summaryrefslogtreecommitdiffstats
path: root/Modules/Compiler/XL-CXX.cmake
diff options
context:
space:
mode:
authorChuck Atkins <chuck.atkins@kitware.com>2017-04-27 14:12:18 (GMT)
committerChuck Atkins <chuck.atkins@kitware.com>2017-04-27 17:29:28 (GMT)
commitf320907b3d288930c80411a8e97f63eef0e8ec8b (patch)
tree1ad829d3c8b3c27886aaeb2d89198f6f8c647bd1 /Modules/Compiler/XL-CXX.cmake
parentf0ae821681a0a0b4f411a0feb4a6f6e458475999 (diff)
downloadCMake-f320907b3d288930c80411a8e97f63eef0e8ec8b.zip
CMake-f320907b3d288930c80411a8e97f63eef0e8ec8b.tar.gz
CMake-f320907b3d288930c80411a8e97f63eef0e8ec8b.tar.bz2
XL: Fix the use of the -qhalt flag for varying OSs
The valid settings to pass to the IBM XL compiler for the -qhalt flag vary widely by language and platform. Based on existing documentation, the following table shows which error levels are valid to pass to -qhalt= since -qhalt=e is not always available. OS | xlc | xlC | xlf | ---------|-------|------------------ AIX | iwes | iw s | ilwesu | BlueGene | iwes | iwes | ilwesu | OS X | iwesu | iwesu | ilwesu | Linux | w | w | ilwesu |
Diffstat (limited to 'Modules/Compiler/XL-CXX.cmake')
-rw-r--r--Modules/Compiler/XL-CXX.cmake3
1 files changed, 1 insertions, 2 deletions
diff --git a/Modules/Compiler/XL-CXX.cmake b/Modules/Compiler/XL-CXX.cmake
index 545d657..4fdc43e 100644
--- a/Modules/Compiler/XL-CXX.cmake
+++ b/Modules/Compiler/XL-CXX.cmake
@@ -4,8 +4,7 @@ string(APPEND CMAKE_CXX_FLAGS_RELEASE_INIT " -DNDEBUG")
string(APPEND CMAKE_CXX_FLAGS_MINSIZEREL_INIT " -DNDEBUG")
# -qthreaded = Ensures that all optimizations will be thread-safe
-# -qhalt=e = Halt on error messages (rather than just severe errors)
-string(APPEND CMAKE_CXX_FLAGS_INIT " -qthreaded -qhalt=e")
+string(APPEND CMAKE_CXX_FLAGS_INIT " -qthreaded")
set(CMAKE_CXX_COMPILE_OBJECT
"<CMAKE_CXX_COMPILER> -+ <DEFINES> <INCLUDES> <FLAGS> -o <OBJECT> -c <SOURCE>")