[e-users] EFL compile problem
Hi, I just did the usual updates on my system today and that included pulling the latest efl sources using AUR (the efl-git package). But i quickly ran into a problem... meson ran fine, but as soon as ninja started the compile process i got errors: ninja: Entering directory `build' [1/2065] Compiling C object src/lib/eina/libeina.so.1.27.99.p/eina_abi.c.o FAILED: src/lib/eina/libeina.so.1.27.99.p/eina_abi.c.o ccache cc -Isrc/lib/eina/libeina.so.1.27.99.p -Isrc/lib/eina -I../src/lib/eina -I. -I.. -I/usr/include/valgrind -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -fvisibility=hidden -Wpointer-arith -Wunused-parameter -Wsign-compare -Wno-missing-field-initializers -DHAVE_CONFIG_H=1 -D_GNU_SOURCE=1 -DEFL_BETA_API_SUPPORT=1 -DELM_INTERNAL_API_ARGESFSDFEFC=1 -march=x86-64 -mtune=generic -O2 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/usr/src/aur/efl-git/src=/usr/src/debug/efl-git -flto=auto -fvisibility=hidden -fPIC -pthread '-DPACKAGE_DATA_DIR="/usr/share/eina"' -DNEED_RUN_IN_TREE=1 -DEFL_BUILD=1 -DEINA_BUILD '' -MD -MQ src/lib/eina/libeina.so.1.27.99.p/eina_abi.c.o -MF src/lib/eina/libeina.so.1.27.99.p/eina_abi.c.o.d -o src/lib/eina/libeina.so.1.27.99.p/eina_abi.c.o -c ../src/lib/eina/eina_abi.c cc: warning: : linker input file unused because linking not done cc: error: : linker input file not found: No such file or directory [2/2065] Compiling C object src/lib/eina/libeina.so.1.27.99.p/eina_accessor.c.o FAILED: src/lib/eina/libeina.so.1.27.99.p/eina_accessor.c.o ccache cc -Isrc/lib/eina/libeina.so.1.27.99.p -Isrc/lib/eina -I../src/lib/eina -I. -I.. -I/usr/include/valgrind -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -fvisibility=hidden -Wpointer-arith -Wunused-parameter -Wsign-compare -Wno-missing-field-initializers -DHAVE_CONFIG_H=1 -D_GNU_SOURCE=1 -DEFL_BETA_API_SUPPORT=1 -DELM_INTERNAL_API_ARGESFSDFEFC=1 -march=x86-64 -mtune=generic -O2 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/usr/src/aur/efl-git/src=/usr/src/debug/efl-git -flto=auto -fvisibility=hidden -fPIC -pthread '-DPACKAGE_DATA_DIR="/usr/share/eina"' -DNEED_RUN_IN_TREE=1 -DEFL_BUILD=1 -DEINA_BUILD '' -MD -MQ src/lib/eina/libeina.so.1.27.99.p/eina_accessor.c.o -MF src/lib/eina/libeina.so.1.27.99.p/eina_accessor.c.o.d -o src/lib/eina/libeina.so.1.27.99.p/eina_accessor.c.o -c ../src/lib/eina/eina_accessor.c cc: warning: : linker input file unused because linking not done cc: error: : linker input file not found: No such file or directory [3/2065] Compiling C object src/lib/eina/libeina.so.1.27.99.p/eina_binshare.c.o FAILED: src/lib/eina/libeina.so.1.27.99.p/eina_binshare.c.o ccache cc -Isrc/lib/eina/libeina.so.1.27.99.p -Isrc/lib/eina -I../src/lib/eina -I. -I.. -I/usr/include/valgrind -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -fvisibility=hidden -Wpointer-arith -Wunused-parameter -Wsign-compare -Wno-missing-field-initializers -DHAVE_CONFIG_H=1 -D_GNU_SOURCE=1 -DEFL_BETA_API_SUPPORT=1 -DELM_INTERNAL_API_ARGESFSDFEFC=1 -march=x86-64 -mtune=generic -O2 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/usr/src/aur/efl-git/src=/usr/src/debug/efl-git -flto=auto -fvisibility=hidden -fPIC -pthread '-DPACKAGE_DATA_DIR="/usr/share/eina"' -DNEED_RUN_IN_TREE=1 -DEFL_BUILD=1 -DEINA_BUILD '' -MD -MQ src/lib/eina/libeina.so.1.27.99.p/eina_binshare.c.o -MF src/lib/eina/libeina.so.1.27.99.p/eina_binshare.c.o.d -o src/lib/eina/libeina.so.1.27.99.p/eina_binshare.c.o -c ../src/lib/eina/eina_binshare.c cc: warning: : linker input file unused because linking not done cc: error: : linker input file not found: No such file or directory [4/2065] Compiling C object src/lib/eina/libeina.so.1.27.99.p/eina_benchmark.c.o FAILED: src/lib/eina/libeina.so.1.27.99.p/eina_benchmark.c.o ccache cc -Isrc/lib/eina/libeina.so.1.27.99.p -Isrc/lib/eina -I../src/lib/eina -I. -I.. -I/usr/include/valgrind -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -fvisibility=hidden -Wpointer-arith -Wunused-parameter -Wsign-compare -Wno-missing-field-initializers -DHAVE_CONFIG_H=1 -D_GNU_SOURCE=1 -DEFL_BETA_API_SUPPORT=1 -DELM_INTERNAL_API_ARGESFSDFEFC=1 -march=x86-64 -mtune=generic -O2 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/usr/src/aur/efl-git/src=/usr/src/debug/efl-git -flto=auto -fvisibility=hidden -fPIC -pthread '-DPACKAGE_DATA_DIR="/usr/share/eina"' -DNEED_RUN_IN_TREE=1 -DEFL_BUILD=1
[e-users] Problem with conky in Enlightenment
I'm having trouble running conky under Enlightenment recently (it
seems to have started a month or two ago). After a while, CPU usage
shoots up to 100% on all cores. The more conky instances i run, the
sooner it happens, but even with only one it happens after a few
minutes.
When i hear my CPU fan start sounding like a jet engine, top shows
that it's Enlightenment that's the main culprit, followed by every
other Xorg process:
PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND
82569 conrad20 0 1238632 155060 84116 R 100.0 1.0 1:36.25
enlightenment
82530 conrad20 0 587940 84100 55764 R 98.3 0.5 1:29.13 Xorg
82695 conrad20 0 760092 87076 72356 S 83.4 0.5 1:14.68
kdeconnectd
82593 conrad22 2 257904 61420 33344 R 74.2 0.4 1:07.38
terminology
82649 conrad22 2 847524 116036 82764 R 63.2 0.7 0:56.30 keepassxc
82646 conrad23 3 427004 49676 43788 R 57.3 0.3 0:50.50
kdeconnect-indi
82748 conrad23 3 186032 22472 18504 S 30.5 0.1 0:26.80 conky
82747 conrad23 3 404964 22100 18260 R 29.8 0.1 0:26.37 conky
82746 conrad23 3 2996660 23856 18584 R 28.8 0.1 0:25.80 conky
82556 conrad20 0 73360 5224 1920 R 19.9 0.0 0:15.30 xbindkeys
In normal usage, none of these are above 3% CPU usage, most 1% or lower.
I've tested the same conkyrc files under Gnome and don't run into this
situation. And before whatever conky update started this, i didn't
have a problem under Enlightenment either.
Can anyone suggest how to figure out what's going wrong? I attached to
a conky process with strace immediately after starting it and saw
nothing unusual. Then once the CPU went nuts i tried again and got
several MB of output in a few seconds full of the same, over and over:
strace: Process 74691 attached
recvmsg(4, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\2\3\27\0\0\0\0#\0\200\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\>
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\3\27\2\0\340\2-\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\3\27\2\0\340\2.\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
recvmsg(4, {msg_namelen=0}, 0) = -1 EAGAIN (Resource
temporarily unavailable)
poll([{fd=4, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=4,
revents=POLLIN|POLLOUT}])
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\3\27\2\0\340\2-\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
writev(4, [{iov_base="\31\0\v\0\371\2\0\0L\200@\0!
\0\0\2\0\300\2.\2\0\0\2\0\300\2\0\0\0\0"..., iov_len=48}, {iov_base=>
poll([{fd=4, events=POLLIN}], 1, -1)= 1 ([{fd=4, revents=POLLIN}])
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\4\27\2\0\300\2.\2\0\0\2\0\300\2\0\0\0\0\0\0\0\0\0\0>
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\5\27\2\0\340\2-\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
recvmsg(4, {msg_namelen=0}, 0) = -1 EAGAIN (Resource
temporarily unavailable)
poll([{fd=4, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=4,
revents=POLLIN|POLLOUT}])
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\5\27\2\0\340\2.\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
writev(4, [{iov_base="\31\0\v\0\371\2\0\0L\200@\0!
\0\0\2\0\300\2-\2\0\0\2\0\300\2\0\0\0\0"..., iov_len=48}, {iov_base=>
poll([{fd=4, events=POLLIN}], 1, -1)= 1 ([{fd=4, revents=POLLIN}])
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\5\27\2\0\340\2-\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
poll([{fd=4, events=POLLIN}], 1, -1)= 1 ([{fd=4, revents=POLLIN}])
recvmsg(4, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\2\7\27\0\0\0\0#\0\200\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\>
recvmsg(4, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\241
\7\27\2\0\340\2-\2\0\0\2\0\340\2\0\0\0\0\0\0\0\0\0\0>
recvmsg(4, {msg_namelen=0}, 0) = -1 EAGAIN (Resource
temporarily unavailable)
...ad infinitum. None of this looks particularly useful.
Any suggestions?
Thanks,
-Conrad.
--
Shine like thunder
Cry like rain
___
enlightenment-users mailing list
enlightenment-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] keyboard repeat rate
On Tue, Sep 12, 2023 at 12:41 AM Carsten Haitzler wrote: > it's actually a missing feature > that;s been missing for a very long time. efl too has no code to do this in x > (in ecore_x). :) And yet, something is setting lower repeat rate numbers every few days. I suspected E, because other WMs sometimes have settings for those, and i saw the values in the config file. Is there any way you can think of to track down what might be causing this instead? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] keyboard repeat rate
When i log in to an Xorg session, i have a file sourced to set up a few things, including the line: xset r rate 250 30 This works, for a while, but at some indeterminate point later, E changes the keyboard repeat rate and it slows down. I can't figure out what triggers the rate to reset, but it's annoying to have to switch to a terminal and re-source my startup script when i notice it. I did find the following: $ eet -d ~/.e/e/config/standard/e.cfg config ... value "keyboard.repeat_delay" int: 400; value "keyboard.repeat_rate" int: 25; So i know at least where E is getting the slower numbers from. But when does E use these values, and can i prevent it? Or, failing that, is there anywhere in the Settings menus where i can change these default values? I cannot seem to find any mention of repeat rate there. Thanks! -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Chromium shrinks when changing virtual desktops/de-iconifying
> From: Dave > Thought I'd get some feedback from E25 users out there, before I report this > issue to > the Chromium devs. Anyone seen this issue? Got any idea what may be causing > it? I saw this same thing a while back. See this thread: https://sourceforge.net/p/enlightenment/mailman/enlightenment-users/thread/20220207195913.00f2ba4cb694384ee47f0566%40rasterman.com/#msg376 The solution is to enable system title bars, to prevent chromium using it's own, which don't play nice :) -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Conky on E (probably yet again)
From: Jon Craig
> I have 2 Conky instances I used on Cinnamon, one on my main center monitor
> that shows the usual system stats, and another on my right-hand monitor
> that shows the status of a few Docker containers. They're transparent on
> Cinnamon.
Don't know if this will help, but here's how i start up my conkys. On
the rare occasion that E gets wobbly and has to restart 2 of the 3 are
redrawn in the wrong position, so the startup script first kills any
existing conky instances. Then i switch to the virtual desktop i want
them to appear on, pause, and start them up. I don't remember the
reason for the pause, it might be because when this script runs at
login, other startup programs are running and the virtual desktop
switch didn't complete before conky started.
