Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
Revert "evas: Fix use of invalid EO object"
This reverts commit
9b1e785f8ebba618016dd217aa18835b4fb675b7.
this breaks devices with vt switching under drm
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
elput: fix seat cleanup
ensure seat deletion does not conflict with elput shutdown
@fix
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
elput: unref devices in post event cb
@fix
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
elm_table: do not propagate max size
evas table does not set max size, propagating this triggers unnecessary
callbacks
@fix
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
elm_box: use correct aspect ratio for HORIZONTAL layout
ratio should be flipped in this case
@fix
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
elm_box: use homogeneous layout if only one child exists
this should be a small speedup for a corner case of box use
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
elm_win: do not implement aspect_get()
this breaks aspect hints.
@fix
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
efl-wl: escape command args from test
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
efl-wl: hide popup surface before unsetting attrs during destructor
ensure focus remains with the popup parent
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
efl-wl: simplify refocus on popup hide
Mike Blumenkrantz [Fri, 28 Jul 2017 19:44:43 +0000 (15:44 -0400)]
efl-wl: use correct window when requesting x11 clipboard selection
Guilherme Iscaro [Fri, 28 Jul 2017 18:22:07 +0000 (15:22 -0300)]
Ecore_Con: Fix wrong comparison.
Otherwise it will always evaluate to false.
Guilherme Iscaro [Fri, 28 Jul 2017 14:04:23 +0000 (11:04 -0300)]
Ecore_Conn: Set EOS only if the connection is not closed.#algo.
This check prevents that EOS is set on a closed connection.
Fixes T5180
Guilherme Iscaro [Fri, 28 Jul 2017 14:02:42 +0000 (11:02 -0300)]
Ecore_Con: Fix CLOEXEC documentation.
The socket attributes are inherited by the child.
Derek Foreman [Fri, 28 Jul 2017 14:40:40 +0000 (09:40 -0500)]
ecore_drm2: Allow picking a sequence number for blanktime_get
This lets us do a blocking wait for a vsync. Something we should try to
do as infrequently as possible, but in some cases we need it one time at
startup to catch graphics driver bugs.
Daniel Hirt [Fri, 28 Jul 2017 09:33:55 +0000 (12:33 +0300)]
Elm entry: fix call to set text via entry_set
Accidentally called the super here, so did not get routed to
'_elm_entry_text_set'.
Fixes T5803.
Amitesh Singh [Fri, 28 Jul 2017 07:07:21 +0000 (16:07 +0900)]
elm test: img.zoomable: allow top rect to pass events
Felipe Magno de Almeida [Thu, 27 Jul 2017 16:19:49 +0000 (13:19 -0300)]
eo-cxx: Add overload for Eina_Bool inout handling interoperability
Add convert_inout_impl overload to handle bool/Eina_Bool conversion in inout direction.
Guilherme Iscaro [Wed, 26 Jul 2017 21:48:45 +0000 (18:48 -0300)]
Ecore_Conn: Enable CLOEXEC by default.
This flag should be enabled by default in order to avoid socket leaks.
Guilherme Iscaro [Wed, 26 Jul 2017 15:11:29 +0000 (12:11 -0300)]
Ecore_IPC: Preserve Ecore_Con legacy behaviour.
This patch sets some Ecore_Con flags that were missing after
the EO migration. These flags must be set in order
maintain the Ecore_IPC behaviour before Ecore_Con EO was implemented.
Fixes T5722
Amitesh Singh [Thu, 27 Jul 2017 10:03:51 +0000 (19:03 +0900)]
efl.ui.image.zoomable: Add missing edje.group_size_min/max_get
This supresses the warnings when photocam is used as
an external edje object.
Jean-Philippe Andre [Thu, 27 Jul 2017 06:45:37 +0000 (15:45 +0900)]
evas: Always call show/hide intercept
Ref T5370
Marcel Hollerbach [Wed, 26 Jul 2017 16:56:59 +0000 (18:56 +0200)]
elm_widget: change to @inout
according to q66 this is correcter than ptr(..).
Anyhow this still leads to a bug in cxx that crashes compilation in cxx
examples, people are notified!
Marcel Hollerbach [Wed, 26 Jul 2017 13:52:12 +0000 (15:52 +0200)]
elm_widget: fix warning
WooHyun Jung [Wed, 26 Jul 2017 09:57:37 +0000 (18:57 +0900)]
ecore_events: inarray should be flushed before return
@fix
Jiyoun Park [Wed, 26 Jul 2017 06:28:02 +0000 (15:28 +0900)]
ecore_evas_wayland: fix bug ecore_evas cannot update the evas when rotation
usecase:
show -> rotation -> hide -> show
ecore_evas_wayland didn't check the rotation.
when ee is landscapemode, it cannot update the right area of evas.
Andy Williams [Tue, 25 Jul 2017 22:54:21 +0000 (23:54 +0100)]
elm_code: Fix missing middle button selection paste
Resolves task T5520
@fix
Derek Foreman [Tue, 25 Jul 2017 21:51:21 +0000 (16:51 -0500)]
elput: Close fds when asked to
Our close callback tells logind we're done with a device, but it should
also actually close the fd it's passed, or we end up leaking piles of
fds on VC switch.
see weston commit
8f5acc2f3a29c3831af4ddd6bed57f703c98dc77
and subsequent regression in commit
72dea06d7952e3ce8dd8057f7106186da4fa2678
and pending fix in https://patchwork.freedesktop.org/patch/168992/
Sungtaek Hong [Tue, 25 Jul 2017 05:44:39 +0000 (14:44 +0900)]
edje_cc: fix wrong comment about lazEDC norequired.
Summary: - norequired; stands for required: 0
Reviewers: conr2d, jpeg
Reviewed By: jpeg
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D5036
Jean-Philippe Andre [Thu, 20 Jul 2017 09:00:15 +0000 (18:00 +0900)]
elm: Move calls to efl_destructor(efl_super()) to the end
When destroying any object, its parent class destructor should
be called after the subclass destructor has been called. Only
some extremely limited work may be done after the super call.
This commit makes sure that all efl_destructor() overrides in
elementary are doing operations in the right order.
Also, remove a return void.
Marcel Hollerbach [Mon, 24 Jul 2017 21:35:26 +0000 (23:35 +0200)]
elm_widget: only emit events when there is really a changed manager
A manager change to NULL is to be used as "this object is not registered
anymore" call, which will happen quite regular during shutdown.
Marcel Hollerbach [Mon, 24 Jul 2017 15:58:13 +0000 (17:58 +0200)]
elm_widget: call the events after eval outself with new manager/parent
otherwise we might end up with a node_get error, since the widget where
the event is called on is still registered in the wrong manager
Marcel Hollerbach [Mon, 24 Jul 2017 15:54:31 +0000 (17:54 +0200)]
elm_widget: only set focus.manager if we are really registering
otherwise we might set focus.manager even if we are not registered
Marcel Hollerbach [Thu, 20 Jul 2017 15:37:03 +0000 (17:37 +0200)]
elm_widget: give feedback if registration was successfull or not
with this we can whipe out the focus.manager field in elm.widget so for
the case that something goes wrong we only get the error message where
actually something went wrong, and not the whole bunch of follow ups
where the code assumes its registered but it isnt.
Derek Foreman [Mon, 24 Jul 2017 21:03:10 +0000 (16:03 -0500)]
elput: Fix multiple open/close of drm devices
When I added the code to probe drm devices to ensure they're
modeset capable (ref
414d406b3b442216543cdaef112787696ae09898)
I didn't realize elput didn't allow us to open and close more than
one drm device at startup without blowing up libinput.
This is a somewhat dirty hack to rough that in.
The problem is that open/close the device during startup will
result in an async "gone" callback from logind, which then kicks
off an input shutdown. We need to try harder to only do that
shutdown when it makes sense.
Mike Blumenkrantz [Mon, 24 Jul 2017 12:28:39 +0000 (08:28 -0400)]
ecore-wl2: handle null offer drags more effectively
this is valid and refers to an offer with no types; a leave event
with no enter is a protocol error, however
fix T5770
@fix
Mike Blumenkrantz [Mon, 24 Jul 2017 12:28:39 +0000 (08:28 -0400)]
Revert "ecore wl2 - dnd - handle NULL drags... shouldnt happen but does"
This reverts commit
85e585846600742a7f395751816f348b4152269f.
please do not randomly add null checks for the purpose of quickly closing tickets.
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 09:53:33 +0000 (18:53 +0900)]
ecore wl2 - dnd - handle NULL drags... shouldnt happen but does
this should fix T5770
@fix
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 08:44:22 +0000 (17:44 +0900)]
efl net server udp - report erro on alloc failure
unwind nicely and complain
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 08:35:13 +0000 (17:35 +0900)]
ecore con proxy helper - complain when realloc fails
so you know why things might be failing... complain about out of
memory errors.
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 07:59:53 +0000 (16:59 +0900)]
ecore exe - pisix - handle malloc fails
handle malloc fails better
@fix
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 07:57:22 +0000 (16:57 +0900)]
ecore win32 exe handling - check realloc and malloc returns
handle out of memory errors better
@fix
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 05:10:17 +0000 (14:10 +0900)]
ecore_exe - handle realloc failures by complaining and rolling back
@fix
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 04:46:30 +0000 (13:46 +0900)]
embryo_cc - be sensible about realloc returns so analysers are happy
assingto tmp var then assign to real one after a check to analysers
dont complain.
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 04:37:57 +0000 (13:37 +0900)]
efreetd cache create - fix reallocs to bail on out of memory cleanly
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 04:26:28 +0000 (13:26 +0900)]
edje_cc - make svg loader (still beta) handle out of memory nicely
print ERR and abort.
Carsten Haitzler (Rasterman) [Mon, 24 Jul 2017 04:09:26 +0000 (13:09 +0900)]
edje convert - make realloc fatal and complain about it
cleaner error handling
Amitesh Singh [Mon, 24 Jul 2017 00:50:05 +0000 (09:50 +0900)]
edje_cc out: fix div by 0 correctly
Andy Williams [Sun, 23 Jul 2017 20:30:35 +0000 (21:30 +0100)]
elm_code: Fix crash with long lines
Also fixes issue where widget would sometimes blank when scrolling
@fix
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:48:16 +0000 (19:48 +0900)]
edje_cc out - fix theoretical div by 0
fixes CID 1371125
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:46:17 +0000 (19:46 +0900)]
edje_cc out ... fix theoretical div by zero...
if files size was 0... which would basically make a lot fail before...
it could be a divide by zero. fix and just assume minimum size of 1.
fix CID 1371127
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:42:06 +0000 (19:42 +0900)]
efl ui grid - handle invalid 0x0 fully if the grid size fails to do it
if efl_pack_grid_size_set is overridden and somehow allows 0x0 to
still happen then handle this caase and skip the layout.
fix CID 1374317
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:36:41 +0000 (19:36 +0900)]
edje_cc out ... check more iui list element ptrs to help coverity
fix CID 1374482 ... and others i hope.
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:33:58 +0000 (19:33 +0900)]
edje_cc out - double check list element ptr to make coverity happy
fix CID 1374483
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:30:31 +0000 (19:30 +0900)]
evas seat focus add - handle other case where layer may be null
fix CID 1377522
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:26:56 +0000 (19:26 +0900)]
eina debug timer - handle fail wakeup write with error complaint
fix CID 1377528
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:18:55 +0000 (19:18 +0900)]
add comment about poll high/medium/low implementation of events
not brilliant as it stands. :(
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:12:32 +0000 (19:12 +0900)]
ecore main - fix division to remain a double/float as expected
1/60 != 1.0/60.0 ... the first is an integer which is 0... the second
is 0.
016666666 .... a float/double. fix CID 1377532
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 10:09:11 +0000 (19:09 +0900)]
eina debug - don't try and collect stats on more than a million threads
sanity check... just in case.
