lib/msun: Allow building tests with WARNS=6The only change needed is to mark a few variables as static.
Update libm tests from NetBSDI did this without a full vendor update since that would cause too manyconflicts. Since these files now almost match the NetBSD sources thenext git subtree merge shou
Update libm tests from NetBSDI did this without a full vendor update since that would cause too manyconflicts. Since these files now almost match the NetBSD sources thenext git subtree merge should work just fine.Reviewed By: lwhsuDifferential Revision: https://reviews.freebsd.org/D28797
show more ...
Fix lib/msun/test builds on platforms without 80-bit long doublesAfter d3338f3355a612cf385632291f46c5777bba8d18, the lib/msun test case'hypotl_near_underflow' would fail to compile on platforms wh
Fix lib/msun/test builds on platforms without 80-bit long doublesAfter d3338f3355a612cf385632291f46c5777bba8d18, the lib/msun test case'hypotl_near_underflow' would fail to compile on platforms where longdoubles weren't 80 bit, like on x86. Disable this particular test onsuch platforms for now.PR: 253313MFC after: 1 weekX-MFC-With: d3338f3355a612cf385632291f46c5777bba8d18
Fix incorrect hypotl(3) result with subnormal numbersThis adjusts the factor used to scale the subnormal numbers, so itbecomes the right value after adjusting its exponent. Thanks to SteveKargl f
Fix incorrect hypotl(3) result with subnormal numbersThis adjusts the factor used to scale the subnormal numbers, so itbecomes the right value after adjusting its exponent. Thanks to SteveKargl for finding the most elegant fix.Also enable the hypot tests, and add a test case for this bug.PR: 253313MFC after: 1 week
Add test case for 93fc67896550 (incorrect powf(3) result)This adds the test case to contrib/netbsd-tests/lib/libm/t_pow.c, as itis currently the only place testing pow(3) and friends.MFC after:
Add test case for 93fc67896550 (incorrect powf(3) result)This adds the test case to contrib/netbsd-tests/lib/libm/t_pow.c, as itis currently the only place testing pow(3) and friends.MFC after: 1 week
Un-XFAIL two tests with Clang > 10SVN r343917 fixed this for in-tree clang, but when building with a newerout-of-tree clang the test was still marked as XFAIL.Reviewed By: dimDifferential Revis
Un-XFAIL two tests with Clang > 10SVN r343917 fixed this for in-tree clang, but when building with a newerout-of-tree clang the test was still marked as XFAIL.Reviewed By: dimDifferential Revision: https://reviews.freebsd.org/D28390
Amend r343442, by only expecting the lib.msun.cbrt_test.cbrtl_powl andtrig_test.reduction test cases to fail, if the fixes from r343916 havenot yet been applied to the base compiler.Reported by:
Amend r343442, by only expecting the lib.msun.cbrt_test.cbrtl_powl andtrig_test.reduction test cases to fail, if the fixes from r343916 havenot yet been applied to the base compiler.Reported by: lwhsuPR: 234040Upstream PR: https://bugs.llvm.org/show_bug.cgi?id=40206MFC after: 1 week
Temporarily mark lib.msun.{cbrt_test.cbrtl_powl,trig_test.reduction}expected failure after clang700-import mergePR: 234040Reviewed by: ngie, markjApproved by: markj (mentor)Sponsored by: The F
Temporarily mark lib.msun.{cbrt_test.cbrtl_powl,trig_test.reduction}expected failure after clang700-import mergePR: 234040Reviewed by: ngie, markjApproved by: markj (mentor)Sponsored by: The FreeBSD FoundationDifferential Revision: https://reviews.freebsd.org/D18938
Oops... put the atf_tc_expect_fail in the testcase definition, not thetest suite definition
Expect the t_precision long double checks to fail on FreeBSD/i386There are some potential issues with the test (as brd@ has pointed outelsewhere) with precision, etc not being set before the test,
Expect the t_precision long double checks to fail on FreeBSD/i386There are some potential issues with the test (as brd@ has pointed outelsewhere) with precision, etc not being set before the test, but asalways, more research is required.
Merge in changes from ^/vendor/NetBSD/tests/dist@r313245Diff reduce lib/libm/t_pow.c per upstream changes
Revert t_pow@r312154This hunk hasn't been accepted yet
Diff reduce with upstream post-accepted contributions
Merge content currently under test from ^/vendor/NetBSD/tests/dist/@r312123Sponsored by: Dell EMC Isilon
Upgrade NetBSD tests to 01.11.2017_23.20 snapshotThis contains some new testcases in /usr/tests/...:- .../lib/libc- .../lib/libthr- .../lib/msun- .../sys/kernTested on: amd64, i386MFC after
Upgrade NetBSD tests to 01.11.2017_23.20 snapshotThis contains some new testcases in /usr/tests/...:- .../lib/libc- .../lib/libthr- .../lib/msun- .../sys/kernTested on: amd64, i386MFC after: 1 month
Don't expect :fmod to fail on FreeBSDSponsored by: EMC / Isilon Storage Division
Checkpoint initial integration work- Some of the lib/libc and lib/thr tests fail- lib/msun/exp_test:exp2_values now passes with clang 3.8.0The Makefiles in contrib/netbsd-tests were pruned as th
Checkpoint initial integration work- Some of the lib/libc and lib/thr tests fail- lib/msun/exp_test:exp2_values now passes with clang 3.8.0The Makefiles in contrib/netbsd-tests were pruned as they have no valueSponsored by: EMC / Isilon Storage Division
The fmodl compat shims on arm/mips/powerpc aren't completeDisable the test code for now on those architecturesMFC after: 1 weekPR: 199422
Reset errno in :scalbnf_val and :scalbnl_val before running the tests so thetested errno isn't staleThis was needed in order for the test to pass on amd64 with stable/10MFC after: 3 days
- Skip over the testcases that call cbrtl on platforms where LDBL_PREC == 53(arm, mips, powerpc). This fixes the build on these platforms, based on somead hoc tinderbox runs I did a while ago- Ski
- Skip over the testcases that call cbrtl on platforms where LDBL_PREC == 53(arm, mips, powerpc). This fixes the build on these platforms, based on somead hoc tinderbox runs I did a while ago- Skip cast the arguments to powl as long double so powl properly interpretsthose arugments at compile-time when picking the typeSponsored by: EMC / Isilon Storage Division
- Expect exp2_powers to fail on FreeBSD/i386- Expect exp2_values to fail on FreeBSD due to the small epsilonSponsored by: EMC / Isilon Storage Division
Alias isinff to isinf on FreeBSDisinf on FreeBSD automatically picks the appropriate type per math.hSponsored by: EMC / Isilon Storage Division
Reset errno to 0 before running scalbn to be sure that the tested errno isvalidSponsored by: EMC / Isilon Storage Division
Skip the long-double epsilon checks on FreeBSD/i386Sponsored by: EMC / Isilon Storage Division
Import proper fix for misc/49356 (/usr/include/atf-c/config.h) after atf-c/config.hwas removed from the buildPointyhat to: me (again, for not running make delete-old after running test builds)
12