make introducing parallelism in existing codebases easier.
This is an initial cut, any opinions and ideas for improvement are welcome!