the big switch?
I've just returned from a week's holiday, so I haven't read all traffic; but I fail to find information on the big abi switch. However, it seems that 1.5.3 is now [curr]ent, but a lot of TEST/1.5.x packages do not seem to be switched? I assume this, eg, will break badly. Is this an oversight, or anything required from package maintainers? @ cygwin sdesc: The UNIX emulation engine category: Base requires: libintl2 _update-info-dir base-passwd version: 1.5.3-1 [..] @ tetex-bin sdesc: The TeX text formatting system (binaries). [..] version: 2.0.2-2 install: release/tetex/tetex-bin/tetex-bin-2.0.2-2.tar.bz2 2917523 cfc223e930bb1d83fbbfbcdf875967e4 [test] version: 2.0.2-12 -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
Re: the big switch?
On Mon, Sep 08, 2003 at 03:05:02PM +0200, Jan Nieuwenhuizen wrote: I've just returned from a week's holiday, so I haven't read all traffic; but I fail to find information on the big abi switch. However, it seems that 1.5.3 is now [curr]ent, but a lot of TEST/1.5.x packages do not seem to be switched? I assume this, eg, will break badly. Older applications work FINE with 1.5.x. I think I need to change my signature. Is this an oversight, or anything required from package maintainers? There is nothing new. If the package maintainers want to change their packages from test to current, they have to notify us here and someone will flip the switch. cgf
[ITP] GMP NTL libraries
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I didn't even try to compile them, yet, but being mathematical packages I fear not they may not compile. So I'm beginning to ask the list if there is interest in those two number theory / big integer libraries. http://www.swox.com/gmp/ http://www.shoup.net/ntl/ (I think I'll compile NTL in the use GMP for faster base arithmetic method so NTL will depend upon GMP, while GMP should depend on nothing else) - -- Lapo 'Raist' Luchini [EMAIL PROTECTED] (PGP X.509 keys available) http://www.lapo.it (ICQ UIN: 529796) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (Cygwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAj9chP4ACgkQaJiCLMjyUvuw1QCgndU8n7rJAV/69SC8fV1Ux6zy 664AoIq27KzVm8McJOzPe8xnWY7i8Te9 =MJ4X -END PGP SIGNATURE-
PING! PING! PING! [setup PATCH] Rework do_download [next_dialog removal (2b)]
PING! PING! PING!
Original Message:
http://sources.redhat.com/ml/cygwin-apps/2003-07/msg00695.html
First Ping:
http://sources.redhat.com/ml/cygwin-apps/2003-08/msg00047.html
Second Ping:
http://sources.redhat.com/ml/cygwin-apps/2003-08/msg00219.html
Robert, did you miss this one?
I ask because it should be fairly simple to review, being an application of
the general pronciples already approved in 2a to another region of code.
If it is queued waiting for you to have time, please let me know.
After this, there is only one more trivial commit to actually remove the
definition of next_dialog itself.
Max.
+2003-07-30 Max Bowsher [EMAIL PROTECTED]
+
+ * download.cc (do_download_thread): Return bool. Move program flow
+ logic to threebar.cc.
+ (do_download_thread_reflector): Put returned bool in lParam.
+ * threebar.cc: Include msg.h.
+ (WM_APP_DOWNLOAD_THREAD_COMPLETE): Rework for new meaning of lParam,
+ incorporating program flow logic from download.cc.
Index: download.cc
===
RCS file: /home/max/cvsmirror/cygwin-apps-cvs/setup/download.cc,v
retrieving revision 2.37
diff -u -p -r2.37 download.cc
--- download.cc 7 Apr 2003 12:46:55 - 2.37
+++ download.cc 30 Jul 2003 22:03:05 -
@@ -221,7 +221,7 @@ download_one (packagesource pkgsource,
return 1;
}
-static void
+static bool
do_download_thread (HINSTANCE h, HWND owner)
{
int errors = 0;
@@ -302,25 +302,7 @@ do_download_thread (HINSTANCE h, HWND ow
}
}
- if (errors)
-{
- if (yesno (owner, IDS_DOWNLOAD_INCOMPLETE) == IDYES)
- {
- next_dialog = IDD_SITE;
- return;
- }
-}
-
- if (source == IDC_SOURCE_DOWNLOAD)
-{
- if (errors)
- exit_msg = IDS_DOWNLOAD_INCOMPLETE;
- else if (!unattended_mode)
- exit_msg = IDS_DOWNLOAD_COMPLETE;
- next_dialog = 0;
-}
- else
-next_dialog = IDD_S_INSTALL;
+ return (errors == 0);
}
static DWORD WINAPI
@@ -329,10 +311,11 @@ do_download_reflector (void *p)
HANDLE *context;
context = (HANDLE *) p;
- do_download_thread ((HINSTANCE) context[0], (HWND) context[1]);
+ bool succeeded = do_download_thread ((HINSTANCE) context[0],
+ (HWND) context[1]);
// Tell the progress page that we're done downloading
- Progress.PostMessage (WM_APP_DOWNLOAD_THREAD_COMPLETE, 0,
next_dialog);
+ Progress.PostMessage (WM_APP_DOWNLOAD_THREAD_COMPLETE, 0, succeeded);
ExitThread(0);
}
Index: threebar.cc
===
RCS file: /home/max/cvsmirror/cygwin-apps-cvs/setup/threebar.cc,v
retrieving revision 2.7
diff -u -p -r2.7 threebar.cc
--- threebar.cc 30 Jul 2003 21:49:53 - 2.7
+++ threebar.cc 30 Jul 2003 22:05:48 -
@@ -27,6 +27,7 @@
#include threebar.h
#include cistring.h
#include state.h
+#include msg.h
bool ThreeBarProgressPage::Create ()
{
@@ -148,21 +149,32 @@ ThreeBarProgressPage::OnMessageApp (UINT
}
case WM_APP_DOWNLOAD_THREAD_COMPLETE:
{
- if (lParam == IDD_S_INSTALL)
+ if (!lParam)
{
- // Download is complete and we want to go on to the install.
- Window::PostMessage (WM_APP_START_INSTALL);
+ if (yesno (GetHWND(), IDS_DOWNLOAD_INCOMPLETE) == IDYES)
+ {
+ // Download failed for some reason,
+ // go back to site selection page
+ GetOwner ()-SetActivePageByID (IDD_SITE);
+ break;
+ }
}
- else if (lParam != 0)
+
+ if (source == IDC_SOURCE_DOWNLOAD)
{
- // Download failed for some reason, go back to site selection page
- GetOwner ()-SetActivePageByID (lParam);
+ if (!lParam)
+ exit_msg = IDS_DOWNLOAD_INCOMPLETE;
+ else if (!unattended_mode)
+ exit_msg = IDS_DOWNLOAD_COMPLETE;
+ // Was a download-only, and is complete or failed.
+ GetOwner ()-PressButton (PSBTN_CANCEL);
}
else
{
- // Was a download-only, and is complete or failed.
- GetOwner ()-PressButton (PSBTN_CANCEL);
+ // Download is complete and we want to go on to the install.
+ Window::PostMessage (WM_APP_START_INSTALL);
}
+
break;
}
case WM_APP_START_INSTALL:
PING! PING! Re: [mknetrel suggestion] Move --enable-haifa from main script to extra/gcc2
Max Bowsher wrote: Max Bowsher wrote: Max Bowsher wrote: Max Bowsher wrote: --enable-haifa is only applicable to gcc2, therefore having it in the main script can be misleading. Thanks for the commit to bin/mknetrel. --enable-haifa should probably also be added into CONFIGOPTS in extra/gcc2, to record the fact that Cygwin gcc2 was built with that option. This was not committed. Since there is a possibility gcc2 may be rebuilt for 1.5.0, this should *definitely* be committed. Ping? If there is disagreement, please let me know, so at least I am sure this mail hasn't disassappeared into some hyperaticve spam filter, or something. Since the last ping, CGF has said he will not be rebuilding gcc for 1.5.x. That's fine, but in that case, maybe mknetrel/extra/gcc2 should be removed, to document the fact that it is no longer an accurate documentation of the gcc2 package build process? Max.
Re: PING! PING! Re: [mknetrel suggestion] Move --enable-haifa from main script to extra/gcc2
On Mon, Sep 08, 2003 at 02:48:05PM +0100, Max Bowsher wrote: Max Bowsher wrote: Max Bowsher wrote: Max Bowsher wrote: Max Bowsher wrote: --enable-haifa is only applicable to gcc2, therefore having it in the main script can be misleading. Thanks for the commit to bin/mknetrel. --enable-haifa should probably also be added into CONFIGOPTS in extra/gcc2, to record the fact that Cygwin gcc2 was built with that option. This was not committed. Since there is a possibility gcc2 may be rebuilt for 1.5.0, this should *definitely* be committed. Ping? If there is disagreement, please let me know, so at least I am sure this mail hasn't disassappeared into some hyperaticve spam filter, or something. Since the last ping, CGF has said he will not be rebuilding gcc for 1.5.x. That's fine, but in that case, maybe mknetrel/extra/gcc2 should be removed, to document the fact that it is no longer an accurate documentation of the gcc2 package build process? Since this is my package, you can let me worry about it.
Re: [ITP] GMP NTL libraries
I didn't even try to compile them, yet, but being mathematical packages I fear not they may not compile. So I'm beginning to ask the list if there is interest in those two number theory / big integer libraries. http://www.swox.com/gmp/ http://www.shoup.net/ntl/ (I think I'll compile NTL in the use GMP for faster base arithmetic method so NTL will depend upon GMP, while GMP should depend on nothing else) - -- I imagine gmp should be not too bad - i'll express interest in it since i think its going to be needed to compile g95 fortran... and ofcourse that its cool... Gareth
promote to curr: tetex, guile, lilypond
I'd like all my test packages to be promoted to curr; and the corresponding curr packages moved to prev. There are two packages of which there is no test version: libguile12abi13 and libkpathsea3abi13, which contain libraries for 1.3.22 binaries. I suppose these need to stay in curr, as long as they are listed in requires:, similar to libjpeg6[b2]. Greetings, Jan. curr: tetex-2.0.2-12 tetex-base-2.0.2-12 tetex-extra-2.0.2-12 tetex-doc-2.0.2-12 tetex-tiny-2.0.2-12 tetex-devel-2.0.2-12 libkpathsea3-2.0.2-12 and libkpathsea3abi13-2.0.2-2 tetex-x11-2.0.2-12 tetex-bin-2.0.2-12 guile-1.6.4-12 libguile12-1.6.4-12 and libguile12abi13-1.6.4-2 guile-devel-1.6.4-12 guile-doc-1.6.4-12 lilypond-1.8.1-12 lilypond-doc-1.8.1-12 prev: tetex-2.0.2-1 tetex-base-2.0.2-1 tetex-extra-2.0.2-1 tetex-doc-2.0.2-1 tetex-tiny-2.0.2-1 tetex-devel-2.0.2-2 tetex-x11-2.0.2-2 tetex-bin-2.0.2-2 guile-1.6.4-2 guile-devel-1.6.4-2 guile-doc-1.6.4-2 lilypond-1.6.11-1 lilypond-doc-1.6.11-1 The massaged setup hints are at: http://lilypond.org/cygwin/uploads/tetex/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-devel/setup.hint http://lilypond.org/cygwin/uploads/tetex/libkpathsea3/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-x11/setup.hint http://lilypond.org/cygwin/uploads/tetex/libkpathsea3abi13/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-bin/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-base/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-extra/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-doc/setup.hint http://lilypond.org/cygwin/uploads/tetex/tetex-tiny/setup.hint http://lilypond.org/cygwin/uploads/guile/setup.hint http://lilypond.org/cygwin/uploads/guile/libguile12/setup.hint http://lilypond.org/cygwin/uploads/guile/guile-devel/setup.hint http://lilypond.org/cygwin/uploads/guile/guile-doc/setup.hint http://lilypond.org/cygwin/uploads/guile/libguile12abi13/setup.hint http://lilypond.org/cygwin/uploads/lilypond/setup.hint http://lilypond.org/cygwin/uploads/lilypond/lilypond-doc/setup.hint -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
Re: promote to curr: tetex, guile, lilypond
On Mon, Sep 08, 2003 at 10:49:23PM +0200, Jan Nieuwenhuizen wrote: I'd like all my test packages to be promoted to curr; and the corresponding curr packages moved to prev. There are two packages of which there is no test version: libguile12abi13 and libkpathsea3abi13, which contain libraries for 1.3.22 binaries. I suppose these need to stay in curr, as long as they are listed in requires:, similar to libjpeg6[b2]. I think I've set up everything correctly. It looks like most of the packages just needed to have their curr/test entries removed to do the right thing. tetex/tetex-beta refers to a curr and prev, neither of which are available, and, in fact, results in an empty setup.ini setting. Should this package just be removed? There is a tetex-beta-20020530-3 package available but it isn't clear if this is what should be used. cgf
php cli?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi to all. Only want to ask a question: to have the CLI version of PHP (i.e. as a shell scripting language), which builds from the very same sources of course, should a new package be built? Should I do the patch and send it to mod_php maintaner? (actually is just a little different line in the configure, as far a sI remember) Of course PHP can be both useful as a web scripting language and as a shell scripting language and PEAR and common parts should be installed only once, maybe with a separate package required by both mod_php and cli_php. Well I can do that I guess, expecially if I'm not the only one to find it useful (I'm just too lazy to learn bash scripting better, e.g. to do a numeric for). - -- Lapo 'Raist' Luchini [EMAIL PROTECTED] (PGP X.509 keys available) http://www.lapo.it (ICQ UIN: 529796) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (Cygwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAj9c9FEACgkQaJiCLMjyUvvu3ACglEScMZaNeisXXSHxFP4vxC4j M60AoNUNGLHkqWWC8st8XjE+4U0RXOfu =vxIc -END PGP SIGNATURE-
Re: promote to curr: tetex, guile, lilypond
Christopher Faylor writes: I think I've set up everything correctly. It looks like most of the packages just needed to have their curr/test entries removed to do the right thing. Thanks very much. tetex/tetex-beta refers to a curr and prev, neither of which are available, and, in fact, results in an empty setup.ini setting. Should this package just be removed? There is a tetex-beta-20020530-3 package available but it isn't clear if this is what should be used. Yes, please do. This package is a leftover from the split tetex-beta and texmf times. It is empty and served as an upgrade helper. The equivalent texmf upgrade helpers (texmf, texmf-base, texmf-doc, texmf-extra, texmf-tiny) can be remove too. Jan. -- Jan Nieuwenhuizen [EMAIL PROTECTED] | GNU LilyPond - The music typesetter http://www.xs4all.nl/~jantien | http://www.lilypond.org
Re: php cli?
On Mon, 8 Sep 2003, Lapo Luchini wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi to all. Only want to ask a question: to have the CLI version of PHP (i.e. as a shell scripting language), which builds from the very same sources of course, should a new package be built? Should I do the patch and send it to mod_php maintaner? (actually is just a little different line in the configure, as far a sI remember) I think that's what the external-source: directive in setup.hint is for... Of course PHP can be both useful as a web scripting language and as a shell scripting language and PEAR and common parts should be installed only once, maybe with a separate package required by both mod_php and cli_php. I'm guessing there's nothing wrong with cli_php requiring mod_php, at least initially, as anyone who's likely to use php for shell scripting will probably have apache with mod_php installed anyway... Well I can do that I guess, expecially if I'm not the only one to find it useful (I'm just too lazy to learn bash scripting better, e.g. to do a numeric for). for ((i=0; i10; i++)); do echo $i; done for i in `seq 0 9`; do echo $i; done The latter one also works in sh... :-p Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton
Re: php cli?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Igor Pechtchanski wrote: | I think that's what the external-source: directive in setup.hint is | for... But that wuoln't use the same exact sources as the other package? (which are enough but with have no command to build them both, i.e. the configure is set only to generate mod_php, in mod_php) In that case I'd have to bother mod_php mantainer and/or send him a patch, anyway. | Of course PHP can be both useful as a web scripting language and as | a shell scripting language and PEAR and common parts should be | installed only once, maybe with a separate package required by both | mod_php and cli_php. | | | I'm guessing there's nothing wrong with cli_php requiring mod_php, at | least initially, as anyone who's likely to use php for shell | scripting will probably have apache with mod_php installed anyway... Yeah, I guess so. | for i in `seq 0 9`; do echo $i; done Mhhh... yummi seq command =) Anyway that was just an example, I actually feel more at ease in a language I already had to learn well 0=) - -- Lapo 'Raist' Luchini [EMAIL PROTECTED] (PGP X.509 keys available) http://www.lapo.it (ICQ UIN: 529796) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (Cygwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAj9c/iQACgkQaJiCLMjyUvsMrgCg5Dr4J0e4ssyiWEUpDEuwHMrr CloAoM7VQEAsI/F19m4Hl5oA5pqjRfnw =WxHu -END PGP SIGNATURE-
RE: cygwin 1.5.3-1
Hi Harold,
What do you mean? Do you mean that even after adding a call to check if
shm is supported and running that the extension is still installed, even
if you return?
