353 Commits

Author SHA1 Message Date
jca
15f9099d0a Move to libelf from base
Initial diff from jsg@, ok jasper@ (maintainer of
libdwarf/ctftools/libelf).
2019-02-04 23:34:10 +00:00
ajacoutot
f75f8368d7 Rework the gunixmount.c patch and add more checks. 2019-01-23 10:06:21 +00:00
ajacoutot
a88625b0b2 sysutils/toad is coming back, teach glib to use pkexec(1). 2019-01-23 03:37:09 +00:00
ajacoutot
54401f636c Update to glib2-2.58.3. 2019-01-21 21:52:58 +00:00
jca
b445abf2ab Unbreak on gcc archs after the update to 2.58.2
glib2 checks for __sync_bool_compare_and_swap, and requires
__GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 to be defined if the function is
available... except with special cases like Linux armv5.  base gcc
implements __sync_bool_compare_and_swap on many of our archs but doesn't
provide __GCC_HAVE_* macros, so glib2 is confused.  Extend the
existing workaround to unbreak.

ok ajacoutot@ (maintainer)
2019-01-10 13:36:26 +00:00
sthen
5ddfd9e61b unbreak some glib2 users on 32-bit arches, ok aja@
The old autoconf build has the following additional check after
determining a valid type for gsize, which is not present in meson.build.:

dnl If int/long are the same size, we see which one produces
dnl warnings when used in the location as size_t. (This matters
dnl on AIX with xlc)
dnl
AS_IF([test $ac_cv_sizeof_size_t = $ac_cv_sizeof_int &&
       test $ac_cv_sizeof_size_t = $ac_cv_sizeof_long],

It matters on OpenBSD too. Since my meson-fu is lacking, the patch
simply reorders the if statement to put the one we need earlier.
2019-01-05 22:00:20 +00:00
ajacoutot
58d7383993 Needs shells/bash at build time to install the completion snippets. 2019-01-01 12:48:06 +00:00
ajacoutot
f2f94c039a Update to glib2-2.58.2.
This is the first step towards the GNOME 3.30 update which will happen
this month.
This and subsequent updates have been tested in an amd64 bulk.
2019-01-01 11:32:15 +00:00
ajacoutot
343054bed8 Update to glib2-2.56.4. 2018-12-19 10:36:04 +00:00
jasper
9c62776c59 /usr/share is not a valid data directory on OpenBSD for xdg/glib purposes
so remove it from the list returned by g_get_system_data_dirs().

ok aja@
2018-09-24 20:00:59 +00:00
ajacoutot
706d1c70f3 Update to glib2-2.56.3. 2018-09-22 08:39:31 +00:00
espie
f4b7f81318 convert to PKGSTEM 2018-09-04 12:46:09 +00:00
ajacoutot
6353ef3978 Put back the "un-harcode /usr/local" fix in. This will leave an empty
${PREFIX}/<...>/usr directory at delete time in case PREFIX != /usr/local but
espie@ says it's OK.

prodded by espie@
2018-08-23 08:36:07 +00:00
ajacoutot
8aab46a20b We're using autoreconf(1) etc. so we must patch Makefile.am, not .in. 2018-08-23 07:16:36 +00:00
ajacoutot
a5faaf5992 Update to glib2-2.56.2. 2018-08-17 08:03:57 +00:00
ajacoutot
2ede011167 Err, revert previous. There's still /usr to handle. 2018-08-15 15:33:39 +00:00
ajacoutot
31d6a953a8 Un-hardcode /usr/local.
spotted by espie@
2018-08-15 15:27:35 +00:00
espie
bb921d2165 use AUTOCONF_ENV and add correct env variables, so that those ports work
even if /usr/local is not part of the builder's path.
2018-08-10 10:19:18 +00:00
ajacoutot
6d59cf24f8 Fix issues number. 2018-08-09 07:29:38 +00:00
ajacoutot
1250129c77 Re-add @tag at the top (next to define-tag).
This way the associated command will be run on update (at-end) and at
delete time (at the time it is encountered in the PLIST).

discussed wth espie@
2018-07-30 08:34:07 +00:00
ajacoutot
ddd7fdcb77 Make sure gint64 matches int64_t so that C++ is happy
with and OK espie@
2018-07-11 12:03:22 +00:00
espie
72bc1e0558 when we delete this, we don't need to run tag a last time 2018-07-11 11:21:12 +00:00
espie
cd12df3943 misc small tags:
- gio-querymodules
- gtk-query-immodules*
2018-07-07 22:39:42 +00:00
espie
2a95b6e111 tag glib-compile-schemas 2018-07-06 06:04:41 +00:00
ajacoutot
966f798401 Committed upstream. 2018-04-23 19:51:21 +00:00
ajacoutot
78f5a5025d Update to glib2-2.56.1.
First step of the GNOME 3.28.2 update work that will happen at p2k18.
2018-04-22 13:27:25 +00:00
mpi
0a4cc523f3 Reorder operations in _kqsub_cancel() to prevent races.
Should prevent the timeout to refill the list of depenencies while
another thread is freeing it.  Should help with segfault reported
on bugs@.

