This should help getting static builds working on mingw.
Based on a patch by Volker Grabsch, bug 619126.
At the same time, drop the unnecessary GLIB_RT_LIBS variable;
we are already adding -lrt to G_THREAD_LIBS.
If we don't do the cast to the proper size in 32 bits, things like below
doesn't work:
uint8_t u = 20;
void *p;
p = GUINT_TO_POINTER(u);
Signed-off-by: Colin Walters <walters@verbum.org>
https://bugzilla.gnome.org/show_bug.cgi?id=661546
Add GNetworkMonitor and its associated extension point, provide a base
implementation that always claims the network is available, and a
netlink-based implementation built on top of that that actually tracks
the network state.
https://bugzilla.gnome.org/show_bug.cgi?id=620932
For some reason, the setting of g_atomic_lock_free wasn't making it down
to the lower part of the configure script where glibconfig.h was being
generated when building using mingw32-configure.
If we prefix glib_cv_ to the start of the variable name (like everyone
else is doing) then it magically starts working.
I love you, automake.
We clean up the detection of if we should do 'real' atomic operations or
mutex-emulated ones with the introduction of a new (public) macro:
G_ATOMIC_LOCK_FREE. If defined, our atomic operations are guaranteed to
be done in hardware.
We need to use __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 to determine if our
compiler supports GCC-style atomic operations from the gatomic.h header
because we might be building a program against GLib using a different
set of compiler options (or a different compiler) than was used to build
GLib itself.
Unfortunately, this macro is not available on clang, so it has currently
regressed to using the mutex emulation. A bug about that has been
opened here:
http://llvm.org/bugs/show_bug.cgi?id=11174
These were meant to provide the ability to use GLib directly out of its
build tree but have a couple of rather serious limitations that prevent
their serious use:
- these files only worked with programs built using libtool
- for non-trivial uses, other environment variables will need to be
setup to allow GLib to function properly
- the system doesn't stack well when attempting to use multiple
libraries all uninstalled
The jhbuild --prefix= style system works much better in every way, so we
remove this old hack.
We can just assume that strerror/strsignal are available
nowadays. At the same time, drop use of thread-private storage.
Instead, always return interned strings.
https://bugzilla.gnome.org/show_bug.cgi?id=660849
Now that GMutex is exposed we can avoid the dance we did in ./configure
to allocate the correct amount of space for it within the GStaticMutex.
Remove the checks and move the definitions to gthread.h, trying very
hard to keep ABI-stable (even though we will be deprecating this soon).
Implement g_ascii_strto{d,ll,ull} and g_ascii_formatd using
xlocale functions where available. This is slightly faster
and a lot less icky than our homegrown code.
https://bugzilla.gnome.org/show_bug.cgi?id=640293
configure.ac defined G_THREAD_SOURCE and gthread-impl would #include it.
Instead, since we only have two thread implementations now, and since we
always use the Windows one only on Windows, move the logic to the
Makefile, predicated on 'if OS_WIN32'. Then have the chosen backend do
the #include "gthread-impl.c" from there.
Remove the G_THREAD_SOURCE define from configure.ac.
In particular, remove the libasyncns import, which was only used by
GUnixResolver, which is only used when threads are not available.
Likewise remove GWin32Resolver, and the hacky broken non-threaded
parts of GIOScheduler.
https://bugzilla.gnome.org/show_bug.cgi?id=616754
G_THREADS_ENABLED still exists, but is always defined. It is still
possible to use libglib without threads, but gobject (and everything
above it) is now guaranteed to be using threads (as, in fact, it was
before, since it was accidentally impossible to compile with
--disable-threads).
https://bugzilla.gnome.org/show_bug.cgi?id=616754
We had the 12 hour time format hard-coded to "%02d:%02d:%02d %s" but it
actually changes depending on the locale. Just with the other formats,
use nl_langinfo() if we have it, otherwise fall back on gettext().
- getmntinfo can take struct statfs or statvfs depending on the
OS. Use getvfsstat and if not found getfsstat instead. Idea from
Dan Winship.
- g_local_file_query_filesystem_info(): use statvfs.f_fstypename
if available
https://bugzilla.gnome.org/show_bug.cgi?id=617949
Improve a few situations where g_date_time_format() was getting the
padding wrong when displaying alt digits (eg: Arabic numerals) for
formatting time.
We now depend on nl_langinfo (_NL_CTYPE_OUTDIGITn_WC) to do the
conversion, which is very likely glibc-specific, but our previous method
relied on a glibc-specific printf() feature, so no harm done there.
Add a configure check for nl_langinfo (_NL_CTYPE_OUTDIGITn_WC).
