summaryrefslogtreecommitdiff
path: root/tests
AgeCommit message (Collapse)Author
2014-03-04Make the abs() use check non-fatal for a bogus use in the LFS cleanerd.joerg
2014-03-03Remove the print of the rounding mode.dsl
It was added in case it was non-zero - which it isn't. It still isn't clear why acos(-1) gives the wrong result on some amd64 systems.
2014-03-03Define HAVE_FENV_H on architectures that support it.martin
What a mess, we need a central place for this!
2014-03-03Remove all cargo-cult #ifndef __vax__ from the tests, that just papermartin
over bugs in the vax libm.
2014-03-03Fix build for platforms w/o fenv.h.martin
Remove some bogus #ifdef __vax__ (and add a few, hopefully non-bogus, new ones).
2014-03-02Include the subtest number in any error output.dsl
Also temporarily print the rounding mode. I think that acos(-1) is ending up with the wrong sign because the test is being run with 'round towards -ve infinity' set. I think it getting set somewhere and causing this test to fail. The acos() code probably needs fixing - it shouldn't depend on the round mode like this. But first I want to know if this if the error.
2014-03-02Make cleanup routines actually work.jmmv
The only way to pass global state from the body to the cleanup is via the file system. Fixes leaks of global system resources (in all cases, given that the body does not by itself clean things up).
2014-03-02Remove unnecessary and broken cleanup routine.jmmv
2014-03-01Fix sign.joerg
2014-03-01Some of the acos() tests seem to fail on some systems.dsl
Sorting out why isn't helped by the tests not reporting the erronous value. Change the 'boilerplate' pattern used so that all the values are output. Reduce the amount of faffy red tape as well. Some of these reductions could be shared with other libm tests, but for the moment they are defined in this file. All these tests pass on my amd64 system, and when I run amd64 qemu.
2014-02-27Avoid promotion in subexpressions.joerg
2014-02-27Don't use integer downcasts, use correctly sized constants for each test.joerg
2014-02-27Use an abs wide enough to cover the types involved.joerg
2014-02-26Remove hardcoded VM_MIN_ADDRESS and instead query current value viamartin
sysctl vm.minaddress.
2014-02-21Remove excemption for i386 (PR lib/48601) - missing aliases will bemartin
added to the C code.
2014-02-18For now excempt i386 from __sync_lock_test_and_set_8 and point tomartin
PR lib/48601. Probably more architectures to follow.
2014-02-17Since clang doesn't recognize the nand variants of the sync builtins andjoerg
gcc doesn't support __has_builtin, use a plain #ifndef __clang__ check as workaround.
2014-02-17Add a link-time test for __sync_* primitives (see PR 48368) - thismartin
will allow us to notice missing functions during the build, instead of when compiling arbitrary pkgsrc stuff later.
2014-02-14ARM now works with integrated-as.joerg
2014-02-09Use compiler builtins instead of atf_arch and atf_machine.jmmv
The atf_arch and atf_machine configuration variables were removed from atf-0.19 without me realizing that some tests were querying them directly. Instead of reintroducing those variables, just rely on compiler builtins as many other tests already do. Should fix PR bin/48582.
2014-02-07Change vnode operation lookup to return the resulting vnode *vpp unlocked.hannken
Change cache_lookup() to return an unlocked vnode. Discussed on tech-kern@ Welcome to 6.99.31
2014-02-06Pass the correct buffer size.joerg
2014-02-04Handle another declaration after statement case for lint in c99 mode.njoly
Add the corresponding testcase.
2014-02-02Add a few more test values from the range that sparc64 previously wouldmartin
have got wrong.
2014-02-01Skip testcase inside its body for architectures w/o long double supportmartin
2014-02-01Print a slightly more helpfull message in case of test failuremartin
2014-01-30No unused functions if there is no long double support.joerg
2014-01-30Add test for uint64 to real long double conversion.joerg
2014-01-26Support using fenv instead of fpsetmask if HAVE_FENV is defined.matt
2014-01-26Make this real even though it isn't yet used.matt
2014-01-25Use English spelling of alignment.skrll
2014-01-21fix comment typos pointed out by uebayasiyamt
2014-01-20- fix funopen usageyamt
- some more checks - remove a bogus test case (bad_eucJP_getwc) PR/47660 (Julio Merino) - add XXX comments
2014-01-19Remove -O0 from CPPFLAGS, and instead setapb
COPTS.testlang_parse.c += -Wno-uninitialized. Without the old -O0, and without the new -Wno-uninitialized, gcc issues this bogus error: testlang_parse.y:1400:8: error: 'cmd' may be used uninitialized in this function
2014-01-19Remove "-g" from CPPFLAGS and LDFLAGS in curses tests.apb
The debug information contains embedded references to the src directory, which is bad for reproducible builds.
2014-01-19Sprinkle in some UNCONST()pgoyette
2014-01-18Include more of the test cases from RFC 1321.pgoyette
XXX cases 3, 6, and 9 are currently disabled because we don't seem to handle plain-text in other than 8-byte chunks.
2014-01-18Remove some debugging code.pgoyette
2014-01-18Free each crypto(4) session when we're done with it. (A new sessionpgoyette
is created for each test case.)
2014-01-18Fix verification to use correct buffer.pgoyette
Use test cases defined in RFC 1321
2014-01-18Add the test cases for sha1_hmacpgoyette
2014-01-18Needs some unconst.joerg
2014-01-17Additional tests for SHA1_HMAC (with test cases from RFC2202)pgoyette
2014-01-17Add additional test cases from RFC2202pgoyette
2014-01-17Use RFC 3713 vector, add comment to identify the sourcepgoyette
2014-01-17Reenable h_null (NULL_CBC) test. It works if we correctly pass a NULL iv.pgoyette
2014-01-17Clean up the helper programs. Remove the code to print the results, andpgoyette
make sure that all helpers actually check the results against the correct values.
2014-01-17Disable the arc4 and null_cbc tests for now. The swcrypto driver doespgoyette
not (yet) support them.
2014-01-16Use __arraycount and size_t as the format strings are wrong anyway.joerg
2014-01-16Since sizeof() is unsigned, use an unsigned int for the loop index.pgoyette
Fixes latest build break.