While here sync the kqueue rewrite with upstream.

ok ajacoutot@
2018-04-12 10:03:17 +00:00
mpi
d7944fdcc9 Sync with latest bugzilla submission. 2018-02-20 16:59:19 +00:00
mpi
b45078bdb2 GIO kqueue(2) backend redesign.
- Stop using a socketpair(2) to communicate between two threads of
  the same process.

- Call kevent() in non-blocking mode such that it can be integrated
  in the poller thread of glib.

- Use glib's abstraction to poll(2) the kqueue instead of adding it
  to itself Oo.

- Remove the secondary thread, the associated global hash table, its
  broken locking and a lot of abstraction.

- Properly refcount sources.

/!\ This backend is still not generating the events Glib is expecting.

Please report any problem such that this can be pushed upstream.

ok ajacoutot@
2018-01-23 10:43:42 +00:00
ajacoutot
0a26bdfd51 Update to glib2-2.54.3. 2018-01-09 12:31:54 +00:00
bentley
868e9749a3 Pledge glib-compile-schemas, which is @exec'd by 100+ packages.
ok ajacoutot@ jasper@
2017-11-03 13:16:05 +00:00
ajacoutot
7ebe7bae3e Update to glib2-2.54.2. 2017-10-28 06:19:08 +00:00
ajacoutot
40dff462c8 Update to glib2-2.54.1. 2017-10-26 08:48:59 +00:00
ajacoutot
a79597aace Update to glib2-2.52.3. 2017-06-23 07:29:17 +00:00
ajacoutot
42f8de2788 Update to glib2-2.52.2. 2017-05-08 16:57:08 +00:00
mpi
b0d2c8685b Fix a mutex recursion introduced in my last diff and protect another
code path with the same lock.

ok ajacoutot@
2017-05-05 12:04:58 +00:00
mpi
0a83fc77ae Prevent a use-after-free resulting in a crash of all apps depending
on glib/GIO mostly at "pkg_add -u" time.

You can now run "pkg_add -u" inside GNOME 3 without any crash.

Diff already submitted upstream.

ok ajacoutot@
2017-04-28 13:29:22 +00:00
ajacoutot
46dfc195ab "Everything is a Freaking GLib/GTK+ problem"
Compile with debug symbols and don't strip on install(1). This will ease
debugging all the many things that depend on and crash within these libs. We'll
see how this works out in the long run.
req. by mpi@
2017-04-28 11:16:29 +00:00
ajacoutot
3d1fc8c9f8 Stop patching for /var/lib/dbus -> /var/db/dbus; nowadays, most apps look
into /etc/machine-id *before* looking into /var/lib/dbus/machine-id and for
the few that don't patch them to look directly there.

Thanks sthen@ for the source ports seach.
2017-04-22 13:43:21 +00:00
ajacoutot
65421852af Update to glib2-2.52.1.
- survived a bulk
- this officially starts the work on updating GNOME to 3.24.1
2017-04-11 07:25:38 +00:00
ajacoutot
052d8274b1 Update to glib2-2.50.3. 2017-02-13 16:07:52 +00:00
ajacoutot
116e571341 Update to glib2-2.50.2. 2016-11-07 16:24:59 +00:00
ajacoutot
9626203092 Update to glib2-2.50.1.
This starts the GNOME 3.22 update which will be finished next week at b2k16.
2016-10-28 13:40:04 +00:00
ajacoutot
14b95f2427 Update to glib2-2.48.2.
- drop the gettext MODULE
2016-08-21 12:39:47 +00:00
ajacoutot
2a83d0db61 Update to glib2-2.48.1. 2016-05-10 10:02:58 +00:00
ajacoutot
e0d27de83e Update to glib2-2.48.0. 2016-04-18 06:46:30 +00:00
ajacoutot
f9083360bb Committed upstream. 2015-12-23 14:43:58 +00:00
ajacoutot
485961c8e7 Sync patch with what's in the BZ. 2015-12-22 08:52:02 +00:00
ajacoutot
5722e8cb06 Update to glib2-2.46.2. 2015-11-07 05:52:09 +00:00
ajacoutot
c4d48f1460 Better workaround for BZ #753378. 2015-10-18 06:33:54 +00:00