Or do you mean that looking at the existing code shows
that they were never checking for shm support in the first place?
this one.
Harold
--- old ---
void
XFree86BigfontExtensionInit()
{
if ((extEntry = AddExtension(XF86BIGFONTNAME, /* add extension */
.. /* check for shm support and return if isn't
avaliable */
--- new ---
.. /* check for shm support and return if isn't
avaliable */
if ((extEntry = AddExtension(XF86BIGFONTNAME,/* add extension */
Cheers
Ralf
Problem with 'setxkbmap fr' since cygwin 1.5.x
Hello. I've got a little problem with my keyboard. - When I run cygwin-bach shell, my french keyboard is ok. - When I run rxvt without X11, my keyboard is OK too. - When I start XWin for XDM session, my keyboard is seen as us one (I never managed to change this with an XF86Config file, but no matter). So, I use 'setxkbmap fr' to have my french keyboard. This was working well since I updaded my cygwin installation from 1.3.xx to 1.5.yy cygwin pacages suite. Now, the keys '' and '' are no more recognised under X11, when all the others are ok. What should I do ? Thanks. Gilles.
Re: Problem with 'setxkbmap fr' since cygwin 1.5.x
On Mon, 8 Sep 2003, gilles civario wrote: Hello. I've got a little problem with my keyboard. - When I run cygwin-bach shell, my french keyboard is ok. - When I run rxvt without X11, my keyboard is OK too. - When I start XWin for XDM session, my keyboard is seen as us one (I never managed to change this with an XF86Config file, but no matter). So, I use 'setxkbmap fr' to have my french keyboard. This was working well since I updaded my cygwin installation from 1.3.xx to 1.5.yy cygwin pacages suite. Now, the keys '' and '' are no more recognised under X11, when all the others are ok. What should I do ? can you please send me /tmp/Xwin.log bye ago
Re: Problem with 'setxkbmap fr' since cygwin 1.5.x
gilles civario wrote: Alexander Gottwald wrote: On Mon, 8 Sep 2003, gilles civario wrote: Hello. I've got a little problem with my keyboard. - When I run cygwin-bach shell, my french keyboard is ok. - When I run rxvt without X11, my keyboard is OK too. - When I start XWin for XDM session, my keyboard is seen as us one (I never managed to change this with an XF86Config file, but no matter). So, I use 'setxkbmap fr' to have my french keyboard. This was working well since I updaded my cygwin installation from 1.3.xx to 1.5.yy cygwin pacages suite. Now, the keys '' and '' are no more recognised under X11, when all the others are ok. What should I do ? can you please send me /tmp/Xwin.log bye ago Here you are. Plus my XWin starting shell. Gilles. I've tested this XF86Config file, and with it, I do not need the setxkbmap. I've got directly the french keyboard, but steel without the '' and '' keys. What could be the trouble ? Gilles. Section InputDevice Identifier RuKeyboard Driver Keyboard Option XkbRules xfree86 Option XkbModel pc101 Option XkbLayout fr EndSection Section Device Identifier dummy Driver dummy EndSection Section Monitor Identifier dummy EndSection Section Screen Identifier dummy Device dummy Monitor dummy EndSection ddxProcessArgument - Initializing default screens winInitializeDefaultScreens - w 1280 h 1024 winInitializeDefaultScreens - Returning OsVendorInit - Creating bogus screen 0 _XSERVTransmkdir: Owner of /tmp/.X11-unix should be set to root (==) Using config file: /etc/X11/XF86Config Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (??) unknown. (WW) No Layout section. Using the first Screen section. (==) FontPath set to /usr/X11R6/lib/X11/fonts/misc/,/usr/X11R6/lib/X11/fonts/Speedo/,/usr/X11R6/lib/X11/fonts/Type1/,/usr/X11R6/lib/X11/fonts/CID/,/usr/X11R6/lib/X11/fonts/75dpi/,/usr/X11R6/lib/X11/fonts/100dpi/ (==) RgbPath set to /usr/X11R6/lib/X11/rgb winDetectSupportedEngines - Windows NT/2000/XP winDetectSupportedEngines - DirectDraw installed winDetectSupportedEngines - Allowing PrimaryDD winDetectSupportedEngines - DirectDraw4 installed winDetectSupportedEngines - Returning, supported engines 001f InitOutput - g_iNumScreens: 1 iMaxConsecutiveScreen: 1 winSetEngine - Using Shadow DirectDraw NonLocking winAdjustVideoModeShadowDDNL - Using Windows display depth of 32 bits per pixel winCreateBoundingWindowWindowed - User w: 1280 h: 1024 winCreateBoundingWindowWindowed - Current w: 1280 h: 1024 winAdjustForAutoHide - Original WorkArea: 0 0 996 1280 winAdjustForAutoHide - Adjusted WorkArea: 0 0 996 1280 winCreateBoundingWindowWindowed - WindowClient w 1280 h 996 r 1280 l 0 b 996 t 0 winCreateBoundingWindowWindowed - Returning winCreatePrimarySurfaceShadowDDNL - Creating primary surface winCreatePrimarySurfaceShadowDDNL - Created primary surface winCreatePrimarySurfaceShadowDDNL - Attached clipper to primary surface winAllocateFBShadowDDNL - lPitch: 5120 winAllocateFBShadowDDNL - Created shadow pitch: 5120 winAllocateFBShadowDDNL - Created shadow stride: 1280 winFinishScreenInitFB - Masks: 00ff ff00 00ff winInitVisualsShadowDDNL - Masks 00ff ff00 00ff BPRGB 8 d 24 bpp 32 winCreateDefColormap - Deferring to fbCreateDefColormap () winFinishScreenInitFB - returning winScreenInit - returning InitOutput - Returning. (==) winConfigKeyboard - Layout: 040C (**) Using keyboard RuKeyboard as primary keyboard (**) Option XkbRules xfree86 (**) XKB: rules: xfree86 (**) Option XkbModel pc101 (**) XKB: model: pc101 (**) Option XkbLayout fr (**) XKB: layout: fr Rules = xfree86 Model = pc101 Layout = fr Variant = (null) Options = (null) winPointerWarpCursor - Discarding first warp: 640 498 winBlockHandler - Releasing pmServerStarted winBlockHandler - pthread_mutex_unlock () returned
Re: Problem with 'setxkbmap fr' since cygwin 1.5.x
On Mon, 8 Sep 2003, gilles civario wrote: I've tested this XF86Config file, and with it, I do not need the setxkbmap. I've got directly the french keyboard, but steel without the '' and '' keys. What could be the trouble ? It must be a proble with the keyboard definition. From a debian host setxkbmap works fine, from cygwin the keys are missing. XFree86 on debian is 4.1.0 XFree86 on cygwin is 4.3.0 I'll check the xfree cvs if there have been patches to fix this problem. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
solved
Ok, I solved my problem. Sorry for the noise. In the XF86Config file, I've changed the XkbModel from 'pc102' to 'pc105'. Now, it's working perfecly... It proofs that questioning others helps to answering itself. Gilles. ddxProcessArgument - Initializing default screens winInitializeDefaultScreens - w 1280 h 1024 winInitializeDefaultScreens - Returning OsVendorInit - Creating bogus screen 0 _XSERVTransmkdir: Owner of /tmp/.X11-unix should be set to root (==) Using config file: /etc/X11/XF86Config Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (??) unknown. (WW) No Layout section. Using the first Screen section. (==) FontPath set to /usr/X11R6/lib/X11/fonts/misc/,/usr/X11R6/lib/X11/fonts/Speedo/,/usr/X11R6/lib/X11/fonts/Type1/,/usr/X11R6/lib/X11/fonts/CID/,/usr/X11R6/lib/X11/fonts/75dpi/,/usr/X11R6/lib/X11/fonts/100dpi/ (==) RgbPath set to /usr/X11R6/lib/X11/rgb winDetectSupportedEngines - Windows NT/2000/XP winDetectSupportedEngines - DirectDraw installed winDetectSupportedEngines - Allowing PrimaryDD winDetectSupportedEngines - DirectDraw4 installed winDetectSupportedEngines - Returning, supported engines 001f InitOutput - g_iNumScreens: 1 iMaxConsecutiveScreen: 1 winSetEngine - Using Shadow DirectDraw NonLocking winAdjustVideoModeShadowDDNL - Using Windows display depth of 32 bits per pixel winCreateBoundingWindowWindowed - User w: 1280 h: 1024 winCreateBoundingWindowWindowed - Current w: 1280 h: 1024 winAdjustForAutoHide - Original WorkArea: 0 0 996 1280 winAdjustForAutoHide - Adjusted WorkArea: 0 0 996 1280 winCreateBoundingWindowWindowed - WindowClient w 1280 h 996 r 1280 l 0 b 996 t 0 winCreateBoundingWindowWindowed - Returning winCreatePrimarySurfaceShadowDDNL - Creating primary surface winCreatePrimarySurfaceShadowDDNL - Created primary surface winCreatePrimarySurfaceShadowDDNL - Attached clipper to primary surface winAllocateFBShadowDDNL - lPitch: 5120 winAllocateFBShadowDDNL - Created shadow pitch: 5120 winAllocateFBShadowDDNL - Created shadow stride: 1280 winFinishScreenInitFB - Masks: 00ff ff00 00ff winInitVisualsShadowDDNL - Masks 00ff ff00 00ff BPRGB 8 d 24 bpp 32 winCreateDefColormap - Deferring to fbCreateDefColormap () winFinishScreenInitFB - returning winScreenInit - returning InitOutput - Returning. (==) winConfigKeyboard - Layout: 040C (**) Using keyboard RuKeyboard as primary keyboard (**) Option XkbRules xfree86 (**) XKB: rules: xfree86 (**) Option XkbModel pc105 (**) XKB: model: pc105 (**) Option XkbLayout fr (**) XKB: layout: fr Rules = xfree86 Model = pc105 Layout = fr Variant = (null) Options = (null) winPointerWarpCursor - Discarding first warp: 640 498 winBlockHandler - Releasing pmServerStarted winBlockHandler - pthread_mutex_unlock () returned Section InputDevice Identifier RuKeyboard Driver Keyboard Option XkbRules xfree86 Option XkbModel pc105 Option XkbLayout fr EndSection Section Device Identifier dummy Driver dummy EndSection Section Monitor Identifier dummy EndSection Section Screen Identifier dummy Device dummy Monitor dummy EndSection
How to get xdm working?
CYGWIN_95-4.0 myhost 1.5.3(0.90/3/2) 2003-09-01 13:15 i586 unknown unknown Cygwin I am running these tests on a Samba partition which is mounted on the F: drive. I can make an XDMCP connection using startxdmcp.bat or the command line Xwin.exe -query remotehost. (Having learnt to mount the /tmp as a bin type mount! ) This works fine and I can get the indirect list of machines on my net served up to me via the remotename machine. Connect to that machine and get the remote xdm to give me a login prompt. Is it possible set up cywin to run the equivalent of my Linux Boxes' inittab lines: x1:5:respawn:/usr/X11/bin/X -indirect thislocalhost x2:5:respawn:/etc/X11/xdm -nodaemon I have tried to do this in two ways: The first is to run xdm without removing the line in /etc/xdm/Xservers specifying the server name: :0 local /usr/X11R6/bin/X Infact I modified this to: localhost:0 local /usr/X11R6/bin/X Because I needed that name to fix the auth problem. (see below) I could then run xdm from the command line. It returns: Only root wants to run xdm I worked around this in one of two ways: either running xdm -debug 10 or creating a root account and logging in as root. The other files I altered were all in /etc/X11/xdm Xaccess * #any host can get a login window * CHOOSER BROADCAST #any indirect host can get a chooser xdm-config !DisplayManager.errorLogFile: /var/log/xdm.log DisplayManager.errorLogFile: /tmp/xdm.log ! removing this makes life simpler for now ! DisplayManager.willing: su nobody -c /usr/X11R6/lib/X11/xdm/Xwilling DisplayManager.*.authFile:Xauthority DisplayManager.RandomFile:/tmp/XWin.log !DisplayManager.requestPort: 0 Some comments on the above. I changed the errorLogFile so that I did not have to monitor logs in more than one dir. I removed the su nobody -c because it just complicates things! Removing the whole line saves running a script which includes programs which do not exist on my installation of cygwin. I created an Xauthority file because the one generated by xdm has a colon in it and the file system barfs at that. When creating the entries for the Xauthority file I used a modified version of the which appeared in May 2002 in this mail list: XAUTHORITY=/usr/X11R6/lib/X11/xdm/authdir/authfiles/Xauthority DISPLAYNUM=:0 mcookie=`dd if=/dev/urandom count=1 | md5sum | cut -f1 -d\ '` xauth -f $XAUTHORITY -v add $DISPLAYNUM . $mcookie xauth -f $XAUTHORITY -v add 127.0.0.1$DISPLAYNUM . $mcookie xauth -f $XAUTHORITY -v add $HOSTNAME$DISPLAYNUM . $mcookie NB; For belt and braces I mounted the file authfiles dir as type bin. This is why I changed the line in the Xserver file to include localhost as this script generates that entry: $ xauth list myhost/unix:0 MIT-MAGIC-COOKIE-1 7042...fc localhost:0 MIT-MAGIC-COOKIE-1 7042...fc myhost.mydomain.com:0 MIT-MAGIC-COOKIE-1 7042...fc DM.RandomFile was used because the default is /dev/mem which does not appear to exist. I would be grateful for a better solution than the log file as a file for sum. DM.requestPort is commented out because I want other machines to get XDMCP messages from this machine. If I run xdm with the Xservers commented out and from separate window with the line: Xwin -indirect myhost The host name appears in the indirect list on a Linux machine. But when I connect to it I get a grey cross hatched screen and nothing else. The same is true for the local X screen. At the same time as the cross hatch appears xdm core dumps Whether I run xdm -debug 10 as root or another user whether there is an entry in Xserver or chooser or xterm inplace of chooser xdm always core dumps usually with the KERNEL32.DLL or less often CYGWIN1.DLL depending on the settings in xdm-config I noted the entry 2003/7/102 that other have been doing far more work on his than I have. Has anyone got a version which runs (with the release which I am using) and if so what am I doing that is not correct? Two useful mail list refrences: http://cygwin.com/ml/cygwin-xfree/2003-07/msg00102.html http://sources.redhat.com/ml/cygwin-xfree/2002-05/msg00226.html Here is one example of a dump cut from the popup window. XDM caused an invalid page fault in module KERNEL32.DLL at 0137:bff85fe5. Registers: EAX=c0013f4c CS=0137 EIP=bff85fe5 EFLGS=00010202 EBX=8160e5a0 SS=013f ESP=006d EBP=006d009c ECX=c0013f48 DS=013f ESI=006d00d0 FS=2877 EDX=006d01f8 ES=013f EDI=006d00b4 GS= Bytes at CS:EIP: 57 8b 30 55 83 7e 54 00 0f 84 1c 01 00 00 f6 46 Stack dump: 006d00d0 006cda98 bff85812 0001 000e 006d00b4 006d00d0 006d0300 000b 006cda98 Here is a example taken from the XDM.EXE.stackdump Excption: STATUS_ACCESS_VIOLATION at eip=610788D1 eax=20202020 ebx=008CF380 ecx=61621714 edx=61621718 esi=0039 edi= ebp=008CF3A8 esp=008CF340 program=F:\CYGWIN\USR\X11R6\BIN\XDM.EXE cs=0137 ds=013F es=013F
Re: Alt Gr special characters on german keyboard work only sometimes
Hi all,
with all the Experts on keybords, could anybody give me a hint what might be wrong
here?
