Simplify the conditionals for architectures which do not need any particular
md files lists.
ok guenther@ deraadt@

Unplug assembly implementations of trig functions on x86 platforms

The same change was done by NetBSD some time back as:

Disable x87 implementations of sin, cos, tan.

The x87 hardware uses a bad approximation to pi for argument
reduction, and consequently yields bad answers for inputs near pi or
pi/2.

OK mbuhl@, tested by mbuhl@ and phessler@

riscv64 additions for libm
ok kettenis@

Fix ilogb(3) implementation. The results have to match FP_ILOGB0 and
FP_ILOGBNAN which isn't the case for the amd64 and i386 assembly versions.
Drop these in favour of C implementations. Als reimplement ilogbl(3)
by providing separate ld80 and ld128 implementations that replace the
existing implementation which may hit an infinite loop when built for
quad-precision long double.

ok patrick@, gkoehler@

Add powerpc64 support.

Remove the unused leftovers of the 4.4BSD libm, which was only used
on non-IEEE platforms. Since the VAX port was discontinued, all
our remaining architectures use IEEE floating point, as will any
future ones.
ok millert@ tb@

Implement sicos(3), sincosf(3) and sincosl(3). These functions are common
extensions and modern compilers (such as clang) will use them to optimize
separate calculations of sine and cosine.

ok tom@, patrick@, deraadt@, jmc@

Remove CFLAGS+=-mfloat-abi=softfp that I accidentally committed.

Change floating-point mode/environment control functions to operate on
both the hardfloat (FPU registers) and softfloat (software) state. This
makes these functions work correctly when mixing hardfloat and softfloat
code. Disable trapping for softfloat since most (if not all) ARMv7 and
ARMv8 hardware that is on the market doesn't support trapping of
floating-point exceptions.

ok patrick@

Add support for AArch64. Based on FreeBSD's msun header, modified
to look and feel like our code.

Reduce libm's exports and make internal calls go direct.
All dependencies on libc are now via reserved/standardized names.

ok kettenis@ millert@ deraadt@

remove references to sparc

remove hppa64 port, which we never got going beyond broken single users.
hppa reverse-stack gives us a valuable test case, but most developers don't
have a 2nd one to proceed further with this.
ok kettenis

for some time now mandoc has not required MLINKS to function
correctly - logically complete that now by removing MLINKS from base;

authors need only to ensure there is an entry in NAME for any function/
util being added. MLINKS will still work, and remain for perl to ease
upgrades;

ok nicm (curses) bcook (ssl)
ok schwarze, who provided a lot of feedback and assistance
ok tb natano jung

no need for SUBDIR now that vax is gone; ok deraadt

We are done providing support for the vax.
lots of agreement.

Move s_signgam.c to PURE_SRCS, so that it also gets built on vax.

Provide a sparc64 version of sqrtl(3) that simply calls _Qp_sqrt.
The generic sqrtl(3) is not nearly accurate enough for quad-precision
floating point.

Retire hp300, mvme68k and mvme88k ports. These ports have no users, keeping
this hardware alive is becoming increasingly difficult, and I should heed the
message sent by the three disks which have died on me over the last few days.

Noone sane will mourn these ports anyway. So long, and thanks for the fish.

simple inaccurate implementation of lrint* functions on vax, just so
we have them.
Compiles on vax. okay martynas@

use some generic code on vax, to do round/roundf/scalbln...
suggested by martynas@

okay miod@

(just rearranges the !vax makefile, no compile change there)

reorganize NOIEEE_SRCS so that files not compiled on vax because
of assembly version are obvious.
Move cproj/cprojf to the PURE club, as the asm support code provides
copysignf

okay martynas@

with DBL vs LDBL fixed, tedu some source in noieee which is actually the
same as code in generic src.
introduce PURE_SRCS for completely MI math code.
okau martynas@

m88k does not have the m68k long double type actually (long double == double
on this platform), so don't pretend it does, and don't build long double libm
routines.

remove lint leftovers; ok guenther@

remove leftover NOLINT, WANTLINT, LINTFLAGS, LOBJ vars and lint targets.

ok guenther@

install infnan(3/VAX) into man3/vax/ on all archs;
clean make build tested on i386;
ok deraadt@, no objections from martynas@

Bring libc and libm in line with the compiler now that we no longer have a
128-bit long double. Diff committed on behalf of martynas@

Document fma, fmaf, fmal, nexttoward, nexttowardf, nexttowardl.

Document complex math library.

Remove the GCC2 optimizer workaround for sparc. After switching
to GCC4 it doesn't generate broken code anymore.
From Brad. Tested and verified by myself.

- Fix the cabs, and cabsf prototypes in the manual pages; it's
been C99 complex (not struct complex) for couple of years.
- Remove BUGS section; cabs is finally prototyped in complex.h.
- Also document cabsl.

Don't reimplement the same thing twice. OK matthew@.

Move fabs(3) manual page from libm to libc, for consistency. The
modf(3), frexp(3), ldexp(3), fpclassify(3) pages are in libc, too.

Remove the stupid commented out fabs(3), frexp(3), and modf(3)
entries and unused implementations. It is clear that this situation
won't change.

Revert (leaving the complex math part alone). Some stuff is depending
on this historical behavior; so we're stuck in this stupid situation.
No cookie for me.

Finalize work on complex math routines, now that we have the
extended-precision support. Mostly from Cephes.

Move fabs(3), frexp(3), and modf(3) to libm--nothing has been using
them in libc for a very long time. OK guenther@.

Remove math(3). OK theo.

Document extended-precision routines.

Don't pull bsd.own.mk for NOPROFILE & NOPIC -- the libm_i387 thing
is gone since 1999.

Finalize work on the math library. It's time to do this monster
commit, and deal with problems (if any) in tree.

Note that this adds the following functions. Ports with hacks might
need adjustments.

nexttoward(3), fma(3), nexttowardf(3), fmaf(3), acoshl(3), asinhl(3),
atanhl(3), coshl(3), sinhl(3), tanhl(3), expl(3), expm1l(3), logl(3),
log10l(3), log1pl(3), log2l(3), modfl(3), cbrtl(3), hypotl(3),
powl(3), erfl(3), erfcl(3), lgammal(3), tgammal(3), ceill(3),
floorl(3), lrintl(3), llrintl(3), roundl(3), lroundl(3), llroundl(3),
truncl(3), fmodl(3), remainderl(3), remquol(3), nextafterl(3),
nexttowardl(3), fmal(3).

With this commit, our library implements all functionality required
by C99. Documentation bits will follow.

Adapt m68k MD functions to extended-precision.
Go for it miod@.

Adapt m68k MD functions to single-precision.
Looked over by miod@.

Document nearbyint, nearbyintf, nearbyintl.

Document C99 floating-point environment.

fix build ok martynas@

- enable fenv
- add nearbyint, nearbyintf and nearbyintl implemented using fenv

Add barebones manual pages for cimag(3), conj(3) and cproj(3)
OK kettenis@ jmc@

Use MACHINE_CPU instead of MACHINE_ARCH to pick the correct machine dependent
files or directories when applicable.
The inspiration and name of MACHINE_CPU come from NetBSD, although the way to
provide it to Makefiles is completely different.
ok kettenis@

Provide assembly version of sqrtl(3).

ok pirofti@

Provide assembly version of sqrtl(3).

ok otto@

Add man pages for the fdim/fmin/fmax family of functions and make sure fdiml
is defined on all architectures.