$ cat conky-startup.sh
killall conky
enlightenment_remote -desktop-show 2 1
sleep 5
cd "/home/conrad/.conky/openweather"
conky -q -c "conkyrc" &
cd "/home/conrad/.conky/conn_tar"
conky -q -c "conkyrc_seamod" &
cd "/home/conrad/.conky/conky_rings"
conky -q -c "rings" &
As for transparency, all my conkyrc files include these lines:
conky.config = {
-- Create own window instead of using desktop
own_window = true,
own_window_type = 'normal',
own_window_transparent = true,
own_window_hints = 'undecorated,below,skip_taskbar,skip_pager',
own_window_argb_visual = true,
no_buffers = true,
own_window_class = 'Conky',
-- Use double buffering (reduces flicker, may not work for everyone)
double_buffer = true,
Hope this at least gives you some clues for your setup!
-Conrad.
--
Shine like thunder
Cry like rain
___
enlightenment-users mailing list
enlightenment-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] tooltip problems
On Sun, Feb 12, 2023 at 3:06 PM Carsten Haitzler wrote: > i'm using default... i don't see it. :/ perhaps its the focus rect i added .. > but i've made that ignore mouse events... so i'm not sure how it could dause > it > atm. Some more investigating... I tried reverting some of the changes through git, even all the way back to before any of the focus-related parts were added, but couldn't get those changes to "take". I finally figured out that, despite choosing the System themes > default, E was consistently using my Personal themes > default! Not until i renamed my own copy to default2.edj did E correctly pick the theme under System. It turns out my copy was missing a handful of mouse_events: 0; lines in default/edc/comp.edc. As well as other changes, but i'm guessing that's what was causing the issue. I must have used an older version when i changed the splash.wav file. So, new question: how come E was not letting me choose the System theme "default" when i also had a Personal theme with the same name? Clicking on the radio buttons at the top of the theme selection dialog box was showing two different lists, but E wasn't picking the right "default". Some kind of caching problem related to using the same name? I'll be sure to use a different name from now on... Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] tooltip problems
On Sun, Feb 12, 2023 at 4:58 AM Carsten Haitzler wrote: > i'm wondering if this is theme related with CSD + oversized gtk windows for > the > tooltips that render a shadow as part of the tooltip? can you change gtk theme > to something without shadows? I tried a few different GTK themes, but it didn't help. And it's not just GTK, but QT-based applications, too. This did give me an idea, though... I've made a very minor tweak (startup sound) to E's default theme, so tried the system version instead, and then two others I happened to have installed. The problem disappears entirely with "Dark" and "Dimensions" themes! It seems to be caused by the "default" theme, both the as-installed from latest git (via AUR) and my tweaked versions. But why is it affecting only me? Is the "default" theme picking up some weird efl setting i have messed up that the others aren't? > so it might be that it isn't focus loss but the app gets a mouse out event on > its window because mouse went INTO the tooltip window which s bigger than it > looks? Some more experimenting: the tooltip flickering on and off (or, more commonly, just one on/off flicker) occurs only when any part of the cursor is over the tooltip. The tooltip is stable in some GTK applications where the tip is rendered below the hovered item, and the item (e.g. button) is large enough that i can hover over the top part of it and have the cursor not touch the tooltip when it appears. There needs to be what looks like 3 or 4 pixels separation between the edge of the tooltip and the edge of the cursor, but i can't tell if that's an invisible border on the tooltip or the cursor. Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] tooltip problems
Just wondering if anyone had any thoughts on this, as it continues to cause problems. It seems to be an Enlightenment-only problem, as i've checked it doesn't happen with Gnome or Plasma. As far as i can tell, what's happening is: - after a mouse-over/hover, a tooltip pops up. - E sees this as a new window, sets focus to the new window - the cursor changes to E's cursor - focus is lost from the underlying window, so the element that previously had a mouse-over event no longer does - tooltip closes due to above - focus reverts to the original window - cursor changes back to what the application in that window previously had it set to. - depending on the application, and any cursor movement, this sometimes triggers the first point again, starting a loop. > Date: Tue, 31 Jan 2023 12:31:27 -0500 > From: Conrad Knight > Subject: [e-users] tooltip problems > I've recently started having trouble with tooltips displayed by > applications under Enlightenment. Where a tooltip would normally pop > up after hovering the cursor over an element for a second or so, it > instead pops up and immediately disappears. The tooltip lasts only a > fraction of a second, too quick to read more than a word or two of the > tip's text. > > I've also noticed this happens mostly in applications that use their > own mouse cursors (but not all). When the tooltip first appears, the > application's cursor is replaced by Enlightenment's cursor, at which > point the tooltip disappears and the cursor reverts to the > application's own one. Examples of where this happens: Brave browser, > GTK applications, KDE applications. Curiously, it does NOT occur in > Firefox, even though that uses its own cursors, too. The tooltips in > Gmail (displayed in Brave) are also unaffected, probably because Gmail > is using its own version of tooltips instead of what everyone else is. > Maybe Firefox is, too, as those tooltips do not have the same border > as the "system" ones. > > This seems to be a recent development... I'm using the latest git > versions of EFL and enlightenment available from AUR. Is there a fix > for this? I didn't realise how much i rely on some tooltips until i > couldn't read them! Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] tooltip problems
I've recently started having trouble with tooltips displayed by applications under Enlightenment. Where a tooltip would normally pop up after hovering the cursor over an element for a second or so, it instead pops up and immediately disappears. The tooltip lasts only a fraction of a second, too quick to read more than a word or two of the tip's text. I've also noticed this happens mostly in applications that use their own mouse cursors (but not all). When the tooltip first appears, the application's cursor is replaced by Enlightenment's cursor, at which point the tooltip disappears and the cursor reverts to the application's own one. Examples of where this happens: Brave browser, GTK applications, KDE applications. Curiously, it does NOT occur in Firefox, even though that uses its own cursors, too. The tooltips in Gmail (displayed in Brave) are also unaffected, probably because Gmail is using its own version of tooltips instead of what everyone else is. Maybe Firefox is, too, as those tooltips do not have the same border as the "system" ones. This seems to be a recent development... I'm using the latest git versions of EFL and enlightenment available from AUR. Is there a fix for this? I didn't realise how much i rely on some tooltips until i couldn't read them! Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] polkit problem
> From: Carsten Haitzler > so - build e/efl without asan and it'll all work fine. or ... get digging > deep, > built pam with asan and find out just what on earth is really going on. Wow, that is a weird one! At least i know now it's not E. As i haven't had any crashes or restarts lately (fingers crossed), i'll just switch to a non-asan build. -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] polkit problem
I'm having a problem with E's polkit module. I was running pamac-manager and it asked for authentication. The password dialog box popped up, i entered my password, but then got an authentication failure error message. There's some output in .e-log.log that looks relevant: AUTH: POLKIT: [3-85675816164cbd241ec6e8b4f92b0e4f-3-a4ac0b07e510e40440dbf83b5e4bcfb8] AUTH: UID: [1000] AUTH: readpass... AddressSanitizer:DEADLYSIGNAL = ==99121==ERROR: AddressSanitizer: SEGV on unknown address 0x (pc 0x bp 0x7fffdece1aa0 sp 0x7fffd ece1228 T0) ==99121==Hint: pc points to the zero page. ==99121==The signal is caused by a READ memory access. ==99121==Hint: address points to the zero page. #0 0x0 () #1 0x7fb5b6ceb74b (/usr/lib/security/pam_unix.so+0x574b) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV () ==99121==ABORTING If i unload the polkit module and manually start polkit-gnome-authentication-agent-1 instead, and then try again, everything works: i get a gnome password dialog box instead of an E one, i enter my password, and pamac starts its update process. I don't know when this problem may have started, as i rarely use pamac. Hmm, it just occurred to me that gparted also asks for authentication with polkit. Tried running that, and the same thing happens. After entering my password, i get an E error dialog telling me "Enlightenment was unable to run the application...". And there's another, similar message in .e-log.log. Any ideas what might be wrong? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] enlightenment-users Digest, Vol 196, Issue 8
On Sat, Nov 12, 2022 at 1:41 PM Carsten Haitzler wrote:
> the default is just print. not shift-print.perhaps this is your problem... you
> THINK it's shift-print... ?
Hmm, Print was definitely not bound to anything before i looked at
this yesterday, and shift-Print was. Maybe left over from a
long-forgotten previous attempt at getting screenshots to work? I see
there's a "reset to default" button Before i try that, do you
foresee any problem with me backing up
~/.e/e/config/standard/e_bindings.cfg and restoring it afterward?
(Sidenote: What's the difference between e_bindings.cfg.{1,2,etc.} and
e_bindings.{1,2,etc.}.cfg? I seem to have both.)
> I'm wondering if you have something odd with your keyboard layout OR you're
> running some other software that has stolen the print key?
But surely that other software would also steal it from xev and
xbindkey? And when the key bindings settings screen prompts me to
press a key to bind?
The only other thing possibly affecting that key might be the SysRq
function, alt-Print, which works even in the text console. I could try
temporarily disabling sysrq, maybe?
-Conrad.
--
Shine like thunder
Cry like rain
___
enlightenment-users mailing list
enlightenment-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] enlightenment-users Digest, Vol 196, Issue 8
> From: Ross Vandegrift > No idea- but I can confirm it works for me. I remap screen shot to > C-Print, and map Print to play/pause on a laptop without media keys. I just tried C-Print as well, still nothing. Shift-Print was already in the default bindings, and Alt-Print is SysRq, at a much lower level than X. I've also noticed (while using xbindkeys) that Fn-Print shows up as Ins. This is a laptop keyboard, and the F keys and a few others have alternative labels (such as media keys) used when the Fn key is held down, and the Print key does have a tiny Ins label. Hmm, just thought of something, and tried it out: binding that Ins to the screenshot function DOES work. I can use Fn-Print as a work-around. And the other Ins key, on the numpad, does not trigger the screenshot. xev shows these as keycode 118 (keysym Insert), and keycode 90 (keysym KP_Insert), respectively, so i still have an Ins key to use separately, should i ever need one. E, for some reason, is not intercepting keycode 218 or keysym Print (don't know which it should be looking for). With the Fn-Print binding active, xev doesn't even see that keypress. But it always sees the Print keypress, regardless of that binding state. And yet... E does see it when adding the binding, while asking for a keypress to bind. Weird... -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Key binding PrtSrn not working
Revisiting something i mentioned here quite some time ago, but never got around to fixing... The key binding for the Print Screen button doesn't seem to work for me. The default key binding of Shift + Print Screen (bound to "take screenshot") doesn't either. I added the Print Screen binding from the Key Bindings Settings screen, pressed "Add" and then when the pop up dialog box appeared, i pressed the Print Screen button on my keyboard. Then i selected the "Take Screenshot" function on the right. Screenshot (had to use the Enlightenment menu to take one) is at: https://mega.nz/file/YOFWQZCY#JWLyB_t98x-1Ao2PeTmXuBV2KqKzaFHtqJv-wT_xeIE After hitting Apply or OK, of course, pressing the same Print Screen button on my keyboard... has no effect at all. Neither does the default binding of Shift + Print Screen, trying with both shift keys. I can bind any other key (combination) to the same function, and it works perfectly: i press the key combo and the screenshot window opens, exactly as if i had used the Enlightenment menu to take one. Something seems to be going wrong with binding this particular key. xev shows: $xev -event keyboard Outer window is 0x341, inner window is 0x342 KeymapNotify event, serial 24, synthetic NO, window 0x0, keys: 199 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 KeymapNotify event, serial 24, synthetic NO, window 0x0, keys: 199 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 KeymapNotify event, serial 24, synthetic NO, window 0x0, keys: 199 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 KeyPress event, serial 25, synthetic NO, window 0x341, root 0x301, subw 0x0, time 509100241, (83,28), root:(1150,744), state 0x10, keycode 218 (keysym 0xff61, Print), same_screen YES, XKeysymToKeycode returns keycode: 107 XLookupString gives 0 bytes: XmbLookupString gives 0 bytes: XFilterEvent returns: False KeyRelease event, serial 28, synthetic NO, window 0x341, root 0x301, subw 0x0, time 509100366, (83,28), root:(1150,744), state 0x10, keycode 218 (keysym 0xff61, Print), same_screen YES, XKeysymToKeycode returns keycode: 107 XLookupString gives 0 bytes: XFilterEvent returns: False xbindkeys also sees the keypress: $ xbindkeys -k Press combination of keys or/and click under the window. You can use one of the two lines after "NoCommand" in $HOME/.xbindkeysrc to bind a key. "(Scheme function)" m:0x10 + c:218 Mod2 + Print (Mod2 just indicates that numlock is on.) >From all appearances, it looks like E just doesn't see that particular keypress. Yet it does when i add a keybinding, as it added that entry in the left column shown in the screenshot when i pressed the button. Any ideas what might be going on here? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Frequent crashes when closing windows
For the past few weeks (month or two?) i've been experiencing a lot of crashes with enlightenment (from git). It's not exactly reproducible, but i've noticed it only happens when closing windows. Usually when closing a program's dialog window (file open or save location, for example). But sometimes it happens when closing a program's main window, too (quitting emacs, for example). I finally got around to recompiling enlightenment with debugging support (using the enlightenment-git-asan package from AUR). I triggered a crash and got the flashing red rectangle at the top of the screen telling me to press F1, as usual. I now have e-log.log and e-crashdump.txt files... Are these of any use to someone? As far as i can tell from the log file, the error occurred somewhere in the procstats module. Anyway, i've uploaded the two files here: https://mega.nz/file/8WciETKT#iVo_zmj2TZtOOE5yUobz1zHBnOVFa2pX2TKVGO3AayI https://mega.nz/file/UGdy3bBK#CV3QJjBmQTRI60ALV1Q27kbLJ2nz9hBHeNvLaJW5lkY If there's any way i can generate more debugging output, please let me know. As i mentioned, it's not always reproducible, but i can be fairly confident in triggering this crash a few times a day. Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] efl-git build fails, missing file
Hi, Just updating efl-git and the ninja build process gave an error message about a missing file. When running makepkg, the meson configure proceeded as usual, efl started to build, and then near the end: ninja: Entering directory `build' [2055/2059] Generating data/elementary/objects/edje_cc_elm_test_focus_style.edc with a custom command FAILED: data/elementary/objects/test_focus_style.edj /usr/bin/env EFL_RUN_IN_TREE=1 /usr/src/aur/efl-git/src/efl/build/src/bin/edje/edje_cc -beta -fastcomp -no-warn-unused-images -sd /usr/src/aur/efl-git/src/efl/data/elementary/themes/snd -id /usr/src/aur/efl-git/src/efl/data/elementary/themes/img -id /usr/src/aur/efl-git/src/efl/data/elementary/objects -fd /usr/src/aur/efl-git/src/efl/data/elementary/objects ../data/elementary/objects/test_focus_style.edc data/elementary/objects/test_focus_style.edj edje_cc: Critical. Unable to load image "ic/person-bust-060.png" used by file "data/elementary/objects/test_focus_style.edj": File (or file path) does not exist. Check if path to file "ic/person-bust-060.png" is correct (both directory and file name). [2058/2059] Generating data/elementary/themes/edje_cc_default.edc with a custom command FAILED: data/elementary/themes/default.edj /usr/bin/env EFL_RUN_IN_TREE=1 /usr/src/aur/efl-git/src/efl/build/src/bin/edje/edje_cc -beta -fastdecomp -deps data/elementary/themes/default.edj.d -sd /usr/src/aur/efl-git/src/efl/data/elementary/themes/snd -id /usr/src/aur/efl-git/src/efl/data/elementary/themes/img -id /usr/src/aur/efl-git/src/efl/data/elementary/themes/fdo -fd /usr/src/aur/efl-git/src/efl/data/elementary/themes/fnt ../data/elementary/themes/default.edc data/elementary/themes/default.edj edje_cc: Critical. Unable to load image "ic/person-bust-060.png" used by file "data/elementary/themes/default.edj": File (or file path) does not exist. Check if path to file "ic/person-bust-060.png" is correct (both directory and file name). ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting... -> error making: efl-git And, indeed, i can't seem to find "person-bust*" anywhere. Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Black terminology window on starting E
On Sun, Jul 10, 2022 at 12:43 PM Carsten Haitzler wrote: > the notes in the readme are a result of hearing > feedback from people on nvidia. it would seem their egl/gles support is > certainly sub-par from what I hear. but this is a change to the readme, not a > change to any default config Oh, no, what i meant is that i did change those build options, based on the README. I have a long-running issue with emacs windows not updating after i switch virtual desktops. I saw "nvidia driver" and "sometimes windows not updating" and hoped it would help (it didn't). It seems in my case at least, using nvidia, i do need to use the default opengl=es-egl. But either way, i still have the emacs window not updating problem... Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Black terminology window on starting E
Hi, I recently updated EFL and E from git and have noticed some weirdness when E starts up. (I updated on July 3rd, previous build was from June 14th, just checked again now and git didn't update anything for me.) I have a terminology window set to open when E starts, using the window remember options. Since the update, the window opens as a solid black rectangle instead of the semi-transparent terminal it should be. It's completely black, no text in it, but it does have the outside shadow decoration of an active window. Any new terminology window opens correctly. Also, if i minimize the black window and then restore it, it's drawn correctly as a terminal. At a guess, there's something out-of-sequence with creating the remembered window and E rendering its contents. I've downgraded to my previous EFL build and the problem goes away. The only difference that might be relevant is the notes in the README say to use "-Ddrm=false -Dopengl=full" whereas before i had used "-Ddrm=true" (and no opengl option). The README says this is to prevent windows not updating correctly with nvidia. Could this instead be causing this one window to not update at the right time? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] screen blanking stops working
On Mon, Jul 4, 2022 at 3:42 AM Carsten Haitzler wrote: > close all your apps and see. i bet you one is holding a lock on screen > blanking > - most likely your web browser. they now start doing this whenever a browser > is > playing media (video - but it may not apparently be video. Ugh, yeah, that was it! I closed Brave and the screen blanking started again. And, of course, all those other "restarts" i had tried left the browser actually running. And it's not just having the browser open... I just re-opened Brave to reply to this email, and blanking is working fine. Which is why i hadn't noticed the correlation between having the browser open and the blanking stopping before... It must be only when certain tabs are (re)loaded. > this lock is > between the browser and xserver. Hrm... is there anything in X i can do to prevent this lock? I don't see anything in Brave's settings. > xset q will show > the screensaver/dpms config and you'll probably find its what e set it to... Yes, i checked this before. The timeout settings are 10, 11, and 12 seconds after e's, and change when i make changes in e's settings. So i knew e was communicating with X correctly about this, and using xset to manually blank the screen worked... Well, thanks for clearing up the mystery! Now to figure out how to prevent it :) Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] screen blanking stops working
When i first run enlightenment, screen blanking works fine. In settings > all > screen > blanking i have "enable screen blanking" and "use power saving" turned on, and the screen fades out and blanks after the specified timeout. No other settings are turned on, nor in the "wakeups" tab. However after some indeterminate time (a few hours?) it stops working, and the screen will no longer blank. Going back into the settings and toggling the "enable" option on and off doesn't poke it back into working again. I've also tried "enlightenment_remote -restart", and that doesn't start it again either. Enlightenment has also crashed a few times with the "press F1" error message, and pressing F1 to restart also does not bring the blanking back. The only thing that works is to log out and back in again so that gdm restarts enlightenment. At that point it works again, for "a while". When the automatic blanking stops, i am able to manually do it with: xset dpms force suspend (but i don't get the nifty fade-out enlightenment gives me). I'm using enlightenment from git, and i just updated today to check if it's any better (it isn't). Is there a way to figure out what is going wrong, or if anything else is interfering with the blanking procoess? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Odd placement of procstats module after full screen
I have the procstats module active, and it typically displays in the title bar of the window. I can click on it to get more information in a pop-up window. A weird bug seems to happen when the window gets sized to full screen, though. Using a web browser (Brave), i visited a site (stadia.google.com) that switched to full screen. As the module is displayed in the title bar, i fully expected it to disappear along with the title bar itself and the rest of the window decorations. Instead, it hovered sort of vaguely in the top left of my screen. When i exited full screen mode, it stayed in the same position instead of being put back in the now-visible title bar, just sort of floating on my desktop. Clicking on the box opened up the more info pop-up, but also the main desktop menu (as i was clicking on the desktop, after all). A second click closed the desktop menu, leaving just the pop-up. A third click closed the pop-up and opened the desktop menu again. Finally, a fourth click closed that, too. I'm attaching two screen shots, one showing the correct placement of the procstats module in the Brave window, and the other showing the weird, floating behaviour after switching to full screen and back again. I don't know if this mailing list allows attachments, so if they go missing let me know where to upload them instead :) Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Chromium window resizing every time it's displayed
On Mon, Feb 7, 2022 at 7:12 AM Carsten Haitzler wrote: > i suggest enabling the use of system titlebars and all will work fine. :) it > also makes it look consistent as well then. That does seem to function as a workaround. It looks REALLY ugly though! :) And now i'm missing the minimize and close buttons that Chrome had in its own title bar. Oh, and no more resize handles, either, i've just noticed! -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Chromium window resizing every time it's displayed
I just updated Brave to the latest version, based on Chromium 98, and almost immediately noticed an odd problem that hadn't been there before with Chromium 97. Every time the browser window is re-displayed, it is resized smaller. There’s a visible flicker when the window is displayed at the original size and then redrawn smaller, by approximately twice the width of the scroll bar, and the height is reduced by double that distance. The top left corner remains in the same spot, the bottom right corner gets moved inward every time the window is displayed. This happens when I change to a different virtual desktop and then switch back. But it also happens simply by minimizing the window and restoring it. I installed the latest Chrome-stable, too, and verified it's happening there, so it would appear Chromium-related and not just a bug in Brave. It also does not happen when switching desktops or minimizing the window in Gnome. I don't think this is related to the problem posted here a few months(?) back about an off-by-16-pixels pointer focus in Chrome, i think it was. The window is shrinking by more than 16 pixels, and happens repeatedly. After switching desktops a few times, the browser window becomes unusably small unless I keep resizing it back. But given the problem at the time was found to be E not knowing how to handle a combination of circumstances that shouldn't have happened to begin with, could something similar be happening here? Oh, and I also tried to use E's window lock settings, but they had no effect. Well, other than to prevent me from resizing the window larger again if i selected the User locks > size option. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] full-screen image viewer in Telegram
Recent versions of Telegram desktop seem to have issues when running under Enlightenment. The full-screen image viewer will display correctly the first time it is used, but the moment you try to use it a second time, the whole Telegram interface freezes. Fortunately the systray icon still works to quit the program. I have verified this does not happen under Gnome. I posted on Telegram's github issues page, and was directed to this discussion: https://github.com/telegramdesktop/tdesktop/issues/17296 The OP there is also using Enlightenment. It's possibly a Qt6 thing? Is it an actual bug in Qt6 that should be reported? Is there a workaround in Enlightenment? Thanks! -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Terminology 1.12.0
> From: J?r?my Zurcher > the themes are in INSTALL_DIR/share/terminology/themes/ > should be /usr/share/terminology/themes/ on most systems > you'll find the 2 installed themes : default.edj and nyanology.edj Yes, that's what i was doing. But this: > > > From: Carsten Haitzler > > > you can make a theme that ONLY covers this edje group and > > > nothing else - it'll fall back to using default elements for everything > > > else. seems to imply that in addition to an installed theme, you can have another edj file that overrides certain elements. How do you tell terminology to also use this extra file besides the chosen theme? Or does the new theme begin with an "include" statement to import another theme, and then proceed with the overriding elements? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Terminology 1.12.0
> From: Carsten Haitzler > change your theme ... :) I simply replaced terminology's default.edj with the one from 1.11.0. That visual bell was just WAY too much, and I preferred the old cursor, too. But i did like the handles to adjust the text selection in the new theme. > you can make a theme that ONLY covers this edje group and > nothing else - it'll fall back to using default elements for everything else. I had initially decompiled the original default.edj, and after some digging and experimenting with changing some numbers, found how to reduce the radius of the new flashing circles. Each time i made a change, i used the included build.sh to generate a new default.edj and replace the existing one. Are you saying you can also keep the existing one, and create a new theme that will override just those elements? Where should that new file be placed? I looked around on enlightenment.org but couldn't find any documentation (which was why i was using trial and error to adjust the existing default.edj). Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Terminology 1.12.0
> From: Boris Faure > == Additions == > * New default theme! Is there supposed to be a Default.eet? I see a Default.ini with the other colour schemes, but the meson.build file doesn't include it. When i run terminology, i get the following error: $ terminology --version ERR<199698>:terminology ../src/bin/colors.c:703 _color_scheme_get() failed find colorscheme 'Default' ## Copy & Paste the below (until EOF) into a terminal, then hit Enter eina_btlog << EOF /usr/lib/libeina.so.1 0x7f3009e67b84 0x7f3009e3d000 /usr/lib/libeina.so.1 0x7f3009e68dc1 0x7f3009e3d000 /usr/lib/libeina.so.1 0x7f3009e6a469 0x7f3009e3d000 /usr/bin/terminology 0x55ac528704eb 0x55ac5285b000 /usr/bin/terminology 0x55ac52874457 0x55ac5285b000 /usr/bin/terminology 0x55ac52879bdb 0x55ac5285b000 /usr/bin/terminology 0x55ac5286e0df 0x55ac5285b000 /usr/lib/libc.so.6 0x7f30095edb25 0x7f30095c6000 /usr/bin/terminology 0x55ac5286e12e 0x55ac5285b000 EOF Version: 1.11.0 (Note that i do have 1.12.0 installed, not 1.11.0.) Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Christmas releases!
It was a nice surprise yesterday morning, thank you to everyone involved! I have a quick question about the new palette selector: the dialog box seems to be missing the usual "ok" and "apply" buttons. There is only the "close" button to cancel. How do you go about choosing a new palette? Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] efreetd not picking up all icons
Hi, I've just done a fresh install of Arch linux, and so, of course, added Enlightenment as my wm. I've noticed, though, that my applications menu is missing quite a few icons in its entries. These seem to be almost all with the Gnome apps included with the bare-bones Gnome desktop i also added for testing purposes. I've tried deleting efreetd's cache and recreating it with: /usr/lib/efreet/v-1.25/efreet_desktop_cache_create -d /usr/share/applications/ /usr/lib/efreet/v-1.25/efreet_icon_cache_create -d /usr/share/icons/ -e .png .svg .jpg .xpm but this results in no change. Just to pick an example of an entry that doesn't have an icon: /usr/share/applications/org.gnome.FileRoller.desktop contains Icon=org.gnome.ArchiveManager and /usr/share/icons/hicolor/scalable/apps/org.gnome.ArchiveManager.svg does exist, and is the icon that shows up for the program in Gnome's desktop. What could be the reason for efreetd not picking up this (and other) icon? As i mentioned, this is a new Arch install, so it's quite possible i missed an essential package, but i can't think what it might be. Thanks! -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] emacs problem persists
On Fri, Aug 20, 2021 at 5:33 PM Carsten Haitzler wrote: > well to rule out it being the hidden property. > src/bin/e_hints.c >if (ec->netwm.state.hidden) > state[num++] = ECORE_X_WINDOW_STATE_HIDDEN; > comment those out. like: Yup, that "fixed" the problem. What else is likely to be affected, though? -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] emacs problem persists
> From: Carsten Haitzler > I wonder if they are using the netwm visibility state oddly? E will set > the netwm state to invisible if a window is iconified or when screen blanks > and > set it back when it is uniconified or screen unblanks. Also if you switch > virtual desktops windows on other desktops get the hidden state set and > removed when they appear again. When does it freeze and stop rendering? what > do > you do before that to make it happen? Just to throw another data point in here... this is exactly what i'm seeing since using the git version of enlightenment (it was fine with e 0.24.2). If i move to another virtual desktop, and then return to the one with the emacs window, it looks like emacs never gets focus. Enlightenment's borders do change colours to indicate it has focus, but the cursor remains an empty box instead of a solid rectangle. Typing in the window also shows nothing happening. Until... If i minimize the window and bring it back, or if i shade and unshade the window, it all looks ok again, and everything i typed is there! So i can be typing blindly with nothing appearing in the window, then shade and unshade, and see that everything i typed (more often than not on the wrong line, because i couldn't see it). This is with emacs 27.1, as packaged in 1:27.1+1-3ubuntu3. -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Solved! (Was: Re: font settings don't work)
> From: Carsten Haitzler > enlightenment will set xft.dpi to 75 if scale is 1.0. it didn't set this > before. efl itself always rendered fonts at 75dpi. it's compiled-in. scaling > is > not done by faking/changing dpi but by applying a scale factor. to make sure > everything matches it now sets the dpi to 75 (multiplied by scale factor), > thus > you may need to set a different font size. I had been meaning to ask if there were any other changes (besides the cursor scaling) that involved the scale factor. So are there changes to dpi values to take into account, too? I've noticed that fonts in my Chrome window are now a bit smaller, and my conky displays need tweaking (the spacings being affected by font sizes?). Is the fix to just go through every program and change font sizes if things look different? It's not universal... other windows i have open right now (Firefox, Telegram, Terminology...) all look fine. My display is 96 dpi, and i use a scale factor in E's settings of 1.2. By what factor should i scale up font sizes in the now-distorted programs? I've got cursor size and font size (in some programs). Are there any other sizing surprises to watch out for in moving from E24? :) Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Theme tweaking help
On Sun, Apr 18, 2021 at 5:42 AM Carsten Haitzler wrote: > colors are everywhere. grep for color: I was hoping a few theme defaults were set somewhere, but yes, everywhere :) I used "sed -i" extensively... > it's the same size it was before... though scaling applies if you use a scale > setting other than 1.0. Maybe that's it... was scaling not applied to the cursor before? I found the size of the default/color pointer and reduced it. The resize pointers with the pulsing boxes seem to inherit sizes from that, but i can't seem to find the size of the box itself. Oddly, the size of the floating box on the move pointer did adjust itself accordingly. > but .. the clock is there... in a logically named file... :) $ find -name "*clock*" ./edc/elm/uiclock.edc ./edc/elm/clock.edc ./edc/efl/uiclock.edc ./edc/clock.edc The gadget itself can be sized, of course, and I replaced the digit images with the old ones, so the clock doesn't appear stretched out any more. Everything appears to be in edc/clock.edc, except for the size of the panel that opens with the calendar display when clicking on the clock. The clock that is duplicated in that panel is huge, but i figure it should automatically scale down if i can find where the panel size is set...? One more thing: is there a way to set a corner radius on the top window decoration bar that E adds? I don't like the square corners... to me, it looks too much like what i've seen on some Windows users' desktops. Thanks! -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Theme tweaking help
I built EFL and E from git and now want to make some adjustments to the default theme... I've unpacked /usr/local/share/elementary/themes/default.edj and now i'm looking for some hints on where to start. It looks like the theming for absolutely everything is in there :) Things i'd like to tweak: - colours (e's menus and dialog boxes, highlighting...) - size of the (now huge) mouse pointer. - size of the time display in the digital clock desktop gadget (i know the digits themselves are images, so just need to find the scaling factors). Thanks! -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Is it possible that E 0.24 broke xkbmap?
I posted to this list on 18th July about E 0.24 breaking xmodmap (possibly, i was looking for confirmation and a fix). Could whatever changed to break xkb also be affecting xmodmap? The tl;dr version of that message is: Compose key still works (separately set by X), but xmodmap -e "keycode 133 = Multi_key dead_greek" has no effect any more, shift-compose does nothing. Still works with other WMs, or none at all, only broken in E. Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] "Night Light" Mode?
