On Wed, 23 Aug 2006 18:28:08 +0200 mathieu <[EMAIL PROTECTED]> babbled:
> Hi, > > Please find some more informations about this bug. I compile the last > cvs version of e17 today, and the good thing is that it seems a big more > stable when xmms is running. I needed to spend some time to crash e17 > today, but it was still possible. The good point is that I have > debugging informations and I had also the window indicating that e > crashed thing I didn't have before. The debugging informations are > included in the attach file. Unfortunately some of the variables are not > accessible. > > This time this problem arrives when I press ctrl+a combinason with xmms > playlist open. that is weird. looked at the backtrace - something has corrupted libc's memory alloc keeping data - but what - don't know. that free should work as between the calloc and the free clients is not played with. I have put in some extra count checks and error dialogs that should avoid any memory buffer overruns - if they ever can happen and detect mis-matches of count and lists. so cvs update (and yes - live with the pain that themes broke etc. etc. etc.) and see if you get the dialogs. if you don't then its quite likely a deeper memory corruption issue that affects libc's memory accounting data. then you need to dig out valgrind to catch that :( > Mathieu > > >> it seems that there are a lot of problems with xmms, that some people > >> are experiencing. > >> > >> Here, enlightenment just freezes when xmms is started. It doesn't > >> segfault, but it can't be restarted, or recover in any other way. > >> > > > > gdb attach - find out where its hung. > > > > also cvs update - i could be trynig to segv but is spinlocked out. > > > > > >> В нд, 2006-08-20 в 13:27 +0200, Mathieu Taillefumier написа: > >> > >>> Hello, > >>> > >>> I have some question about a very strange thing with e17. Does anyone > >>> experience any problems using xmms and e17 at the same time on system > >>> based on glibc 2.4 ? In fact I cannot use both of them because I have a > >>> crash of X free and e17 otherwise with a very strange result. All > >>> programs are dead excepted enlightenment. The debug session indicates > >>> that this problem occurs when the program want to free some memory, but > >>> that's all. The glibc indicates me an abort signal some times with a > >>> list of all libraries which are running on memory. Valgrind session > >>> gives me now informations excepted few possible memory leaks (). It is > >>> even worse than that It works perfectly with the combination > >>> Valgrind+enlightenment+xmms. Do you have any ideas where can come from > >>> Rasterman suggested that it is a bug in the C library, it is possible > >>> but I am not sure. > >>> > >>> For those how want to try to reproduce this bug on others systems (glibc > >>> 2.3.6 for instance). > >>> > >>> Run e17 normally and run xmms using the menu or any thing else. It seems > >>> to be not important. After that click on xmms to open and close the > >>> principal window. With chance you may crash e17 very fast. Before > >>> reporting bugs to e17 or glibc dev, it would be good to have an other > >>> person how can reproduce this error. If some of you have some ideas to > >>> obtain more informations about this problem please manifest yourself ;-). > >>> > >>> of course e17 without xmms works perfectly and xmms without e17 works > >>> also perfectly. > >>> > >>> Regards > >>> > >>> Mathieu > >>> > >>> > >>> > >> -- > >> Виктор Кожухаров /Viktor Kojouharov/ > >> > >> > > > > > > > > -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) [EMAIL PROTECTED] 裸好多 Tokyo, Japan (東京 日本) ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ enlightenment-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-users
