Re: [PATCH] inform user if any postinstall script failed to run
On Fri, 2010-08-27 at 18:15 +0100, Jon TURNEY wrote:
+ // Remove anything which we just tried to run (so we don't try
twice)
+ for (i = packages.begin (); i != packages.end (); ++i)
+{
+ packagemeta pkg = **i;
+ for (std::vectorScript::const_iterator j =
pkg.installed.scripts().begin();
+j != pkg.installed.scripts().end();
+j++)
+ {
+ std::vectorScript::iterator p = find(scripts.begin(),
scripts.end(), *j);
+ if (p != scripts.end())
+ {
+ scripts.erase(p);
+ }
+ }
+}
+
This code does not compile with i686-pc-mingw32 gcc-4.5.1:
postinstall.cc: In function ‘std::string
do_postinstall_thread(HINSTANCE__*, HWND__*)’:
postinstall.cc:178:85: error: no matching function for call to
‘find(std::vectorScript::iterator, std::vectorScript::iterator,
const Script)’
Yaakov
Re: [ITA] ocaml 3.12.0
On Mon, 2010-09-06 at 15:48 +0200, Damien Doligez wrote: 1. The tcl/tk bindings: they don't work out of the box with Cygwin's tcl/tk, and I don't think there is much demand anyway. Interesting, given that OCaml presumes that Cygwin's tcl/tk is Win32 (which it currently is), but Ports' X11 tcl/tk works with OCaml with some patches: http://cygwin-ports.git.sourceforge.net/git/gitweb.cgi?p=cygwin-ports/ports;a=tree;f=lang/ocaml Given the recent issues on the list, I think it's about time I ITA tcl/tk. 2. Dynamic linking: this needs FlexDLL http://alain.frisch.fr/flexdll/ , which isn't packaged for cygwin. I guess I'll do a cygwin package of FlexDLL later, when I have a bit more experience. I have a flexdll package in Ports which includes several patches necessary for Cygwin, which I used to build OCaml with dynlink support: http://cygwin-ports.git.sourceforge.net/git/gitweb.cgi?p=cygwin-ports/ports;a=tree;f=devel/flexdll I will be busy for the remainder of the week, but let's work on this further next week. Yaakov
Re: [PATCH] inform user if any postinstall script failed to run
On 07/09/2010 10:45, Yaakov (Cygwin/X) wrote:
On Fri, 2010-08-27 at 18:15 +0100, Jon TURNEY wrote:
+ // Remove anything which we just tried to run (so we don't try
twice)
+ for (i = packages.begin (); i != packages.end (); ++i)
+{
+ packagemeta pkg = **i;
+ for (std::vectorScript::const_iterator j =
pkg.installed.scripts().begin();
+j != pkg.installed.scripts().end();
+j++)
+ {
+ std::vectorScript::iterator p = find(scripts.begin(),
scripts.end(), *j);
+ if (p != scripts.end())
+ {
+ scripts.erase(p);
+ }
+ }
+}
+
This code does not compile with i686-pc-mingw32 gcc-4.5.1:
postinstall.cc: In function ‘std::string
do_postinstall_thread(HINSTANCE__*, HWND__*)’:
postinstall.cc:178:85: error: no matching function for call to
‘find(std::vectorScript::iterator, std::vectorScript::iterator,
const Script)’
Hmm... perhaps a '#include algorithm' is needed?
Re: [ITA] ocaml 3.12.0
On Tue, Sep 07, 2010 at 05:03:13AM -0500, Yaakov (Cygwin/X) wrote: On Mon, 2010-09-06 at 15:48 +0200, Damien Doligez wrote: 1. The tcl/tk bindings: they don't work out of the box with Cygwin's tcl/tk, and I don't think there is much demand anyway. Interesting, given that OCaml presumes that Cygwin's tcl/tk is Win32 (which it currently is), but Ports' X11 tcl/tk works with OCaml with some patches: http://cygwin-ports.git.sourceforge.net/git/gitweb.cgi?p=cygwin-ports/ports;a=tree;f=lang/ocaml Given the recent issues on the list, I think it's about time I ITA tcl/tk. I know that it isn't obvious but I really am still here and that sorta means that you can't take tcl/tk away just yet. I have been trying to talk to the insight maintainer about what to do with tcl/tk but he hasn't been really responsive and I haven't been extremely motivated to push the issue. I actually have had a native cygwin version of tcl ready for quite some time but insight really messes things up. cgf
Re: RFU: googlecl-0.9.10-1
On Sep 6 13:29, Chris Sutcliffe wrote: Please upload: --- wget -x -nH --cut-dirs=1 \ http://emergedesktop.org/cygwin/googlecl/googlecl-0.9.10-1.tar.bz2 \ http://emergedesktop.org/cygwin/googlecl/googlecl-0.9.10-1-src.tar.bz2 Done. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat
Re: [ITA] ocaml 3.12.0
On 9/7/2010 6:03 AM, Yaakov (Cygwin/X) wrote: Given the recent issues on the list, I think it's about time I ITA tcl/tk. More power to you, but I don't think cgf has gone anywhere...so tcltk isn't yet orphaned. -- Chuck
Updated: googlecl-0.9.10-1
Version 0.9.10-1 of googlecl has been uploaded. GoogleCL brings Google services to the command line. For examples see: http://code.google.com/p/googlecl/wiki/ExampleScripts Change include: * v2/v3 support for Docs and Contacts. Manipulate arbitrary uploads, list many more details of your contacts. * Support for non-latin alphabets * Adheres to XDG base directory specification *** CYGWIN-ANNOUNCE UNSUBSCRIBE INFO *** If you want to unsubscribe from the cygwin-announce mailing list, 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. -- Chris Sutcliffe http://emergedesktop.org http://www.google.com/profiles/ir0nh34d
Re: [PATCH] inform user if any postinstall script failed to run
On Tue, 2010-09-07 at 15:01 +0100, Jon TURNEY wrote: On 07/09/2010 10:45, Yaakov (Cygwin/X) wrote: This code does not compile with i686-pc-mingw32 gcc-4.5.1: postinstall.cc: In function ‘std::string do_postinstall_thread(HINSTANCE__*, HWND__*)’: postinstall.cc:178:85: error: no matching function for call to ‘find(std::vectorScript::iterator, std::vectorScript::iterator, const Script)’ Hmm... perhaps a '#include algorithm' is needed? Yep, that fixes it. Do you want the honours of committing the fix? Yaakov
Re: [ITA] ocaml 3.12.0
Am 07.09.2010, 16:42 Uhr, schrieb Christopher Faylor: On Tue, Sep 07, 2010 at 05:03:13AM -0500, Yaakov (Cygwin/X) wrote: On Mon, 2010-09-06 at 15:48 +0200, Damien Doligez wrote: 1. The tcl/tk bindings: they don't work out of the box with Cygwin's tcl/tk, and I don't think there is much demand anyway. Interesting, given that OCaml presumes that Cygwin's tcl/tk is Win32 (which it currently is), but Ports' X11 tcl/tk works with OCaml with some patches: http://cygwin-ports.git.sourceforge.net/git/gitweb.cgi?p=cygwin-ports/ports;a=tree;f=lang/ocaml Given the recent issues on the list, I think it's about time I ITA tcl/tk. I know that it isn't obvious but I really am still here and that sorta means that you can't take tcl/tk away just yet. I have been trying to talk to the insight maintainer about what to do with tcl/tk but he hasn't been really responsive and I haven't been extremely motivated to push the issue. Insight is dead for all practical purposes, but Tcl/Tk in Cygwin depending on X11 rather than Win32 would be major regression. Not that I'd have time to help though. -- Matthias Andree
Re: [PATCH] inform user if any postinstall script failed to run
On 07/09/2010 17:57, Yaakov (Cygwin/X) wrote: On Tue, 2010-09-07 at 15:01 +0100, Jon TURNEY wrote: On 07/09/2010 10:45, Yaakov (Cygwin/X) wrote: This code does not compile with i686-pc-mingw32 gcc-4.5.1: postinstall.cc: In function ‘std::string do_postinstall_thread(HINSTANCE__*, HWND__*)’: postinstall.cc:178:85: error: no matching function for call to ‘find(std::vectorScript::iterator, std::vectorScript::iterator, const Script)’ Hmm... perhaps a '#includealgorithm' is needed? Yep, that fixes it. Do you want the honours of committing the fix? No, I don't have that compiler installed to test the fix, please go ahead.
Re: [ITA] ocaml 3.12.0
On Tue, Sep 07, 2010 at 08:40:43PM +0200, Matthias Andree wrote: Am 07.09.2010, 16:42 Uhr, schrieb Christopher Faylor: On Tue, Sep 07, 2010 at 05:03:13AM -0500, Yaakov (Cygwin/X) wrote: On Mon, 2010-09-06 at 15:48 +0200, Damien Doligez wrote: 1. The tcl/tk bindings: they don't work out of the box with Cygwin's tcl/tk, and I don't think there is much demand anyway. Interesting, given that OCaml presumes that Cygwin's tcl/tk is Win32 (which it currently is), but Ports' X11 tcl/tk works with OCaml with some patches: http://cygwin-ports.git.sourceforge.net/git/gitweb.cgi?p=cygwin-ports/ports;a=tree;f=lang/ocaml Given the recent issues on the list, I think it's about time I ITA tcl/tk. I know that it isn't obvious but I really am still here and that sorta means that you can't take tcl/tk away just yet. I have been trying to talk to the insight maintainer about what to do with tcl/tk but he hasn't been really responsive and I haven't been extremely motivated to push the issue. Insight is dead for all practical purposes, but Tcl/Tk in Cygwin depending on X11 rather than Win32 would be major regression. Not that I'd have time to help though. This has already been decided. I was just waiting for some sign of life from my friend the insight maintainer. I guess I'll go ahead and pull insight from the release. That should make things easier. cgf
Re: [ITA] ocaml 3.12.0
On Tue, 2010-09-07 at 14:55 -0400, Christopher Faylor wrote: This has already been decided. I was just waiting for some sign of life from my friend the insight maintainer. I guess I'll go ahead and pull insight from the release. That should make things easier. FWIW, Debian, Fedora, and Gentoo all ship gdb and insight based on separate source packages (using the latest official releases of each). Packaging a TUI-only gdb-7.2 now would therefore be the first step towards that. I was able to build a separate insight CVS snapshot against my X11 tcl/tk with some patches: http://cygwin-ports.git.sourceforge.net/git/gitweb.cgi?p=cygwin-ports/ports;a=tree;f=devel/insight HTH, Yaakov
Re: [ITA] ocaml 3.12.0
On 2010-09-06, at 16:43, Charles Wilson wrote: Are you sure that your new version requires libncurses7? The only current libncurses-dev package will cause you to link against libncurses10... Try cygcheck /usr/bin/name-of-ocaml-exe and see what DLLs it actually uses... Also, new policy is NOT to list 'cygwin' in the requires... You're absolutely right. I have fixed both problems and updated some stale information in the README file. The new version is here: wget http://yquem.inria.fr/~doligez/cygwin/ocaml-3.12.0-2-src.tar.bz2 wget http://yquem.inria.fr/~doligez/cygwin/ocaml-3.12.0-2.tar.bz2 -- Damien
[ITP] flexdll
FlexDLL creates binaries whose symbols can be resolved at runtime, despite the limitations of the PE format. It is required by OCaml for dynamic linking support on Cygwin. This package contains patches to fix data auto-imports, shared libgcc, and to remove some Win32-isms: ftp://ftp.cygwinports.org/pub/cygwinports/release-2/OCaml/flexdll/ FlexDLL itself is written in OCaml but is compiled into native code, so it needs an OCaml (with or without dynlink support) at build time only. Due to its unique nature, it is obviously not in any *NIX distro, so I suppose this needs to be voted on. Yaakov
Re: [ITP] flexdll
On Tue, Sep 07, 2010 at 07:06:02PM -0500, Yaakov (Cygwin/X) wrote: FlexDLL creates binaries whose symbols can be resolved at runtime, despite the limitations of the PE format. It is required by OCaml for dynamic linking support on Cygwin. This package contains patches to fix data auto-imports, shared libgcc, and to remove some Win32-isms: ftp://ftp.cygwinports.org/pub/cygwinports/release-2/OCaml/flexdll/ FlexDLL itself is written in OCaml but is compiled into native code, so it needs an OCaml (with or without dynlink support) at build time only. Due to its unique nature, it is obviously not in any *NIX distro, so I suppose this needs to be voted on. I think it can go in transitively if OCaml needs it. cgf
Re: [ITA] ocaml 3.12.0
On 9/7/2010 2:55 PM, Christopher Faylor wrote: On Tue, Sep 07, 2010 at 08:40:43PM +0200, Matthias Andree wrote: Insight is dead for all practical purposes, but Tcl/Tk in Cygwin depending on X11 rather than Win32 would be major regression. Not that I'd have time to help though. This has already been decided. Maybe this is a dumb question, but I've been paying pretty close attention and this statement confuses me. What /exactly/ was the decision that was reached? (I *think* I know, I *believe* I know, but cgf is the maintainer of tcl/tk, not to mention his position wrt to cygwin as a whole -- so he pretty much gets final say about tcl/tk, regardless of what I or anybody else might say. So...) -- Chuck
Re: XServer draws to incorrect window when using VirtuaWin
On 03/09/2010 10:01, Pete wrote: On 13 August 2010 11:38, Petesneakypet...@gmail.com wrote: VirtuaWin (http://virtuawin.sourceforge.net/) is a virtual desktop manager for Windows that lets you switch between several virtual desktops, similar to those provided in KDE Gnome. When switching between desktops that have CygwinX windows open, occasionally the Xserver draws to the wrong window. This is difficult to describe, so will continue with an example: I'm afraid I think this falls into the category of known problems, see [1]. The tricks that XWin uses to implement multiwindow mode don't seem to be compatible with the tricks that VirtuaWin uses to implement multiple desktops. [1] https://bugs.freedesktop.org/show_bug.cgi?id=21540 Using Windows 7, Cygwin/X v1.8.0 Steps to reproduce: 1) Install VirtuaWin from http://virtuawin.sourceforge.net/ 2) Start the CygwinX server 3) Open a (DOS) cygwin window 4) Type xterm twice, to open two xterm windows. Maximise these two windows to full screen. 5) Move one of these windows to desktop2 6) Type ping google.com -n 1000 to get a stream of data appearing in the xterm window on desktop2 7) Go back to desktop1, and make sure the DOS cygwin window is selected 8) Switch back to desktop2. The ping xterm window should be selected. 9) Switch back to desktop1. The cygwin window should be selected. What should happen: The empty xterm session on desktop1 should be displayed in the window behind the cygwin window What happens: The ping data stream appears in the xterm window on desktop1, and continues receiving updates every second. Selecting the xterm window causes the ping data to disappear and the empty xterm session to be displayed correctly. This is reproducible every time. The critical thing is that if the xterm window on desktop1 is not selected after a desktop switch, it shows the data from the xterm window on desktop 2. FWIW, this problem doesn't exist with xming, and I haven't seen the issue with any other applications. Xming version number, please? Is there anything else I can do to help debug this? Unfortunately it is necessary to install VirtuaWin to reproduce the issue, but it's very reproducible. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- 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/
Re: SIGSEGV in xorg-1.8.2.0 during -resize operation
On 04/09/2010 00:10, Ryan Johnson wrote: On 8/31/2010 7:00 PM, Jon TURNEY wrote: Okay, I think I have worked out the correct thing to do do to handle bpp changes in the RANDR code, and I've uploaded a test build at [1]. Perhaps you could try it and see if it works for you? Note that you will need to use -resize with this build to turn on RANDR in any mode. If you can make this crash, with or without -resize, a backtrace would be very helpful. [1] ftp://cygwin.com/pub/cygwinx/XWin.20100831-git-5fa9c90425fb1d68.exe.bz2 So far so good... I've switched monitors several times without problems (both with -resize and without). Good news, thanks for testing! :-) -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- 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/
Re: XServer draws to incorrect window when using VirtuaWin
On 7 September 2010 14:38, Jon TURNEY jon.tur...@dronecode.org.uk wrote: On 03/09/2010 10:01, Pete wrote: On 13 August 2010 11:38, Petesneakypet...@gmail.com wrote: VirtuaWin (http://virtuawin.sourceforge.net/) is a virtual desktop manager for Windows that lets you switch between several virtual desktops, similar to those provided in KDE Gnome. When switching between desktops that have CygwinX windows open, occasionally the Xserver draws to the wrong window. This is difficult to describe, so will continue with an example: I'm afraid I think this falls into the category of known problems, see [1]. The tricks that XWin uses to implement multiwindow mode don't seem to be compatible with the tricks that VirtuaWin uses to implement multiple desktops. [1] https://bugs.freedesktop.org/show_bug.cgi?id=21540 Thanks for your reply. You're right, this is the same issue. I'll update the bug. Using Windows 7, Cygwin/X v1.8.0 Steps to reproduce: 1) Install VirtuaWin from http://virtuawin.sourceforge.net/ 2) Start the CygwinX server 3) Open a (DOS) cygwin window 4) Type xterm twice, to open two xterm windows. Maximise these two windows to full screen. 5) Move one of these windows to desktop2 6) Type ping google.com -n 1000 to get a stream of data appearing in the xterm window on desktop2 7) Go back to desktop1, and make sure the DOS cygwin window is selected 8) Switch back to desktop2. The ping xterm window should be selected. 9) Switch back to desktop1. The cygwin window should be selected. What should happen: The empty xterm session on desktop1 should be displayed in the window behind the cygwin window What happens: The ping data stream appears in the xterm window on desktop1, and continues receiving updates every second. Selecting the xterm window causes the ping data to disappear and the empty xterm session to be displayed correctly. This is reproducible every time. The critical thing is that if the xterm window on desktop1 is not selected after a desktop switch, it shows the data from the xterm window on desktop 2. FWIW, this problem doesn't exist with xming, and I haven't seen the issue with any other applications. Xming version number, please? It was 6.9.0.31. -- 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/
winsup/cygwin ChangeLog
CVSROOT:/cvs/uberbaum Module name:winsup Changes by: c...@sourceware.org 2010-09-07 15:13:23 Modified files: cygwin : ChangeLog Log message: fix two typos Patches: http://sourceware.org/cgi-bin/cvsweb.cgi/winsup/cygwin/ChangeLog.diff?cvsroot=uberbaumr1=1.5026r2=1.5027
Re: Cygwin slow on x64 systems
On Tue, Sep 07, 2010 at 01:54:14AM -0400, Christopher Faylor wrote: On Thu, Sep 02, 2010 at 01:10:56AM -0400, Christopher Faylor wrote: On Wed, Sep 01, 2010 at 10:28:35PM -0500, Yaakov (Cygwin/X) wrote: On Wed, 2010-09-01 at 17:20 -0400, Christopher Faylor wrote: On Wed, Sep 01, 2010 at 03:19:15PM -0500, Heath Kehoe wrote: If I put the original cygwin1.dll (1.7.7) back, everything works again. I also have sources, and built the latest from CVS, and that cygwin1.dll fails in the same way. Sorry about that. It should be fixed now. http://cygwin.com/snapshots/ Services refuse to start with the 20100901 snapshot. fork() does seem to be faster, though (Win7 x64 RTM). sshd WJFFM, on XP at least. ...but I can duplicated it on Vista 64. I'll fix it tomorrow. duplicate -- 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: permission denied when removing files on remote file share that is offline
On Sep 6 18:34, Thielemans, Kris wrote: On Sep 6 15:33, Thielemans, Kris wrote: I'm using a file share with Offline Folders such that I can access it when not connected to our intranet. In cygwin I cannot remove any files that I create on the share when it's offline: $ touch /xxx/testfile $ rm /xxx/testfile rm: cannot remove `/xxx/testfile': Permission denied Strange. I cannot reproduce this problem. [...] Can you run rm under strace? Maybe there's a status code which gives some helpful information. Look for the unlink_nt function. --- I did the following $ strace rm /h/TEMP/testfile /h/TEMP/rm.log rm: cannot remove `/h/TEMP/testfile': Permission denied The few lines around unlink_nt are as follows (full log file attached): 87 230535 [main] rm 5200 path_conv::check: this-path(H:\TEMP\testfile), has_acls(0) 1107 231642 [main] rm 5200 unlink_nt: Setting delete disposition failed, status = 0xC022 117 231759 [main] rm 5200 seterrno_from_nt_status: /ext/build/netrel/src/cygwin-1.7.7-1/winsup/cygwin/syscalls.cc:714 status 0xC022 - windows error 5 94 231853 [main] rm 5200 geterrno_from_win_error: windows error 5 == errno 13 77 231930 [main] rm 5200 unlink: -1 = unlink (/h/TEMP/testfile) Does this help? Not really. In the default case, the unlink_nt function just opens the file for deletion and then sets the delete disposition for the file. That's pretty much all what happens. The above snippet shows that the OS function to set the delete disposition fails with status 0xc022, which is STATUS_ACCESS_DENIED. I had hoped for some other status code, but, no luck here. There's no hint why this occurs. I examined what happens in my case using sysinternal's process monitor. Windows Explorer uses the same way to delete the file(*) as Cygwin does. Maybe process monitor helps to figure out what happens on your machine but I doubt it. I'm out of ideas, sorry. Corinna (*) In case you wonder. This is Windows. There are three ways to delete a file, not just one. -- 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
TCL exec kills its owning tty (was: Unable to initialize fd 0 for /dev/tty1)
On 9/6/2010 11:34 PM, Ryan Johnson wrote: The last couple days I've gotten some really strange errors. Sometimes I'll alt-tab to an xterm window and start typing, only to have it disappear at the first keystroke. Sometimes `emacs -nw' will get hit instead, leaving both emacs and the xterm pretty much unusable. I haven't been able to get a stack trace I can reproduce this bug reliably with both 1.7.7 and 1.7.6 versions of cygwin1.dll (which makes me suspect the problem is not there). The reason I couldn't get a stack trace is because the xterm doesn't receive an error... it just gets an EOF and dutifully exits. Sometimes you can see 'exit' appear just before the window closes. After some digging, it looks like TCL is the culprit (recall I was running tkdiff). Here's a minimal test case, which will close its controlling tty when it exits: $ echo 'exec ls' tty-killer; tclsh tty-killer Running the script in the foreground, passing it as a -c argument to bash, or piping it into tclsh, does *not* expose the bug. Ideas? Ryan -- 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: TCL exec kills its owning tty (was: Unable to initialize fd 0 for /dev/tty1)
On Sep 7 11:22, Ryan Johnson wrote: On 9/6/2010 11:34 PM, Ryan Johnson wrote: The last couple days I've gotten some really strange errors. Sometimes I'll alt-tab to an xterm window and start typing, only to have it disappear at the first keystroke. Sometimes `emacs -nw' will get hit instead, leaving both emacs and the xterm pretty much unusable. I haven't been able to get a stack trace I can reproduce this bug reliably with both 1.7.7 and 1.7.6 versions of cygwin1.dll (which makes me suspect the problem is not there). Actually, this is a Cygwin bug. I can reproduce it and I can only reproduce it starting with Cygwin 1.7.6, which introduced a change in the way ttys are handled under the hood. Apparently I managed to break something along the line. $ echo 'exec ls' tty-killer; tclsh tty-killer Running the script in the foreground, passing it as a -c argument to bash, or piping it into tclsh, does *not* expose the bug. Unfortunately it also doesn't happen under strace. So, still digging for the culprit. Stay tuned... 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
Re: OpenGL linking problems
On 04/09/2010 17:29, André Bleau wrote: On 05/09/2010 00:16, David Doria wrote: Well, we need more info to help us help you. What kind of application are you trying to build? X11 or native? There are 2 ways to build OpenGL-related programs: 1- Using native OpenGL, with driver supported hardware acceleration. For that, you need some way to get a graphic device context. One portable way to do that is to use the GLUT library, provided by the opengl package that I maintain. Look at the doc in /usr/share/doc/opengl-1.1.0 to learn how to do that. The OpenGL implementation in that case is provided by Window's opengl32.dll. Cygwin's w32api package provides the headers and link library for that version of OpenGL. LD_LIBRARY_PATH has no effect if you go that way. 2- Using X11, which provides an opengl implementation. At one point, there was support for hardware acceleration; then, it went away; I don't know if it came back or if it is coming back soon. I will let the maintainer of the libGL* packages answer questions you may have about it. You would better send them to cygwin-xfree at cygwin dot com . [1] discusses the differences between using OpenGL with WGL or with GLX. AIGLX should be back soon, indeed :-) I've had precisely zero feedback on the test releases I've done, so I assume nobody is very interested in this feature, and it's very 2-steps-forward-1-step-back keeping the necessary patches to mesa libGL working, but hopefully it will make an appearance in Xserver 1.9 [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-opengl-link-problems - André Bleau, Cygwin's opengl package maintainer. Hi André, I think I'm trying to build an X11 application - I'm trying to use a library called VTK (http://www.vtk.org/). I don't believe it uses glut, and I do see a bunch of X11 options in the CMake files. Even if there is no support for hardware acceleration with this method, it should still build, right? I do have C:\Windows\System32\opengl32.dll . I should also have the w32api package because I installed all of the cygwin packages. Any other thoughts of things that would causes these linking errors? Since you don't give the link line which is causing the problem, I'd tend to assume the problem is [2] [2] http://cygwin.com/faq/faq-nochunks.html#faq.programming.linker -- 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 with mouse events in XWin server
Dear all, At work I use my portable as X server for a remote Linux machine. I can use two brands of software, Reflection X, which is installed by our IT department, or Cygwin XWin, which I install my self. The application which I need to access is Continuus, which is Motif based. What is now my problem? When I start Continuus using Reflection, I am missing some fonts. I think it are the TTF and Type1 fonts that are missing, but Reflection cannot handle them. When I start XWin using 'startxwin', I can connect to Continuus and I get my missing fonts. However, there is something else that is now missing. When one right-clicks on a screen element in Continuus, then a context dependent menu should pop-up. However, with XWin this does not happen. I used 'xev' to see the differences in mouse events, and in XWin I see the following : ButtonPress event, serial 21, synthetic NO, window 0x21, root 0x44, subw 0x0, time 27589328, (123,42), root:(185,627), state 0x10, button 3, same_screen YES while using Reflection X I see ButtonPress event, serial 22, synthetic NO, window 0x41, root 0x3b, subw 0x0, time 16229, (132,53), root:(184,187), state 0x0, button 3, same_screen YES Btw. I had the same problem with our Sun server. Other parts of the handling of mouse events also differ. Reflection X only reports the ButtonPress and ButtonRelease events, while XWin also leaves a MotionNotify event upon releasing the button. Is there any way I can configure or influence this? I looked at xmodmap, but that seems to be only concerned with keyboard mappings. Regards, Jurgen -- 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
likely stupid question but name conflicts with foo and foo.exe?
Hi, I'm trying to copy of bunch of files from debian over to windoze using scp -r. This was working fine except for one error where it complains foo is not a directory and apparently I have foo and foo.exe in one dir but foo is a directory. I think this replicates the problem. It seems that .exe is confused with no extention. Is this true and what can I do or am I just doing something really stupid? Thanks. $ mkdir acrap mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ echo x1 mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ echo x1.exe mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ cp x1* acrap cp: will not overwrite just-created `acrap/x1.exe' with `x1.exe' mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ ls acrap x1.exe mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ mv x1.exe x1.fexe mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ cp x1* acrap cp: cannot create regular file `acrap/x1': File exists mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ ls -al acrap total 6 drwxr-xr-x+ 1 mmarchywka None0 2010-09-07 08:56 . drwxr-xr-x+ 1 mmarchywka None 4096 2010-09-07 08:56 .. -rw-r--r--+ 1 mmarchywka None1 2010-09-07 08:51 x1.exe -rw-r--r--+ 1 mmarchywka None1 2010-09-07 08:56 x1.fexe mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ rm acrap/* mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ cp x1* acrap mmarchy...@phlulap01 /cygdrive/c/old/download/cdarchive $ -- 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: Problem with mouse events in XWin server
On 07/09/2010 14:31, Jurgen Defurne wrote: At work I use my portable as X server for a remote Linux machine. I can use two brands of software, Reflection X, which is installed by our IT department, or Cygwin XWin, which I install my self. The application which I need to access is Continuus, which is Motif based. What is now my problem? When I start Continuus using Reflection, I am missing some fonts. I think it are the TTF and Type1 fonts that are missing, but Reflection cannot handle them. When I start XWin using 'startxwin', I can connect to Continuus and I get my missing fonts. However, there is something else that is now missing. When one right-clicks on a screen element in Continuus, then a context dependent menu should pop-up. However, with XWin this does not happen. I used 'xev' to see the differences in mouse events, and in XWin I see the following : ButtonPress event, serial 21, synthetic NO, window 0x21, root 0x44, subw 0x0, time 27589328, (123,42), root:(185,627), state 0x10, button 3, same_screen YES while using Reflection X I see ButtonPress event, serial 22, synthetic NO, window 0x41, root 0x3b, subw 0x0, time 16229, (132,53), root:(184,187), state 0x0, button 3, same_screen YES X related questions belong on the cygwin-xfree list. I think you have num-lock on. See [1]. [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#numlock-modifier -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- 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: Problem with mouse events in XWin server
On 9/7/2010 9:31 AM, Jurgen Defurne wrote: Dear all, At work I use my portable as X server for a remote Linux machine. I can use two brands of software, Reflection X, which is installed by our IT department, or Cygwin XWin, which I install my self. Since this is a Cygwin-X issue, please send it to the cygwin-xfree list. That's the list to handle X issues/questions with Cygwin (see http://cygwin.com/lists.html). -- 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? -- 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
incredibly slow file listing script on windoze 7 pro 4 core 64 bit
this takes a few minutes on old debian machine, taking much longer
here on same data-
about 19k file about 24Gb total size.
Windoze finally has better perf stuff but still no help- one core at
25 pct all kernel
time disk not exactly busy. All the time is in the ls loop not the
find command.
Now obbviously I expect the ls per file has to make a bunch of OS calls for
each file but still even with cygwin layer seems a bit much.
Thanks.
$ more tots
foo=
#ls -al `find -type f| awk '{print \$0}' ` | awk '{tot=tot+\$5; }END{print tot}
'
#ls -al `find -type f ` | awk '{tot=tot+$5; }END{print tot}'
find -type f xxx
cat xxx | while read
do
ls -al $REPLY
done | awk '{tot=tot+$5; }END{print tot}'
--
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: 1.7.7: Cannot unmount certain user bind mounts
On Sep 04 01:50 Corinna Vinschen wrote: On Sep 3 17:19, John Carey wrote: A user mount whose only non-default option is bind cannot be unmounted if its target is a system mount; please see the end of this email for a test case. It looks to me as if the MOUNT_SYSTEM bit is copied from the bind target by mount() in winsup/cygwin/mount.cc. Right. I fixed this in CVS. I also fixed the problem that bind mounts in the user's fstab file were preserving the targets MOUNT_SYSTEM flag. And while I was at it, I found that `mount -a' also adds the MOUNT_SYSTEM flag if the added mount is from /etc/fstab. That wasn't intended either. Every mount added via mount(2) is supposed to be a user mount. Thanks for the fix! -- John -- 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: can't compile setup.exe
On Thu, 2010-09-02 at 14:38 +0100, Jon TURNEY wrote: This was broken by the recent w32api-3.15 update, which seems to have made those PropSheet macros C++ aware, so the global scoping operator is no longer needed. Patch attached to fix it, but I couldn't work out how to also get it to build with w32api-3.14. Ping? Yaakov -- 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
Portable Cygwin 1.5 compiled using XP fails to run on W7 64
Dear All, Subject says all. Could a portable version of Cygwin 1.5 mounted on a USB stick reasonably be expected to run on W7 64? I know there are recommendations but can't remember outright prohibitions or constraints. If not, I would have expected dumb failure. But actually we are getting: grasp denied (actually Zugriff verweigert, it's a German machine) /cygdrive/f/bin/sed: can`t read c:/cygmnts.log: No such file or directory I've never heard of grasp (or Zugriff) and I cannot remember ever before having a requirement for c:/cygmnts.log. Thanks. I've got another stick with Cygwin 1.7 on it and am not expecting problems (but haven't yet tried) running it on the W7 64 machine. Fergus -- 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: ssh access violation [was: ssh simply prints Aborted]
End of story: *From:* gonzalo diethelm *Sent:* Monday, September 06, 2010 5:13 PM I downgraded openssh to v5.5p1-2 and now it is working again! *From:* Letbetter, Jason That fixed my problem too!!! Thanks a lot Gonzalo, I really appreciate you keeping me in the loop and helping to resolve this issue. -- René Berber -- 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: incredibly slow file listing script on windoze 7 pro 4 core 64 bit
On 9/7/2010 12:05 PM, mike marchywka wrote: this takes a few minutes on old debian machine, taking much longer here on same data- about 19k file about 24Gb total size. Windoze finally has better perf stuff but still no help- one core at 25 pct all kernel time disk not exactly busy. All the time is in the ls loop not the find command. Now obbviously I expect the ls per file has to make a bunch of OS calls for each file but still even with cygwin layer seems a bit much. Thanks. The problem is that Cygwin has to open all the files to stat when you use the '-l' flag. Try without the flag to get the lower bound and see if that's better for you. If so, maybe '-s' will suit your purpose. Or even 'du'? -- 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? -- 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: can't compile setup.exe
On 07/09/2010 18:15, Yaakov (Cygwin/X) wrote: On Thu, 2010-09-02 at 14:38 +0100, Jon TURNEY wrote: This was broken by the recent w32api-3.15 update, which seems to have made those PropSheet macros C++ aware, so the global scoping operator is no longer needed. Patch attached to fix it, but I couldn't work out how to also get it to build with w32api-3.14. Ping? How about adding the following to my previous patch? That at least turns the OPs problem into something intelligible by mere mortals :-) #include w32api.h #if (__W32API_MAJOR_VERSION*100 + __W32API_MINOR_VERSION) 315 #error w32api3.15 or later is needed #endif -- 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: Portable Cygwin 1.5 compiled using XP fails to run on W7 64
On Tue, Sep 07, 2010 at 06:30:44PM +0100, Fergus wrote: Dear All, Subject says all. Could a portable version of Cygwin 1.5 mounted on a USB stick reasonably be expected to run on W7 64? No, it is not expected to run. cgf -- 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: can't compile setup.exe
On Tue, 2010-09-07 at 19:50 +0100, Jon TURNEY wrote: How about adding the following to my previous patch? That at least turns the OPs problem into something intelligible by mere mortals :-) #include w32api.h #if (__W32API_MAJOR_VERSION*100 + __W32API_MINOR_VERSION) 315 #error w32api3.15 or later is needed #endif Is this necessary? Is there any reason we can't just require the latest w32api and mingw-runtime in order to build setup.exe? Yaakov -- 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
Quad precision floating point package for gfortran
We are developing a quad precision floating point library for use with gfortran. When attempting to build the library under Cygwin, we are stopped because: __float128 is not supported when running the ./configure for the package. I suspect the issue may be related to newlib vs glibc, but I really don't know. Do any of the Cygwin maintainers have any hints about this? I am still investigating. It would be nice to have this quad precision capability under Cygwin. The library is a software floating point library and does not depend on having special hardware. We have it running on i686 and x86-64 Linux, but so far no luck under Windows. If I learn more in the next few days, I will report here. Any suggestions would be appreciated. Best regards, Jerry -- 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