> From: Qingxiang Jia > How do you change the color of the display (not > physically) so that it makes eyes more comfortable? Not specific to Enlightenment, but i use redshift. Look for redshift-gtk, an add-on that hangs out in your systray. If your package manager doesn't have it, here's the website: http://jonls.dk/redshift/ -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] xmodmap under E
This is an odd one... I recently tried to enter a Greek letter in a text input field and it didn't work as it should. I checked and my key mapping was not correct. I have the following in my ~/.xprofile: xmodmap -e "keycode 133 = Multi_key" xmodmap -e "keysym Multi_key = Multi_key dead_greek" This has always worked in the past. The windows button is my compose key, and shift-windows should let me type the next letter in Greek. To check it wasn't a problem with my .xprofile being sourced, i typed in a terminal: $ xmodmap -e "keycode 133 = Multi_key dead_greek" $ xmodmap -pk | grep 133 1330xff20 (Multi_key) 0xff20 (Multi_key) 0xff20 (Multi_key) 0xff20 (Multi_key) So... xmodmap runs without error, but doesn't actually DO anything. I tried a couple of other keymappings to make sure it wasn't just the windows key, and none of them took. I ran another X session with no window manager, just an xterm, and the xmodmap commands worked as expected. I even logged into the default Ubuntu gnome (X session, not wayland), and it worked there. The output from the above grep is then: 1330xff20 (Multi_key) 0xfe8c (dead_greek) 0xff20 (Multi_key) 0xfe8c (dead_greek) and a quick test gives the expected results. So it appears xmodmap is not working for me, but only under Enlightenment (0.24.1). It _used_ to work, i'm guessing this changed with either the latest efl or enlightenment releases? I haven't had a need for Greek characters in the last few months and didn't notice until today. Any ideas? Thanks! -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] efreetd error message
On Thu, Jun 11, 2020 at 3:59 PM Massimo Maiurana wrote: > Yes, i can try it increasing the timeout, but since you already found it > can you tell where it is set? :) It's in enlightenment-0.24.1/src/bin/e_startup.c, line 33 or thereabouts: timer = ecore_timer_add(20.0, _e_startup_time_exceeded, NULL); -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] efreetd error message
On Wed, Jun 10, 2020 at 2:31 PM Carsten Haitzler wrote: > yes. possibly. if you change it to 20 sec instead of 10 - does the problem go > away? Yes, i had a chance to reboot (kernel update) and so changed the timer to 20s first. It completed in 12s this time, and there was no error. > but both of > these mean more of a "time to rewrite efreet" which is on my long todo list :) Well, the 20s timeout works for me, and if it does for Massimo Maiurana, too, then there's no need to bump this up that list :) It was only a minor annoyance to begin with, having to clear the error message... Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] efreetd error message
Another data point: I've noticed this error only occurs when E is started the first time after the computer boots. If i log out and back in again without rebooting it doesn't. I checked with the time stamp debug output i added, and the time from setting the timer to clearing it is then 1s. Down from 14s! Probably because everything is already in memory? On Wed, Jun 10, 2020 at 4:57 AM Carsten Haitzler wrote: > as no one who reports > the issues actually has provided detailed debug and analysis of what is going > on. I can certainly add a few more debug print statements... As you pointed out, setting the timer to a larger value would avoid the error message, but if this would help figure out a faster startup i'd rather do that. I just need to know where, and what output you'd like to see. Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] efreetd error message
On Sun, Jun 7, 2020 at 11:22 AM Carsten Haitzler wrote: > is your disk particularly slow? I don't think so...? I mean, it's getting on in years, and it's 5400rpm, but nothing that seems to have caused problems elsewhere. > all it has to do is look at the icon and > desktop file entry dirs and their dir trees and basically look at timestamps > to > see if anything changed Hmm, the pastebin i linked to shows that it's also loading modules in between the time stamps: ESTART: 4.5 [0.2] - Run Startup Apps E: timer started Sat Jun 6 20:15:30 2020 . ESTART: 4.54471 [0.1] - Load Modules ESTART: 16.84508 [12.30037] - Load Modules Done . ESTART: 17.19181 [0.6] - MAIN LOOP AT LAST E: efreet didn't notify about cache update E: sent profile set req to 0xe00023 [standard] E: timer cleared Sat Jun 6 20:15:44 2020 Is something running out of the expected sequence? Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] efreetd error message
On Fri, Jun 5, 2020 at 12:34 PM Carsten Haitzler wrote: > well putting printfs in the code to begin getting logs. Ok, i found where the timer is set and cleared, added a few debug lines, with time stamps. It looks like everything goes as expected, it just takes too long: ESTART: 4.54442 [0.00038] - E_Comp_Canvas Keys Grab Done ESTART: 4.5 [0.2] - Run Startup Apps E: timer started Sat Jun 6 20:15:30 2020 ESTART: 4.54457 [0.00013] - Run Startup Apps Done ESTART: 4.54458 [0.1] - E_Comp Thaw ESTART: 17.34377 [0.15196] - SLEEP RRR: CB screen change... RRR: CB output change... RRR: CB screen change... RRR: CB output change... E: efreet didn't notify about cache update E: sent profile set req to 0xe00023 [standard] E: timer cleared Sat Jun 6 20:15:44 2020 RETRY backlight set as 0.05 != 0.41 (expected) try=0 RRR: CB screen change... So it took 14s to get to that point, and the error is triggered a little before that (but still longer than the 10s set for the timer). Should the timer be set for longer? Or would that just be hiding the fact that something is taking too long? (The full output from E while starting is at https://pastebin.com/DnpN5DQ7 in case anyone wants it.) Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Screenshots (was Re:efreetd error message)
On Fri, Jun 5, 2020 at 12:34 PM Carsten Haitzler wrote: > quality > 90% == png. 90% or less == jpeg. so just slide the quality up. :) ... > click on a screen or window and it'll cycle through crops for exactly that > region (first click will select that screen, next click will select the window > you clicked on + padding, click again and it selects that window without any > padding. Wow, both of these are exactly what i was hoping for! Never knew they were there :) > i have no > idea why it doesnt work for you. delete it and add it back i guess? Tried that, too, and adding other key-bindings, like ctrl-print, or shift-print. In each case, the "add" or "modify" buttons will detect the key press and let me add the screenshot action, but outside of the settings window the key press doesn't do anything. But if i add, for example, alt-shift-p as screenshot, it works. It's just the actual print screen key whose binding doesn't work. xev shows the following when i press the key: KeyPress event, serial 36, synthetic NO, window 0x321, root 0x2f9, subw 0x0, time 78417058, (102,40), root:(1173,91), state 0x0, keycode 218 (keysym 0xff61, Print), same_screen YES, XKeysymToKeycode returns keycode: 107 XLookupString gives 0 bytes: XmbLookupString gives 0 bytes: XFilterEvent returns: False KeyRelease event, serial 39, synthetic NO, window 0x321, root 0x2f9, subw 0x0, time 78417270, (102,40), root:(1173,91), state 0x0, keycode 218 (keysym 0xff61, Print), same_screen YES, XKeysymToKeycode returns keycode: 107 XLookupString gives 0 bytes: XFilterEvent returns: False Does anything there look suspect? Again, the key bindings settings window does detect the key press... Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Screenshots (was Re:efreetd error message)
On Fri, Jun 5, 2020 at 9:53 AM Carsten Haitzler wrote: > On Thu, 4 Jun 2020 15:44:40 -0400 Conrad Knight said: > > (Oh, and another question... Why is it that the text of the error > > message dialogs can't be selected for copy-paste? Seems like being > > able to report them should be one of the more important things to do > > with error messages.) > you can just screenshot it like any window... :) Well, you can't search Google for errors using a screenshot! It just seems that having error message text selectable should be a no-brainer. Taking a proper screenshot would mean firing up GIMP or something. The new version of the built-in screenshot module is an improvement, but i was wondering if you might entertain some feature requests? 1) Save options. At the very least, the option to set the file format to something more appropriate for screenshots. Jpeg is probably the worst choice. 2) An option to screenshot a single window. The crop tool is useful, but requires some pretty fine mouse movement while holding a button down. This is a usability and accessibility problem, and isn't as easy as "click on window". Another big problem i've had since... e20-or-so? I simply cannot get the keyboard shortcut for screenshots to work. I was hoping the new module might fix this, but it seems to be a problem with the actual key-binding part of it. In key bindings settings, i click on any key in the left pane to enable the "modify" button (is this a bug?). Press the "modify" button and, when prompted, press the print-screen key. This highlights "Print" in the left pane, which would seem to indicate the keypress is correctly detected. After some scrolling back-and-forth in the right pane (another bug?) i can find the current binding. "Take Screenshot" is highlighted, with the "Action Params" box below it blank. This is the default binding, i haven't changed anything here. If i click on the "Take Screenshot" entry, the box fills with a long line of options. Should one of these options be selected? And the screenshot isn't working because the default is blank? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] efreetd error message
On Fri, Jun 5, 2020 at 9:53 AM Carsten Haitzler wrote: > well you ge that because of a timeout in hearing back from efreetd. so during > startup e setups up a 10 second timeout > i can't tell you what is going wrong as you'd need to follow the > chain of events as to what is and is not actually happening Hmm, ok... Any suggestions on how to do this? Is there a verbose mode i can start E with, that will log these set and receive timer events? > > Should I manually delete those? Should i run a script > > to empty this directory when E quits? > no. it's meant to stay around. Ok, just wondering, as some of the older files have names ending in ".lock" and i thought perhaps they might be interfering. Thanks, -Conrad. ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] efreetd error message
I just updated to Enlightenment 0.24.1 and i'm still getting this error message when enlightenment starts up: Efreet did not update cache. Please check your Efreet setup. Is efreetd running? Can ~/.cache/efreet be written to? I checked, and the answer to both questions is yes. Almost every file in that cache directory is written to when E starts, but there are a few older ones. Should I manually delete those? Should i run a script to empty this directory when E quits? (Oh, and another question... Why is it that the text of the error message dialogs can't be selected for copy-paste? Seems like being able to report them should be one of the more important things to do with error messages.) Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] appindicator (was Re: competitors to E?)
From: Carsten Haitzler (The Rasterman) > everyone has or is on the way to dropping xmbed - so if youw ant xmbed .. you > need something like stanlonetray. e supports app indicator protocol. it's a > dbus protocol instead of x windows reparented. On this subject... I'm having trouble with E's appindicator with certain programs, dropbox is the most recent example. When i run dropbox (from the command line instead of auto-start at login, so i can see the output), the following warning is displayed: (dropbox:13514): libappindicator-WARNING **: 11:21:34.670: Unable to connect to the Notification Watcher: GDBus.Error:org.kde.StatusNotifierWatcher.Item.AlreadyRegistered: The icon that shows in E's systray is not the usual dropbox icon that displays if i use a different wm (Ubuntu's default gnome, for example). It shows a generic warning triangle instead. Dropbox also displays an xembed icon, i assume for backwards compatibility purposes. If i run stalonetray, the extra tray opens and displays the correct dropbox icon. Another symptom i'm running into of something broken with my systray is that telegram will not show any appindicator icon at all. And i've verified that the setting is checked in telegram's options. Instead, i get these warnings: $ /opt/telegram/telegram QApplication: invalid style override passed, ignoring it. Available styles: Windows, Fusion error: : cannot open error: : cannot open error: : cannot open Running the same in gnome produces no warnings, and the icon displays in the systray. Any ideas what in my set up might be broken, and how to fix it? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] edgar
Hi, I'm trying to get the edgar E module to work, and everything seems to go ok until i load the actual module. At that point an error message pops up: Python-efl not found. Your python gadgets will not work. Please install or update python-efl. (By the way, why can't you select and copy text in error message dialogs?) Obviously i do have python-efl, otherwise i wouldn't have been able to build the module. I'm using efl-1.23.3, python-efl-1.23.0, and edgar doesn't have a version number, but it was cloned from git yesterday. Python is version 3.7, Cython 0.29.10. I saw an archived post mentioning that the cython version could be an issue, but that was from years ago. Just in case, i tried building python-efl with the included C files, and with ENABLE_CYTHON=1, with the same results. Any ideas where to look for the problem, or how to fix it? Thank! -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Wayland problems
On Sat, Jan 12, 2019 at 9:48 AM Carsten Haitzler wrote: > it's been working a charm for me for months now... :) little things here and > there have been fixed, but it's pretty much good. Ok, got it working :) I now have a build that will run under either Wayland or Xorg, based on which session i choose to log in. This is based on the efl and enlightenment from git, using the efl configuration parameters from that AUR archive (without any of the disabler ones set). However, now i still have my original two problems: an extra cursor image in the middle of the screen, and no control over the window borders. In fact, under the window menu, the "border" entry is missing. If i go into "locks", "do not allow border change" is checked. Unchecking it brings back the "border" entry, but top option, "select border style", is missing. The ghost cursor appears where the cursor was positioned on the login screen before hitting enter after typing my password. If i switch back to gdm's login screen, move the cursor around, then switch back, the ghost cursor is in the new position. I can get rid of it by switching to a vt that doesn't have this cursor and then back to e (wayland). This may be a gdm3 problem, but a gnome wayland session doesn't show it. The window border problem is the bigger issue: e seems to draw borders whether or not windows should have them, and there doesn't seem to be any way to remove them. Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Wayland problems
On Fri, Jan 11, 2019 at 3:24 AM Carsten Haitzler wrote: > that's the meson based build - you can just use it as-is and copy & paste... > :) I was under the impression that meson building wasn't quite working yet for efl. I just gave it a go, and it wants a more recent version of meson, anyway... > you did want an explanation... :) Thanks for the info-dump, that's exactly what i was looking for! Would it be safe to just enable "everything" in efl, or would there be conflicts? And, related, i just checked "glxinfo" in a wayland session (Ubuntu has a login option for gnome in either wayland or xorg) and noticed it was running the "mesa project and sgi" version instead of the nvidia one i have in xorg. Does that mean not hardware accelerated (or at least, not by my video card)? It looks like there are some more things for me to work out before i give wayland another shot. On switching back, i seem to have messed something up... i've deleted everything in /usr/local/ that got installed in the past few days, did a "make clean", and "configure" with my usual options, "make", "make install" for efl-1.21.1 and enlightenment-0.22.4 (released versions), and now i can't login! I can start a bare X session, and run enlightenment_start from an xterm, but this is hardly ideal. Trying to start it from gdm directly just dumps me back at the login screen. Any suggestions? I made sure to remove the /usr/share/wayland-sessions/enlightenment.desktop file i had added, so it's not trying to run a wayland session and failing (i think...?). And "enlightenment_remote -desktop-show 2 1" now gives an error: Error org.freedesktop.DBus.Error.UnknownMethod: Method "Show" with signature "ii" on interface "org.enlightenment.wm.Desktop" doesn't exist Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Wayland problems
On Thu, Jan 10, 2019 at 1:39 PM Carsten Haitzler wrote: > As usual - wayland stuff is kind of new and experimental so I'd advise using > git master for that. > as for how to compile - look at the aur git pkgs: Ok, i'll give it a shot :) Couple of quick questions: For EFL, are "--enable-egl" and "--enable-gl-drm" not needed, then? What configure option does "-Dbuffer=true" correspond to? Does "-Dopengl=es-egl" mean "--with-opengl=es"? Because i don't see any other options besides "full" or "none". I have to admit, i'm a little lost on the differences and requirements for opengl, gl-drm, egl, es-egl, etc. Is there by any chance a short reference to all these terms? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Wayland problems
I've followed the instructions on this page: https://www.enlightenment.org/about-wayland adding the necessary configure options. (Note: --enable-wayland-clients and --enable-wayland-egl for enlightenment's configure are not recognised. I used --enable-wayland instead.) I'm using efl 1.21.1 and enlightenment 0.22.4. Everything seems to work, except for these two problems: 1) There's a big, fat arrow cursor in the center of the screen. I would say it's the default X cursor, but this is wayland... It's separate from the normal cursor, which functions as expected. It's a static cursor image that appears on top of everything else. 2) Window decorations appear to be broken! I have a borderless terminology window, and completely decoration-free conky windows open on startup, and they all have borders and drop shadow. Programs (such as Chrome) that manage their own decorations and positioning also have an enlightenment border added, creating a double-border effect, and their positioning over-ridden. Are there fixes or configuration settings to correct either of these? Another quick question -- that same web page mentions a wayland-only option, but the link to those instructions says it's not safe for everyday use. Is this still the case? Obviously, until the above problems are fixed, i'm not about to compile enlightenment without X support. Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] enlightenment "freezing" when opening a new window... sometimes
On Mon, Dec 24, 2018 at 4:54 AM Carsten Haitzler wrote: > https://www.enlightenment.org/contrib/enlightenment_debugging > > if you want to attach without there being a crash.. first tell > enlightenment_start to stop monitoring: > > killall -USR1 enlightenment_start Well that was easier than i had thought it might be :) I set the breakpoint, and stepped through after, with and without the freeze-up, but couldn't see any difference. After not seeing much of anything useful, i recompiled EFL and E with debugging info and... the problem's gone! I suspect left over old versions of EFL libraries may have been the problem. The "make install" and "ldconfig" don't always seem to work as expected. Thanks for your patience. -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] enlightenment "freezing" when opening a new window... sometimes
On Wed, Dec 19, 2018 at 4:39 AM Carsten Haitzler wrote: > oh - mouse pointer didn't change e.g. when over chrome? (when things froze). > zero changes to pointer? in your previous mail you said the pointer changed > when you clicked ... did it still do this? The cursor remains the default E cursor, as if the mouse was not hovering over any window. When clicking, the outline of the cursor blinks with a blue outline, just as it does when clicking on the desktop, except no menu appears. This happens no matter which window (or the desktop) the cursor is over. > nothing that leaps out at me... - try mousing over chrome - does the pointer > change? Nope, as mentioned above. This is part of the reason i thought it looked like no windows were getting focus. > the fact the restart request works means e's event loop is working and > processing events. you didn't try the gdb thing i mentioned right? No, i'm not sure how to do this: > get e attached to gdb > and set breakpoints for the buffer swapping. set a breakpoint for > eng_outbuf_flush (function name - gdb commant: b eng_outbuf_flush). the best > way to do this is ssh in from somewhere else as you will need e to keep > running > normally without it having screenblacked etc. I do have access to a Windows laptop with Putty to ssh in. Should the "attach" part happen before or during a freeze-up? And how do i actually attach a running E process to gdb and set the breakpoint? Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] enlightenment "freezing" when opening a new window... sometimes
On Sun, Dec 16, 2018 at 8:31 AM Carsten Haitzler wrote: > actually you may be surprised that focus may be getting set.. you just don't > SEE the results. if you have something that produces sounds when u type... > e.g. > you have speakers or headphones set up and volume up and maybe a terminology > terminal (and you haven't disabled the visual bell) and you have a prompt - > hit > "ctrl+g" and you'll get a "bing" sound as well as the red bell. It just happened again, so i was able to test this. There's definitely no focus -- i had just opened a terminal window with a password prompt, and no amount of clicking in that window would allow me to enter the password. I also had a terminology window open on the same virtual desktop, and there was no "bing" or visual bell. I used the miniature pager in the i-bar to switch to the desktop with Chrome open, and clicking on different tabs had no effect. The whole time, the mouse cursor never changed from its shape on the desktop, where it would normally change to a different shape when hovering over different windows. > FYI e is unaware of vt switches (in x11). it doesn't look, know or care and in > fact there isn't any X11 way to do this. Oh, i realise this, it's just that i was able to issue the restart command from a different vt, and that (sometimes, it did i this most recent instance) recovers. The vt switch itself doesn't do anything (if i switch back immediately, E is still mostly frozen). I'm including some syslog entries in case they're related... These are from immediately before i switched to a text vt to restart E, and as i mentioned, i did use the pager to attempt focussing other windows first. The freeze-up happened on opening a gnome-terminal window with a password prompt, so i've included the errors right before the terminal opened at Dec 18 13:57:23, which is followed by fprintd asking me to swipe the finger print reader. Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: ERR<3009>:eina_safety src/bin/e_pixmap.c:613 e_pixmap_size_get() safety check failed: cp == NULL Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: ## Copy & Paste the below (until EOF) into a terminal, then hit Enter Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: eina_btlog << EOF Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libeina.so.1#011 0x7fdd9fe101bc 0x7fdd9fde5000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libeina.so.1#011 0x7fdd9fe10f51 0x7fdd9fde5000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libeina.so.1#011 0x7fdd9fe1257b 0x7fdd9fde5000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libeina.so.1#011 0x7fdd9fe29c60 0x7fdd9fde5000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/bin/enlightenment#011 0x55693cb0a3c3 0x55693ca01000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/bin/enlightenment#011 0x55693ca7b9a6 0x55693ca01000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/bin/enlightenment#011 0x55693ca9026b 0x55693ca01000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/bin/enlightenment#011 0x55693ca9094b 0x55693ca01000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/enlightenment/modules/pager/linux-gnu-x86_64-0.22/module.so#011 0x7fdd8cc9865c 0x7fdd8cc9 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/enlightenment/modules/pager/linux-gnu-x86_64-0.22/module.so#011 0x7fdd8cc9887b 0x7fdd8cc9 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/enlightenment/modules/pager/linux-gnu-x86_64-0.22/module.so#011 0x7fdd8cc988f9 0x7fdd8cc9 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/enlightenment/modules/pager/linux-gnu-x86_64-0.22/module.so#011 0x7fdd8cc9900e 0x7fdd8cc9 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fea1399 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fea8adf 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fea43d9 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fea3017 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fe9d962 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fe9ddab 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fea4309 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fea32c7 0x7fdd9fe7d000 Dec 18 13:56:24 inanna /usr/lib/gdm3/gdm-x-session[2832]: /usr/local/lib/libecore.so.1#011 0x7fdd9fe9de47 0x7fdd9fe7d000 D
[e-users] enlightenment "freezing" when opening a new window... sometimes
Hi, I recently upgraded from EFL 1.20.7 to 1.21.1. At the same time i also enabled wayland support, however i'm still running Enlightenment under Xorg. I also ugraded E itself from 0.22.3 to 0.22.4. There were no other changes to my set up. Since this upgrade i've been running into this problem: every now and then, when opening a new window, enlightenment will "freeze". It's not reproducible as far as i can tell... sometimes E will be running for days, and suddenly stop when opening up one more window, sometimes it will freeze the moment i open up the first window afer logging in. The symptoms are that the mouse cursor still moves, but will not give focus to any window under it. If i click the mouse button, the cusor blinks, but nothing else happens. The exception seems to be that the un-hiding of the IBar at the bottom of my screen still works, as do the buttons in it, allowing me to at least cleanly log out. One other thing that does allow me to recover is switching to a text termial and running: DISPLAY=:0 enlightenment_remote -restart However this sometimes causes a whole other problem (again, not reproducible, but about 40% of the time) -- the windows get redrawn without any content. There is the faint outline of the window itself, and the top window bar for windows that have one, but the windows themselves are all empty. Minimizing, resizing the windows has no effect. I have to guess where the pull-down menus are (which DO get drawn, if i happen to find them) to exit the programs. Any clues? :) Thanks, -Conrad. -- Shine like thunder Cry like rain ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] multiple terminology windows
On Sat, Mar 31, 2018 at 10:48 PM, Carsten Haitzler wrote: > 1. you enabled "only launch single instance" under settings->desktop > environments in enlightenment Ugh, that was it! All fixed now. No idea how it got set, i hate that behaviour in other desktop/window managers... Thanks for your help, -Conrad. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] multiple terminology windows
On Wed, Mar 28, 2018 at 1:02 AM, Carsten Haitzler wrote: > On Tue, 27 Mar 2018 22:49:15 -0400 Conrad Knight said: >> Ok, rebooted. Now the desktop file is being read as usual on every >> login. But even with the -s flag, the problem persists. > > is it actually running terminology -s ? As far as i can tell, selecting terminology from the applications menu doesn't run anything, just changes focus to the existing one. There's no additional process. > are you sure there isn't yet another desktop file with terminology commands in > it? None that i can find, unless the ones in the terminology source directories count (~/src/terminology-1.1.1/data/desktop/terminology.desktop). > it may be your environment is missing env vars or has different XDG vars like > XDG_RUNTIME_DIR? if the 2 processes have differing XDG_RUNTIME_DIR dirs... > then > this might happen,. they think they are in differing "runtimes". maybe one > with > and one without XDG_RUNTIME_DIR ? XDG_RUNTIME_DIR is set to /run/user/1000. I don't know what enviroment variables the menu entry uses, but if it inherits them from enlightenment or enlightenment_start, those both have the same value. Is there perhaps a way of logging exactly what runs when i select the terminology menu entry? Thanks, -Conrad. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] multiple terminology windows