The problem is that the keys work only sometimes so you press them 10 times,
you get 8 backslashs an 2 times nothing, or 6-4. It seems to follow no pattern.
Thanks
Holger
Cygwin List wrote:
Hi,
since the last update I have a very particular keyboard problem. The
special characters reachable with the 'alt gr' key on the german
keyboard like: {[]}\~|@ etc. work only sometimes. When pressed
repeatedly they may work at one command at the prompt after hitting
CR they don't work anymore. This is the same in different
applications kdevelop, bash, various editors, even the XDMCP login
prompt. Sometimes after hitting ^q (ctrl q) they work again. Within
vi they are sometimes displayed as ^\ (ctrl backslash) when pressing
\ (backslash) sometimes as \ (backslash) only, as they should.
I'm using cygwin-xfree to connect to a debian box via XDMCP from a
WinXP Pro Machine. My Keyboard is the german version of the
Microsoft Natural Keyboard Pro with the appropriate drivers
installed. Detailed conf, startbat logfile follow (comments
removed).
Any help is appreciated
Holger
-- C:\cygwin\etc\X11\XF86Config-4
Section Files
RgbPath /usr/X11R6/lib/X11/rgb
EndSection
Section InputDevice
Identifier Keyboard1
Driver Keyboard
Option AutoRepeat 500 30
Option XkbModel microsoft
Option XkbLayout de
Option XkbVariant nodeadkeys
EndSection
Section Device
Identifier dummy
Driver dummy
EndSection
Section Monitor
Identifier dummy
EndSection
Section Screen
Identifier dummy
Device dummy
Monitor dummy
EndSection
Section ServerLayout
Identifier dummy
Screen dummy
InputDevice Keyboard1 CoreKeyboard
EndSection
-- startxdmcp.bat
@echo off
SET DISPLAY=127.0.0.1:0.0
SET REMOTE_HOST=myhost
SET CYGWIN_ROOT=\cygwin
SET PATH=.;%CYGWIN_ROOT%\bin;%CYGWIN_ROOT%\usr\X11R6\bin;%PATH%
REM Cleanup after last run.
if not exist %CYGWIN_ROOT%\tmp\.X11-unix\X0 goto CLEANUP-FINISH
attrib -s %CYGWIN_ROOT%\tmp\.X11-unix\X0
del %CYGWIN_ROOT%\tmp\.X11-unix\X0
CLEANUP-FINISH
if exist %CYGWIN_ROOT%\tmp\.X11-unix rmdir %CYGWIN_ROOT%\tmp\.X11-unix
if %OS% == Windows_NT goto OS_NT
echo startxdmcp.bat - Starting on Windows 95/98/Me
goto STARTUP
OS_NT
echo startxdmcp.bat - Starting on Windows NT/2000/XP
STARTUP
start XWin -query %REMOTE_HOST% -nodecoration -lesspointer
--
ddxProcessArgument - Initializing default screens
winInitializeDefaultScreens - w 1600 h 1200
winInitializeDefaultScreens - Returning
OsVendorInit - Creating bogus screen 0
(==) Using config file: /etc/X11/XF86Config-4
Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (??) unknown.
(==) FontPath set to
/usr/X11R6/lib/X11/fonts/misc/,/usr/X11R6/lib/X11/fonts/Speedo/,/usr/X11R6/lib/X11/fonts/Type1/,/usr/X11R6/lib/X11/fonts/CID/,/usr/X11R6/lib/X11/fonts/75dpi/,/usr/X11R6/lib/X11/fonts/100dpi/
(**) RgbPath set to /usr/X11R6/lib/X11/rgb
winDetectSupportedEngines - Windows NT/2000/XP
winDetectSupportedEngines - DirectDraw installed
winDetectSupportedEngines - Allowing PrimaryDD
winDetectSupportedEngines - DirectDraw4 installed
winDetectSupportedEngines - Returning, supported engines 001f
InitOutput - g_iNumScreens: 1 iMaxConsecutiveScreen: 1
winSetEngine - Using Shadow DirectDraw NonLocking
winAdjustVideoModeShadowDDNL - Using Windows display depth of 32 bits
per pixel
winCreateBoundingWindowWindowed - User w: 1600 h: 1200
winCreateBoundingWindowWindowed - Current w: 1600 h: 1200
winAdjustForAutoHide - Original WorkArea: 0 32 1200 1600
winAdjustForAutoHide - Adjusted WorkArea: 0 32 1200 1600
winCreateBoundingWindowWindowed - WindowClient w 1568 h 1200 r 1568 l
0 b 1200 t 0
winCreateBoundingWindowWindowed - Returning
winCreatePrimarySurfaceShadowDDNL - Creating primary surface
winCreatePrimarySurfaceShadowDDNL - Created primary surface
winCreatePrimarySurfaceShadowDDNL - Attached clipper to primary
surface
winAllocateFBShadowDDNL - lPitch: 6272
winAllocateFBShadowDDNL - Created shadow pitch: 6272
winAllocateFBShadowDDNL - Created shadow stride: 1568
winFinishScreenInitFB - Masks: 00ff ff00 00ff
winInitVisualsShadowDDNL - Masks 00ff ff00 00ff BPRGB 8 d
24 bpp 32
winCreateDefColormap - Deferring to fbCreateDefColormap ()
winFinishScreenInitFB - returning
winScreenInit - returning
InitOutput - Returning.
(==) Using preset keyboard for German (Germany) (0407), type 4
(**) Using keyboard Keyboard1 as primary keyboard
(**) Option XkbModel microsoft
(**) XKB: model: microsoft
(**) Option XkbLayout de
(**) XKB: layout: de
(**) Option XkbVariant nodeadkeys
(**) XKB: variant: nodeadkeys
Rules = xfree86 Model = microsoft Layout = de Variant =
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Holger wrote: with all the Experts on keybords, could anybody give me a hint what might be wrong here? The problem is that the keys work only sometimes so you press them 10 times, you get 8 backslashs an 2 times nothing, or 6-4. It seems to follow no pattern. I'm not expiriencing such problems. But you can start the program xev, enter some backslashes into the window and send me the output, especially the KeyPress and KeyRelease events when the blackslash does not work. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: How to use Multi_key
On Mon, 8 Sep 2003, Hans Werner Strube wrote: Is it possible to actually compose characters not on the keyboard using Multi_key and XLookupString() in Cygwin X11? So far I had no success, except for a single, nonreproducible case where Multi_key ' a - á . (On my German keyboard, Multi_key is Shift Alt-Gr.) Or is there a similar feature at the MS Windows level which I don't know? Everything which is possible with keyboard input in the linux XFree server should be possible with Cygwin/XFree. A short test revealed that at least these characters can be produced with the multikey: äáàâåÂòè The keyboard input does not rely on Windows highlevel functions. The only input we get from Windows is the press and the release of the keys on the keyboard. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Alt Gr special characters on german keyboard work only sometimes
Hi Alexander On Mon, 8 Sep 2003, Holger wrote: with all the Experts on keybords, could anybody give me a hint what might be wrong here? The problem is that the keys work only sometimes so you press them 10 times, you get 8 backslashs an 2 times nothing, or 6-4. It seems to follow no pattern. I'm not expiriencing such problems. But you can start the program xev, enter some backslashes into the window and send me the output, especially the KeyPress and KeyRelease events when the blackslash does not work. Here you go. What I did is holding the 'AltGr' key continously and then pressing the '? ß \' (questionmark (with shift), sharps (normal) backslash (ArtGr)) key several times. I double checked the events generated by Windows earlier with Spy++, they do not differ whether cygwin get's the right key or not. thanks for your help Holger KeyPress event, serial 22, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387568, (-402,-377), root:(212,591), state 0x10, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387818, (-402,-377), root:(212,591), state 0x90, keycode 37 (keysym 0xffe3, Control_L), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387818, (-402,-377), root:(212,591), state 0x94, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387818, (-402,-377), root:(212,591), state 0x14, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387859, (-402,-377), root:(212,591), state 0x94, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387859, (-402,-377), root:(212,591), state 0x14, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387899, (-402,-377), root:(212,591), state 0x94, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387899, (-402,-377), root:(212,591), state 0x14, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387939, (-402,-377), root:(212,591), state 0x94, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387939, (-402,-377), root:(212,591), state 0x14, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387979, (-402,-377), root:(212,591), state 0x94, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387979, (-402,-377), root:(212,591), state 0x14, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387999, (-402,-377), root:(212,591), state 0x94, keycode 20 (keysym 0x5c, backslash), same_screen YES, XLookupString gives 1 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9388109, (-402,-377), root:(212,591), state 0x94, keycode 20 (keysym 0x5c, backslash), same_screen YES, XLookupString gives 1 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9388339, (-402,-377), root:(212,591), state 0x94, keycode 37 (keysym 0xffe3, Control_L), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9388339, (-402,-377), root:(212,591), state 0x90, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9390122, (-402,-377), root:(212,591), state 0x10, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
remote login to linux host?
I have cygwin installed on my XP, and I can succesfully login to my host if I am home. I use 'xwin -query host'. If I am outside the network and try it over the internet I get only the grey Xwin screen. First I thought my NAT and firewall was blocking, but that is not true. I defined my linux host as defserver in my NAT, so everything is routed to the host. Telnetting is okay, I can use SWAT to remotely manage samba, VNC works, only cygwin refuses to show my login screen... What could be wrong?
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Holger wrote: Here you go. What I did is holding the 'AltGr' key continously and then pressing the '? ß \' (questionmark (with shift), sharps (normal) backslash (ArtGr)) key several times. KeyPress event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387818, (-402,-377), root:(212,591), state 0x90, keycode 37 (keysym 0xffe3, Control_L), same_screen YES, XLookupString gives 0 characters: KeyRelease event, serial 27, synthetic NO, window 0x41, root 0x3a, subw 0x0, time 9387818, (-402,-377), root:(212,591), state 0x94, keycode 113 (keysym 0xfe03, ISO_Level3_Shift), same_screen YES, XLookupString gives 0 characters: This seem to be the problem. We had some reports that alt-gr does not work properly on (some) WinXP machines. Do you have WinXP? I double checked the events generated by Windows earlier with Spy++, they do not differ whether cygwin get's the right key or not. There must be a subtile difference. Maybe with the timestamp of the message. Or the Control down and alt down messages come in reversed order. I don't have WinXP and can't test this. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: remote login to linux host?
On Mon, 8 Sep 2003, Jan van Veldhuizen wrote: I have cygwin installed on my XP, and I can succesfully login to my host if I am home. I use 'xwin -query host'. If I am outside the network and try it over the internet I get only the grey Xwin screen. First I thought my NAT and firewall was blocking, but that is not true. I defined my linux host as defserver in my NAT, so everything is routed to the host. Telnetting is okay, I can use SWAT to remotely manage samba, VNC works, only cygwin refuses to show my login screen... If you have NAT between your host and the remotehost then xdmcp will not work. Use the X11Forwarding technique of ssh instead. DISPLAY=localhost:0.0 ssh -X remotehost bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: remote login to linux host?
If you have NAT between your host and the remotehost then xdmcp will not work. Use the X11Forwarding technique of ssh instead. DISPLAY=localhost:0.0 ssh -X remotehost This is what my Cygwin window shows: [EMAIL PROTECTED] ~ $ DISPLAY=localhost:0.0 [EMAIL PROTECTED] ~ $ ssh -X van-veldhuizen.nl bash: ssh: command not found [EMAIL PROTECTED] ~ $ DISPLAY=localhost:0.0 ssh -X van-veldhuizen.nl bash: ssh: command not found [EMAIL PROTECTED] ~ $ what next?
Re: remote login to linux host?
On Mon, 8 Sep 2003, Jan van Veldhuizen wrote: [EMAIL PROTECTED] ~ $ DISPLAY=localhost:0.0 ssh -X van-veldhuizen.nl bash: ssh: command not found what next? start cygwin setup and install openssh from the Net category. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: remote login to linux host?
Great! Works...but: I get a normal telnet session. No graphical login screen. Sorry, I am a newbie. Alexander Gottwald [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] On Mon, 8 Sep 2003, Jan van Veldhuizen wrote: [EMAIL PROTECTED] ~ $ DISPLAY=localhost:0.0 ssh -X van-veldhuizen.nl bash: ssh: command not found what next? start cygwin setup and install openssh from the Net category. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: remote login to linux host?
On Mon, 8 Sep 2003, Jan van Veldhuizen wrote: Great! Works...but: I get a normal telnet session. No graphical login screen. Sorry, I am a newbie. After login, you can start xterm, or the windowmanager or with startkde the whole kde desktop (if it is installed) bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Alt Gr special characters on german keyboard work only sometimes
Hi Alexander This seem to be the problem. We had some reports that alt-gr does not work properly on (some) WinXP machines. Do you have WinXP? Yes, XP Prof., no Service Pack c:\ver Microsoft Windows XP [Version 5.1.2600] There must be a subtile difference. Maybe with the timestamp of the message. Or the Control down and alt down messages come in reversed order. I don't have WinXP and can't test this. I attached the logfile of spy++ I pushed: altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out nothing bye Holger msg.log Description: Binary data
Re: remote login to linux host?
Thanks!!! It's working now. Alexander Gottwald [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] On Mon, 8 Sep 2003, Jan van Veldhuizen wrote: Great! Works...but: I get a normal telnet session. No graphical login screen. Sorry, I am a newbie. After login, you can start xterm, or the windowmanager or with startkde the whole kde desktop (if it is installed) bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: How to get xdm working?
