On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley <a...@openjdk.org> wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable >> denormal arithmetic. This breaks Java's floating-point semantics. >> >> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522 >> >> One solution is to save and restore the floating-point control word around >> System.loadLibrary(). This isn't perfect, because some shared library might >> load another shared library at runtime, but it's a lot better than what we >> do now. >> >> However, this fix is not complete. `dlopen()` is called from many places in >> the JDK. I guess the best thing to do is find and wrap them all. I'd like to >> hear people's opinions. > > Andrew Haley has updated the pull request incrementally with one additional > commit since the last revision: > > 8295159: DSO created with -ffast-math breaks Java floating-point arithmetic Thanks a lot for the data, Andrew. I agree with you that the overhead `stmxcsr` introduces makes it almost prohibitive to be unconditionally placed on the hot path on the way back from a JNI call. In contrast, your idea to detect the corruption by checking the result of a carefully chosen FP expression has very modest impact while still being able to catch important types of MXCSR corruption. I fully support having it turned on by default for JNI calls. ------------- PR: https://git.openjdk.org/jdk/pull/10661
