From 8c510c901e7bb2371445f68b1ffe4da3cdbca8d6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Pawe=C5=82=20Bylica?= Date: Fri, 16 Apr 2021 17:35:08 +0200 Subject: [PATCH] test: Disable failing ui32_to_f64/min floattest This test fails in i386 + SSE2 builds by GCC. See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100119. --- test/testfloat/CMakeLists.txt | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/test/testfloat/CMakeLists.txt b/test/testfloat/CMakeLists.txt index c9ed02c24..6cb75c7df 100644 --- a/test/testfloat/CMakeLists.txt +++ b/test/testfloat/CMakeLists.txt @@ -30,6 +30,11 @@ if(TESTFLOAT_GEN) list(APPEND IGNORE_LIST ui64_to_f64/min) endif() + if(CMAKE_SYSTEM_PROCESSOR STREQUAL x86_64 AND CMAKE_SIZEOF_VOID_P EQUAL 4 AND CMAKE_CXX_COMPILER_ID MATCHES GNU) + # TODO: GCC 32-bit i386 build produces -0 for input 0, https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100119. + list(APPEND IGNORE_LIST ui32_to_f64/min) + endif() + set(ROUNDING_MODES near_even minMag min max) # Arithmetic functions: sensitive to rounding and NaNs.