ok jmc@ (man pages) and martynas@

document log2() and log2f(); from thomas pfaff
ok otto millert martynas

ieee, and ieeef aren't real, and the amount of functions it documents
is getting ridiculous. split them into groups of copysign, ilogb,
nextafter, scalbn. discussed long ago with millert@

document scalbln, scalblnf, scalblnl

int is big enough to fully represent exponents of all supported fp
formats. which even for 80-bit & 128-bit long doubles is only 15
bits. therefore, scalbln, scalblnf, scalblnl are essentially the
same as scalbn, scalbnf, scalbnl with bounds checking so that
LONG_MIN..INT_MIN, and INT_MAX..LONG_MAX ranges properly raise
exceptions & yield correct values. looks good to millert@

round, roundf, trunc, truncf for hppa; ok kettenis@

make ldexpf behavior consistent with the double and extended-precision
versions; spotted by kettenis@
while here also remove unused ldexp; it lives in libc
ok kettenis@, "looks good" millert@

lrint, llrint, lrintf, llrintf for amd64. ok kettenis@, oga@

implement and use sqrt, sqrtf & fabsf in hardware; since it's
faster that way. tested by myself. discussed w/ & ok miod@, millert@

add C99-conformant nan, nanf, nanl for vax. always return zero,
since its fp does not have distinguished values for qnans. tested
by naddy@; fixes libnova. ok theo

space

document and mlink long double functions. ok jmc@

pass CPPFLAGS+= -D__STDC__ for vax, to get the right definitions
from asm.h. discussed w/ millert@

- 80-bit and quad precision trigonometric and other most
important functions: acosl, asinl, atanl, atan2l, cosl,
sinl, tanl, exp2l, frexpl, ilogbl, ldexpl, logbl, scalbnl,
fabsl, hypotl, powl, sqrtl, rintl, copysignl, nanl, fdiml,
fmaxl, fminl. mostly taken from freebsd, needed alot of
changes to adapt. note, these are all c versions; and are
quite slow when architectures have, e.g. sqrt. assembly
versions will be added afterwards
- make them .weak/__weak_alias to the double precision
versions on other archs
- no need to have two finites. finite() and finitef() are
non-standard 3BSD obsolete versions of isfinite. remove
from libm. make them weak_alias in libc to __isfinite and
__isfinitef instead. similarly make 3BSD obsolete versions
of isinf, isinff, isnan, isnanf weak_aliases to C99's
__isinf, __isinff, __isnan, __isnanf
- remove unused infinity.c. the c library has infinities
for each supported platform
- use STRICT_ASSIGN cast hack for _kernel_rem_pio2, so that
the double version has a chance of working on i386 with
extra precision
- avoid storing multiple copies of the pi/2 array, since
it won't vary
- bump major due to removed finite/finitef. although they
will be in libc, which anything is linked to, minor bump
might be enough
ok millert@. tested by sthen@, jsg@, ajacoutot@, kili@, naddy@

- noieee_src: adapt complex versions of the functions it already
supports (which is all, except the float ones)
ok millert@

gc unused files, the functions are in libc. ok millert@

since these got no ieee-specific code anymore, add fmin, fmax, fdim,
fminf, fmaxf, fdimf, fdiml for VAX. looks good to millert@

sigh, gamma got lost

- replace dtoa w/ David's gdtoa, version 2008-03-15
- provide proper dtoa locks
- use the real strtof implementation
- add strtold, __hdtoa, __hldtoa
- add %a/%A support
- don't lose precision in printf, don't round to double anymore
- implement extended-precision versions of libc functions: fpclassify,
isnan, isinf, signbit, isnormal, isfinite, now that the ieee.h is
fixed
- separate vax versions of strtof, and __hdtoa
- add complex math support. added functions: cacos, casin, catan,
ccos, csin, ctan, cacosh, casinh, catanh, ccosh, csinh, ctanh, cexp,
clog, cabs, cpow, csqrt, carg, cimag, conj, cproj, creal, cacosf,
casinf, catanf, ccosf, csinf, ctanf, cacoshf, casinhf, catanhf,
ccoshf, csinhf, ctanhf, cexpf, clogf, cabsf, cpowf, csqrtf, cargf,
cimagf, conjf, cprojf, crealf
- add fdim, fmax, fmin
- add log2. (adapted implementation e_log.c. could be more acruate
& faster, but it's good enough for now)
- remove wrappers & cruft in libm, supposed to work-around mistakes
in SVID, etc.; use ieee versions. fixes issues in python 2.6 for
djm@
- make _digittoint static
- proper definitions for i386, and amd64 in ieee.h
- sh, powerpc don't really have extended-precision
- add missing definitions for mips64 (quad), m{6,8}k (96-bit) float.h
for LDBL_*
- merge lead to frac for m{6,8}k, for gdtoa to work properly
- add FRAC*BITS & EXT_TO_ARRAY32 definitions in ieee.h, for hdtoa&ldtoa
to use
- add EXT_IMPLICIT_NBIT definition, which indicates implicit
normalization bit
- add regression tests for libc: fpclassify and printf
- arith.h & gd_qnan.h definitions
- update ieee.h: hppa doesn't have quad-precision, hppa64 does
- add missing prototypes to gdtoaimp
- on 64-bit platforms make sure gdtoa doesn't use a long when it
really wants an int
- etc., what i may have forgotten...
- bump libm major, due to removed&changed symbols
- no libc bump, since this is riding on djm's libc major crank from
a day ago

discussed with / requested by / testing theo, sthen@, djm@, jsg@,
merdely@, jsing@, tedu@, brad@, jakemsr@, and others.
looks good to millert@
parts of the diff ok kettenis@

this commit does not include:
- man page changes

various markup/grammar fixes for previous, and a missing MLINK
for remquof.3;

- add man pages for nan and remainder, from freebsd
- mlink drem, dremf, remainderf to remainder; nanf to nan
- describe that drem and dremf are obsolete aliases
- remove descriptions for remainder and remainderf from ieee
looked over by jmc@, but it's easier for him to work on when they
are in tree

describe exp2, exp2f and mlink them

remove finite from ieee.3, and unmlink from makefile

- remove isinff, isnanf. this has been moved to libc
- remove never-enabled signbit. libc has is now
- add C99 implementations, from freebsd, for nan(), nanf() (needed
to write _digittoint for it), exp2(), exp2f(), remquo(), remquof(),
needed STRICT_ASSIGN macro for math_private.h
- bump major
man pages will follow
exp2{,f} has been requested by chl@
ok millert@

NetBSD's C library supplies these functions -> OpenBSD's C library
supplies these functions. in comments

rename lround.c lroundf.c llround.c llroundf.c to s_lround.c
s_lroundf.c s_llround.c s_llroundf.c, for naming consistency
looks fine to millert@

install infnan.3 only for vax. pointed out by millert@
ok millert@

Makefile: fix MLINKS
lgamma.3: fix STANDARDS and HISTORY

- on non-ieee, rename gamma to tgamma, the 'true' gamma
- make gamma an alias of lgamma
- on ieee, add tgamma, based on gamma from non-ieee
- fixes for tgamma/lgamma/exp/log, esp. special cases (some from
freebsd); properly raise invalid fp operations on vax
- also some general cleanup, ansification, man page (which was ok
jmc@)
- bump minor
this makes some ports using tgamma possible; also consistifies
behavior across openbsd/ieee and openbsd/non-ieee, and other operating
systems
much thanks sthen@, johan@, steven@, Simon Kuhnle, Wiktor Izdebski
for testing
ok millert@

drem{,f} are aliases for remainder{,f}, so:
- document and mlink drem{,f}
- drem->remainder on noieee, and make drem an alias
- finite returns integer, n_support.c
- general n_support.c ansification and cleanup
- drem is now an ALTENTRY for remainder, in n_support.S
help with man page from jmc@ and millert@
ok millert@

add C99 round(), based on ieee_src, for noieee_src. tested on VAX
ok millert@

Move llrint.c, llrintf.c, lrint.c and lrintf.c to s_llrint.c, s_llrintf.c,
s_lrint.c and s_lrintf.c, and add i387-specific assembler versions.

ok deraadt@, otto@

add some more C99 functions: round(3) and roundf(3).

By Steven G. Kargl <kargl at troutmask dot apl dot washington.edu>
From FreeBSD

New amd64 exp(3).
tested by steven@, ok otto@

add some more C99 functions: trunc(3) and truncf(3).

