summaryrefslogtreecommitdiffstats
path: root/lib/msun
Commit message (Collapse)AuthorAgeFilesLines
* MFC r312213,r313713:ngie2017-02-211-2/+20
| | | | | | | | | | | | | | | | | | | | | | r312213: Turn COMPILER_VERSION/COMPILER_TYPE make check into a compile-time check of the clang version This works around breakage on ^/stable/10 when running installworld from a ^/stable/10 host where the test wouldn't be compiled on the first go-around and would be missing when make installworld is run. PR: 208703 r313713: Handle clang 4.x+ with the compile-time exception added in r312213 It also fails the assertions noted in bug 208703 PR: 208703 PR: 217084
* MFC r311925,r311968,r311969,r312008:ngie2017-02-101-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | r311925: Import testcase updates with code contributed back to NetBSD This also (inadvertently) contains an update to contrib/netbsd-tests/lib/libc/sys/t_wait.c (new testcases). In collaboration with: christos@NetBSD.org r311968: Fix lib/libc/sys/access_test after r311925 sys/param.h needs to be #included in order for __FreeBSD_version to be checked r311969: Remove __HAVE_LONG_DOUBLE #define from t_strtod.c and place it in Makefile This is to enable support in other testcases Inspired by lib/msun/tests/Makefile . r312008: Upgrade NetBSD tests to 01.11.2017_23.20 snapshot This contains some new testcases in /usr/tests/...: - .../lib/libc - .../lib/libthr - .../lib/msun - .../sys/kern Tested on: amd64, i386
* MFC r305358,r305449,r305451,r306367,r306397,r309474:ngie2017-02-101-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This also contains a merge of ^/projects/netbsd-tests-update-12@r304035 . This change never hit ^/head because bin/cat's behavior was changed (on ^/head) to match NetBSD. PR: 210607 r305358: Update contrib/netbsd-tests with new content from NetBSD This updates the snapshot from 09/30/2014 to 08/11/2016 This brings in a number of new testcases from upstream, most notably: - bin/cat - lib/libc - lib/msun - lib/libthr - usr.bin/sort lib/libc/tests/stdio/open_memstream_test.c was moved to lib/libc/tests/stdio/open_memstream2_test.c to accomodate the new open_memstream test from NetBSD. Tested on: amd64 (VMware fusion VM; various bare metal platforms); i386 (VMware fusion VM); make tinderbox r305449: Install h_db to unbreak some of the lib/libc/db testcases after r305358 r305451: Fix lib/libc/rpc test assumptions added in r305358 - Require root in the tcp/udp subtests (it's needed on FreeBSD when registering services). - Skip the tests if service registration fails. r306367 (by br): Allow up to 6 arguments only on MIPS. r306397 (by br): Use right piece of code for FreeBSD. r309474: Don't build :strvis_locale if VIS_NOLOCALE is undefined The copy of contrib/libc-vis on ^/stable/10 doesn't contain all of the features in the ^/stable/11 // ^/head version, including VIS_NOLOCALE. The risk is lower in conditionally running the test instead of backporting the newer version of libc-vis
* MFC r289172,r290254:ngie2017-02-091-4/+0
| | | | | | | | | | | | | | | | r289172: Refactor the test/ Makefiles after recent changes to bsd.test.mk (r289158) and netbsd-tests.test.mk (r289151) - Eliminate explicit OBJTOP/SRCTOP setting - Convert all ad hoc NetBSD test integration over to netbsd-tests.test.mk - Remove unnecessary TESTSDIR setting - Use SRCTOP where possible for clarity r290254: Remove unused variable (SRCDIR)
* MFC r309298: libm: remove duplicate version script entriesemaste2016-12-091-5/+0
| | | | | | | These symbols already appear in the common lib/msun/Symbol.map. Duplicate entries produce an error with LLVM's LLD linker. Sponsored by: The FreeBSD Foundation
* MFC r301761:trasz2016-12-031-1/+1
| | | | Fix frexpl() declaration to not include the field name.
* MFC r297836:ngie2016-04-221-3/+3
| | | | | | | | Fix appending -O0 to CFLAGS The previous method would completely nerf CFLAGS once bsd.progs.mk had recursed into the per-PROG logic and make the CFLAGS for tap testcases to -O0, instead of appending to CFLAGS for all of the tap testcases.
