(based on patch from Boris Faure)
NB: This is untested/unsupported code. Ymmv, but using/compiling
ecore_x with versions of xcb > 0.3.6 is not supported yet (until such
time that standard distros support 0.3.8 out of the box).
SVN revision: 61971
Thread safety is disabled by default.
Enable it with --enable-thread-safety
Should cover timers, events, animators, idlers and fd handlers.
Tested with Enlightenment and elementary_test.
Signed-off-by: Mike McCormack <mj.mccormack@samsung.com>
SVN revision: 61851
near finished yet). Add working OpenGL for XCB engine.
NB: wrt Opengl...Raster, this is the env var/dlsym version you
requested this morning ;)
NB: Basically what happens is, if you know you do not ever want/use
opengl, you can export ECORE_X_NO_XLIB env variable, and ecore_x will
use pure xcb to establish it's X connection. However, if you do use
OpenGL and this env var is not exported, then ecore_x(cb) will use
XOpenDisplay to init the connection.
SVN revision: 61724
NB: IF you are going to try this, build evas with
`--enable-software-xcb` AND build ecore with `--enable-ecore-x-xcb
--disable-ecore-evas-opengl-x11`.
NB: OpenGL does NOT work with the xcb stuff yet. E itself does NOT
work with this yet either (still have to commit those changes).
SVN revision: 61385
glib only allows millisecond resolution in g_main_loop.
To avoid this limitation, use timerfd to wake up the main loop.
Signed-off-by: Mike McCormack <mj.mccormack@samsung.com>
SVN revision: 61079
Subject: [E-devel] XRender engine causes ecore build failure
while building ecore. The problem is that this engine was removed from evas
but not yet completely from ecore. I was on IRC with Vincent Torri (vtorri)
and Daniel Juyung Seo (SeoZ) and the consensus was to remove the code for the
XRender engines, both the Xlib and XCB versions.
There is a switch over the different engine types, where there are still a few
places left where XRender is handled, grep for "xrender" or "XRENDER" and you
will find them. The question is whether to just return NULL in order to signal
that this engine is not supported or to remove the whole thing. The latter
could break binary compatibility, therefore I left those stubs in.
SVN revision: 60502
The ecore_con module needed a port of the local connections
with named pipes. The other connections (TCP, UDP) are using
BSD sockets, which are also used on Windows.
No abstract sockets on Windows.
NB: Should I backport that commit to 1.0 ?
SVN revision: 59385
Use isfinite() if available, otherwise use finite() on
compilers != vc++, otherwise use _finite()
and a bit of formatting too (i know, it's bad)
SVN revision: 58566
* check if clock_gettime() is in libc first, then in librt
some systems have these functions in libc, or in a specific lib.
This allows to correctly set dlopen_libs and rt_libs variables.
SVN revision: 55821
simply a pain. no more tests within src trees. talk to gentoo if you
don't like it. i wasted enough of my day already trying to talk sense
into them. if we dont tempt them with stuff they dont comprehend then
its less pain for us having to answer their questions.
SVN revision: 55635
This is an attempt to fix the epoll/fork() issue reported to me where
we end up with a single epoll fd shared between two processes after a
fork() in E.
I've tested with elementary test in epoll and non-epoll combinations,
and all appears to be well. Please check it solves the issue you saw,
and reformat the code as you see fit... ;-)
SVN revision: 53670
AC_CHECK_FUNCS checks for the existence of functions in the C standard
library, so we don't need it. Instead, we need to define
HAVE_CLOCK_GETTIME if the function was found inside the librt.
Moreover, in source file check if HAVE_CLOCK_GETTIME is defined rather
than of checking if it's 0.
SVN revision: 52863
Instead of relying on unix time, use a monotonic clock provided by
clock_gettime(). If a monotonic clock is not available, it will fallback
to CLOCK_REALTIME or unix time if neither is available.
The impact is that now it only makes sense to call ecore_time_get() or
ecore_time_loop_get() if the value retrieved is intended to be used as
relative to previous/posterior measurements. If an absolute value is
needed, the right function to call now is ecore_time_unix_get() which
will give the number of seconds since Jan 1st, 1970, 12:00AM.
SVN revision: 52824