From: Axel S. <Axe...@in...> - 2010-07-16 11:46:14
|
Hi Vivian, this bug is now fixed in darcs. See previous email! Cheers, Axel On 16.07.2010, at 12:56, Axel Simon wrote: > Hi Vivian, > > On 16.07.2010, at 11:36, Alexander McPhail wrote: > >> Hi Axel, >> >> Yes, I think I have reproduced on linux x86-64, P4, debian sid: >> >> http://hackage.haskell.org/trac/gtk2hs/ticket/1183. >> >> With the hackage version of gtk there was a variable time delay >> between the window being drawn (I could resize and move, also) and >> the segfault crash (as described in the bug ticket). >> >> After installing from darcs (darcs get --lazy), the crash now occurs >> immediately (which I think might be a different bug). >> >> The attached file is a patch against Graphics/UI/Gtk/General/ >> hsgthread.c to use pthreads as documented on the ticket. However, >> the immediate crash still occurs. >> > > There is a bug somewhere in that g_thread_init fills a structure with > the addresses of lock und unlock functions but later on these values > are read as something different. I don't know if this is a ghci fault, > but I think it is. > >> Now I know this is a bug, I shall attempt to investigate further. >> Any helpful hints would be appreciated. Is there a toplevel option >> to define -DDEBUG? >> > > Do read: > > http://www.haskell.org/pipermail/glasgow-haskell-users/2010-July/019020.html > > You could compile Gtk+ with debugging support. I for now have merely > installed the debug symbols. I don't think debugging support would > help much since this is a very low-level problem and has nothing to do > with widget drawing or some such. > > Cheers, > Axel > >> Cheers, >> >> Vivian >> >> On 16 July 2010 18:57, Axel Simon <Axe...@in...> wrote: >> Hi Vivian, >> >> >> On 16.07.2010, at 01:40, Alexander McPhail wrote: >> >> Dear gtk2hs-users, >> >> Is there any fundamental reason why Gtk can not be used from ghci? >> I get crashes when (i) running from ghci, (ii) using forkIO to wrap >> Gtk code, and (iii) using forkOS to wrap Gtk code. >> >> >> The gtk package does work in ghci. There is a weird bug we haven't >> understood that I suspect is a problem in ghci. It only occurs on >> x86-64 platforms. I've only reproduced this on Linux, it might also >> happen on 64-bit Windows and Mac OS 10.6. >> >> I suspect you hit this bug: >> >> http://hackage.haskell.org/trac/gtk2hs/ticket/1183 >> >> >> Is that it? >> >> Cheers, >> Axel >> >> Thanks, >> >> Vivian >> >> -- >> --- >> yolar et elver. >> --- >> >> DISCLAIMER >> >> This transmission contains information that may be confidential. It >> is intended for the named addressee only. Unless you are the named >> addressee you may not copy or use it or disclose it to anyone else. >> >> ------------------------------------------------------------------------------ >> This SF.net email is sponsored by Sprint >> What will you do first with EVO, the first 4G phone? >> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first_______________________________________________ >> Gtk2hs-users mailing list >> Gtk...@li... >> https://lists.sourceforge.net/lists/listinfo/gtk2hs-users >> >> >> >> >> -- >> --- >> yolar et elver. >> --- >> >> DISCLAIMER >> >> This transmission contains information that may be confidential. It >> is intended for the named addressee only. Unless you are the named >> addressee you may not copy or use it or disclose it to anyone else. >> >> <pthread.patch> > > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by Sprint > What will you do first with EVO, the first 4G phone? > Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first > _______________________________________________ > Gtk2hs-users mailing list > Gtk...@li... > https://lists.sourceforge.net/lists/listinfo/gtk2hs-users |