aboutsummaryrefslogtreecommitdiff
path: root/openmp/docs
diff options
context:
space:
mode:
authorJohannes Doerfert <johannes@jdoerfert.de>2023-10-21 14:49:30 -0700
committerGitHub <noreply@github.com>2023-10-21 14:49:30 -0700
commitd3921e467005733daa8e63df4b553052c8ec72c1 (patch)
tree66e7aedcca08569165da521b337dc44cf8d35b8c /openmp/docs
parentd571af7f627491841fab7c456f774d7b8f546159 (diff)
downloadllvm-d3921e467005733daa8e63df4b553052c8ec72c1.zip
llvm-d3921e467005733daa8e63df4b553052c8ec72c1.tar.gz
llvm-d3921e467005733daa8e63df4b553052c8ec72c1.tar.bz2
[OpenMP] Basic BumpAllocator for (AMD)GPUs (#69806)
The patch contains a basic BumpAllocator for (AMD)GPUs to allow us to run more tests. The allocator implements `malloc`, both internally and externally, while we continue to default to the NVIDIA `malloc` when we target NVIDIA GPUs. Once we have smarter or customizable allocators we should consider this choice, for now, this allocator is better than none. It traps if it is out of memory, making it easy to debug. Heap size is configured via `LIBOMPTARGET_HEAP_SIZE` and defaults to 512MB. It allows to track allocation statistics via `LIBOMPTARGET_DEVICE_RTL_DEBUG=8` (together with `-fopenmp-target-debug=8`). Two tests were added, and one was enabled. This is the next step towards fixing https://github.com/llvm/llvm-project/issues/66708
Diffstat (limited to 'openmp/docs')
-rw-r--r--openmp/docs/design/Runtimes.rst1
1 files changed, 1 insertions, 0 deletions
diff --git a/openmp/docs/design/Runtimes.rst b/openmp/docs/design/Runtimes.rst
index 4c848ca..62ed757 100644
--- a/openmp/docs/design/Runtimes.rst
+++ b/openmp/docs/design/Runtimes.rst
@@ -1465,3 +1465,4 @@ debugging features are supported.
* Enable debugging assertions in the device. ``0x01``
* Enable diagnosing common problems during offloading . ``0x4``
+ * Enable device malloc statistics (amdgpu only). ``0x8``