* MFC r292497:ngie2016-01-188-1/+1978
| | | | | Integrate the remaining tools/regression/lib/msun testcases into the FreeBSD test suite under lib/msun/tests
* MFC r292328:ngie2016-01-0615-0/+3425
| | | | | | | | | | | | Integrate a number of testcases from tools/regression/lib/msun into the FreeBSD test suite There's no functional change with these testcases; they're purposely being left in TAP format for the time being Other testcases which crash on amd64/i386 as-is have not been integrated yet (they need to be retested on a later version of CURRENT, as I haven't used i386 in some time)
* MFC r284289,r288267:bdrewery2015-12-041-0/+2
| | | | | | | r284289: Misc fixes from projects/bmake r288267: Remove redundant .NOPATH.
* MFC r286515:dim2015-08-121-2/+2
| | | | | | | In libm's exp2(3), avoid left-shifting a negative integer, which is undefined. Replace it with the intended value, in a defined way. Reviewed by: bde
* MFC r271651, r271719, r272138, r272457, r272845, r275476, r275518, r275614,tijl2015-06-2543-511/+1297
| | | | | | | | r275819, r276176, r278154, r278160, r278339, r279127, r279240, r279491, r279493, r279856, r283032, r284423, r284426, r284427, r284428 Merge changes to libm from the past 9 months. This includes improvements to the Bessel functions and adds the C99 function lgammal.
* MFC r282057,r282092,r282106:ngie2015-05-134-17/+4
| | | | | | | | | | | | | | | | | | | | | | | | r282057: Build/install libc, librt, libthr, and msun NetBSD test suites on all architectures r282092 (by andrew): Correct the spelling of MACHINE_CPUARCH, MACHINE_CPU is not set on arm64. r282106: Remove per-architecture checks for enabling HAVE_FENV_H The conditional came from NetBSD, where only select architectures have this header/support All architectures on FreeBSD have the necessary support though, so the conditional's completely unnecessary make tinderbox done on all architectures (including arm64, where the issue occurred before) this time
* MFC r277948:pfg2015-03-231-22/+4
| | | | | | | | | msun: use previously ignored "in" value. This fixes evaluation of exceptional values in scalblnl(). While here, simplify the code as suggested by Bruce Evans. Reported by: clang static analyzer
* MFC r274618,r276521:ngie2015-01-054-0/+79
| | | | | | | | | | | | | | | | | | | | | | MFC Note: text corrected for r274618 r274618: Add reachover Makefiles for contrib/netbsd-tests/lib/msun A variant of this code has been tested on amd64/i386 for some time by EMC/Isilon on 10-STABLE/11-CURRENT. It builds on other architectures, but the code will remain off until it's proven it works on virtual hardware or real hardware on other architectures Sponsored by: EMC / Isilon Storage Division r276521: Reset errno in :scalbnf_val and :scalbnl_val before running the tests so the tested errno isn't stale This was needed in order for the test to pass on amd64 with stable/10
* MFC r269956:andrew2014-10-291-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | From https://sourceware.org/ml/newlib/2014/msg00113.html By Richard Earnshaw at ARM > >GCC has for a number of years provides a set of pre-defined macros for >use with determining the ISA and features of the target during >pre-processing. However, the design was always somewhat cumbersome in >that each new architecture revision created a new define and then >removed the previous one. This meant that it was necessary to keep >updating the support code simply to recognise a new architecture being >added. > >The ACLE specification (ARM C Language Extentions) >(http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.set.swdev/index.html) >provides a much more suitable interface and GCC has supported this >since gcc-4.8. > >This patch makes use of the ACLE pre-defines to map to the internal >feature definitions. To support older versions of GCC a compatibility >header is provided that maps the traditional pre-defines onto the new >ACLE ones. Stop using __FreeBSD_ARCH_armv6__ and switch to __ARM_ARCH >= 6 in the couple of places in tree. clang already implements ACLE. Add a define that says we implement version 1.1, even though the implementation isn't quite complete.
