CVS log for src/lib/libm/src/math_private.h

Up to [cvs.NetBSD.org] / src / lib / libm / src

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

libm/src/math_private.h: Tidy up comments and whitespace.

Clarify which words are which when ambiguous.

Say `significand', not `mantissa'; we're not talking about books of
log tables here.  (Strictly speaking, this may refer either to what
IEEE 754 calls the trailing significand field of the binary encoding,
which is a sequence of bits rather than a real number in [1,2), or,
for ld80 with an explicit 1-bit, what might be called the significand
field since it's not just the trailing part after the dot.  But the
term `significand' will serve as a proxy for these cases.)

libm: Fix order of significand words in EXTRACT_LDBL128_WORDS.

PR lib/58337

For i386, if ft[sg]etprec() are to be used, ensure there's a
prototype for them in scope (so include <ieeefp.h>)

Might fix the i386 build.

move inclusion of <machine/ieee.h> to the ieee source files instead of
including it in math_private.h which is also used by the vax which is not ieee.

Simply an overly complex way of saying "0 or 1" into a cleaner version.

While here add a commment about how the __i386__ magic LD80C macro
might no longer work on all other arch's as well (if someone can verify
that it still does, or can fix it so that it still works on i386, but
also works everywhere else, then please remove that XXX comment.
The issue I am (perhaps imagining) is the (uint32_t) cast.

This set of changes seem to be sufficient to make at least the
amd64 port build again after the long double math function infusion
from yesterday.

Feel free (without consulting me, I'm just hacking about) to revert
any of this and do it a better way.

Note: this is not intended to be complete, in particular, in <math.h>
I added prototypes only for the functions that are needed for a
successful build, not all of the new ones...

Bring in more long double functions from FreeBSD.

fix typos in comments and one definition, mainly s/sucket/socket/.

Add sincos{,f,l} from FreeBSD

sync with FreeBSD; bring some more inline functions.

lib: remove CONSTCOND comment

Since 2021-01-31, lint doesn't need it anymore for the common pattern of
'do ... while (0)'.

Sync with HEAD

Add the complex trig functions from FreeBSD

Fix prototypes to match definitions

Rebase to HEAD as of a few days ago.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

There is some very odd code in s_exp2.c that only works if 'double'
   values are rounded to 53 bit mantissa in teh x87 registers.
This hasn't been true since Nov 11 2013.
Forcing a store-load for 'double' in STRICT_ASSIGN() seems to fix things
  (at least enough for the few random test cases).
I suspect the code could be changed so that the number of mantissa bits
  didn't matter.

Add cbrtl(3) and sqrtl(3), from FreeBSD.

Initial version of fmodl from FreeBSD.
Basic test case for the fmod family.

resync with head

hook fma to the build

sync with head.

Pull up following revision(s) (requested by christos in ticket #232):
	distrib/sets/lists/comp/mi: revision 1.1756
	lib/libm/Makefile: revision 1.124
	lib/libm/src/math_private.h: revision 1.17
	lib/libm/src/b_tgamma.c: revision 1.1
	lib/libm/man/lgamma.3: revision 1.22
	lib/libm/src/b_log.c: revision 1.1
	lib/libm/src/b_exp.c: revision 1.1
	include/math.h: revision 1.58
	lib/libm/src/s_tgammaf.c: revision 1.1
add tgamma{,f}
Add tgamma{,f} from FreeBSD via  rudolf, netbsd at eq dot cz
oops, forgot to commit: add tgamma and tgammaf

Add tgamma{,f} from FreeBSD via  rudolf, netbsd at eq dot cz

fix an obvious botch which made eg conjl(3) a no-op

Commit SoC long double support from Stathis Kamperis

Add exp2 and exp2m

sync with head.

pass a little lint.

Add log2 and log2f from FreeBSD PR 83845

Add support for ARM VFP-format doubles, conditional on __VFP_FP__.

__arm26__ || __arm32__ -> __arm__
Not least because the unified NetBSD/arm compiler probably won't predefine
__arm26__ or __arm32__.

Initial commit of arm26 port

More trailing white space.

Use __arm32__, not arm32, for egcs.

rcsid facism, WARNSify

remove inclusion of <machine/endian.h>

Get the word order of doubles correct for the arm32 port.
Note: The arm32 port is little endian except for the FP word order.

Fix overlooked FIXME: unsigned int -> u_int32_t.

Change int -> int32_t and unsigned int -> u_int32_t to improve portabilty
to machines where int != 32 bits.

Float versions of math functions.  From Ian Taylor (ian@cygnus.com), with
minor changes by me.