On 01/09/2017 08:50 AM, Jeff Hoogland wrote: > How does valgrind differ from Xephyr? Is there a wiki page explaining how > to do what you are suggesting? >
valgrind and Xephyr are too completely different things doing unrelated things, in enlightenments case its best that if you are running enlightenment under valgrind you run it in Xephyr because its going to be close to too slow to use. Xephyr lets you run an xsession such as enlightenment within a window of another running xsession so you can run a instance of e in a window inside another instance of e. Valgrind is a memory debugging tool it tracks when blocks of memory are allocated and deallocated by doing this it can tell you when a block of code read / writes to invalid memory regions causing memory corruption which is likely whats causing your bug. Generally in code this happens when you have a pointer stored in 2 or more places and place one frees its memory but place 2 then decides to use it again later once the memory has been allocated to something else. > On Fri, Jan 6, 2017 at 2:39 AM, Carsten Haitzler <ras...@rasterman.com> > wrote: > >> On Fri, 6 Jan 2017 02:21:10 -0600 Jeff Hoogland <jeffhoogl...@gmail.com> >> said: >> >>> Having some users report segfaults with our E17 fork using the latest EFL >>> under our Ubuntu 16.04 base. >>> >>> Had someone run a back trace and post it here >>> <http://forums.bodhilinux.com/index.php?/topic/14036-moksha- >> segfault/?p=102195>. >>> Looks like it is dying in the EFL somewhere? I'm honestly not great at >>> debugging C code - anyone help point me in the right direction? >> >> the backtrace is "long after the bug already happened". the malloc heap is >> corrupt. libc has detected something is wrong and aborted during an >> allocation. >> what caused the issue is unknown. it could be efl. could be e. could be >> libpng. could be just about anything. it is unknown. the best way to find >> out >> who is doing this is to use valgrind to run execution and it can trap >> out-of-allocation writes and tell you the exact point where this write is >> done. >> that would provide the info needed to fix if it isn;'t already fixed in >> git efl >> - if the issue is there. if the issue is in e17 it'd point that out too. or >> wherever else it is. >> >> -- >> ------------- Codito, ergo sum - "I code, therefore I am" -------------- >> The Rasterman (Carsten Haitzler) ras...@rasterman.com >> >> > > -- Simon Lees (Simotek) http://simotek.net Emergency Update Team keybase.io/simotek SUSE Linux Adeliade Australia, UTC+9:30 GPG Fingerprint: 5B87 DB9D 88DC F606 E489 CEC5 0922 C246 02F0 014B
signature.asc
Description: OpenPGP digital signature
------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