From FreeBSD

ok deraadt@ millert@ otto@ jmc@

amd64 asm code is not correct for exp(+/-Inf), so until somebody
comes up with a correct asm version, use the C version of exp(3).
ok steven@ kettenis@

use WANTLINT= (on all architectures)

Introduce a few c99 functions: {l,ll}{rint,round}{,f}. From NetBSD via
jason@. Man page fixes by jmc@, prodding by jsg@. ok mickey@

remove optimization workaround for sparc64 now that we're using gcc3.

ok deraadt@

reach into new amd64 directory

change amd64's MACHINE_ARCH from x86_64 to amd64. There are many many
reasons for this, quite a few of them technical, and not all of them
in response to Intel's broken ia32e crud. The gcc toolchain stays at
x86_64 for now.

add libm parts for amd64; much from netbsd, integration by mickey.
Instead of mixing the i387 & xmm register components, we decide to
explicitly seperate them. libm is already confusing enough as it is,
thank you very much

ieee_test isn't real, rename it to logb.3 and make everything link to that instead.

Get the m88k libm to say no to drugs.

add reminder and roundings

md sqrt() implementation

Add native implementations of copysign() and copysignf() on alpha.
From NetBSD; ok millert@

The sparc workaround is needed for sparc64 too.

Correctly handle mvme88k settings here.

put infnan in there, even though it is depricated

I think that libm_i387 has lived too long

provide MLINKS for "f" versions, and make sure they show up in whatis.db

Bye, bye work-around.

egcs generates broken code for -O2 on sparc

Work-around: a few files trigger Internal Compiler Errors if compiled with
-O2 on m68k.

pull in bsd.own.mk for NOPROFILE and NOPIC

Never break the source tree builds if a _non-default compilation option_
has an issue. Fix it right, without breaking the source tree.

previous commit broke MLINKS install

move .include <bsd.lib.mk> up before NOPROFILE and NOPIC defines are used; evanc@concer.to

use _USE_WRITE

protect internal "struct excpption" from conflicting one in gcc.

missing VAX manual for infnan(3), from 4.4 BSD

Ok here is the mvme88k userland, minus a few important pieces.
as, ld change need to be merged. It is known that shared libraries and C++
do not work.
If anyone wants to step forward and take over the mvme88k port please do.
It was dumped on me when the original developer, Nivas, left his previous
job and started a new one.

I am trying to get the port in a shape that someone could take over.

correct LINKS usage

nopic

NOPROFILE; from Mathieu.Herrb@mipnet.fr

We do i387 by default now.

Add OpenBSD tag. -moj

Allow the Makefile to work on strange machines like VAX :-) -moj

Somewhat better support for building lint library

from netbsd:
Changed w_gamma.c, w_gamma_r.c, w_gammaf.c, w_gammaf_r.c to call
__ieee754_lgamma_r or __ieee754_lgammaf_r.
Removed e_gamma.c, e_gamma_r.c, e_gammaf.c, e_gammaf_r.c, e_lgamma.c,
e_lgamma_f.c.

branches: 1.1.1;
Initial revision

Unplug assembly implementations of trig functions on x86 platforms

The same change was done by NetBSD some time back as:

Disable x87 implementations of sin, cos, tan.

The x87 hardware uses a bad approximation to pi for argument
reduction, and consequently yields bad answers for inputs near pi or
pi/2.

OK mbuhl@, tested by mbuhl@ and phessler@

riscv64 additions for libm
ok kettenis@

Fix ilogb(3) implementation. The results have to match FP_ILOGB0 and
FP_ILOGBNAN which isn't the case for the amd64 and i386 assembly versions.
Drop these in favour of C implementations. Als reimplement ilogbl(3)
by providing separate ld80 and ld128 implementations that replace the
existing implementation which may hit an infinite loop when built for
quad-precision long double.

ok patrick@, gkoehler@

Add powerpc64 support.

Remove the unused leftovers of the 4.4BSD libm, which was only used
on non-IEEE platforms. Since the VAX port was discontinued, all
our remaining architectures use IEEE floating point, as will any
future ones.
ok millert@ tb@

Implement sicos(3), sincosf(3) and sincosl(3). These functions are common
extensions and modern compilers (such as clang) will use them to optimize
separate calculations of sine and cosine.

ok tom@, patrick@, deraadt@, jmc@

Remove CFLAGS+=-mfloat-abi=softfp that I accidentally committed.

Change floating-point mode/environment control functions to operate on
both the hardfloat (FPU registers) and softfloat (software) state. This
makes these functions work correctly when mixing hardfloat and softfloat
code. Disable trapping for softfloat since most (if not all) ARMv7 and
ARMv8 hardware that is on the market doesn't support trapping of
floating-point exceptions.

ok patrick@

Add support for AArch64. Based on FreeBSD's msun header, modified
to look and feel like our code.

Reduce libm's exports and make internal calls go direct.
All dependencies on libc are now via reserved/standardized names.

ok kettenis@ millert@ deraadt@

remove references to sparc

remove hppa64 port, which we never got going beyond broken single users.
hppa reverse-stack gives us a valuable test case, but most developers don't
have a 2nd one to proceed further with this.
ok kettenis

for some time now mandoc has not required MLINKS to function
correctly - logically complete that now by removing MLINKS from base;

authors need only to ensure there is an entry in NAME for any function/
util being added. MLINKS will still work, and remain for perl to ease
upgrades;

ok nicm (curses) bcook (ssl)
ok schwarze, who provided a lot of feedback and assistance
ok tb natano jung

no need for SUBDIR now that vax is gone; ok deraadt

We are done providing support for the vax.
lots of agreement.

Move s_signgam.c to PURE_SRCS, so that it also gets built on vax.

Provide a sparc64 version of sqrtl(3) that simply calls _Qp_sqrt.
The generic sqrtl(3) is not nearly accurate enough for quad-precision
floating point.

Retire hp300, mvme68k and mvme88k ports. These ports have no users, keeping
this hardware alive is becoming increasingly difficult, and I should heed the
message sent by the three disks which have died on me over the last few days.

Noone sane will mourn these ports anyway. So long, and thanks for the fish.

simple inaccurate implementation of lrint* functions on vax, just so
we have them.
Compiles on vax. okay martynas@

use some generic code on vax, to do round/roundf/scalbln...
suggested by martynas@

okay miod@

(just rearranges the !vax makefile, no compile change there)

reorganize NOIEEE_SRCS so that files not compiled on vax because
of assembly version are obvious.
Move cproj/cprojf to the PURE club, as the asm support code provides
copysignf

okay martynas@

with DBL vs LDBL fixed, tedu some source in noieee which is actually the
same as code in generic src.
introduce PURE_SRCS for completely MI math code.
okau martynas@

m88k does not have the m68k long double type actually (long double == double
on this platform), so don't pretend it does, and don't build long double libm
routines.

remove lint leftovers; ok guenther@

remove leftover NOLINT, WANTLINT, LINTFLAGS, LOBJ vars and lint targets.

ok guenther@

install infnan(3/VAX) into man3/vax/ on all archs;
clean make build tested on i386;
ok deraadt@, no objections from martynas@

Bring libc and libm in line with the compiler now that we no longer have a
128-bit long double. Diff committed on behalf of martynas@

Document fma, fmaf, fmal, nexttoward, nexttowardf, nexttowardl.

Document complex math library.

Remove the GCC2 optimizer workaround for sparc. After switching
to GCC4 it doesn't generate broken code anymore.
From Brad. Tested and verified by myself.

- Fix the cabs, and cabsf prototypes in the manual pages; it's
been C99 complex (not struct complex) for couple of years.
- Remove BUGS section; cabs is finally prototyped in complex.h.
- Also document cabsl.

Don't reimplement the same thing twice. OK matthew@.

Move fabs(3) manual page from libm to libc, for consistency. The
modf(3), frexp(3), ldexp(3), fpclassify(3) pages are in libc, too.

Remove the stupid commented out fabs(3), frexp(3), and modf(3)
entries and unused implementations. It is clear that this situation
won't change.

Revert (leaving the complex math part alone). Some stuff is depending
on this historical behavior; so we're stuck in this stupid situation.
No cookie for me.

Finalize work on complex math routines, now that we have the
extended-precision support. Mostly from Cephes.

