Commit Graph

1375 Commits

Author SHA1 Message Date
Dimitry Andric
b74cf6dcf1 Revert r231673 and r231682 for now, until we can run a full make
universe with them.  Sorry for the breakage.

Pointy hat to:	     me and brooks
2012-02-14 21:48:46 +00:00
David Chisnall
7780c181c5 Fix a misplaced __NO_TLS locations, and change a GNUism to a C11ism for
consistency.

Approved by:	brooks (mentor)
2012-02-14 14:24:37 +00:00
Tijl Coosemans
265f940acc Change some headers such that lang/gcc* ports no longer patch them.
The lang/gcc* ports patch headers where they think something is
non-standard. These patched headers override the system headers which means
you have to rebuild these ports whenever you do installworld to make sure
they contain the latest changes.
2012-02-14 12:50:20 +00:00
David Chisnall
82dd5016bd Cleanup of xlocale:
- Address performance regressions encountered by das@ by caching per-thread
  data in TLS where available.
- Add a __NO_TLS flag to cdefs.h to indicate where not available.
- Reorganise the xlocale.h definitions into xlocale/*.h so that they can be
  included from multiple places.
- Export the POSIX2008 subset of xlocale when POSIX2008 says it should be
  exported, independently of whether xlocale.h is included.
- Fix the bug where programs using ctype functions always assumed ASCII unless
  recompiled.
- Fix some style(9) violations.

Reviewed by:	brooks (mentor)
Approved by:	dim (mentor)
2012-02-14 12:03:23 +00:00
Dimitry Andric
5e0a19bdee Consistently set RPCGEN_CPP when running rpcgen, so the C preprocessor
set via ${CPP} is used, instead of always using hardcoded /usr/bin/cpp.

MFC after:	1 week
2012-02-07 09:27:07 +00:00
Ed Schouten
76f11d88aa Only use the static assertion when __generic is available.
Reported by:	tijl
2012-01-17 20:22:10 +00:00
David Chisnall
4ecabf31f3 Fix clang atomic to use for atomic_is_lock_free().
Reviewed by:	ed
Approved by:	dim (mentor)
2012-01-17 15:20:41 +00:00
David Chisnall
03c142e762 Use the signal fence builtin in stdatomic.h when using the clang atomic
builtins, rather than the __asm hack.  Somehow I missed the existence of this
builtin originally and only noticed that it was there when I went to implement
it...

Note: Trunk clang now has support for (most of) the C[++]11 atomics stuff.
Please test!

Approved by:	brooks (mentor)
2012-01-16 18:19:53 +00:00
Andriy Gapon
ea3282935d xlocale.h: consistently use __va_list
Plain 'va_list' in this header seems to cause troubles with non-base GCC
which creates and uses "tortured" versions of some sysem header files
including stdio.h (installed in a private 'include-fixed' directory).

Reviewed by:	theraven
X-MFC with:	r227753
2012-01-15 20:37:39 +00:00
Ed Schouten
a6a53dc8f7 Correct mistake in atomic_flag macros.
The _explicit versions only have two parameters, namely the object and
the order. There is no need to pass the values of the atomic variable.
2012-01-13 16:01:34 +00:00
Ed Schouten
9e16bab42a Add aligned_alloc(3).
The C11 folks reinvented the wheel by introducing an aligned version of
malloc(3) called aligned_alloc(3), instead of posix_memalign(3). Instead
of returning the allocation by reference, it returns the address, just
like malloc(3).

Reviewed by:	jasone@
2012-01-09 06:36:28 +00:00
Ed Schouten
5d8c5f69f1 Fix spelling of C11 and sort functions by name. 2012-01-08 12:03:46 +00:00
Ulrich Spörlein
2f98c0ba01 Spelling fixes for include/ 2012-01-07 16:10:15 +00:00
Konstantin Belousov
5d8aec9c02 Implement fdlopen(3), an rtld interface to load shared object by file
descriptor.

Requested and tested by:	des (previous version)
Reviewed by:  des, kan (previous version)
MFC after:    2 weeks
2012-01-07 10:33:01 +00:00
Ed Schouten
305bf119f8 Last attempt at <tgmath.h>: do enable the new code for C11 compilers.
I was thinking by myself, if the new code doesn't work with GCC 4.2, why
not simply turn it into an efficient version for C11 compilers? By
changing the code to use _Generic() directly in that case, I can build
the tgmath regression test in a matter of milliseconds with Clang,
instead of the 8 seconds it used to take.

