Reproduced something at least similar. Terminator now vomits this sort of thing continually for me:
4 [main] bash 4984 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AD08 6102792B (0022AD08, 00000000, 00000000, 00000000)
0022AFF8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C028 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C058 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C078 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C0A8 610678AD (0022C0D4, 0022C0B0, 00000000, 00000000)
0022C228 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 3320 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AD08 6102792B (0022AD08, 00000000, 00000000, 00000000)
0022AFF8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C028 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C058 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C078 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C0A8 610678AD (0022C0D4, 0022C0B0, 00000000, 00000000)
0022C228 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
The mirror site I use is slightly out-of-date, meaning that it gave me all of the recent packages but not the base files update first. That worked fine. Then I switched mirrors, got just the base file update and boom. Diff was:
<<<YourSubjectHere>>> Terminator problem (1635/3246/1.6.0_24-b07/Windows XP 5.1 Cygwin 1.7.7/x86 x1)
<<<YourSubjectHere>>> Terminator problem (1635/3246/1.6.0_24-b07/Windows XP 5.1 Cygwin 1.7.8/x86 x1)
I am not seeing a smoking gun at http://cygwin.com/ml/cygwin-announce/2011-03/msg00000.html.
6 The handle is invalid. ERROR_INVALID_HANDLE
Googling for the errors didn't find me anything. Nothing's jumping out of http://www.cygwin.com/ml/cygwin/2011-03/ in the few hours since 1.7.8 was released. Wow, we're on the leading edge this time.
--
You received this message because you are subscribed to the Google Groups
"terminator-users" group.
To post to this group, send email to terminat...@googlegroups.com.
To unsubscribe from this group, send email to
terminator-use...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/terminator-users?hl=en.
I thought I'd try building on Cygwin 1.7.8f. Three-underscores xpg_strerror_r seems to be defined:
$ for file in /lib/lib*.a; do echo $file; objdump --syms $file | grep " "___xpg_strerror_r; done
...
/lib/libc.a
[ 2](sec 1)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x00000000 ___xpg_strerror_r
...
/lib/libcygwin.a
[ 2](sec 1)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x00000000 ___xpg_strerror_r
...
/lib/libg.a
[ 2](sec 1)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x00000000 ___xpg_strerror_r
But two-underscores xpg_strerror_r does not, and that's apparently what's needed to compile now:
$ gcc strerror.c
/cygdrive/c/DOCUME~1/martind/LOCALS~1/Temp/ccypDt0q.o:strerror.c:(.text+0x42): undefined reference to `__xpg_strerror_r'
collect2: ld returned 1 exit status
$ cat strerror.c
#include <string.h>
int main() {
return strerror_r(0, 0, 0);
}
$ gcc strerror.c -c
$ objdump --syms strerror.o | grep xpg_strerror_r
[ 12](sec 0)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x00000000 __xpg_strerror_r
$
A cvs update -dP gets me a last mention of strerror in:
2011-02-10 Eric Blake <ebl...@redhat.com>
* errno.cc (includes): Avoid compilation failure if <string.h>
settles on wrong strerror_r signature.
* errno.cc (__xpg_strerror_r): New function.
(strerror_r): Update to copy newlib's fixes.
(strerror): Set errno on failure.
(_sys_errlist): Cause EINVAL failure for reserved values.
* cygwin.din: Export new function.
* include/cygwin/version.h (CYGWIN_VERSION_API_MINOR): Combine
this into minor 236.
I imagine someone with posting rights on the Cygwin mailing list will report it soon enough.
I wouldn't like to claim this is the right fix but it lets that trivial example link:
$ gcc strerror.c
$ diff -u /usr/include/string.h.orig /usr/include/string.h
--- /usr/include/string.h.orig 2011-03-02 10:49:29.510836800 -0800
+++ /usr/include/string.h 2011-03-02 10:50:04.324450800 -0800
@@ -74,7 +74,7 @@
char *_EXFUN(strerror_r,(int, char *, size_t));
#else
# ifdef __GNUC__
-int _EXFUN(strerror_r,(int, char *, size_t)) __asm__ ("__xpg_strerror_r");
+int _EXFUN(strerror_r,(int, char *, size_t)) __asm__ ("___xpg_strerror_r");
# else
int _EXFUN(__xpg_strerror_r,(int, char *, size_t));
# define strerror_r __xpg_strerror_r
$
My thoughts too.
With that small hack, our stuff compiles again, but it still goes wrong at runtime in pretty much the same way.
I went ahead and posted on the cygwin mailing list, stating that an
unspecified developer/maintainer of terminator (not gnome terminator)
felt confident enough to propose a patch to solve the problem in
cygwin and posted that patch -- hope that's ok...
I really like terminator :-)
That and cygwin makes Windoze actually useful for something other than
just generating heat :-)
Thanks,
Ken Wolcott
Here's the response from Corinna on the cygwin mailing list:
Thanks for the report. I just fixed that in CVS. The problem here
is that we have to take several targets into account, some of them
using a labe lprefix, some of them not. I applied a patch which
does that:
int _EXFUN(strerror_r,(int, char *, size_t)) __asm__ (__ASMNAME
("__xpg_strerror_r"));
The __ASMNAME macro is now defined in cdefs.h. I tested to build
a testcase which calls the POSIX version of strerror_r and it now
works fine. Well, there was a small bug in __xpg_strerror_r, too,
but that should be fixed in CVS as well. I'll create a new developer
snapshot later today.
Thanks again,
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
On Wed, Mar 2, 2011 at 18:49, Martin Dorey <mdo...@bluearc.com> wrote:
...
The DLL from the 2011-03-03 16:17 GMT one, which doesn't include this header file fix, doesn't fix the runtime problem.
A header file problem, with a pathological example and a patch to fix, albeit one that wasn't fit to apply... that's one thing. A big hairy application provoking others to crash... that's not going to be so well-received. I fear we're going to have to build from CVS until we can pin it on a particular change.
The setup tool will probably let you downgrade to 1.7.7 in the meantime.
It doesn't seem to have updated to include the header file change and, even when that is fixed, I can't see it fixing the runtime issue.
> I really don't have time to back down to 1.7.7 either.
I did it this afternoon. The work of seconds. You only have to downgrade the "Base", "cygwin" component. A few clicks and that had done it. Mind, it needs cygwin1.dll not to be in use, but I have a couple of batch files to stop and restart the Cygwin services to cope with that. Alright, a minute, maybe :).
Still...
> I started up mintty
... fair enough.
It seems that the problem has vanished upon subsequent (and very
recent) update of Cygwin.
uname.exe -a
CYGWIN_NT-5.1 KWOLCOTT-T61 1.7.8(0.236/5/3) 2011-03-01 09:36 i686 Cygwin
So I really don't know if there was a fix that really solved the
problem and/or whether I had some unusual corruption that the update
cleared...
Thanks,
Ken Wolcott
> CYGWIN_NT-5.1 KWOLCOTT-T61 1.7.8(0.236/5/3) 2011-03-01 09:36 i686 Cygwin
It's still broken for me, with the same version. I don't remember it terminating in a prompt as quickly as it does below, but isn't this the same version as you originally reported this problem with?
Hmm, in http://cygwin.com/ml/cygwin/2011-03/msg00744.html, you have a snapshot installed. It still dies for me with:
martind@vm-martind ~
$ uname -a
CYGWIN_NT-5.1 vm-martind 1.7.9s(0.236/5/3) 20110327 15:26:47 i686 Cygwin
martind@vm-martind ~
$
> whether I had some unusual corruption that the update
> cleared...
If it's corruption, then it's not that unusual, because I have it too! I know, from earlier in this thread, that your mirror is ahead of mine but I've been away for a few days, so surely by now I must have the same version as you of everything like bash. Can you think of anything else you might have done to make the problem go away? Perhaps I should try a reboot. I haven't, and it is Windows, after all.
The complete log that follows is with stock 1.7.8:
12 [main] bash 4976 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AC78 6102792B (0022AC78, 00000000, 00000000, 00000000)
0022AF68 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022BF98 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022BFC8 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022BFE8 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C018 610678AD (0022C044, 0022C020, 00000000, 00000000)
0022C198 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
8 [main] bash 3144 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AC78 6102792B (0022AC78, 00000000, 00000000, 00000000)
0022AF68 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022BF98 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022BFC8 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022BFE8 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C018 610678AD (0022C044, 0022C020, 00000000, 00000000)
0022C198 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 1196 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AC78 6102792B (0022AC78, 00000000, 00000000, 00000000)
0022AF68 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022BF98 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022BFC8 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022BFE8 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C018 610678AD (0022C044, 0022C020, 00000000, 00000000)
0022C198 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 5428 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AC78 6102792B (0022AC78, 00000000, 00000000, 00000000)
0022AF68 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022BF98 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022BFC8 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022BFE8 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C018 610678AD (0022C044, 0022C020, 00000000, 00000000)
0022C198 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 3488 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AC78 6102792B (0022AC78, 00000000, 00000000, 00000000)
0022AF68 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022BF98 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022BFC8 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022BFE8 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C018 610678AD (0022C044, 0022C020, 00000000, 00000000)
0022C198 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
3 [main] bash 5844 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022AC78 6102792B (0022AC78, 00000000, 00000000, 00000000)
0022AF68 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022BF98 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022BFC8 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022BFE8 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C018 610678AD (0022C044, 0022C020, 00000000, 00000000)
0022C198 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
3 [main] bash 4708 fork: child -1 - died waiting for longjmp before initialization, retry 0, exit code 0x600, errno 11
bash: fork: retry: Resource temporarily unavailable
5 [main] bash 4736 fork: child -1 - died waiting for longjmp before initialization, retry 0, exit code 0xC0000005, errno 11
bash: fork: retry: Resource temporarily unavailable
4 [main] bash 4632 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022ADE8 6102792B (0022ADE8, 00000000, 00000000, 00000000)
0022B0D8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C108 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C138 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C158 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C188 610678AD (0022C1B4, 0022C190, 00000000, 00000000)
0022C308 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 2792 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022ADE8 6102792B (0022ADE8, 00000000, 00000000, 00000000)
0022B0D8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C108 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C138 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C158 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C188 610678AD (0022C1B4, 0022C190, 00000000, 00000000)
0022C308 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 5248 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022ADE8 6102792B (0022ADE8, 00000000, 00000000, 00000000)
0022B0D8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C108 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C138 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C158 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C188 610678AD (0022C1B4, 0022C190, 00000000, 00000000)
0022C308 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 5392 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022ADE8 6102792B (0022ADE8, 00000000, 00000000, 00000000)
0022B0D8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C108 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C138 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C158 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C188 610678AD (0022C1B4, 0022C190, 00000000, 00000000)
0022C308 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 720 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022ADE8 6102792B (0022ADE8, 00000000, 00000000, 00000000)
0022B0D8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C108 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C138 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C158 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C188 610678AD (0022C1B4, 0022C190, 00000000, 00000000)
0022C308 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
4 [main] bash 696 C:\cygwin-1.7\bin\bash.exe: *** fatal error - SetEvent failed
Stack trace:
Frame Function Args
0022ADE8 6102792B (0022ADE8, 00000000, 00000000, 00000000)
0022B0D8 6102792B (6117DC60, 00008000, 00000000, 6117F977)
0022C108 61004F3B (611C3CE0, 00000000, 00000000, 00000000)
0022C138 610C6CA2 (00020938, 00000000, 00000000, 00000000)
0022C158 6106773D (611B4AFD, 00000001, 00000000, 00000000)
0022C188 610678AD (0022C1B4, 0022C190, 00000000, 00000000)
0022C308 61068854 (00000000, 00000000, 00000000, 00000000)
End of stack trace
7 [main] bash 5204 fork: child -1 - died waiting for longjmp before initialization, retry 0, exit code 0x600, errno 11
bash: fork: retry: Resource temporarily unavailable
martind@vm-martind ~
$ uname -a
CYGWIN_NT-5.1 vm-martind 1.7.8(0.236/5/3) 2011-03-01 09:36 i686 Cygwin
martind@vm-martind ~
$
I just am 100% puzzled. It seems to fail earlier and I don't know
why and now it works and I don't know why. The bottom line (over used
!) is that it works now.
Thanks,
Ken Wolcott
> ...
>
> [Message clipped]