[flang][cuda] Convert cuf.sync_descriptor to runtime call (#121524)

Convert the op to a new entry point in the runtime
`CUFSyncGlobalDescriptor`

Revert "Reland '[flang] Allow to pass an async id to allocate the descriptor (#118713)' and #118733" (#121029)

This still cause issue for device runtime build.

Reland '[flang] Allow to pass an async id to allocate the descriptor (#118713)' and #118733 (#120997)

Device runtime build have been fixed. Attempt to re-land these patches
that have been approved

Reland '[flang] Allow to pass an async id to allocate the descriptor (#118713)' and #118733 (#120997)

Device runtime build have been fixed. Attempt to re-land these patches
that have been approved before.

https://github.com/llvm/llvm-project/pull/118713
https://github.com/llvm/llvm-project/pull/118733

show more ...

[flang][cuda][NFC] Fix type in CUFFreeDescriptor (#120799)

[flang][cuda][NFC] Fix typo in CUFAllocDescriptor (#120797)

Missing `r` in the function name.

Revert "[flang] Allow to pass an async id to allocate the descriptor (#118713)" (#119109)

This reverts commit 7d1c661381d36018fd105f4ad4c2d6dc45e7288b.

This commit breaks some device runtime buil

Revert "[flang] Allow to pass an async id to allocate the descriptor (#118713)" (#119109)

This reverts commit 7d1c661381d36018fd105f4ad4c2d6dc45e7288b.

This commit breaks some device runtime builds. Need time to investigate.

show more ...

[flang][CUF] fix missing header after #112188 (#118993)

Otherwise, builds with `-DFLANG_CUF_RUNTIME` hits:

```
runtime/CUDA/descriptor.cpp:44:24: error: invalid use of incomplete type 'const cla

[flang][CUF] fix missing header after #112188 (#118993)

Otherwise, builds with `-DFLANG_CUF_RUNTIME` hits:

```
runtime/CUDA/descriptor.cpp:44:24: error: invalid use of incomplete type 'const class Fortran::runtime::Descriptor'
44 | std::size_t count{src->SizeInBytes()};
```

show more ...

[flang] Allow to pass an async id to allocate the descriptor (#118713)

This is a patch in preparation for the support stream ordered memory
allocator in CUDA Fortran.

This patch adds an asynchro

[flang] Allow to pass an async id to allocate the descriptor (#118713)

This is a patch in preparation for the support stream ordered memory
allocator in CUDA Fortran.

This patch adds an asynchronous id to the AllocatableAllocate runtime
function and to Descriptor::Allocate so it can be passed down to the
registered allocator. It is up to the allocator to use this value or
not.

A follow up patch will implement that asynchronous allocator for CUDA
Fortran.

show more ...

[flang][cuda] Add function to allocate and deallocate device module variable (#109213)

This patch adds new runtime entry points that perform the simple
allocation/deallocation of module allocatable

[flang][cuda] Add function to allocate and deallocate device module variable (#109213)

This patch adds new runtime entry points that perform the simple
allocation/deallocation of module allocatable variable with cuda
attributes.
When the allocation is initiated on the host, the descriptor on the
device is synchronized. Both descriptors point to the same data on the
device.

This is the first PR of a stack.

show more ...

[flang][cuda][NFC] Disambiguate namespace with cuf dialect (#102194)

Rename namespace `Fortran::runtime::cuf` to `Fortran::runtime::cuda` to
avoid embiguity with the namespace `::cuf` that is defin

[flang][cuda][NFC] Disambiguate namespace with cuf dialect (#102194)

Rename namespace `Fortran::runtime::cuf` to `Fortran::runtime::cuda` to
avoid embiguity with the namespace `::cuf` that is defined in the CUF
dialect.

show more ...

[flang][cuda] Allocate local descriptor in managed memory (#102060)

This patch adds entry point in the runtime to be able to allocate
descriptors in managed memory. These entry points currently onl

[flang][cuda] Allocate local descriptor in managed memory (#102060)

This patch adds entry point in the runtime to be able to allocate
descriptors in managed memory. These entry points currently only call
`CUFAllocManaged` and `CUFFreeManaged` but could be more complicated in
the future.

`cuf.alloc` and `cuf.free` related to local descriptors are converted
into runtime calls.

show more ...