Move fabs(3), frexp(3), and modf(3) to libm--nothing has been using
them in libc for a very long time. OK guenther@.

Remove math(3). OK theo.

Document extended-precision routines.

Don't pull bsd.own.mk for NOPROFILE & NOPIC -- the libm_i387 thing
is gone since 1999.

Finalize work on the math library. It's time to do this monster
commit, and deal with problems (if any) in tree.

Note that this adds the following functions. Ports with hacks might
need adjustments.

nexttoward(3), fma(3), nexttowardf(3), fmaf(3), acoshl(3), asinhl(3),
atanhl(3), coshl(3), sinhl(3), tanhl(3), expl(3), expm1l(3), logl(3),
log10l(3), log1pl(3), log2l(3), modfl(3), cbrtl(3), hypotl(3),
powl(3), erfl(3), erfcl(3), lgammal(3), tgammal(3), ceill(3),
floorl(3), lrintl(3), llrintl(3), roundl(3), lroundl(3), llroundl(3),
truncl(3), fmodl(3), remainderl(3), remquol(3), nextafterl(3),
nexttowardl(3), fmal(3).

With this commit, our library implements all functionality required
by C99. Documentation bits will follow.

Adapt m68k MD functions to extended-precision.
Go for it miod@.

Adapt m68k MD functions to single-precision.
Looked over by miod@.

Document nearbyint, nearbyintf, nearbyintl.

Document C99 floating-point environment.

fix build ok martynas@

- enable fenv
- add nearbyint, nearbyintf and nearbyintl implemented using fenv

Add barebones manual pages for cimag(3), conj(3) and cproj(3)
OK kettenis@ jmc@

Use MACHINE_CPU instead of MACHINE_ARCH to pick the correct machine dependent
files or directories when applicable.
The inspiration and name of MACHINE_CPU come from NetBSD, although the way to
provide it to Makefiles is completely different.
ok kettenis@

Provide assembly version of sqrtl(3).

ok pirofti@

Provide assembly version of sqrtl(3).

ok otto@

Add man pages for the fdim/fmin/fmax family of functions and make sure fdiml
is defined on all architectures.

ok jmc@ (man pages) and martynas@

document log2() and log2f(); from thomas pfaff
ok otto millert martynas

ieee, and ieeef aren't real, and the amount of functions it documents
is getting ridiculous. split them into groups of copysign, ilogb,
nextafter, scalbn. discussed long ago with millert@

document scalbln, scalblnf, scalblnl

int is big enough to fully represent exponents of all supported fp
formats. which even for 80-bit & 128-bit long doubles is only 15
bits. therefore, scalbln, scalblnf, scalblnl are essentially the
same as scalbn, scalbnf, scalbnl with bounds checking so that
LONG_MIN..INT_MIN, and INT_MAX..LONG_MAX ranges properly raise
exceptions & yield correct values. looks good to millert@

round, roundf, trunc, truncf for hppa; ok kettenis@

make ldexpf behavior consistent with the double and extended-precision
versions; spotted by kettenis@
while here also remove unused ldexp; it lives in libc
ok kettenis@, "looks good" millert@

lrint, llrint, lrintf, llrintf for amd64. ok kettenis@, oga@

implement and use sqrt, sqrtf & fabsf in hardware; since it's
faster that way. tested by myself. discussed w/ & ok miod@, millert@

add C99-conformant nan, nanf, nanl for vax. always return zero,
since its fp does not have distinguished values for qnans. tested
by naddy@; fixes libnova. ok theo

space

document and mlink long double functions. ok jmc@

pass CPPFLAGS+= -D__STDC__ for vax, to get the right definitions
from asm.h. discussed w/ millert@

- 80-bit and quad precision trigonometric and other most
important functions: acosl, asinl, atanl, atan2l, cosl,
sinl, tanl, exp2l, frexpl, ilogbl, ldexpl, logbl, scalbnl,
fabsl, hypotl, powl, sqrtl, rintl, copysignl, nanl, fdiml,
fmaxl, fminl. mostly taken from freebsd, needed alot of
changes to adapt. note, these are all c versions; and are
quite slow when architectures have, e.g. sqrt. assembly
versions will be added afterwards
- make them .weak/__weak_alias to the double precision
versions on other archs
- no need to have two finites. finite() and finitef() are
non-standard 3BSD obsolete versions of isfinite. remove
from libm. make them weak_alias in libc to __isfinite and
__isfinitef instead. similarly make 3BSD obsolete versions
of isinf, isinff, isnan, isnanf weak_aliases to C99's
__isinf, __isinff, __isnan, __isnanf
- remove unused infinity.c. the c library has infinities
for each supported platform
- use STRICT_ASSIGN cast hack for _kernel_rem_pio2, so that
the double version has a chance of working on i386 with
extra precision
- avoid storing multiple copies of the pi/2 array, since
it won't vary
- bump major due to removed finite/finitef. although they
will be in libc, which anything is linked to, minor bump
might be enough
ok millert@. tested by sthen@, jsg@, ajacoutot@, kili@, naddy@

- noieee_src: adapt complex versions of the functions it already
supports (which is all, except the float ones)
ok millert@

gc unused files, the functions are in libc. ok millert@

since these got no ieee-specific code anymore, add fmin, fmax, fdim,
fminf, fmaxf, fdimf, fdiml for VAX. looks good to millert@

sigh, gamma got lost

- replace dtoa w/ David's gdtoa, version 2008-03-15
- provide proper dtoa locks
- use the real strtof implementation
- add strtold, __hdtoa, __hldtoa
- add %a/%A support
- don't lose precision in printf, don't round to double anymore
- implement extended-precision versions of libc functions: fpclassify,
isnan, isinf, signbit, isnormal, isfinite, now that the ieee.h is
fixed
- separate vax versions of strtof, and __hdtoa
- add complex math support. added functions: cacos, casin, catan,
ccos, csin, ctan, cacosh, casinh, catanh, ccosh, csinh, ctanh, cexp,
clog, cabs, cpow, csqrt, carg, cimag, conj, cproj, creal, cacosf,
casinf, catanf, ccosf, csinf, ctanf, cacoshf, casinhf, catanhf,
ccoshf, csinhf, ctanhf, cexpf, clogf, cabsf, cpowf, csqrtf, cargf,
cimagf, conjf, cprojf, crealf
- add fdim, fmax, fmin
- add log2. (adapted implementation e_log.c. could be more acruate
& faster, but it's good enough for now)
- remove wrappers & cruft in libm, supposed to work-around mistakes
in SVID, etc.; use ieee versions. fixes issues in python 2.6 for
djm@
- make _digittoint static
- proper definitions for i386, and amd64 in ieee.h
- sh, powerpc don't really have extended-precision
- add missing definitions for mips64 (quad), m{6,8}k (96-bit) float.h
for LDBL_*
- merge lead to frac for m{6,8}k, for gdtoa to work properly
- add FRAC*BITS & EXT_TO_ARRAY32 definitions in ieee.h, for hdtoa&ldtoa
to use
- add EXT_IMPLICIT_NBIT definition, which indicates implicit
normalization bit
- add regression tests for libc: fpclassify and printf
- arith.h & gd_qnan.h definitions
- update ieee.h: hppa doesn't have quad-precision, hppa64 does
- add missing prototypes to gdtoaimp
- on 64-bit platforms make sure gdtoa doesn't use a long when it
really wants an int
- etc., what i may have forgotten...
- bump libm major, due to removed&changed symbols
- no libc bump, since this is riding on djm's libc major crank from
a day ago

discussed with / requested by / testing theo, sthen@, djm@, jsg@,
merdely@, jsing@, tedu@, brad@, jakemsr@, and others.
looks good to millert@
parts of the diff ok kettenis@

this commit does not include:
- man page changes

various markup/grammar fixes for previous, and a missing MLINK
for remquof.3;

