On Mon, 24 Dec 2012 12:01:31 +0100 Dominic Fandrey <kamik...@bsdforen.de> said:
> On 24/12/2012 03:21, Carsten Haitzler (The Rasterman) wrote: > > On Mon, 24 Dec 2012 01:26:19 +0100 Dominic Fandrey <kamik...@bsdforen.de> > > said: > > > >> enlightenment_start of the e17 alpha just loads up the background window > >> and mouse cursor. Then it hangs. All I can do is move the mouse cursor > >> in front of a black background. > >> > >> ESTART: 0.48024 [0.00001] - E_Test Init > >> ESTART: 0.48025 [0.00001] - E_Test Done > >> ESTART: 0.48026 [0.00001] - E_Shelf Init > >> ESTART: 0.48026 [0.00001] - E_Shelf Init Done > >> ESTART: 0.48027 [0.00001] - E_Shelf Config Update > >> ESTART: 0.48028 [0.00001] - E_Shelf Config Update Done > >> ESTART: 0.48029 [0.00001] - Manage all windows > >> ESTART: 0.48179 [0.00150] - MAIN LOOP AT LAST > >> ESTART: 0.52705 [0.04526] - SLEEP > >> PAUSE ! > >> > >> It's stuck there, nothing happens any more. > >> > >> I'm running FreeBSD 9.1 amd64, if that's relevant somehow. I have no > >> what to do from here. > >> > >> I removed my old ~/.e to make sure it doesn't interfere. > > > > e has segfaulted. e_start is babysitting the e process (it ptraces it) and > > has probably tried running gdb on it etc to make a ~/.e-crashdump file. - > > if you have gdb etc. ... if not u can gdb attch to the enlightenment > > process (not enlightenment_start) and get one too. e_start would/should > > have spawned the white box of death alert process to tell u this - maybe > > for whatever reason its not visible on your xserver... > > > > using efl from svn or released efl (1.7.4) ? > > > > The released 1.7.4. > > Any way, I'm not the only with the problem on FreeBSD and we narrowed > it down to evas-core: > http://docs.freebsd.org/cgi/mid.cgi?50D83078.2020203 > > I didn't do a real debugging session (I didn't build with debugging > symbols), but I also posted a summary of the startup procedure with > all the signals gdb trapped: > http://docs.freebsd.org/cgi/mid.cgi?50D7BDCF.4040401 > > It goes into "PAUSE !" mode after segfaulting. The issue can be redeemed > by building evas-core with gcc 4.2 instead of clang 3.1. I cannot tell > whether that is a problem in evas-core or clang. I might try it with > clang 3.2 later. well there's no backtrace in those links... that doesn't help us much at all. :) change in compiler and it works? odd in the least. options: 1. bug in code that only manifests on a different or specific compiler (clang/llvm) OR 2. bug in that other compiler. that's the extent of the informtion so far. as best i know there are people who compile e and efl using clang on linux and i don't hear any reports of issues there related to that that i remember. so right now its a fbsd+clang+e/efl issues ... which means you're the one reproducing and if you want it fixed... you'll have to provide more details like backtraces... full ones etc. :) (gdb debuging enabled, maybe use valgrind to catch stuff, print of vars and what not). > Regards > > -- > A: Because it fouls the order in which people normally read text. > Q: Why is top-posting such a bad thing? > A: Top-posting. > Q: What is the most annoying thing on usenet and in e-mail? > -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) ras...@rasterman.com ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
