Request flushing denorms to zero in OpenCL

This change adds by default the -cl-denorms-are-zero to the flags used
for kernel compilation. This is done to:
- avoid a large performance penalty on AMD Vega with ROCm (which by
  default handles denorms on GFX9 or later).
- make the defaults uniform across CUDA and OpenCL.

Fixes #2593

Change-Id: I9e6183c4367b5960e0e21f1dd342d7695acfbc44

diff --git a/docs/release-notes/2018/2018.3.rst b/docs/release-notes/2018/2018.3.rst
index 58d8bef243f97710d0b1ae7c6beb2c12763fe675..2bda2a55c8c28a8fcf5fb355b8bc9c1a4a1e7260 100644 (file)
--- a/docs/release-notes/2018/2018.3.rst
+++ b/docs/release-notes/2018/2018.3.rst
@@ -82,3 +82,12 @@ Fixes to improve portability
 
 Miscellaneous
 ^^^^^^^^^^^^^
+
+Improve OpenCL kernel performance on AMD Vega GPUs
+""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
+The OpenCL kernel optimization flags did not explicitly turn off denorm handling
+which could lead to performance loss. The optimization is now explicitly turned
+on both for consistency with CUDA and performance reasons.
+On AMD Vega GPUs (with ROCm) kernel performance improves by up to 30%.
+
+

diff --git a/src/gromacs/gpu_utils/ocl_compiler.cpp b/src/gromacs/gpu_utils/ocl_compiler.cpp
index 957c6bc45387d0832fc20d850e6a6addb47cefab..f54a94fa42dca39c6f4e2683933c4da7df178659 100644 (file)
--- a/src/gromacs/gpu_utils/ocl_compiler.cpp
+++ b/src/gromacs/gpu_utils/ocl_compiler.cpp
@@ -179,6 +179,11 @@ selectCompilerOptions(ocl_vendor_id_t deviceVendorId)
     if (getenv("GMX_OCL_DISABLE_FASTMATH") == NULL)
     {
         compilerOptions += " -cl-fast-relaxed-math";
+
+        // Hint to the compiler that it can flush denorms to zero.
+        // In CUDA this is triggered by the -use_fast_math flag, equivalent with
+        // -cl-fast-relaxed-math, hence the inclusion on the conditional block.
+        compilerOptions += " -cl-denorms-are-zero";
     }
 
     if ((deviceVendorId == OCL_VENDOR_NVIDIA) && getenv("GMX_OCL_VERBOSE"))