summaryrefslogtreecommitdiffstats
path: root/config
diff options
context:
space:
mode:
Diffstat (limited to 'config')
-rw-r--r--config/cmake/ConfigureChecks.cmake6
1 files changed, 5 insertions, 1 deletions
diff --git a/config/cmake/ConfigureChecks.cmake b/config/cmake/ConfigureChecks.cmake
index b2078e2..0a4d190 100644
--- a/config/cmake/ConfigureChecks.cmake
+++ b/config/cmake/ConfigureChecks.cmake
@@ -190,7 +190,7 @@ ENDIF (WINDOWS)
# something not quite correct with the actual test implementation. This affects
# the 'dt_arith' test and most likely lots of other code
# ----------------------------------------------------------------------------
-SET (H5_FP_TO_ULLONG_RIGHT_MAXIMUM "" CACHE INTERNAL "")
+#SET (H5_FP_TO_ULLONG_RIGHT_MAXIMUM "" CACHE INTERNAL "")
# ----------------------------------------------------------------------
# END of WINDOWS Hard code Values
@@ -1029,7 +1029,9 @@ H5MiscConversionTest (H5_SIZEOF_LONG_DOUBLE H5_LDOUBLE_TO_INTEGER_ACCURATE "chec
# integers except 'unsigned long long'. Other HP-UX systems are unknown
# yet. (1/8/05 - SLU)
#
+IF (NOT MSVC)
H5ConversionTests (H5_LDOUBLE_TO_INTEGER_WORKS "Checking IF converting from long double to integers works")
+ENDIF (NOT MSVC)
# -----------------------------------------------------------------------
# Set flag to indicate that the machine can handle conversion from
# integers to long double. (This flag should be set "yes" for all
@@ -1103,7 +1105,9 @@ ENDIF (H5_LLONG_TO_FP_CAST_WORKS MATCHES ^H5_LLONG_TO_FP_CAST_WORKS$)
# where the last 2 bytes of mantissa are lost when compiler tries to do
# the conversion, and Cygwin where compiler doesn't do rounding correctly.)
#
+IF (NOT MSVC)
H5ConversionTests (H5_ULLONG_TO_LDOUBLE_PRECISION "Checking IF converting unsigned long long to long double with precision")
+ENDIF (NOT MSVC)
# ----------------------------------------------------------------------
# Set the flag to indicate that the machine can handle overflow converting
# all floating-point to all integer types.