-
Notifications
You must be signed in to change notification settings - Fork 22
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
Resource Integration #89
base: develop
Are you sure you want to change the base?
Conversation
Conflicts: src/tpl/umpire tests/unit/CMakeLists.txt
…CHAI into feature/device-context-integration Conflicts: examples/context.cpp
…CHAI into feature/device-context-integration
…om/llnl/CHAI into feature/device-context-integration
|
||
camp::resources::Resource* res; | ||
if (space == chai::CPU){ | ||
res = record->m_last_resource; |
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.
It seems weird that we might ignore the resource passed in.
@@ -830,7 +830,9 @@ GPU_TEST(ManagedArray, PodTestGPU) | |||
|
|||
TEST(ManagedArray, ExternalConstructorUnowned) |
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.
@davidbeckingsale, @mdavis36, I can't get this test to pass.
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.
It also seems that using pinned memory for the CPU space conflicts with the ability to construct using a simple malloc'ed pointer (say from a 3rd party library).
Adding support for streaming context resources to work with Chai/Umpire/Camp. Current support added for host devices, cuda devices and asynchronous data migration from GPU to Host.
This new feature is demonstrated in new examples using Chai forall utility.