[Openmp-commits] [PATCH] D154523: [OpenMP][AMDGPU] Tracking of busy HSA queues

[Kevin Sala via Phabricator via Openmp-commits]

kevinsala added inline comments.


================
Comment at: openmp/libomptarget/plugins-nextgen/amdgpu/src/rtl.cpp:603
+  /// Returns if this queue is considered busy
+  bool isBusy() const { return NumUsers.load() > 0; }
+
----------------
ye-luo wrote:
> kevinsala wrote:
> > Do we need these three atomic operations with `memory_order_seq_cst`? Or a more relaxed memory order could be enough?
> Is this function already under the stream manager mutex when being called? I feel NumUsers doesn't need to be atomic.
Yes, as it is now (everything protected by the stream manager mutex), it seems it shouldn't be atomic.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D154523/new/

https://reviews.llvm.org/D154523