GCC 4.5 has been released, what does this mean for cygwin?
Hello, as I'm sure many of you have noticed, GCC 4.5 has been released. I think I recall Mr Dave Korn saying that he would skip releasing GCC 4.4 for Cygwin and instead focus on getting fixes in for GCC 4.5. I'm just curious to where things stand now. Dave? Btw, I'm very much grateful for your hard work. I've said so in the past and I'm happy to repeat it. Wish I had the technical skills required to help. :( - EL -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: git stopped working with 1.7.1
The problem still exists but it appears infrequently with small repos. But larger repositories constantly fail even with the environment variable below defined. I currently use the putty ssh client for git. Nevertheless I consider this a major issue since the ssh protocol is the most important use case for git and as far as I know the cygwin version is the recommended one under windows. Corinna Vinschen-2 wrote: On Jan 18 00:17, ol42 wrote: But obviously the environment variable has still an effect! Without binmode I get the following error $ git clone ssh://o...@simulacron/home/git/gen-dsp.git Initialized empty Git repository in /home/ol/tmp/gen-dsp/.git/ remote: Counting objects: 979, done. remote: Compressing objects: 100% (966/966), done. fatal: The remote end hung up unexpectedly fatal: early EOF fatal: index-pack failed If I have CYGWIN=binmode defined it obviously works --- $ git clone ssh://o...@simulacron/home/git/gen-dsp.git Initialized empty Git repository in /home/ol/tmp/gen-dsp/.git/ remote: Counting objects: 979, done. remote: Compressing objects: 100% (966/966), done. remote: Total 979 (delta 483), reused 0 (delta 0) Receiving objects: 100% (979/979), 950.64 KiB | 1269 KiB/s, done. Resolving deltas: 100% (483/483), done. Checking out files: 100% (196/196), done. Dunno why you see two different results, but the CYGWIN=binmode setting most certainly doesn't exist anymore in Cygwin 1.7. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple -- View this message in context: http://old.nabble.com/git-stopped-working-with-1.7.1-tp26905956p28606567.html Sent from the Cygwin list mailing list archive at Nabble.com. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Missing stack_chk type functions
Hi all, I tried to install the Win32::GuiTest from cpan. It did not work because g++ -shared GuiTest.o DibSect.o -o blib/arch/auto/Win32/GuiTest/GuiTest.dll \ /usr/lib/perl5/5.10/i686-cygwin/CORE/cygperl5_10.dll -L/usr/lib/w32api -lgdi32\ GuiTest.o:GuiTest.cpp:(.text+0x91f0): undefined reference to `___stack_chk_guard' GuiTest.o:GuiTest.cpp:(.text+0x96c1): undefined reference to `___stack_chk_fail' [...] I found Re: glibc-2.4 __stack_chk_guard/__pointer_chk_guard from 2006 which is pretty old and refers to using distcc So I don't know whats the problem today. Any hints ?? Norbert -- Dipl. Phys. Norbert Zacharias Wind Measurements Power Curve Measurements DEWI GmbH Ebertstrasse 96 26382 Wilhelmshaven Germany Tel.: +49 4421 4808 876 Fax:+49 4421 4808 843 Email: n.zachar...@dewi.de Home: http://http://www.dewi.de DEWI GmbH - Deutsches Windenergie-Institut, Wilhelmshaven Commercial Register No.: Amtsgericht Oldenburg, HRB 130241 Managing Director: Jens Peter Molly Chairman of the supervisory board: Ministerialrat Dr. Niels Kämpny P Please consider the environment before printing this email. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: vfork always fail problem
Huang Bambo wrote on 2010-05-19: http://cygwin.com/snapshots/ To the OP: please check it out and verify that it solves your problem. Thanks a lot. I've changed my work directory to a full English name directory and it dosn't matter me. And this problem is first reported by gcc, strange :) . Huang, you - as the person who first saw and documented the problem in public (thank you!) - are in the best position to test it, if you can recreate the original situation (with the GBK(?) directory names). It would be beneficial to all of us, and you would be doing everybody a favour (if you wish, which is your choice), if you could test if the problem reproduces with Cygwin 1.7.5, and goes away with the snapshot. Thank you. Best -- Matthias Andree -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Cygwin enviroment !C:=
Hello, * On Wed, May 19, 2010 at 12:25:15PM +0200 s.baun...@ifw-dresden.de wrote: When I display the environment variables in the bash shell using $ set On the PCs with IMOD running correctly I found these entries [...] !D:='D:\cygwin\bin' This is the path to Cygwin !X:='X:\' This is HOMEDRIVE of Windows (PC: Windows XP Professional SP3 in a domain) [...] What does !drive mean and how can I insert such expressions in the environment variable? I can only guess because I read just a few days ago this: http://blogs.msdn.com/oldnewthing/archive/2010/05/06/10008132.aspx Did you start cygwin from cmd.exe (or via a .BAT or .CMD file) on the machines where these entries are? Best regards, Spiro. -- Spiro R. Trikaliotis http://opencbm.sf.net/ http://www.trikaliotis.net/ http://www.viceteam.org/ -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: NCurses and Cygwin
On 5/19/2010 2:04 AM, rushojp wrote: u...@localhost /usr/include $ for h in curses.h eti.h form.h menu.h ncurses.h panel.h term.h unctrl.h ; do ln -s ncurses/$h $h ;done I think libncurses-devel package should have these symlink files. Is anybody going to read the documentation? From /usr/share/doc/Cygwin/ncurses.README: ... All libraries (libncurses, libpanel, libmenu, libform, libncurses++) come in both static (.a) and dynamic (.dll) forms. To link your project with the C libraries: #0) Use -I /usr/include/ncurses when compiling ... Port Notes: -- ncurses-5.7-18 -- 20090102 --- * Remove symlinks in /usr/include/ to /usr/include/ncurses/* The reasons the symlinks in /usr/include were removed was to treat ncurses and ncursesw the same, rather than stamping one as the approved version with official symlinks in /usr/include. Actually, I'd prefer if people started using -I/usr/include/ncursesw and linking against the wide version of the library instead. -- Chuck -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: 1.7.5: Bug with bash read in /etc/profile.d invocation
OK, I changed my script to have: read -p How are you today? Ans /dev/stdin But I now get bash: /dev/stdin: No such file or directory Since profile is redirecting stdin stdout, wouldn't it make more sense for profile to redirect stdin and stdout back to normal when sourcing the profile.d scripts? Thanks, Dave -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Garber, Dave (GE Infra, Energy, Non-GE) Sent: Tuesday, May 18, 2010 4:30 PM To: cygwin@cygwin.com Subject: RE: 1.7.5: Bug with bash read in /etc/profile.d invocation Thanks for the info. This didn't happen in 1.5.25 so something with 1.7.5 is different. I'll go back to my 1.5.25 setup and look at /etc/profile and see what's different. Dave -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Steven Collins Sent: Tuesday, May 18, 2010 4:23 PM To: cygwin@cygwin.com Subject: Re: 1.7.5: Bug with bash read in /etc/profile.d invocation Look at /etc/profile where it runs the profile.d scripts. The scripts are run with standard input redirected to a here document generated by a find command. That is the source of the /etc/profile.d/xinit.sh you're seeing as the answer. The read statement in your script is actually consuming one of the arguments intended to be processed by the read in /etc/profile. Because the scripts are sourced by the current shell your #! line has no affect (-x isn't getting set.) In other words, the shell is doing exactly what it has been told to do. Don't use a read in your profile.d scripts unless you make sure to reroute standard input back to the terminal. On Thu, May 13, 2010 at 07:16, Garber, Dave (GE Infra, Energy, Non-GE) wrote: #!/usr/bin/bash -x echo In p.sh read -p How are you today? Ans echo Ans is $Ans -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
bash - command - PATH question
Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? What is the shebang (first line) of foo? BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? -Jeremy -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
On 05/19/2010 06:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? Usually this means that foo is in DOS mode and contains extra carriage returns. So your she-bang line (i.e. #!/bin/bash) is actually #!/bin/bashcr, a file that doesn't exist. Do a dos2unix foo and you should be fine. -- Andrew DeFaria http://defaria.com You have the right to remain silent. Anything you say will be misquoted, then used against you. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: bash - command - PATH question
On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? I get exactly the same error. The error is correct. ./bin/foo doesn't exist. (I'm not in home directory when I issue the command.) What is the shebang (first line) of foo? #!/bin/bash I thought it might have something to do with this and tried commenting It out but nothing changed. BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? /cygdrive/c/DOCUME~1/harryr/bin/foo -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
Again, have you tried dos2unix foo? On 05/19/2010 08:16 AM, Rockefeller, Harry wrote: On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? I get exactly the same error. The error is correct. ./bin/foo doesn't exist. (I'm not in home directory when I issue the command.) What is the shebang (first line) of foo? #!/bin/bash I thought it might have something to do with this and tried commenting It out but nothing changed. BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? /cygdrive/c/DOCUME~1/harryr/bin/foo -- Andrew DeFaria http://defaria.com The easiest way to find something lost around the house is to buy a replacement. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: bash - command - PATH question
Again, have you tried dos2unix foo? Yes this didn't help. [snip] -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
On 05/19/2010 08:31 AM, Rockefeller, Harry wrote: Again, have you tried dos2unix foo? Yes this didn't help. That's funny because this is the usual cause. Are you sure there is no extra carriage return line endings. I usually check by going into vim and seeing if it says [DOS} at the bottom (there are other ways). [snip] Thanks dude I was totally lost wondering where the rest of the quote was until I saw your [snip] comment... :-) -- Andrew DeFaria http://defaria.com Why do we play in recitals and recite in plays? -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: bash - command - PATH question
I found that if I give the simple 'bash' command to create a new shell then type 'foo' it does work. [quote on] -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Andrew DeFaria Sent: Wednesday, May 19, 2010 10:28 AM To: cygwin@cygwin.com Subject: Re: bash - command - PATH question Again, have you tried dos2unix foo? On 05/19/2010 08:16 AM, Rockefeller, Harry wrote: On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? I get exactly the same error. The error is correct. ./bin/foo doesn't exist. (I'm not in home directory when I issue the command.) What is the shebang (first line) of foo? #!/bin/bash I thought it might have something to do with this and tried commenting It out but nothing changed. BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? /cygdrive/c/DOCUME~1/harryr/bin/foo [quote off] -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
On 5/19/2010 10:16 AM, Rockefeller, Harry wrote: On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? I get exactly the same error. The error is correct. ./bin/foo doesn't exist. (I'm not in home directory when I issue the command.) What is the output of: echo $PATH Where does bin/foo exist, and from where are you trying to run it in these tests? What is the shebang (first line) of foo? #!/bin/bash I thought it might have something to do with this and tried commenting It out but nothing changed. I'm not aware of any way to comment out a shebang line aside from making it not be a shebang line anymore. In any case that line looks good, assuming it has a Unix line ending rather than a Windows line ending. You told Andrew that running dos2unix on foo didn't fix anything, so I assume it has a Unix line ending. BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? /cygdrive/c/DOCUME~1/harryr/bin/foo Did you run which foo from inside your home directory or from the same directory in which you attempted to run foo and ./bin/foo? -Jeremy -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
On 5/19/2010 10:37 AM, Andrew DeFaria wrote: On 05/19/2010 08:31 AM, Rockefeller, Harry wrote: Again, have you tried dos2unix foo? Yes this didn't help. That's funny because this is the usual cause. Are you sure there is no extra carriage return line endings. I usually check by going into vim and seeing if it says [DOS} at the bottom (there are other ways). Running od -c on the file is a pretty quick and simple way to check. If you see \r in the output, you most likely have at least a few Windows line endings in the file. -Jeremy -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: bash - command - PATH question
I double and triple checked for DOS things \r, ^M, etc. I use emacs to edit and so it's pretty clear about DOS things. Anyway, I ran od -c on files and outputs of 'echo $PATH', etc. I compared PATH in normal shell with PATH after spawning a new bash, since the command worked in the new bash. The only difference in PATH is that :/cygdrive/c/DOCUME~1/harryr/bin was appended to PATH. This is understandable since my .bashrc does export PATH=$PATH:~/bin. So, in my normal shell where the error was happening I next tried this: PATH=$PATH:~/bin and low and behold foo ran correctly. FWIW, /cygdrive/c/DOCUME~1/harryr/bin is in PATH twice now, both at the end. I agree that the shebang is normaly the problem. But all this difference in behavior is external to foo. -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Rockefeller, Harry Sent: Wednesday, May 19, 2010 10:38 AM To: cygwin@cygwin.com Subject: RE: bash - command - PATH question I found that if I give the simple 'bash' command to create a new shell then type 'foo' it does work. [quote on] -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Andrew DeFaria Sent: Wednesday, May 19, 2010 10:28 AM To: cygwin@cygwin.com Subject: Re: bash - command - PATH question Again, have you tried dos2unix foo? On 05/19/2010 08:16 AM, Rockefeller, Harry wrote: On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? I get exactly the same error. The error is correct. ./bin/foo doesn't exist. (I'm not in home directory when I issue the command.) What is the shebang (first line) of foo? #!/bin/bash I thought it might have something to do with this and tried commenting It out but nothing changed. BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? /cygdrive/c/DOCUME~1/harryr/bin/foo [quote off] -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
[ANNOUNCEMENT] Updated: {neon/libneon27/libneon-devel}-0.29.3-1: HTTP and WebDAV library
Hi New versions of 'neon/libneon27/libneon-devel' have been uploaded to a server near you. o Update to latest upstream neon NEWS: === * Change ne_sock_close() to no longer wait for SSL closure alert: - fixes possible hang with IIS servers when closing SSL connection - this reverts the behaviour with OpenSSL to match 0.28.x, and changes the behaviour with GnuTLS to match that with OpenSSL * Fix memory leak with GnuTLS * API clarification in ne_sock_close(): - SSL closure handling now documented - return value semantics fixed to describe the implementation CYGWIN-ANNOUNCE UNSUBSCRIBE INFO If you want to unsubscribe from the cygwin-announce mailing list, please use the automated form at: http://cygwin.com/lists.html#subscribe-unsubscribe If this does not work, then look at the List-Unsubscribe: tag in the email header of this message. Send email to the address specified there. It will be in the format: cygwin-announce-unsubscribe-you=yourdomain@cygwin.com If you need more information on unsubscribing, start reading here: http://sourceware.org/lists.html#unsubscribe-simple Please read *all* of the information on unsubscribing that is available starting at this URL. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
[ANNOUNCEMENT] Updated: xpdf-3.02-10: An open source viewer for Portable Document Format (PDF) files
Hi A new version of 'xpdf' has been uploaded to a server near you. o Rebuild for cygwin 1.7 with gcc-4 o Incorporated latest upstream patches CYGWIN-ANNOUNCE UNSUBSCRIBE INFO If you want to unsubscribe from the cygwin-announce mailing list, please use the automated form at: http://cygwin.com/lists.html#subscribe-unsubscribe If this does not work, then look at the List-Unsubscribe: tag in the email header of this message. Send email to the address specified there. It will be in the format: cygwin-announce-unsubscribe-you=yourdomain@cygwin.com If you need more information on unsubscribing, start reading here: http://sourceware.org/lists.html#unsubscribe-simple Please read *all* of the information on unsubscribing that is available starting at this URL. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
[ANNOUNCEMENT] Updated: tzcode-2010j-1: The time zone package.
Hi A new version of 'tzcode' has been uploaded to a server near you. CYGWIN NEWS: * Update to latest upstream release tzcode/tzdata NEWS == * Sorry no changelog available. You have to do the diff yourself. CYGWIN-ANNOUNCE UNSUBSCRIBE INFO If you want to unsubscribe from the cygwin-announce mailing list, please use the automated form at: http://cygwin.com/lists.html#subscribe-unsubscribe If this does not work, then look at the List-Unsubscribe: tag in the email header of this message. Send email to the address specified there. It will be in the format: cygwin-announce-unsubscribe-you=yourdomain@cygwin.com If you need more information on unsubscribing, start reading here: http://sourceware.org/lists.html#unsubscribe-simple Please read *all* of the information on unsubscribing that is available starting at this URL. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Missing stack_chk type functions
DEWI - N. Zacharias schrieb: Hi all, I tried to install the Win32::GuiTest from cpan. It did not work because g++ -shared GuiTest.o DibSect.o -o blib/arch/auto/Win32/GuiTest/GuiTest.dll \ /usr/lib/perl5/5.10/i686-cygwin/CORE/cygperl5_10.dll -L/usr/lib/w32api -lgdi32\ GuiTest.o:GuiTest.cpp:(.text+0x91f0): undefined reference to `___stack_chk_guard' GuiTest.o:GuiTest.cpp:(.text+0x96c1): undefined reference to `___stack_chk_fail' [...] I found Re: glibc-2.4 __stack_chk_guard/__pointer_chk_guard from 2006 which is pretty old and refers to using distcc So I don't know whats the problem today. You need to install gcc4 and set it as default with set-gcc-default-4.sh -- Reini Urban http://phpwiki.org/ http://murbreak.at/ -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Terminal windows
On Wed, May 19, 2010 at 06:29:03PM +0100, Luis Vital wrote: Hi, I am running the text version of cygwin. What I want to do is, while in a terminal window launch a script wich opens another terminal window and launches inside this one a program. Thanks in advance for any help. Best regards, if not using xterm, replace it with your terminal of choice and adjust parameters accordingly $ cat ./launchit.sh #!/bin/bash nohup xterm -e $@ -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: bash - command - PATH question
After more testing, where export PATH=$PATH:~/bin only exists in .bash_login. If I run 'foo' from my login directory it works. If I then cd to a different place I get the error bash: ./bin/foo: No such file or directory which is true since I cd'd away from home *BUT* when I *first* ran foo ~/bin was the same as ./bin and is where foo lives. If I don't run foo from my home directory first, i.e., cd away and then run foo it works. The problem seems to be an execuatable system Memory which essentially says yes I found foo at ./bin and then catalogs this to save time later when foo is run. But, when cd'd away from home ./bin no longer has foo and the error occurs. This dynamic system memory to reexecute commands faster seems to be confusing ~/bin with ./bin. -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Rockefeller, Harry Sent: Wednesday, May 19, 2010 11:22 AM To: cygwin@cygwin.com Subject: RE: bash - command - PATH question I double and triple checked for DOS things \r, ^M, etc. I use emacs to edit and so it's pretty clear about DOS things. Anyway, I ran od -c on files and outputs of 'echo $PATH', etc. I compared PATH in normal shell with PATH after spawning a new bash, since the command worked in the new bash. The only difference in PATH is that :/cygdrive/c/DOCUME~1/harryr/bin was appended to PATH. This is understandable since my .bashrc does export PATH=$PATH:~/bin. So, in my normal shell where the error was happening I next tried this: PATH=$PATH:~/bin and low and behold foo ran correctly. FWIW, /cygdrive/c/DOCUME~1/harryr/bin is in PATH twice now, both at the end. I agree that the shebang is normaly the problem. But all this difference in behavior is external to foo. -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Rockefeller, Harry Sent: Wednesday, May 19, 2010 10:38 AM To: cygwin@cygwin.com Subject: RE: bash - command - PATH question I found that if I give the simple 'bash' command to create a new shell then type 'foo' it does work. [quote on] -Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Andrew DeFaria Sent: Wednesday, May 19, 2010 10:28 AM To: cygwin@cygwin.com Subject: Re: bash - command - PATH question Again, have you tried dos2unix foo? On 05/19/2010 08:16 AM, Rockefeller, Harry wrote: On 5/19/2010 8:50 AM, Rockefeller, Harry wrote: Given that 'foo' is a bash script, why is it that: $ foo returns the error: bash: ./bin/foo: No such file or directory What happens when you directly run ./bin/foo? I get exactly the same error. The error is correct. ./bin/foo doesn't exist. (I'm not in home directory when I issue the command.) What is the shebang (first line) of foo? #!/bin/bash I thought it might have something to do with this and tried commenting It out but nothing changed. BUT since foo is *really in* PATH, e.g., $ `which foo` runs correctly? What is the output of which foo in this case? /cygdrive/c/DOCUME~1/harryr/bin/foo [quote off] -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: CygWin-1.7.3-1 fails to provide pipe. (merge back into trunk please)
On 5/19/2010 1:04 PM, Gary wrote: benczur writes: Sorry for bringig up this thread again, but was this issue fixed in 1.7.5-1? When I do: $ mkfifo /tmp/pipe $ ls -l /tmp/pipe cygwin hangs... and cannot be stopped by Ctrl-C. WFM (in the sense it responds to Ctrl-C). The function of Ctrl-C is conditional, based on whether you use the Windows terminal (cmd.exe) or a Cygwin terminal (mintty/rxvt). Under the Windows terminal, Ctrl-C is ignored, but it works as expected under a Cygwin terminal. Alternatively after $ ls -l /tmp/pipe $ cat /tmp/pipe nothing happens, so I interrupt cat with Ctrl-C, and finally try: $ cat /tmp/pipe cat: -: Communication error on send I see this behavior under both kinds of terminals. Under Linux, all of the above works as you expect, so this appears to be a Cygwin defect. -Jeremy -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
[please don't top-post] On 05/19/2010 12:45 PM, Rockefeller, Harry wrote: After more testing, where export PATH=$PATH:~/bin only exists in .bash_login. If I run 'foo' from my login directory it works. If I then cd to a different place I get the error bash: ./bin/foo: No such file or directory What does 'type foo' say? Is it hashed? What is $HOME? Could you accidentally have set home to a relative path, in which case ~ is relative instead of absolute? What does 'echo $PATH' say; are there any relative paths in that listing? -- Eric Blake ebl...@redhat.com+1-801-349-2682 Libvirt virtualization library http://libvirt.org signature.asc Description: OpenPGP digital signature
RE: bash - command - PATH question
On 05/19/2010 12:45 PM, Rockefeller, Harry wrote: After more testing, where export PATH=$PATH:~/bin only exists in .bash_login. If I run 'foo' from my login directory it works. If I then cd to a different place I get the error bash: ./bin/foo: No such file or directory Eric Blake What does 'type foo' say? Is it hashed? What is $HOME? Could you accidentally have set home to a relative path, in which case ~ is relative instead of absolute? What does 'echo $PATH' say; are there any relative paths in that listing? $ type mysvn mysvn is hashed (./mysvn) Cat's out of the bag. foo really is a bash script to run my common svn commands. $ echo $HOME /cygdrive/c/DOCUME~1/harryr $ cd ~ ; pwd /cygdrive/c/DOCUME~1/harryr $ echo $PATH [gives a very long path. Yes, ./ and ./bin are present but are the only relative paths in the listing.] -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: bash - command - PATH question
On 05/19/2010 01:42 PM, Rockefeller, Harry wrote: $ type mysvn mysvn is hashed (./mysvn) There's your problem. Bash remembers the hashed location of where it last found the command, but that location is relative. You either need to disable bash's hashing, or force it to re-evaluate its hash as soon as you are in a different directory. 'shopt -s checkhash'. $ echo $PATH [gives a very long path. Yes, ./ and ./bin are present but are the only relative paths in the listing.] Obligatory comments about having '.' in your PATH being a security hole, opening the door for trojan horse applications. (Well, I do it, but only as the LAST item in my PATH, so that any application I am likely to run occurs earlier in the PATH). But ./bin? Seriously, change your path to call out an absolute path of ~/bin instead of using ./bin and relying on the first invocation being from ~. That way, bash will hash the absolute name in the first place. At which point, this is no longer a cygwin-specific question - you would get the same behavior on Linux. -- Eric Blake ebl...@redhat.com+1-801-349-2682 Libvirt virtualization library http://libvirt.org signature.asc Description: OpenPGP digital signature
Re: Terminal windows
Hi Reid, Thanks for your answer. Nevertheless I don't have my problem solved yet. If I make: $ cat ./launchit.sh I obtain : cat: ./launchit.sh: No such file or directory Besides it seems I am not using xterm. If I make: $ nohup xterm -e $@ I obtain : nohup: ignoring input and appending output to `nohup.out' nohup: failed to run command `xterm': No such file or directory Could it be passible that you explain a bit better? Thanks in advance. Luis Vital On Wed, May 19, 2010 at 06:29:03PM +0100, Luis Vital wrote: Hi, I am running the text version of cygwin. What I want to do is, while in a terminal window launch a script wich opens another terminal window and launches inside this one a program. Thanks in advance for any help. Best regards, if not using xterm, replace it with your terminal of choice and adjust parameters accordingly $ cat ./launchit.sh #!/bin/bash nohup xterm -e $@ -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Terminal windows
On 5/19/2010 3:59 PM, Luis Vital wrote: Hi Reid, Thanks for your answer. Nevertheless I don't have my problem solved yet. If I make: $ cat ./launchit.sh I obtain : cat: ./launchit.sh: No such file or directory The idea was for you to create a launchit.sh file with the contents given, starting the the #!/bin/bash line. The poster ran cat specifically to show you what needed to be in that file. Note that launchit.sh should have execute permissions set (e.g., via chmod +x launchit.sh. Besides it seems I am not using xterm. If I make: $ nohup xterm -e $@ I obtain : nohup: ignoring input and appending output to `nohup.out' nohup: failed to run command `xterm': No such file or directory Indeed. Either you don't have xterm installed, or else its directory is on on your PATH. You can check your path by typing echo $PATH, for example. But it sounds to me as if you are new to Unix-like commands, which makes me wonder why you're using cygwin in the first place ... (Not meaning to be rude; just wondering if maybe there's a better ay to get at what you really want.) Regards -- Eliot MOss -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: bash - command - PATH question [not cygwin issue last post]
-Original Message- From: cygwin-ow...@cygwin.com [mailto:cygwin-ow...@cygwin.com] On Behalf Of Eric Blake Sent: Wednesday, May 19, 2010 2:51 PM To: cygwin@cygwin.com Subject: Re: bash - command - PATH question On 05/19/2010 01:42 PM, Rockefeller, Harry wrote: $ type mysvn mysvn is hashed (./mysvn) There's your problem. Bash remembers the hashed location of where it last found the command, but that location is relative. You either need to disable bash's hashing, or force it to re-evaluate its hash as soon as you are in a different directory. 'shopt -s checkhash'. $ echo $PATH [gives a very long path. Yes, ./ and ./bin are present but are the only relative paths in the listing.] Obligatory comments about having '.' in your PATH being a security hole, opening the door for trojan horse applications. (Well, I do it, but only as the LAST item in my PATH, so that any application I am likely to run occurs earlier in the PATH). But ./bin? Seriously, change your path to call out an absolute path of ~/bin instead of using ./bin and relying on the first invocation being from ~. That way, bash will hash the absolute name in the first place. At which point, this is no longer a cygwin-specific question - you would get the same behavior on Linux. Thanks for your help in tracking this down. I found out that ./ and ./bin were being prepended to PATH by Mortens Cygwin X-Launcher. FWIW, ./ is in the default PATH in Mortens Cygwin X-Launcher. I removed them from there and added export PATH=$PATH:~/bin/:./bin/:./ to ~/.bash_login. Everything began working correctly afterward. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Minor patch to /etc/profile
Le 01/05/2010 05:25, Chris Sutcliffe a écrit :
The PS1 definition for ksh in /etc/profile uses the literal ascii
characters '^[' for the escape sequence as opposed to the single byte
escape control code. The attached patch corrects that as well as
HOSTNAME not being set correctly (at least for mksh).
the right way to go would be something like :
the current way :
PS1=$(print '\033]0;${pwd}\n\033[32m${us...@${hostname}
\033[33m${PWD}\033[0m\n$ ')
w/ tilde expansion as bash does :
TILDE=\~
PS1=$(print
'\033]0;${TILDE[(1-0${PWD%%@([!/]*|${HOME}*)}1)]:-}${pwd#${home}}\n\033[32m${us...@${hostname}
\033[33m${TILDE[(1-0${PWD%%@([!/]*|${HOME}*)}1)]:-}${PWD#${HOME}}\033[0m\n$
')
so, you can still print the file w/o side effects
Regards,
Cyrille Lefevre
--
mailto:cyrille.lefevre-li...@laposte.net
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: 1.7.5: sshd environment variables
Le 03/05/2010 02:40, Larry Hall (Cygwin) a écrit :
On 5/2/2010 2:33 AM, Vincent Pelletier wrote:
Le dimanche 02 mai 2010 02:52:55, Larry Hall (Cygwin) a écrit :
You can add whatever variables you want and need to a script you can run
as part of the login or just after, depending on your needs.
So my question becomes:
Is it possible to get windows environment vars in a shell obtained via
ssh
(to avoid duplicating their definition in my .bashrc) ? Or is there a
way to
extend the set of vars sshd lets through to shell (I'm ok with having
to name
the variables I need, as long as I don't have to set their value
manually) ?
You can get the environment variables using 'env' (or 'set' in 'cmd.exe')
prior to invoking 'ssh'. You can then trim down the list to those that
you want. Putting those in a file that you can invoke shouldn't be hard.
around february 19, there was a discussion about to limit ssh
environment variables not including some windows ones !
since then, I use the following script to set the missing ones at
connection time...
you may comment the lines containing SECONDS.
it based on Kurt Franke's similar script in the idea.
Regards,
Cyrille Lefevre
--
mailto:cyrille.lefevre-li...@laposte.net
#!/bin/sh
#
# ssh-session-env.sh - script for installation in /etc/profile.d
#
# because in sessions started from sshd the windows system environment
# variables in general are not set except for some which are possible
# special handled (like PATH, etc.) and the windows user environment
# variables are not set from the actual user but from the user of the
# sshd server this script is used to build this environment settings in
# shells with bournish syntax which uses /etc/profile for initialization.
#
# authors: Kurt Franke, Cyrille Lefevre
#
# date: 06 march 2010
if ps -fp ${PPID} | grep -q /usr/sbin/sshd$; then
_SECONDS_=${SECONDS:-$(date +%s)}
_IFS_=${IFS}
IFS='
'
eval $( (
regtool -qv list /machine/SYSTEM/CurrentControlSet/Control/Session\
Manager/Environment
regtool -qv list /user/Environment
regtool -qv list /user/Volatile\ Environment
) |
awk -v q=' -v mp=$(mount --show-cygdrive-prefixes) '
function s2a(str, a, sep, i, t) {
if (sep == ) sep = +
split(str, t, sep)
for (i in t) a[t[i]] =
}
function uniqp(p, i, j, k, o, n, a, s) {
k = split(p, o, :)
for (i = j = 1; i = k; i++)
if (!(tolower(o[i]) in a))
a[tolower(n[j++] = o[i])] =
p = s =
for (i = 1; i j; i++) {
p = p s n[i]
s = :
}
return p
}
BEGIN {
s2a(APPDATA CLASSPATH QTJAVA LOCALAPPDATA USERPROFILE VS90COMNTOOLS, \
noconvert)
s2a(HOMEPATH PATH TEMP TMP, override)
sub(/.*\n/, , mp)
sub(/[ \t].*/, , mp)
}
! /\\ \(\)$/ {
var = toupper($1)
$1 = $2 = $3 =
sub(^ +, )
environ[var] = var == PATH var in environ ? \
environ[var] ; $0 : $0
}
END {
flag = 1
while (flag) {
flag = 0
for (var in environ) {
val = environ[var]
if (match(val,/%[^%]+%/)) {
flag = 1
subvar = substr(val, RSTART+1, RLENGTH-2)
subvar = toupper(subvar)
subval = subvar in environ ? \
environ[subvar] : ENVIRON[subvar]
if (subval !~ /%[^%]%/) {
head = substr(val, 1, RSTART-1)
tail = substr(val, RSTART+RLENGTH)
environ[var] = head subval tail
}
}
}
}
for (var in environ) {
if (!(var in override) var in ENVIRON)
continue
val = environ[var]
if (!(var in noconvert) (val ~ /;/ || val ~ /^.:/)) {
gsub(/([a-zA-Z]):/, mp /, val)
gsub(/:/,,val)
gsub(/\\/,/,val)
gsub(/;/,:,val)
}
if (var == PATH) val = uniqp(ENVIRON[var] : val)
gsub(q, \\q, val)
print export, var =q val q
}
}
')
echo elapsed: $(( ${SECONDS:-$(date +%s)} - $_SECONDS_ ))s
IFS=${_IFS_}
unset _IFS_ _SECONDS_
fi
# eof
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Protecting msmtp passwords
Le 27/04/2010 09:44, Gary . a écrit : Unfortunately when I use msmtp with gnus, it seems that msmtp decides it can't ask me for a password. Is it possible to save them in some file and have them protected somehow? try msmtp-config and/or read the msmtp manual page Regards, Cyrille Lefevre -- mailto:cyrille.lefevre-li...@laposte.net -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: vfork always fail problem
Huang, you - as the person who first saw and documented the problem in public (thank you!) - are in the best position to test it, if you can recreate the original situation (with the GBK(?) directory names). It would be beneficial to all of us, and you would be doing everybody a favour (if you wish, which is your choice), if you could test if the problem reproduces with Cygwin 1.7.5, and goes away with the snapshot. fork() now work well but find new problem after use the 0518 patch. Usually I use mintty to as the default term tool. After use this patch, I can't directly use mintty to open the first shell( like nothing happened), I must use Cygwin.bat to create a fist shell, then use short cut of mintty in the start menu to open others. I'll find out why if I have time. Thak you. Best -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: ncurses problem with mvcur function
On 5/19/2010 9:09 PM, Joe Java wrote: I have a very old game that uses ncurses. lines 602-603 are /* this moves curses to bottom right corner */ mvcur(curscr-_cury, curscr-_curx, LINES - 1, 0); lines 738-739 are the same as above. ncurses is now compiled with reentrant support, which had the unfortunate effect of changing the API slightly, by making the WINDOW object an opaque pointer. Instead of accessing the members of curscr directly, you use accessor functions: curscr-_cury --- getcury(curscr) curscr-_curx --- getcurx(cursrc) -- Chuck -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: vfork always fail problem
2010/5/20 Huang Bambo bambo.hu...@gmail.com: Huang, you - as the person who first saw and documented the problem in public (thank you!) - are in the best position to test it, if you can recreate the original situation (with the GBK(?) directory names). It would be beneficial to all of us, and you would be doing everybody a favour (if you wish, which is your choice), if you could test if the problem reproduces with Cygwin 1.7.5, and goes away with the snapshot. fork() now work well but find new problem after use the 0518 patch. Usually I use mintty to as the default term tool. After use this patch, I can't directly use mintty to open the first shell( like nothing happened), I must use Cygwin.bat to create a fist shell, then use short cut of mintty in the start menu to open others. I'll find out why if I have time. terminate while calling forkpty(). Thak you. Best -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Updated: {jasper/libjasper1/libjasper-devel}-1.900.1-10: JPEG 2000 library
Hi New versions of 'jasper/libjasper1/libjasper-devel' have been uploaded to a server near you. o Rebuild for cygwin 1.7 with gcc-4 CYGWIN-ANNOUNCE UNSUBSCRIBE INFO If you want to unsubscribe from the cygwin-announce mailing list, please use the automated form at: http://cygwin.com/lists.html#subscribe-unsubscribe If this does not work, then look at the List-Unsubscribe: tag in the email header of this message. Send email to the address specified there. It will be in the format: cygwin-announce-unsubscribe-you=yourdomain@cygwin.com If you need more information on unsubscribing, start reading here: http://sourceware.org/lists.html#unsubscribe-simple Please read *all* of the information on unsubscribing that is available starting at this URL.
Updated: {neon/libneon27/libneon-devel}-0.29.3-1: HTTP and WebDAV library
Hi New versions of 'neon/libneon27/libneon-devel' have been uploaded to a server near you. o Update to latest upstream neon NEWS: === * Change ne_sock_close() to no longer wait for SSL closure alert: - fixes possible hang with IIS servers when closing SSL connection - this reverts the behaviour with OpenSSL to match 0.28.x, and changes the behaviour with GnuTLS to match that with OpenSSL * Fix memory leak with GnuTLS * API clarification in ne_sock_close(): - SSL closure handling now documented - return value semantics fixed to describe the implementation CYGWIN-ANNOUNCE UNSUBSCRIBE INFO If you want to unsubscribe from the cygwin-announce mailing list, please use the automated form at: http://cygwin.com/lists.html#subscribe-unsubscribe If this does not work, then look at the List-Unsubscribe: tag in the email header of this message. Send email to the address specified there. It will be in the format: cygwin-announce-unsubscribe-you=yourdomain@cygwin.com If you need more information on unsubscribing, start reading here: http://sourceware.org/lists.html#unsubscribe-simple Please read *all* of the information on unsubscribing that is available starting at this URL.