- add man pages for nan and remainder, from freebsd
- mlink drem, dremf, remainderf to remainder; nanf to nan
- describe that drem and dremf are obsolete aliases
- remove descriptions for remainder and remainderf from ieee
looked over by jmc@, but it's easier for him to work on when they
are in tree

describe exp2, exp2f and mlink them

remove finite from ieee.3, and unmlink from makefile

- remove isinff, isnanf. this has been moved to libc
- remove never-enabled signbit. libc has is now
- add C99 implementations, from freebsd, for nan(), nanf() (needed
to write _digittoint for it), exp2(), exp2f(), remquo(), remquof(),
needed STRICT_ASSIGN macro for math_private.h
- bump major
man pages will follow
exp2{,f} has been requested by chl@
ok millert@

NetBSD's C library supplies these functions -> OpenBSD's C library
supplies these functions. in comments

rename lround.c lroundf.c llround.c llroundf.c to s_lround.c
s_lroundf.c s_llround.c s_llroundf.c, for naming consistency
looks fine to millert@

install infnan.3 only for vax. pointed out by millert@
ok millert@

Makefile: fix MLINKS
lgamma.3: fix STANDARDS and HISTORY

- on non-ieee, rename gamma to tgamma, the 'true' gamma
- make gamma an alias of lgamma
- on ieee, add tgamma, based on gamma from non-ieee
- fixes for tgamma/lgamma/exp/log, esp. special cases (some from
freebsd); properly raise invalid fp operations on vax
- also some general cleanup, ansification, man page (which was ok
jmc@)
- bump minor
this makes some ports using tgamma possible; also consistifies
behavior across openbsd/ieee and openbsd/non-ieee, and other operating
systems
much thanks sthen@, johan@, steven@, Simon Kuhnle, Wiktor Izdebski
for testing
ok millert@

drem{,f} are aliases for remainder{,f}, so:
- document and mlink drem{,f}
- drem->remainder on noieee, and make drem an alias
- finite returns integer, n_support.c
- general n_support.c ansification and cleanup
- drem is now an ALTENTRY for remainder, in n_support.S
help with man page from jmc@ and millert@
ok millert@

add C99 round(), based on ieee_src, for noieee_src. tested on VAX
ok millert@

Move llrint.c, llrintf.c, lrint.c and lrintf.c to s_llrint.c, s_llrintf.c,
s_lrint.c and s_lrintf.c, and add i387-specific assembler versions.

ok deraadt@, otto@

add some more C99 functions: round(3) and roundf(3).

By Steven G. Kargl <kargl at troutmask dot apl dot washington.edu>
From FreeBSD

New amd64 exp(3).
tested by steven@, ok otto@

add some more C99 functions: trunc(3) and truncf(3).

From FreeBSD

ok deraadt@ millert@ otto@ jmc@

amd64 asm code is not correct for exp(+/-Inf), so until somebody
comes up with a correct asm version, use the C version of exp(3).
ok steven@ kettenis@

use WANTLINT= (on all architectures)

Introduce a few c99 functions: {l,ll}{rint,round}{,f}. From NetBSD via
jason@. Man page fixes by jmc@, prodding by jsg@. ok mickey@

remove optimization workaround for sparc64 now that we're using gcc3.

ok deraadt@

reach into new amd64 directory

change amd64's MACHINE_ARCH from x86_64 to amd64. There are many many
reasons for this, quite a few of them technical, and not all of them
in response to Intel's broken ia32e crud. The gcc toolchain stays at
x86_64 for now.

add libm parts for amd64; much from netbsd, integration by mickey.
Instead of mixing the i387 & xmm register components, we decide to
explicitly seperate them. libm is already confusing enough as it is,
thank you very much

ieee_test isn't real, rename it to logb.3 and make everything link to that instead.

Get the m88k libm to say no to drugs.

add reminder and roundings

md sqrt() implementation

Add native implementations of copysign() and copysignf() on alpha.
From NetBSD; ok millert@

The sparc workaround is needed for sparc64 too.

Correctly handle mvme88k settings here.

put infnan in there, even though it is depricated

I think that libm_i387 has lived too long

provide MLINKS for "f" versions, and make sure they show up in whatis.db

Bye, bye work-around.

egcs generates broken code for -O2 on sparc

Work-around: a few files trigger Internal Compiler Errors if compiled with
-O2 on m68k.

pull in bsd.own.mk for NOPROFILE and NOPIC

Never break the source tree builds if a _non-default compilation option_
has an issue. Fix it right, without breaking the source tree.

previous commit broke MLINKS install

move .include <bsd.lib.mk> up before NOPROFILE and NOPIC defines are used; evanc@concer.to

use _USE_WRITE

protect internal "struct excpption" from conflicting one in gcc.

missing VAX manual for infnan(3), from 4.4 BSD

Ok here is the mvme88k userland, minus a few important pieces.
as, ld change need to be merged. It is known that shared libraries and C++
do not work.
If anyone wants to step forward and take over the mvme88k port please do.
It was dumped on me when the original developer, Nivas, left his previous
job and started a new one.

I am trying to get the port in a shape that someone could take over.

correct LINKS usage

nopic

NOPROFILE; from Mathieu.Herrb@mipnet.fr

We do i387 by default now.

Add OpenBSD tag. -moj

Allow the Makefile to work on strange machines like VAX :-) -moj

Somewhat better support for building lint library

from netbsd:
Changed w_gamma.c, w_gamma_r.c, w_gammaf.c, w_gammaf_r.c to call
__ieee754_lgamma_r or __ieee754_lgammaf_r.
Removed e_gamma.c, e_gamma_r.c, e_gammaf.c, e_gammaf_r.c, e_lgamma.c,
e_lgamma_f.c.

branches: 1.1.1;
Initial revision

riscv64 additions for libm
ok kettenis@

Fix ilogb(3) implementation. The results have to match FP_ILOGB0 and
FP_ILOGBNAN which isn't the case for the amd64 and i386 assembly versions.
Drop these in favour of C implementations. Als reimplement ilogbl(3)
by providing separate ld80 and ld128 implementations that replace the
existing implementation which may hit an infinite loop when built for
quad-precision long double.

ok patrick@, gkoehler@

Add powerpc64 support.

Remove the unused leftovers of the 4.4BSD libm, which was only used
on non-IEEE platforms. Since the VAX port was discontinued, all
our remaining architectures use IEEE floating point, as will any
future ones.
ok millert@ tb@

Implement sicos(3), sincosf(3) and sincosl(3). These functions are common
extensions and modern compilers (such as clang) will use them to optimize
separate calculations of sine and cosine.

ok tom@, patrick@, deraadt@, jmc@

Remove CFLAGS+=-mfloat-abi=softfp that I accidentally committed.

Change floating-point mode/environment control functions to operate on
both the hardfloat (FPU registers) and softfloat (software) state. This
makes these functions work correctly when mixing hardfloat and softfloat
code. Disable trapping for softfloat since most (if not all) ARMv7 and
ARMv8 hardware that is on the market doesn't support trapping of
floating-point exceptions.

ok patrick@

Add support for AArch64. Based on FreeBSD's msun header, modified
to look and feel like our code.

Reduce libm's exports and make internal calls go direct.
All dependencies on libc are now via reserved/standardized names.

ok kettenis@ millert@ deraadt@

remove references to sparc

remove hppa64 port, which we never got going beyond broken single users.
hppa reverse-stack gives us a valuable test case, but most developers don't
have a 2nd one to proceed further with this.
ok kettenis

for some time now mandoc has not required MLINKS to function
correctly - logically complete that now by removing MLINKS from base;

authors need only to ensure there is an entry in NAME for any function/
util being added. MLINKS will still work, and remain for perl to ease
upgrades;

ok nicm (curses) bcook (ssl)
ok schwarze, who provided a lot of feedback and assistance
ok tb natano jung

no need for SUBDIR now that vax is gone; ok deraadt

We are done providing support for the vax.
lots of agreement.

Move s_signgam.c to PURE_SRCS, so that it also gets built on vax.

Provide a sparc64 version of sqrtl(3) that simply calls _Qp_sqrt.
The generic sqrtl(3) is not nearly accurate enough for quad-precision
floating point.

Retire hp300, mvme68k and mvme88k ports. These ports have no users, keeping
this hardware alive is becoming increasingly difficult, and I should heed the
message sent by the three disks which have died on me over the last few days.

Noone sane will mourn these ports anyway. So long, and thanks for the fish.

simple inaccurate implementation of lrint* functions on vax, just so
we have them.
Compiles on vax. okay martynas@

use some generic code on vax, to do round/roundf/scalbln...
suggested by martynas@

okay miod@

(just rearranges the !vax makefile, no compile change there)

reorganize NOIEEE_SRCS so that files not compiled on vax because
of assembly version are obvious.
Move cproj/cprojf to the PURE club, as the asm support code provides
copysignf

okay martynas@

with DBL vs LDBL fixed, tedu some source in noieee which is actually the
same as code in generic src.
introduce PURE_SRCS for completely MI math code.
okau martynas@

m88k does not have the m68k long double type actually (long double == double
on this platform), so don't pretend it does, and don't build long double libm
routines.

remove lint leftovers; ok guenther@

remove leftover NOLINT, WANTLINT, LINTFLAGS, LOBJ vars and lint targets.

ok guenther@

install infnan(3/VAX) into man3/vax/ on all archs;
clean make build tested on i386;
ok deraadt@, no objections from martynas@

Bring libc and libm in line with the compiler now that we no longer have a
128-bit long double. Diff committed on behalf of martynas@

Document fma, fmaf, fmal, nexttoward, nexttowardf, nexttowardl.

Document complex math library.

Remove the GCC2 optimizer workaround for sparc. After switching
to GCC4 it doesn't generate broken code anymore.
From Brad. Tested and verified by myself.

- Fix the cabs, and cabsf prototypes in the manual pages; it's
been C99 complex (not struct complex) for couple of years.
- Remove BUGS section; cabs is finally prototyped in complex.h.
- Also document cabsl.

Don't reimplement the same thing twice. OK matthew@.

Move fabs(3) manual page from libm to libc, for consistency. The
modf(3), frexp(3), ldexp(3), fpclassify(3) pages are in libc, too.

Remove the stupid commented out fabs(3), frexp(3), and modf(3)
entries and unused implementations. It is clear that this situation
won't change.

Revert (leaving the complex math part alone). Some stuff is depending
on this historical behavior; so we're stuck in this stupid situation.
No cookie for me.

Finalize work on complex math routines, now that we have the
extended-precision support. Mostly from Cephes.

Move fabs(3), frexp(3), and modf(3) to libm--nothing has been using
them in libc for a very long time. OK guenther@.

Remove math(3). OK theo.

Document extended-precision routines.

Don't pull bsd.own.mk for NOPROFILE & NOPIC -- the libm_i387 thing
is gone since 1999.

Finalize work on the math library. It's time to do this monster
commit, and deal with problems (if any) in tree.

Note that this adds the following functions. Ports with hacks might
need adjustments.

nexttoward(3), fma(3), nexttowardf(3), fmaf(3), acoshl(3), asinhl(3),
atanhl(3), coshl(3), sinhl(3), tanhl(3), expl(3), expm1l(3), logl(3),
log10l(3), log1pl(3), log2l(3), modfl(3), cbrtl(3), hypotl(3),
powl(3), erfl(3), erfcl(3), lgammal(3), tgammal(3), ceill(3),
floorl(3), lrintl(3), llrintl(3), roundl(3), lroundl(3), llroundl(3),
truncl(3), fmodl(3), remainderl(3), remquol(3), nextafterl(3),
nexttowardl(3), fmal(3).

With this commit, our library implements all functionality required
by C99. Documentation bits will follow.

Adapt m68k MD functions to extended-precision.
Go for it miod@.

Adapt m68k MD functions to single-precision.
Looked over by miod@.

Document nearbyint, nearbyintf, nearbyintl.

Document C99 floating-point environment.

fix build ok martynas@

- enable fenv
- add nearbyint, nearbyintf and nearbyintl implemented using fenv

Add barebones manual pages for cimag(3), conj(3) and cproj(3)
OK kettenis@ jmc@

Use MACHINE_CPU instead of MACHINE_ARCH to pick the correct machine dependent
files or directories when applicable.
The inspiration and name of MACHINE_CPU come from NetBSD, although the way to
provide it to Makefiles is completely different.
ok kettenis@

Provide assembly version of sqrtl(3).

ok pirofti@

Provide assembly version of sqrtl(3).

ok otto@

Add man pages for the fdim/fmin/fmax family of functions and make sure fdiml
is defined on all architectures.

ok jmc@ (man pages) and martynas@

document log2() and log2f(); from thomas pfaff
ok otto millert martynas

ieee, and ieeef aren't real, and the amount of functions it documents
is getting ridiculous. split them into groups of copysign, ilogb,
nextafter, scalbn. discussed long ago with millert@

document scalbln, scalblnf, scalblnl

int is big enough to fully represent exponents of all supported fp
formats. which even for 80-bit & 128-bit long doubles is only 15
bits. therefore, scalbln, scalblnf, scalblnl are essentially the
same as scalbn, scalbnf, scalbnl with bounds checking so that
LONG_MIN..INT_MIN, and INT_MAX..LONG_MAX ranges properly raise
exceptions & yield correct values. looks good to millert@

round, roundf, trunc, truncf for hppa; ok kettenis@

make ldexpf behavior consistent with the double and extended-precision
versions; spotted by kettenis@
while here also remove unused ldexp; it lives in libc
ok kettenis@, "looks good" millert@

lrint, llrint, lrintf, llrintf for amd64. ok kettenis@, oga@

implement and use sqrt, sqrtf & fabsf in hardware; since it's
faster that way. tested by myself. discussed w/ & ok miod@, millert@

add C99-conformant nan, nanf, nanl for vax. always return zero,
since its fp does not have distinguished values for qnans. tested
by naddy@; fixes libnova. ok theo

space

document and mlink long double functions. ok jmc@

pass CPPFLAGS+= -D__STDC__ for vax, to get the right definitions
from asm.h. discussed w/ millert@

- 80-bit and quad precision trigonometric and other most
important functions: acosl, asinl, atanl, atan2l, cosl,
sinl, tanl, exp2l, frexpl, ilogbl, ldexpl, logbl, scalbnl,
fabsl, hypotl, powl, sqrtl, rintl, copysignl, nanl, fdiml,
fmaxl, fminl. mostly taken from freebsd, needed alot of
changes to adapt. note, these are all c versions; and are
quite slow when architectures have, e.g. sqrt. assembly
versions will be added afterwards
- make them .weak/__weak_alias to the double precision
versions on other archs
- no need to have two finites. finite() and finitef() are
non-standard 3BSD obsolete versions of isfinite. remove
from libm. make them weak_alias in libc to __isfinite and
__isfinitef instead. similarly make 3BSD obsolete versions
of isinf, isinff, isnan, isnanf weak_aliases to C99's
__isinf, __isinff, __isnan, __isnanf
- remove unused infinity.c. the c library has infinities
for each supported platform
- use STRICT_ASSIGN cast hack for _kernel_rem_pio2, so that
the double version has a chance of working on i386 with
extra precision
- avoid storing multiple copies of the pi/2 array, since
it won't vary
- bump major due to removed finite/finitef. although they
will be in libc, which anything is linked to, minor bump
might be enough
ok millert@. tested by sthen@, jsg@, ajacoutot@, kili@, naddy@

