-
Notifications
You must be signed in to change notification settings - Fork 218
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
Avoid capturing AbstractArray
s in BoundsError
#2314
Conversation
This quirk allows bounds checking of `MArray`s on device in more cases. This is a workaround to address <JuliaGPU/CUDA.jl#2313>. This change was proposed upstream here <JuliaGPU/CUDA.jl#2314> and is no longer needed if accepted upstream.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you instead change
Line 40 in 7f725c0
@device_override @noinline Base.throw_boundserror(A, I) = |
to be @inline
and move the error report to a second function?
Is this what you are thinking? Feel free to tweek it. |
@maleadt are you okay with that extra call frame? And should we add the Static array test case? |
Sure. I would just call it
I don't think so, because I hope we can get it working under |
This avoid capturing `MArray`s in `BoundsError` to allow bounds checking of `MArray`s on device in more cases. This is a workaround to address <JuliaGPU#2313>.
MArray
s in BoundsError
AbstractArray
s in BoundsError
This quirk allows bounds checking of `MArray`s on device in more cases. This is a workaround to address <JuliaGPU/CUDA.jl#2313>. This change was proposed upstream here <JuliaGPU/CUDA.jl#2314> and is no longer needed if accepted upstream.
This quirk allows bounds checking of
MArray
s on device in more cases. This is a workaround to address #2313.