summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Merge branch 'master' into elm_code_wrapdevs/ajwillia-ms/elm_code_wrapAndy Williams2018-07-222375-75720/+130092
|\
| * elm_code: (cherry-pick) Fix cusor pos after newline when indenting with tabsAndy Williams2018-07-221-2/+2
| | | | | | | | @fix
| * elm_code: (cherry-pick) Support indentation styles that are purely tab basedAndy Williams2018-07-227-13/+133
| | | | | | | | | | read: Allow non-EFL style indentation. This is off by default but is switched on if you turn 'tabs insert spaces' off
| * Ctxpopup super slow with some more itemsDave Andreoli2018-07-211-0/+27
| | | | | | | | | | | | | | If you put some (like 100) items in a ctxpopup it will take 5 seconds to show up and throw all kind of errors on console. This is an example that show the issue, I opened T7176 to track this issue
| * ecore: remove exit_signal_received conditional from windows buildMike Blumenkrantz2018-07-201-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: windows does not include ecore_signal.c, so this is not a defined symbol lib/ecore/.libs/lib_ecore_libecore_la-ecore_anim.o: In function `timer_tick_notify':D:\Documents\MSYS2\home\vtorri\gitroot\efl3\src/lib/ecore/ecore_anim.c:372: undefined reference to `exit_signal_received'lib/ecore/.libs/lib_ecore_libecore_la-ecore_anim.o: In function `ecore_animator_custom_tick':D:\Documents\MSYS2\home\vtorri\gitroot\efl3\src/lib/ ecore/ecore_anim.c:940: undefined reference to `exit_signal_received' ref 6405a5a68c0ad618e8b8ab169a0ac85bf43c65c1 Reviewers: vtorri, devilhorns Reviewed By: vtorri Subscribers: cedric, #committers Tags: #efl_build Differential Revision: https://phab.enlightenment.org/D6648
| * efl: fix some warnings in examplesXavi Artigas2018-07-203-9/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: - Check return vaule of fread() - Uninitialized var (seriously!) - Weird struct assignment Still one warning remains, pending evaluation of T7166. Fixes T6718 Test Plan: Those 3 warnings have disappeared and the related example still works as expected. Reviewers: zmike Reviewed By: zmike Subscribers: cedric, #committers Tags: #efl Maniphest Tasks: T6718 Differential Revision: https://phab.enlightenment.org/D6631
| * eina: Add doxygen in/out tags for cow, cpu, and debugBryce Harrington2018-07-203-63/+69
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: devilhorns Reviewed By: devilhorns Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6642
| * evas clip: stop move events feeds.Hermet Park2018-07-201-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: there is no point to feeding mouse events if clipped object is invisible because previsou/current situation won't be different. fyi, move events won't be triggered if prev/cur coordinates are same. Reviewers: #committers, zmike Reviewed By: #committers, zmike Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6645
| * elementary list: add a todo comment.Hermet Park2018-07-191-0/+3
| |
| * elm/list: block recursive recalc loopsMike Blumenkrantz2018-07-192-1/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: this size hints callback is triggered by both list objects and list item objects, meaning that setting size hints on the item objects during recalc will trigger a recursive recalc, potentially infinitely this blocks recursive recalcs unless triggered by size hint changes from a main list object, since those will eventually resolve themselves fix T7121 Reviewers: devilhorns, Hermet Reviewed By: Hermet Subscribers: netstar, DaveMDS, cedric, #committers Tags: #efl_widgets Maniphest Tasks: T7121 Differential Revision: https://phab.enlightenment.org/D6572
| * ecore_main: silence warning when compiling.Alastair Poole2018-07-191-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Silence compilation warning. There is an ifdef'd block of code which accesses obj but I don't think it's in use in production? Test Plan: Build EFL and watch for warning. Reviewers: #committers, zmike, Hermet Reviewed By: #committers, Hermet Subscribers: cedric Tags: #efl Differential Revision: https://phab.enlightenment.org/D6628
| * eina: Fix doxygen in/out tags for eina_bezierBryce Harrington2018-07-191-9/+9
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: Hermet Reviewed By: Hermet Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6630
| * ecore_evas: when window resized in ecore_evas, check evas rotate state.Umesh Tanwar2018-07-181-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: when ecore_evas is resized, get evas' size, but angle not checked, so put a check. @fix Signed-off-by: Umesh Tanwar <umesh.tanwar@samsung.com> Reviewers: Hermet, devilhorns Reviewed By: devilhorns Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6627
| * examples/evas: do not attempt to free animator on window deleteMike Blumenkrantz2018-07-181-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: animators are implicitly destroyed for this case, and attempting to manually destroy them just results in an error/failure and invalid reads since this is a poorly designed api which can internally destroy itself fix T7000 @fix Reviewers: Hermet, devilhorns Subscribers: segfaultxavi, cedric Tags: #efl Maniphest Tasks: T7000 Differential Revision: https://phab.enlightenment.org/D6488
| * eina: Add doxygen in/out tagsBryce Harrington2018-07-185-53/+53
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: Hermet Reviewed By: Hermet Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6624
| * emile image: close file properly.Hermet Park2018-07-181-0/+2
| | | | | | | | @fix
| * efl_ui_layout_object: Fix not to use background/shadow part interface in legacyJaehyun Cho2018-07-181-15/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: "background" and "shadow" part interface was designed to be used in efl ui interface only. Therefore, "background" and "shadow" part interface is not used in legacy. Test Plan: Calls elm_object_part_content_set(layout, "background", content); for legacy layout. Reviewers: Hermet Reviewed By: Hermet Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6595
| * elput: Remove unused and useless event typeChris Michael2018-07-171-4/+1
| | | | | | | | | | | | | | | | | | f220f08 introduced this event, however this event never gets raised or used. I believe this was an accidental inclusion. '#betabreak' Reviewed-By: Derek Foreman <derek.foreman.samsung@gmail.com>
| * Evas text textblock: call evas_font_free even on null fontsDaniel Hirt2018-07-162-7/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A null font set can potentially be stored and passed on. There is no specific fallback in our codebase to roll-back if something like that occurs. This fixes a case where an entry of a null font was not removed from the fonts_cache and was leaked to subsequent tests (the null font entry was not removed from the fonts_cache). Note that this was only apparent when EFL was built with `--disable-fontconfig`, likely due to fontconfig succeeding to always retrieve some non-null font. @fix
| * Eina binbuf: add explicit null-check before magic-checkDaniel Hirt2018-07-161-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Using `--disable-magic-debug` defines EINA_MAGIC_CHECK to always evaluate to `1`. Eina_Binbuf's null-check relied solely on the EINA_MAGIC_CHECK, and failed to pick up on the cases where `buf` was passed as NULL. This has led to a segfault when a NULL `buf` was passed in the test suite (see referenced ticket). This fix adds an explicit NULL check. Fixes https://phab.enlightenment.org/T7147 Reviewers: zmike Reviewed By: zmike Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6592
| * elm_spinner: Do not focus to text button on non-editable modeYeongjong Lee2018-07-161-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: text button don't need to be focused if spinner is non-editable mode. ref T6798 Test Plan: 1. elementary_test -to spinner 2. Focus second spinner 3. move focus using key. 4. turn on access mode and test 1-3 again. Reviewers: CHAN Reviewed By: CHAN Subscribers: cedric, #committers, zmike Tags: #efl Maniphest Tasks: T6798 Differential Revision: https://phab.enlightenment.org/D6581
| * efl ui dnd - further fixes on drop list data key handlingCarsten Haitzler (Rasterman)2018-07-161-0/+1
| | | | | | | | | | | | the efl ui dnd code didn't reset the drop list key to null (deleting it) once it had freed the list of frop items. this also fixes another issue in T7113 which causes crashes on exit.
| * efl selection manager + elm dnd test fix with bad string handlingCarsten Haitzler (Rasterman)2018-07-163-5/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | so there are 2 problems behind T7113. first is a problem in the efl selection manager being "sloppy" with selection data. it's doing a strlen on the data but it's not a normal c string. it's a blob of binary data + length value. this fixes that "sloppiness" by using the len field. there is also another bug in the dnd test code that again has to do with "sloppy" handling of data buffers and assuming nul byte termination and not using the len field properly. this fixes T7113.
