On Fri, 31 Jul 2020 19:23:12 -0700 Jose R R <jose....@metztli.com> said:
> On Fri, Jul 31, 2020 at 6:54 PM Carsten Haitzler <ras...@rasterman.com> wrote: > > > > On Fri, 31 Jul 2020 16:34:05 -0700 Jose R R <jose....@metztli.com> said: > > > > you may want to install those... > It is overkill as I only install efl-dev to build Terminology, rage, > ephoto, etc., but if you suggest... ok. to report bugs with backtraces that are more useful.... otherwise i cant do much with this report.. even with full vt info i probably can't but i can do much less with minimal bt debug info. > > > Apologies. > > > > > > On Fri, Jul 31, 2020 at 2:46 PM Carsten Haitzler <ras...@rasterman.com> > > > wrote: > > > > > > > > On Fri, 31 Jul 2020 11:44:22 -0700 Jose R R <jose....@metztli.com> said: > > > > > > > > > On Fri, Jul 31, 2020 at 5:31 AM Carsten Haitzler > > > > > <ras...@rasterman.com> wrote: > > > > > > > > > > > > On Fri, 31 Jul 2020 03:03:13 -0700 Jose R R <jose....@metztli.com> > > > > > > said: > > > > > > > > > > > > well e got locked up in reading async events ... because evas was > > > > > > still seemingly rendering... i see no rendering thread for evas... > > > > > > what was on the screen? using gl for compositing or software? > > > > > > software i suspect. > > > > > A large message across the top of the screen with blinking red letters > > > > > including something like 'software failure'; sorry I did not write > > > > > down the whole message but I had a couple of those experiences, as > > > > > well, with the immediately previous version of E. > > > > > > > > well if you sent a segv signal.. that's what you would expect... a segv > > > > handler kicked in... :) the crashdump is what counts. dont worry about > > > > the message. > > > > > > > > > E 0.24.2 built nicely with e17_0.24.1-2.debian.tar.xz (dch -v 0.24.2-1 > > > > > ...) which only has the debian/changelog minor text strings difference > > > > > with current: > > > > > < > > > > > http://deb.debian.org/debian/pool/main/e/e17/e17_0.24.2-1.debian.tar.xz > > > > > > > > > > > > > > > > Accordingly, I installed E 0.24.2 from the freshly built Debian > > > > > packaging for E 0.24.1-2 -generated Debian Buster backports: > > > > > enlightenment_0.24.2-1_amd64.deb > > > > > < > > > > > https://metztli.it/readOnlyEphemeral/enlightenment_0.24.2-1_amd64.deb > > > > > > > > > > > > > > > > and enlightenment-data_0.24.2-1_all.deb > > > > > > > > well it is missing full debug symbols so i'm limited in what i can say > > > > from the backtrace... :) > > > > > > This? > > > < > > > https://metztli.it/readOnlyEphemeral/enlightenment-dbgsym_0.24.2-1_amd64.deb > > > > > > > > > > I also missed this other package -- for development: > > > < > > > https://metztli.it/readOnlyEphemeral/enlightenment-dev_0.24.2-1_amd64.deb > > > > > > > > > > > > > > > > < > > > > > https://metztli.it/readOnlyEphemeral/enlightenment-data_0.24.2-1_all.deb > > > > > > > > > > > > > > > > > > why is here still rendering items in the rendering list? it's > > > > > > waiting for it to clear out by processing events/input from pending > > > > > > render thread(s). his item is modified by the mainloop code not the > > > > > > thread so it doesnt need locks - a pipe is used to pass events back > > > > > > to the mainloop... > > > > > > > > > > > > so i'm unsure why this is happening other than the above. there is > > > > > > no render thread i see in the bt. well no active blocked one. > > > > > > thread 2 is the render thread sleeping waiting for commands to go > > > > > > do its thing... thread 1 is the mainloop stuck waiting for return > > > > > > input from the render thread. without lots of debug to show what > > > > > > all the sequence of events was on all threads.. i can't say much > > > > > > more as it;'s a snapshot of a system that's indeed locked up > > > > > > waiting on something that is not actually busy, but it certainly > > > > > > thinks it is. > > > > > > > > > > > > > Normally, > > > > > > > > > > > > > > < http://www.enlightenment.org/ss/e-5f23ea745ad658.07505304.jpg > > > > > > > > > > > > > > > But I was building a kernel, etc., and E became unresponsive... > > > > > > > > > > > > > > I had to exit E via triplet CTRL + ALT + F2; logging at shell then > > > > > > > executed killall -SEGV enlightenment. > > > > > > > > > > > > > > Attached is the log: > > > > > > > > > > > > > > ---------- Forwarded message --------- > > > > > > > From: Carsten Haitzler <ras...