On Mon, Mar 26, 2018 at 10:52 PM, Carsten Haitzler wrote: > it should be. though it may have found another one somewhere more preferable > first? I did find one other, very old one, oddly called "terminology:terminology.desktop". But i deleted that before logging out and back in again. Ok, rebooted. Now the desktop file is being read as usual on every login. But even with the -s flag, the problem persists. > disable the multiple instance thing in behavior and it should work. Oh, i see! it's a terminology setting in the settings menu itself. Sorry, being dense :) I found the setting... and it was off. I tried turning it on, and running terminology from the command line again: new window appears, with a new process. So it seems that setting is being ignored when run from the command line (in an existing terminology window)? I turned the setting off again and the behaviour is unchanged: running from the command line gives me a new window, running from the applications menu just moves focus to the existing window. Anywhere else i can look for possible causes??/ Thanks, -Conrad. -- Shine like thunder Cry like rain -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] multiple terminology windows
> -s should bypass this and force a new process. it sees multiple > instances one process has become broken somehow. disabling that should > get you back to normal. Hmm, using "terminology" or "terminology -s" from the command line both start a new process. But then again, starting it from the command line has never been a problem. I changed the line in my terminology.desktop file from "Exec=terminology" to "Exec=terminology -s", but for some reason, that file is not being re-read. I logged out and back in again, and every *.desktop file in /usr/local/share/applications/ was accessed (looking at "ls -lu") EXCEPT terminology.desktop. And it's not a permissions problem, every file there has 0644. Why would E not be looking at this file, and only this file? Another symptom? E doesn't like terminology at all? :) I'll wait for the next EFL release and try again... Unless it's going to take a while and use what's in git instead. Thanks, -Conrad. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] multiple terminology windows
Hi, I've been having this issue for a while now. If i attempt to run terminology from Enlightenment's applications menu, it does not open a new terminal. Instead it will simply put the focus on an existing terminology window. If i happen to attempt to open a new terminology window on another virtual desktop, it also moves me to the desktop where that window happens to be. And it doesn't even open a new tab in that window. Every once in a while, though, i get the expected behaviour of a new terminology window opening. Unfortunately, this happens so rarely that i can't figure out under what conditions this works. The terminology help has the following (and i'm not sure if this appies) : -s, --singleForce single executable if multi-instance is enabled. But obviously i'm not using command line switches when trying to open in from the menu. I've also checked any terminology.desktop files to make sure the relevant line (Exec=terminology) does not have the -s option. Is this even what this option would do? Also, "if multi-instance is enabled" implies there must be some way of enabling it, but there don't appear to be any compile-time options that mention anything resembling this. Further, if i run terminology from the command line (either terminology's own, or another terminal such as uxterm) i DO get the expected behaviour. A new window opens every time. The only other thing i can think of that might be affecting this are the existing terminology window's remember options. I have these settings: Identifiers tab: window class: terminology Properties tab: position, size, border style, virtual desktop are checked Options tab: match only one window start this program on login The idea is to open a terminology window on login with the desired properties. I'm running terminology 1.1.1 (this may have started when i upgraded from a previous version, come to think of it) and Enlightenment 0.22.1. Does anyone have an idea what might be happening? Any more information i can provide? Thanks, -Conrad. -- Shine like thunder Cry like rain -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Odd desktop gadget behaviour
I've been struggling with this for a while, and really can't figure out the problem. On my desktop, i have a few gadgets added through E's menu (Settings > Gadgets). I've added the clock and the systray, and resized / positioned them to my liking. They're near the top right of the screen, but i've tried other positions as well. When E restarts, (either by logging out, or using the Enlightenment > Restart menu entry) the systray gadget appears in the top-left corner, extremely small. If i've logged in again, the icons in it are one or two pixels in size. If i simply restart E, the systray icons don't appear at all. This makes it very difficult to right-click on the gadget to resize it again, as there seems to be a single pixel sweet spot. Now, if i delete the systray gadget, leaving only the clock, the same happens to the clock when restarting. Although it's not quite as small and the displayed time is almost legible. But it is tiny and moved over to the top left of the screen. Wondering if it's a problem with the last gadget added, i've tried adding more after the systray, but it's always the systray the gets so drastically shrunk on restart. Even weirder: seeing as both the clock and the systray exhibit similar behaviour, i tried switching the order in which they're added. I added the systray and resized it (it defaults to the previous one's size, the tiny one in the top left corner). Then i add the clock, and as it appears, the systray shrinks back to near nothing in size in the top left corner. This is without restarting E! I've tried all the appearance options for the systray (fixed, horizontal, and vertical; plain and inset, though they look exactly the same). I can't think of anything else to get a systray to stick, without re-adding it every time i start E. I'm running E 0.21.10 at the moment, but i've had this problem for a long time, probably since 0.19 days. And every now and then, with some versions, i get lucky and i don't have the problem. Although it's quite likely by stumbling around adding/removing gadgets i accidentally found a combination that worked. I've also completely nuked the ~/.e directory a few times in case any stray, leftover settings were to blame. Any ideas if there's some E settings issue? Or perhaps a ./configure flag i'm leaving out (or adding) when building E? Any more information i can provide to track this down? Thanks anyone/all! -Conrad. -- Shine like thunder Cry like rain -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] cfg files
Hi, How do i read/edit E's *.cfg files? Or even just find out what settings are stored in each one? My settings seem to have gotten messed up over time (perhaps due to numerous upgrades?). An easy solution i've used before is just the config directory and start all over with a blank slate. However, there are quite a few settings i'd like to preserve. Starting completely over is a pain (as i metioned, i've had to do this before). Thanks! -- Shine like thunder Cry like rain -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] EFL build problem
I'm running into a problem while trying to build efl-1.20.5. I'm working with a system that pretty much had no header files, or anything needed for the compile to work. Many repeated runs of the "configure" script gradually helped me find the ones needed. Until it came the "make" step. Apparently I'm missing hwcomposer.h: In file included from modules/evas/engines/eglfs/evas_outbuf.c:1:0: modules/evas/engines/eglfs/evas_engine.h:18:25: fatal error: hwcomposer.h: No such file or directory I've tried looking for this header file, but have so far only found references to Android. I'm trying to get EFL working on a Linux box. Is there a development package i'm missing that "configure" didn't check for? Is there a "configure" or "make" option i should have used that i missed? Thanks, -Conrad. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] GLX Pixmap problem (SOLVED)
On Mon, Jun 13, 2016 at 3:21 AM, Carsten Haitzler wrote: > On Sat, 14 May 2016 22:39:28 -0400 Conrad Knight said: > >> I seem to have run into a conflict with efl, enlightenment, and nvidia >> GLX libraries, and i'm not sure where the problem lies. While running >> E-0.20.6, or E-0.21-beta, any windows opened have completely black >> contents. > > you could switch to egl/gles ... :) So, the problem doesn't happen with an older nvidia driver (358), and i wanted to try again with a more recent driver (367) and the released E-21 (previously i had been using the beta E-21). I had the exact same problem... So i tried the above suggestion. After digging through the configure options for the various libraries and a few false starts, i found you have to add: --enable-egl --with-opengl=es to the configure options for efl. I then recompiled elementary and enlightenment to be on the safe side and it works! Window contents are back, using the newer driver. Quick question, though: What possible consequences are there from switching to egl? Besides the obvious, that it works while gl didn't :) Specifcally, setting the above opengl option to "es" instead of the default "full"? Thanks, -Conrad. -- Whenever and wherever you want. -- Attend Shape: An AT&T Tech Expo July 15-16. Meet us at AT&T Park in San Francisco, CA to explore cutting-edge tech and listen to tech luminaries present their vision of the future. This family event has something for everyone, including kids. Get more information and register today. http://sdm.link/attshape ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] GLX Pixmap problem
I seem to have run into a conflict with efl, enlightenment, and nvidia GLX libraries, and i'm not sure where the problem lies. While running E-0.20.6, or E-0.21-beta, any windows opened have completely black contents. The window decorations are fine, and E's menus, etc. appear normally. But if i open a settings window from the menu, it's black. Terminology is likewise completely black. There are error messages in my .xsession-errors: ERR<11215>:evas-gl_x11 modules/evas/engines/gl_x11/evas_engine.c:2783 eng_image_native_set() GLX Pixmap create fail This does not occur if i switch my glx libraries to the Intel ones for this laptop. It also does not occur if i run terminology under another window manager. The nvidia drivers appear to be working correctly otherwise (i can run Steam games that depend on glx 1.4, for instance). I've tried recompiling efl, and enlightenment, to no avail. Would anyone have any suggestions what to try next? :) TIA, -Conrad. -- Whenever and wherever you want. -- Mobile security can be enabling, not merely restricting. Employees who bring their own devices (BYOD) to work are irked by the imposition of MDM restrictions. Mobile Device Manager Plus allows you to control only the apps on BYO-devices by containerizing them, leaving personal data untouched! https://ad.doubleclick.net/ddm/clk/304595813;131938128;j ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Weird problem on waking after suspend
On Wed, May 4, 2016 at 8:18 PM, Carsten Haitzler wrote: > never seen that. sounds like randr has reconfigured the screen to 10x10 or > whatever. e does have logs. look for RRR: in stdout/err logs from e. I didn't think of those... I ran E in a bare-bones X session and captured the output and errors (attached). There are a lot of RRR lines i hope someone can make sense of. At a first glance i can see where the lid closure and open events are detected, but in the end it settles back on the 1600x900 resolution. As a test, i had a unity session on another screen during the suspend/wake cycle and that didn't have resizing problems (though horrible redrawing errors). And i've certainly never had this problem with E 0.20. -Conrad. -- Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Weird problem on waking after suspend
Just ran into a problem i've never seen before... E0.21-b has been working fine for me so far, but today was the first time since switching that i had to take my laptop somewhere, so i closed the lid to suspend it. When i later opened the lid, the laptop woke up, but the display was blank, or so i thought at first. Near the top left corner was the cursor, apparently pointing at a small (10 x 10 pixels, maybe?), blue-ish square. Moving my finger over the touchpad did not move the cursor. Experimenting a little, i realised that if i continually slid across the touchpad in the same direction, the mouse cursor would very gradually move in that direction. When i got to an edge of the tiny square, the cursor would appear to stutter, and bounce back a pixel -- edge flip! The tiny square was my desktop, so the colour made sense, the cursor movement was scaled to the same miniature size... it made sense. But why? And, by the way, the cursor itself was full sized, much larger than the tiny box itself. No clues in any log files i could find. E responded to enlightenment_remote (-restart didn't fix it, -exit exited cleanly and reset the screen). X itself seems to have the correct resolution (the cursor was the right size). I haven't had any driver issues, and haven't changed them any time recently. Does anyone have an idea what might be going on? As i mentioned, this is the 0.21-beta release, so if it's a Known Problem, is there maybe a command line work-around to use for now? Thanks, -Conrad. -- Whenever and wherever you want. -- Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] Enlightenment DR 0.21-beta Release
Looking for some clarification on the build instructions... On the main download page, the build order no longer mentions the Elementary libs. But the README in the E21 source does... Are these libs required or not? Thanks! -Conrad. -- Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] E, xscreensaver, and memory issues
Hi, I wrote a while back about an apparent memory leak while running Enlightenment, but only have time to look into it every few weeks. Here is something I've found... The run-away memory usage only occurs while xscreensaver is running, and then not every time. I figured it might have to do with specific modes of the screensaver (i have it set to cycle through them randomly). A few quick tests found one that reliably causes memory usage to spike, "zoom", the last hack (in xscreensaver parlance) in the list for the random selections. These "hacks" can also be run manually without the xscreensaver executable, and i was able to verify that "zoom" run by itself in a window very quickly caused the memory problem. I had to switch to a text virtual terminal to kill the process. Even after this, "ps" shows that enlightenment continues to use over 40% of my system's memory. I can kind of see why the memory usage of a badly behaving program (or is it my video driver?) would show up as being used by enlightenment instead -- enlightenment has to actually display the pretty graphics on the screen. But why, after the process has been killed, is this memory not freed? And is there perhaps a compile option for E (or xscreensaver, for that matter) that would prevent all of this in the first place? Side note: "enlightenment_remote -restart" does free up the memory, but often this times out before the restart occurs. Also, the restart triggers the not-remembering-size-or-location problem with the clock module, a separate issue that has persisted into e20. Thanks, -Conrad. -- Whenever and wherever you want. -- Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] tracking down memory issue