| * evas: fix minor typo in smart class color_set method docMike Blumenkrantz2018-07-131-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: vtorri Reviewed By: vtorri Subscribers: cedric, #committers Tags: #efl_docs Differential Revision: https://phab.enlightenment.org/D6591
| * build: fix distcheck when elua is disabled in main treeMike Blumenkrantz2018-07-133-33/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: elua example files were not being distributed correctly as a result of improper use of build conditionals Depends on D6582 Reviewers: devilhorns, ManMower Reviewed By: ManMower Subscribers: #committers, cedric Tags: #efl_build Differential Revision: https://phab.enlightenment.org/D6583
| * build: fix distcheck building when main tree is not compiledMike Blumenkrantz2018-07-131-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: when running 'make distcheck' immediately after configure, the build will fail because some example files were incorrectly being distributed instead of being compiled normally in the dist build Reviewers: ManMower Reviewed By: ManMower Subscribers: cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6582
| * ecore: immediately block animator ticking upon receiving an exit signalMike Blumenkrantz2018-07-132-2/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: when an exit (SIGINT, SIGQUIT, SIGTERM) signal is received, the ui should immediately stop updating in order to present the user with an instant response this uses a simple volatile int to block any ticks which begin after the signal is received; if a signal is received during a tick then it will complete normally fix T7000 Depends on D6589 Reviewers: ManMower Reviewed By: ManMower Subscribers: cedric, #committers Tags: #efl_main_loop Maniphest Tasks: T7000 Differential Revision: https://phab.enlightenment.org/D6590
| * ecore/animator: break out _ecore_animator_flush() from tick functionMike Blumenkrantz2018-07-131-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: no functional changes Reviewers: ManMower Reviewed By: ManMower Subscribers: cedric, #committers Tags: #efl_main_loop Differential Revision: https://phab.enlightenment.org/D6589
| * elm_test_focus: fix a testcaseMarcel Hollerbach2018-07-131-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: The test was checking that multiple register calls fail. However, we made this more stable and are making register only error if the configuration is not the same. This fixes that. fix T7141 Reviewers: zmike Subscribers: cedric, #committers Tags: #efl Maniphest Tasks: T7141 Differential Revision: https://phab.enlightenment.org/D6588
| * eina: Add doxygen in/out tags for strbuf's and ustrbuf's parametersBryce Harrington2018-07-132-173/+173
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: devilhorns, Hermet Reviewed By: Hermet Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6586
| * Selection manager: properly clean in drop_target_delDaniel Hirt2018-07-131-11/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: The last lines in _efl_selection_manager_drop_target_del were skipped when `!pd->drop_list` was met. These lines clear the selection event handlers. Thus, events kept firing and causing erratic behavior. Fixes T7130 Subscribers: cedric, #committers, zmike Tags: #efl Maniphest Tasks: T7130 Differential Revision: https://phab.enlightenment.org/D6579
| * elm_fileselector: move to focus compostitionMarcel Hollerbach2018-07-122-14/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: elm_fileselector was using efl_ui_widget_state_apply to call for internal chain update. This internal chain update was relying on efl_ui_focus_object_manager_get, which results in a wrong value at this time, since the value is still the old one and not the current one. (This behaviour will be updated after the release). Focus composition solves this issue by not relying on the properties of efl.ui.focus_object, but rather take the values from the configuration in the state_apply call. fix T6147 fix T7105 Reviewers: zmike, stefan_schmidt, segfaultxavi, devilhorns, herdsman Reviewed By: zmike Subscribers: herdsman, cedric, #committers Tags: #efl Maniphest Tasks: T7105, T6147 Differential Revision: https://phab.enlightenment.org/D6552
| * eo: Fix efl_new without initializing parametersXavi Artigas2018-07-121-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: efl_new(CLASS), i.e., without initializers, never worked before. The problem is the dangling comma: efl_add_ref(klass, NULL, ); Fortunately the C preprocessor concatenation operator # # has a special case just for this occasion: When you do A ## B, and A is a comma and B is empty, it removes also A. https://gcc.gnu.org/onlinedocs/cpp/Variadic-Macros.html This trick is already in use in several other EFL headers using variadic macros. Reviewers: zmike, bu5hm4n, devilhorns, herdsman Reviewed By: zmike Subscribers: cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6553
| * elm_spinner: Text button and entry both gets hidden when navigated through ↵Prasoon Singh2018-07-121-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | arrow keys. Summary: Summary When focus rect move from inc/dec button using keyboard arrow keys, entry show callback is called and focus is set on entry and then text button is hidden. Hiding focus button triggers focus cycle to revert focus on entry, due to which entry gets unfocused callback which subsequently hides entry. So, entry and text button both goes to hidden state. To fix this setting focus on entry after text button is hidden. Signed-Off By: Prasoon Singh<prasoonsingh16@gmail.com> Test Plan Move focus rect from inc/dec button to text button, and observe both text button and entry gets invisible. Expected behavior is entry should become active. Test Plan: Move focus rect from inc/dec button to text button, and obesrve both text button and entry gets invivible. Expected behaviour is entry should become active. Reviewers: cedric, CHAN, shilpasingh, zmike Reviewed By: shilpasingh, zmike Subscribers: zmike, #committers, govi, rajeshps, jpeg Tags: #efl Differential Revision: https://phab.enlightenment.org/D5153
| * ci: set number of make jobs globally and decrease to 5 jobsMike Blumenkrantz2018-07-123-27/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: travis docs explicitly state that the expectation for builds is to have 2 cpus, meaning that 10 jobs is wayyyy too many and was actually causing some build failures due to strain on the virtual hw this sets the number of jobs using a global variable to avoid having to set it separately for each build https://docs.travis-ci.com/user/reference/overview/#Virtualization-environments Reviewers: devilhorns, ManMower Reviewed By: ManMower Subscribers: ManMower, cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6558
| * ecore: fix order of signal fd shutdownXavi Artigas2018-07-121-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: When closing the fd handler, it checks if the fd is already closed and prints an annoying warning: "fd %d closed, can't remove from epoll - reinit!" We need to close the handler first and then the actual fd. I am not familiar with this part of the code, but this fix removes the warnings and does not seems to have adverse effects. Test Plan: It had warnings before and now it doesn't, haven't observed any other adverse effect. Reviewers: raster, zmike, devilhorns Reviewed By: zmike Subscribers: cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6561
| * canvas render: stop render_pre/post cb if it didn't render.Hermet Park2018-07-121-7/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Basically, render_pre/post should be called only if rendering happens. Unfortunately, this behavior had been broken by some mistakes. As far as I understand after reading history carefully, this issue introduced for fixing unpair of pre/post cbs. In case of async render, the post callback would be triggered in evas_render_pipe_wakeup(), it means POST callbake shouldn't be tiggered in evas_render_updates_internal(). If that post callback were removed in evas_render_updates_internal(), then in case of sync_render, POST callback won't be triggered. So previously, the PRE/POST couldn't get paired. I guess cedric put a mistake that nothing2rendering which brought the pre/post pair issue, even he said that patch fix the unpair of them. (But actually that patch brought the issue!) See this: dc841ed9b2ccf880df2d7590ea46d95e03e8752f Whatever it was intented or not, that patch totally insane, even non-descripting. If we just trigger post cb only if sync render or no_rendered case, We could make satisfication for both cases. @fix Reviewers: #committers, devilhorns, zmike Reviewed By: #committers, zmike Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6573
| * evas textblock: adds missing legacy typesYoungbok Shin2018-07-121-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Adds missing legacy types. The following types were generated in "*.eo.legacy.h" by Eolian. - Evas_Textblock evas_textblock.eo.legacy.h:7:typedef Eo Evas_Textblock; - Evas_Textblock_Node_Format evas_textblock.eo.legacy.h:14:typedef struct _Evas_Textblock_Node_Format Evas_Textblock_Node_Format; @fix Test Plan: N/A Reviewers: raster, cedric, herdsman, devilhorns Subscribers: #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6529
| * tests/edje: add test to verify that swallow objects exist during DEL callbackMike Blumenkrantz2018-07-121-0/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: this mimics expected behavior by enlightenment for 2.0, this behavior should likely be changed since the INVALIDATE event should be the last expected point at which a parent<->child relationship continues to exist in any form ref D6540 Depends on D6554 Reviewers: devilhorns, Hermet Reviewed By: Hermet Subscribers: cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6555
