openbsd/src/regress/lib/libc/ieeefp, branch libressl-v2.4.2 A mirror of https://github.com/libressl/openbsd.git https://git.lua4.win/openbsd/atom?h=libressl-v2.4.2 2010-05-08T19:16:33+00:00 * add #includes to pull in missing prototypes, as pointed out by gcc4 2010-05-08T19:16:33+00:00 naddy 2010-05-08T19:16:33+00:00 urn:sha1:774ac5b3a4d8ea16943c1f374c06c5e2c21a72c2 * fix CFLAGS syntax for make depend ok blambert@ Only test one condition per invocation flavour, instead of testing all 2007-10-22T21:07:10+00:00 miod 2007-10-22T21:07:10+00:00 urn:sha1:744bfb5fee0dbac6a1a2189140e8d871d8542065 of them with exceptions disabled and then only one with exceptions enabled. ok kettenis@ Try harder to trigger an exception. On architectures where floating-point 2006-05-15T14:00:22+00:00 kettenis 2006-05-15T14:00:22+00:00 urn:sha1:18756a40ff21d843e407ceffbdf5be0c1c941b86 exceptions are deferred, we might not see the exception before another floating-point instruction is executed. This is especially true on the i387 where we don't see underflow/overflow until the result is stored into memory. some -Wimplicit-function-declaration cleaning; ok millert@ 2004-12-22T00:54:39+00:00 david 2004-12-22T00:54:39+00:00 urn:sha1:114b1aac46cd445dc7c11bd8a6e5f38730542316 use sig_atomic_t instead of int 2004-10-10T03:06:42+00:00 mickey 2004-10-10T03:06:42+00:00 urn:sha1:8a16cf0bfbdc91bb8136c9cedf795be3e0496e8c Split and modify tests to prevent infinite loops on platforms with precise 2004-07-22T19:29:42+00:00 kettenis 2004-07-22T19:29:42+00:00 urn:sha1:399e1958904c4c167e3bc4f17daca5a4926f5edf floating-point exceptions like amd64. ok deraadt@, david@ check the signals for exceptions and do siginfo in the signal handler; also some style 2004-04-02T03:06:12+00:00 mickey 2004-04-02T03:06:12+00:00 urn:sha1:8fcf616fc089002b1b804fc9fe84b0c79e2b4f2e Test more ways of producing a positive infinity, and then test negative 2004-01-16T19:34:37+00:00 miod 2004-01-16T19:34:37+00:00 urn:sha1:1d7384540d665cdb91806c51a5b917ad6e152388 infinity as well to prevent entropy leak; the usual suspects still fail all tests. Add a new regression test, which checks that we handle fp overflow correctly, 2004-01-15T18:53:24+00:00 miod 2004-01-15T18:53:24+00:00 urn:sha1:b4594fd38896c2ffb37c430cb680d9a4a2c6ad0d and produce a correct infinity. Currently, this tests fails on 68060 (060sp is to blame) and 88100 processors, and maybe more. add missing includes 2003-09-02T23:52:16+00:00 david 2003-09-02T23:52:16+00:00 urn:sha1:907b17c42a136132a2e911e8c66258755cad4391 ok deraadt@ tedu@ dhartmei@ krw@