-
Notifications
You must be signed in to change notification settings - Fork 209
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
Host-side CUTENSOR #243
Host-side CUTENSOR #243
Conversation
Wait, how does this work? This isn't unified memory, so does CUTENSOR just do a |
It just needs to be memory pinned by CUDA iirc, not a |
Also, shouldn't we be more strict with the type signatures? e.g. |
Can do, let's see how this round of tests does. |
|
d77e727
to
7bcb2e0
Compare
Codecov Report
@@ Coverage Diff @@
## master #243 +/- ##
==========================================
+ Coverage 80.33% 80.98% +0.65%
==========================================
Files 155 155
Lines 10275 10408 +133
==========================================
+ Hits 8254 8429 +175
+ Misses 2021 1979 -42
Continue to review full report at Codecov.
|
Why would these synchronizations matter? Memory copies should be synchronizing already. |
We discussed this in private, but for the record it's because this is host-side memory. |
I'm losing my mind at the fact that tests are passing |
Anyway @maleadt do you mind reviewing this? Now that tests pass can we merge? |
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.
LGTM! Some minor nits.
🎉 |
finally done! 🚀 |
Seems to have been forgotten in #243
@maleadt @vchuravy this is the PR with problems on cyclops. This should enable us to call CUTENSOR on host memory using memory pinning, and is a port of JuliaGPU/CuArrays.jl#683 (where these tests passed, and I checked that they still do on latest CuArrays).