So by the time C11 becomes the default, it will pick up the new code
automatically. And now I will refrain from making more changes to
<tgmath.h>.
2012-01-06 19:04:59 +00:00
Ed Schouten
81ec6b6b47 Add an even faster implementation of <tgmath.h> for the future.
Instead of using an exponential number of cases with respect to the
number of arguments, this version only uses a linear number.
Unfortunately, it works with Clang, GCC 4.6 and GCC 4.7, but not GCC
4.2. Therefore, leave it commented out.
2012-01-06 16:20:17 +00:00
Ed Schouten
ff3e248a90 Remove redundant inclusion of <sys/cdefs.h>.
Mea culpa.
2012-01-05 12:09:39 +00:00
Ed Schouten
cede7d22fe Remove _Complex_I workaround from <tgmath.h>.
This removes the use of __typeof() from this header, thus making it work
with a plain C11 compiler.
2012-01-05 12:06:49 +00:00
Ed Schouten
f92d9d7d37 Make _Complex_I a proper float _Complex when using GCC 4.2.
It turns out our GCC has quite an interesting bug:

	typeof(1.0fi) != float _Complex
	typeof((float _Complex)1.0fi) != float _Complex
	typeof((float _Complex)1.0i) == float _Complex

In other words: if casting to an equal size, GCC seems to take a
shortcut. By casting down from a double to a float, GCC doesn't take
this shortcut, yielding the proper type.

To prevent foot-shooting, add a _Static_assert() to guarantee that
_Complex_I is always a float _Complex. I'm not going to MFC this part of
the diff.

MFC after:	2 weeks
2012-01-05 12:05:48 +00:00
Ed Schouten
069a43171f Reimplement <tgmath.h> on top of __generic().
The macro construction used now, is almost identical to the code
provided in C11 proposal N1404. This new version doesn't seem to
introduce any regressions according to the regression test in tools/,
but still seems to malfunction with Clang on certain aspects.

The new code does work successfully with GCC 4.2, 4.6 and 4.7. With 4.7,
it also works when __generic() is implemented on top of _Generic().

Discussed with:	stefanf
2012-01-05 10:46:22 +00:00
Ed Schouten
112f210647 Simply disallow <stdnoreturn.h> to be used in combination with C++.
There is no way one could possibly use this header file in combination
with C++ code. The problem is that in C11 the `noreturn' macro expands
to the `_Noreturn' function specifier, while in C++11 the `noreturn'
keyword is an attribute.

So in C11 you have to write:

	noreturn void exit(int status);

While in C++11 you have to write:

	[[noreturn]] void exit(int status);

It is impossible to #define noreturn for C++ in such a way that it
allows both conventions.

By intentionally breaking this header this way, we prevent people from
using this header in their C++<11 sources.
2012-01-03 23:05:23 +00:00
Ed Schouten
1e234009e3 Remove extraneous semicolons.
These macros are supposed to be invoked as regular functions, so remove
them.
2012-01-02 22:58:32 +00:00
Ed Schouten
092c20cb19 Extend <stdatomic.h> to support GCC 4.7's __atomic.
The development version of GCC also supports an atomics interface
similar to Clang's. Change the header file to work as follows:

- __CLANG_ATOMICS: Use Clang's new atomics interface,
- __GNUC_ATOMICS: Use GCC's new atomics interface,
- else: fall back to GCC's __sync interface.
2011-12-30 01:37:25 +00:00
Ed Schouten
ffa01562fd Don't define static_assert for C++.
Even though _Static_assert() is pretty robust for C code, it cannot work
correctly with C++ code.  This is due to the fact that C++ template
parameters may contain commas that are not enclosed in parentheses. For
example:

	static_assert(foo<int, int>::bar == baz, "...");

This causes _Static_assert to be called with an excessive number of
parameters.  If you want to use static_assert in C++, just use a C++11
compiler.

Reported on:	current@, ports@
2011-12-29 14:41:17 +00:00
Xin LI
c8fb747dc1 In POSIX.1-2008:
P_tmpdir [OB XSI]  Default directory prefix for tempnam().

This macro is used in a lot of places in legacy applications,
and is why we see a lot of programs written for e.g. Linux
store volatile temporary files in /var/tmp and not /tmp.

MFC after:	2 months
2011-12-28 05:35:33 +00:00
Ed Schouten
d4ae5eb992 Fix some bugs in <stdatomic.h>.
- Make atomic_init() work for GCC, as assigning to structs doesn't work.
- Fix misplaced parenthesis in atomic_is_lock_free() for GCC.
- Make atomic_compare_exchange_strong() for GCC return the proper
  boolean value, whether object == expected.
- Fix argument passing in atomic_exchange_explicit() for GCC.
2011-12-26 23:33:41 +00:00
Ed Schouten
c6d3530d85 As per C11, add static_assert() to <assert.h>. 2011-12-26 18:57:59 +00:00
Ed Schouten
b1214a51cc Improve C11 bits in <stdlib.h>:
- Add missing semicolon to quick_exit(),
- Remove `func' parameter name from at_quick_exit(),
- Fix indentation.
- Compare against 2011 value.
2011-12-26 18:55:37 +00:00
Ed Schouten
36893c3ded Make white space in this file a bit more consistent.
Remove trailing whitespace and place all macro definitions at the same
column.
2011-12-25 21:05:35 +00:00
Ed Schouten
d9171fdcc9 Fix field name.
The value field in the atomic structure is called __val; not value.
2011-12-25 20:59:39 +00:00
Ed Schouten
367bebd402 Add <stdalign.h> and <stdnoreturn.h>.
Even though these header files make little sense to me, they are part of
the standard. By including these header files, you can simply use
`alignas', `alignof' and `noreturn' instead of the underscore-prefixed
versions.
2011-12-25 20:51:40 +00:00
Ed Schouten
576fb69bc9 Remove unneeded guard.
There is no reason why <stdbool.h> needs an include guard. It is already
protected by __bool_true_false_are_defined.
2011-12-25 20:15:41 +00:00
David Chisnall
eb1f614be5 Restore __is_threaded in C++ mode. Some Google stuff needs it apparently.
Reported by:	swills
Approved by:	dim (mentor)
2011-12-25 16:03:54 +00:00
David Chisnall
d1b02beefa ...and actually install it.
Approved by:	dim (mentor)
2011-12-24 15:31:06 +00:00
David Chisnall
50af8167d6 Initial implementation of stdatomic.h. Works (at least in the parts that were
tested) with clang and gcc, and more efficiently with clang+a big out-of-tree
diff that I need to commit soon (once it's been tidied and reviewed a bit).

Large portions by: ed
Reviewed by:	ed
Approved by:	dim (mentor)
2011-12-24 15:17:01 +00:00
Colin Percival
3e65b9c6e6 Fix a problem whereby a corrupt DNS record can cause named to crash. [11:06]
Add an API for alerting internal libc routines to the presence of
"unsafe" paths post-chroot, and use it in ftpd. [11:07]

Fix a buffer overflow in telnetd. [11:08]

Make pam_ssh ignore unpassphrased keys unless the "nullok" option is
specified. [11:09]

Add sanity checking of service names in pam_start. [11:10]

Approved by:    so (cperciva)
Approved by:    re (bz)
Security:       FreeBSD-SA-11:06.bind
Security:       FreeBSD-SA-11:07.chroot
Security:       FreeBSD-SA-11:08.telnetd
Security:       FreeBSD-SA-11:09.pam_ssh
Security:       FreeBSD-SA-11:10.pam
2011-12-23 15:00:37 +00:00
Eitan Adler
0d6d372c03 - Add restrict keyword to glob(3)
PR:		kern/161958
Submitted by:	Henning Petersen <henning.petersen@t-online.de>
Approved by:	jilles
MFC after:	3 days
2011-12-20 22:56:13 +00:00
Ed Schouten
f6ab8089c6 Replace __const by const in all non-contributed source code.
As C1X is close to being released, there is no need to wrap around a
feature that is already part of C90. Most of these files already use
`const' in different placed as well.
2011-12-13 13:32:56 +00:00
Matthew D Fleming
108cd49471 - Define true and false in sys/types.h for _KERNEL consumers, and
typedef bool.  Due to macro expansion it seemed better to use a
   typedef for kernel consumers (specifically ofed won't compile
   without more changes if a define is used).
 - <stdbool.h> should also not re-define bool/true/false if they are
   defined by <sys/types.h>.  It would probably be a programming error
   to define _KERNEL for user-space code, but downstream consumers
   like Isilon have already been including <stdbool.h> in kernel
   sources, and this protects that usage.
 - sizeof(_Bool) is not necessarily the same as sizeof(int), so kernel
   modules should be rebuild with this change.  Bump __FreeBSD_version.

