Have memory::type_of treat non-CUDA-allocated memory as a non-error #468
Assignees
Comments
eyalroz
added a commit
that referenced
this issue
Feb 18, 2023
* Added a memory::type_t enum value, `non_cuda`, for memory not allocated by CUDA * `memory::type_of()` will no longer fail for non-CUDA-allocated memory * `copy_parameters_t<N>` is not out of `detail_`. * The `endpoint_t` enum is now also out of `detail_` * Beefed up `copy_parameters_t<N>` with additional convenience methods *`copy_parameters_t<N>` now "recognizes" non-CUDA-allocated memory, treating it like CUDA host memory * Our array-to-raw-mem copy function now determines the destination address' context * Add an array-to-raw-mem copy function which takes the destination context explicitly
eyalroz
added a commit
that referenced
this issue
Feb 18, 2023
* Added a memory::type_t enum value, `non_cuda`, for memory not allocated by CUDA * `memory::type_of()` will no longer fail for non-CUDA-allocated memory * `copy_parameters_t<N>` is not out of `detail_`. * The `endpoint_t` enum is now also out of `detail_` * Beefed up `copy_parameters_t<N>` with additional convenience methods *`copy_parameters_t<N>` now "recognizes" non-CUDA-allocated memory, treating it like CUDA host memory * Our array-to-raw-mem copy function now determines the destination address' context * Add an array-to-raw-mem copy function which takes the destination context explicitly
eyalroz
added a commit
that referenced
this issue
Mar 9, 2023
* Added a memory::type_t enum value, `non_cuda`, for memory not allocated by CUDA * `memory::type_of()` will no longer fail for non-CUDA-allocated memory * `copy_parameters_t<N>` is not out of `detail_`. * The `endpoint_t` enum is now also out of `detail_` * Beefed up `copy_parameters_t<N>` with additional convenience methods *`copy_parameters_t<N>` now "recognizes" non-CUDA-allocated memory, treating it like CUDA host memory * Our array-to-raw-mem copy function now determines the destination address' context * Add an array-to-raw-mem copy function which takes the destination context explicitly
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
If we call
cuPointerGetAttribute()withCU_POINTER_ATTRIBUTE_MEMORY_TYPE, and pass a non-CUDA-allocated pointer, we get an error status:CUDA_ERROR_INVALID_VALUE. But actually - that is a perfectly reasonable thing to do, given an arbitrary pointer and wanting to check what CUDA has to say about it.So, let's add support for doing this:
type_of()detect this error and return the non-CUDA memory type instead of passing on an error.The text was updated successfully, but these errors were encountered: