You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Before we go down a rabbit hole, are buildProgram and createKernel parallelized, where for the same context, we can run several kernel compilations concurrently and see benefits from parallel execution under-the-hood? We've been working on cutting down our load times, and while we achieved much via preloading our kernels, we still have a few load-time kernels where compilation introduces a noticeable bottleneck.
If cl compilation is believed to work but not well-tested, happy to contribute to the test suite.
Frankly, I have never tested that and that would depend not on nodejs or node-opencl but on the native compiler. According to spec these methods are not thread safe but if your kernels are independent, it may work (?)
--mike
From: lmeyerov
Sent: Sunday, February 5, 20:48
Subject: [mikeseven/node-opencl] Parallel execution of concurrent kernel compilation calls? (#54)
To: mikeseven/node-opencl
Cc: Bourges-sevenier, Mikael, Mention
Before we go down a rabbit hole, are buildProgram and createKernel parallelized, where for the same context, we can run several kernel compilations concurrently and see benefits from parallel execution under-the-hood? We've been working on cutting down our load times, and while we achieved much via preloading our kernels, we still have a few load-time kernels where compilation introduces a noticeable bottleneck.
If cl compilation is believed to work but not well-tested, happy to contribute to the test suite.
(Been awhile, @<https://github.com/mikeseven>mikeseven<https://github.com/mikeseven> !)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub<#54>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AAxYLIBJ__86T0_f4mI-FbDOy9Gm5gDrks5rZqYPgaJpZM4L3xg6>.
Before we go down a rabbit hole, are
buildProgram
andcreateKernel
parallelized, where for the same context, we can run several kernel compilations concurrently and see benefits from parallel execution under-the-hood? We've been working on cutting down our load times, and while we achieved much via preloading our kernels, we still have a few load-time kernels where compilation introduces a noticeable bottleneck.If cl compilation is believed to work but not well-tested, happy to contribute to the test suite.
(Been awhile, @mikeseven !)
The text was updated successfully, but these errors were encountered: