Hide Forgot
I just installed the gnome-builder package on Fedora Rawhide, and when I try to run it, it fails with a SIGSEGV: |> gnome-builder Segmentation fault (core dumped) Running using gdb and getting a back trace: gdb>r Starting program: /usr/bin/gnome-builder [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". [New Thread 0x7fffea3bc700 (LWP 5792)] [New Thread 0x7fffe9bbb700 (LWP 5793)] [New Thread 0x7fffe393c700 (LWP 5794)] Thread 1 "gnome-builder" received signal SIGSEGV, Segmentation fault. 0x0000000000000000 in ?? () gdb>bt #0 0x0000000000000000 in () #1 0x00007ffff6370ac9 in __pthread_once_slow () at /lib64/libpthread.so.0 #2 0x00007fffcbed9aef in bmalloc::PerThread<bmalloc::Cache>::getSlowCase() () at /lib64/libjavascriptcoregtk-4.0.so.18 #3 0x00007fffcbed98f9 in bmalloc::Cache::allocateSlowCaseNullCache(unsigned long) () at /lib64/libjavascriptcoregtk-4.0.so.18 #4 0x00007fffcbec23a7 in WTF::StringImpl::createFromLiteral(char const*, unsigned int) () at /lib64/libjavascriptcoregtk-4.0.so.18 #5 0x00007fffcbec2401 in WTF::StringImpl::createFromLiteral(char const*) () at /lib64/libjavascriptcoregtk-4.0.so.18 #6 0x00007fffcbeccbd0 in WTF::String::String(WTF::ASCIILiteral) () at /lib64/libjavascriptcoregtk-4.0.so.18 #7 0x00007fffcc8428c7 in _GLOBAL__sub_I_PasteboardHelper.cpp () at /lib64/libwebkit2gtk-4.0.so.37 #8 0x00007ffff7de78d1 in call_init.part () at /lib64/ld-linux-x86-64.so.2 #9 0x00007ffff7de79db in _dl_init () at /lib64/ld-linux-x86-64.so.2 #10 0x00007ffff7dec946 in dl_open_worker () at /lib64/ld-linux-x86-64.so.2 #11 0x00007ffff7de7774 in _dl_catch_error () at /lib64/ld-linux-x86-64.so.2 #12 0x00007ffff7debf19 in _dl_open () at /lib64/ld-linux-x86-64.so.2 #13 0x00007ffff615ef09 in dlopen_doit () at /lib64/libdl.so.2 #14 0x00007ffff7de7774 in _dl_catch_error () at /lib64/ld-linux-x86-64.so.2 #15 0x00007ffff615f594 in _dlerror_run () at /lib64/libdl.so.2 #16 0x00007ffff615efa2 in dlopen@@GLIBC_2.2.5 () at /lib64/libdl.so.2 #17 0x00007ffff5d587d1 in g_module_open () at /lib64/libgmodule-2.0.so.0 #18 0x00007ffff716120d in peas_object_module_load () at /lib64/libpeas-1.0.so.0 #19 0x00007ffff2ce9ae4 in g_type_module_use () at /lib64/libgobject-2.0.so.0 #20 0x00007ffff7163923 in peas_plugin_loader_c_load () at /lib64/libpeas-1.0.so.0 #21 0x00007ffff715daad in peas_engine_load_plugin_real () at /lib64/libpeas-1.0.so.0 #22 0x00007ffff2cc616f in g_cclosure_marshal_VOID__BOXEDv () at /lib64/libgobject-2.0.so.0 #23 0x00007ffff2cc3617 in _g_closure_invoke_va () at /lib64/libgobject-2.0.so.0 #24 0x00007ffff2cdde08 in g_signal_emit_valist () at /lib64/libgobject-2.0.so.0 #25 0x00007ffff2cde45f in g_signal_emit () at /lib64/libgobject-2.0.so.0 #26 0x00007ffff715d92b in peas_engine_load_plugin () at /lib64/libpeas-1.0.so.0 #27 0x00007ffff76c944d in ide_application_load_plugins () at /usr/lib64/gnome-builder/libide-1.0.so #28 0x00007ffff76c7575 in ide_application_local_command_line () at /usr/lib64/gnome-builder/libide-1.0.so #29 0x00007ffff358d575 in g_application_run () at /lib64/libgio-2.0.so.0 #30 0x0000555555554f73 in main () gdb>
You'll want to post a full backtrace (bt full) with debugging symbols. (Easiest way is to let ABRT do it for you.)
I just tried letting the automated bug reporting system report on this, but it said it was a duplicate of: https://bugzilla.redhat.com/show_bug.cgi?id=1251535 https://bugzilla.redhat.com/show_bug.cgi?id=1217952 so would not create a new report.
I tried "bt full" as a command to gdb but it only gave the same data as above. I guess I am using gdb wrongly.
Huh, it's clearly wrong... got to get it manually, then. :(
If you give me guidance as to what to do and collect, I will do so. Sorry to appear such a newb.
We need a stacktrace like you've already posted, but taken with 'bt full' instead of simply 'bt', and with debuginfo installed first. gdb prints a 'dnf debuginfo-install' command you can use to install the relevant debuginfo. Thanks!
I used an alias I picked up from the Web: https://blog.cryptomilk.org/2010/12/23/gdb-backtrace-to-file/. The result is the attached file. Hopefully this is the right stuff.
Created attachment 1151474 [details] Putative full backtrace
That looks good, thanks. I'm aware of another bmalloc crasher that was fixed recently; let's wait for the 2.12.2 update and if the crash is still occurring with that, then please leave a reminder here and I will forward this upstream.
This bug appears to have been reported against 'rawhide' during the Fedora 25 development cycle. Changing version to '25'.
Hi, have you seen this crash anytime recently (since updating to 2.12.2)? If not, we can probably assume it's the bmalloc crash that was fixed around the time you reported this.
(In reply to Michael Catanzaro from comment #11) > Hi, have you seen this crash anytime recently (since updating to 2.12.2)? If > not, we can probably assume it's the bmalloc crash that was fixed around the > time you reported this. not really.
I just started gnome-builder (3.22.2) and it seems to be working fine on this cursory check.