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:56:43 +0000 (18:56 +0900)]
elm cnp - check if selbuf is null or not and skip copying if null
this should fix CID 1377536
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:53:46 +0000 (18:53 +0900)]
evas callbacks - check if obj->events is valid because it's cow alloced
it COULD be null and we do have other checks for it being not-null and
we dont check if the alloc fails... so check. this SHOULD fix
CID 1377547
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:43:58 +0000 (18:43 +0900)]
eina debug - add comments to cpu debug using locks to pause sysmon
this is intended so leave comments saying so. coverity doesn't like it
with CID 1377549
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:40:04 +0000 (18:40 +0900)]
eina debug - check return value of syscall for errors
this should fix CID 1377550
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:36:44 +0000 (18:36 +0900)]
efl canvas image - dbug ingo get - dont get if no root dbg info supplied
if root dbg info is null we'll leak and not have anything useful
stored somewhere for fetching anyway.
this should fix CID 1365656
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:29:57 +0000 (18:29 +0900)]
evas module load add comments pointing out this is intentional ptr loss
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:11:26 +0000 (18:11 +0900)]
efreetd cache - fix leak of strbuf when error conditions abort save
fix CID 1374951
Carsten Haitzler (Rasterman) [Sun, 23 Jul 2017 09:09:39 +0000 (18:09 +0900)]
efreetd cache - handle queue case leaking strbuf
fix CID 1374954
Carsten Haitzler (Rasterman) [Sat, 22 Jul 2017 08:59:11 +0000 (17:59 +0900)]
ecore_audio headers conflict with app have ifdefs
Ecore_Audio.h had #ifdef HAVE_PULSE ... and that is just so wrong as
this is something an ap[p may or may not set in its config.h ... but
this certainly hase no place in our public headers. the api's there
should always be there... symbols always... just imtplementation may
be empty ... though ecore_audio doesnt build at all without pulse...
so it's moot.
so fix this build issue i found when fixing e build issues with meson
changes etc.
@fix
Carsten Haitzler (Rasterman) [Fri, 21 Jul 2017 08:37:33 +0000 (17:37 +0900)]
efl_wl - also add xkbcommon as it's actually used
Derek Foreman [Fri, 21 Jul 2017 21:40:17 +0000 (16:40 -0500)]
ecore_drm2: Ensure device we find can mode set
Some systems have dri devices that can't mode set, and if they're first in
the directory they'll get picked by our code and things fall apart later.
So, open the potential device and ensure it has basic functionality before
selecting it.
This is a little inefficient as it gets the device via elput twice before
it can be used - this will be addressed later as the changes are a little
more invasive to optimize.
Derek Foreman [Fri, 21 Jul 2017 21:36:06 +0000 (16:36 -0500)]
ecore_drm2: Open elput manager before drm device
We're going to need to access the device as part of the find process,
so we need a manager.
Jean-Philippe Andre [Fri, 21 Jul 2017 03:14:44 +0000 (12:14 +0900)]
hover: Return valid slot for main_menu_submenu
Fixes T5760
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
ecore-evas-wayland: destroy frame callback on hide
ensure cursor surfaces can render again after being hidden
#TheDisappointer
@fix
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
eldbus: print full filename when codegen write fails
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
wayland: stop creating 1x1 surfaces on init
fix T5226
#IGot99TicketsBut1x1AintOne
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
elm_win: instantiate csd frame after creating pointer surface
csd attempts to create sw cursors if the pointer surface doesn't exist,
so create the pointer surface before the csd to avoid duplicate cursors
#TheDisappointer
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
elm_win: do not set invalid geometries on the window object
clamping to 1x1 is not valid, neither is trying to always position at 0,0
#IGot99TicketsBut1x1AintOne
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
elm_win: hide pointer surface on mouse out
this should be unmapped so edje stops trying to render to it, though
at some point it could be optimized to not hide and just reuse the buffer
when showing again
#IGot99TicketsBut1x1AintOne
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
elm_win: show pointer canvas after setting cursor
setting cursor calls resize, which is what sets the surface size
#IGot99TicketsBut1x1AintOne
Mike Blumenkrantz [Fri, 21 Jul 2017 20:17:55 +0000 (16:17 -0400)]
evas sw x11: formatting
Derek Foreman [Fri, 21 Jul 2017 19:00:51 +0000 (14:00 -0500)]
gl_drm: half fix eglGetPlatformDisplayEXT usage
Just because the #define is present doesn't mean the extension is, so we're
BAILing on egl completely on some systems for no good reason at all. (saw
this on an SGX stack)
This is still wrong. I don't want to try too hard until after the upcoming
release, though.
