Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Use GPUArrays caching allocator #2593

Draft
wants to merge 5 commits into
base: master
Choose a base branch
from

Conversation

pxl-th
Copy link
Member

@pxl-th pxl-th commented Dec 15, 2024

@maleadt maleadt added cuda array Stuff about CuArray. performance How fast can we go? labels Dec 16, 2024
@maleadt maleadt force-pushed the master branch 15 times, most recently from 5d585c4 to c850163 Compare December 20, 2024 08:18
src/array.jl Outdated
Comment on lines 82 to 84
# Do not use caching allocator if it is not set or
# the buffer is not a device memory.
return if !(M <: DeviceMemory) || name == :none
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I take it you're only caching device memory because the cache doesn't differentiate between memory types and could thus return the wrong array? Would it be possible to resolve that by including the array type in the key, or do you want to be able to alias e.g. a [1, 100] and [10, 10] array?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cuda array Stuff about CuArray. performance How fast can we go?
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants