http://bugs.winehq.org/show_bug.cgi?id=9581
Summary: winecfg crashes on freebsd early Product: Wine Version: 0.9.44. Platform: PC OS/Version: FreeBSD Status: UNCONFIRMED Severity: critical Priority: P2 Component: wine-kernel AssignedTo: wine-bugs@winehq.org ReportedBy: dmitry@karasik.eu.org
Clear setup, freshly compiled v0.44_1 from freebsd ports, no ~/.wine
$ winecfg wine: creating configuration directory '/home/dk/.wine'... err:seh:raise_exception Unhandled exception code c0000005 flags 0 addr 0x9c05230d wine: wineprefixcreate failed while creating '/home/dk/.wine'.
then says "starting debugger" and hangs.
http://bugs.winehq.org/show_bug.cgi?id=9581
Tijl Coosemans tijl@ulyssis.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |tijl@ulyssis.org
--- Comment #1 from Tijl Coosemans tijl@ulyssis.org 2007-09-05 13:45:15 --- What version of FreeBSD do you have?
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #2 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-06 00:06:44 --- (In reply to comment #1)
What version of FreeBSD do you have?
6.2 (sorry should've said that). If other info is needed (kernel conf, dmesg, recompile wine with some extra debug, whatever), tell me.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #3 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-06 09:33:37 --- update: found this in dmesg, might help?
Warning: pid 1619 used static ldt allocation. See the i386_set_ldt man page for more info
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #4 from Vitaliy Margolen vitaliy@kievinfo.com 2007-09-06 10:05:35 --- You will need to apply patches from http://wiki.freebsd.org/Wine if you want Wine to work.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #5 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-07 04:21:43 --- (In reply to comment #4)
You will need to apply patches from http://wiki.freebsd.org/Wine if you want Wine to work.
Patches applied, but wine still crashes identical results, crash and 'static ldt' message
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #6 from Vitaliy Margolen vitaliy@kievinfo.com 2007-09-07 08:05:38 --- It seems this problem is not solved yet on FreeBSD. This is work in progress and there are several people who working on it.
There is not much Wine can do about this issue - $fs register absolutely required for win apps to work.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #7 from Tijl Coosemans tijl@ulyssis.org 2007-09-07 14:49:29 --- (In reply to comment #5)
Patches applied, but wine still crashes identical results, crash and 'static ldt' message
The address in crash messages is now 0x7cxxxxxx instead of 0x9c05230d though, right?
The static allocation of LDT messages are normal. The warning was added to FreeBSD back in the day the threading libraries also used the LDT and so programs using it could interfere with that if they didn't use dynamic allocation. Those days are long gone though, so there isn't any problem anymore.
I've had a similar case like this where Wine always crashed. That was on FreeBSD 7-CURRENT with a multicore processor, so what CPU do you have?
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #8 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-07 15:42:26 --- No, actually the address is exactly the same:
wine: Unhandled page fault on write access to 0x00000000 at address 0x9c05230d (thread 0009), starting debugger... ^Cerr:seh:raise_exception Unhandled exception code c000013a flags 0 addr 0x9c117959
and the cpu is:
CPU: Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz (2392.28-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x6f6 Stepping = 6
Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,C Features2=0xe3bd<SSE3,RSVD2,MON,DS_CPL,VMX,EST,TM2,<b9>,CX16,<b14>,<b15>> AMD Features=0x20100000<NX,LM> AMD Features2=0x1<LAHF> Cores per package: 2
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #9 from Tijl Coosemans tijl@ulyssis.org 2007-09-07 16:33:31 --- (In reply to comment #8)
No, actually the address is exactly the same:
Hmm, in that case could you check you applied all patches on the wiki properly? There's one patch for your version of FreeBSD and one for the Wine port. You also need to patch, build and install FreeBSD first, and then Wine.
CPU: Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz (2392.28-MHz 686-class
Ok, in that case I assume it's the same problem as in CURRENT (race in mmap(2)), so I'll see if the fix for that can be merged to 6-STABLE.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #10 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-08 01:06:21 ---
Hmm, in that case could you check you applied all patches on the wiki properly? There's one patch for your version of FreeBSD and one for the Wine port. You also need to patch, build and install FreeBSD first, and then Wine.
Yes, that was done - patch applied, freebsd recompiled, then Wine recompiled. Actually I also found strange that results were the same, so I grep'd 'strings /kernel' for those new syscalls, like thr_kill2, and yes, these are present in the kernel.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #11 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-10 01:44:00 --- update: 'options SMP' also is not set in the kernel config
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #12 from Tijl Coosemans tijl@ulyssis.org 2007-09-10 12:34:07 --- Would it be possible for you to update to 6-STABLE and see if the problem has been fixed there or not? There have been a couple bug fixes to the virtual memory subsystem (c0000005 means memory access violation) in the last couple of weeks. Be sure to also use the 6-STABLE patch from the wiki then.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #13 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-10 13:14:04 --- Uhm.. I'm on 6.2-STABLE. Do you mean downgrade to 6.0?
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #14 from Tijl Coosemans tijl@ulyssis.org 2007-09-10 14:19:35 --- Oh, I thought you had 6.2-RELEASE. In that case, could you update to today's STABLE?
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #15 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-11 06:53:45 --- Today's 6.2-STABLE steps one bit further, but not far really. Winecfg worked(!), some other apps too, but far not all of them. The one among these is (native wine) notepad.exe:
dk@gromit ~> wine ~/.wine/drive_c/windows/notepad.exe err:wgl:X11DRV_WineGL_InitOpenglInfo couldn't initialize OpenGL, expect problems wine: Unhandled page fault on read access to 0x000000a1 at address 0x9caa2b3e (thread 0009), starting debugger... Unhandled exception: page fault on read access to 0x000000a1 in 32-bit code (0x9caa2b3e). file_set_error: Bad address file_set_error: Bad address Register dump: CS:0033 SS:003b DS:003b ES:003b FS:1007 GS:001b EIP:9caa2b3e ESP:0034ea78 EBP:0034ea78 EFLAGS:00010202( - 00 - -RI1) EAX:9ca45030 EBX:9ca449e8 ECX:00000009 EDX:9ca440e4 ESI:00000000 EDI:7bf17800 Stack dump: 0x0034ea78: 0034f0c8 9ca04d4b 00000009 9ca45030 0x0034ea88: 9ca45034 00000000 9c13d060 00000000 0x0034ea98: 00000000 00000010 9c11b46f 7bf28480 0x0034eaa8: 9bf085c9 9c0689a3 9c0689a3 9cb626c6 0x0034eab8: 9bf08559 9cb637f8 00000b1e 01000000 0x0034eac8: 9bf26018 9bf2e070 0034e6d0 0034eb14 0200: sel=1007 base=00112000 limit=00001fff 32-bit rw- Backtrace: =>1 0x9caa2b3e (0x0034ea78) 2 0x9ca04d4b (0x0034f0c8) 3 0x9ca1f1db (0x0034f278) 4 0x9ca2da7a (0x0034f2a8) 5 0x9c17c809 (0x0034f2c8) 6 0x9c17e852 (0x0034f348) 7 0x9c17ea71 (0x0034f378) 8 0x9c180ae7 (0x0034f398) 9 0x9c252134 in kernel32 (+0x42134) (0x0034f5f8) 10 0x9c252369 in kernel32 (+0x42369) (0x0034f618) 11 0x9c252437 in kernel32 (+0x42437) (0x0034f638) 12 0x9c252463 in kernel32 (+0x42463) (0x0034f658) 13 0x9c663298 (0x0034f7d8) 14 0x9c65db6a (0x0034fa78) 15 0x9c554752 (0x0034fb18) 16 0x9c554dfa (0x0034fb78) 17 0x9c556e60 (0x0034fc08) 18 0x9c5574bc (0x0034fca8) 19 0x9c55766b (0x0034fce8) 20 0x9c54d896 (0x0034fd18) 21 0x9c54d8e0 (0x0034fd28) 22 0x9c5aed08 (0x0034fda8) 23 0x9c5bfea6 (0x0034fdd8) 24 0x9c17c809 (0x0034fdf8) 25 0x9c17e852 (0x0034fe78) 26 0x9c17ea71 (0x0034fea8) 27 0x9c17eaf4 (0x0034fec8) 28 0x9c17eaf4 (0x0034fee8) 29 0x9c17eaf4 (0x0034ff08) 30 0x9c17eaf4 (0x0034ff28) 31 0x9c180d86 (0x0034ff68) 32 0x9c25ce33 in kernel32 (+0x4ce33) (0x0034ffe8) 0x9caa2b3e: movl 0x98(%ecx),%edx Modules: Module Address Debug info Name (12 modules) PE 9c150000-9c154000 Deferred ntdll PE 9c210000-9c27c000 Export kernel32 PE 9c310000-9c328000 Deferred notepad PE 9c340000-9c39f000 Deferred comdlg32 PE 9c3e0000-9c448000 Deferred shell32 PE 9c4d0000-9c4d6000 Deferred shlwapi PE 9c530000-9c541000 Deferred user32 PE 9c640000-9c644000 Deferred gdi32 PE 9c6d0000-9c6d4000 Deferred advapi32 PE 9c700000-9c711000 Deferred comctl32 PE 9c7b0000-9c7b6000 Deferred winspool PE 9c7e0000-9c820000 Deferred msvcrt Threads: process tid prio (all id:s are in hex) 00000008 (D) c:\windows\system32\notepad.exe 00000009 0 <==
I don't know if this backtrace makes any sense, so if you tell me how to recompile wine with debug info I'll do that and repost.
http://bugs.winehq.org/show_bug.cgi?id=9581
Dmitry Karasik dmitry@karasik.eu.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |FIXED
--- Comment #16 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-11 07:14:32 ---
err:wgl:X11DRV_WineGL_InitOpenglInfo couldn't initialize OpenGL, expect problems
after stumbling on this, I've enabled GLX extension on my X (which thanks to old uninstalled nvidia drivers was not loading), and the problems disappeared, both on notepad.exe and other applications.
Tijl Coosemans, I'd very much like to thank you for your time and patience! I guess I'll close the ticket.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #17 from Dmitry Timoshkov dmitry@codeweavers.com 2007-09-11 07:23:48 ---
err:wgl:X11DRV_WineGL_InitOpenglInfo couldn't initialize OpenGL, expect problems
after stumbling on this, I've enabled GLX extension on my X (which thanks to old uninstalled nvidia drivers was not loading), and the problems disappeared, both on notepad.exe and other applications.
It's still worrying that simple applications like winecfg and notepad that don't use OpenGL at all crash due to problem above.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #18 from Dmitry Karasik dmitry@karasik.eu.org 2007-09-11 07:27:09 ---
It's still worrying that simple applications like winecfg and notepad that don't use OpenGL at all crash due to problem above.
My proposition is still valid - if someone shall teach me how, I'm willing to disable GLX again and dig myself into the problems, but with these purely binary backtracks I have no chance of doing so.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #19 from Dmitry Timoshkov dmitry@codeweavers.com 2007-09-11 07:50:50 --- What do you mean by "purely binary backtracks"?
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #20 from Dmitry Timoshkov dmitry@codeweavers.com 2007-09-11 07:53:18 --- Ah, if that's the back trace debugger prints, I have no idea, perhaps winedbg problem under FreeBSD.
http://bugs.winehq.org/show_bug.cgi?id=9581
--- Comment #21 from Tijl Coosemans tijl@ulyssis.org 2007-09-11 08:21:12 --- Ah, good to hear the problem has been fixed. About the backtrace, the FreeBSD ports system strips binaries by default. If you want a more useful backtrace you can build and install the Wine port using "make WITH_DEBUG=yes install".
http://bugs.winehq.org/show_bug.cgi?id=9581
Dan Kegel dank@kegel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #22 from Dan Kegel dank@kegel.com 2008-01-28 05:19:39 --- Closing all RESOLVED FIXED bugs more than three months old.
https://bugs.winehq.org/show_bug.cgi?id=9581
Manny20444@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |Manny20444@gmail.com
-
wine-bugs＠winehq.org