* MFC r257770 r257818 r257823 r260066 r260067 r260089 r260145 r268587 r268588tijl2014-09-1829-723/+2140
| | | | | | | | | | | r268589 r268590 r268593 r268597 r269758 r270845 r270847 r270893 r270932 r270947 r271147 Merge libm work by kargl, bde and das from the past few months. Besides optimisations and small bug fixes this includes new implementations for C99 functions expl, coshl, sinhl, tanhl, erfl and erfcl. Approved by: re (kib)
* MFC 262989, 263210, 263230, 263231, 263239, 263242, 263243,ian2014-05-177-10/+398
| | | | | | | | | | | | | | | | | Export _libc_arm_fpu_present as a private symbol to be used by other system libraries, for example libm. On armv6 access both the softfloat and, when available, the vfp to get and set the floating-point environment. Build fenv-vfp.c with the softfp float abi. Without this gcc generates an incorrect assembly file that doesn't allow for vfp instructions. Only build the vfp/softfp switching code on armv6 as we don't support vfp on anything earlier than this. This should fix the armeb and arm builds when using gcc. Add an optimised version of the float and double helper functions.
* MFC r257207, r261161, r261163:ian2014-05-152-42/+63
| | | | | | | | | Update the hard-float version of the fenv functions to use the VFP unit. Any other floating-point unit is unsupported on ARM. Use __fenv_static for all static inline functions. Correctly shift the mask when masking/unmasking exceptions.
* MFC r262613:dim2014-03-261-0/+1
| | | | | | | | | | | | | | | | | | | | | Merge the projects/clang-sparc64 branch back to head. This brings in several updates from the llvm and clang trunks to make the sparc64 backend fully functional. Apart from one patch to sys/sparc64/include/pcpu.h which is still under discussion, this makes it possible to let clang fully build world and kernel for sparc64. Any assistance with testing this on actual sparc64 hardware is greatly appreciated, as there will unavoidably be bugs left. Many thanks go to Roman Divacky for his upstream work on getting the sparc64 backend into shape. MFC r262985: Repair a few minor mismerges from r262261 in the clang-sparc64 project branch. This is also to minimize differences with upstream.
* After r255294, building lib/msun's symbol map (using clang as thedim2013-09-121-1/+1
| | | | | | | | | | | | | | preprocessor) gives the following error: --- Version.map --- <stdin>:287:4: error: invalid preprocessing directive # Implemented as weak aliases for imprecise versions ^ 1 error generated. Change the comment to a C-style one, to prevent this error. Approved by: re (hrs)
* On ARM EABI double precision floating point values are stored in theandrew2013-09-071-1/+1
| | | | | | endian the CPU is in, i.e. little-endian on most ARM cores. This allows ARMv4 and ARMv5 boards to boot with the ARM EABI.
* Add stub implementations of the missing C++11 math functions.theraven2013-09-063-0/+79
| | | | | | | | | | These are weak and so can be replaced by other versions in applications that choose to do so, and will give a linker warning when used so that applications that rely on the extra precision can avoid them. Note that since the C/C++ specs only guarantee that long double has precision equal to double, code that actually relies on these functions having greater precision is unportable at best and broken at worst.
* * Whitespace.kargl2013-08-281-1/+1
|
* * s_erf.c:kargl2013-08-272-99/+71
| | | | | | | | | | | | | | | | | | | | | | . Use integer literal constants instead of double literal constants. * s_erff.c: . Use integer literal constants instead of casting double literal constants to float. . Update the threshold values from those carried over from erf() to values appropriate for float. . New sets of polynomial coefficients for the rational approximations. These coefficients have little, but positive, effect on the maximum error in ULP in the four intervals, but do improve the overall speed of execution. . Remove redundant GET_FLOAT_WORD(ix,x) as hx already contained the contents that is packed into ix. . Update the mask that is used to zero-out lower-order bits in x in the intervals [1.25, 2.857143] and [2.857143, 12]. In tests on amd64, this change improves the maximum error in ULP from 6.27739 and 63.8095 to 3.16774 and 2.92095 on these intervals for erffc(). Reviewed by: bde
* Restore the longer form of the _Generic. The short form does not work in C++.theraven2013-07-291-2/+11
|
* Reenable the isnan(double) / isinf(double) declarations when targeting C89 + ↵theraven2013-07-291-0/+15
| | | | SUSv2 mode.
* Cleaner support for type qualifiers.theraven2013-07-131-11/+2
| | | | Submitted by: Pasi Parviainen
* Ensure that the _Generic() macro in math.h works with qualified types.theraven2013-07-131-17/+26
| | | | tgmath.h contains the same bugs and so should be fixed in the same way.