That is a pretty large and well-detailed request. Unfortunately, I don't know enough to answer your question. Perhaps someone else does... if not, it looks like you would be the one that knows the most about the subject (a position I often find myself in :). Harold [EMAIL PROTECTED] wrote: CYGWIN_95-4.0 myhost 1.5.3(0.90/3/2) 2003-09-01 13:15 i586 unknown unknown Cygwin I am running these tests on a Samba partition which is mounted on the F: drive. I can make an XDMCP connection using startxdmcp.bat or the command line Xwin.exe -query remotehost. (Having learnt to mount the /tmp as a bin type mount! ) This works fine and I can get the indirect list of machines on my net served up to me via the remotename machine. Connect to that machine and get the remote xdm to give me a login prompt. Is it possible set up cywin to run the equivalent of my Linux Boxes' inittab lines: x1:5:respawn:/usr/X11/bin/X -indirect thislocalhost x2:5:respawn:/etc/X11/xdm -nodaemon I have tried to do this in two ways: The first is to run xdm without removing the line in /etc/xdm/Xservers specifying the server name: :0 local /usr/X11R6/bin/X Infact I modified this to: localhost:0 local /usr/X11R6/bin/X Because I needed that name to fix the auth problem. (see below) I could then run xdm from the command line. It returns: Only root wants to run xdm I worked around this in one of two ways: either running xdm -debug 10 or creating a root account and logging in as root. The other files I altered were all in /etc/X11/xdm Xaccess * #any host can get a login window * CHOOSER BROADCAST #any indirect host can get a chooser xdm-config !DisplayManager.errorLogFile: /var/log/xdm.log DisplayManager.errorLogFile: /tmp/xdm.log ! removing this makes life simpler for now ! DisplayManager.willing: su nobody -c /usr/X11R6/lib/X11/xdm/Xwilling DisplayManager.*.authFile:Xauthority DisplayManager.RandomFile:/tmp/XWin.log !DisplayManager.requestPort: 0 Some comments on the above. I changed the errorLogFile so that I did not have to monitor logs in more than one dir. I removed the su nobody -c because it just complicates things! Removing the whole line saves running a script which includes programs which do not exist on my installation of cygwin. I created an Xauthority file because the one generated by xdm has a colon in it and the file system barfs at that. When creating the entries for the Xauthority file I used a modified version of the which appeared in May 2002 in this mail list: XAUTHORITY=/usr/X11R6/lib/X11/xdm/authdir/authfiles/Xauthority DISPLAYNUM=:0 mcookie=`dd if=/dev/urandom count=1 | md5sum | cut -f1 -d\ '` xauth -f $XAUTHORITY -v add $DISPLAYNUM . $mcookie xauth -f $XAUTHORITY -v add 127.0.0.1$DISPLAYNUM . $mcookie xauth -f $XAUTHORITY -v add $HOSTNAME$DISPLAYNUM . $mcookie NB; For belt and braces I mounted the file authfiles dir as type bin. This is why I changed the line in the Xserver file to include localhost as this script generates that entry: $ xauth list myhost/unix:0 MIT-MAGIC-COOKIE-1 7042...fc localhost:0 MIT-MAGIC-COOKIE-1 7042...fc myhost.mydomain.com:0 MIT-MAGIC-COOKIE-1 7042...fc DM.RandomFile was used because the default is /dev/mem which does not appear to exist. I would be grateful for a better solution than the log file as a file for sum. DM.requestPort is commented out because I want other machines to get XDMCP messages from this machine. If I run xdm with the Xservers commented out and from separate window with the line: Xwin -indirect myhost The host name appears in the indirect list on a Linux machine. But when I connect to it I get a grey cross hatched screen and nothing else. The same is true for the local X screen. At the same time as the cross hatch appears xdm core dumps Whether I run xdm -debug 10 as root or another user whether there is an entry in Xserver or chooser or xterm inplace of chooser xdm always core dumps usually with the KERNEL32.DLL or less often CYGWIN1.DLL depending on the settings in xdm-config I noted the entry 2003/7/102 that other have been doing far more work on his than I have. Has anyone got a version which runs (with the release which I am using) and if so what am I doing that is not correct? Two useful mail list refrences: http://cygwin.com/ml/cygwin-xfree/2003-07/msg00102.html http://sources.redhat.com/ml/cygwin-xfree/2002-05/msg00226.html Here is one example of a dump cut from the popup window. XDM caused an invalid page fault in module KERNEL32.DLL at 0137:bff85fe5. Registers: EAX=c0013f4c CS=0137 EIP=bff85fe5 EFLGS=00010202 EBX=8160e5a0 SS=013f ESP=006d EBP=006d009c ECX=c0013f48 DS=013f ESI=006d00d0 FS=2877 EDX=006d01f8 ES=013f EDI=006d00b4 GS= Bytes at CS:EIP: 57 8b 30 55 83 7e 54 00 0f 84 1c 01 00 00 f6 46 Stack dump: 006d00d0 006cda98 bff85812 0001 000e 006d00b4 006d00d0 006d0300 000b 006cda98
Re: Alt Gr special characters on german keyboard work only sometimes
Holger, We have also had reports that one of the XP PowerToys messes up the AltGr key handling. Uninstalling this PowerToy fixes the problem. Have you got any of the PowerToys installed? Harold Holger Böhnke wrote: Hi Alexander This seem to be the problem. We had some reports that alt-gr does not work properly on (some) WinXP machines. Do you have WinXP? Yes, XP Prof., no Service Pack c:\ver Microsoft Windows XP [Version 5.1.2600] There must be a subtile difference. Maybe with the timestamp of the message. Or the Control down and alt down messages come in reversed order. I don't have WinXP and can't test this. I attached the logfile of spy++ I pushed: altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out nothing bye Holger
Re: Xwin stackdump while logging into RH8 with KDE as window manager
On Sunday 07 September 2003 19:59, Alexander Gottwald wrote: Alexander Gottwald wrote: I was hunting the problem with textmode /tmp and xkb today. The fix I've submitted long ago never got into the xfree cvs, so the problem is still there. I'll check it into xoncygwin cvs after a few tests. The change for the server is submitted. I'm not sure if xkbcomp is already patched in the last 4.3 release. Harold, can you please check if the attached patch is applied to your buildtree? (file is xc/programs/xkbcomp/xkbcomp.c) Hit the nail on the head! Works like a charm now. Sorry I always keep everything mounted in binmode, but when I built the new machines that slipped through the cracks. Many thanks to you, Harold and the xfree team. Great product and great support. ciao, joel bye ago NP: Lacrimosa - Schakal
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Holger Böhnke wrote: I attached the logfile of spy++ Where can I get this spy++ program? Is it part of the VC++ package? Or is it freely available from the microsoft website? bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: How to get xdm working?
On Mon, 8 Sep 2003 [EMAIL PROTECTED] wrote: CYGWIN_95-4.0 myhost 1.5.3(0.90/3/2) 2003-09-01 13:15 i586 unknown unknown Cygwin I am running these tests on a Samba partition which is mounted on the F: drive. I can make an XDMCP connection using startxdmcp.bat or the command line Xwin.exe -query remotehost. (Having learnt to mount the /tmp as a bin type mount! ) This works fine and I can get the indirect list of machines on my net served up to me via the remotename machine. Connect to that machine and get the remote xdm to give me a login prompt. Is it possible set up cywin to run the equivalent of my Linux Boxes' inittab lines: x1:5:respawn:/usr/X11/bin/X -indirect thislocalhost x2:5:respawn:/etc/X11/xdm -nodaemon There is a sysvinit package for Cygwin that might do exactly what you want. See the package documentation in /usr/doc/sysvinit* (or /usr/share/doc/sysvinit*) and the Cygwin-specific documentation in /usr/doc/Cygwin/sysvinit-*.README (or /usr/share/doc/...). I have tried to do this in two ways: The first is to run xdm without removing the line in /etc/xdm/Xservers specifying the server name: :0 local /usr/X11R6/bin/X Infact I modified this to: localhost:0 local /usr/X11R6/bin/X Because I needed that name to fix the auth problem. (see below) I could then run xdm from the command line. It returns: Only root wants to run xdm I worked around this in one of two ways: either running xdm -debug 10 or creating a root account and logging in as root. This is probably a bug in xdm. It's been discussed on this list before. Cygwin (or, rather, Windows) doesn't have a root user, but does have users with various permissions usually attributed to root. On Win9x you can simply add a line for root to /etc/passwd (on WinNT/2k/XP you can copy the line for system and change the user name in that new line to root). The other files I altered were all in /etc/X11/xdm Xaccess * #any host can get a login window * CHOOSER BROADCAST #any indirect host can get a chooser xdm-config !DisplayManager.errorLogFile: /var/log/xdm.log DisplayManager.errorLogFile: /tmp/xdm.log ! removing this makes life simpler for now ! DisplayManager.willing: su nobody -c /usr/X11R6/lib/X11/xdm/Xwilling DisplayManager.*.authFile:Xauthority DisplayManager.RandomFile:/tmp/XWin.log !DisplayManager.requestPort: 0 Some comments on the above. I changed the errorLogFile so that I did not have to monitor logs in more than one dir. I removed the su nobody -c because it just complicates things! Removing the whole line saves running a script which includes programs which do not exist on my installation of cygwin. I created an Xauthority file because the one generated by xdm has a colon in it and the file system barfs at that. When creating the entries for the Xauthority file I used a modified version of the which appeared in May 2002 in this mail list: XAUTHORITY=/usr/X11R6/lib/X11/xdm/authdir/authfiles/Xauthority DISPLAYNUM=:0 mcookie=`dd if=/dev/urandom count=1 | md5sum | cut -f1 -d\ '` xauth -f $XAUTHORITY -v add $DISPLAYNUM . $mcookie xauth -f $XAUTHORITY -v add 127.0.0.1$DISPLAYNUM . $mcookie xauth -f $XAUTHORITY -v add $HOSTNAME$DISPLAYNUM . $mcookie NB; For belt and braces I mounted the file authfiles dir as type bin. This is why I changed the line in the Xserver file to include localhost as this script generates that entry: $ xauth list myhost/unix:0 MIT-MAGIC-COOKIE-1 7042...fc localhost:0 MIT-MAGIC-COOKIE-1 7042...fc myhost.mydomain.com:0 MIT-MAGIC-COOKIE-1 7042...fc DM.RandomFile was used because the default is /dev/mem which does not appear to exist. I would be grateful for a better solution than the log file as a file for sum. Cygwin has a /dev/mem (but it won't be shown by ls /dev, since /dev is a virtual filesystem in Cygwin. Try ls -l /dev/mem, and you'll see that it does exist. I've posted a script to the main Cygwin list that would create the /dev directory and the various entries in it so that ls and tab completion works. Another possibility is using /dev/urandom, which you've used for mcookie above. DM.requestPort is commented out because I want other machines to get XDMCP messages from this machine. If I run xdm with the Xservers commented out and from separate window with the line: Xwin -indirect myhost The host name appears in the indirect list on a Linux machine. But when I connect to it I get a grey cross hatched screen and nothing else. The same is true for the local X screen. At the same time as the cross hatch appears xdm core dumps Whether I run xdm -debug 10 as root or another user whether there is an entry in Xserver or chooser or xterm inplace of chooser xdm always core dumps usually with the KERNEL32.DLL or less often CYGWIN1.DLL depending on the settings in xdm-config If you'd like to debug this yourself, you could
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Alexander Gottwald wrote: On Mon, 8 Sep 2003, Holger Bohnke wrote: I attached the logfile of spy++ Where can I get this spy++ program? Is it part of the VC++ package? Or is it freely available from the microsoft website? bye ago Spy++ is part of the Windows Platform SDK, which comes with Visual Studio, and is also (according to Microsoft) available as a free download on the web (at http://www.microsoft.com/msdownload/platformsdk/sdkupdate/). Hope this helps, Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Holger Böhnke wrote: I attached the logfile of spy++ ca you do the log again with all output checkboxes marked (esp raw parameter and time). I want to compare with the log I got. Thanks, ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Alt Gr special characters on german keyboard work only sometimes
Ooops, forgot to attach the logfile win2k... Holger Holger wrote: Meanwhile I'm at another workplace, where I run win2k. Here the Altgr key works. I spyed the windows messages and came up with something peculiar: w2k: (pressed 3*backslash) 1 002C029C P WM_KEYDOWN nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:0 fRepeat:0 fUp:0 2 002C029C P WM_KEYDOWN nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:1 fRepeat:0 fUp:0 3 002C029C P WM_KEYDOWN nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:1 fRepeat:1 fUp:0 4 002C029C P WM_KEYDOWN nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:1 fRepeat:1 fUp:0 5 002C029C P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 6 002C029C P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 7 002C029C P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 8 002C029C P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 9 002C029C P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 00010 002C029C P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 00011 002C029C P WM_SYSKEYUP nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:1 fRepeat:1 fUp:1 00012 002C029C P WM_KEYUP nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:0 fRepeat:1 fUp:1 see the first 4 messages. apparently w2k sends 1 WM_KEYDOWN nVirtKey:VK_CONTROL ... fAltDown:0 fRepeat:0 2 WM_KEYDOWN nVirtKey:VK_MENU ... fAltDown:1 fRepeat:0 3 WM_KEYDOWN nVirtKey:VK_CONTROL ... fAltDown:1 fRepeat:1 4 WM_KEYDOWN nVirtKey:VK_MENU ...fAltDown:1 fRepeat:1 It looks like it repeats the VK_CONTROL VK_MENU msgs with altdown flag = 1 5 6 are the real key that gets pressed 'ß?\' while XP sends: (pressed 1*backslash) 1 005202C4 P WM_KEYDOWN nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:0 fRepeat:0 fUp:0 2 005202C4 P WM_KEYDOWN nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:1 fRepeat:0 fUp:0 3 005202C4 P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 4 005202C4 P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 5 005202C4 P WM_SYSKEYUP nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:1 fRepeat:1 fUp:1 6 005202C4 P WM_KEYUP nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:0 fRepeat:1 fUp:1 here the keypress of 'ß?\' is sent directly after only 2 messages: 1 WM_KEYDOWN nVirtKey:VK_CONTROL ... fAltDown:0 fRepeat:0 2 WM_KEYDOWN nVirtKey:VK_MENU ... fAltDown:1 fRepeat:0 3 WM_KEYDOWN nVirtKey:00DB ... fAltDown:1 fRepeat:0 fUp:0 4 WM_KEYUP nVirtKey:00DB ... fAltDown:1 fRepeat:1 fUp:1 5 WM_SYSKEYUP nVirtKey:VK_CONTROL ... 6 WM_KEYUP nVirtKey:VK_MENU ... with the first 2 like the first 2 messages 2k sends. Therefore the keydown message with altdown=1 for VK_CONTROL and VK_MENU are missing. Could it be that cygwin is confused by the missing messages? Holger Holger Böhnke wrote: Hi Alexander This seem to be the problem. We had some reports that alt-gr does not work properly on (some) WinXP machines. Do you have WinXP? Yes, XP Prof., no Service Pack c:\ver Microsoft Windows XP [Version 5.1.2600] There must be a subtile difference. Maybe with the timestamp of the message. Or the Control down and alt down messages come in reversed order. I don't have WinXP and can't test this. I attached the logfile of spy++ I pushed: altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out nothing bye Holger msg.log Description: Binary data
Re: Alt Gr special characters on german keyboard work only sometimes
Alexander Gottwald wrote: On Mon, 8 Sep 2003, Holger Böhnke wrote: I attached the logfile of spy++ ca you do the log again with all output checkboxes marked (esp raw parameter and time). I want to compare with the log I got. Thanks, ago Yupp, I'll send it as soon as I get back home. Do you need the Log of W2K as well with all the flags enabled? Holger
Re: Alt Gr special characters on german keyboard work only sometimes
Stupid me got it confused. :-) It was the other way round, XP sent the extra 2 Messages W2K did not. And it looks like it's the repeat function of windows generating the extra key events. When I simply hold the AltGr key on W2k I get the same amount of messages. Holger Holger wrote: Meanwhile I'm at another workplace, where I run win2k. Here the Altgr key works. I spyed the windows messages and came up with something peculiar: w2k: (pressed 3*backslash) 1 002C029C P WM_KEYDOWN nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:0 fRepeat:0 fUp:0 2 002C029C P WM_KEYDOWN nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:1 fRepeat:0 fUp:0 3 002C029C P WM_KEYDOWN nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:1 fRepeat:1 fUp:0 4 002C029C P WM_KEYDOWN nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:1 fRepeat:1 fUp:0 5 002C029C P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 6 002C029C P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 7 002C029C P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 8 002C029C P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 9 002C029C P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 00010 002C029C P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 00011 002C029C P WM_SYSKEYUP nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:1 fRepeat:1 fUp:1 00012 002C029C P WM_KEYUP nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:0 fRepeat:1 fUp:1 see the first 4 messages. apparently w2k sends 1 WM_KEYDOWN nVirtKey:VK_CONTROL ... fAltDown:0 fRepeat:0 2 WM_KEYDOWN nVirtKey:VK_MENU ... fAltDown:1 fRepeat:0 3 WM_KEYDOWN nVirtKey:VK_CONTROL ... fAltDown:1 fRepeat:1 4 WM_KEYDOWN nVirtKey:VK_MENU ...fAltDown:1 fRepeat:1 It looks like it repeats the VK_CONTROL VK_MENU msgs with altdown flag = 1 5 6 are the real key that gets pressed 'ß?\' while XP sends: (pressed 1*backslash) 1 005202C4 P WM_KEYDOWN nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:0 fRepeat:0 fUp:0 2 005202C4 P WM_KEYDOWN nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:1 fRepeat:0 fUp:0 3 005202C4 P WM_KEYDOWN nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:0 fUp:0 4 005202C4 P WM_KEYUP nVirtKey:00DB cRepeat:1 ScanCode:0C fExtended:0 fAltDown:1 fRepeat:1 fUp:1 5 005202C4 P WM_SYSKEYUP nVirtKey:VK_CONTROL cRepeat:1 ScanCode:1D fExtended:0 fAltDown:1 fRepeat:1 fUp:1 6 005202C4 P WM_KEYUP nVirtKey:VK_MENU cRepeat:1 ScanCode:38 fExtended:1 fAltDown:0 fRepeat:1 fUp:1 here the keypress of 'ß?\' is sent directly after only 2 messages: 1 WM_KEYDOWN nVirtKey:VK_CONTROL ... fAltDown:0 fRepeat:0 2 WM_KEYDOWN nVirtKey:VK_MENU ... fAltDown:1 fRepeat:0 3 WM_KEYDOWN nVirtKey:00DB ... fAltDown:1 fRepeat:0 fUp:0 4 WM_KEYUP nVirtKey:00DB ... fAltDown:1 fRepeat:1 fUp:1 5 WM_SYSKEYUP nVirtKey:VK_CONTROL ... 6 WM_KEYUP nVirtKey:VK_MENU ... with the first 2 like the first 2 messages 2k sends. Therefore the keydown message with altdown=1 for VK_CONTROL and VK_MENU are missing. Could it be that cygwin is confused by the missing messages? Holger Holger Böhnke wrote: Hi Alexander This seem to be the problem. We had some reports that alt-gr does not work properly on (some) WinXP machines. Do you have WinXP? Yes, XP Prof., no Service Pack c:\ver Microsoft Windows XP [Version 5.1.2600] There must be a subtile difference. Maybe with the timestamp of the message. Or the Control down and alt down messages come in reversed order. I don't have WinXP and can't test this. I attached the logfile of spy++ I pushed: altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out \\\ altgr (holding down) - 3x backsl - release altgr : out nothing bye Holger
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Holger wrote: with the first 2 like the first 2 messages 2k sends. Therefore the keydown message with altdown=1 for VK_CONTROL and VK_MENU are missing. Could it be that cygwin is confused by the missing messages? The duplicated messages are repeats of the first two. This happens when you wait too long before pressing the ß-key. This happens on XP too (see you first posted log) bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Alt Gr special characters on german keyboard work only sometimes
On Mon, 8 Sep 2003, Holger wrote: Yupp, I'll send it as soon as I get back home. Do you need the Log of W2K as well with all the flags enabled? I've already done such a dump. bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: Xwin stackdump while logging into RH8 with KDE as window manager
On Sunday 07 September 2003 03:11, Harold L Hunt II wrote: Joel, [snip/snip] That is interesting... are you using an XF86Config file? Or, do you have your keyboard in Windows set to something other than the default layout (the answer is probably no if you don't know what this is)? What about the keyboard in your KDE setup. I am not sure about the XF86Config file, if I was where would it be sitting. As for the keyboards, yes we use UK keyboards. Up until a few months ago I had to run xmodmap /usr/share/xmodmap/xmodmap.uk to reset the key bindings. This problem disappeared at some point a few months back but reapeared with the new release. The fix of changing the mount types to binmode also re-fixed this problem. I am really very curious as to what all of the interactions are, but I rather afraid you guys would have to write a small book to explain it all. Perhaps better to simply leave it with a hearty thank you and be happy that you are on top of it all ;-) - joel
Re: Xwin stackdump while logging into RH8 with KDE as window manager
On Mon, 8 Sep 2003, Joel Rosi-Schwartz wrote: I am really very curious as to what all of the interactions are, but I rather afraid you guys would have to write a small book to explain it all. The book is already written :) But who would like to search the archive for all the chapters? bye ago -- [EMAIL PROTECTED] http://www.gotti.org ICQ: 126018723
Re: cygwin 1.5.3-1 and xfree devel
Just to humor me, can someone else running Cygwin 1.5.3-1 confirm that a
simple X11 based hello world program is broken. If confirmed, I will
start digging to locate the offending functions, but I'd rather just wait
for Harold to release a 1.5.3-1 compiled Xfree86 set of packages. Thanks.
