Re: Multimedia keys interpreted as alphabetic keys
On 14/04/2010 15:37, Chris Bird wrote:
I wonder if anyone could shed some light on this minor problem I have:
I'm using Cygwin/X as my X client for several linux boxes at work.
Everything's working fine (after some tweaking), but there is one small
thing I find very annoying:
The multimedia keys on my keyboard generate keypresses as though they
were normal keys. Specifically, Play/Pause generates a 'g', volume up
generates 'c', volume down generates 'b', and mute generates 'd'. The
keyboard is a logitech Internet 350. As I listen to music all day to
keep my sanity, and occasionally need to alter the volume or stop or
mute the track to join in a conversation in the office (or stop driving
other people mad with my relentless beats!), I often find these
keypresses being entered into whatever program I'm working on. Not ideal.
I used xev to find out what keycodes were being generated by these keys,
and by the keys the output would suggest is being pressed, ie. g, c, b,
and d, expecting to find different keycodes perhaps being mapped to the
same keysyms, so I could alter .Xmodmap to assign them to XF86Play, etc.
But I found that they were exactly the same, as if Cygwin was
interpreting them as the same keys at some very basic level, whereas
Windows can tell the difference between pressing g (outputs a 'g') and
pressing Play/Pause (it plays or pauses the current track in my media
player). Here is the output from xev when pressing Play/Pause, then g:
[snip]
They both output 'keycode 42 (keysym 0x67, g)'. I tried various settings
in the arguments to XWin, to try to enable multimedia keys, and set the
model of my keyboard and layout correctly:
%RUN% XWin -multiwindow -clipboard -silent-dup-error -xkbmodel
logitech_base
REM -xkbvariant extd
REM -xkbmodel logii350
REM -xkblayout gb
but none of these made any positive difference, in fact one of them (I
can't remember which), actually disabled some other keys (e.g. the arrow
keys).
This behaviour doesn't happen in non-X Cygwin. I tried using Xming get
some ideas as suggested in this similar-looking post
http://sourceware.org/ml/cygwin-xfree/2004-12/msg00249.html
('Xming.exe :1 -multiwindow -logverbose 10'), and got no useful messages
about the keyboard.
Has anyone got a clue what's going on? Hope someone can help.
Thanks for the clear problem report.
After doing a bit of testing, the issue here seems to be that XWin doesn't
know to generate the correct X keycodes for these multimedia keys, so this is
a bug in XWin, not a problem with the XKB layout selected.
I've added some some entries to the table which translates Windows Virtual
Keycodes into X keycodes which hopefully fixes this.
Perhaps you could try out the test build I have uploaded at [1] and see if
that resolves this problem for you?
I've also turned on a bit more debug, so even if this doesn't work correctly,
-logverbose should indicate a bit more clearly what's going on...
[1] ftp://cygwin.com/pub/cygwinx/XWin.20100418-git-9594fb420a063608.exe.bz2
--
Jon TURNEY
Volunteer Cygwin/X X Server maintainer
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://x.cygwin.com/docs/
FAQ: http://x.cygwin.com/docs/faq/
[PATCH 1/3] Cygwin/X: Generate X keycodes for multimedia keys
Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
hw/xwin/winkeybd.h| 14 +++---
hw/xwin/winkeynames.h | 13 +
2 files changed, 16 insertions(+), 11 deletions(-)
diff --git a/hw/xwin/winkeybd.h b/hw/xwin/winkeybd.h
index 5b2a589..662392b 100644
--- a/hw/xwin/winkeybd.h
+++ b/hw/xwin/winkeybd.h
@@ -216,13 +216,13 @@ g_iKeyMap [] = {
/* 170 */0, 0, 0,
/* 171 */0, 0, 0,
/* 172 */0, 0, 0,
- /* 173 */0, 0, 0,
- /* 174 */0, 0, 0,
- /* 175 */0, 0, 0,
- /* 176 */0, 0, 0,
- /* 177 */0, 0, 0,
- /* 178 */0, 0, 0,
- /* 179 */0, 0, 0,
+ /* 173 */VK_VOLUME_MUTE, 0, KEY_Mute,
+ /* 174 */VK_VOLUME_DOWN, 0, KEY_AudioLower,
+ /* 175 */VK_VOLUME_UP, 0, KEY_AudioRaise,
+ /* 176 */VK_MEDIA_NEXT_TRACK,0, KEY_NEXTSONG,
+ /* 177 */VK_MEDIA_PREV_TRACK,0, KEY_PREVIOUSSONG,
+ /* 178 */VK_MEDIA_STOP, 0, KEY_STOPCD,
+ /* 179 */VK_MEDIA_PLAY_PAUSE,0, KEY_PLAYPAUSE,
/* 180 */0, 0, 0,
/* 181 */0, 0, 0,
/* 182 */0, 0, 0,
diff --git a/hw/xwin/winkeynames.h b/hw/xwin/winkeynames.h
index 7c16337..5a3129f 100644
--- a/hw/xwin/winkeynames.h
+++ b/hw/xwin/winkeynames.h
@@ -23,10 +23,6 @@
*
*/
-#define XK_TECHNICAL
-#defineXK_KATAKANA
-#include X11/keysym.h
-
#define GLYPHS_PER_KEY 4
#define NUM_KEYCODES 248
#define MIN_KEYCODE 8
@@ -195,6 +191,15 @@
#define KEY_HKTG /* Hirugana/Katakana tog 0xc8 */ 200
#define KEY_BSlash2 /* \ _ 0xcb */ 203
+#define KEY_Mute /* Audio Mute */ 152
+#define KEY_AudioLower /* Audio Lower */ 168
+#define KEY_AudioRaise /* Audio Raise */ 166
+
+#define KEY_NEXTSONG /* Media next */ 145
+#define KEY_PLAYPAUSE/* Media play/pause toggle */ 154
+#define KEY_PREVIOUSSONG /* Media previous */ 136
+#define KEY_STOPCD /* Media stop */ 156
+
/* These are for notused and unknown entries in translation maps. */
#define KEY_NOTUSED 0
#define KEY_UNKNOWN255
--
1.7.0.4
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://x.cygwin.com/docs/
FAQ: http://x.cygwin.com/docs/faq/
[PATCH 2/3] Cygwin/X: Generate X keycodes for a few 'internet' keys
Add a few more special keys I have on my keyboard: mail, search, browser
These are normally trapped by Windows, so -keyhook needs to be taught
how to catch them as well
---
hw/xwin/winkeybd.h|6 +++---
hw/xwin/winkeyhook.c | 12 +++-
hw/xwin/winkeynames.h |4
3 files changed, 14 insertions(+), 8 deletions(-)
diff --git a/hw/xwin/winkeybd.h b/hw/xwin/winkeybd.h
index 662392b..2784673 100644
--- a/hw/xwin/winkeybd.h
+++ b/hw/xwin/winkeybd.h
@@ -213,9 +213,9 @@ g_iKeyMap [] = {
/* 167 */0, 0, 0,
/* 168 */0, 0, 0,
/* 169 */0, 0, 0,
- /* 170 */0, 0, 0,
+ /* 170 */VK_BROWSER_SEARCH,KEY_SEARCH, 0,
/* 171 */0, 0, 0,
- /* 172 */0, 0, 0,
+ /* 172 */VK_BROWSER_HOME,KEY_BROWSER,0,
/* 173 */VK_VOLUME_MUTE, 0, KEY_Mute,
/* 174 */VK_VOLUME_DOWN, 0, KEY_AudioLower,
/* 175 */VK_VOLUME_UP, 0, KEY_AudioRaise,
@@ -223,7 +223,7 @@ g_iKeyMap [] = {
/* 177 */VK_MEDIA_PREV_TRACK,0, KEY_PREVIOUSSONG,
/* 178 */VK_MEDIA_STOP, 0, KEY_STOPCD,
/* 179 */VK_MEDIA_PLAY_PAUSE,0, KEY_PLAYPAUSE,
- /* 180 */0, 0, 0,
+ /* 180 */VK_LAUNCH_MAIL, KEY_MAIL, 0,
/* 181 */0, 0, 0,
/* 182 */0, 0, 0,
/* 183 */0, 0, 0,
diff --git a/hw/xwin/winkeyhook.c b/hw/xwin/winkeyhook.c
index 2d6ed18..0c2e65b 100755
--- a/hw/xwin/winkeyhook.c
+++ b/hw/xwin/winkeyhook.c
@@ -101,11 +101,13 @@ winKeyboardMessageHookLL (int iCode, WPARAM wParam,
LPARAM lParam)
switch (wParam)
{
case WM_KEYDOWN: case WM_SYSKEYDOWN:
- case WM_KEYUP:case WM_SYSKEYUP:
- fPassKeystroke =
- (fPassAltTab
-(p-vkCode == VK_TAB) ((p-flags LLKHF_ALTDOWN) != 0))
- || (p-vkCode == VK_LWIN) || (p-vkCode == VK_RWIN)
+ case WM_KEYUP:case WM_SYSKEYUP:
+ fPassKeystroke =
+ (fPassAltTab
+ (p-vkCode == VK_TAB) ((p-flags LLKHF_ALTDOWN) != 0))
+|| (p-vkCode == VK_LWIN) || (p-vkCode == VK_RWIN)
+|| (p-vkCode == VK_BROWSER_SEARCH) || (p-vkCode ==
VK_BROWSER_HOME)
+|| (p-vkCode == VK_LAUNCH_MAIL)
;
break;
}
diff --git a/hw/xwin/winkeynames.h b/hw/xwin/winkeynames.h
index 5a3129f..2f45ce6 100644
--- a/hw/xwin/winkeynames.h
+++ b/hw/xwin/winkeynames.h
@@ -200,6 +200,10 @@
#define KEY_PREVIOUSSONG /* Media previous */ 136
#define KEY_STOPCD /* Media stop */ 156
+#define KEY_SEARCH /* Search */ 221
+#define KEY_MAIL /* Mail*/ 228
+#define KEY_BROWSER /* Browser */ 170
+
/* These are for notused and unknown entries in translation maps. */
#define KEY_NOTUSED 0
#define KEY_UNKNOWN255
--
1.7.0.4
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://x.cygwin.com/docs/
FAQ: http://x.cygwin.com/docs/faq/
[PATCH 3/3] Cygwin/X: Better keyboard debugging output
---
hw/xwin/winkeybd.c |8
hw/xwin/winkeyhook.c |4 +---
2 files changed, 5 insertions(+), 7 deletions(-)
diff --git a/hw/xwin/winkeybd.c b/hw/xwin/winkeybd.c
index ad9e66a..860e356 100644
--- a/hw/xwin/winkeybd.c
+++ b/hw/xwin/winkeybd.c
@@ -73,6 +73,8 @@ winTranslateKey (WPARAM wParam, LPARAM lParam, int
*piScanCode)
int iParam = HIWORD (lParam);
int iParamScanCode = LOBYTE (iParam);
+ winDebug(winTranslateKey: wParam %08x lParam %08x\n, wParam, lParam);
+
/* WM_ key messages faked by Vista speech recognition (WSR) don't have a
* scan code.
*
@@ -488,10 +490,8 @@ winSendKeyEvent (DWORD dwKey, Bool fDown)
for (i = 0; i nevents; i++)
mieqEnqueue(g_pwinKeyboard, events[i].event);
-#if CYGDEBUG
- ErrorF(winSendKeyEvent: dwKey: %d, fDown: %d, nEvents %d\n,
- dwKey, fDown, nevents);
-#endif
+ winDebug(winSendKeyEvent: dwKey: %d, fDown: %d, nEvents %d\n,
+ dwKey, fDown, nevents);
}
BOOL winCheckKeyPressed(WPARAM wParam, LPARAM lParam)
diff --git a/hw/xwin/winkeyhook.c b/hw/xwin/winkeyhook.c
index 0c2e65b..5314684 100755
--- a/hw/xwin/winkeyhook.c
+++ b/hw/xwin/winkeyhook.c
@@ -94,9 +94,7 @@ winKeyboardMessageHookLL (int iCode, WPARAM wParam, LPARAM
lParam)
/* Pass keystrokes on to our main message loop */
if (iCode == HC_ACTION)
{
-#if 0
- ErrorF (vkCode: %08x\tscanCode: %08x\n, p-vkCode, p-scanCode);
-#endif
+ winDebug(winKeyboardMessageHook: vkCode: %08x scanCode: %08x\n,
p-vkCode, p-scanCode);
switch (wParam)
{
--
1.7.0.4
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://x.cygwin.com/docs/
FAQ: http://x.cygwin.com/docs/faq/
Re: Keyboard and start options problems
On 07/04/2010 20:04, Denis Beauchemin wrote: My French-Canadian keyboard is not recognized: 2010-04-06 14:42:35 (--) winConfigKeyboard - Layout: 1009 (1009) 2010-04-06 14:42:35 (EE) Keyboardlayout Canadian French (1009) is unknown 2010-04-06 14:42:35 Rules = base Model = pc105 Layout = us Variant = Options = I was going to add this to the list of automatically recognized layouts, but there's something I don't quite understand here. Perhaps you can help clear up the confusion so I can do that? According to the list of locale IDs [2], 0x0c0c is Canadian French (for which we already automatically select the XKB layout ca variant fr) and 0x1009 is Canadian English. Assuming that is correct, at the very least, it looks like the middle line of the log output where we map the locale ID to keyboard layout name is wrong somehow. Is it the case that you have multiple keyboard layouts installed? Is it really the case that 0x1009 should be mapped to ca-fr? Is ca-en more appropriate, but you happen to want to use ca-fr? [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-submit-layout [2] http://msdn.microsoft.com/en-us/library/0h88fahh%28VS.85%29.aspx -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: bug report/suggested temp. patch: handling bursts of sent keys
On 08/04/2010 04:47, Mark Lillibridge wrote: Can you pull this patch and include it in the next X server release, please. Excellent. Thank you. Do we need to do anything special for this to get picked up for Xming as well? Assuming I don't receive too much hatemail about this breaking stuff after the next X server release, I'll push it upstream to X.Org. Colin Harrison, the Xming maintainer, periodically syncs from X.Org, but it's his decision what he includes. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: XWin does not delete the lock files
On 06/04/2010 22:16, Rodrigo Medina wrote: On 06/04/2010 10:15:44, Jon TURNEY wrote: I've uploaded a corrected build at [3], please try that instead. [3] ftp://cygwin.com/pub/cygwinx/XWin.20100406153607.exe.bz2 XWin.20100406153607 also stops with a Fatal Error and then segmentation fault See attachments [ 4157,578] (EE) XKB: Could not invoke xkbcomp [ 4157,578] (EE) XKB: Couldn't compile keymap [ 4157,578] XKB: Failed to compile keymap [ 4157,578] Keyboard initialization failed. This could be a missing or incorrect setup of xkeyboard-config. [ 4157,578] Fatal server error: [ 4157,578] Failed to activate core devices. Hmm... I'm still not doing something right. I've uploaded another build at [1], could you please give that a try? [1] ftp://cygwin.com/pub/cygwinx/XWin.20100413145147.exe.bz2 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: emacs-x11 takes 30-40 sec to open after upgrading to cygwin-1.7
On 13/04/2010 00:52, Dan Tsafrir wrote: On Tue, Apr 6, 2010 at 00:39, Jon TURNEY wrote: I've conducted a few repeated measurements and it looks as though setting LANG to be en_US somewhat reduces the start time of emacs-x11: instead of ~30 seconds with LANG=C.UTF-8, it take ~27 seconds LANG=en_US. While this is ~10% less, waiting 27 seconds for emacs to open still seems unreasonable. Any other ideas? Hmm You don't have any emacs fonts being set via ~/.Xdefaults or ~/.Xresources? Actually, I do. However, following the suggestion of Ken Brown http://www.mail-archive.com/cyg...@cygwin.com/msg107126.html I've invoked emacs with -Q (and also, just to make sure, removed my ~/.Xdefaults). It did not change anything: emacs still takes ~30 seconds to open. The mechanism that loads ~/.Xdefaults into the X server resource database is nicely obscure, so I'm not sure that -q actually avoids that. The definitive way to check would be to move ~/.Xdefaults aside, restart the X server then start emacs (I'm not sure if your clean install test would have done this or not) Another thing which might be worth trying is to see if the behaviour changes in a non-UTF-8 locale, e.g. export LANG=C and then run emacs. So far it seems that (1) emacs takes a long time to start up (2) xterm starts up quickly. Are there other X applications that you use and how do they behave? -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: XWin does not delete the lock files
On 06/04/2010 05:56, Rodrigo Medina wrote: On 05/04/2010 17:14:51, Jon TURNEY wrote: Perhaps you could try out the test build I have uploaded at [1] and see if that resolves this problem for you? [1] ftp://cygwin.com/pub/cygwinx/XWin.20100405215019.exe.bz2 XWin.20100405215019 crashes immediately. See XWin.0.log Thanks for testing this. I wouldn't describe this problem as a crash. Compare and contrast [1] and [2]. [ 8658,171] (EE) XKB: Couldn't open rules file /opt/wip/jhbuild/install/share/X11/xkb/rules/base [ 8658,171] XKB: Failed to compile keymap [ 8658,171] Keyboard initialization failed. This could be a missing or incorrect setup of xkeyboard-config. [ 8658,171] Fatal server error: [ 8658,171] Failed to activate core devices. Oops, this test build is looking in the wrong place for the keyboard configuration data. I've uploaded a corrected build at [3], please try that instead. [1] http://en.wikipedia.org/wiki/Abort_%28computing%29 [2] http://en.wikipedia.org/wiki/Crash_%28computing%29 [3] ftp://cygwin.com/pub/cygwinx/XWin.20100406153607.exe.bz2 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: XWin does not delete the lock files
On 30/03/2010 22:02, Rodrigo Medina wrote: I have found that XWin -multiwindow -clipboard dos not remove the lock files when it is stopped with the 'Exit' button of the XWin menu [...] Thanks for the reproduction steps here, although I wasn't able to reproduce the problem without also connecting an xterm before shutting down the X server. At least part of the problem seems to be awesomely bad multi-threaded code: It looks like the code thinks XSetIOErrorHandler() sets a per-thread handler, rather than the global handler it actually sets, so each client thread sets it to a handler which longjmp()s somewhere, and we end up getting the last one which happened to get installed if the handler ever actually gets called, with hilarious consequences... (longjmp()ing to a setjmp() buffer allocated in another thread being undefined, and even if it happens to work, it isn't going to do the right thing) This also seems to be the cause of the segfault on shutdown you have reported in another thread. I've had a go at working around these issues a bit, although the only way of handling the multithreading correctly looks to be to re-write the clipboard and multiwindow WM to use xcb. Perhaps you could try out the test build I have uploaded at [1] and see if that resolves this problem for you? [1] ftp://cygwin.com/pub/cygwinx/XWin.20100405215019.exe.bz2 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Conflict between cliupboard and xmodmap
On 31/03/2010 15:52, Rodrigo Medina wrote: Apparently here is a conflict between clipboard and xmodmap: [...] winClipboardFlushXEvents - unexpected event type 34 This is a not a conflict or an error. This is merely a warning that the clipboard thread wasn't written to expect the MappingNotify events, which xmodmap causes the server to send to all clients. There is no mechanism to express disinterest in that event, so I suppose we should add something to ignore those events, to prevent alarm and confusion... -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: problem about keyboard
On 02/04/2010 01:40, kyle @ gmail wrote: I downloaded and installed cygwin on this Thuesday, but I found that I can not type anything in X-server console. I read the QA, and it said this problem is caused by old version of startxwin.bat But since this is my first time to install cygwin, there is no startxwin.bat in my cygwin directory. I've also checked if there is anything setting XKEYSYMDB, but it replied nothing. Could you give me any suggest to fix that problem? (My OS is windows vista) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html ^^^ You might get some more useful replies if you attach your cygcheck output and XWin.0.log as suggested on that page. Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: USER/GDI Objects leak with (XWin.exe) Cygwin/X X Server Verion 1.7.6, Build Date 2010-03-18
On 29/03/2010 18:20, Amal Khailtash wrote: I do have native TCL installed as well, but I am not really running anything locally. The only client I run is a Linux Konsole window launched by an SSH command. Once I have a remote terminal, I launch my simulation tool (ModelSim). I watch the resources using windows Task Manager. Do you think that ModelSim or Konsole are eating up resources? Although, even when I quit those programs, until I close XWin.exe, the resources are not freed! Well, how about approaching the problem methodically? If you boot up and just start the X server, do you see these leaks? Does starting you X client (Konsole via ssh) cause these leaks? Does starting your TCL application cause these leaks? - Original Message Sent: Mon, March 29, 2010 12:55:18 PM Subject: Re: USER/GDI Objects leak with (XWin.exe) Cygwin/X X Server Verion 1.7.6, Build Date 2010-03-18 On 22/03/2010 18:38, Amal Khailtash wrote: My Windows XP USER and GDI object counts, for process XWin.exe, keep going up even if I leave my PC running and doing nothing. The increase eventually makes my windows title-bars and and menu bars disappear and I need to reboot to be able to work again. How are you measuring these object counts? It's unclear from what you write if you have any X clients connected to your X server while the leak is occurring. It's interesting that, from your cygcheck output, you seem to have a native TCL installed, as that has been mentioned previously associated with some source of resource leak problem (see [1]), although I've never been able to reproduce it. [1] http://sourceware.org/ml/cygwin-xfree/2009-02/msg00184.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
[PATCH] Store pthread_ids and chain IOError handlers to avoid longjmp across threads
Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
hw/xwin/winclipboardthread.c | 24 +-
hw/xwin/winmultiwindowwm.c | 54 --
2 files changed, 53 insertions(+), 25 deletions(-)
diff --git a/hw/xwin/winclipboardthread.c b/hw/xwin/winclipboardthread.c
index f61037c..be8126f 100644
--- a/hw/xwin/winclipboardthread.c
+++ b/hw/xwin/winclipboardthread.c
@@ -58,6 +58,9 @@ extern Window g_iClipboardWindow;
*/
static jmp_buf g_jmpEntry;
+static XIOErrorHandler g_winClipboardOldIOErrorHandler;
+static pthread_t g_winClipboardProcThread;
+
Bool g_fUnicodeSupport = FALSE;
Bool g_fUseUnicode = FALSE;
@@ -122,6 +125,11 @@ winClipboardProc (void *pvNotUsed)
ErrorF (winClipboardProc - Warning: Locale not supported by X.\n);
}
+ /* Set error handler */
+ XSetErrorHandler (winClipboardErrorHandler);
+ g_winClipboardProcThread = pthread_self();
+ g_winClipboardOldIOErrorHandler = XSetIOErrorHandler
(winClipboardIOErrorHandler);
+
/* Set jump point for Error exits */
iReturn = setjmp (g_jmpEntry);
@@ -144,10 +152,6 @@ winClipboardProc (void *pvNotUsed)
/* Use our generated cookie for authentication */
winSetAuthorization();
- /* Set error handler */
- XSetErrorHandler (winClipboardErrorHandler);
- XSetIOErrorHandler (winClipboardIOErrorHandler);
-
/* Initialize retry count */
iRetries = 0;
@@ -452,8 +456,14 @@ winClipboardIOErrorHandler (Display *pDisplay)
{
ErrorF (winClipboardIOErrorHandler!\n\n);
- /* Restart at the main entry point */
- longjmp (g_jmpEntry, WIN_JMP_ERROR_IO);
-
+ if (pthread_equal(pthread_self(),g_winClipboardProcThread))
+{
+ /* Restart at the main entry point */
+ longjmp (g_jmpEntry, WIN_JMP_ERROR_IO);
+}
+
+ if (g_winClipboardOldIOErrorHandler)
+g_winClipboardOldIOErrorHandler(pDisplay);
+
return 0;
}
diff --git a/hw/xwin/winmultiwindowwm.c b/hw/xwin/winmultiwindowwm.c
index 43c047c..3f231bc 100644
--- a/hw/xwin/winmultiwindowwm.c
+++ b/hw/xwin/winmultiwindowwm.c
@@ -203,7 +203,11 @@ winUpdateWindowPosition (HWND hWnd, Bool reshape, HWND
*zstyle);
*/
static jmp_buf g_jmpWMEntry;
+static XIOErrorHandler g_winMultiWindowWMOldIOErrorHandler;
+static pthread_t g_winMultiWindowWMThread;
static jmp_buf g_jmpXMsgProcEntry;
+static XIOErrorHandler g_winMultiWindowXMsgProcOldIOErrorHandler;
+static pthread_t g_winMultiWindowXMsgProcThread;
static Boolg_shutdown = FALSE;
static BoolredirectError = FALSE;
static Boolg_fAnotherWMRunning = FALSE;
@@ -900,9 +904,14 @@ winMultiWindowXMsgProc (void *pArg)
ErrorF (winMultiWindowXMsgProc - pthread_mutex_unlock () returned.\n);
+ /* Install our error handler */
+ XSetErrorHandler (winMultiWindowXMsgProcErrorHandler);
+ g_winMultiWindowXMsgProcThread = pthread_self();
+ g_winMultiWindowXMsgProcOldIOErrorHandler = XSetIOErrorHandler
(winMultiWindowXMsgProcIOErrorHandler);
+
/* Set jump point for IO Error exits */
iReturn = setjmp (g_jmpXMsgProcEntry);
-
+
/* Check if we should continue operations */
if (iReturn != WIN_JMP_ERROR_IO
iReturn != WIN_JMP_OKAY)
@@ -918,10 +927,6 @@ winMultiWindowXMsgProc (void *pArg)
pthread_exit (NULL);
}
- /* Install our error handler */
- XSetErrorHandler (winMultiWindowXMsgProcErrorHandler);
- XSetIOErrorHandler (winMultiWindowXMsgProcIOErrorHandler);
-
/* Setup the display connection string x */
winGetDisplayName(pszDisplay, (int)pProcArg-dwScreen);
@@ -1282,9 +1287,14 @@ winInitMultiWindowWM (WMInfoPtr pWMInfo, WMProcArgPtr
pProcArg)
ErrorF (winInitMultiWindowWM - pthread_mutex_unlock () returned.\n);
+ /* Install our error handler */
+ XSetErrorHandler (winMultiWindowWMErrorHandler);
+ g_winMultiWindowWMThread = pthread_self();
+ g_winMultiWindowWMOldIOErrorHandler = XSetIOErrorHandler
(winMultiWindowWMIOErrorHandler);
+
/* Set jump point for IO Error exits */
iReturn = setjmp (g_jmpWMEntry);
-
+
/* Check if we should continue operations */
if (iReturn != WIN_JMP_ERROR_IO
iReturn != WIN_JMP_OKAY)
@@ -1300,10 +1310,6 @@ winInitMultiWindowWM (WMInfoPtr pWMInfo, WMProcArgPtr
pProcArg)
pthread_exit (NULL);
}
- /* Install our error handler */
- XSetErrorHandler (winMultiWindowWMErrorHandler);
- XSetIOErrorHandler (winMultiWindowWMIOErrorHandler);
-
/* Setup the display connection string x */
winGetDisplayName(pszDisplay, (int)pProcArg-dwScreen);
@@ -1426,12 +1432,18 @@ winMultiWindowWMIOErrorHandler (Display *pDisplay)
{
ErrorF (winMultiWindowWMIOErrorHandler!\n\n);
- if (g_shutdown)
-pthread_exit(NULL);
+ if (pthread_equal(pthread_self(),g_winMultiWindowWMThread))
+{
+ if (g_shutdown)
+pthread_exit(NULL);
+
+ /* Restart at the main entry point
Re: emacs-x11 takes 30-40 sec to open after upgrading to cygwin-1.7
On 29/03/2010 23:41, Dan Tsafrir wrote: On Mon, Mar 29, 2010 at 19:14, Jon TURNEY wrote: On 29/03/2010 11:33, Dan Tsafrir wrote: After I've upgraded to cygwin-1.7 my emacs takes 30-40 seconds to open. Sorry, I hadn't noticed what you'd written about xterm starting up normally, I guess that tends to suggest this is an emacs-specific problem in some way rather than the general X fontset problem I assumed it was... I've conducted a few repeated measurements and it looks as though setting LANG to be en_US somewhat reduces the start time of emacs-x11: instead of ~30 seconds with LANG=C.UTF-8, it take ~27 seconds LANG=en_US. While this is ~10% less, waiting 27 seconds for emacs to open still seems unreasonable. Any other ideas? Hmm You don't have any emacs fonts being set via ~/.Xdefaults or ~/.Xresources? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: emacs-x11 takes 30-40 sec to open after upgrading to cygwin-1.7
On 29/03/2010 11:33, Dan Tsafrir wrote: After I've upgraded to cygwin-1.7 my emacs takes 30-40 seconds to open. Following http://x.cygwin.com/docs/faq/cygwin-x-faq.html#poor-performance, disabling my antivirus has no affect on emacs's opening time. As far as I can tell other X programs behave similarly to the way they did before the upgrade. Any help would be appreciated. The output of 'cygcheck -s -v -r' is attached. LANG = 'C.UTF-8' You are probably being bitten by [1] As far as I can tell, this is a general problem with X, but cygwin unfortunately encounters it in a default install because (a) the default locale is a UTF-8 locale, and (b) we don't install the CJK fonts by default. Workarounds you might try are (a) install the font packages font-isas-misc, font-jis-misc and font-daewoo-misc (and restart your server), or (b) set your locale to a non-UTF-8 one. We need to do something about this, but as the discussion in the bug should make clear, we're not sure what :-) [1] http://sourceware.org/bugzilla/show_bug.cgi?id=10948 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Xwin crashes in multiwindow
On 29/03/2010 06:46, Rodrigo Medina wrote: I have found that with some programs XWin crashes giving segmentation Fault. The problem arises when the client program is stopped using the X button of the Windows window. The cygcheck output and the XWin log file are attached. 2010-03-29 00:06:53 winInitMultiWindowWM - Warning: Locale not supported by X. I'm kind of puzzled to see this, and I'm wondering what locale you have set, but your cygcheck output seems to have been generated using cygcheck -s rather than cygcheck -svr, as recommended by the problem reporting page linked at the bottom of this email, so it doesn't record your environment. In case you need it, I can also attach the source of of some of the programs that fail. Unfortunately the problem does not always appear, particulary for very simple programs it does not arise. What would be most useful would be using gdb to get a backtrace for the X server when it crashes. Apart from the XWin.0.log the following messages appear in the terminal window used for starting the server Segmentation fault at address 0x1c Fatal server error: Caught signal 11 (Segmentation fault). Server aborting $ [mi] EQ overflowing. The server is probably stuck in an infinite loop. The message says that the server is aborting, but actually it does not disappear. It cannot be stopped with the exit button in the servers menu. After trying that the following message appears $ winDeinitMultiWindowWM - Noting shutdown in progress [mi] EQ overflowing. The server is probably stuck in an infinite loop. [mi] EQ overflowing. The server is probably stuck in an infinite loop. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Problems getting started in Cgywin/X
On 28/03/2010 20:40, David Sicks wrote: I want to run DDLab and I have Windows XP. DDLab can be run using Cygwin/X among other things, and I decided to use it. Coming out of the Cygwin/X setup, I got the following error messages: warning: locale is not supported by c library, locale unchanged Also, I tried s.t.a.r.t.x.d.m.c.p..b.a.t. in bash 3.2 and got the following fatal error message: /usc/bin/xwin -query 10.0.0.1 -nodirection -lesspointer I've not used Linux before. When I googled the locale is not supported by c library message, I found a bewildering array of prior similar problems, numerous futile requests for help, and (I think) some fixes that may have worked that I didn't understand. Can you tell me how to get on track? Chapter 4 of the Cygwin/X user's guide [1] should tell you how to do whatever it is you want to do. If you still have problems, please read [2] [1] http://x.cygwin.com/docs/ug/using.html [2] http://cygwin.com/problems.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: 1.7.2: ssh -X unresponsive until xauth PPID killed
On 27/03/2010 21:53, Mark McConnell wrote: (Please forgive the repost - this version has the attachments mentioned) After installing Cygwin 1.7.2 for the first time on Windows Server 2008 SP2, I discovered that I could not connect to a remote host using Cygwin's ssh, if X is forwarded. ssh -X host or ssh -Y host : program becomes unresponsive. I've attached the output of cygcheck -s -v -r to this message, and /var/log/XWin.0.log. I should mention as an aside that there are other clues that my Cygwin installation is flawed (a local gvim gui will not display, man dumps core). Given that, I'm thinking that the problem you describe with the xauth subprocess is probably symptomatic of some deeper problem with you installation or environment. You should probably ask for help on the main cygwin list. Below, I describe two experiments that indicate that my X forwarding problem is somehow related to xauth* and its PPID: ### 'ssh -X host gui' displays if the PPID of xauth* is killed ### First, I open a new mintty window within the Cygwin environment and issue the command to start gvim on the remote server, using the forwarded X display. No gui is displayed. The cursor returns below the prompt. ^d or ^c do not terminate the process ( however, I can suspend it from that window by typing ^z, and I can return it to the foreground with fgjob ). -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: USER/GDI Objects leak with (XWin.exe) Cygwin/X X Server Verion 1.7.6, Build Date 2010-03-18
On 22/03/2010 18:38, Amal Khailtash wrote: My Windows XP USER and GDI object counts, for process XWin.exe, keep going up even if I leave my PC running and doing nothing. The increase eventually makes my windows title-bars and and menu bars disappear and I need to reboot to be able to work again. How are you measuring these object counts? It's unclear from what you write if you have any X clients connected to your X server while the leak is occurring. It's interesting that, from your cygcheck output, you seem to have a native TCL installed, as that has been mentioned previously associated with some source of resource leak problem (see [1]), although I've never been able to reproduce it. [1] http://sourceware.org/ml/cygwin-xfree/2009-02/msg00184.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: bug report/suggested temp. patch: handling bursts of sent keys