* Fix the build with C++ where __builtin_types_compatible_p is not allowed.theraven2013-07-121-2/+3
|
* Fix some typoes in math.h cleanup.theraven2013-07-111-3/+3
|
* Cleanups to math.h that prevent namespace conflicts with C++.theraven2013-07-111-28/+49
| | | | | Reviewed by: bde MFC after: 3 days
* Make the order of operations for lib/msun more clear.eadler2013-06-242-2/+2
| | | | | | | Tested with md5 sum of object code Reported by: swildner@DragonFlyBSD.org Submitted by: bde
* Add implementations of acoshl(), asinhl(), and atanhl(). This is adas2013-06-1012-48/+344
| | | | merge of the work done by bde and myself.
* Style fixes.das2013-06-054-54/+58
| | | | Submitted by: bde
* Add man links for expl(3) and expm1l(3).das2013-06-041-2/+2
|
* Change a comma to a semicolon.kargl2013-06-031-3/+2
| | | | | | Remove a blank line that crept into the declarations. Fix a comment to show a sign on a NaN.
* ld80 and ld128 implementations of expm1l(). This code started lifekargl2013-06-036-4/+373
| | | | | | | | | | | | | | as a fairly faithful implementation of the algorithm found in PTP Tang, "Table-driven implementation of the Expm1 function in IEEE floating-point arithmetic," ACM Trans. Math. Soft., 18, 211-222 (1992). Over the last 18-24 months, the code has under gone significant optimization and testing. Reviewed by: bde Obtained from: bde (most of the optimizations)
* Fix two comments that got lost in the disentanglement of the larger diff.kargl2013-06-031-1/+6
|
* ld80/s_expl.c:kargl2013-06-032-8/+11
| | | | | | | | | | | | | | * Use integral numerical constants, and let the compiler do the conversion to long double. ld128/s_expl.c: * Use integral numerical constants, and let the compiler do the conversion to long double. * Use the ENTERI/RETURNI macros, which are no-ops on ld128. This however makes the ld80 and ld128 identical. Reviewed by: bde (as part of larger diff)
* Micro-optimization: move the unary mius operator to operatekargl2013-06-032-3/+3
| | | | | | on a literal constant. Obtained from: bde
* Add a comment to note that bde supplied most, if not all,kargl2013-06-031-0/+2
| | | | of the optimizations.
* ld80/s_expl.c:kargl2013-06-032-13/+9
| | | | | | | | | | | | | | | | * In the special case x = -Inf or -NaN, use a micro-optimization to eliminate the need to access u.xbits.man. * Fix an off-by-one for small arguments |x| < 0x1p-65. ld128/s_expl.c: * In the special case x = -Inf or -NaN, use a micro-optimization to eliminate the need to access u.xbits.manh and u.xbits.manl. * Fix an off-by-one for small arguments |x| < 0x1p-114. Obtained from: bde
* ld80/s_expl.c:kargl2013-06-032-27/+55
| | | | | | | | | | | | | | * Update the evaluation of the polynomial. This allows the removal of the now unused variables t23 and t45. ld128/s_expl.c: * Update the evaluation of the polynomial and the intermediate result t. This update allows several numerical constants to be written as double rather than long double constants. Update the constants as appropriate. Obtained from: bde
* Rename a few P2, P3, ... coefficients to A2, A3, ... missed inkargl2013-06-031-3/+3
| | | | my previous commit.
* Update a comment to reflect that we are using an endpoint ofkargl2013-06-032-2/+2
| | | | an interval instead of a midpoint.
* Add a u suffix to the IEEEl2bits unions o_threshold and u_threshold,kargl2013-06-031-4/+7
| | | | | | | and use macros to access the e component of the unions. This allows the portions of the code in ld80 to be identical to the ld128 code. Obtained from: bde
* Introduce the macro LOG2_INTERVAL, which is log2(number of intervals).kargl2013-06-032-2/+5
| | | | | | | Use the macroi as a micro-optimization to convert a subtraction and division to a shift. Obtained from: bde
* Whitespace.kargl2013-06-032-4/+4
|
* * Rename the polynomial coefficients from P2, P3, ... to A2, A3, ....kargl2013-06-032-25/+24
| | | | | | | | | | The names now coincide with the name used in PTP Tang's paper. * Rename the variable from s to tbl to better reflect that this is a table, and to be consistent with the naming scheme in s_exp2l.c Reviewed by: bde (as part of larger diff)
OpenPOWER on IntegriCloud