This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Product Version = NetBeans IDE 8.1 (Build 201510222201) Operating System = Linux version 3.16.7-29-desktop running on amd64 Java; VM; Vendor = 1.8.0_60 Runtime = Java HotSpot(TM) 64-Bit Server VM 25.60-b23 Reproducibility: Happens every time STEPS: Run the program with breakpoint enabled ACTUAL: Warning message displayed and then gdb exits EXPECTED: Step out to the next instruction in the thread... (was working in 2014)...
Created attachment 157316 [details] IDE log
I tested in depth the issue with different code I had I know they are ANSI compliant and respected the good way of programming multithreaded application. I compared on two computers: one that debug threads without any problem and this one that can't go further than two or three steps without displaying a warning message and stopping the debugger (gdb). The only difference between the two environments was the kernel I/O scheduler parameter set to "deadline" (opensuse 12.2 on kernel 3.4.63)... I then set the kernel I/O scheduler to deadline on the other computer (opensuse 13.2 on kernel 3.16.7) and... it worked !! Now, I can step on a thread without having the issue...
I shout it worked too early, sorry ! It improved the number of steps before the issue came again :{ I stepped in console mode and I got more informations: internal-error: switch_back_to_stepped_thread: Assertion `!schedlock_applies (1)' failed. That may mean that I would set the gdb's parameter scheduler-locking to another value ? YES !! I put those lines in the /etc/gdbinit file set libthread-db-search-path /usr/lib64 set debug libthread-db 1 set scheduler-locking off And now I can REALLY say it works !!
I have more additional informations to give to avoid being very frustrated: In your project properties, you have to point /etc/gdbinit instead of /root/.gdbinit. Or create a symbolic link from /root/.gdbinit to /etc/gdbinit to make it work, otherwise you should have same behaviour ^^
Do you speak about GDB 7.8 bug? https://lists.gnu.org/archive/html/info-gnu/2014-10/msg00018.html https://sourceware.org/bugzilla/show_bug.cgi?id=17408
Yes, I use gdb 7.8.