We should actually be testing for the presence of client extensions before
attempting to do any of this. It's entirely possible that a gl stack will
return bogus functions for these from eglGetProcAddress
Carsten Haitzler (Rasterman) [Fri, 21 Jul 2017 08:02:43 +0000 (17:02 +0900)]
efreet - dont complain if INTERNAL x extension hash of desktop is NULL
asking for an x- field from a desktop shouldnt result in null safety
checks IF the x hash is null - it's fine if it's null - that field
just doesn't eixst. if other fields were in the x hash but the desired
one wasn't it'd not complain, so why complain if there is just no x
hash?
Carsten Haitzler (Rasterman) [Fri, 21 Jul 2017 07:35:21 +0000 (16:35 +0900)]
evas xpm loader - don't error print for files that cant be mmaped
no point complaining because we can't mmap... it'd be an unloadable
file.
Carsten Haitzler (Rasterman) [Fri, 21 Jul 2017 03:28:43 +0000 (12:28 +0900)]
efl_wl - add wayland-server pkgconfig dep for headers from wayland
Carsten Haitzler (Rasterman) [Thu, 20 Jul 2017 09:13:53 +0000 (18:13 +0900)]
efreet - remove noisy debug printfs
Daniel Hirt [Thu, 20 Jul 2017 16:18:20 +0000 (19:18 +0300)]
Evas textblock: add definition to 'cursor_equal'
This is a follow-up to
c4ea25cbdbfb027d5119880ddcf0bf5ba6a5720b to fix
legacy API.
The definition was missing, so adding it.
Stefan Schmidt [Thu, 20 Jul 2017 14:15:32 +0000 (16:15 +0200)]
eldbus: codegen: free allocated strbuf in error path
fname would have been leaked if we failed to get a file_handler.
Also make sure we always create a new fname strbuf even without
a output_dir present to fix a potential unbalance between creating
and freeing.
CID: 1378112
Stefan Schmidt [Thu, 20 Jul 2017 11:46:36 +0000 (13:46 +0200)]
release: Update NEWS and bump version for 1.20.0-beta3 release
Marcel Hollerbach [Thu, 20 Jul 2017 10:00:13 +0000 (12:00 +0200)]
elm_widget: do not check for the visibility
it seems like the visibility is evalulated from different parent
relations, Just because the box (that is a container of a element) is
not visible, does not mean that the element in it is also not visible,
this somehow comes back to the box beeing deleted, the children beeing
reparented in evas, but not in elm. This is for now just a quickfix,
since those calls are just for optimization.
This fixes a load of errors once the elm_test window is closed
@fix
Jean-Philippe Andre [Thu, 20 Jul 2017 08:42:14 +0000 (17:42 +0900)]
eina_list: Ensure single init
This fixes a crash in make check when --profile=dev is explicitely
enabled. eina_list_init() is called by the standard eina_init() loop
and by eina_debug_init() as well.
Honestly I'm not sure why it doesn't crash for other people as
well...
The crash was in eolian_suite during the second eina_init (called
from eolian_init).
Jean-Philippe Andre [Thu, 20 Jul 2017 07:43:52 +0000 (16:43 +0900)]
evas: Fix a crash in emotion with GL
Ooops! When mapping for writing we can't use the same code
path as when tofree is true. This restores the path for image
writing. This basically fixes rage with GL engine.