MFC after:	2 weeks
Sponsored by:	Isilon Systems, LLC
2011-12-12 18:44:17 +00:00
David Chisnall
57979d1bd8 As per das@'s suggestion, s/__noreturn/_Noreturn/, since the latter is an
identifier reserved for the implementation in C99 and earlier so there is
no sensible reason for introducing yet another reserved identifier when we
could just use the one C1x uses.

Approved by:	brooks (mentor)
2011-12-07 21:17:50 +00:00
David Chisnall
0a31efe016 Implement quick_exit() / at_quick_exit() from C++11 / C1x. Also add a
__noreturn macro and modify the other exiting functions to use it.

The __noreturn macro, unlike __dead2, must be used BEFORE the function.
This is in line with the C and C++ specifications that place _Noreturn (c1x)
and [[noreturn]] (C++11) in front of the functions.  As with __dead2, this
macro falls back to using the GCC attribute.

Unfortunately, clang currently sets the same value for the C version macro
in C99 and C1x modes, so these functions are hidden by default.  At some
point before 10.0, I need to go through the headers and clean up the C1x /
C++11 visibility.

Reviewed by:	brooks (mentor)
2011-12-07 15:25:48 +00:00
David Chisnall
3c87aa1d3d Implement xlocale APIs from Darwin, mainly for use by libc++. This adds a
load of _l suffixed versions of various standard library functions that use
the global locale, making them take an explicit locale parameter.  Also
adds support for per-thread locales.  This work was funded by the FreeBSD
Foundation.

Please test any code you have that uses the C standard locale functions!

Reviewed by:    das (gdtoa changes)
Approved by:    dim (mentor)
2011-11-20 14:45:42 +00:00
David Chisnall
c3b43c58d0 Hide some more macros that will break C++ when compiling in C++ mode.
Approved by:	dim (mentor)
2011-11-13 17:07:26 +00:00
David Chisnall
c9e7ce2fe1 The spec says that FILE must be defined in wchar.h, but it wasn't. It
is now.  Also hide some macros in C++ mode that will break C++
namespaced calls.

Approved by:	dim (mentor)
2011-11-13 16:18:48 +00:00
David Chisnall
22e24cfdc1 Don't expose the wctype.h macros in C++ mode. They cause problems when
people try to invoke the namespaced versions of the functions of the
same names.

Approved by:	dim (mentor)
2011-11-12 20:45:10 +00:00
David Schultz
a50079b7ff People porting FreeBSD to new architectures ought not have to
implement a deprecated FPU control interface in addition to the
standard one.  To make this clearer, further deprecate ieeefp.h
by not declaring the function prototypes except on architectures
that implement them already.

Currently i386 and amd64 implement the ieeefp.h interface for
compatibility, and for fp[gs]etprec(), which doesn't exist on
most other hardware.  Powerpc, sparc64, and ia64 partially implement
it and probably shouldn't, and other architectures don't implement it
at all.
2011-10-21 06:41:46 +00:00
David Schultz
3daee1d6c3 Add c{cos,sin,tan}{,h}{,f} functions. This is joint work with
bde and kargl.
2011-10-17 05:41:03 +00:00
Jung-uk Kim
7331c16912 Avoid accidental conflicts with C++ operator keywords.
MFC after:	1 week
2011-09-27 16:33:17 +00:00
Konstantin Belousov
a01fdfcef1 Install ciss(4) ioctl header (together with other .h files from sys/dev/ciss).
PR:	kern/109813
Discussued with:	Alex Samorukov <samm os2 kiev ua>
	(smartmontools maintainer)
MFC after:	1 week
2011-09-27 12:14:43 +00:00
Kevin Lo
d0b2bb9e86 Revert the previous change and add xdr_sizeof
Requested by:	bde
2011-07-10 14:03:13 +00:00