@rasterman.com> > > > > > > > Date: Sun, Jul 26, 2020 at 10:05 AM > > > > > > > Subject: [E-devel] Release: Enlightenment 0.24.2 > > > > > > > To: e <enlightenment-devel@lists.sourceforge.net>, e-users > > > > > > > <enlightenment-us...@lists.sourceforge.net>, > > > > > > > <enlightenment-annou...@lists.sourceforge.net> > > > > > > > > > > > > > > > > > > > > > Here is the follow-on beta tarball tarball: > > > > > > > > > > > > > > https://download.enlightenment.org/rel/apps/enlightenment/enlightenment-0.24.2.tar.xz > > > > > > > > > > > > > > Hilights: > > > > > > > * Fixed null exec command running > > > > > > > * Fixed handling of screen with no edid > > > > > > > * Fixed signal.h, execinfo.h include for some libc's > > > > > > > * Sped up pager thumb loading significant;y > > > > > > > * Handle zone changes during startup > > > > > > > * Longer efreetd timeout for update event > > > > > > > * Fixed e's xsettings support on 64bit to use the right > > > > > > > types > > > > > > > * Fixed preloading of icon > > > > > > > * Fixed e's wl backlight/dimming logic to be correct and > > > > > > > reliable > > > > > > > * Fixed noisy logging on invalid exec handle found > > > > > > > * Fixed shot blurriness if size didn't quite match > > > > > > > * Fixed battery on openbsd to use right constants > > > > > > > * Fixed cpu freq setting on freebsd etc. > > > > > > > * Handle error exits from dependency libs as an error and > > > > > > > bring up alert > > > > > > > * E system - isolate stdio so it doesn't affect ipc on > > > > > > > stdin/out > > > > > > > * Polkit & Askpass password - handle escaping right > > > > > > > * Fix tiling issue with float/unfloat > > > > > > > * Fixed shell autohide when clock data up > > > > > > > * Handle x io error with proper exit code and exit > > > > > > > > > > > > > > -- > > > > > > > ------------- Codito, ergo sum - "I code, therefore I am" > > > > > > > -------------- Carsten Haitzler - ras...@rasterman.com > > > > > > > > Possibly relevant: the prior E 0.24.1 version *sometimes* booted with > dialog complaining about efreet having directory writing issue: > < https://metztli.it/readOnlyEphemeral/shot-2020-07-15_10-31-21.jpg > > I verified permissions of directory but there were no anomalies. > ls -ltd ~/.cache/efreet > drwxr-xr-x 2 nehual nehual 20 Jul 31 10:58 /home/nehual/.cache/efreet/ > *nehual is a placeholder for the actual username under which E is running > > Notwithstanding, since I built and installed E 0.24.2, the efreet > issue dialog did not splash anymore when instantiating E environment. > > Nevertheless, today -- as I was building Terminology 1.8.0-1.1 -- > pertinent section of log called my attention: > ... > ERR<29441>:efreet_cache ../src/lib/efreet/efreet_cache.c:332 > efreet_cache_init() Failed to create directory > '/tmp/tmp.4aR9KO8e7j/.cache/efreet' > ## Copy & Paste the below (until EOF) into a terminal, then hit Enter > > eina_btlog << EOF > /usr/lib/x86_64-linux-gnu/libeina.so.1 0x7f67f722731c 0x7f67f71fa000 > /usr/lib/x86_64-linux-gnu/libeina.so.1 0x7f67f72280c1 0x7f67f71fa000 > /usr/lib/x86_64-linux-gnu/libeina.so.1 0x7f67f72296e3 0x7f67f71fa000 > /usr/lib/x86_64-linux-gnu/libefreet.so.1 0x7f67f6e08930 0x7f67f6ded000 > /usr/lib/x86_64-linux-gnu/libefreet.so.1 0x7f67f6df40ef 0x7f67f6ded000 > /usr/lib/x86_64-linux-gnu/libedje.so.1 0x7f67f6d88f34 0x7f67f6ce5000 > /usr/bin/edje_cc 0x5620b1c120f5 0x5620b1c01000 > /lib/x86_64-linux-gnu/libc.so.6 0x7f67f687b09b 0x7f67f6857000 > /usr/bin/edje_cc 0x5620b1c12dba 0x5620b1c01000 > EOF > ... > > *copied and pasted into another terminal and... > > < https://metztli.it/readOnlyEphemeral/efreet_cache-issues_07-31-2020_PST.png > > Yet: > ls -ltd > drwxrwxrwt 21 root root 24 Jul 31 19:15 /tmp/ > > Best Professional Regards. and... why cant it create that directory? what part of that dir path is missing? it looks like somehow $HOME and he cache dir has been redirected to /tmp/tmp.xxxxx which is really odd... -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- Carsten Haitzler - ras...@rasterman.com _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
