Copy stable/10@r272459 to releng/10.1 as part of
the 10.1-RELEASE process.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Copy head (r256279) to stable/10 as part of the 10.0-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

The getcontext() from the __fillcontextx() call in the
check_deferred_signal() returns twice, since handle_signal() emulates
the return from the normal signal handler by sigreturn(2)ing the
passed context. Second return is performed on the destroyed stack
frame, because __fillcontextx() has already returned. This causes
undefined and bad behaviour, usually the victim thread gets SIGSEGV.

Avoid nested frame and the need to return from it by doing direct call
to getcontext() in the check_deferred_signal() and using a new private
libc helper __fillcontextx2() to complement the context with the
extended CPU state if the deferred signal is still present.

The __fillcontextx() is now unused, but is kept to allow older
libthr.so to be used with the new libc.

Mark __fillcontextx() as returning twice [1].

Reported by: pgj
Pointy hat to: kib
Discussed with: dim
Tested by: pgj, dim
Suggested by: jilles [1]
MFC after: 1 week

libc/amd64: Do not export .cerror.

For some reason, libc exports the symbol .cerror (HIDENAME(cerror)), albeit
in the FBSDprivate_1.0 version. It looks like there is no reason for this
since it is not used from other libraries. Given that it cannot be accessed
from C and its strange calling convention, it is rather unlikely that other
things rely on it. Perhaps it is from a time when symbols could not be
hidden.

Most of the amd64 assembler code jumps to .cerror using the GOT. It can jump
to it directly now, as in non-PIC mode.

There are also some minor size optimizations to instructions but they yield
virtually no benefit in the size of libc.so.7 due to padding.

Reviewed by: kib

Make the sys/ucontext.h self-contained by changing the return type
of __getcontextx_size(3) from size_t to int.

PR: ports/164654
MFC after: 1 month

Add API for obtaining extended machine context states that cannot be
fit into existing mcontext_t.

On i386 and amd64 do return the extended FPU states using
getcontextx(3). For other architectures, getcontextx(3) returns the
same information as getcontext(2).

Tested by: pho
MFC after: 1 month

Make sure that stack is 16-byte aligned before calling a function,
as it is required by amd64 ABI. Add a comment for the places were
the stack is accidentally properly aligned already.

PR: amd64/162214
Submitted by: yamayan <yamayan kbh biglobe ne jp>
MFC after: 1 week

Replace a proliferation of buggy MD implementations of modf() with a
working MI one. The MI one only needs to be overridden on machines
with non-IEEE754 arithmetic. (The last supported one was the VAX.)
It can also be overridden if someone comes up with a faster one that
actually passes the regression tests -- but this is harder than it sounds.

Add section .note.GNU-stack for assembly files used by 386 and amd64.

Retire the amd64 and i386 specific inline assembly versions of ldexp.c,
as they are slower than the generic version in C, at least on modern
hardware. This leaves us with just five implementations.

Suggested by: bde
Approved by: rpaulo (mentor)

Apply the same workaround for clang to amd64's version of ldexp.c (as in
r212976): order the incoming arguments to fscale as st(0), st(1), and
mark temp2 volatile (only in case of compilation with clang) to force
clang to pop it correctly. No binary change when compiled with gcc.

This fixes ldexp() when compiled with clang on amd64, which makes
drand48() and friends work correctly again, and this in turn fixes
perl's tempfile().

Reported by: Renato Botelho, Derek Tattersall
Approved by: rpaulo (mentor)

We've been lax about matching END() macros in asm code for some time. This
is used to set the ELF size attribute for functions. It isn't normally
critical but some things can make use of it (gdb for stack traces).
Valgrind needs it so I'm adding it in. The problem is present on all
branches and on both i386 and amd64.

Two FP-related setjmp/longjmp changes:

1. Save and restore the control part of the MXCSR in addition to the
i387 control word to ensure that the two are consistent.

Note that standards don't require longjmp to restore either control
word, and none of Linux, MacOS X 10.3 and earlier, NetBSD, OpenBSD,
or Solaris do it. However, it is historical FreeBSD behavior, and
bde points out that it is needed to make longjmping out of a signal
handler work properly, given the way FreeBSD clobbers the FPU state
on signal handler entry.

2. Don't clobber the FPU exception flags in longjmp. C99 requires them
to remain unchanged.

Remove silly n that crept in

Per Regents of the University of Calfornia letter, remove advertising
clause.

# If I've done so improperly on a file, please let me know.

Remove fpsetsticky(). This was added for SysV compatibility, but due
to mistakes from day 1, it has always had semantics inconsistent with
SVR4 and its successors. In particular, given argument M:

- On Solaris and FreeBSD/{alpha,sparc64}, it clobbers the old flags
and *sets* the new flag word to M. (NetBSD, too?)
- On FreeBSD/{amd64,i386}, it *clears* the flags that are specified in M
and leaves the remaining flags unchanged (modulo a small bug on amd64.)
- On FreeBSD/ia64, it is not implemented.

There is no way to fix fpsetsticky() to DTRT for both old FreeBSD apps
and apps ported from other operating systems, so the best approach
seems to be to kill the function and fix any apps that break. I
couldn't find any ports that use it, and any such ports would already
be broken on FreeBSD/ia64 and Linux anyway.

By the way, the routine has always been undocumented in FreeBSD,
except for an MLINK to a manpage that doesn't describe it. This
manpage has stated since 5.3-RELEASE that the functions it describes
are deprecated, so that must mean that functions that it is *supposed*
to describe but doesn't are even *more* deprecated. ;-)

Note that fpresetsticky() has been retained on FreeBSD/i386. As far
as I can tell, no other operating systems or ports of FreeBSD
implement it, so there's nothing for it to be inconsistent with.

PR: 75862
Suggested by: bde

Add support for TLS in statically linked programs.

1. Use correct alignment mask, -15 != ~15
2. Make end of frames

Set fpu context flags to known values, zero is illegal.

Make FLT_ROUNDS correctly reflect the dynamic rounding mode.

Replace seven nominally MD implementations of frexp() that are broken
for subnormals with one implementation that works.

Implement the classification macros isfinite(), isinf(), isnan(), and
isnormal() the hard way, rather than relying on fpclassify(). This is
a lose in the sense that we need a total of 12 functions, but it is
necessary for binary compatibility because we have never bumped libm's
major version number. In particular, isinf(), isnan(), and isnanf()
were BSD libc functions before they were C99 macros, so we can't
reimplement them in terms of fpclassify() without adding a dependency
on libc.so.5. I have tried to arrange things so that programs that
could be compiled in FreeBSD 4.X will generate the same external
references when compiled in 5.X. At the same time, the new macros
should remain C99-compliant.

The isinf() and isnan() functions remain in libc for historical
reasons; however, I have moved the functions that implement the macros
isfinite() and isnormal() to libm where they belong. Moreover,
half a dozen MD versions of isinf() and isnan() have been replaced
with MI versions that work equally well.

Prodded by: kris

Avoid to touch red zone, in libpthread, ucontext may be saved by kernel's
get_mcontext, and libpthread will use signalcontext to deliver signal in
userland, it looks same as kernel's send_sig does.

Reviewed by: deischen, tjr

Fix Yet Another 16 byte stack alignment bug. Thankfully, this one is
solved by a simple 'make world'. The signalcontext function was going
to the trouble of generating an even 16 byte alignment, but in fact it
needed to be odd aligned to simulate the 8-byte return address having
been pushed by the caller. This fixes yet another group of crashes in
applications using libpthread. And yet again, it was my fault all along.

While here, rename the duplicate internal ctx_wrapper() functions to
makectx_wrapper() and sigctx_wrapper() so that traces aren't ambiguous.

Fix a typo in the !__GNUC__ case and remove an obsolete comment.

Fix some aliasing problems.

Rewrite fabs.S to use pure SSE2 operations. I got the clues how to do
this from looking at code generated by gcc.

"Fix" makecontext() so that the C code begins execution with its
ABI-required stack alignment. C code expects that the push of the
return address disturbed the 16 byte alignment and it will take corrective
measures to fix it before making another call. Of course, if its wrong
to start with, then all hell breaks loose. Essentially we "fix" this
by making the stack alignment odd to start with.

This was one of the things that broke on libkse with apps that use
floating point/varargs/etc.

Approved by: re (scottl)

Add rfork_thread(3).

Fix fabs(). This commit brought to you by the letter 'l'.
(fstp stores a mem32 value, fstpl stores a mem64 value)

This fixes ghostscript for 'make release' on amd64. Ghostscript for some
reason thinks it is a good idea to use -fno-builtin, which means it is
vulnerable to bugs in libc that are normally hidden by the builtin gcc
functions. Oops.

Set mc_len to sizeof(mcontext_t), otherwise it is an invalid context.

Instantiate explicit callable versions of the machine/ieeefp.h inlines
for the use of non-GCC compilers and C++ code.

Add amd64 versions of makecontext() and signalcontext() needed
for libkse (makecontext() is also needed for libthr).
These probably will need some tweaking.

This is unusable on amd64. Remove it before it causes more confusion.
It is only possible to do this on an ABI that has a compulsory frame
pointer, which the amd64 ABI does not. Thus, it is only possible to
implement this as a compiler builtin.

Use C99 compatible ASM statements.
(untested, but existing state breaks
http://triangle.rtp.freebsd.org/~des/tinderbox-CURRENT-amd64-amd64.brief)

Cosmetic; record offsets used within jmpbuf

Fix sigsetjmp(). It helps if we do not try to save the old signal mask
to a random memory location. Perl works a lot better with this.

Update ldexp.c for amd64.

SIG_SETMASK is 3, not 1. Sigh.

Fix an embarresing transcription error from i386 to amd64. Put the arguments
to sigprocmask(2) int the correct order. *blush*.
For sigsetjmp(), match up the pushq/popq in the non-savemask case.

Tidy up modf.S and make it actually work. It wasn't extracting
the value out of ST(0) before copying it to %xmm0. Also remove bogus stack
frame and work in the red zone.

Fix typo, even though this is unused.

Turn off alloca.S - it cannot possibly work like this since on AMD64, gcc
doesn't use stack frames. It uses offsets relative to %rsp, not %rbp. So
we cannot just change %rsp like this.

Approved by: re (blanket amd64)

Convert the i386 alloca(3) to the x86-64 ISA.

Alignment hack from: NetBSD

Update for AMD64 contexts. Note that this still has some x87-style FPU
instructions here, these may need to be updated for SSE.
(Repocopied from i386/gen)

Reduce the SRCS to what compiles

Floating point libc functions traditionally written in ASM.
AMD64 does away with the x87 in 64-bit long mode, so we have to play the
SSE/SSE2 game now.

Obtained from: NetBSD/x86-64

o Implement C99 classification macros isfinite(), isinf(), isnan(),
isnormal(). The current isinf() and isnan() are perserved for
binary compatibility with 5.0, but new programs will use the macros.
o Implement C99 comparison macros isgreater(), isgreaterequal(),
isless(), islessequal(), islessgreater(), isunordered().

Submitted by: David Schultz <dschultz@uclink.Berkeley.EDU>

Implement fpclassify():
o Add a MD header private to libc called _fpmath.h; this header
contains bitfield layouts of MD floating-point types.
o Add a MI header private to libc called fpmath.h; this header
contains bitfield layouts of MI floating-point types.
o Add private libc variables to lib/libc/$arch/gen/infinity.c for
storing NaN values.
o Add __double_t and __float_t to <machine/_types.h>, and provide
double_t and float_t typedefs in <math.h>.
o Add some C99 manifest constants (FP_ILOGB0, FP_ILOGBNAN, HUGE_VALF,
HUGE_VALL, INFINITY, NAN, and return values for fpclassify()) to
<math.h> and others (FLT_EVAL_METHOD, DECIMAL_DIG) to <float.h> via
<machine/float.h>.
o Add C99 macro fpclassify() which calls __fpclassify{d,f,l}() based
on the size of its argument. __fpclassifyl() is never called on
alpha because (sizeof(long double) == sizeof(double)), which is good
since __fpclassifyl() can't deal with such a small `long double'.

This was developed by David Schultz and myself with input from bde and
fenner.

PR: 23103
Submitted by: David Schultz <dschultz@uclink.Berkeley.EDU>
(significant portions)
Reviewed by: bde, fenner (earlier versions)

Disconnect the userland get/set/swapcontext() functions from
libc. I want to keep these in some version for the thread
library/ies, but don't know whether to have them repo-copied
to libc_r or renamed and kept in libc.

Change the name of an alpha macro that was changed with the
system call commit.

Re-apply the previously backed-out commit that fixes the problem where
HUGE_VAL is not properly aligned on some architectures. The previous
fix now works because the two versions of 'math.h' (include/math.h
and lib/msun/src/math.h) have since been merged into one.

PR: bin/43544

Revert previous commit to unbreak world until we figure out the
right way to do it.

Fix a problem with the definition of HUGE_VAL causing the gcc warning
"cast increases required alignment of target type" on some platforms.

Reviewed by: bde

Fixed editing error in previous commit (*blush*).

Fixed unsorting of SRCS.

Save and restore FPU state properly in ucontext_t's.

Reviewed by: deischen, julian
Approved by: -arch

Restore CSRG ID's lost in January 1995.

Fix the style of the SCM ID's.
I believe have made all of libc .c's as consistent as possible.

Move swapcontext.c to the ${MACHINE_ARCH}/gen/Makefile.inc area, otherwise
it causes undefined references to getcontext() and setcontext() on
platforms other than i386 and alpha.

Fixed unsorting and splitting of SRCS in previous commit.

Add getcontext, setcontext, makecontext, and swapcontext. These
functions are defined in SUSv2 and the latest POSIX spec.

Thanks to Bernd Walter <ticso@cicely8.cicely.de> for helping debug my
alpha assembly.

Approved by: -arch

De-orbit DEFS.h - the other arches do not use it, and it got replaced
with <machine/asm.h>.

Reviewed by: bde

Removed duplicate VCS ID tags, as per style(9).

Remove _THREAD_SAFE and make libc thread-safe by default by
adding (weak definitions to) stubs for some of the pthread
functions. If the threads library is linked in, the real
pthread functions will pulled in.

Use the following convention for system calls wrapped by the
threads library:
__sys_foo - actual system call
_foo - weak definition to __sys_foo
foo - weak definition to __sys_foo

Change all libc uses of system calls wrapped by the threads
library from foo to _foo. In order to define the prototypes
for _foo(), we introduce namespace.h and un-namespace.h
(suggested by bde). All files that need to reference these
system calls, should include namespace.h before any standard
includes, then include un-namespace.h after the standard
includes and before any local includes. <db.h> is an exception
and shouldn't be included in between namespace.h and
un-namespace.h namespace.h will define foo to _foo, and
un-namespace.h will undefine foo.

Try to eliminate some of the recursive calls to MT-safe
functions in libc/stdio in preparation for adding a mutex
to FILE. We have recursive mutexes, but would like to avoid
using them if possible.

Remove uneeded includes of <errno.h> from a few files.

Add $FreeBSD$ to a few files in order to pass commitprep.

Approved by: -arch

rfork(2) wrapper for simple rfork-style threads. I have lost count of
the number of times I have given this to people and got asked: why isn't
it in libc? It is impossible to do this without assembler glue to reset
the stack for the new child process.

int rfork_thread(flags, stack_addr, start_fnc, start_arg)
int flags; Flags to rfork system call. See rfork(2).
void *stack_addr; Top of stack for thread.
int (*start_fnc)(void *); Address of thread function to call in child.
void *start_arg; Argument to pass to the thread function in child.

This is deliberately not documented or prototyped in includes until the
corresponding alpha version is written.

Use assembler directives rather than ALTENTRY() so that longjmp() and
siglongjmp() are weak symbols. This is necessary to allow static linking
with the linuxthreads library port.

Remove unnecessary alternate entry points for *setjmp(). Make the main
entry point the standard name when not compiling libc_r (for example,
longjmp is the main entry point instead of __longjmp).

Suggested by: bde

Make minor entry point changes to support libc_r.

Add three-tier symbol naming in support of POSIX thread cancellation
points. For library functions, the pattern is __sleep() <--
_libc_sleep() <-- sleep(). The arrows represent weak aliases. For
system calls, the pattern is _read() <-- _libc_read() <-- read().

Unbreak profiling. bde says this is not the cleanest way to fix the
problem, but that it works.

Submitted by: bde

Creating weak symbols doesn't work correctly when building an aout libc.
Doing the "right thing" here is difficult, so create two ENTRY points for
each function (for example, __setjmp and setjmp are equivalent). This
isn't pretty, but it works for both aout and ELF.

libc symbol naming needs an overhaul in order to properly support function
wrapping, specifically in the case of a real libpthread, and these
duplicate entry points should be fixed as part of that overhaul.

Pointed out by: bde

Make setjmp, longjmp, sigsetjmp, and siglongjmp weak aliases for
__setjmp, __longjmp, __sigsetjmp, and __siglongjmp, respectively.
This supports cancellation in the linuxthreads port. In the long run,
a much more comprehensive solution will necessitate more dramatic changes
to libc symbol naming, and these aliases will probably need modification
at that time.

\begin{quote}
setjmp() gets the jmp_buf pointer from the wrong place (the place
where the return address is) in the shlib case, and uses it (only)
to fetch the current signal mask to address (return_address + 28).
This address is normally read-only (I hope), so the sigprocmask()
call has no effect except to return an error code.
\end{quote}

Submitted by: bde

sigset_t change (part 5 of 5)
-----------------------------

Most of the userland changes are in libc. For both the alpha
and the i386 setjmp has been changed to accomodate for the
new sigset_t. Internally, libc is mostly rewritten to use the
new syscalls. The exception is in compat-43/sigcompat.c

The POSIX thread library has also been rewritten to use the
new sigset_t. Except, that it currently only handles NSIG
signals instead of the maximum _SIG_MAXSIG. This should not
be a problem because current applications don't use any
signals higher than NSIG.

There are version bumps for the following libraries:
libdialog
libreadline
libc
libc_r
libedit
libftpio
libss

These libraries either a) have one of the modified structures
visible in the interface, or b) use sigset_t internally and
may cause breakage if new binaries are used against libraries
that don't have the sigset_t change. This not an immediate
issue, but will be as soon as applications start using the
new range to its fullest.

NOTE: libncurses already had an version bump and has not been
given one now.

NOTE: doscmd is a real casualty and has been disconnected for
the moment. Reconnection will eventually happen after
doscmd has been fixed. I'm aware that being the last one
to touch it, I'm automaticly promoted to being maintainer.
According to good taste this means that I will receive a
badge which either will be glued or mechanically stapled,
drilled or otherwise violently forced onto me :-)

NOTE: pcvt/vttest cannot be compiled with -traditional. The
change cause sys/types to be included along the way which
contains the const and volatile modifiers. I don't consider
this a solution, but more a workaround.

$Id$ -> $FreeBSD$

Stop renaming these in libc_r because wrappered versions don't make sense.
PR: i386/4826, bin/5953

Moved `SRCS+= frexp.c' to the correct Makefile.inc.

Sorted SRCS.

Revert $FreeBSD$ to $Id$

Make the long-awaited change from $Id$ to $FreeBSD$

This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.

Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.

General -Wall warning cleanup, part I.
Submitted-By: Kent Vander Velden <graphix@iastate.edu>

Fix a bug caused by the collision of a local assembler label with another
use of the same label in a recently-introduced PIC_PROLOGUE. This
should solve the recent core dumps from pdksh.

Add support to enable libc to be compiled in ELF format. (#ifdef __ELF__)
In a nutshell, this macroizes the local/global symbol scoping rules
that are different in a.out and ELF. It also makes the i386 assembler
stubs conform to i386 PIC calling conventions - the a.out ld.so didn't
object, but the ELF one needs it as it implements PIC jumps via PLT's as
well as calls. The a.out rtld only worked because it was accidently
snooping the grandparent calling function's return address off the stack..

This also affects the libc_r code a little, because of cpp macro nesting.

Put in missing '_' in call to wrapper _thread_sys_sigsetmask() function.

Reviewed by: julian and (hsu?)
Submitted by: John Birrel(L?)

changes for threadsafe operations

remove GCC support functions from libc.
Should never have been here in the first place.

First round of changes to clean up the RCSID mess in libc:

1) Changed LIB_SCCS and SYSLIB_SCCS to LIB_RCS and SYSLIB_RCS.
2) Changed sccsid[] variables to rcsid[]
3) Moved all RCSID strings into .text
4) Converted all SCCSID's to RCS $Id$'s
5) Added missing $Id$'s after copyright.

sigsetjmp.S:
Remove unnecessary .text statement.

Remove unnecessary .align statement.

Reenable sigsetjmp.S. Preserve the FP state. Rearrange offsets
to match setjmp.S.

This is weird. I *added this*, but it went away again! Ummm.. Mumble.
I'm confused..
Submitted by: jkh

Put __infinity back here again until someone does the right thing and
repartitions libc into something human again. I don't have that kind of
time right now myself, unfortunately.
Submitted by: jkh

First crack at making libc work with the new make macros. It compiles on
my machine, and a simple static (genassym) and shared (sysctl) executable
both work. Still to be done: RPCand YP merge.

Pull in GNU2 fix for this from FreeBSD, allows ldexp.c to compile with
gcc2.x

BSD 4.4 Lite Lib Sources