On 20/03/2010 16:32, Mark Lillibridge wrote: On February 23, 2010, Jon Turney wrote: On 13/02/2010 20:24, Mark Lillibridge wrote: Jon wrote: Thanks for the patch. Have you actually tested that this resolves your problem? Yes. Of course, really, really large bursts will still fail, but they should be very rare. Perhaps you might try the attached patch, instead, and see if that helps. Yes, your (previously) attached patch seems to solve the problem without wasting memory or only working for a finite number of keys. Nice job. What's the next step towards getting this patch added to the source/sent upstream/to Xming? Yaakov, Can you pull this patch and include it in the next X server release, please. Sadly this isn't quite a complete solution: there's still the problem in multiwindow mode that Windows uses a modal loop when resizing/moving a window, so the X server code gets no chance to do anything then. You can easily demonstrate this by resizing the frame of an X window, moving the mouse continuously for a number of seconds, and then notice that X window doesn't react to the queued size changes until the mouse button is released. The following changes since commit 579715f830fbbca9e1ecb17dc18176132f5969e7: Rami Ylimaki (1): os: Prevent backtrace from being stopped in noreturn functions. are available in the git repository at: git://anongit.freedesktop.org/~jturney/xserver for-yaakov Jon TURNEY (1): Cygwin/X: Process one Windows message per wakeup, rather than all of them. hw/xwin/winblock.c | 23 --- hw/xwin/winwakeup.c |4 ++-- 2 files changed, 6 insertions(+), 21 deletions(-) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: problems starting xterm
On 19/03/2010 09:54, Markus Hoenicka wrote: I've set up Cygwin 1.7.1 lately, along with X and some applications. The X server works ok and claims to be Version 1.7.5 Build Date: 2010-02-05. I can run Emacs as an X application without hassles. However, when I start the X server, or when I try to run xterm from the X server systray icon, xterm apparently fails to find a shell to run, so it terminates (no pun intended) after a couple of seconds: xterm: Could not exec 4.90.2.20040617: No such file or directory Google suggests '4.90.2.20040617' is a novell netware client version. So, erm... yeah. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: xterm no longer accepts keyboard input
On 13/03/2010 21:47, Bob Kline wrote: On 3/13/2010 1:05 PM, Bob Kline wrote: I have used X under Cygwin successfully for years. Yesterday I did an upgrade of my Cygwin installation and X stopped working. After struggling to get things working again, I decided to just install Cygwin fresh and start over. Now I can get the X server started (following the instructions in the Cygwin/X User's Guide, running /usr/bin/startxwin from a Cygwin shell), but xterm windows that I open do not respond to keyboard input. I looked at the FAQ and found the entry (3.2 I upgraded my X server and now I can't type anything into any X application) which warns against using any older scripts to launch X, but as noted above, I'm invoking the startxwin binary directly. Any thoughts as to what could be going wrong, given that this is an installation of Cygwin and X from scratch? looking at your cygcheck.out, you have the following variables in your environment XKEYSYMDB = '/usr/X11R6/lib/X11/XKeysymDB' XAPPLRESDIR = '/usr/X11R6/lib/X11/app-defaults' XCMSDB = '/usr/X11R6/lib/X11/Xcms.txt' XNLSPATH = '/usr/X11R6/lib/X11/locale' These have probably been set by Microsoft Services for Unix, and as the FAQ tries to make clear, at least XKEYSYMDB must be unset for keyboard input to XWin to work correctly (FAQ 3.2 refers to FAQ 8.8) We used to have something in startxwin.bat to unset this, so I guess it is a regression that we don't do that anymore. Following up on my earlier post, I see that although I am able to start the X server if I invoke /usr/bin/startxwin directly from a Cygwin shell, that's not *exactly* what the FAQ was suggesting. I'm not sure if the difficulties starting a server mean you have some other problem as well. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: xterm no longer accepts keyboard input
On 15/03/2010 05:42, David Barr wrote: On Sun, Mar 14, 2010 at 7:17 AM, Jeff Spirkospi...@gmail.com wrote: The icon runs from the Windows environment, so the run.exe and bash.exe wrapper programs are needed to set up the proper cygwin environment. Once you're in a cygwin shell, you already have the cygwin environment set up. That's why you don't need to type them in from the command line. So, they really should be equivalent. I've probably missed a prior discussion, but if startxwin.exe doesn't set up the Cygwin environment, then what good is it? I thought that this was about the only advantage that startxwin.bat had over startx. No. startx and startxwin are different tools to solve different problems. I keep adding more text to [1] to clarify this, but this doesn't appear to help, perhaps because nobody actually reads it... [1] http://x.cygwin.com/docs/ug/using.html#using-starting Considering that startx also sets up the .Xauthority file and startxwin.exe doesn't, I see absolutely no reason to use startxwin.exe instead of startx. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Incorrect Cygwin/X behavior leads to .X0-lock related fatal error
On 12/03/2010 02:31, Ken Klein wrote: Here is tool behavior that seems like a bug in 1.7.1 related to the Cygwin/X FAQ item 3.4 (A2)called Fatal server error: Can't read lock file /tmp/.X0-lock. I am posting this, because the faq item states that the buggy behavior happens For reasons which are not well understood at present. This post presents what I consider to be incorrect behavior that leads to this error. If I log on as a user without admin rights (on Windows XP), and I start Cygwin/X using the XWin Server program shortcut, the referenced lock file gets created and gives me read/write access. The Everyone user just has read access. When I then close my XWin session, the lock file remains. I log out of Windows, then someone else logs on (also without admin rights), and they get the fatal error as described in the FAQ item, since they do not have write access to the lock file. If, however, I had originally started Cygwin/X by opening a Cygwin window then entering startx, when I eventually end my Cygwin session, the lock file gets deleted. So in this scenario, the next person to log on will not get the fatal error, since Cygwin didn't leave the lock file behind. My opinion is that the startx scenario behaves correctly, and the XWin Server shortcut scenario presents a bug, namely, it leaves behind the lock file after the session is ended. Thanks for the detailed bug report. In fact that FAQ is slightly out of date and we do now understand that the problem is shown when running XWin as both Administrator and non-Adminstrator users as you describe. I'm not able to reproduce the difference in behaviour you describe (of leaving the log file behind) between running startx (which then starts XWin) and directly starting XWin when running as a non-administrator user. Can you be specific about how you are exiting XWin which causes it to leave the lock file behind? This shouldn't happen, and is the cause of the problem. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris
On 04/03/2010 01:39, Mark Veneziano wrote: Confirmed, I see the same behavior on FC12 with xorg-x11-server-Xorg-1.7.5-1.fc12.x86_64. In XDMCP mode, clicking a CDE panel up-triangle for the second time in a session causes a mouse/keyboard freeze. Thanks for the confirmation. I can only suggest you monitor the upstream bug [1], when a fix is available we will integrate it. [1] http://bugs.freedesktop.org/show_bug.cgi?id=25400 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: cygwin\var\log\XWin.0.log
On 05/03/2010 03:34, Mayank Jain Nawal wrote: 2010-03-05 08:55:29 (EE) XKB: Couldn't open rules file /usr/share/X11/xkb/rules/base 2010-03-05 08:55:29 XKB: Failed to compile keymap 2010-03-05 08:55:29 Keyboard initialization failed. This could be a missing or incorrect setup of xkeyboard-config. 2010-03-05 08:55:29 Fatal server error: 2010-03-05 08:55:29 Failed to activate core devices. It sometimes helps if you read the log. Note the first error. The file /usr/share/X11/xkb/rules/base should exist, installed by the xkeyboard-config package. Try 'ls -al /usr/share/X11/xkb/rules/base' and 'cygcheck -c xkeyboard-config' Also, please attach you cygcheck.out as requested by Problem reports: http://cygwin.com/problems.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Fatal error
On 08/03/2010 18:31, evgar...@ull.es wrote: my name is Eva GarcĂa Llorente, I have installed a new version of cygwin (that one which is necessary for installing the latest Singular version (3-1-1)). It is supposed to be installed,but when I try to execute Singular (Emacs and X-server) a fatal error occurs, and in /var/log/Xwinlog is reported the following error: Fatal server error: Can't read lock file /tmp/.X0-lock I have tried these possible solutions (FAQ): 1. startxwin -- -nolock 2. rm -rf /tmp; mkdir /tmp ; chmod 1777 /tmp and they have not worked. Can you attach the log file from running the X server with the additional -nolock option? What about the facts which FAQ 3.4 mentions as relevant? Are you using a FAT filesystem? Does your account have local Administrator privileges? -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Problem running Xwin.exe in Windows 7
On 02/03/2010 05:19, Babak Fallah wrote: A fatal error has happened and Cygwin/X will now exit. Please open /var/log/XWin.0.log for more information. Vendor: The Cygwin/X project. Release 1.7.5.0 (10705000) contact: cygwin-xfree@cygwin.com Build date: 2010-02-25 X-Win was started with the following command line: (null) This is a bit odd, as you seem to have managed to start XWin without getting argv[0] set. Whilst this is certainly possible, it seems to suggest you are either launching XWin in some non-standard manner, or perhaps have some kind of BLODA problem... Looking again, that extraneous hyphen probably means you haven't actually cut-and-pasted the XWin.0.log, so I guess all bets are off... -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: What is replacement for the -kb Xwin startup parameter
On 07/01/2010 05:13, Bennett, Tony wrote: I a using Xwin to connect to an AIX system using XDMCP. This was working fine on an old version of Xwin, As long as I used the -kb command line argument (there was an old post in the FAQ, suggesting that parameter As the workaround for AIX not handling keyboard extensions). I just downloaded 1.7.3.0 on another Win-XP system, and discovered That -kb is no longer supported. Without that argument, I am able To successfully connect using XDMCP, but after opening my 2nd Xterm, the window freezes up... neither mouse or keyboard Seems to function with it (which was the same symptom I had on The old version prior to adding the -kb startup parm. Thanks for pointing out the FAQ 5.1.5 is out of date, I'll fix that. The Xming troubleshooting page has some more up-to-date advice [1], but I can't personally check that. [1] http://www.straightrunning.com/XmingNotes/trouble.php#head-142 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris
On 03/03/2010 00:27, Mark Veneziano wrote: I started looking at the X.Org bugzilla when I first noticed the keyboard/mouse freeze, but then I realized I could not duplicate the problem using Xorg on Linux. The freeze only occurs with Cygwin/XWin. This has made me doubt that a fix could come from upstream, but perhaps that is not a good conclusion to draw. I don't think you were comparing like with the like. The dtterm problem you describe, I could reproduce with FC12 xserver 1.7.1-7, but not with 1.7.4-6 or 1.7.5-1. This is not surprising as the patch from fd.o bugzilla #25400 is present in FC12 1.7.4-6 and later. We added this patch in Cygwin XServer 1.7.5-1. I am still able to reproduce the CDE problem you describe with FC12 xserver 1.7.5-1 (as well as the cygwin Xserver 1.7.5.-1), perhaps you could confirm that observation? - switch to a spare VT (e.g. ctrl-alt-f2) - login - run X -query solaris on a spare display number (e.g. X :1 -query solaris) - login, start CDE, use panel to launch something as before... -- From: Colin Harrison Sent: Tuesday, March 02, 2010 12:32 PM To: cygwin-xfree@cygwin.com Subject: Re: Re: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris Hi, Is the patch in X.Org Bugzilla 25400 superseded by this...? http://lists.x.org/archives/xorg-devel/2010-February/005845.html Peter and Keith have changed their minds..maybe of some help (or a red herring!) Thanks for pointing that out. I had noticed that patch, despite it not being in bugzilla, and that was included in my 20100302123309 build. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris
On 01/03/2010 16:26, Bennett, Tony wrote: Parenthetically, I am currently using XMING due to this problem, but would like to go back to using cygwin-X. Which version of Xming are you using? If any X11-developer would like to give me a debug version of xorg-server to try capture the location of the problem, I would be assist. The upstream patch for this issue has been updated, so I've built a new XWin with that incorporated [1]. If you would like to try that out, that would be helpful. [1] ftp://cygwin.com/pub/cygwinx/XWin.20100302123309.exe.bz2 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: fvwm2 and startxwin
On 01/03/2010 22:16, Stephen More wrote: Let me ask this question to clarify things: multiwindow mode (startxwin) -- Is this the same thing as --rootless http://x.cygwin.com/docs/ug/configure.html#configure-cygwin-x-windowing-modes I read the thread, and there is alot about changing .xinitrc to add a window manager. But I did not read how to add a window manager to .startxwinrc or a parameter. I need to run fvwm rootless like: http://x.cygwin.com/screenshots/cygx-rootless-fvwm2-20031224-0010.png Start the x server using 'startx -- -rootless' and start fvwm2 from your .xinitrc file (typically by making 'exec fvwm2' the last line. NB this does not end with an '') If you aren't interested in starting other programs from your .xinintrc, you can just do 'startx fvwm2 -- -rootless' for the all in one solution :-) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: fvwm2 and startxwin
On 02/03/2010 01:22, Jon TURNEY wrote: Start the x server using 'startx -- -rootless' and start fvwm2 from your .xinitrc file (typically by making 'exec fvwm2' the last line. NB this does not end with an '') If you aren't interested in starting other programs from your .xinintrc, you can just do 'startx fvwm2 -- -rootless' for the all in one solution :-) Oops. I mean 'startx /usr/bin/fvwm2 -- -rootless' (as man xinit says, the client supplied to startx or xinit must begin with / or ., otherwise it's treated as client arguments to the default xterm) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: [ANNOUNCEMENT] [1.7] Updated: xorg-server 1.7.5-1, 1.7.99.901-1 (TEST)
On 25/02/2010 08:21, Yaakov (Cygwin/X) wrote: The following package has been updated for Cygwin 1.7: * xorg-server-1.7.5-1 The following patches has been added in the stable release: I believe this also contains a patch for the issue where input would freeze in running a motif application in an XDMCP session (due to a passive grab triggering, but the event the application was waiting for to release it not being sent) http://sourceware.org/bugzilla/show_bug.cgi?id=11301 https://bugs.freedesktop.org/show_bug.cgi?id=25400 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris
On 20/02/2010 04:02, Mark Veneziano wrote: Thanks very much for investigating this issue. Please let me know if I can help in any way. If you can confirm this is fixed with xorg-server 1.7.5-1, that would be helpful :-) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: autogen.sh error
On 24/02/2010 20:01, Yaakov (Cygwin/X) wrote: On 2010-02-24 10:41, J. Offerman wrote: The rdynamic thing is harmless. The error was about undefined symbols. That's a different story entirely... This is what the build log says: -- Making all in hw Making all in xwin Making all in vfb LINK Xvfb.exe gcc: unrecognized option `-rdynamic' ../../mi/.libs/libmi.a(miexpose.o): In function `miPaintWindow': /usr/src/xorg-server-1.7.3-1/src/xorg-server-1.7.3/mi/miexpose.c:555: undefined reference to `_IsFramedWindow' /usr/src/xorg-server-1.7.3-1/src/xorg-server-1.7.3/mi/miexpose.c:556: undefined reference to `_RootlessStartDrawing' /usr/src/xorg-server-1.7.3-1/src/xorg-server-1.7.3/mi/miexpose.c:557: undefined reference to `_RootlessDamageRegion' /usr/src/xorg-server-1.7.3-1/src/xorg-server-1.7.3/mi/miexpose.c:562: undefined reference to `_RootlessSetPixmapOfAncestors' collect2: ld returned 1 exit status make[3]: *** [Xvfb.exe] Error 1 make[2]: *** [all] Error 2 make[1]: *** [all-recursive] Error 1 make: *** [all-recursive] Error 1 -- I can see those functions in miext/rootless so it looks like librootless is somehow not linked in? This is some upstream breakage, courtesy of XQuartz :-) Yes, known issue with 1.7+. Currently XWin (if -mwextwm support is enabled) needs to be built in a separate pass from the other DDXs, as my .cygport does. If you don't care about about -mwextwm, then uninstall the windowswmproto package and rebuild. Or if you're not interested in building xvfb and xnest, you can ./configure with --disable-xnest --disable-xvfb. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Vista, new install, x server does not start, startxwin.bat missing
On 19/02/2010 15:40, Dr. M. C. Nelson wrote: I just did a fresh install on Vista, and confirmed the following 1) From an administrator account, it works out-of-the-box. Beautiful. 2) From a regular user account, clicking the x-server or shell icons, produces a UAC prompt. The behavior, without changing anything whatsoever from the installed state, is that the regular user cannot run x or the bash shell console window without first entering a password for an administrator account. If you can't start the bash shell (in a windows console or mintty window or whatever), then your problem isn't X specific, so I suggest you raise it on the main cygwin list. From the way you describe it, it sounds like you are getting a UAC prompt just to start these executables, rather than from anything they are trying to do? (although, from my recollection, the 'details' provided in a UAC prompt is some UUID, exactly how one is supposed to decode that to determine what action was taken which requires privilege, I don't know...) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: How do I start WM?
On 23/02/2010 14:11, Joseph Ess wrote: How do I start XWindows in full screen mode with OpenBox window manager? It used to be that I'd just put the command for the window manager at the end of startxwin.bat. Now that file's gone away. I can't use startxwin.exe as it is multi-window only. I can run XWin.exe, but the only command line or .XWinrc options I can find for that program are for Windows-specific functions(i.e. system tray icon). Read the documentation. http://x.cygwin.com/docs/ug/using.html#using-starting-startx -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: How do I start WM?
On 23/02/2010 15:11, Joseph Ess wrote: Sent: Tue, February 23, 2010 9:36:55 AM Subject: Re: How do I start WM? On 23/02/2010 14:11, Joseph Ess wrote: How do I start XWindows in full screen mode with OpenBox window manager? It used to be that I'd just put the command for the window manager at the end of startxwin.bat. Now that file's gone away. I can't use startxwin.exe as it is multi-window only. I can run XWin.exe, but the only command line or .XWinrc options I can find for that program are for Windows-specific functions(i.e. system tray icon). Read the documentation. http://x.cygwin.com/docs/ug/using.html#using-starting-startx I've read the documentation. The page you refer to mentions startxwin and startx. As I mentioned before, startxwin is for multi-window only. startx apparently requires a Cygwin shell. I'm trying to start Cygwin/XWindows in full screen mode. Is there something I missed? From your response, it sounds like you haven't actually tried using startx. Perhaps you should follow my suggestion and try it. Then please explain how it differs from what you want to achieve. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: bug report/suggested temp. patch: handling bursts of sent keys
On 13/02/2010 20:24, Mark Lillibridge wrote: Jon wrote: Thanks for the patch. Have you actually tested that this resolves your problem? Yes. Of course, really, really large bursts will still fail, but they should be very rare. Perhaps you might try the attached patch, instead, and see if that helps. On 23/01/2010 22:02, Mark Lillibridge wrote: I am not a Windows programmer. Can someone tell me if it's okay for winWindowProc to block? In particular, could we make it block until the mieq queue is not full? I think blocking would just result in a deadlock, as the X server is only single-threaded. The windows message pump is called when the server has no other work to do. This should be documented in [1], although perhaps that is lacking in detail. I notice that winWakeHandler()/winBlockHandler() try to completely empty the windows message queue, which leads to this problem as the server won't get a chance to process anything (draining the mieq queue) until they return. It might be enough to resolve this problem to allow those functions to complete after processing a limited number of events (chosen so as to not overflow the mieq queue), or if they notice that the event queue has crossed some high-water mark threshold. This is an interesting idea; I spent a while looking at WaitFor.c/WaitForSomething, but it's pretty opaque -- couldn't figure out when/under what conditions winWakeHandler is called by it. E.g., what's the actual priority between emptying the queue and processing window messages? Let's see: right at the end of hw/xwin/InitInput.c we open /dev/windows (a special device which becomes ready when there is anything on the windows message queue) and add that to the select mask. When the select returns, all the wakeup handlers are run, including winWakeHandler() which checks the WM message queue. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer 0001-Process-one-Windows-message-per-wakeup-rather-than-a.patch Description: application/itunes-itlp -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: How do I start WM?
On 23/02/2010 18:29, Joseph Ess wrote: Perhaps you should follow my suggestion and try it. I double-clicked on c:\cygwin\bin\startx. It opens a dialog titled Open With that instructs one to Choose the program you want to use to open this file. Does http://x.cygwin.com/docs/ug/using.html#using-starting-startx say to run startx by double clicking on it? Not it does not. Hint: that will not work as startx is bash script, which Windows explorer has no idea how to run. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: How do I start WM?
On 23/02/2010 19:04, Joseph Ess wrote: On 23/02/2010 18:29, Joseph Ess wrote: Perhaps you should follow my suggestion and try it. I double-clicked on c:\cygwin\bin\startx. It opens a dialog titled Open With that instructs one to Choose the program you want to use to open this file. Does http://x.cygwin.com/docs/ug/using.html#using-starting-startx say to run startx by double clicking on it? Not it does not. Hint: that will not work as startx is bash script, which Windows explorer has no idea how to run. I say that I'm trying to replace a .bat file, which one could presumable run by double-clicking on it, and you propose using a bash script, which one cannot double-click on and then berate me for trying it? Your question was How do I start XWindows in full screen mode with OpenBox window manager?, not How do I start XWindows in full screen mode with OpenBox window manager by double-clicking on something? Copy the XWin Server link on the start menu, and change the target of the copy from: C:\cygwin\bin\run.exe /usr/bin/bash.exe -l -c /usr/bin/startxwin.exe to: C:\cygwin\bin\run.exe /usr/bin/bash.exe -l -c /usr/bin/startx You probably want to change the name of the shortcut to something like XWin server (windowed) or whatever... Customize ~/.xinitrc to launch openbox by having it end 'exec openbox' and remove the line which starts twm. Double-click away! -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: How do I start WM?
On 23/02/2010 16:18, Christopher Faylor wrote: On 23/02/2010 14:11, Joseph Ess wrote: How do I start XWindows in full screen mode with OpenBox window manager? It used to be that I'd just put the command for the window manager at the end of startxwin.bat. Now that file's gone away. I can't use startxwin.exe as it is multi-window only. I can run XWin.exe, but the only command line or .XWinrc options I can find for that program are for Windows-specific functions(i.e. system tray icon). Read the documentation. http://x.cygwin.com/docs/ug/using.html#using-starting-startx -- Jon TURNEY Volunteer Cygwin/X X Server maintainer From: Jon TURNEYjon.tur...@dronecode.org.uk I've read the documentation. The page you refer to mentions startxwin and startx. As I mentioned before, startxwin is for multi-window only. startx apparently requires a Cygwin shell. I'm trying to start Cygwin/XWindows in full screen mode. Is there something I missed? From my reading of the documentation, it sounds like you'd get this behavior by adding server options to startxwin: startxwin -- [ server options ] This isn't actually the case, as startxwin always supplies the '-multiwindow' option to XWin, and there is no XWin option to specify the default (windowed) mode you can add after it. This is almost by design. There are a few differences between startx and startxwin, appropriate to running in windowed or multiwindow mode, and mixing them up would be a bad idea. * They use different scripts to start clients (~/.startxwinrc and ~/.xinitrc). This is because we expect the .xinitrc to end by starting a WM, which would a bad idea for .startxwinrc (as it would discover the internal WM is already running (hopefully)) and exit immediately. * startxwin exits after ~/.startxwinrc has completed and leaves XWin running, whereas startx waits until ~/.xinitrc exits (which is usually waiting for the WM started by it to exit) and then kills XWin. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: autogen.sh error
On 24/02/2010 01:44, J. Offerman wrote: I'm trying to compile Xserver. Following the guide on http://x.cygwin.com/docs/cg/prog-obtaining-source.html, I tried to run autogen.sh and got errors. First, I got this error saying: configure.ac:42: error: must install xorg-macros 1.2 or later before running autoconf/autogen So I ran the cygwin setup again and obtained xorg-macros and ran autogensh again, then I got this: -- joffer...@dell ./autogen.sh -V autoreconf-2.65: Entering directory `.' autoreconf-2.65: configure.ac: not using Gettext autoreconf-2.65: running: aclocal --force -I m4 configure.ac:84: warning: PKG_PROG_PKG_CONFIG is m4_require'd but not m4_defun'd I'd guess you don't have pkg-config installed? If so, you should go ahead and install everything mentioned on the previous page (http://x.cygwin.com/docs/cg/prog-build-prerequisites.html), you are going to need them all... -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Vista, new install, x server does not start, startxwin.bat missing
On 18/02/2010 03:38, Dr. M. C. Nelson wrote: And that is the problem (server, shell and xterm have to run as administrator). Even if the programs are set to run at elevated privilege, they will still throw up a UAC dialog, and the user will have to type in an administrator password. I think, as I said before, that this is Vista 'functioning as designed'. If you set an application to run as administrator, you will get a UAC prompt. If you tell us what the problems are with the programs which cause them to require administrator privileges, then perhaps we can fix them. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: 1.7.1 - problem with moving mouse programmatically with xte
On 16/02/2010 20:04, Radu Berinde wrote: It's unfortunate that this problem hasn't been fixed and Cygwin/X 1.7 is now part of the stable distribution. It's even more unfortunate that there's no way (none I could find anyway) to install the older version which worked for me. On Thu, Dec 17, 2009 at 9:37 AM, Radu Berinderadu.beri...@gmail.com wrote: I am using Cygwin/X 1.7.1 and I am running applications on this X server from a Slackware virtual machine. I am using xte v1.03 to move the mouse using keyboard shortcuts. Using ion3 as the window manager. When running e.g. # xte 'mousermove +250 0' the mouse cursor should move to the right; indeed, the window manager changes focus as if moved to the right, but the cursor doesn't visibly move. When mouse is moved, movement continues from the visible location (the xte movement is 'forgot'). This is an interesting and obscure use case. :-) The reason this doesn't work as expected seem to be this [1] upstream change. This prevents cursor moves inserted via XTEST from getting propagated to the native cursor. I'd be interested to know if this works with X server 1.7.x on any platform, I'd suspect possibly not. Using XWarpCursor() should move the native cursor correctly, so you might be able to use that instead? [1] http://cgit.freedesktop.org/xorg/xserver/commit/?id=dfb7de6571345061d1cd88ab915064fc376529fd -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris
-- From: Mark Veneziano Sent: Thursday, February 04, 2010 9:22 PM To: cygwin-xfree@cygwin.com Subject: Cygwin/X 1.7 keyboard/mouse freeze shortly after Xwin login to Solaris I am having a problem similar to the one mentioned by Tony Bennett in a couple of messages to this list back in January. I use XWin -query to establish an XDMCP connection to a Solaris 10 host (Tony's was AIX). I log in to Solaris CDE without any issues. Then as soon as I open and close one of the CDE panels, XWin no longer responds to keystrokes or mouse clicks. Moving the mouse around works fine, and hovering the mouse over a window with a tooltip (for example the perfmeter) still causes the tooltip to be displayed, but clicking and keystrokes are ignored. The only recourse is to terminate XWin from the X icon in the system tray or from Task Manager. As an experiment, I tried running GNOME rather than CDE as my desktop session, and that worked a little better, but as soon as I started a Motif-based application (IBM/Rational Apex Ada development environment), the keyboard and mouse clicks died again. Running CDE and Apex on the same Solaris box via Cygwin/X 1.5 works flawlessly. This problem only started occurring with 1.7. I even tried compiling xorg-xserver 1.7.4.0 from source to see if there was a problem specific to 1.7.3.0, but it behaved the exact same way. Thanks in advance for any suggestions! On 11/02/2010 13:41, Mark Veneziano wrote: Also, I tested using Xorg 1.7.4-1 on Fedora 12 Linux to remotely open CDE applications (dtterm, dtpad, etc.) on a Solaris box. The problems that occur under XWin 1.7.x with Cygwin do not occur with Xorg 1.7.x on Linux. So this keyboard/mouse hang issue seems specific to Cygwin, not to Xorg in general. The problem is reproducible even in multiwindow mode. Just remotely open a Solaris dtterm from Cygwin XWin and Linux Xorg. With XWin, right clicking in the dtterm window causes all keyboard/mouse input to stop without ever showing the context menu (the mouse pointer rotates from 10:00 to 2:00 and stays in that orientation). With Xorg, the context menu is shown correctly and input continues to work. Note that in multiwindow mode, keyboard/mouse input can be restored by closing the dtterm window from the native window manager and opening a new one. XWin does not have to be completely terminated to restore input. Not sure if that's a clue for any of the developers. Thanks for the bug report and clear reproduction steps. I can reproduce the problem. It's a bit disappointing that this has stopped working, after going to the trouble to make it work in Xserver 1.6.x. :-( It's not immediately obvious what's changed to make it stop working. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: X11 assertion fails
On 19/01/2010 16:01, Carl R. Crawford wrote: I just upgraded to the latest version of X11. A program that previously worked fine now reports the following error: assertion ret != inval_id failed: file /usr/src/ports/xorg/libX11/libX11-1.3.2-2/src/libX11-1.3.2/src/xcb_io .c, line 385, function: _XAllocID Aborted (core dumped) The program uses the xview toolkit from http://www.physionet.org/physiotools/xview/ This is a bug in Xview [1], and the answer to your problem is actually hidden on that page: If you are using Fedora 8, 9, or 10, or any distribution that includes X.org's libX11 version 1.1, you will need to use the i386-FC8 RPMs (or to compile XView from the sources included in xview-3.2p1.4-21.1.fc8.src.rpm) to avoid XAllocID runtime failures in XView applications. Applications that do not use this toolkit work fine. I did not recompile the xview library. You will need to recompile the xview library with the latest patches from that src rpm applied (or otherwise obtain an updated cygwin binary) [1] http://lists.freedesktop.org/archives/xcb/2009-October/005094.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/X and Vista and UAC
On 10/02/2010 17:01, Dr. M. C. Nelson wrote: I am trying to setup Cygwin/X for a Vista machine After setting some executables to run as administrator You shouldn't need to do this. Isn't this the reason for your UAC prompt? and then after cleaning up some lock files and log files, the x server and applications can be run by a user with administrator privileges. I think I understand the problem with the log file in this situation, and have a patch (see [1]). But the lock files should be removed when the X server is cleanly shutdown, so there's some bit of information I'm missing here. BUT, 1) The x startup (by any method) throws a UAC prompt 2) A non-privileged user cannot get past the UAC prompt without the administrator password Perhaps some .manifest files would cure this? I wasn't able to reproduce this problem doing some brief testing on W7 (I don't have access to a Vista system right now) There is a manifest embedded in the XWin application [2], but it doesn't do anything to affect UAC at the moment. Do you have some suggestion as to what you think it should contain? Are there any other ways to allow a non-privileged user to run cygwin in vista? [1] http://sourceware.org/bugzilla/show_bug.cgi?id=9778 [2] http://cgit.freedesktop.org/xorg/xserver/tree/hw/xwin/XWin.exe.manifest -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Graphics not appearing in Gaussview 3
On 09/02/2010 16:17, Jon Rhoad wrote:
I have been using the 1.5 version of cygwin/X installed in Windows XP to
connect to a SLES9 server running Gaussview with little problem. When I
upgrade to 1.7 everything works fine except the graphics do not show in the
Gaussview windows. The only e-mail from the archive I found related to
Gaussview is:
___
Dear Sir / Madam,
I need to use GView (a software for molecules visualization
constructed by Gaussian). An error occurs when I use the GView through
cygwin/X. This error is that the GView appears in my desktop. However,
when I try to display a molecule, no molecule is observed in the window
of GView.
The Gaussian's reply is :
{{{ I see the following line in the setup,
xorg-x11-xwin-gl (761 KiB; optional, the Cygwin/X X Server with
***EXPERIMENTAL*** GLX acceleration)
and the fact that GaussView starts but does not show molecules, which
require GLX while the splash screen and the basic dialog windows to
not, can be the problem. Can you turn off the experimental GLX
acceleration? }}}
How can I do it?
Regards,
Patrick Pang
___
There was no reply to this e-mail that I could find. I am not a programmer,
just an end user, so please put any reply in layman's terms. Thank you.
You are almost certainly not using experimental GLX acceleration, which is
the experimental XWin_GL.exe X server, as that hasn't been included in recent
Cygwin/X releases and never in Cygwin 1.7.
So that is probably not the cause of your problem.
Unless you can reproduce your GLX problem with a free application, your best
idea is to contact the supplier of the application you have a problem with for
support.
--
Jon TURNEY
Volunteer Cygwin/X X Server maintainer
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://x.cygwin.com/docs/
FAQ: http://x.cygwin.com/docs/faq/
Re: Downgrading cygwin/x and/or cygwin
On 10/02/2010 12:36, Humann, Wolfram wrote: I'm not sure I can provide something much more useful than the lame it worked in a previous installation -- that's why I consider downgrading. Here's what I can provide: - My current cygwin is version 1.7.1-1, my Cygwin/X is 1.7.3.0 - I run 'XWin -multiwindow -clipboard -emulate3buttons' - Then I use ssh to log into a remote Linux machine and start a konsole window - From the konsole I start various applications all of which work fine except one large company-owned application which crashes with a SIGSEGV - Only the application crashes, the konsole and all other running remote apps continue to work Sad to say, this probably indicates a bug in your company application, so you should report it to those responsible for maintaining it. Even if the X server is returning something invalid, it could do something more helpful, like exit with an error, rather than crash :-) - The application works fine when I'm locally on the Linux machine I log into - It also works fine when I use Xming (version 6.9.0.31) instead of Xwin, but Xming has several other problems (e.g. clipboard) - It also worked fine on my previous PC where I had cygwin 1.5.24 with Xwin 6.8.99.901 Unless someone can provide instructions how to find the rootcause for the SIGSEGV I would consider downgrading to something close to my older version and see if that helps. 1. gdb --args your-mystery-application 2. Wait for the '(gdb)' prompt, type 'run' 3. Do whatever you do to make the crash happen. You should get another '(gdb)' prompt, type 'bt full' If the backtrace shows the crash is in a toolkit or X library, then this might be worth addressing in the X server. I already tried to downgrade just Xwin and downloaded XWin.20081204180806.exe.bz2 and XWin.20090629103603.exe.bz2 form ftp://cygwin.com/pub/cygwinx/ but that alone did not help. A colleague has the same problem, so it's not just my installation. Is there an archive of old versions? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Can't start X after upgrading to cygwin1.7
On 03/02/2010 19:43, Jon TURNEY wrote: I had the same symptoms. Non-administrator WinXP users couldn't start the XWin Server, and no /var/log/XWin.0.log was created. The log file could get created if there wasn't one already present, which led me to put the log file in the user's directory instead of /var/log. The other problem was that users couldn't create the /tmp/.X0-lock file and /tmp/.X11-unix/X0 socket if they were left by another user. The solution was to give each user their own /tmp directory. I fixed it with the following: 1. Added a line in /etc/profile, just before the chmod 1777 /tmp line: mount -f $USERPROFILE/Local Settings/Temp /tmp 2. Changed the XWin Server icon so instead of just startxwin.exe, it says: ... startxwin.exe -- -logfile ~/XWin.log (Don't take out the beginning of the command that runs startxwin through bash, or the /etc/profile won't get run. The quotes are needed because this is the argument to bash's -c option.) Could any of the guru's comment on how good or bad these solutions are? Thanks. It's long been suspected that there is some problem with lockfiles running XWin as a non-administrator user (See FAQ 3.4,[1]), but I've never had a sufficient clear description to understand what the problem is, until now. Okay, thanks to having some useful reproduction steps I've done a bit of testing and generated some patches (at [1]) which should hopefully address some of these issues. The thing I don't understand at this stage is that the /tmp/.X0-lock file and /tmp/.X11-unix/X0 socket should be removed when the X server shuts down, so shouldn't be causing problems. Is there some specific way the server is being exited which doesn't remove these files? [1] http://sourceware.org/bugzilla/show_bug.cgi?id=9778 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: [ANNOUNCEMENT] Updated: xterm-255-1
On 05/02/2010 18:57, Yaakov (Cygwin/X) wrote: The following package has been updated for the Cygwin distribution: *** xterm-255-1 xterm is a terminal emulator for the X Window System. It provides DEC VT102 and Tektronix 4014 compatible terminals for programs that can't use the window system directly. This is an update to the latest upstream version. I've adjusted the dependency information in the setup.hint for the xterm package on sourceware.org from libncurses9 to libncurses10, which this updated xterm requires. Thanks to the people who reported this problem, which should now be corrected once the mirrors are updated. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: startxwin.exe and ssh-agent
On 05/02/2010 12:35, Joe Warren-Meeks wrote: I've googled extensively and searched the mail archives, but I can't find anything which describes how to configure cygwin to use ssh-agent. In the 1.5 world, I modified startxwin.bat to prefix the XWin.exe line with ssh-agent, and this worked a treat. How should I do this now? I basically want my whole Xserver to run as a child of my ssh-agent so all xterms and other processes will collect the credentials. Creating a shortcut which starts the X server with: 'C:\cygwin\bin\run.exe /usr/bin/bash.exe -l -c ssh-agent startxwin' should probably do what you want. Note that (as before), processes you start other than via the X traymenu (for example, from a start menu item) aren't children of ssh-agent, so won't inherit SSH_AUTH_SOCK and SSH_AGENT_PID from it. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: new cygwin/x11 and backingstore
On 04/02/2010 15:25, anne crespy wrote: Hi, Before updating Cygwin/X with the new version (X11R7) I launched Xwin with this command : XWin -multiwindow -clipboard -silent-dup-error +bs and the backingStore option worked correctly in a graphic window (built with tcl/tk8.4 and Xlib). You were just lucky :-) The backing store attribute is a hint; the server doesn't have to honour a request to use backing store. Note that even if you have an X server which does provide backingstore, the server doesn't have to provide an unlimited amount of memory for it, so it may discard backingstore at any time. Any X client which *requires* backingstore to work is broken, not correctly redrawing windows when they are exposed. See [1] Now, it doesn't work. When the graphic window is overlapped its contents is corrupted by the top window. I didn't have this problem with the beta-version in november 2009. What have changed since? backingstore is implemented in terms of Composite in the X.Org server since X server 1.5.0 Unfortunately, we have had to disable Composite in the current X server due to crashes. If/when these problems with Composite are fixed, things may work a little better... [1] http://www.mail-archive.com/x...@lists.freedesktop.org/msg04440.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: bug report/suggested temp. patch: handling bursts of sent keys
On 12/01/2010 18:44, Mark Lillibridge wrote: Background: I use Nuance's Dragon NaturallySpeaking voice recognition software to control remote Linux systems from a Windows box. I open up xterms and emacs windows on a local Cygwin X server. Occasionally the voice recognizer makes a mistake, which has to be corrected. The mechanics of how this is done are unimportant; what matters is how Dragon applies the correction. It does this by sending a large burst of keystrokes via something like SendKeys. (E.g., many (shift) arrow keys to select the text to be changed, a backspace to delete it, then the replacement text.) Unfortunately, there is a bug in the current released Cygwin X server that causes it to drop sent keystrokes when the burst exceeds a given size (roughly 64 or 128 keys depending on the version). This breaks correction, and forces painful manual fix up of the text. Note that what is really dropped are key events so the system can get into the state where the shift key remains pressed or a key starts repeating forever because the key up event was dropped. [repro snipped] Mark, Thanks for the detailed bug report and clear reproduction steps. On 23/01/2010 22:02, Mark Lillibridge wrote: I am not a Windows programmer. Can someone tell me if it's okay for winWindowProc to block? In particular, could we make it block until the mieq queue is not full? I think blocking would just result in a deadlock, as the X server is only single-threaded. The windows message pump is called when the server has no other work to do. This should be documented in [1], although perhaps that is lacking in detail. I notice that winWakeHandler()/winBlockHandler() try to completely empty the windows message queue, which leads to this problem as the server won't get a chance to process anything (draining the mieq queue) until they return. It might be enough to resolve this problem to allow those functions to complete after processing a limited number of events (chosen so as to not overflow the mieq queue), or if they notice that the event queue has crossed some high-water mark threshold. On 05/02/2010 00:45, Mark Lillibridge wrote: Given there doesn't seem to be anyone competent enough to attempt a more ambitious patch, I think we should apply the temporary patch. How should we/I. go about doing this? The Cygwin X website and contributors documentation doesn't actually say how to do this. The patch is enclosed at the end. - Thanks, Mark $ diff -u mieq.c~ mieq.c --- mieq.c~ 2009-10-15 21:38:27.0 -0700 +++ mieq.c 2010-02-04 16:42:30.773405200 -0800 @@ -58,7 +58,7 @@ # includeX11/extensions/dpmsconst.h #endif -#define QUEUE_SIZE 512 +#define QUEUE_SIZE 25000 #define EnqueueScreen(dev) dev-spriteInfo-sprite-pEnqueueScreen #define DequeueScreen(dev) dev-spriteInfo-sprite-pDequeueScreen Thanks for the patch. Have you actually tested that this resolves your problem? I'll add some words about contributing patches to the CG guide documentation, thanks for pointing out this oversight. Perhaps that's the reason we hardly ever get any :-) [1] http://x.cygwin.com/docs/cg/prog-server-architecture.html#prog-server-architecture-input -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Vista, new install, x server does not start, startxwin.bat missing