| * eina: Spelling fixesBryce Harrington2018-07-1237-112/+111
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: devilhorns, Hermet Reviewed By: Hermet Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6570
| * tests/edje: split edje tests into separate filesMike Blumenkrantz2018-07-1210-955/+1041
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: this will speed up tests and make it slightly less intimidating to add new tests Reviewers: devilhorns Subscribers: cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6554
| * efl_ui_focus_manager_calc: do not error when the configuraiton is equalMarcel Hollerbach2018-07-121-9/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Before this was erroring. However, this error was not reporting a bad situation, thus there is no need to error. This patch fixes errors on the Popup test when scrollable is set to 1. Reviewers: devilhorns, zmike Reviewed By: zmike Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6560
| * eina: Add doxygen in/out tags for binbuf's parametersBryce Harrington2018-07-121-42/+42
| | | | | | | | | | | | | | | | | | | | | | | | Reviewers: devilhorns, Hermet Reviewed By: Hermet Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6571
| * elm/win: don't force opaque bg if alpha is setMike Blumenkrantz2018-07-111-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: alpha prevents an opaque bg, so ensure that this is being correctly detected worth noting here is that the eo api for win calls through _elm_win_standard_init() during finalize which ensures that need_bg_solid is set; the legacy api does not call this during "normal" construction fix T7129 Depends on D6568 Reviewers: devilhorns Subscribers: cedric, #committers Tags: #efl Maniphest Tasks: T7129 Differential Revision: https://phab.enlightenment.org/D6569
| * elm/win: correctly track and set x11 shaped stateMike Blumenkrantz2018-07-111-9/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: when applying alpha in a non-composited environment, shaped should be set. when removing alpha in a non-composited environment, shaped should be unset if it has not previously been explicitly set @fix Reviewers: devilhorns Subscribers: cedric, #committers Tags: #efl Differential Revision: https://phab.enlightenment.org/D6568
| * canvas vg_loader: close opened file after using it.Hermet Park2018-07-111-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Here opened eina file is just leaked. close it properly. @fix. Reviewers: devilhorns, #committers, zmike Reviewed By: #committers, zmike Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6549
| * canvas vg: check expcetional case.Hermet Park2018-07-111-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: if file is not exists. Reviewers: devilhorns, #committers, zmike Reviewed By: #committers, zmike Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6546
| * evas_drm: Trim the queue of buffers if we've had extra for too longDerek Foreman2018-07-102-2/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: In fixing T7099 I've also allowed the buffer queue to grow quite large, so now we should prune it back if it's bigger than it needs to be for a long time. ref T7099 Depends on D6565 Reviewers: devilhorns Reviewed By: devilhorns Subscribers: cedric, #committers, zmike Tags: #efl Maniphest Tasks: T7099 Differential Revision: https://phab.enlightenment.org/D6566
| * evas_drm: Move MAX_BUFFERS macro from header to source fileDerek Foreman2018-07-102-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: It's no longer needed in the header because it doesn't change the size of the structures there anymore. Depends on D6564 Reviewers: devilhorns Reviewed By: devilhorns Subscribers: cedric, #committers, zmike Tags: #efl Differential Revision: https://phab.enlightenment.org/D6565
| * evas_drm: Allocate buffers on demandDerek Foreman2018-07-102-62/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Instead of allocating a fixed number of buffers immediately, allocate buffers if needed to render to. Normally we only need 2 buffers, but we've been allocating 3 to handle worse case behaviour. As T7099 shows, this is not always enough. We now cap at a max of 10. For the normal case where we always use 2 this results in a slight memory reduction (1 buffer) and a slight renering load reduction because we pick the oldest buffer to render into. A future patch will trim the buffer queue if it's been too large for a long time. fix T7099 Depends on D6563 Reviewers: devilhorns Reviewed By: devilhorns Subscribers: cedric, #committers, zmike Tags: #efl Maniphest Tasks: T7099 Differential Revision: https://phab.enlightenment.org/D6564