> From: Cedric BAIL > As for tracking memory leak > the right tool this day is valgrind. It is going to slow down the > tracked software quite a lot ! > enlightenment_start -massif -display :1.0 Actually, the slowing wasn't too bad, and it was easier to catch it at the right moment running it directly this way in my regular X session. A few days of trial-and-error produced a massif output file with the run-away memory problem. > You can open > that file using massif-visualizer and look at what is constantly > increasing. I'm not sure what i'm looking at, but the main culprit seems to be "e_main.c:1062" with 1.1GiB. Unless i'm reading it backward, in which case it's something in libX11.so.6.3.0. At the same time, there's a second 1.0GiB pile of "e_main.c:1062" with "_ecore_x_event_handle_any_event (in /usr/local/lib/libecore_x.so.1.16.1)" at the other end of the chain. I can provide the massif output file (about 240k) if anyone would like to take a look. Thanks, -Conrad. -- Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] Window positions not being remembered
Hi, I'm Enlightenment 20.2, and am having trouble with window positions being remembered. I have a terminology window set to open on E's start-up, and i set it up as desired, then selected the following options from the window -> remember settings: First screen: none of the radio buttons are checked Press Advanced, Identifiers tab: check box for Window name is checked. Properties tab: Position, Size, Border style, and Virtual Desktop are checked. Options tab: Match only one window, and Start this program on login are checked. When i log out and back in, the terminology window is opened and all of the settings are applied except for position. The window appears on the desired virtual desktop (not the one immediately visible on logging in), but in the top left corner of that desktop instead of where i had previously placed it. This was previously working in E19, but i deleted my entire ~/.e directory, as well as ~/.cache/efreet and ~/.cache/evas_gl_common_caches when upgrading to avoid conflicts from outdated and incompatible settings. Are there any others i possibly missed, considering i also had E18 installed at one point? In addition, i'm also having trouble with the clock module. I have it added as a desktop gadget, and it first appears as a miniscule clock in the top left corner. It is just large enough to allow me to right-click on it, so that i can resize it and move it where i'd like it (just below another window near the top right corner). However, whenever i restart E, it forgets both this size and position and returns to its tiny corner. I remember having this problem in E19 as well, but not how i solved it. As the size is also being forgotten, and there aren't explicit remember settings for the desktop gadget, i don't know if this is related to the terminology position issue. Are there any config files i can check to see if they're being correctly updated? Is there any way to manually correct these config files? The only files i can find are in a binary format, and do not appear to be written to at the same time any settings are changed. Thanks, -Conrad. -- Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] tracking down memory issue
Hi, I posted before about build problems with E19, and seem to have solved those. For the record (and anyone searching the mailing list archives)... - edje segfaults during elementary build was caused by left over modules in /usr/local/lib/edje/. Have to run "make uninstall" in _both_ efl and elementary source directories before building. Although edje is installed with efl, some of its modules are installed by elementary. These are apparently not compatible when a newer efl's edje tries to load them. - constant crashing and respawning of efreetd was caused by loading contents of a cache directory. As this was presumably first created by E18, format change or corrupted contents were likely to blame. Nuking the cache directory (~/.cache/efreet, i think it was) fixed this. Anyway, with these problems solved, I decided to have a go at enlightenment 0.20.2, with efl/elementary 1.16.1. It all works, but i'm left with what appears to be an intermittent memory leak. Currently I have the following status for enlightenment, having just restarted it: ~$ ps u -p 3303 USER PID %CPU %MEMVSZ RSS TTY STAT START TIME COMMAND conrad3303 5.9 1.4 790196 119480 ? Rl 11:39 1:45 /usr/local/bin/enlightenment Leaving this running ovenight, with pretty much just chrome and xscreensaver running, that %MEM value went up to just over 50% some 12 hours later. Issuing "enlightenment_remote -restart" brought that right back down to 1.2%. Then, barely some 3 hours later with more heavy use (opening/closing windows, etc.) it was up over 80%. So much so, that the enlightenment desktop was unresponsive except for showing cursor movement. I had to switch to a text terminal to kill enlightenment. I tracked down the efreetd crashing problem by running it under gdb and using a backtrace. I'm afraid i have no idea how to go about finding what's causing this memory usage, though. Any suggestions for debugging welcome! :) Thanks, -Conrad. -- Whenever and wherever you want. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] build problem
> From: Carsten Haitzler (The Rasterman) > edje_cc doesnt - but libedje can load modules but directory existence does > not mean a module loads. it needs a module.so inside to do anything... without > a backtrace we dont have enouhg info to go on Ok, so that's not the problem... More digging had shown that i AM able to get edje_cc to run, as long as the "-fastdecomp" option is omitted. I've also noticed since compiling 1.16 that the cpu usage for enlightenment (and terminology) is consistently high. And i also discovered my syslog was rapidly filling with lines such as: kernel: [ 3574.458558] traps: efreetd[24960] general protection ip:7f3768a6c568 sp:7ffcc3cd1390 error:0 in libemile.so.1.16.0[7f3768a62000+11000] and that efreetd was constantly crashing and respawning. I managed to get backtraces and found both efreetd and edje_cc (when using -fastdecomp) were segfaulting for the same reason. Here's one of them: Program received signal SIGSEGV, Segmentation fault. 0x7454e894 in LZ4HC_compress_generic.constprop.2 () from /usr/local/lib/libemile.so.1 (gdb) bt #0 0x7454e894 in LZ4HC_compress_generic.constprop.2 () from /usr/local/lib/libemile.so.1 #1 0x7454fc60 in LZ4_compressHC2 () from /usr/local/lib/libemile.so.1 #2 0x74546eb0 in emile_compress (data=data@entry=0x869380, t=t@entry=EMILE_LZ4HC, l=l@entry=10) at lib/emile/emile_compress.c:67 #3 0x7710ff59 in eet_data_image_lossless_compressed_convert (compression=10, alpha=1, h=64, w=64, data=, size=0x7fff775c) at lib/eet/eet_image.c:735 #4 eet_data_image_encode_cipher (data=0x77fd6000, cipher_key=cipher_key@entry=0x0, w=64, h=64, alpha=alpha@entry=1, comp=comp@entry=10, quality=0, lossy=EMILE_IMAGE_LOSSLESS, size_ret=0x7fff7824) at lib/eet/eet_image.c:1589 #5 0x771102b6 in eet_data_image_encode (data=, size_ret=size_ret@entry=0x7fff7824, w=, h=, alpha=alpha@entry=1, comp=comp@entry=10, quality=0, lossy=EMILE_IMAGE_LOSSLESS) at lib/eet/eet_image.c:1649 #6 0x77110312 in eet_data_image_write_cipher (ef=0x1d56a30, name=name@entry=0x7fff7940 "edje/images/68", cipher_key=cipher_key@entry=0x0, data=, w=, h=, alpha=1, comp=10, quality=0, lossy=EMILE_IMAGE_LOSSLESS) at lib/eet/eet_image.c:1350 #7 0x77110395 in eet_data_image_write (ef=, name=name@entry=0x7fff7940 "edje/images/68", data=, w=, h=, alpha=, comp=10, quality=0, lossy=EMILE_IMAGE_LOSSLESS) at lib/eet/eet_image.c:1374 #8 0x00405ce7 in data_thread_image (data=data@entry=0x302b700, thread=thread@entry=0x0) at bin/edje/edje_cc_out.c:1011 #9 0x00407d3c in data_image_preload_done (data=0x302b700, e=, o=0x8024a126, event_info=) at bin/edje/edje_cc_out.c:1081 #10 0x0040b0eb in data_write_images (ef=ef@entry=0x1d56a30, image_num=image_num@entry=0x7fffbad8) at bin/edje/edje_cc_out.c:1143 #11 0x0040c146 in data_write () at bin/edje/edje_cc_out.c:2336 #12 0x0040443f in main (argc=8, argv=0x7fffdcf8) at bin/edje/edje_cc.c:399 I can recompile with any necessary flags to get more information, just let me know :) Thanks, -Conrad. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] build problem
Sorry to bring this up again, but i ran into the same issue with the efl-1.16 release. During the elementary build process, the freshly-compiled edje_cc (from efl) crashes with a segfault. This time i wasn't able to finish building no matter what combination of CFLAGS i used. I making sure there were absolutely no left-over efl-1.15 libraries left, i manually searched through /usr/local/lib/. Previously i had been relying on "make uninstall" in the previous version's source directory. I found an empty directory: /usr/local/lib/edje/modules/elm/v-1.15 After deleting this directory, the build completed without problems! Is it possible that edje_cc loads modules dynamically at build time, based only on the names of directories it finds? In this case, edje_cc is built first (with the efl libraries) and then used to build the elementary libraries, which apparently includes some of these modules. (After the build finished and i did a "make install" for elementary, i found a new edje/modules directory, elm/v-1.16.) If this is indeed what caused the segfault, is this a bug in edje_cc? Trying to load a non-existent, or incompatible module? Or at least a problem with the logic in the build process/order? -Conrad. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] memory usage
Hi, After resolving my build problems earlier (-O3 in the CFLAGS caused edje_cc to segfault while elementary was building, -O2 worked), i'm now noticing extreme memory usage with Enlightenment. Not at first, it gradually goes up over time. I first noticed that after Enlightenment runs for a while, the machine gets a bit "laggy" for want of a better description. Windows take a while to respond when switching focus, clicks take longer to register, stuttering while scrolling in Chrome... Wondering what was up, I ran "top" just to see if some process had run away, and found this: PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND 8867 root 20 0 316872 128708 29988 S 26.1 1.6 1765:45 Xorg 17873 conrad23 3 1627332 318176 84944 R 10.6 3.9 1876:52 chrome 1174 conrad23 3 1611356 555672 162412 S 5.9 6.9 80:20.43 chrome 20550 conrad23 3 2926800 738060 49996 S 5.9 9.1 284:31.02 firefox 9718 conrad20 0 1975404 1.179g 86444 S 5.6 15.3 604:09.09 enlightenment 3307 conrad23 3 1225216 457652 69544 S 3.6 5.7 15:17.08 chrome ... (That looked better with a mono-spaced font...) The memory usage goes right back down again after selecting "Restart" from the Enlightenment menu. This was much worse at first, after I used the "-O2 -ffast-math -m=native" CFLAGS, going up to 4.5g under RES overnight, the VIRT number had to be abbreviated with a 'g' as well, and the %MEM value was over 50%! While tracking down the previous build problem I tried with just "-O3" and then just "-O2" to verify that was the cause of the previous segfault, and that last build is the one that's currently running. This seems to be using up memory at a slower rate. I gave up trying to get the libraries freshly from git to build... there were too many conflicting libwhatever-1.so -> 1.15.so / 1.16.so conflicts, and some enlightenment modules were looking for non-existent files in E20 vs E19 directories. Should i just wait for the next release, or will i simply run into the same problems? Should I ditch this new gcc 5.2.1? :) Is there any other way to track down / solve this excessive memory usage? Thanks, -Conrad. -- Whenever and wherever you want. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] build problem
> From: Philippe Jean Guillaumie > Same error building elm on Ubuntu 15.10. > As a workaround use this: CFLAGS="-O2 -ffast-math -march=native" Uninstalled everything and started over with the tarballs for efl-1.15.2, etc. and used these flags instead. That seems to have worked! > From: Carsten Haitzler (The Rasterman) > that ... smells of an optimizer bug (-O3 crashes, -O2 is fine), but of course > it's a tall order to blame gcc. this would need deep down serious debugging > ... > but i have found compiler bugs before... it does happen. I'm also using Ubuntu 15.10... They updated gcc from 4.9.something to 5.2.1. Maybe related? Thanks, -Conrad. -- Whenever and wherever you want. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
[e-users] build problem
Hi all, I'm having trouble building the EFL, etc. libraries... I typically download the tarballs and compile from those, but i've also now tried following the directions on: https://www.enlightenment.org/docs-efl-start Compiling all the libraries (in the order specified) still results in the same problem. Since the last time i had a successful build, the instructions have been updated to include (or maybe they were there before, but i just missed them) To ensure your code is compiled with decent optimzations you should also set this up in your environment: export CFLAGS="-O3 -ffast-math -march=native" efl, emotion_generic_players, and evas_generic_loaders all build fine, but when it comes to elementary i get the following error towards the end: make[2]: Entering directory '/home/conrad/src/e19/elementary/data' Making all in themes make[3]: Entering directory '/home/conrad/src/e19/elementary/data/themes' EDJ default.edj /bin/bash: line 3: 9778 Segmentation fault (core dumped) /usr/local/bin/edje_cc -fastdecomp -id . -fd . -id ../../data/themes/img -fd ../../data/themes/fnt -sd ../../data/themes/snd -l ../../COPYING -a ../../AUTHORS -fastdecomp ../../data/themes/default.edc ../../data/themes/default.edj Makefile:1545: recipe for target 'default.edj' failed make[3]: *** [default.edj] Error 139 If i start over (make clean, run configure again, etc.) and don't set those optimization CFLAGS, all goes well. What could be causing this? Is there any way to track down the problem? There is a work-around, but that means everything is built without the aforementioned "decent optimzations". Thanks, -Conrad. -- Whenever and wherever you want. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
Re: [e-users] umount external sub does not work because enlightenment process keeps file open
> From: Carsten Haitzler (The Rasterman) > settings -> advanced -> performance > under "caches" > "cache flush interval" - default is 512 ticks. top slider. I've run into this problem before, too. After reading this, I checked this setting and it is still set to the default, which would be slightly over a minute. I have been unable to unmount a USB stick 20 minutes after closing EFM. Either this setting is being ignored, or something else is keeping the file open... -Conrad. -- ___ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