- noieee_src: adapt complex versions of the functions it already
supports (which is all, except the float ones)
ok millert@

gc unused files, the functions are in libc. ok millert@

since these got no ieee-specific code anymore, add fmin, fmax, fdim,
fminf, fmaxf, fdimf, fdiml for VAX. looks good to millert@

sigh, gamma got lost

- replace dtoa w/ David's gdtoa, version 2008-03-15
- provide proper dtoa locks
- use the real strtof implementation
- add strtold, __hdtoa, __hldtoa
- add %a/%A support
- don't lose precision in printf, don't round to double anymore
- implement extended-precision versions of libc functions: fpclassify,
isnan, isinf, signbit, isnormal, isfinite, now that the ieee.h is
fixed
- separate vax versions of strtof, and __hdtoa
- add complex math support. added functions: cacos, casin, catan,
ccos, csin, ctan, cacosh, casinh, catanh, ccosh, csinh, ctanh, cexp,
clog, cabs, cpow, csqrt, carg, cimag, conj, cproj, creal, cacosf,
casinf, catanf, ccosf, csinf, ctanf, cacoshf, casinhf, catanhf,
ccoshf, csinhf, ctanhf, cexpf, clogf, cabsf, cpowf, csqrtf, cargf,
cimagf, conjf, cprojf, crealf
- add fdim, fmax, fmin
- add log2. (adapted implementation e_log.c. could be more acruate
& faster, but it's good enough for now)
- remove wrappers & cruft in libm, supposed to work-around mistakes
in SVID, etc.; use ieee versions. fixes issues in python 2.6 for
djm@
- make _digittoint static
- proper definitions for i386, and amd64 in ieee.h
- sh, powerpc don't really have extended-precision
- add missing definitions for mips64 (quad), m{6,8}k (96-bit) float.h
for LDBL_*
- merge lead to frac for m{6,8}k, for gdtoa to work properly
- add FRAC*BITS & EXT_TO_ARRAY32 definitions in ieee.h, for hdtoa&ldtoa
to use
- add EXT_IMPLICIT_NBIT definition, which indicates implicit
normalization bit
- add regression tests for libc: fpclassify and printf
- arith.h & gd_qnan.h definitions
- update ieee.h: hppa doesn't have quad-precision, hppa64 does
- add missing prototypes to gdtoaimp
- on 64-bit platforms make sure gdtoa doesn't use a long when it
really wants an int
- etc., what i may have forgotten...
- bump libm major, due to removed&changed symbols
- no libc bump, since this is riding on djm's libc major crank from
a day ago

discussed with / requested by / testing theo, sthen@, djm@, jsg@,
merdely@, jsing@, tedu@, brad@, jakemsr@, and others.
looks good to millert@
parts of the diff ok kettenis@

this commit does not include:
- man page changes

various markup/grammar fixes for previous, and a missing MLINK
for remquof.3;

- add man pages for nan and remainder, from freebsd
- mlink drem, dremf, remainderf to remainder; nanf to nan
- describe that drem and dremf are obsolete aliases
- remove descriptions for remainder and remainderf from ieee
looked over by jmc@, but it's easier for him to work on when they
are in tree

describe exp2, exp2f and mlink them

remove finite from ieee.3, and unmlink from makefile

- remove isinff, isnanf. this has been moved to libc
- remove never-enabled signbit. libc has is now
- add C99 implementations, from freebsd, for nan(), nanf() (needed
to write _digittoint for it), exp2(), exp2f(), remquo(), remquof(),
needed STRICT_ASSIGN macro for math_private.h
- bump major
man pages will follow
exp2{,f} has been requested by chl@
ok millert@

NetBSD's C library supplies these functions -> OpenBSD's C library
supplies these functions. in comments

rename lround.c lroundf.c llround.c llroundf.c to s_lround.c
s_lroundf.c s_llround.c s_llroundf.c, for naming consistency
looks fine to millert@

install infnan.3 only for vax. pointed out by millert@
ok millert@

Makefile: fix MLINKS
lgamma.3: fix STANDARDS and HISTORY

- on non-ieee, rename gamma to tgamma, the 'true' gamma
- make gamma an alias of lgamma
- on ieee, add tgamma, based on gamma from non-ieee
- fixes for tgamma/lgamma/exp/log, esp. special cases (some from
freebsd); properly raise invalid fp operations on vax
- also some general cleanup, ansification, man page (which was ok
jmc@)
- bump minor
this makes some ports using tgamma possible; also consistifies
behavior across openbsd/ieee and openbsd/non-ieee, and other operating
systems
much thanks sthen@, johan@, steven@, Simon Kuhnle, Wiktor Izdebski
for testing
ok millert@

drem{,f} are aliases for remainder{,f}, so:
- document and mlink drem{,f}
- drem->remainder on noieee, and make drem an alias
- finite returns integer, n_support.c
- general n_support.c ansification and cleanup
- drem is now an ALTENTRY for remainder, in n_support.S
help with man page from jmc@ and millert@
ok millert@

add C99 round(), based on ieee_src, for noieee_src. tested on VAX
ok millert@

Move llrint.c, llrintf.c, lrint.c and lrintf.c to s_llrint.c, s_llrintf.c,
s_lrint.c and s_lrintf.c, and add i387-specific assembler versions.

ok deraadt@, otto@

add some more C99 functions: round(3) and roundf(3).

By Steven G. Kargl <kargl at troutmask dot apl dot washington.edu>
From FreeBSD

New amd64 exp(3).
tested by steven@, ok otto@

add some more C99 functions: trunc(3) and truncf(3).

From FreeBSD

ok deraadt@ millert@ otto@ jmc@

amd64 asm code is not correct for exp(+/-Inf), so until somebody
comes up with a correct asm version, use the C version of exp(3).
ok steven@ kettenis@

use WANTLINT= (on all architectures)

Introduce a few c99 functions: {l,ll}{rint,round}{,f}. From NetBSD via
jason@. Man page fixes by jmc@, prodding by jsg@. ok mickey@

remove optimization workaround for sparc64 now that we're using gcc3.

ok deraadt@

reach into new amd64 directory

change amd64's MACHINE_ARCH from x86_64 to amd64. There are many many
reasons for this, quite a few of them technical, and not all of them
in response to Intel's broken ia32e crud. The gcc toolchain stays at
x86_64 for now.

add libm parts for amd64; much from netbsd, integration by mickey.
Instead of mixing the i387 & xmm register components, we decide to
explicitly seperate them. libm is already confusing enough as it is,
thank you very much

ieee_test isn't real, rename it to logb.3 and make everything link to that instead.

Get the m88k libm to say no to drugs.

add reminder and roundings

md sqrt() implementation

Add native implementations of copysign() and copysignf() on alpha.
From NetBSD; ok millert@

The sparc workaround is needed for sparc64 too.

Correctly handle mvme88k settings here.

put infnan in there, even though it is depricated

I think that libm_i387 has lived too long

provide MLINKS for "f" versions, and make sure they show up in whatis.db

Bye, bye work-around.

egcs generates broken code for -O2 on sparc

Work-around: a few files trigger Internal Compiler Errors if compiled with
-O2 on m68k.

pull in bsd.own.mk for NOPROFILE and NOPIC

Never break the source tree builds if a _non-default compilation option_
has an issue. Fix it right, without breaking the source tree.

previous commit broke MLINKS install

move .include <bsd.lib.mk> up before NOPROFILE and NOPIC defines are used; evanc@concer.to

use _USE_WRITE

protect internal "struct excpption" from conflicting one in gcc.

missing VAX manual for infnan(3), from 4.4 BSD

Ok here is the mvme88k userland, minus a few important pieces.
as, ld change need to be merged. It is known that shared libraries and C++
do not work.
If anyone wants to step forward and take over the mvme88k port please do.
It was dumped on me when the original developer, Nivas, left his previous
job and started a new one.

I am trying to get the port in a shape that someone could take over.

correct LINKS usage

nopic

NOPROFILE; from Mathieu.Herrb@mipnet.fr

We do i387 by default now.

Add OpenBSD tag. -moj

Allow the Makefile to work on strange machines like VAX :-) -moj

Somewhat better support for building lint library

from netbsd:
Changed w_gamma.c, w_gamma_r.c, w_gammaf.c, w_gammaf_r.c to call
__ieee754_lgamma_r or __ieee754_lgammaf_r.
Removed e_gamma.c, e_gamma_r.c, e_gammaf.c, e_gammaf_r.c, e_lgamma.c,
e_lgamma_f.c.

branches: 1.1.1;
Initial revision

Fix ilogb(3) implementation. The results have to match FP_ILOGB0 and
FP_ILOGBNAN which isn't the case for the amd64 and i386 assembly versions.
Drop these in favour of C implementations. Als reimplement ilogbl(3)
by providing separate ld80 and ld128 implementations that replace the
existing implementation which may hit an infinite loop when built for
quad-precision long double.

ok patrick@, gkoehler@

Add powerpc64 support.

