Re: EOL for Windows 95/98/Me
On Jul 24 18:59, Corinna Vinschen wrote: Now that Microsoft has finally dropped support for Windows 98 and Me, we're going to drop 9x support as well. What we're planning to do is this: The complete net distribution gets copied to a new place. This new place is the distro kept for people running 9x. There is no further development in this distribution. Maintainers may decide whether or not they apply fixes to the packages in the 9x distro, or keep it up to date at all. The normal net distribution will continue to be the normal distro. It might work on 9x, but there's no guarantee at all that it will continue to do so. The setup tool (hello setup developers?) should either be split into two versions, one for 9x, one for NT. Or the setup tool should choose the download path depending on the OS it's running on. Or something completely different. Comments? Ideas? Btw., it just occured to me that I'd rather get rid of the 9x stuff in the 1.7.0 DLL entirely. This would have visible advantages. - The code size of the DLL would shrink by a good amount. - The autoloading of functions could be reduced to the functions not available on all NT versions. This would reduce the autoload overhead by about 90%. - The code complexity would be reduced enormously by stripping off at least 50% of the `if (wincap.foo ()) tests. This would also have some positive effects on the performance. - Long 32K pathname support doesn't exist in 9x. So, when we switch over to using the unicode functions for pathnames, we would have a lot of avoidable hassle to keep 9x running at all. You're all convinced, right? Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat
RE: EOL for Windows 95/98/Me
On 07 February 2007 10:18, Corinna Vinschen wrote: On Jul 24 18:59, Corinna Vinschen wrote: Btw., it just occured to me that I'd rather get rid of the 9x stuff in the 1.7.0 DLL entirely. This would have visible advantages. - The code size of the DLL would shrink by a good amount. - The autoloading of functions could be reduced to the functions not available on all NT versions. This would reduce the autoload overhead by about 90%. - The code complexity would be reduced enormously by stripping off at least 50% of the `if (wincap.foo ()) tests. This would also have some positive effects on the performance. - Long 32K pathname support doesn't exist in 9x. So, when we switch over to using the unicode functions for pathnames, we would have a lot of avoidable hassle to keep 9x running at all. You're all convinced, right? Hell yeah! Let's have a mass-delete-fest! We should tag the repository beforehand, just in case some retro-enthusiasts feel like keeping 1.5.x alive on a branch and keeping it hobbling along on '9x for a while longer. cheers, DaveK -- Can't think of a witty .sigline today
Re: EOL for Windows 95/98/Me
On Feb 7 10:45, Dave Korn wrote: On 07 February 2007 10:18, Corinna Vinschen wrote: Btw., it just occured to me that I'd rather get rid of the 9x stuff in the 1.7.0 DLL entirely. This would have visible advantages. - The code size of the DLL would shrink by a good amount. - The autoloading of functions could be reduced to the functions not available on all NT versions. This would reduce the autoload overhead by about 90%. - The code complexity would be reduced enormously by stripping off at least 50% of the `if (wincap.foo ()) tests. This would also have some positive effects on the performance. - Long 32K pathname support doesn't exist in 9x. So, when we switch over to using the unicode functions for pathnames, we would have a lot of avoidable hassle to keep 9x running at all. You're all convinced, right? Hell yeah! Let's have a mass-delete-fest! We should tag the repository beforehand, just in case some retro-enthusiasts feel like keeping 1.5.x alive on a branch and keeping it hobbling along on '9x for a while longer. We have a branch for 1.5.x already for 9 months. Guess where the recent 1.5.x releases came from? ;) Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat
RE: EOL for Windows 95/98/Me
On 07 February 2007 10:58, Corinna Vinschen wrote: We have a branch for 1.5.x already for 9 months. Guess where the recent 1.5.x releases came from? ;) Ah, didn't notice. Is 1.7 going to branch off 1.5, or are 1.5 changes going to be merged back to mainline and 1.7 series kick off from there? cheers, DaveK -- Can't think of a witty .sigline today
Re: EOL for Windows 95/98/Me
On Feb 7 11:03, Dave Korn wrote: On 07 February 2007 10:58, Corinna Vinschen wrote: We have a branch for 1.5.x already for 9 months. Guess where the recent 1.5.x releases came from? ;) Ah, didn't notice. Is 1.7 going to branch off 1.5, or are 1.5 changes going to be merged back to mainline and 1.7 series kick off from there? Er... Dave? Hello? There's a 1.5 *branch*. Where, do you guess, is 1.7 developed in? The trunk maybe? The trunk is 1.7 since 2006-07-25. Didn't you notice that the snapshots already diverge from 1.5 for a couple of months? I actually thought you're one of the guys following the development... Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat
RE: EOL for Windows 95/98/Me
On 07 February 2007 11:14, Corinna Vinschen wrote: On Feb 7 11:03, Dave Korn wrote: On 07 February 2007 10:58, Corinna Vinschen wrote: We have a branch for 1.5.x already for 9 months. Guess where the recent 1.5.x releases came from? ;) Ah, didn't notice. Is 1.7 going to branch off 1.5, or are 1.5 changes going to be merged back to mainline and 1.7 series kick off from there? Er... Dave? Hello? There's a 1.5 *branch*. Where, do you guess, is 1.7 developed in? The trunk maybe? The trunk is 1.7 since 2006-07-25. Didn't you notice that the snapshots already diverge from 1.5 for a couple of months? I actually thought you're one of the guys following the development... Not for the last few months, no - CES took up 100% of my time (spare and otherwise) for quite a long while this time round I'll go read some change logs. cheers, DaveK -- Can't think of a witty .sigline today
setup.exe: No choice of mount point type?
Guys, I just had an in-shower inspiration which I have to share immediately. Given that we still support textmode mounts for historical reasons, but don't like people actually using them, what if setup.exe would simply not give the choice? We can't keep users from changing the mount points after an installation, but who demands from us to leave the user the choice in the base installation process? Just removing the button group from the options dialog would be sufficient, me thinks... Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat
Re: setup.exe: No choice of mount point type?
On Wed, Feb 07, 2007 at 03:03:00PM +0100, Corinna Vinschen wrote: Guys, I just had an in-shower inspiration which I have to share immediately. Given that we still support textmode mounts for historical reasons, but don't like people actually using them, what if setup.exe would simply not give the choice? We can't keep users from changing the mount points after an installation, but who demands from us to leave the user the choice in the base installation process? Just removing the button group from the options dialog would be sufficient, me thinks... I think we're sending a mixed message about textmode mounts these days when it is one of the solutions to make bash understand CRLF scripts. And, as much as I like not offering people this option, I think we'll be flooded with user requests if we delete it. OTOH, maybe including something like: 'See man mount for details about how to change the characteristics of the cygwin root or to mount additional directories.' would be sufficient. cgf
Re: setup.exe: No choice of mount point type?
On Feb 7 09:32, Christopher Faylor wrote: On Wed, Feb 07, 2007 at 03:03:00PM +0100, Corinna Vinschen wrote: Given that we still support textmode mounts for historical reasons, but don't like people actually using them, what if setup.exe would simply not give the choice? We can't keep users from changing the mount points after an installation, but who demands from us to leave the user the choice in the base installation process? Just removing the button group from the options dialog would be sufficient, me thinks... I think we're sending a mixed message about textmode mounts these days when it is one of the solutions to make bash understand CRLF scripts. Indeed. And, as much as I like not offering people this option, I think we'll be flooded with user requests if we delete it. More than usual? ;) OTOH, maybe including something like: 'See man mount for details about how to change the characteristics of the cygwin root or to mount additional directories.' would be sufficient. That sounds good to me. Just keep a hint which mentions there's something mysterious you can do with a even more mysterious tool called mount. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat
Re: ssh -x
Scott Fordin schrieb: Holger Krull wrote: /usr/X11R6/bin/xhost +linux_machine_name No you don't have to, if you use ssh to tunnel the X11 connection. (It would be localhost anyway and not the linux_machine_name). Try to understand the difference between tunneling and a remote connection. I understand what you're saying. All I know is that for some applications, like Emacs and Arbortext Epic, it seems I have to explicitly add the remote machine to my local X access list for the applications to display on my local machine. Moreover, for Epic to work, I have to explicitly set the DISPLAY variable on the remote machine. Dunno. It works for me, but YMMV. If you do this, you are not using the forwarding mechanism of ssh. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
problem installing Cygwin/X - usr\X11R6\bin\X.exe - Symbolic link points to missing file
hi there, I'm trying to install cygwin (actually Cygwin/X), I downloaded the latested cygwin installer, and I'm trying to install X11 packages, some files are installed but some files are surelly missing (for instance, I select to install X-startup-scripts-1.0.11-1.tar.bz2, but after the installation I still don't have a /usr/X11R6/bin directory)... well, I decided to download the packages locallly and open them to see if something was wrong, and I found that in the file xorg-x11-bin-6.8.99.901-1.tar.bz2, there is an entry called X.exe with 0 bytes... actually I think this is a symbolic link because when I try to extract it, I get an error like this: ! C:\cyginstall\http%3a%2f%2fmirror.anl.gov%2fpub%2fcygwin%2f\release\X11\xorg-x11-bin\xorg-x11-bin-6.8.99.901-1.tar.bz2: Cannot open XWin.exe (usr\X11R6\bin\X.exe -- XWin.exe) ! C:\cyginstall\http%3a%2f%2fmirror.anl.gov%2fpub%2fcygwin%2f\release\X11\xorg-x11-bin\xorg- x11-bin-6.8.99.901-1.tar.bz2: Symbolic link points to missing file any idea of what's wrong? best regards, Fabio. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
late bash CR/LF functionality change problem
I recently updated to the latest bash rev. only to find that most of my scripts do not work anymore. I browsed through the FAQ and mailing lists and found many people having problems with the changed behaviour of treating \r literally. I do not want to argue what is right or wrong though. Fact is that bash until fall last year functioned in a certain way. Many people wrote scripts using this functionality. Now bash functions differently which breaks a lot of jobs. Especially batch processes which did their job silently in the background all over the sudden cause problems. If a change is implemented which has such a big impact it should NEVER be implemented as the default behaviour. Instead this new functionality should be activated by an env setting, start parameter or the like so that the functionality stays backward compatible. What made things even worse is that there is no notification about the changed behaviour. It is just the symptoms and everyone experiencing the problem must then dig to find out the cause. Is there an quick and easy solution to the problem out yet? Peter -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: late bash CR/LF functionality change problem
peter.kielbasiewicz skrev: I recently updated to the latest bash rev. only to find that most of my scripts do not work anymore. I browsed through the FAQ and mailing lists and found many people having problems with the changed behaviour of treating \r literally. I do not want to argue what is right or wrong though. Fact is that bash until fall last year functioned in a certain way. Many people wrote scripts using this functionality. Now bash functions differently which breaks a lot of jobs. Especially batch processes which did their job silently in the background all over the sudden cause problems. If a change is implemented which has such a big impact it should NEVER be implemented as the default behaviour. Instead this new functionality should be activated by an env setting, start parameter or the like so that the functionality stays backward compatible. What made things even worse is that there is no notification about the changed behaviour. It is just the symptoms and everyone experiencing the problem must then dig to find out the cause. Is there an quick and easy solution to the problem out yet? Peter Stop beating a dead horse and use dos2unix on your scripts already. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Bugzilla under Cygwin?
Hi, I'm playing with the thought to install Bugzilla under Cygwin. Did anyone tried that before? Are there any experiences? Any recommendations? Most of the prerequisites are available under Cygwin (Apache, Perl,...), I'm only missing MySQL. It might be possible to combine Bugzilla with the Windows version of MySQL. Thanks for any suggestions! Matthias -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
No manual entry for tar?
Hi, It is normal that there is no man page for tar, or is this a package error? $ man tar No manual entry for tar Regards, Frank -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: No manual entry for tar?
Frank Fesevur schrieb: Hi, It is normal that there is no man page for tar, or is this a package error? $ man tar No manual entry for tar Regards, Frank -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ Just use $ info tar for information about tar. I can reproduce this and I am pretty sure, I have *all* the packages installed. Probably wrong packaging... SE -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
On Feb 6 16:38, Matthew Woehlke wrote: ...anyone have any guesses why rename() can't rename files in the same directory, if the directory is an NFS mount? This causes 'ar' to fail with the error: ar: unable to rename 'libcloudobj.a' reason: Invalid cross-device link I see this in strace with both 'ar' and 'mv': 46 371094 [main] ar 2588 rename: -1 = rename (h:\...path...\src\objects\stWBl4pj, h:\...path...\src\objects\libcloudobj.a) This seems to happen when the file already exists, but rename() only fails on NFS mounts (not local drives). Is this a problem with 'ar', or Cygwin's rename()? Since you're one of the few running NFS, you should also debug the situation further. Apparently the Cygwin rename function fails, but from your one-liner strace snippet there's no hint at all what happens. The error should have been mentioned a couple of lines before. You should also mention if you use a release version of Cygwin or a developer snapshot. You know http://cygwin.com/problems.html, don't you? Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
On Feb 7 11:55, Corinna Vinschen wrote: On Feb 6 16:38, Matthew Woehlke wrote: ...anyone have any guesses why rename() can't rename files in the same directory, if the directory is an NFS mount? This causes 'ar' to fail with the error: ar: unable to rename 'libcloudobj.a' reason: Invalid cross-device link I see this in strace with both 'ar' and 'mv': 46 371094 [main] ar 2588 rename: -1 = rename (h:\...path...\src\objects\stWBl4pj, h:\...path...\src\objects\libcloudobj.a) This seems to happen when the file already exists, but rename() only fails on NFS mounts (not local drives). Is this a problem with 'ar', or Cygwin's rename()? Since you're one of the few running NFS, you should also debug the situation further. Apparently the Cygwin rename function fails, but from your one-liner strace snippet there's no hint at all what happens. The error should have been mentioned a couple of lines before. You should also mention if you use a release version of Cygwin or a developer snapshot. You know http://cygwin.com/problems.html, don't you? I just installed SFU NFS for testing and, here it comes: I can not reproduce rename failing on an NFS share. Maybe it's something with your permissions. http://cygwin.com/acronyms/#SHTDI2(*) Corinna (*) I know, this is an invalid URL. What I mean is Somebody has to debug it, but that's not covered by SHTDI. -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: late bash CR/LF functionality change problem
peter.kielbasiewicz wrote: I recently updated to the latest bash rev. only to find that most of my scripts do not work anymore. I browsed through the FAQ and mailing lists and found many people having problems with the changed behaviour of treating \r literally. I do not want to argue what is right or wrong though. Fact is that bash until fall last year functioned in a certain way. Many people wrote scripts using this functionality. Now bash functions differently which breaks a lot of jobs. Especially batch processes which did their job silently in the background all over the sudden cause problems. If a change is implemented which has such a big impact it should NEVER be implemented as the default behaviour. Instead this new functionality should be activated by an env setting, start parameter or the like so that the functionality stays backward compatible. What made things even worse is that there is no notification about the changed behaviour. It is just the symptoms and everyone experiencing the problem must then dig to find out the cause. The change in bash's behaviour has been announced many times in bash package release announcements which are sent into this mailing list. There even was a transition period when it was only part of experimental bash package. If you have read any of the announcements you would know about it. Reading this mailing list is important. It is the only information channel from Cygwin developers and package maintainers to you, the users. Is there an quick and easy solution to the problem out yet? Use the d2u utility on your scripts. Peter -- Vaclav Haisman signature.asc Description: OpenPGP digital signature
Re: No manual entry for tar?
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Frank Fesevur on 2/7/2007 3:22 AM:
Hi,
It is normal that there is no man page for tar, or is this a package error?
Normal. The upstream tar maintainer doesn't go to the efforts of creating
one in his tarball, so I don't put forth any additional effort to make one
myself. The GNU Coding Standards recommend the info format ('info tar')
over the man format, and although many GNU packages provide both, there is
no obligation that they must. Also try using 'tar --help'.
- --
Don't work too hard, make some time for fun as well!
Eric Blake [EMAIL PROTECTED]
volunteer cygwin tar maintainer
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFycyM84KuGfSFAYARAtxYAKCNWIad5z8NL5KHTLygm9NNOYmMiwCeKesH
00Bgs9xJqvrTCF+2ccyVepo=
=8Xue
-END PGP SIGNATURE-
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Problem with OpenGL support under python: ctypes cannot find GL lib(?)
Hi, I have a python specific question, but the problem only seems to occur under cygwin, so here I am. In trying to get my python OpenGL application to run, I installed the PyOpenGL package, version 3.0.0a5 for python 2.4, and ctypes 1.0.1. Now whenever I try to run it, the following happens: Traceback (most recent call last): File GLvis.py, line 8, in ? from OpenGL.GL import * File /usr/lib/python2.4/site-packages/PyOpenGL-3.0.0a5-py2.4.egg/OpenGL/GL/__init__.py, line 2, in ? from OpenGL.raw.GL import * File /usr/lib/python2.4/site-packages/PyOpenGL-3.0.0a5-py2.4.egg/OpenGL/raw/GL/__init__.py, line 6, in ? from OpenGL.raw.GL.constants import * File /usr/lib/python2.4/site-packages/PyOpenGL-3.0.0a5-py2.4.egg/OpenGL/raw/GL/constants.py, line 7, in ? from OpenGL import platform, arrays File /usr/lib/python2.4/site-packages/PyOpenGL-3.0.0a5-py2.4.egg/OpenGL/platform/__init__.py, line 24, in ? from OpenGL.platform.glx import * File /usr/lib/python2.4/site-packages/PyOpenGL-3.0.0a5-py2.4.egg/OpenGL/platform/glx.py, line 29, in ? mode=ctypes.RTLD_GLOBAL File /usr/lib/python2.4/site-packages/PyOpenGL-3.0.0a5-py2.4.egg/OpenGL/platform/ctypesloader.py, line 37, in loadLibrary return dllType( name, mode ) File /usr/lib/python2.4/site-packages/ctypes/__init__.py, line 312, in __init__ self._handle = _dlopen(self._name, mode) OSError: No such file or directory When this happens, self._name is GL and mode is 4. It would seem as though somehow the GL library is not registered(?) with python? I'm at a loss here - can anybody give me a hint? Has there been any successful attempt at running PyOpenGL under cygwin? Thanks, Martin Cygwin Configuration Diagnostics Current System Time: Wed Feb 07 15:21:29 2007 Windows XP Professional Ver 5.1 Build 2600 Service Pack 2 Path: . ~\sbin ~\bin C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin C:\cygwin\usr\X11R6\bin c:\Programme\MiKTeX 2.5\miktex\bin c:\WINDOWS\system32 c:\WINDOWS Output from C:\cygwin\bin\id.exe (nontsec) UID: 1004(felder)GID: 513(Kein) 0(root) 513(Kein)544(Administratoren) 545(Benutzer) Output from C:\cygwin\bin\id.exe (ntsec) UID: 1004(felder)GID: 513(Kein) 0(root) 513(Kein)544(Administratoren) 545(Benutzer) SysDir: C:\WINDOWS\system32 WinDir: C:\WINDOWS PWD = '/home/felder/python/Neural' HOME = '/home/felder' HOMEPATH = '\Dokumente und Einstellungen\felder' APPDATA = 'C:\Dokumente und Einstellungen\felder\Anwendungsdaten' XKEYSYMDB = '/usr/X11R6/lib/X11/XKeysymDB' PROCESSOR_IDENTIFIER = 'x86 Family 6 Model 15 Stepping 6, GenuineIntel' TERM = 'xterm' WINDIR = 'C:\WINDOWS' WINDOWID = '2097166' OLDPWD = '/home/felder/python' USERDOMAIN = 'LAPKNOLL52' ALLUSERSPROFILE = 'C:\Dokumente und Einstellungen\All Users' OS = 'Windows_NT' XAPPLRESDIR = '/usr/X11R6/lib/X11/app-defaults' XCMSDB = '/usr/X11R6/lib/X11/Xcms.txt' XTERM_SHELL = '/bin/bash' COMMONPROGRAMFILES = 'C:\Programme\Gemeinsame Dateien' TEMP = '/cygdrive/c/DOKUME~1/felder/LOKALE~1/Temp' XNLSPATH = '/usr/X11R6/lib/X11/locale' TERMCAP = 'xterm-r6|xterm|xterm X11R6 version:am:km:mi:ms:xn:co#120:it#8:li#24:AL=\E[%dL:DC=\E[%dP:DL=\E[%dM:DO=\E[%dB:LE=\E[%dD:RI=\E[%dC:UP=\E[%dA:ae=^O:al=\E[L:as=^N:bl=^G:cd=\E[J:ce=\E[K:cl=\E[H\E[2J:cm=\E[%i%d;%dH:cr=^M:cs=\E[%i%d;%dr:ct=\E[3g:dc=\E[P:dl=\E[M:do=^J:ei=\E[4l:ho=\E[H:im=\E[4h:is=\E7\E[r\E[m\E[?7h\E[?1;3;4;6l\E[4l\E8\E:k1=\EOP:k2=\EOQ:k3=\EOR:k4=\EOS:k5=\E[15~:k6=\E[17~:k7=\E[18~:k8=\E[19~:k9=\E[20~:kD=\E[3~:kI=\E[2~:kN=\E[6~:kP=\E[5~:kd=\EOB:ke=\E[?1l\E:kh=\E[1~:kl=\EOD:kr=\EOC:ks=\E[?1h\E=:ku=\EOA:le=^H:md=\E[1m:me=\E[m:mr=\E[7m:nd=\E[C:rc=\E8:sc=\E7:se=\E[m:sf=^J:so=\E[7m:sr=\EM:ta=^I:te=\E[2J\E[?47l\E8:ti=\E7\E[?47h:ue=\E[m:up=\E[A:us=\E[4m:kb=\010:' USERNAME = 'felder' GASCRP = '/cygdrive/c/progs/PCGrADS/lib' PROCESSOR_LEVEL = '6' FP_NO_HOST_CHECK = 'NO' SYSTEMDRIVE = 'C:' CYGWIN_ROOT = '\cygwin' EDITOR = '/usr/X11R6/bin/nedit' CLIENTNAME = 'Console' USERPROFILE = 'C:\Dokumente und Einstellungen\felder' LOGONSERVER = '\\LAPKNOLL52' !C: = 'C:\cygwin\usr\X11R6\bin' PROCESSOR_ARCHITECTURE = 'x86' HISTIGNORE = '[ ]*::bg:fg' XTERM_VERSION = 'Cygwin 6.8.2.0(202)' HISTCONTROL = 'ignoredups' SHLVL = '2' PATHEXT = '.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH' HOMEDRIVE = 'C:' COMSPEC = 'C:\WINDOWS\system32\cmd.exe' PROMPT = '$P$G' SYSTEMROOT = 'C:\WINDOWS' TMP = '/cygdrive/c/DOKUME~1/felder/LOKALE~1/Temp' LOGNAME = 'felder' PYTHONPATH = '.:~/python:/cygdrive/c/Python24/Lib' PROCESSOR_REVISION = '0f06' ODE_BASE = '/home/felder/src/ode-0.8-rc1' RUN = '\cygwin\bin\run -p /usr/X11R6/bin' PROGRAMFILES = 'C:\Programme' DISPLAY = '127.0.0.1:0.0' NUMBER_OF_PROCESSORS = '2' SESSIONNAME = 'Console' !EXITCODE = '' COMPUTERNAME = 'LAPKNOLL52' _ = '/usr/bin/cygcheck' POSIXLY_CORRECT = '1' HKEY_CURRENT_USER\Software\Cygnus Solutions
Vista how to guide or FAQ?
I've attempted to go over the entire Vista related postings and inferred that current cygwin.dll 1.5.24-2 on Vista has the following issues: - Cannot fork another process - Has an issue doing the install The fix is to download the latest snapshot and do a rebaseall, if so, what is the commandline for the rebaseall? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Vista how to guide or FAQ?
On Feb 7 11:02, Tim michals wrote: I've attempted to go over the entire Vista related postings and inferred that current cygwin.dll 1.5.24-2 on Vista has the following issues: - Cannot fork another process You got that wrong. 1.5.24 runs on Vista. - Has an issue doing the install Yes, it's the `install.exe' tool from coreutils. There's no fix for that so far. The right fix would be to add a manifest file for install.exe to the coreutils package. Eric, given that I pasted a matching manifest file in one of my mails about this Vista feature(*), would you mind to create a new coreutils package which contains this manifest file in /usr/bin alongside with install.exe? The problem is that the base-whatever script potentially fails on Vista otherwise, depending on the UAC settings. I know how frustrating this is, but we have no chice, really. Corinna (*) http://sourceware.org/ml/cygwin/2006-12/msg00580.html -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Vista how to guide or FAQ?
On Feb 7 18:08, Corinna Vinschen wrote: Eric, given that I pasted a matching manifest file in one of my mails about this Vista feature(*), would you mind to create a new coreutils package which contains this manifest file in /usr/bin alongside with install.exe? The problem is that the base-whatever script potentially fails on Vista otherwise, depending on the UAC settings. I know how frustrating this is, but we have no chice, really. [...] (*) http://sourceware.org/ml/cygwin/2006-12/msg00580.html Sorry! It's actually the very simple solution from this mail: http://sourceware.org/ml/cygwin/2006-12/msg00583.html Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
(I hope this gets through (and only once), gmane has been misbehaving all day...) Corinna Vinschen wrote: On Feb 7 11:55, Corinna Vinschen wrote: On Feb 6 16:38, Matthew Woehlke wrote: ...anyone have any guesses why rename() can't rename files in the same directory, if the directory is an NFS mount? This causes 'ar' to fail with the error: ar: unable to rename 'libcloudobj.a' reason: Invalid cross-device link I see this in strace with both 'ar' and 'mv': 46 371094 [main] ar 2588 rename: -1 = rename (h:\...path...\src\objects\stWBl4pj, h:\...path...\src\objects\libcloudobj.a) This seems to happen when the file already exists, but rename() only fails on NFS mounts (not local drives). Is this a problem with 'ar', or Cygwin's rename()? Since you're one of the few running NFS, you should also debug the situation further. Apparently the Cygwin rename function fails, but from your one-liner strace snippet there's no hint at all what happens. The error should have been mentioned a couple of lines before. You should also mention if you use a release version of Cygwin or a developer snapshot. You know http://cygwin.com/problems.html, don't you? I just installed SFU NFS for testing and, here it comes: I can not reproduce rename failing on an NFS share. Maybe it's something with your permissions. http://cygwin.com/acronyms/#SHTDI2(*) Thanks for trying. Sorry about forgetting the version info, although (and I forgot to mention this too, more apologies...) 'ren' in cmd.com also fails, which is why I am pretty sure the problem is that for some unknown reason, rename() (meaning Windows' API) fails on NFS if the target already exists, at least under whatever combinations of permissions I have... with error 17 (ERROR_NOT_SAME_DEVICE) of all things (which Cygwin translates to 18=EXDEV). I think the real question is 'should ar and/or Cygwin work around this apparent Windows bug'? D'oh! And I have more information... it only fails on Win2k3 (which means SUA, not SFU). I just tried on my XP machine and 'ar' succeeds. getfacl doesn't have much to say: $ getfacl.exe . # file: . # owner: mwoehlke # group: None user::rwx group::r-x other:r-x mask:rwx Cygwin version is 1.5.24(0.156/4/2) (are any of the other package versions really relevant?) -- Matthew We interrupt this e-mail to bring you a lame signature attempting to be witty. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
On Feb 7 11:48, Matthew Woehlke wrote: Corinna Vinschen wrote: Maybe it's something with your permissions. http://cygwin.com/acronyms/#SHTDI2(*) Thanks for trying. Sorry about forgetting the version info, although (and I forgot to mention this too, more apologies...) 'ren' in cmd.com also fails, which is why I am pretty sure the problem is that for some unknown reason, rename() (meaning Windows' API) fails on NFS if the target already exists, at least under whatever combinations of permissions I have... with error 17 (ERROR_NOT_SAME_DEVICE) of all things (which Cygwin translates to 18=EXDEV). I think the real question is 'should ar and/or Cygwin work around this apparent Windows bug'? If even the native Windows tool fails... Definitely not ar since that's a generic problem, not a binutils tool problem. I have my doubts that we really should workaround anything here. The first question is, what *exactly* is the problem we should workaround? If it's only a permission problem there's nothing to workaround for us. D'oh! And I have more information... it only fails on Win2k3 (which means SUA, not SFU). I just tried on my XP machine and 'ar' succeeds. getfacl doesn't have much to say: Forget getfacl. NFS doesn't support ACLs so there's no such thing as Windows security settings. The actual relevant settings are in the SFU or SUA MMC snap-in. Actually, if you ask me, your goal is to get the native ren working. If you can't get this working, you will have to ask Microsoft, IMHO. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
Corinna Vinschen wrote: On Feb 7 11:48, Matthew Woehlke wrote: Corinna Vinschen wrote: Maybe it's something with your permissions. http://cygwin.com/acronyms/#SHTDI2(*) Thanks for trying. Sorry about forgetting the version info, although (and I forgot to mention this too, more apologies...) 'ren' in cmd.com also fails, which is why I am pretty sure the problem is that for some unknown reason, rename() (meaning Windows' API) fails on NFS if the target already exists, at least under whatever combinations of permissions I have... with error 17 (ERROR_NOT_SAME_DEVICE) of all things (which Cygwin translates to 18=EXDEV). I think the real question is 'should ar and/or Cygwin work around this apparent Windows bug'? If even the native Windows tool fails... Definitely not ar since that's a generic problem, not a binutils tool problem. Right, that's why I was thinking Cygwin (i.e. cygwin1.dll) would need to change. But... (cont'd at the bottom) I have my doubts that we really should workaround anything here. The first question is, what *exactly* is the problem we should workaround? If it's only a permission problem there's nothing to workaround for us. I don't think so, since 'mv' succeeds (but then it also tries harder, i.e. removing the target and doing a copy followed by an unlink). D'oh! And I have more information... it only fails on Win2k3 (which means SUA, not SFU). I just tried on my XP machine and 'ar' succeeds. getfacl doesn't have much to say: Forget getfacl. NFS doesn't support ACLs so there's no such thing as Windows security settings. The actual relevant settings are in the SFU or SUA MMC snap-in. IOW the Unix permissions (as read by SUA)? Those're fine. In fact they *do* match what getfacl is reporting except that group = 0 (wheel), which apparently isn't mapping correctly. Actually, if you ask me, your goal is to get the native ren working. If you can't get this working, you will have to ask Microsoft, IMHO. ...and since I realized that the problem is in (newer) SUA and not (older) SFU, there might be hope there. And thanks anyway. :-) -- Matthew We interrupt this e-mail to bring you a lame signature attempting to be witty. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Vista how to guide or FAQ?
Corinna Vinschen corinna-cygwin at cygwin.com writes: Eric, given that I pasted a matching manifest file in one of my mails about this Vista feature(*), would you mind to create a new coreutils package which contains this manifest file in /usr/bin alongside with install.exe? http://sourceware.org/ml/cygwin/2006-12/msg00583.html Done. http://cygwin.com/ml/cygwin-announce/2006-12/msg00027.html -- Eric Blake -- View this message in context: http://www.nabble.com/Vista-how-to-guide-or-FAQ--tf3188257.html#a8851594 Sent from the Cygwin Users mailing list archive at Nabble.com. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
daylight savings in March
Just curious if the daylight savings time change occurring in the middle of March might be a factor. I suppose if cygwin gets its time from the PC and does no time adjustment, then there's no worries but to ensure that Windows takes care of it. Ken -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: daylight savings in March
On 07 February 2007 18:53, Ken Shaffer wrote: Just curious if the daylight savings time change occurring in the middle of March might be a factor. A factor in what? I suppose if cygwin gets its time from the PC and does no time adjustment, then there's no worries but to ensure that Windows takes care of it. Everything cygwin does relies on the underlying functionality of the windows o/s to get the basics right for it; cygwin is, after all, just a win32 DLL containing utility functions used by win32 programs.. There Is No Kernel In Cygwin! cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
ssh usage banner truncation
The usage banner lines produced by ssh.exe (latest version and previous ones) are unexpectedly truncated like this: usage: ssh [-1246AaCfgkMNnqsTtVvXxY] [-b bind_address] [-c cipher_spec] [-D [bind_address:]port] [-e escape_char] [-F when they ought to say: usage: ssh [-1246AaCfgkMNnqsTtVvXxY] [-b bind_address] [-c cipher_spec] [-D [bind_address:]port] [-e escape_char] [-F configfile] [-i identity_file] [-L [bind_address:]port:host:hostport] [-l login_name] [-m mac_spec] [-O ctl_cmd] [-o option] [-p port] [-R [bind_address:]port:host:hostport] [-S ctl_path] [-w local_tun[:remote_tun]] [EMAIL PROTECTED] [command] The truncated version is 128 bytes long. -- Robin Walker (Junior Bursar), Queens' College, Cambridge CB3 9ET, UK [EMAIL PROTECTED] http://www.queens.cam.ac.uk/ Tel:+44 1223 335528 p7sTrUuuxWiSQ.p7s Description: S/MIME cryptographic signature
Re: ssh usage banner truncation
Robin Walker wrote: The usage banner lines produced by ssh.exe (latest version and previous ones) are unexpectedly truncated like this: usage: ssh [-1246AaCfgkMNnqsTtVvXxY] [-b bind_address] [-c cipher_spec] [-D [bind_address:]port] [-e escape_char] [-F when they ought to say: usage: ssh [-1246AaCfgkMNnqsTtVvXxY] [-b bind_address] [-c cipher_spec] [-D [bind_address:]port] [-e escape_char] [-F configfile] [-i identity_file] [-L [bind_address:]port:host:hostport] [-l login_name] [-m mac_spec] [-O ctl_cmd] [-o option] [-p port] [-R [bind_address:]port:host:hostport] [-S ctl_path] [-w local_tun[:remote_tun]] [EMAIL PROTECTED] [command] The truncated version is 128 bytes long. Works fine for me. Perhaps you should submit a proper problem report if you can't find the source of the problem on your end. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Vista how to guide or FAQ?
On Feb 7 10:45, Eric Blake-1 wrote: Corinna Vinschen corinna-cygwin at cygwin.com writes: Eric, given that I pasted a matching manifest file in one of my mails about this Vista feature(*), would you mind to create a new coreutils package which contains this manifest file in /usr/bin alongside with install.exe? http://sourceware.org/ml/cygwin/2006-12/msg00583.html Done. http://cygwin.com/ml/cygwin-announce/2006-12/msg00027.html *blush* I'm a bit slow lately... Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
On Feb 7 12:21, Matthew Woehlke wrote:
Corinna Vinschen wrote:
Forget getfacl. NFS doesn't support ACLs so there's no such thing
as Windows security settings. The actual relevant settings are in
the SFU or SUA MMC snap-in.
IOW the Unix permissions (as read by SUA)? Those're fine. In fact they
*do* match what getfacl is reporting except that group = 0 (wheel),
which apparently isn't mapping correctly.
Actually, if you ask me, your goal is to get the native ren working.
If you can't get this working, you will have to ask Microsoft, IMHO.
...and since I realized that the problem is in (newer) SUA and not
(older) SFU, there might be hope there. And thanks anyway. :-)
I took the time to install and configure SUA on my R2 server. Then I
tried to rename a file, where the target file name already exists:
$ uname -a
CYGWIN_NT-5.2-WOW64 vmbert64 1.5.24(0.156/4/2) 2007-01-30 22:56 i686 Cygwin
$ cd /cygdrive/h/testdir
$ /my/secret/GetVolumeInformation-tool . | head -6
rootdir: h:\
Volume Name:
Serial Number : 2065
Max Filenamelength : 255
Filesystemname : NFS
Flags : 2
$ echo foo foo
$ echo bar bar
Try the native ren:
$ cmd /c ren foo bar
A duplicate file name exists, or the file
cannot be found.
Well, that's ok if you ask me...
Let's try a simple rename tool:
$ cat rename.c EOF
#include stdio.h
#include string.h
#include errno.h
int main(int argc, char **argv)
{
if (argc != 3)
{
fprintf (stderr, usage: rename old new\n);
return 1;
}
if (rename (argv[1], argv[2]))
{
fprintf (stderr, rename: errne %d %s\n, errno, strerror (errno));
return 2;
}
return 0;
}
EOF
$ make rename
cc rename.c -o rename
$ ./rename foo bar
$ cat bar
foo
$
So it WJFFM. There must be a problem in your setup.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: ar 'unable to rename... Invalid cross-device link'
Corinna Vinschen wrote: I took the time to install and configure SUA on my R2 server. Then I tried to rename a file, where the target file name already exists: [snip test program] So it WJFFM. There must be a problem in your setup. Ok, trying your exact test, I get EXDEV... so I agree with the above. Any guesses *what*, though? (I don't have that volume information toy :-(...) ...it might even be something stupid like if I reboot the machine, it'll work again (it's a lab machine, it isn't rebooted often). I've noticed that I often have to unmount and remount drives before I can write to them (nothing to do with Cygwin, just an annoyance I've noticed that may be related). Oh, well. Chalk it up to the computer being broken. Since it turns out I can build on a different machine, it's not important. Thanks again for the help, sorry it's a dead end. -- Matthew We interrupt this e-mail to bring you a lame signature attempting to be witty. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Using a DLL built with cygwin in VC++
Hello, I'm trying to use a DLL created with cygwin in my VC++ project. I have a DLL file and I need to create .lib in order to use the DLL in VC++, I use the following command sequence: dlltool --export-all-symbols --output-def mylib.def mylib.dll #cygwin's dlltool lib /DEF:mylib.def /OUT:mylib.lib #VC++'s lib Then, I link my VC++ application using the mylib.lib, but the linker says: mylib.lib : fatal error LNK1127: library is corrupt What should I do ? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: sshd exits during login attempt on WinXP 2003 x64 -- even with sshd running as service
Thanks for responding, Corinna. I obviously wasn't clear in my original post -- ssh login attempts to my Win2003 x64 system running Cygwin sshd as a service always fail, which is why I tried running sshd from the command prompt as a test. As far as I know, my Cygwin installation is completely up-to-date. I ran both the ssh-host-config and ssh-user-config scripts after installing Cygwin and configured sshd to run as a service (*without* privilege separation, which [I hope] should keep things simpler). I can start the sshd service using either the Win2003 Services control panel or cygrunsrv. When sshd starts, 4 success audit events appear in the Security event log, and the service appears to start normally. sshd is configured to log on as .\sshd_server, according to the Properties page for the service. If I then start a bash shell on my system and try to ssh localhost, I see the following: ~ 503 $ ssh localhost Connection closed by 127.0.0.1 ~ 504 $ /var/log/sshd.log remains empty, but a failure audit event appears in my Security event log: Event Type: Failure Audit Event Source: Security Event Category: Privilege Use Event ID: 577 Date: 2/7/2007 Time: 8:26:55 AM User: ABF466\sshd_server Computer: ABF466 Description: Privileged Service Called: Server: NT Local Security Authority / Authentication Service Service:LsaRegisterLogonProcess() Primary User Name: ABF466$ Primary Domain: AERO-ORG Primary Logon ID: (0x0,0x3E7) Client User Name: sshd_server Client Domain: ABF466 Client Logon ID:(0x0,0x14B91291) Privileges: SeTcbPrivilege I don't know how to activate output to /var/log/sshd.log -- would that help to diagnose this problem? -Brian Corinna Vinschen wrote: [exerpt of my original post snipped] This won't work. By default, no user has the required permissions to change the user context. Start here: $ less /usr/share/doc/Cygwin/openssh.README and here: http://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-switch and install sshd as service using the ssh-host-config script. Corinna -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Workaround for timeout problems with ssh/scp etc on a dual core Dell Laptop
Using my new Dell Inspiron E1705 laptop, with dual-core processor, I was unable to transfer much data using ssh or scp without getting the error: Disconnecting: Timeout, server not responding. That made the laptop unusable for accessing my CVS server via ssh. The problem seems similar to the one described by: http://sourceware.org/ml/cygwin/2006-07/msg00726.html I tracked the problem down to a conflict with the internet filter application I had installed. I have been using SafeEyes version 4.1, sold by http://www.safebrowse.com. I have submitted a report to their tech service and we'll see what happens. For now I have to run without the filter application installed. So if anyone encounters a similar problem, try uninstalling any filter you may have. It seems obvious now... -- Cliff -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
weird seg. fault related to fhandler_pipe::get_guard()
Hi there,
I checked up all the archives and FAQ but seems not got a resolve. Ok here is
the details:
My pc is HP with XP Pro; the cygwin version is 1.5.24 (Jan.31,2007).
The basic structure of my program is like this:
main() {
:
:
status=A(file_input_name, ..);
:
}
A(input_file,..){
:
:
status=B(input_file,..);
:
status=C(input_file,..);
:
}
B(input_file){
:
status=D(input_file);
:
}
C(input_file){
:
status=D(input_file);
:
}
D(input_file){
:
fp=fopen(input_file);
:
fclose(fp);
}
Building was a success, but when run it I got a segmentation fault immediately
while executing function C (B was fine, and nothing was run in C). The code has
no any problems on linux, solaris and irix. Followings are results from gdb
running and xxx.exe.stackdump.
gdb output:
Program received signal SIGSEGV, Segmentation fault.
0x005e5b63 in fhandler_pipe::get_guard()
xxx.exe.stackdump looks like this:
Exception: STATUS_STACK_OVERFLOW at eip=005E5B63
eax=0003C41C ebx=00A510E8 ecx=000323CC edx=0001 esi=001A77F0 edi=00B06F38
ebp=000773D8 esp=000773C4 program=c:\HEGwin\heg\bin\gdtif.exe, pid 3800, thread
main
cs=001B ds=0023 es=0023 fs=003B gs= ss=0023
Stack trace:
Frame Function Args
000773D8 005E5B63 (001A77F0, 0001, 00B06F38, 00A510E8)
001A41B8 0041D81B (001A77F0, , 0001, )
0022CB98 0040C814 (0022CBC0, 00912930, 00686FB0, 004011BF)
0022CCA8 00401238 (0003, 00912930, 00910090, 6003)
0022CD98 61006198 (, 0022CDD0, 61005510, 0022CDD0)
61005510 61004416 (009C, A02404C7, E8611001, FF48)
823515 [main] gdtif 3800 _cygtls::handle_exceptions: Exception:
STATUS_ACCESS_VIOLATION
838259 [main] gdtif 3800 _cygtls::handle_exceptions: Error while dumping state
(probably corrupted stack)
Thank you very much for your help!
-David
Never miss an email again!
Yahoo! Toolbar alerts you the instant new Mail arrives.
http://tools.search.yahoo.com/toolbar/features/mail/
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Eliminating -mno-cygwin from gcc.
Christopher Faylor wrote in message news:[EMAIL PROTECTED] [snip] I guess this means that I need to generate a cross-binutils, too. Is a cross-binutils really needed? Which (if any) of the binutils act differently when targeted at MingW32 than when targeted at Cygwin? (I've never worked with a cross toolchain where the executable format of the host and target are the same.) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
accessing db2 v9 from cygwin shell
Folks, I used to be able to access db2 (v8 and prior) from within a cygwin bash shell after doing some tricks with the DB2CLP variable (see http://www.cygwin.com/ml/cygwin/2003-02/msg00635.html). e.g. export DB2CLP=$CLP_SAVE # saved during db2cmd session db2 connect to mydb db2 list tables for all db2 terminate With db2 v9 this doesn't work anymore, as the db2 connection seems to be gone after db2 connect to mydb: when issuing db2 list tables for all I get now a nasty: SQL1024N A database connection does not exist. SQLSTATE=08003 does anybody know a workaround? Or do I really have to use that db2cmd abomination? thx, H. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: pasting into an ssh connection painfully slow... sometimes
DePriest, Jason R. wrote: My department manages some Netscreen NS208 firewalls. If I ssh into them with the cygwin ssh client (either with rxvt or with cygwin from a cmd.exe prompt), pasting into the session runs like a 2400 baud modem. If I ssh into them with something else such as PuTTY, then the paste function works just like it does on other systems: fast. I don't have this problem with cygwin ssh on systems other than the Netscreens. PuTTY doesn't have any special configuration done to it so that it works. I just connect and its okay. Has anyone else experienced this sort of slowness and if so, how did you fix it? Perhaps Netscreen recognizes PuTTY as an OK app but not 'ssh'. Worth a check. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
gnome without fonts!!!
Hi! I install x, gnome, etc.. All goes well, work nice, speedy and all, but without fonts!!! Everything is squared, instead of letters... So I wonder what to do,.. Thanks !!! -- View this message in context: http://www.nabble.com/gnome-without-fonts%21%21%21-tf3190947.html#a8858323 Sent from the Cygwin Users mailing list archive at Nabble.com. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
calling cgwin apps from the windows shell
I use rsync and was wondering if there is a way to call rsync directly from a windows batch file where the path arguments would be properly cleaned and slashflipped -- I'm wondering if the windows shell will be more kind to file names with spaces in them... or if rsync will still choke on names.. -- -- David Bear College of Public Programs at Arizona State University -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: daylight savings in March
Dave Korn wrote: On 07 February 2007 18:53, Ken Shaffer wrote: Just curious if the daylight savings time change occurring in the middle of March might be a factor. A factor in what? I suppose if cygwin gets its time from the PC and does no time adjustment, then there's no worries but to ensure that Windows takes care of it. Everything cygwin does relies on the underlying functionality of the windows o/s to get the basics right for it; cygwin is, after all, just a win32 DLL containing utility functions used by win32 programs.. There Is No Kernel In Cygwin! Todays headlines: No Kernel, Korn Claims. Saddened Cygwinners shocked. Related story: Cover-up revealed - Cygwin's Korn doesn't write or use Korn shell. Millions outraged. ;-) -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: stupid spaces in environment vars
David Bear wrote: I would like to have used something like cd $USERPROFILE in a bash script but since windows insists on putting spaces in names, this seems impossible. I did find a usecase where the spaces in c:\Documents and Settings\username were tripple quoted. However, this did not work for me. Bash still wanted to split on the space. Is there any cool utility that could be used like cleanpath=pathcleaner($USERPROFILE) cd $cleanpath I know this is a consistent issue with cygwin. There really needs to be a good solution. $ cat cup #!/bin/bash cd $USERPROFILE pwd $ ./cup /Documents and Settings/me What's the problem again? -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: calling cgwin apps from the windows shell
David Bear wrote: I use rsync and was wondering if there is a way to call rsync directly from a windows batch file where the path arguments would be properly cleaned and slashflipped -- I'm wondering if the windows shell will be more kind to file names with spaces in them... or if rsync will still choke on names.. 'cygpath' can convert paths back and forth between DOS/Windows paths and POSIX. If you're using Cygwin's rsync, it will definitely prefer POSIX paths though. Windows shell (I use the term lightly here) is no better with raw spaces in the paths than any other shell. You'll need to quote them just like you would in other shells. Perhaps you should state the problem you're having. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: pasting into an ssh connection painfully slow... sometimes
On 2/7/07, DePriest, Jason R. wrote: My department manages some Netscreen NS208 firewalls. If I ssh into them with the cygwin ssh client (either with rxvt or with cygwin from a cmd.exe prompt), pasting into the session runs like a 2400 baud modem. If I ssh into them with something else such as PuTTY, then the paste function works just like it does on other systems: fast. I don't have this problem with cygwin ssh on systems other than the Netscreens. PuTTY doesn't have any special configuration done to it so that it works. I just connect and its okay. Has anyone else experienced this sort of slowness and if so, how did you fix it? Windows has some deficiencies when it comes to sending data in certain configurations. They particularly show up due to the way cygwin makes use of winsock. I looked into this a while ago: http://www.cygwin.com/ml/cygwin-patches/2006-q2/msg00031.html When I get some time I'll redo those patches against the latest cygwin version. In the meantime, if netscreen provides a sysctl net.inet.tcp.ackonpush or some way to disable delayed acks, then that might help you. Lev -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
