Bug 766520

Created attachment 545640 [details]
core dump produced by X

I tested with the latest rawhide version (xorg-x11-server-1.11.99.1-9.20111109.fc17), same result.

Funny thing is that after X crashes (and restarts), a lot of other programs crash as well (pidgin, lxdm-greeter, pcmanfm). In $HOME/.xsession-errors a lot of "Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0." entries can be seen.

The X server crashed on me twice this morning, without the Yubikey being attached. So the Yubikey thing might simply be a symptom, not a cause.

The Yubikey is definitely not the cause. Open a terminal and typing, well, pretty much anything can result in a crash. It renders the docking station useless this way.

Did some more digging with gdb:

[root@lptfrwi /]# gdb /usr/bin/X core.1430
GNU gdb (GDB) Fedora (7.3.50.20110722-10.fc16)
Copyright (C) 2011 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /usr/bin/X...Reading symbols from /usr/lib/debug/usr/bin/Xorg.debug...done.
done.
[New LWP 1430]
Missing separate debuginfo for
Try: yum --disablerepo='*' --enablerepo='*-debuginfo' install /usr/lib/debug/.build-id/9f/6a5fda6fec28b048eb78a4340676aa14bf9be5
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `/usr/bin/X -core -verbose -logverbose -background none :0 vt01 -nolisten tcp'.
Program terminated with signal 6, Aborted.
#0  0x000000366e236285 in __GI_raise (sig=6)
    at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
64	  return INLINE_SYSCALL (tgkill, 3, pid, selftid, sig);
(gdb) bt f
#0  0x000000366e236285 in __GI_raise (sig=6)
    at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
        resultvar = 0
        pid = <optimized out>
        selftid = 1430
#1  0x000000366e237b9b in __GI_abort () at abort.c:91
        save_stage = 2
        act = {__sigaction_handler = {sa_handler = 0x7dc880 <tmpBuffer.10170>,
            sa_sigaction = 0x7dc880 <tmpBuffer.10170>}, sa_mask = {__val = {
              4615396, 21111472, 1, 140735199068656, 8285024, 51797696, 0,
              140735199068000, 4614971, 64, 206158430232, 140735199068016,
              140735199067824, 140735199067936, 5747794, 1}},
          sa_flags = 1856038496, sa_restorer = 0x7defe0 <tty_attr>}
        sigs = {__val = {32, 0 <repeats 15 times>}}
#2  0x000000000046a40e in OsAbort () at utils.c:1251
No locals.
#3  0x000000000048193c in ddxGiveUp (error=EXIT_ERR_ABORT) at xf86Init.c:955
        i = <optimized out>
#4  0x0000000000466a52 in AbortServer () at log.c:416
No locals.
#5  0x0000000000466c37 in FatalError (
    f=0x579078 "Caught signal %d (%s). Server aborting\n") at log.c:547
        args = {{gp_offset = 24, fp_offset = 48,
            overflow_arg_area = 0x7fff778c40a0,
            reg_save_area = 0x7fff778c3fd0}}
        beenhere = 1
#6  0x0000000000467dbe in OsSigHandler (sip=<optimized out>, signo=3,
    unused=<optimized out>) at osinit.c:146
No locals.
#7  OsSigHandler (signo=3, sip=<optimized out>, unused=<optimized out>)
    at osinit.c:108
No locals.
#8  <signal handler called>
No symbol table info available.
#9  0x000000366e2e8863 in __select_nocancel ()
    at ../sysdeps/unix/syscall-template.S:82
No locals.
#10 0x0000000000460266 in WaitForSomething (pClientsReady=0x3165ec0)
    at WaitFor.c:232
        i = <optimized out>
        waittime = {tv_sec = 0, tv_usec = 199334}
        wt = 0x7fff778c4740
        timeout = <optimized out>
        clientsReadable = {fds_bits = {0 <repeats 16 times>}}
        clientsWritable = {fds_bits = {67872, 0, 8186520, 233779689216, 98240,
            52829152, 32752, 233779689216, 8186520, 234178491288, 21112336,
            21112336, 52422240, 16384, 21112712, 21112336}}
        selecterr = <optimized out>
        nready = 0
        devicesReadable = {fds_bits = {52422240, 16384, 21112712,
            140716555730464, 21112336, 52422400, 29, 140716558181936,
            21112568, 16384, 29, 18, 8186520, 21060864, 21057520, 21077400}}
        now = <optimized out>
        someReady = <optimized out>
#11 0x000000000043378a in Dispatch () at dispatch.c:381
        clientReady = 0x3165ec0
        result = <optimized out>
        client = <optimized out>
        nready = <optimized out>
        icheck = 0x7e6450
        start_tick = <optimized out>
#12 0x0000000000422db5 in main (argc=<optimized out>, argv=0x7fff778c4968,
    envp=<optimized out>) at main.c:287
        i = <optimized out>
        alwaysCheckForInput = {0, 1}
(gdb) list
59	    if (__builtin_expect (pid <= 0, 0))
60	      pid = (pid & INT_MAX) == 0 ? selftid : -pid;
61	#endif
62
63	#if __ASSUME_TGKILL
64	  return INLINE_SYSCALL (tgkill, 3, pid, selftid, sig);
65	#else
66	# ifdef __NR_tgkill
67	  int res = INLINE_SYSCALL (tgkill, 3, pid, selftid, sig);
68	  if (res != -1 || errno != ENOSYS)
(gdb) list
69	    return res;
70	# endif
71	  return INLINE_SYSCALL (tkill, 2, selftid, sig);
72	#endif
73	}
74	libc_hidden_def (raise)
75	weak_alias (raise, gsignal)

I am slowly going nuts here. It seems that the 'Enter' key on any keyboard except for the laptops internal keyboard causes the crash. No matter in what application I use the Enter key (tried geany, libreoffice-writer, lxterminal). The Yubikey is seen as an usb keyboard and does provide an CRLF after input, so that one triggers it as well. I have tried a Logitech UltraX and a Logitech Cordless Desktop EX110 keyboard.

Been fiddling around with xkb settings to no avail.

When the X server crashes it drops me back to lxdm (logical, cause systemd logs you out). However, attempting to log back in crashes the X server over and over again. Only a reboot would help.

I have tried the latest rawhide packages for mesa/xorg-x11-server, same issue.

I am in desperate need of some help and/or feedback.

This may be a kernel thing. I thought I solved it by using a rawhide 3.2 kernel. However, with kernel 3.2.0-2.fc17.x86_64 X keeps crashing (whether docked or not), kernel 3.2.0-0.rc7.git4.1.fc17.x86_64 works fine.

It does seem to be a whole lot more stable if I remove the 'rhgb quiet' boot parameters.

Sounds (In reply to comment #5)
> I am slowly going nuts here. It seems that the 'Enter' key on any keyboard
> except for the laptops internal keyboard causes the crash.

This symptom sounds similar to F15 Bug 720137 - X crashes when ENTER key is pressed , but worse if it's crashing over and over for you?

It certainly looks like the same problem, intermittent crashes on login when the enter keys is used. As I am using lxdm, this certainly is not a display manager issue.

I can say however that kernel 3.2 does not have this problem, as long as you omit the 'rhgb quiet' boot parameters (I read somewhere the boot splash is an issue on 3.2 so this might not be related to this bug).

I installed kernel-3.1.9-1 this morning and so far the problem has not arisen yet. I even put the 'rhgb quiet' back in. So it looks like the issue was resolved somewhere after kernel-3.1.6.

I will test this for a week or so. If no more crashes occur, I will close this bug.

I can confirm that this bug no longer exists using kernel-3.1.9-1. I am once again a very happy user!