See also
45c8e5e983
Hosang Kim [Thu, 20 Jul 2017 06:21:34 +0000 (15:21 +0900)]
slider: fix jumping handler bug.
Summary:
Actual slider value is not changed.
So when user release mouse or touch, slider value is returned
to the value by user set.
This is a partial revert of
f41e2761608fff7b070d630a738
Fixes T5745
Test Plan: elementary_test -> slider
Reviewers: jpeg
Subscribers: cedric
Maniphest Tasks: T5745
Differential Revision: https://phab.enlightenment.org/D5031
Jean-Philippe Andre [Thu, 20 Jul 2017 05:35:48 +0000 (14:35 +0900)]
evas: Restore flag pre_render_done in render_mapped
Some objects don't go through render_pre (unchanged, child
of mapped parents), even though they will go through
evas_render_mapped. Those were marked as pre_render_done
inside evas_render_mapped since it seemed to fix some issues
a long time ago.
Unfortunately, if those objects are changed their flag may
not be reset to false, which means they never go through
render_pre, leading to render issues.
I believe simply restoring the value of the pre_render_done
flag should be good enough. I don't know why it is set to
true inside evas_render_mapped but I also don't want to find
out :)
See
9ac13e4aec83ea6dadfa62e4253b1db4ffe167e8 (old)
See
87e5e70a9d888ce93ed807bb0d39d1738241ad9b (older)
Patch made with @jiin.moon
Jean-Philippe Andre [Thu, 20 Jul 2017 05:04:10 +0000 (14:04 +0900)]
evas: Avoid infinite loop on evas shutdown
If an object fails to call efl_destructor() on all the parent
classes, then it may never be removed from the layer object
inlist and never would its data be deleted either (eo keeps
it safely alive).
Here's how to test: add a "return;" statement inside an object's
destructor (eg. the window class).
Fixes T5680
Jean-Philippe Andre [Thu, 20 Jul 2017 05:22:07 +0000 (14:22 +0900)]
evas focus: Fix crash when shutting down with errors
Example test: add many evas_object_ref() to a window and click
on its X to close it.
Jean-Philippe Andre [Wed, 19 Jul 2017 08:12:35 +0000 (17:12 +0900)]
elm: Fix a safety error message
Happens when shutting down E
Jean-Philippe Andre [Thu, 20 Jul 2017 04:25:16 +0000 (13:25 +0900)]
evas gl: Rename secsym_ to eglsym_
eglDestroyImage has nothing to do with SEC...
This is a cosmetic change.
Jean-Philippe Andre [Thu, 20 Jul 2017 04:16:50 +0000 (13:16 +0900)]
evas gl: Check EGL version before using dlsym
A recent commit broke texture_from_pixmap for NVIDIA EGL
(again), because eglCreateImage is a symbol in libEGL.so
but isn't in fact implemented by the driver.
That's because eglCreateImage() is exposed by libglvnd but
the underlying EGL implementation is NVIDIA and its version
is only 1.4, not 1.5 (where the API was introduced as core).
Instead of reverting the patch, it's better to cover our
bases properly and use dlsym() only if the version is right.
Note that GetProcAddress() may return garbage function
pointers for ALL functions as dynamic virtual functions may
be created on the fly by libglvnd. So it is absolutely
necessary to check the extension string as well.
See
0255f14dc2189c71776408b00307b8488bfa4dc5
Derek Foreman [Wed, 19 Jul 2017 18:05:53 +0000 (13:05 -0500)]
gl_common: extsn is no longer unused
Also, we should throw an error when it's NULL, since it's important
for finding symbols for extended egl/gl.
Derek Foreman [Wed, 19 Jul 2017 17:24:19 +0000 (12:24 -0500)]
gl_common: Prefer unextended eglCreateImage
eglCreateImage is objectively better than eglCreateImageKHR - it allows
attributes large enough to hold pointer values. We should use it when
available and only use the older extension version as fallback.
Also, eglCreateImage is core EGL functionality so don't depend on extensions
to be present to use it. Theoretically we should be testing for
EGL version >= 1.5 but it's probably safe not to.