Remove the unused leftovers of the 4.4BSD libm, which was only used
on non-IEEE platforms. Since the VAX port was discontinued, all
our remaining architectures use IEEE floating point, as will any
future ones.
ok millert@ tb@

Implement sicos(3), sincosf(3) and sincosl(3). These functions are common
extensions and modern compilers (such as clang) will use them to optimize
separate calculations of sine and cosine.

ok tom@, patrick@, deraadt@, jmc@

Remove CFLAGS+=-mfloat-abi=softfp that I accidentally committed.

Change floating-point mode/environment control functions to operate on
both the hardfloat (FPU registers) and softfloat (software) state. This
makes these functions work correctly when mixing hardfloat and softfloat
code. Disable trapping for softfloat since most (if not all) ARMv7 and
ARMv8 hardware that is on the market doesn't support trapping of
floating-point exceptions.

ok patrick@

Add support for AArch64. Based on FreeBSD's msun header, modified
to look and feel like our code.

Reduce libm's exports and make internal calls go direct.
All dependencies on libc are now via reserved/standardized names.

ok kettenis@ millert@ deraadt@

remove references to sparc

remove hppa64 port, which we never got going beyond broken single users.
hppa reverse-stack gives us a valuable test case, but most developers don't
have a 2nd one to proceed further with this.
ok kettenis

for some time now mandoc has not required MLINKS to function
correctly - logically complete that now by removing MLINKS from base;

authors need only to ensure there is an entry in NAME for any function/
util being added. MLINKS will still work, and remain for perl to ease
upgrades;

ok nicm (curses) bcook (ssl)
ok schwarze, who provided a lot of feedback and assistance
ok tb natano jung

no need for SUBDIR now that vax is gone; ok deraadt

We are done providing support for the vax.
lots of agreement.

Move s_signgam.c to PURE_SRCS, so that it also gets built on vax.

Provide a sparc64 version of sqrtl(3) that simply calls _Qp_sqrt.
The generic sqrtl(3) is not nearly accurate enough for quad-precision
floating point.

Retire hp300, mvme68k and mvme88k ports. These ports have no users, keeping
this hardware alive is becoming increasingly difficult, and I should heed the
message sent by the three disks which have died on me over the last few days.

Noone sane will mourn these ports anyway. So long, and thanks for the fish.

simple inaccurate implementation of lrint* functions on vax, just so
we have them.
Compiles on vax. okay martynas@

use some generic code on vax, to do round/roundf/scalbln...
suggested by martynas@

okay miod@

(just rearranges the !vax makefile, no compile change there)

reorganize NOIEEE_SRCS so that files not compiled on vax because
of assembly version are obvious.
Move cproj/cprojf to the PURE club, as the asm support code provides
copysignf

okay martynas@

with DBL vs LDBL fixed, tedu some source in noieee which is actually the
same as code in generic src.
introduce PURE_SRCS for completely MI math code.
okau martynas@

m88k does not have the m68k long double type actually (long double == double
on this platform), so don't pretend it does, and don't build long double libm
routines.

remove lint leftovers; ok guenther@

remove leftover NOLINT, WANTLINT, LINTFLAGS, LOBJ vars and lint targets.

ok guenther@

install infnan(3/VAX) into man3/vax/ on all archs;
clean make build tested on i386;
ok deraadt@, no objections from martynas@

Bring libc and libm in line with the compiler now that we no longer have a
128-bit long double. Diff committed on behalf of martynas@

Document fma, fmaf, fmal, nexttoward, nexttowardf, nexttowardl.

Document complex math library.

Remove the GCC2 optimizer workaround for sparc. After switching
to GCC4 it doesn't generate broken code anymore.
From Brad. Tested and verified by myself.

- Fix the cabs, and cabsf prototypes in the manual pages; it's
been C99 complex (not struct complex) for couple of years.
- Remove BUGS section; cabs is finally prototyped in complex.h.
- Also document cabsl.

Don't reimplement the same thing twice. OK matthew@.

Move fabs(3) manual page from libm to libc, for consistency. The
modf(3), frexp(3), ldexp(3), fpclassify(3) pages are in libc, too.

Remove the stupid commented out fabs(3), frexp(3), and modf(3)
entries and unused implementations. It is clear that this situation
won't change.

Revert (leaving the complex math part alone). Some stuff is depending
on this historical behavior; so we're stuck in this stupid situation.
No cookie for me.

Finalize work on complex math routines, now that we have the
extended-precision support. Mostly from Cephes.

Move fabs(3), frexp(3), and modf(3) to libm--nothing has been using
them in libc for a very long time. OK guenther@.

Remove math(3). OK theo.

Document extended-precision routines.

Don't pull bsd.own.mk for NOPROFILE & NOPIC -- the libm_i387 thing
is gone since 1999.

Finalize work on the math library. It's time to do this monster
commit, and deal with problems (if any) in tree.

Note that this adds the following functions. Ports with hacks might
need adjustments.

nexttoward(3), fma(3), nexttowardf(3), fmaf(3), acoshl(3), asinhl(3),
atanhl(3), coshl(3), sinhl(3), tanhl(3), expl(3), expm1l(3), logl(3),
log10l(3), log1pl(3), log2l(3), modfl(3), cbrtl(3), hypotl(3),
powl(3), erfl(3), erfcl(3), lgammal(3), tgammal(3), ceill(3),
floorl(3), lrintl(3), llrintl(3), roundl(3), lroundl(3), llroundl(3),
truncl(3), fmodl(3), remainderl(3), remquol(3), nextafterl(3),
nexttowardl(3), fmal(3).

With this commit, our library implements all functionality required
by C99. Documentation bits will follow.

Adapt m68k MD functions to extended-precision.
Go for it miod@.

Adapt m68k MD functions to single-precision.
Looked over by miod@.

Document nearbyint, nearbyintf, nearbyintl.

Document C99 floating-point environment.

fix build ok martynas@

- enable fenv
- add nearbyint, nearbyintf and nearbyintl implemented using fenv

Add barebones manual pages for cimag(3), conj(3) and cproj(3)
OK kettenis@ jmc@

Use MACHINE_CPU instead of MACHINE_ARCH to pick the correct machine dependent
files or directories when applicable.
The inspiration and name of MACHINE_CPU come from NetBSD, although the way to
provide it to Makefiles is completely different.
ok kettenis@

Provide assembly version of sqrtl(3).

ok pirofti@

Provide assembly version of sqrtl(3).

ok otto@

Add man pages for the fdim/fmin/fmax family of functions and make sure fdiml
is defined on all architectures.

ok jmc@ (man pages) and martynas@

document log2() and log2f(); from thomas pfaff
ok otto millert martynas

ieee, and ieeef aren't real, and the amount of functions it documents
is getting ridiculous. split them into groups of copysign, ilogb,
nextafter, scalbn. discussed long ago with millert@

document scalbln, scalblnf, scalblnl

int is big enough to fully represent exponents of all supported fp
formats. which even for 80-bit & 128-bit long doubles is only 15
bits. therefore, scalbln, scalblnf, scalblnl are essentially the
same as scalbn, scalbnf, scalbnl with bounds checking so that
LONG_MIN..INT_MIN, and INT_MAX..LONG_MAX ranges properly raise
exceptions & yield correct values. looks good to millert@

round, roundf, trunc, truncf for hppa; ok kettenis@

make ldexpf behavior consistent with the double and extended-precision
versions; spotted by kettenis@
while here also remove unused ldexp; it lives in libc
ok kettenis@, "looks good" millert@

lrint, llrint, lrintf, llrintf for amd64. ok kettenis@, oga@

implement and use sqrt, sqrtf & fabsf in hardware; since it's
faster that way. tested by myself. discussed w/ & ok miod@, millert@

add C99-conformant nan, nanf, nanl for vax. always return zero,
since its fp does not have distinguished values for qnans. tested
by naddy@; fixes libnova. ok theo

space

document and mlink long double functions. ok jmc@