On Fri, 5 Sep 2003, Brian Ford wrote:
When compiled under cygwin 1.5.3-1, the following hello world Motif
program gets a SEGV in XtAppInitialize. Harold, what's your time
table for a 1.5.3 built Xfree so I can test and update lesstif? Thanks.
#include Xm/Xm.h
#include Xm/Label.h
XtAppContext context;
XmStringCharSet char_set=XmSTRING_DEFAULT_CHARSET;
main(argc, argv)
int argc;
char *argv[];
{
Widget toplevel, label;
XmString S;
/* Create an application shell. */
toplevel = XtAppInitialize(context,,NULL,0,argc,argv,NULL,
NULL,0);
/* Create the Label widget. */
label = XmCreateLabel(toplevel,label,NULL,0);
S = XmStringCreateLtoR(Hello World!, char_set);
XtVaSetValues(label,
XmNlabelString, S,
NULL);
XtRealizeWidget(toplevel);
XtAppMainLoop(context);
}
(gdb) r
Starting program: /home/ford/xtest.exe
Program received signal SIGSEGV, Segmentation fault.
0x in ?? () from
(gdb) bt
#0 0x in ?? () from
--
Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
Phone: 314-551-8460
Fax: 314-551-8444
RE: cygwin 1.5.3-1 and xfree devel
FWIW, It fails the same for me.
Latest 1.5.3-1 (as of this morning.)
Latest XFree86 (as of this morning.)
-David.
-
David Dawson
[EMAIL PROTECTED]
703-367-3885
-Original Message-
From: Brian Ford [mailto:[EMAIL PROTECTED]
Sent: Monday, September 08, 2003 12:16 PM
To: [EMAIL PROTECTED]
Subject: Re: cygwin 1.5.3-1 and xfree devel
Just to humor me, can someone else running Cygwin 1.5.3-1 confirm that a
simple X11 based hello world program is broken. If confirmed, I will
start digging to locate the offending functions, but I'd rather just
wait
for Harold to release a 1.5.3-1 compiled Xfree86 set of packages.
Thanks.
On Fri, 5 Sep 2003, Brian Ford wrote:
When compiled under cygwin 1.5.3-1, the following hello world Motif
program gets a SEGV in XtAppInitialize. Harold, what's your time
table for a 1.5.3 built Xfree so I can test and update lesstif?
Thanks.
#include Xm/Xm.h
#include Xm/Label.h
XtAppContext context;
XmStringCharSet char_set=XmSTRING_DEFAULT_CHARSET;
main(argc, argv)
int argc;
char *argv[];
{
Widget toplevel, label;
XmString S;
/* Create an application shell. */
toplevel = XtAppInitialize(context,,NULL,0,argc,argv,NULL,
NULL,0);
/* Create the Label widget. */
label = XmCreateLabel(toplevel,label,NULL,0);
S = XmStringCreateLtoR(Hello World!, char_set);
XtVaSetValues(label,
XmNlabelString, S,
NULL);
XtRealizeWidget(toplevel);
XtAppMainLoop(context);
}
(gdb) r
Starting program: /home/ford/xtest.exe
Program received signal SIGSEGV, Segmentation fault.
0x in ?? () from
(gdb) bt
#0 0x in ?? () from
--
Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
Phone: 314-551-8460
Fax: 314-551-8444
Re: How to get xdm working?
This doesnt work for me on Win2K. I have an account that is in the Administrators group. I copied the SYSTEM line and changed it to root: xdm fails the same way. I tried with the Administrator line too, no way. [EMAIL PROTECTED] ~ $ grep 544 /etc/passwd SYSTEM:*:18:544:,S-1-5-18:: root:*:18:544:,S-1-5-18:: Administrators:*:544:544:,S-1-5-32-544:: [EMAIL PROTECTED] ~ $ xdm Only root wants to run xdm This is probably a bug in xdm. It's been discussed on this list before. Cygwin (or, rather, Windows) doesn't have a root user, but does have users with various permissions usually attributed to root. On Win9x you can simply add a line for root to /etc/passwd (on WinNT/2k/XP you can copy the line for system and change the user name in that new line to root). = Sylvain Petreolle (spetreolle_at_users_dot_sourceforge_dot_net) ICQ #170597259 Say NO to software patents Dites NON aux brevets logiciels What if tomorrow the War could be over ? Morpheus, in Reloaded. ___ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com
Re: How to get xdm working?
On Mon, 8 Sep 2003, Sylvain Petreolle wrote: This doesnt work for me on Win2K. I have an account that is in the Administrators group. I copied the SYSTEM line and changed it to root: xdm fails the same way. I tried with the Administrator line too, no way. [EMAIL PROTECTED] ~ $ grep 544 /etc/passwd SYSTEM:*:18:544:,S-1-5-18:: root:*:18:544:,S-1-5-18:: Administrators:*:544:544:,S-1-5-32-544:: [EMAIL PROTECTED] ~ $ xdm Only root wants to run xdm This is probably a bug in xdm. It's been discussed on this list before. Cygwin (or, rather, Windows) doesn't have a root user, but does have users with various permissions usually attributed to root. On Win9x you can simply add a line for root to /etc/passwd (on WinNT/2k/XP you can copy the line for system and change the user name in that new line to root). Erm, yes. I forgot that some programs check for root by checking whether UID==0. You might need to also change the UID (3rd field) to 0. If xdm will try to do a seteuid(), use the SYSTEM line as your template (to have the same SID). Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton
Re: How to get xdm working?
On Mon, 8 Sep 2003, Igor Pechtchanski wrote: On Mon, 8 Sep 2003, Sylvain Petreolle wrote: This doesnt work for me on Win2K. I have an account that is in the Administrators group. I copied the SYSTEM line and changed it to root: xdm fails the same way. I tried with the Administrator line too, no way. [EMAIL PROTECTED] ~ $ grep 544 /etc/passwd SYSTEM:*:18:544:,S-1-5-18:: root:*:18:544:,S-1-5-18:: Administrators:*:544:544:,S-1-5-32-544:: [EMAIL PROTECTED] ~ $ xdm Only root wants to run xdm This is probably a bug in xdm. It's been discussed on this list before. Cygwin (or, rather, Windows) doesn't have a root user, but does have users with various permissions usually attributed to root. On Win9x you can simply add a line for root to /etc/passwd (on WinNT/2k/XP you can copy the line for system and change the user name in that new line to root). Erm, yes. I forgot that some programs check for root by checking whether UID==0. You might need to also change the UID (3rd field) to 0. If xdm will try to do a seteuid(), use the SYSTEM line as your template (to have the same SID). Igor Replying to myself here. Just realized: of course, you'll have to be logged in as the user you set as root before you run xdm... Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton
Re: How to get xdm working?
Replying to myself here. Just realized: of course, you'll have to be logged in as the user you set as root before you run xdm... Igor In this case, will a user that is in Administrators group be able to launch xdm ? = Sylvain Petreolle (spetreolle_at_users_dot_sourceforge_dot_net) ICQ #170597259 Say NO to software patents Dites NON aux brevets logiciels What if tomorrow the War could be over ? Morpheus, in Reloaded. ___ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com
Re: How to get xdm working?
This doesnt work for me on Win2K. I have an account that is in the Administrators group. I copied the SYSTEM line and changed it to root: xdm fails the same way. I tried with the Administrator line too, no way. [EMAIL PROTECTED] ~ $ grep 544 /etc/passwd SYSTEM:*:18:544:,S-1-5-18:: root:*:18:544:,S-1-5-18:: Administrators:*:544:544:,S-1-5-32-544:: I have run some tests. It is the UID which is tested not the name. So any user which as the 3rd column in the passwd file of 0 can run xdm as a daemon. NB For testing purposes I suggest that you remove the * in the second column (password) until you have the thing working. Also to login as root I would have thought you would need a home dir and a shell: root::0:544:optional_string:/root:/bin/bash
Re: How to get xdm working?
Replying to myself here. Just realized: of course, you'll have to be logged in as the user you set as root before you run xdm... Igor In this case, will a user that is in Administrators group be able to launch xdm ? Will xdm run under SYSTEM using xdm -nodaemon ?
Re: How to get xdm working?
On Tue, 9 Sep 2003 [EMAIL PROTECTED] wrote: Replying to myself here. Just realized: of course, you'll have to be logged in as the user you set as root before you run xdm... Igor In this case, will a user that is in Administrators group be able to launch xdm ? Will xdm run under SYSTEM using xdm -nodaemon ? xdm has this silly check for UID==0 before it can run. Once you convince xdm that it's running under root, it'll run under any user (even an unprivileged one), FWICS. Since xdm needs to access the desktop, you have to make sure you set up the service (and access permissions) properly, but otherwise nothing prevents xdm from being invoked by any particular user. Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton
Re: Alt Gr special characters on german keyboard work only sometimes
Hi Alexander Yupp, I'll send it as soon as I get back home. Do you need the Log of W2K as well with all the flags enabled? I've already done such a dump. Ok, here's the dump with all codes and times and the parallel dump of the xev output. I pressed '\' 4 times, first 2 worked, second 2 did not. Holger spy++.log Description: Binary data xev.log Description: Binary data
Amerika 'da Yasama Fýrsatýný Kacýrmayýn !!!
Green Card, Amerikan vatandaþý olmayan yabancýlara, Amerika Birleþik Devletleri 'nde süresiz (ömür boyu) oturma ve çalýþma izni saðlayan göçmen vizesidir. Fýrsatlar ülkesi Amerika 'da, 2003 yýlý DV-2005 çekiliþine katýlýp ABD 'de çalýþma, yaþama ve okuma fýrsatýný yakalamak mümkün. ABD heryýl 55.000 kiþiye çekiliþ sonucu göçmen vizesi yani Green Card vermektedir. Greencardline.com Consulting Baþvuru ücreti: Kredi kartlý : 22 Milyon + KDV =25.960.000 TL Posta Ödemeli : 26 Milyon + KDV =30.680.000 TL Ayrýntýlý bilgi için www.greencardline.com týklayýnýz...
src/winsup/w32api ChangeLog include/winuser.h ...
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-08 10:35:57 Modified files: winsup/w32api : ChangeLog winsup/w32api/include: winuser.h wingdi.h Log message: 2003-09-08 Justin Forest [EMAIL PROTECTED] * include/winuser.h (PM_QS_INPUT, PM_QS_POSTMESSAGE, PM_QS_PAINT, PM_QS_SENDMESSAGE): Add defines. 2003-09-08 Filip Navara [EMAIL PROTECTED] * include/wingdi.h (GRADIENT_FILL_RECT_H, GRADIENT_FILL_RECT_V, GRADIENT_FILL_TRIANGLE, GRADIENT_FILL_OP_FLAG): Add defines. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/w32api/ChangeLog.diff?cvsroot=srcr1=1.439r2=1.440 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/w32api/include/winuser.h.diff?cvsroot=srcr1=1.49r2=1.50 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/w32api/include/wingdi.h.diff?cvsroot=srcr1=1.28r2=1.29
src/winsup/cygwin ChangeLog cygwin.din syscall ...
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-08 20:08:53 Modified files: winsup/cygwin : ChangeLog cygwin.din syscalls.cc winsup/cygwin/include/cygwin: version.h Log message: * cygwin.din: Export endusershell, getusershell and setusershell. * syscalls.cc (getusershell): New function. (setusershell): Ditto. (endusershell): Ditto. * include/cygwin/version.h: Bump API minor number. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/ChangeLog.diff?cvsroot=srcr1=1.2046r2=1.2047 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/cygwin.din.diff?cvsroot=srcr1=1.94r2=1.95 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/syscalls.cc.diff?cvsroot=srcr1=1.282r2=1.283 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/include/cygwin/version.h.diff?cvsroot=srcr1=1.130r2=1.131
src/winsup/cygwin ChangeLog sigproc.cc
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-08 21:10:23 Modified files: winsup/cygwin : ChangeLog sigproc.cc Log message: * sigproc.cc (wait_sig_inited): Remove assertion since it is racy. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/ChangeLog.diff?cvsroot=srcr1=1.2047r2=1.2048 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/sigproc.cc.diff?cvsroot=srcr1=1.152r2=1.153
src/winsup/cygwin ChangeLog passwd.cc
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-08 21:44:53 Modified files: winsup/cygwin : ChangeLog passwd.cc Log message: * passwd.cc (getpwuid_r32): Initialize pw_comment field. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/ChangeLog.diff?cvsroot=srcr1=1.2048r2=1.2049 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/passwd.cc.diff?cvsroot=srcr1=1.71r2=1.72
src/winsup/cygwin ChangeLog passwd.cc
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-08 22:03:20 Modified files: winsup/cygwin : ChangeLog passwd.cc Log message: * passwe.cc (getpwnam_r): Initialize pw_comment field. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/ChangeLog.diff?cvsroot=srcr1=1.2049r2=1.2050 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/passwd.cc.diff?cvsroot=srcr1=1.72r2=1.73
src/winsup/cygwin ChangeLog dir.cc include/sys ...
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-09 00:21:16 Modified files: winsup/cygwin : ChangeLog dir.cc winsup/cygwin/include/sys: dirent.h Log message: * dir.cc (readdir): Reinstate setting of old ino field for legacy applications. * dirent.h (dirent): Rename unused field to __ino32. * passwd.cc (getpwnam_r): Initialize pw_comment field. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/ChangeLog.diff?cvsroot=srcr1=1.2050r2=1.2051 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/dir.cc.diff?cvsroot=srcr1=1.72r2=1.73 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/include/sys/dirent.h.diff?cvsroot=srcr1=1.2r2=1.3
src/winsup/cygwin ChangeLog dcrt0.cc exception ...
CVSROOT:/cvs/src Module name:src Changes by: [EMAIL PROTECTED] 2003-09-09 03:11:31 Modified files: winsup/cygwin : ChangeLog dcrt0.cc exceptions.cc fhandler_console.cc fork.cc sigproc.cc sigproc.h winsup/cygwin/include/sys: cygwin.h Log message: * dcrt0.cc (dlL_crt0_1): Set __argc_safe after __argc is absolutely know to be set. * exceptions.cc (sig_handle_tty_stop): Don't reset sigCONT event since it is reset automatically. * fork.cc (fork): Remove obsolete usage of PID_SPLIT_HEAP. * include/sys/cygwin.h: Ditto. * sigproc.cc (sig_send): Use sigframe init method to set frame since it checks for previous ownership of the frame. * sigproc.h (sigframe::init): Accept an is_exception argument. Patches: http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/ChangeLog.diff?cvsroot=srcr1=1.2051r2=1.2052 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/dcrt0.cc.diff?cvsroot=srcr1=1.178r2=1.179 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/exceptions.cc.diff?cvsroot=srcr1=1.164r2=1.165 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/fhandler_console.cc.diff?cvsroot=srcr1=1.112r2=1.113 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/fork.cc.diff?cvsroot=srcr1=1.111r2=1.112 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/sigproc.cc.diff?cvsroot=srcr1=1.153r2=1.154 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/sigproc.h.diff?cvsroot=srcr1=1.56r2=1.57 http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/winsup/cygwin/include/sys/cygwin.h.diff?cvsroot=srcr1=1.45r2=1.46
Re: [PATCH] getusershell, setusershell, endusershell
On Mon, Sep 08, 2003 at 10:26:07PM +0200, Corinna Vinschen wrote: FWIW, I've added these three functions to Cygwin. But there's something which might need another tweak. Getusershell returns shells from an internal list of default shells if /etc/shells isn't available. According to the Linux man pages, these are /bin/sh and /bin/csh. I've implemented this list accordingly, just adding /usr/bin paths: /bin/sh/usr/bin/sh /bin/csh /usr/bin/csh However, I had another look into the Solaris man pages. According to these man pages, the list on Solaris 9 looks a bit more complete: /bin/bash /usr/bin/bash /bin/csh /usr/bin/csh /bin/jsh /usr/bin/jsh /bin/ksh /usr/bin/ksh /bin/pfcsh /usr/bin/pfcsh /bin/pfksh /usr/bin/pfksh /bin/pfsh /usr/bin/pfsh /bin/sh/usr/bin/sh /bin/tcsh /usr/bin/tcsh /bin/zsh /usr/bin/zsh /sbin/jsh /sbin/sh /usr/xpg4/bin/sh Wow. So the default shell list contains any shell which might be shipped with Solaris and perhaps some more. Should we do the same in Cygwin, adding all shells to the internal default shell list, which are part of the Cygwin distro? This would bloat the list to something like /bin/sh/usr/bin/sh /bin/bash /usr/bin/bash /bin/csh /usr/bin/csh /bin/tcsh /usr/bin/tcsh /bin/ksh /usr/bin/ksh /bin/pdksh /usr/bin/pdksh /bin/zsh /usr/bin/zsh Or should we keep the short list as on Linux, knowing that our postinstall scripts add and manipulate /etc/shells as needed? I'm for anything which keeps the size of the DLL down. If we don't need to have an internal table, then I think we should rely on the postinstall scripts. cgf
fix getpwuid_r() and getpwnam_r()
I have not rebuilt cygwin to test this patch, but I think that the problem and the fix are obvious: pw_comment is not returned or initialized by these reentrant functions. The problem was discovered when debugging a perl test failure on cygwin: http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-09/msg00500.html Thanks, Radu Greab --- cygwin-1.5.3-1/winsup/cygwin/passwd.cc.orig Sat Jun 21 03:12:35 2003 +++ cygwin-1.5.3-1/winsup/cygwin/passwd.cc Tue Sep 9 00:22:50 2003 @@ -171,7 +171,8 @@ /* check needed buffer size. */ size_t needsize = strlen (temppw-pw_name) + strlen (temppw-pw_dir) + strlen (temppw-pw_shell) + strlen (temppw-pw_gecos) + - strlen (temppw-pw_passwd) + 5; + strlen (temppw-pw_passwd) + + strlen (temppw-pw_comment) + 6; if (needsize bufsize) return ERANGE; @@ -184,11 +185,13 @@ pwd-pw_shell = pwd-pw_dir + strlen (temppw-pw_dir) + 1; pwd-pw_gecos = pwd-pw_shell + strlen (temppw-pw_shell) + 1; pwd-pw_passwd = pwd-pw_gecos + strlen (temppw-pw_gecos) + 1; + pwd-pw_comment = pwd-pw_passwd + strlen (temppw-pw_passwd) + 1; strcpy (pwd-pw_name, temppw-pw_name); strcpy (pwd-pw_dir, temppw-pw_dir); strcpy (pwd-pw_shell, temppw-pw_shell); strcpy (pwd-pw_gecos, temppw-pw_gecos); strcpy (pwd-pw_passwd, temppw-pw_passwd); + strcpy (pwd-pw_comment, temppw-pw_comment); return 0; } @@ -228,7 +231,8 @@ /* check needed buffer size. */ size_t needsize = strlen (temppw-pw_name) + strlen (temppw-pw_dir) + strlen (temppw-pw_shell) + strlen (temppw-pw_gecos) + - strlen (temppw-pw_passwd) + 5; + strlen (temppw-pw_passwd) + + strlen (temppw-pw_comment) + 6; if (needsize bufsize) return ERANGE; @@ -241,11 +245,13 @@ pwd-pw_shell = pwd-pw_dir + strlen (temppw-pw_dir) + 1; pwd-pw_gecos = pwd-pw_shell + strlen (temppw-pw_shell) + 1; pwd-pw_passwd = pwd-pw_gecos + strlen (temppw-pw_gecos) + 1; + pwd-pw_comment = pwd-pw_passwd + strlen (temppw-pw_passwd) + 1; strcpy (pwd-pw_name, temppw-pw_name); strcpy (pwd-pw_dir, temppw-pw_dir); strcpy (pwd-pw_shell, temppw-pw_shell); strcpy (pwd-pw_gecos, temppw-pw_gecos); strcpy (pwd-pw_passwd, temppw-pw_passwd); + strcpy (pwd-pw_comment, temppw-pw_comment); return 0; }
Re: fix getpwuid_r() and getpwnam_r()
On Tue, Sep 09, 2003 at 12:36:17AM +0300, Radu Greab wrote: I have not rebuilt cygwin to test this patch, but I think that the problem and the fix are obvious: pw_comment is not returned or initialized by these reentrant functions. The problem was discovered when debugging a perl test failure on cygwin: Actually, I think the fix is more obvious that yours since pw_comment isn't used. I think yours would result in a SEGV. cgf Index: passwd.cc === RCS file: /cvs/src/src/winsup/cygwin/passwd.cc,v retrieving revision 1.71 diff -u -p -r1.71 passwd.cc --- passwd.cc 16 Jun 2003 03:24:11 - 1.71 +++ passwd.cc 8 Sep 2003 21:43:42 - @@ -183,6 +183,7 @@ getpwuid_r32 (__uid32_t uid, struct pass pwd-pw_dir = pwd-pw_name + strlen (temppw-pw_name) + 1; pwd-pw_shell = pwd-pw_dir + strlen (temppw-pw_dir) + 1; pwd-pw_gecos = pwd-pw_shell + strlen (temppw-pw_shell) + 1; + pwd-pw_comment = NULL; pwd-pw_passwd = pwd-pw_gecos + strlen (temppw-pw_gecos) + 1; strcpy (pwd-pw_name, temppw-pw_name); strcpy (pwd-pw_dir, temppw-pw_dir);
Re: fix getpwuid_r() and getpwnam_r()
Christopher Faylor wrote: Actually, I think the fix is more obvious that yours since pw_comment isn't used. I'm just a monkey which thought that pw_comment is used. I think yours would result in a SEGV. Yes, but the SEGV would have been seen in cygwin, not in perl :) Thanks, Radu Greab
Fixing a security hole in mount table.
This is the first in a series of patches fixing security holes associated with the file mappings in the core of Cygwin. I hope the explanations below are clear! Background on the mount table: System and user mounts are kept in a FileMapping, shared by all programs started from Windows by a given user, and their descendants, even after seteuid. User mounts are stored permanently under the HKCU registry. The user pointed to by HKCU depends on the user SID in the current thread access token. It changes during the seteuid call. So a seteuid'ed user creating a user mount will make an entry in the mount file mapping of its ancestor, and will also store it in its own user registry. That makes for a weird combination. For consistency the mount file mapping should be recreated following seteuid. The file mapping should be named after the SID (on NT), not the user name, to avoid name aliasing. To make sure HKCU is meaningful loaded, load_registry_hive should be called in seteuid, calling it in spawn is too late. The current code also creates a security gap that is easy to exploit: Telnet in as a non-privileged user. Create a $HOME/etc directory and cp -R /etc to it, give 777 access. Edit $HOME/etc/passwd and remove all the passwords. mount -u `cygpath -m $HOME/etc` /etc Bingo, all the programs running as SYSTEM will now switch to the new mapping for /etc. One can telnet in as any user without a password. (FWIW, here are two extra observations: - programs running as SYSTEM cannot umount -u /etc [because it's not under their HKCU], but they can mount -f -u over it. - programs running as SYSTEM use the user mounts of the .Default user, at least on NT4.) This simple attack would not work with the change described above, but the SYSTEM file mapping could still be manipulated by writing a program that directly accesses it. The solution is to specify appropriate security attributes when creating the file mapping. Using sec_none is fine, don't worry about the name. When initially testing the patch I found that the user mounts were set correctly from sshd but not from telnetd (with or without passwd). Telnetd (login) was producing the following error when reading HKCU: 255 1183303 [main] a 266 mount_table_initialize: initializing mount table 432 1183735 [main] a 266 reg_key::build_reg: failed to create key SOFTWARE in the registry 422 1184157 [main] a 266 reg_key::build_reg: failed to create key SOFTWARE in the registry 772 1184929 [main] a 266 mount_info::read_mounts: RegEnumKeyEx failed, error 6! 778 1185707 [main] a 266 mount_info::add_item: e:\[e:], /[/], 0xA = XXX system mounts OK reg_key::build_reg does not print %E, but gdb reveals it is 5. I then found out that HKCU can behave weirdly http://support.microsoft.com/default.aspx?scid=kb;en-us;199190 and implemented the short term workaround outlined therein. According to MS we should not use HKCU if several thread access it. The attached patch is the minimum that fixes the problem. A follow up patch, touching more files, will cleanup some details. Including it at this stage would be needlessly confusing. In summary this patch: 1) moves load_registry_hive from spawn_guts() to seteuid(). 2) moves all the mount table initialization from memory_init into a new function, user_shared_initialize [this choice of name will become clear later], which is then called from both memory_init and seteuid. 3) Adds an optional security attributes argument to open_shared. 4) fixes an unrelated (and currently non exposed) bug in security.h (sec_user). Pierre 2003-09-08 Pierre Humblet [EMAIL PROTECTED] * shared_info.h: Include security.h. (open_shared): Add psa argument. (user_shared_initialize): New declaration. * security.h: Add _SECURITY_H guard. (sec_user): Use sec_none in the no ntsec case. * spawn.cc (spawn_guts): Remove call to load_registry_hive. * syscalls (seteuid32): If warranted, call load_registry_hive, user_shared_initialize and RegCloseKey(HKEY_CURRENT_USER). * shared.cc (user_shared_initialize): New. (open_shared): Add and use psa argument. (memory_init): Move mount table initialization to user_shared_initialize. Call it.Index: shared_info.h === RCS file: /cvs/src/src/winsup/cygwin/shared_info.h,v retrieving revision 1.30 diff -u -p -r1.30 shared_info.h --- shared_info.h 3 Sep 2003 14:15:55 - 1.30 +++ shared_info.h 9 Sep 2003 00:38:28 - @@ -9,6 +9,7 @@ Cygwin license. Please consult the file details. */ #include tty.h +#include security.h /* Mount table entry */ @@ -189,4 +190,7 @@ struct console_state #endif char *__stdcall shared_name (char *, const char *, int); -void *__stdcall open_shared (const char *name, int n, HANDLE shared_h, DWORD size, shared_locations); +void
Re: Fixing a security hole in mount table.
On Mon, Sep 08, 2003 at 08:46:06PM -0400, Pierre A. Humblet wrote: This is the first in a series of patches fixing security holes associated with the file mappings in the core of Cygwin. I hope the explanations below are clear! Yes they are, thanks. I can't comment on the security stuff but everything else looks good to me. I'll let Corinna have the final say on this. I wonder if it is time to bite the bullet and get rid of user-mode mounts entirely. Or maybe disallow them in suid'ed sessions? They are always going to be a security hole AFAICT. cgf
Re: Fixing a security hole in mount table.
At 09:11 PM 9/8/2003 -0400, you wrote: On Mon, Sep 08, 2003 at 08:46:06PM -0400, Pierre A. Humblet wrote: This is the first in a series of patches fixing security holes associated with the file mappings in the core of Cygwin. I hope the explanations below are clear! Yes they are, thanks. I can't comment on the security stuff but everything else looks good to me. I'll let Corinna have the final say on this. I wonder if it is time to bite the bullet and get rid of user-mode mounts entirely. Or maybe disallow them in suid'ed sessions? They are always going to be a security hole AFAICT. Yep, the same thought has crossed my mind. However I now believe that with the patch the user mounts do not pose a security issue. And they are really useful! If we keep accessing HKCU as we do now, we should make a note in the doc (mount man page ?) that SYSTEM uses the user mounts of .Default. This could be a source of hard to explain behavior. Pierre
Re: Fixing a security hole in mount table.
On Tue, Sep 09, 2003 at 12:12:11AM -0400, Pierre A. Humblet wrote: At 09:11 PM 9/8/2003 -0400, you wrote: On Mon, Sep 08, 2003 at 08:46:06PM -0400, Pierre A. Humblet wrote: This is the first in a series of patches fixing security holes associated with the file mappings in the core of Cygwin. I hope the explanations below are clear! Yes they are, thanks. I can't comment on the security stuff but everything else looks good to me. I'll let Corinna have the final say on this. I wonder if it is time to bite the bullet and get rid of user-mode mounts entirely. Or maybe disallow them in suid'ed sessions? They are always going to be a security hole AFAICT. Yep, the same thought has crossed my mind. However I now believe that with the patch the user mounts do not pose a security issue. I can't see how a feature which allows any user to redefine what /etc or / is could not be a security issue. cgf
Re: [ANNOUNCEMENT] Updated: zsh-4.0.7-2
you've downloaded setup.exe, run it and select Editors and then click on *giggle* - a little too much cut and paste from mine i see ... almost made this mistake myself copying from igors... Gareth -- 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: [PATCH] gcc3/ld patch for direct-linking-to-dll and auto-importsupport
Hi, From: Nick Clifton nickc at redhat dot com Hi Ralf, while compiling trolltechs qt/xfree library with gcc3 (3.2x) on cygwin I recognized, that the auto-import stuff in combination of recent ld does not work in case of const variables in a dll when using direct linking to a dll, because gcc put those variables into a readonly, that means the .text section. The patch and a testcase is appended. A documentation could be found in the patch file. Are you sure that the current linker does not work ? When I tried it, using the test case you supplied, but without the patch applied, I got this: % make ../gcc/g++ -B ../gcc/ -g -save-temps -c -o dll.o dll.cc ../gcc/g++ -B ../gcc/ -g --shared -L ../i686-pc-cygwin/newlib -L ../i686-pc-cygwin/libstdc++-v3/src/.libs dll.o -o dll.dll ../gcc/g++ -B ../gcc/ -g -save-temps -c -o client.o client.cc ../gcc/g++ -B ../gcc/ -g -o client client.o -L ../i686-pc-cygwin/newlib -L ../i686-pc-cygwin/libstdc++-v3/src/.libs -L. -ldll Info: resolving _var by linking to __imp__var (auto-import) Info: resolving test::var by linking to __imp___ZN4test3varE (auto-import) There must be four auto imported vars. Yes, but try running ./client.exe. Since the const data variables are in .text section they are treated as functions, not data. They are imported without warning, but not as data. Ralf, what about a less intrusive approach? Compile dll code with -fdata-sections. This will put const data in .rdata$foo sections. Then, make pe_implied_import_dll() scan .rdata as well as .data and .bss. for data symbols I've tried that with your testcase and it seems to work. What gcc release you are using ? It seems that gcc3.2 (which is the recent cygwin release) does not support this. (See the ! below) Are there any other gcc options to archive this ? $ gcc --version gcc (GCC) 3.2 20020927 (prerelease) $ make g++ -save-temps -fdata-sections-c -o dll.o dll.cc $ less dll.s .file dll.cc .globl _var .section.data$var,w .align 4 _var: .long 1 .globl _constvar !.section.text$constvar, .align 4 _constvar: .long 2 !.section.text$_ZZ4funcvE14staticconstvar, .align 4 _ZZ4funcvE14staticconstvar: .long 2 .text .align 2 .globl __Z4funcv .def__Z4funcv; .scl2; .type 32; .endef __Z4funcv: LFB1: pushl %ebp LCFI0: movl%esp, %ebp LCFI1: subl$4, %esp LCFI2: movl$3, -4(%ebp) movl$1, %eax leave ret LFE1: .globl __ZN4test3varE .section.data$_ZN4test3varE,w .align 4 __ZN4test3varE: .long 4 .globl __ZN4test14staticconstvarE !.section.text$_ZN4test14staticconstvarE, .align 4 __ZN4test14staticconstvarE: .long 5 -- 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/
rxvt hanging after latest cygwin update
Hey all, I just upgraded my cygwin installation to the new release made available at the beginning of the month. Since this upgrade, my rxvt launch script doesn't work anymore. I have the following dos-batch script: start /i /min /separate /b $PATH2CYG\bash --login -c /startrxvt.sh 12 The startrxvt.sh script looks like this: #!/bin/bash #Redirect IO to allow parent terminal to closr exec 0 /dev/null exec 1 ~/.xdmcp-errors exec 21 # Select font size FONTSIZE=12 if [ -n $1]; then FONTSIZE=$1; fi rxvt -bg black -fg green -fn -*-Lucida Console-*-*-*-*-$FONTSIZE-*-*-*-*-*-*-* -sr -sl 1024 -e /bin/bash --login -i -x # The rxvt window opens but hangs. I don't see the scroll bar and there are no prompts. The window doesn't refresh, resize... nothing happens. I have to kill it in order to get rid of it. This script has worked for a long time and works on another machine which I haven't upgraded yet. My system is running WindowsXP-SP1 If i run the commands manually, everything works ok. Thnks for your help --- Thomas Fischer, MCSE mailto:[EMAIL PROTECTED] mailto:[EMAIL PROTECTED] Apple, WinNT, e-Mail, Groupware mailto:[EMAIL PROTECTED] -- 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/
Marvelous CYGWIN
Hi, I love my CyGwin thanx to all concerned. A few brief questions quand-meme! 1) Although I have the following setting in my .profile .bashrc it is apparently ignored:- set completion-ignore-case on 2) The following alias causes the shell to exit if there are two many files x*, can I set more memory is there a workaround? (or should I be using XARGS? alias vewx='gvim.exe `l\s -t c:/aaa/x* | head -1 ` ' zzapper -- vim -c :%s/^/WhfgTNabgureRIvzSUnpxre/|:%s/[R-T]/ /Ig|:normal ggVGg? http://www.vim.org/tips/tip.php?tip_id=305 Best of Vim Tips -- 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: Marvelous CYGWIN
zzapper wrote: 1) Although I have the following setting in my .profile .bashrc it is apparently ignored:- set completion-ignore-case on This goes in $HOME/.inputrc since it's a GNU readline parameter, not a shell setting. The following alias causes the shell to exit if there are two many files x*, can I set more memory is there a workaround? (or should I be using XARGS? alias vewx='gvim.exe `l\s -t c:/aaa/x* | head -1 ` ' I can't reproduce that, what do you mean it exits? Your shell window closes? Do these files that match the glob have spaces or any other weird characters? Brian -- 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/
ps -W not working after latext cygwin DLL update
After upgrading to the latest cygwin DLL, the command ps -W (or ps --windows) returns the same list as ps, i.e. - no Windows native processes are displayed. Is it a known bug? -- 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: cygwin doesn't show .. in all directories
On 08 September 2003 00:12 -0400 Christopher Faylor wrote: Any snapshot greater than or equal to 2003-08-08 should have this fix in it. ^^ 8th August 2003? s/2003-08-08/2003-09-08/ -- Mark Himsley -- 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/
ABC questions
Hi, (B (BSorry my ABC question on sygwin. (BWhen I try to compile X window program (Bby (Bgcc -o target test.c -lX11 (Bld failed. The message is (B/usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../../i686-pc-cygwin/bin/ld: (Bcannot find -lX11 (B (BCould someone let me know the reason? (B (BOne more is can we make target instead of target.exe when we use -o (Btarget option? (B (BThank you, (B (BShige (B (B (B (B (B-- (BUnsubscribe info: http://cygwin.com/ml/#unsubscribe-simple (BProblem reports: http://cygwin.com/problems.html (BDocumentation: http://cygwin.com/docs.html (BFAQ: http://cygwin.com/faq/
Re: ps -W not working after latext cygwin DLL update
On Mon, Sep 08, 2003 at 12:00:46PM +0300, Goren Il wrote: After upgrading to the latest cygwin DLL, the command ps -W (or ps --windows) returns the same list as ps, i.e. - no Windows native processes are displayed. It does for me. Sure you're using the right ps? Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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: Marvelous CYGWIN
On Mon, 08 Sep 2003 01:14:38 -0700, Brian Dessent [EMAIL PROTECTED] wrote: set completion-ignore-case on This goes in $HOME/.inputrc since it's a GNU readline parameter, not a shell setting. Solved thanx The following alias causes the shell to exit if there are two many files x*, can I set more memory is there a workaround? (or should I be using XARGS? alias vewx='gvim.exe `l\s -t c:/aaa/x* | head -1 ` ' I can't reproduce that, what do you mean it exits? Your shell window closes? Do these files that match the glob have spaces or any other weird characters? I run CygWin from a Dos Window, if I get the above problem, I get a message logout . But hold on problem seems to have gone way?!? zzapper -- vim -c :%s/^/WhfgTNabgureRIvzSUnpxre/|:%s/[R-T]/ /Ig|:normal ggVGg? http://www.vim.org/tips/tip.php?tip_id=305 Best of Vim Tips -- 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: ABC questions
On Mon, Sep 08, 2003 at 07:30:25PM +0900, Shigeyoshi Tsutsui wrote: Hi, Sorry my ABC question on sygwin. When I try to compile X window program by gcc -o target test.c -lX11 ld failed. The message is /usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../../i686-pc-cygwin/bin/ld: cannot find -lX11 Could someone let me know the reason? libX11.a is not in the lib search path. Use the -L option or the LD_LIBRARY_PATH environment variable. See `info gcc' for more info. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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: latest cygcheck -c is expensive
Chris, On Sun, Sep 07, 2003 at 12:30:21AM -0400, Christopher Faylor wrote: On Sun, Sep 07, 2003 at 12:28:05AM -0400, Christopher Faylor wrote: On Fri, Sep 05, 2003 at 08:42:49PM -0400, Jason Tishler wrote: On Fri, Sep 05, 2003 at 03:01:27PM -0400, Christopher Faylor wrote: If the only concern is that cygcheck takes a long time, now, then, like I said, that is something that can be rectified. Yes, the above is my real concern. I just hacked cygcheck to avoid calling gzip and use a mingw libz.a (courtesy of Chuck Wilson) and it takes 26 seconds to complete on my dual PIII 733MHZ WinXP system. I have most packages installed. Is that still too slow? I should add that the previous version took 1 minute, 9 seconds. Approximately 3 time faster is a significant improvement, but IMO, anything more than a few seconds is still too expensive. What about an option to disable the sanity checking? In this way, Cygwin users do not need to learn new tricks, but powers user still get the control they desire. BTW, I tried to give cygcheck from the 2003-Sep-08 snapshot a ride. The ChangeLog and diff seem to indicate the required changes have been completed. Unfortunately, my timing showed no difference and objdump showed no dependency on Mingw zlib: $ objdump -p cygcheck.exe | fgrep 'DLL Name:' DLL Name: msvcrt.dll DLL Name: msvcrt.dll DLL Name: ADVAPI32.DLL DLL Name: KERNEL32.dll Is the above cockpit error on my part? Thanks, Jason -- PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers Fingerprint: 7A73 1405 7F2B E669 C19D 8784 1AFD E4CC ECF4 8EF6 -- 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: Pondering on cygcheck -c output; Some Incomplete packages + multiple versions of libxerces
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski Hannu, Replies inline below: On Sun, 7 Sep 2003, Hannu E K Nevalainen (garbage mail) wrote: Hmm... should I be concerned about these? I've seen diff and XFree86-base mentioned - and disregarded as important in previous messages here. But not the others. $ uname -a CYGWIN_NT-5.0 P450 1.5.3(0.90/3/2) 2003-09-01 13:15 i686 unknown unknown Cygwin $ cygcheck -c ~/c_c.txt gzip: stdout: Bad file descriptor ^ Hmm, I'm not sure I like this. I'll investigate. $ grep -i incomplete ~/c_c.txt diff1.0-1 Incomplete mod_auth_mysql 1.11-1 Incomplete mod_auth_ntsec 1.7-1 Incomplete mod_dav 1.0.3-1.3.6-1 Incomplete mod_php44.2.0-2Incomplete mod_ssl 2.8.8-1.3.24-1 Incomplete newlib-man 20020801 Incomplete tetex-base 2.0.2-1Incomplete tetex-beta 20020911-1 Incomplete tetex-tiny 2.0.2-1Incomplete texmf 20020911-1 Incomplete texmf-base 20020911-1 Incomplete texmf-doc 20020911-1 Incomplete texmf-extra 20020911-1 Incomplete texmf-tiny 20020911-1 Incomplete XFree86-base4.3.0-1Incomplete I tried a re-install of the above packages, but that doesn't seem to change anything... [snip] Try adding a -v (--verbose) option -- it should tell you exactly why those packages are deemed incomplete, e.g., what files are missing. FWIW, I have tetex-* packages installed on my system with no missing files. I believe the texmf-* packages are upgrade helpers and might be empty. Cygwin Package Information Last downloaded files to: E:\_SystemBackup\Programvara_fri\www.cygwin.com\cygwin-downloads Last downloaded files from: http://mirrors.sunsite.dk/cygwin Package VersionStatus snip Can't open file list /etc/setup/diff.lst.gz for package diff diff1.0-1 Incomplete snip Missing file: /usr/lib/apache/new/mod_auth_mysql.dll from package mod_auth_mysql mod_auth_mysql 1.11-1 Incomplete Missing file: /usr/lib/apache/new/mod_auth_ntsec.dll from package mod_auth_ntsec mod_auth_ntsec 1.7-1 Incomplete Missing file: /usr/lib/apache/new/libdav.dll from package mod_dav mod_dav 1.0.3-1.3.6-1 Incomplete Missing file: /usr/lib/apache/new/libphp4.dll from package mod_php4 mod_php44.2.0-2Incomplete Missing file: /usr/lib/apache/new/libssl.dll from package mod_ssl mod_ssl 2.8.8-1.3.24-1 Incomplete snip Can't open file list /etc/setup/newlib-man.lst.gz for package newlib-man newlib-man 20020801 Incomplete snip Can't open file list /etc/setup/tetex-beta.lst.gz for package tetex-beta tetex-beta 20020911-1 Incomplete snip Can't open file list /etc/setup/texmf.lst.gz for package texmf texmf 20020911-1 Incomplete Can't open file list /etc/setup/texmf-base.lst.gz for package texmf-base texmf-base 20020911-1 Incomplete Can't open file list /etc/setup/texmf-doc.lst.gz for package texmf-doc texmf-doc 20020911-1 Incomplete Can't open file list /etc/setup/texmf-extra.lst.gz for package texmf-extra texmf-extra 20020911-1 Incomplete Can't open file list /etc/setup/texmf-tiny.lst.gz for package texmf-tiny texmf-tiny 20020911-1 Incomplete snip Can't open file list /etc/setup/XFree86-base.lst.gz for package XFree86-base XFree86-base4.3.0-1Incomplete snip Additionally, there seems to be multiple versions of libxerces installed. I'm not worried as I don't use it (at least currently), but others might... $ grep -i libxe c_c-reinstalled.txt libxerces-c21 2.1.0-1OK libxerces-c22 2.2.0-1OK libxerces-c23 2.3.0-2OK These indicate that you've used setup.exe to install various versions of libxerces from somewhere other than the official mirror -- check your /var/log/setup.log, it should have a record of this. Igor $ wc -l /var/log/setup.log 13836 /var/log/setup.log $ grep -i libxerces /var/log/setup.log snip 2003/09/07 13:35:38 Uninstalling libxerces-c21 2003/09/07 13:35:38 Uninstalling libxerces-c22 2003/09/07 13:35:38 Uninstalling libxerces-c23 2003/09/07 13:43:47 Installing file://E:\_SystemBackup\Programvara_fri\www.cygwin.com\cygwin-downloads/http %3a%2f%2fmirrors.sunsite.dk%2fcygwin/release/xerces-c/libxerces-c21/libxerce s-c21-2.1.0-1.tar.bz2 2003/09/07 13:43:48 Installing
Re: pthread_mutex_lock error
Sad but true. When I finally did what I should have done to start with, I quickly found that I had managed to include the wrong version of cygwin1.dll in my binary. I then found I also had to add cygreadline5.dll and cygncurses6.dll, but that was easy enough. No doubt the latter was documented somewhere else I didn't look, ;) but it didn't take any time to fix anyway. The reason for not running cygcheck was that this was someone else's box and didn't have cygwin installed. Once it was installed, it didn't take me long to spot that the two versions of cygwin1.dll had different timestamps, which quickly led me to the resolution. So thank you for your help and patience. Rob Christopher Faylor wrote: On Thu, Sep 04, 2003 at 04:31:35PM +0100, Rob Clack wrote: So I'm stuck now as to how to proceed. Or does your return post, Christopher, mean the answer is there if I only look in the right place? You said that you had no idea how to diagnose the problem. I pointed you at the page which talks about diagnosing problems. You haven't read it thoroughly, obviously. It talks about information that should be in every problem report, like attached cygcheck output. -- Please use the resources at cygwin.com rather than sending personal email. Special for spam email harvesters: send email to [EMAIL PROTECTED] and be permanently blocked from mailing lists at sources.redhat.com -- ~~ The mouth: natural home of the foot ~~ Rob ClackAcedb Development, Informatics Group email: [EMAIL PROTECTED]Wellcome Trust Sanger Institute Tel: +44 1223 494780 Wellcome Trust Genome Campus Fax: +44 1223 494919 Hinxton CambridgeCB10 1SA -- 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: bug in ar when creating libraries on cross linked filesystems
On Mon, 2003-09-08 at 23:18, [EMAIL PROTECTED] wrote: Hello, r Services for Unix from Microsoft is buggy. Maybe it is more successful to make the rename command in libc more robust than waiting for ... Why not just stop using SFU, and use cygwin alone? Cygwin's ranlib doesn't have this problem, AFAIK. Rob -- GPG key available at: http://members.aardvark.net.au/lifeless/keys.txt. signature.asc Description: This is a digitally signed message part
