AMA: Explaining my 750 line compiler+runtime designed to GPU self-host APL

Re:AMA: Explaining my 750 line compiler+runtime designed to ... by joe_the_user on Sunday March 05, 2017 @08:15PM (#13800027)

I'm scanning the documents.

Basically, group nodes and map/operate on them.

It seem like that would be fairly simple if you are dynamically creating kernels.

-- Edit: Which implies you're doing a pretty "bare metal" interface to the GPU which still lets you do that and there's interesting story there, if I'd not totally off-base.

[ Parent | on HN ]

Re:AMA: Explaining my 750 line compiler+runtime designed to ... by arcfide on Sunday March 05, 2017 @10:52PM (#13800577)

I try to avoid going bare metal as much as I can because of the difficulty involved with working with GPUs at that layer. I try to focus on higher level concerns, and use the excellent low level work of people such as the ArrayFire developers to handle the low level stuff. I still have to understand it and work with it at some level sometimes, but for the most part I stick to my core abstractions.

[ Parent | on HN ]

Re:AMA: Explaining my 750 line compiler+runtime designed to ... by joe_the_user on Monday March 06, 2017 @06:08PM (#13806930)

So ArrayFire allows the creation of kernels/the calling of functions in parallel/ on the fly?

[ Parent | on HN ]

Re:AMA: Explaining my 750 line compiler+runtime designed to ... by arcfide on Monday March 06, 2017 @07:12PM (#13807294)

In some limited, restricted cases, yes.

[ Parent | on HN ]