From 87a40a9aea2512722f70d3cbf8a4dea5f00e7952 Mon Sep 17 00:00:00 2001 From: Daniel Arndt Date: Mon, 13 May 2019 19:53:23 +0200 Subject: [PATCH] Warn about using CUDA+Threads+march=native --- doc/external-libs/cuda.html | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/doc/external-libs/cuda.html b/doc/external-libs/cuda.html index 68222e1d31..fda03f4c9b 100644 --- a/doc/external-libs/cuda.html +++ b/doc/external-libs/cuda.html @@ -72,5 +72,19 @@ will lead to segmentation faults in MPI calls.

+

+ Using CUDA in combination with architecture-specific C++ compiler flags + like -march=native is known to be fragile and there might be + compatibility issues with other libraries, e.g. using CUDA 10.1 with + -DDEAL_II_WITH_THREADS=ON and + -DDEAL_II_CXX_FLAGS=-march=native results in compile time + errors like: +

+
+        /usr/lib/gcc/x86_64-linux-gnu/7/include/avx512fintrin.h(11265): error: identifier "__builtin_ia32_scalefsd_round" is undefined
+        /usr/lib/gcc/x86_64-linux-gnu/7/include/avx512fintrin.h(11274): error: identifier "__builtin_ia32_scalefss_round" is undefined
+      
+

+ -- 2.39.5