Re: cygwin doesn't show .. in all directories
On Mon, Sep 08, 2003 at 10:30:12AM +0100, Mark Himsley wrote: On 08 September 2003 00:12 -0400 Christopher Faylor wrote: Any snapshot greater than or equal to 2003-08-08 should have this fix in it. ^^ 8th August 2003? s/2003-08-08/2003-09-08/ Sigh. Yes. Sorry if my typo caused panic or confusion. -- 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: Pondering on cygcheck -c output; Some Incomplete packages + multiple versions of libxerces
On Mon, 8 Sep 2003, Hannu E K Nevalainen (garbage mail) wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski Hannu, Replies inline below: On Sun, 7 Sep 2003, Hannu E K Nevalainen (garbage mail) wrote: Hmm... should I be concerned about these? I've seen diff and XFree86-base mentioned - and disregarded as important in previous messages here. But not the others. ^ I assume you mean unimportant. $ uname -a CYGWIN_NT-5.0 P450 1.5.3(0.90/3/2) 2003-09-01 13:15 i686 unknown unknown Cygwin $ cygcheck -c ~/c_c.txt gzip: stdout: Bad file descriptor ^ Hmm, I'm not sure I like this. I'll investigate. As CGF said, the latest version won't exec the gzip program, so never mind. $ grep -i incomplete ~/c_c.txt diff1.0-1 Incomplete mod_auth_mysql 1.11-1 Incomplete mod_auth_ntsec 1.7-1 Incomplete mod_dav 1.0.3-1.3.6-1 Incomplete mod_php44.2.0-2Incomplete mod_ssl 2.8.8-1.3.24-1 Incomplete newlib-man 20020801 Incomplete tetex-base 2.0.2-1Incomplete tetex-beta 20020911-1 Incomplete tetex-tiny 2.0.2-1Incomplete texmf 20020911-1 Incomplete texmf-base 20020911-1 Incomplete texmf-doc 20020911-1 Incomplete texmf-extra 20020911-1 Incomplete texmf-tiny 20020911-1 Incomplete XFree86-base4.3.0-1Incomplete I tried a re-install of the above packages, but that doesn't seem to change anything... [snip] Try adding a -v (--verbose) option -- it should tell you exactly why those packages are deemed incomplete, e.g., what files are missing. FWIW, I have tetex-* packages installed on my system with no missing files. I believe the texmf-* packages are upgrade helpers and might be empty. Cygwin Package Information Last downloaded files to: E:\_SystemBackup\Programvara_fri\www.cygwin.com\cygwin-downloads Last downloaded files from: http://mirrors.sunsite.dk/cygwin Package VersionStatus snip Can't open file list /etc/setup/diff.lst.gz for package diff diff1.0-1 Incomplete snip Missing file: /usr/lib/apache/new/mod_auth_mysql.dll from package mod_auth_mysql mod_auth_mysql 1.11-1 Incomplete Missing file: /usr/lib/apache/new/mod_auth_ntsec.dll from package mod_auth_ntsec mod_auth_ntsec 1.7-1 Incomplete Missing file: /usr/lib/apache/new/libdav.dll from package mod_dav mod_dav 1.0.3-1.3.6-1 Incomplete Missing file: /usr/lib/apache/new/libphp4.dll from package mod_php4 mod_php44.2.0-2Incomplete Missing file: /usr/lib/apache/new/libssl.dll from package mod_ssl mod_ssl 2.8.8-1.3.24-1 Incomplete Ok, I see. These packages are most likely performing the same postinstall directory renaming trick as the one I complained about in http://cygwin.com/ml/cygwin/2003-09/msg00334.html. As before, suggestions on how to deal with this are welcome. Stipe? snip Can't open file list /etc/setup/newlib-man.lst.gz for package newlib-man newlib-man 20020801 Incomplete snip Can't open file list /etc/setup/tetex-beta.lst.gz for package tetex-beta tetex-beta 20020911-1 Incomplete snip Can't open file list /etc/setup/texmf.lst.gz for package texmf texmf 20020911-1 Incomplete Can't open file list /etc/setup/texmf-base.lst.gz for package texmf-base texmf-base 20020911-1 Incomplete Can't open file list /etc/setup/texmf-doc.lst.gz for package texmf-doc texmf-doc 20020911-1 Incomplete Can't open file list /etc/setup/texmf-extra.lst.gz for package texmf-extra texmf-extra 20020911-1 Incomplete Can't open file list /etc/setup/texmf-tiny.lst.gz for package texmf-tiny texmf-tiny 20020911-1 Incomplete snip Can't open file list /etc/setup/XFree86-base.lst.gz for package XFree86-base XFree86-base4.3.0-1Incomplete snip All of the above should be reported as OK by the latest version. I'll add some more checking, though. Igor Additionally, there seems to be multiple versions of libxerces installed. I'm not worried as I don't use it (at least currently), but others might... $ grep -i libxe c_c-reinstalled.txt libxerces-c21 2.1.0-1OK libxerces-c22 2.2.0-1OK libxerces-c23 2.3.0-2OK These indicate that you've used
mozilla displays wrong headers from uw-imap (12/31/1969, blank subject)
Hey, all. Fetchmail gets my mail via pop3 and sends them to either tmail or dmail (I've used both.) The logs look just fine, but says bare LF only and I don't know if that is a problem. The mail ends up where the uw-imap readme says it will, in /var/mail/username. I log in successfully with both Mozilla and Pegasus Mail and the messages have blank headers except for the date, which is 12/31/1969. When I open the messages, the headers look okay, the date and time are accurate, but moving the message to another folder corrupts the message and, it appears, only moves the body over, not the headers. I have tried all kinds of fixes: - mounted mail directories in binary mode - messed with tmail and dmail, different arguments, no arguments - used procmail instead of fetchmail to send to tmail/dmail - tried formatting the messages with procmail before sending them to tmail/dmail in case the bare LF is a problem - looked at the mailbox to find any inconsitencies in the format I have read everything I could find and the answer eludes me. Any advice, please? Thanks for reading, Ali -- 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: cygwin build problems...(libintl)
[Crossposted to binutils, gdb, and cygwin for information purposes] Marcus Binny said: I am trying to build a cross compiler from the latest checked out sources. When I configure into the combined sources, it says that it doesn't need to build libintl.a (in intl/Makefile all: is all-no), but then it is trying to link binutils with ../intl/libintl.a instead of -lintl. I tried -lintl and it works. If I use configure --with-included-getttext, linking with the built libintl.a gives unresolved symobols problem like ./../intl/libintl.a(dcigettext.o)(.text+0xad7): In function `_nl_find_msg': /cygdrive/d/work/new_src/build/intl/../../combined/intl/dcigettext.c:940: undefined reference to `_libiconv' ./../intl/libintl.a(loadmsgcat.o)(.text+0x4e9): In function `_nl_init_domain_con v': /cygdrive/d/work/new_src/build/intl/../../combined/intl/loadmsgcat.c:863: undefined reference to `_libiconv_open' ./../intl/libintl.a(loadmsgcat.o)(.text+0x542): In function `_nl_free_domain_conv': /cygdrive/d/work/new_src/build/intl/../../combined/intl/loadmsgcat.c:896: undefined reference to `_libiconv_close' ./../intl/libintl.a(relocatable.o)(.text+0xfe): In function `libintl_set_relocation_prefix': /cygdrive/d/work/new_src/build/intl/../../combined/intl/relocatable.c:148: undefined reference to `_libiconv_set_relocation_prefix' To be sure, I had updated my cygwin installations also today. Now I am clueless. You will have two problems, with combined tree Cygwin build. The first is related to the current intl 0.12.1 directory the Zack Weinberg imported into the GCC sources. Neither binutils nor gdb have been updated for the new version and still are expecting version 0.10.4. This probably could be fixed by applying some configury wizardry to the binutils/gdb sources similar to those that were necessary to get cygwin to build on GCC again after it was imported. But no one has volunteered the time to do it yet. See http://gcc.gnu.org/ml/gcc-patches/2003-07/msg00781.html The workaround is to use the older copy of intl directory that comes with binutils or gdb instead of the copy that comes with GCC. Then GCC will ignore intl, but unless you actually require the international error messages, the resulting compiler will work. The second problem is with GCC itself and is related to -funit-at-a-time now being turned on at -O2. A cygwin bootstrap will ICE when the stage2 compiler attempts to compile _popcountsi2 in libgcc2. It might be a problem with the loop optimizer being miscompiled; I am in the process of filing a more detailed bug report. The work around for this is to make BOOT_CFLAGS=-O2 -g -fno-unit-a-time bootstrap Cheers, Kelley Cook -- 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: bug in ar when creating libraries on cross linked filesystems
Robert Collins wrote: Why not just stop using SFU, and use cygwin alone? Cygwin's ranlib doesn't have this problem, AFAIK. And how should I mount my NFS-home-directory on a windows PC without Services for Unix? Is it possible to mount an NFS directory with Cygwin? How can I access this directory from a non-cygwin application? Daniel -- 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: [ANNOUNCEMENT] NEW: mingw-zlib-1.1.4-4
On Mon, Sep 08, 2003 at 10:01:14AM -0400, Igor Pechtchanski wrote: This is good for a temp measure, but wouldn't this normally go into the mingw-runtime and gcc-mingw packages? Or is this going to be a separate package forever and ever? I don't mean to speak for Chuck but, after some discussion, it's intended to be a separate package. The intent was to not put a zlib directory in winsup, which would be the case with mingw-runtime. Ditto for the gcc sources. -- 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: [ANNOUNCEMENT] NEW: mingw-zlib-1.1.4-4
On Mon, 8 Sep 2003, Christopher Faylor wrote: On Mon, Sep 08, 2003 at 10:01:14AM -0400, Igor Pechtchanski wrote: This is good for a temp measure, but wouldn't this normally go into the mingw-runtime and gcc-mingw packages? Or is this going to be a separate package forever and ever? I don't mean to speak for Chuck but, after some discussion, it's intended to be a separate package. The intent was to not put a zlib directory in winsup, which would be the case with mingw-runtime. Ditto for the gcc sources. So basically this means that everyone wanting to compile Cygwin from CVS or source tarball will have to install this package, right? Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_[EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_[EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton -- 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: ps -W not working after latext cygwin DLL update
/usr/bin/ps.exe -v ps (cygwin) 1.11 Process Statistics Copyright 1996, 1997, 1998, 1999, 2000, 2001, 2002 Red Hat, Inc. Compiled on Sep 1 2003 On Mon, Sep 08, 2003 at 12:00:46PM +0300, Goren Il wrote: After upgrading to the latest cygwin DLL, the command ps -W (or ps --windows) returns the same list as ps, i.e. - no Windows native processes are displayed. It does for me. Sure you're using the right ps? Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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: mozilla displays wrong headers from uw-imap (12/31/1969, blank subject)
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Mr. Ali Nabavi Fetchmail gets my mail via pop3 and sends them to either tmail or dmail (I've used both.) The logs look just fine, but says bare LF only and I don't know if that is a problem. The mail ends up where the uw-imap readme says it will, in /var/mail/username. I log in successfully with both Mozilla and Pegasus Mail and the messages have blank headers except for the date, which is 12/31/1969. When I open the messages, the headers look okay, the date and time are accurate, but moving the message to another folder corrupts the message and, it appears, only moves the body over, not the headers. I'm no expert, but to me this looks as fetchmail and the other (cygwin!) tools uses LF line endings (Unix style). While Mozilla and Pegasus Mail (Windows executables most likely) expect it to be the Windows way (i.e. CR+LF). IMO a Mozilla and/or Pegasus Mail bug. /Hannu E K Nevalainen, B.Sc. EE - 59°16.37'N, 17°12.60'E --END OF MESSAGE-- -- 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: Pondering on cygcheck -c output; Some Incomplete packages + multiple versions of libxerces
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Mon, 8 Sep 2003, Hannu E K Nevalainen (garbage mail) wrote: From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Igor Pechtchanski On Sun, 7 Sep 2003, Hannu E K Nevalainen (garbage mail) wrote: Hmm... should I be concerned about these? I've seen diff and XFree86-base mentioned - and disregarded as important in previous messages here. But not the others. ^ I assume you mean unimportant. Regarded as unimportant might had been a better wording from me. Sigh: IMO dis... + un... = double negation :-] Well... English is a picky language, translate that into Swedish and it will be acceptable AND understandable ;-). SNIP Missing file: /usr/lib/apache/new/libssl.dll from package mod_ssl mod_ssl 2.8.8-1.3.24-1 Incomplete Ok, I see. These packages are most likely performing the same postinstall directory renaming trick as the one I complained about in http://cygwin.com/ml/cygwin/2003-09/msg00334.html. Right... I'll stay behind on this from now on. As before, suggestions on how to deal with this are welcome. Stipe? /Hannu E K Nevalainen, B.Sc. EE - 59?16.37'N, 17?12.60'E --END OF MESSAGE-- -- 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: [ANNOUNCEMENT] NEW: mingw-zlib-1.1.4-4
On Mon, Sep 08, 2003 at 10:30:08AM -0400, Igor Pechtchanski wrote: On Mon, 8 Sep 2003, Christopher Faylor wrote: I don't mean to speak for Chuck but, after some discussion, it's intended to be a separate package. The intent was to not put a zlib directory in winsup, which would be the case with mingw-runtime. Ditto for the gcc sources. So basically this means that everyone wanting to compile Cygwin from CVS or source tarball will have to install this package, right? If they want this functionality, yes. cgf -- 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: mozilla displays wrong headers from uw-imap (12/31/1969, blan k subject)
Apologies to Hannu, I clicked reply and not reply to all... -Original Message- From: Hannu E K Nevalainen (garbage mail) Subject: RE: mozilla displays wrong headers from uw-imap (12/31/1969, blank subject) Of Mr. Ali Nabavi Fetchmail gets my mail via pop3 and sends them to either tmail or dmail (I've used both.) The logs look just fine, but says bare LF only and I don't know if that is a problem. The mail ends up where the uw-imap readme says it will, in /var/mail/username. I log in successfully with both Mozilla and Pegasus Mail and the messages have blank headers except for the date, which is 12/31/1969. When I open the messages, the headers look okay, the date and time are accurate, but moving the message to another folder corrupts the message and, it appears, only moves the body over, not the headers. I'm no expert, but to me this looks as fetchmail and the other (cygwin!) tools uses LF line endings (Unix style). While Mozilla and Pegasus Mail (Windows executables most likely) expect it to be the Windows way (i.e. CR+LF). IMO a Mozilla and/or Pegasus Mail bug. Not definately true. The Mail/MIME encoding RFC's state that all headers should be CRLF no matter what the platform. It is open to dispute quite whether the RFC's should still apply once the mail has finished it's journey through the ether and has been collected from the email server by such a tool as fetchmail. It seems likely that Mozilla is expecting the CRLF's to remain on the headers which would probably be true if Mozilla was collecting the emails itself. It is my guess that fetchmail is stripping the CR's from the message as it retrieves it which could well be the accepted practice for UNIX boxes, but under a cygwin text mount it should perhaps behave differently. This is purely speculation I've not looked at the sources, but it is certainly true that the pop3 server should be presenting the headers with CRLF line endings to the client. HTH, Steve Fairbairn. -- -- LONG DISCLAIMER -- *** This email has originated from Perwill plc (Registration No. 1906964) Office registered at: 13A Market Square, Alton, Hampshire, GU34 1UR, UK Tel: +44 (0)1420 545000 Fax: +44 (0)1420 545001 www.perwill.com *** Privileged, confidential and/or copyright information may be contained in this email, and is only for the use of the intended addressee. To copy, forward, disclose or otherwise use it in any way if you are not the intended recipient or responsible for delivering to him/her is prohibited. If you receive this email by mistake, please advise the sender immediately, by using the reply facility in your email software. We may monitor the content of emails sent and received via our network for the purposes of ensuring compliance with policies and procedures. This message is subject to and does not create or vary any contractual relationships between Perwill plc and the recipient. *** Any opinions expressed in the email are those of the sender and not necessarily of Perwill plc. *** This email has been scanned for known viruses using McAfee WebShield 4.5 MR1a *** -- 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/