Uncomment a few testcases that were failing previously.
https://bugzilla.gnome.org/show_bug.cgi?id=658107
Also, link libgio to -lresolv explicitly, rather than depending on
getting it implicitly via the libasyncns build (which should
eventually be going away).
https://bugzilla.gnome.org/show_bug.cgi?id=645336
This makes the source efficient on Linux.
Tested on Fedora 15 x86_64 + updates, kernel-2.6.40-4.fc15.x86_64
Also tested fallback code for unsupported flag TFD_TIMER_CANCEL_ON_SET
on kernel 2.6.38.
https://bugzilla.gnome.org/show_bug.cgi?id=655129
This makes g_date_time_format() react to LC_TIME, which is
what people expect.
Translators: this change means that the GDateTime strings
are only used when the C library does not already provide
suitable translated strings for these (month names, etc).
We need to test the case of eventfd in the libc but no kernel support.
In order to do that, we add a separate compile of the GWakeup testcase
that interposes an 'eventfd' symbol that always returns -1 with errno
set. That will trigger the fallback case.
- move choice of statfs vs statvfs from gio/glocalfile.c to configure.ac
- if statvfs is the choice, then don't check number of arguments to statfs()
- use choice in gio/gunixmounts.c as well
https://bugzilla.gnome.org/show_bug.cgi?id=617949
The Linux eventfd() call is basically tailor made for the main loop
wake up pipe - all we want is a threadsafe way to write to a file
descriptor, and wake up the context on the other end; we don't care
about the content at all.
The eventfd manual page basically explains the benefits:
Applications can use an eventfd file descriptor instead of a
pipe (see pipe(2)) in all cases where a pipe is used simply to
signal events. The kernel overhead of an eventfd file
descriptor is much lower than that of a pipe, and only one file
descriptor is required (versus the two required for a pipe).
When writing my multithreaded spawn test case I actually hit the 1024
file descriptor limit quickly, because we used 2 fds per main context.
This brings that down to 1.
https://bugzilla.gnome.org/show_bug.cgi?id=653140
We can use AC_CHECK_FUNCS to detect if qsort_r is available on
the system or not since it will unconditionnally define
HAVE_QSORT_R, which we don't want since on BSD, qsort_r isn't usable
for us, so we don't want to have HAVE_QSORT_R defined on such platforms.
By using AC_CHECK_FUNC instead, we can defer defining HAVE_QSORT_R until
we have actually tested it's usable.
https://bugzilla.gnome.org/show_bug.cgi?id=651920
When cross compiling, test programs cannot be run. In order to make it
easier to cross compile for targets that do not have qsort_r(), check
for the function before trying to run the test program.
This avoid having to populate cache with glib_cv_have_qsort_r=no with
e.g. mingw cross compiler.
https://bugzilla.gnome.org/show_bug.cgi?id=651920
This avoids the generated types (e.g. ExampleAnimal, ExampleCat,
ExampleObject and ExampleObjectManagerClient) being referenced in the
core gio docs. This was requested by Matthias.
Signed-off-by: David Zeuthen <davidz@redhat.com>
Commit 22e7fc34c4 introduced a regression:
futexes were always disabled and the emulated codepath was always being
used. That commit was in response to an originally buggy
implementationt hat wrote junk into config.h (but happened to be working
properly).
Fix up the mess and while we're at it, close bug #631231 by including
syscall.h from the correct location and using __NR_futex instead of
SYS_futex.
Closes#631231.
When using gcc builtins for atomic operations, provide them
as macros, so gcc can see the builtins and do optimizations.
This change gives considerable speedups in bitlocks, which
use atomic operations heavily, see bug 650458.
Also, don't define G_ATOMIC_OP_MEMORY_BARRIER_NEEDED unconditionally
when using gcc builtins.
https://bugzilla.gnome.org/show_bug.cgi?id=617491
_GNU_SOURCE must be defined before including any other (system)
header, so defining it in glib-unix.h (and hoping no one has included
anything else before that) is wrong. And the "#define _USE_GNU"
workaround for this problem in gnetworkingprivate.h is even wronger
(and still prone to failure anyway due to single-include guards).
Fix this by defining _GNU_SOURCE in config.h when building against
glibc. In theory this is bad because new releases of glibc may include
symbols that conflict with glib symbols, which could then cause
compile failures. However, most people only see new releases of glibc
when they upgrade their distro, at which point they also generally get
new releases of gcc, which have new warnings/errors to clean up
anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=649201
autoconf 2.68 is very insistent that AC_LANG_SOURCE/AC_LANG_PROGRAM
must be used in certain places, to avoid quoting/lack-of-quoting
problems, or something. Fix.