On 03/02/2010 01:18, Dr. M. C. Nelson wrote: We installed cygwin/x on a new acer laptop (vista). The X server does not start from the icon, and the x server does not start from the command line startx The batch script startxwin.bat is not present. Don't panic about the absence of startxwin.bat. startxwin.(bat|sh) have been retired in favour of a shiny new startxwin.exe [1] in an effort to make... X server startup more reliable :-) I've finally got around to updating the Cygwin/X User Guide [2] to take note of this fact [1] http://cygwin.com/ml/cygwin-xfree-announce/2009-12/msg5.html [2] http://x.cygwin.com/docs/ug/using.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: startxwin.exe and emulate3buttons
On 02/02/2010 15:23, Anthony Becker wrote: In the previous version of Cygwin/X, when XWin.exe was started by startxwin.bat, I was able to specify the emulate3buttons option to simulate a middle mouse button press by pressing the right and left buttons simultaneously. Now that I must use startxwin.exe to start the X Server, how do I specify this? The syntax for specifying additional server options is documented in 'man startxwin' You want 'startxwin.exe -- -emulate3buttons' -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: screen pixel density VS startxwin.exe
On 19/01/2010 14:45, Alexandre Beaugy wrote: For many years now, I used to set my screen pixel density to 90 pixels per inch, instead of the default 75 ppi, in the startxwin.bat file. My startxwin.bat used to be executed, then, at windows startup, to get a full functional Cygwin-X environment (at 90 ppi, by the way). Here is the commandline I did put in my startxwin.bat file: [...] run XWin -multiwindow -clipboard -dpi 90 [...] Then, I recently upgraded my Cygwin-X installation from 1.5.x to 1.7.x, and now the startxwin.bat script disappeared, replaced by the startxwin.exe binary. That did not trouble me much. Nevertheless, I'm not able to set my screen pixel density anymore, or at least, I did not found the way (config file, commandline argument, ...) to set it again. I tried, as target of XWin Server shortcut: `run C:\cygwin\bin\run.exe /usr/bin/bash.exe -l -c /usr/bin/startxwin.exe -- -dpi 90' ^^^ In order to specify particular server options, append a double dash (−−) to the startxwin command line (after any client and arguments) followed by the desired server options., as it is said in the `man 1 startxwin'. Nevertheless, my option seems not to reach the server arguments, because, the screen pixel density of my Cygwin-X session remains stuck to 75ppi (as `xdpyinfo' reports). You really deserved to succeed after paying such good attention to the man-pages. I think you will find that setting the shortcut to 'run C:\cygwin\bin\run.exe /usr/bin/bash.exe -l -c /usr/bin/startxwin.exe -- -dpi 90' should succeed in passing the dpi option down to startxwin. Note that the quotes around the entire command after bash's -c option, which are needed, otherwise bash will interpret the '-- -dpi 90' as options for itself. I think perhaps we should arrange for the shortcut to be created with these quotation marks in place to make it a bit more obvious where to add extra options. I also heard about a file called ~/.xserverrc to alter the way the Xserver is launched. But I did not found anything detailed about its content and how it shall be addressed. .xserverrc is discussed in the xinit manpage, and while you could achieve the same ends using that, I wouldn't recommend it, as this will also affect X servers started via startx or xinit. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Can't start X after upgrading to cygwin1.7
On 23/01/2010 15:37, Jeff Spirko wrote: On Sat, Jan 16, 2010 at 12:56 PM, Andrew Seniora...@andrewsenior.com wrote: I've had cygwin installed for a year on my Thinkpad T61, running Windows XP professional, and just ran the latest setup.exe from cygwin.com. I now can't run X with startxwin.exe (no process appears, no icon in the system tray, clients won't start) No /var/log/Xwin.0.log is written, nor anywhere else I can see in /var/log I think the original poster had a broken installation, there's no other reason for it to get fixed by a reinstall :-) I had the same symptoms. Non-administrator WinXP users couldn't start the XWin Server, and no /var/log/XWin.0.log was created. The log file could get created if there wasn't one already present, which led me to put the log file in the user's directory instead of /var/log. The other problem was that users couldn't create the /tmp/.X0-lock file and /tmp/.X11-unix/X0 socket if they were left by another user. The solution was to give each user their own /tmp directory. I fixed it with the following: 1. Added a line in /etc/profile, just before the chmod 1777 /tmp line: mount -f $USERPROFILE/Local Settings/Temp /tmp 2. Changed the XWin Server icon so instead of just startxwin.exe, it says: ... startxwin.exe -- -logfile ~/XWin.log (Don't take out the beginning of the command that runs startxwin through bash, or the /etc/profile won't get run. The quotes are needed because this is the argument to bash's -c option.) Could any of the guru's comment on how good or bad these solutions are? Thanks. It's long been suspected that there is some problem with lockfiles running XWin as a non-administrator user (See FAQ 3.4,[1]), but I've never had a sufficient clear description to understand what the problem is, until now. The only problem I see with your solution is that you are assuming only a single user is logged in. This might not be the case on a Terminal Server (although it would probably take a specific sequence of privileged/unprivileged logins and XWin startups/shutdowns to show a problem) This would be better fixed in XWin itself by arranging to give these files the correct permissions, if we could work out how to do that :-) [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-cant-read-lock-file -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: screen not updated outside 1680x1050
On 18/12/2009 06:11, Frédéric Bron wrote: XWin -multiwindow -clipboard -silent-dup-error ddxProcessArgument - Initializing default screens winInitializeDefaultScreens - w 1680 h 1050 ^ Looks like the system doesn't see your whole resolution at startup. I'd pass -screen 0 1920x1200 to XWin.exe. I tried this to reduce my screen resolution, and saw things similar to what you describe outside the declared screensize. Works fine, thanks a lot! In fact I saw only the following lines so that I thought it was OK with the resolution: 2009-12-17 13:47:46 winWindowProc - WM_DISPLAYCHANGE - orig bpp: 32, last bpp: 32, new bpp: 32 2009-12-17 13:47:46 winWindowProc - WM_DISPLAYCHANGE - new width: 1920 new height: 1200 As you have discovered, XWin doesn't understand about display resolution changes. This is actually mentioned in the XWin man page, and at [1] [1] http://x.cygwin.com/devel/todo.html -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Windows Batch File to start xterm on Cygwin/X 1.7.x
On 25/01/2010 22:53, Zheng-Da Tan wrote: I like to have one icon in Windows that I can click on to open a new xterm window: - if the X server is not running, the icon would start X and launch a xterm window - if the X server is running, it will simply open a new xterm window on the existing X server This was how it used to work by default in a very old version of Cygwin. It had been broken once before (search silent-dup-error in the mailing list archives), fixed, and now broken again when startxwin.bat became startxwin.exe. How about saying how it's broken, so it can be fixed, and we can get rid of horrible batch files for ever? -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: startxwin.exe without lingering DOS window
On 06/01/2010 23:14, Raul Acevedo wrote: How do you run startxwin.exe so that it doesn't leave a lingering MS-DOS window after it starts the programs in ~/.startwinrc? As pointed out by others, this is achieved by using 'run' (which is designed for exactly that purpose) to invoke startxwin. The 'XWin Server' start menu entry created by the installation of the xinit package should contain a command line to do that: 'C:\cygwin\bin\run.exe /usr/bin/bash.exe -l -c /usr/bin/startxwin.exe' (this also gives XWin and all your clients a login shell environment) If that isn't working to hide the DOS window, that would probably be a bug in run. Are there plans to update http://x.cygwin.com/docs/ug/using.html so people like me don't waste a few hours finding out that startxwin.bat has been replaced by startxwin.exe and that it is documented in its man page? Yes, the plan is to update that documentation either: i/ When someone sends me a patch, or ii/ When I have some time and the inclination to do so -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Option to disable PRIMARY clipboard
On 12/11/2009 14:27, Corentin Chary wrote: On Thu, Nov 12, 2009 at 2:53 PM, Jon TURNEYjon.tur...@dronecode.org.uk wrote: On 09/11/2009 16:48, Corentin Chary wrote: Is there a way to disable the PRIMARY clipboard ? Would a patch like that http://sourceware.org/ml/cygwin-xfree/2003-06/msg00148.html (with a command line option) be accepted ? It's far more likely to be accepted if you say what problem you are trying to solve and why a patch like that is the right solution. We use nx to display X application hosted on a Linux servers. Clients could be on Linux, Windows or MacOS. The PRIMARY clipboard really confuse windows (and MacOS) users. An option (XWin.exe -no-primary-clipboard) would really be great for that. You haven't really answered my question. So far I understand users working in some way involving nx get confused trying to do something with the clipboard because something happens and making something diferent happen will not confuse them. Note that patch also needs to exist to be accepted. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: problem with startxwin.exe and a possible solution
On 21/01/2010 23:13, Marco Atzeri wrote: Hi, after the transition from startxwin.bat to startxwin.exe, I had the Xwin Server menu pratically unusable. The server start but the xterm usually don't show up while ps reports it running. Running a second xterm could appear but most of the time as if the window manager is absent: no border, no possibility to move the program and similar. The Xserver is also unstable, the icon on the toolbar usually disappear and to terminate the server I really need to kill it with Process Explorer. I can reproduce this if I have CYGWIN=tty (which I note in your cygcheck.out) in my Windows environment. Without that it seems to work fine. I'm not sure that I believe that this is the underlying cause of the problem, and if it is, I have no idea how it could make difference or why it wasn't a problem previously. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: xserver bug?
On 20/01/2010 09:13, Sylvain RICHARD wrote: Charles Wilson wrote: I've noticed that with XWin 1.7.3 (and perhaps earlier versions; not sure), the key combination CTRL-SHIFT-0 (zero) doesn't generate any events. CTRL-SHIFT-1 thru -9, alphabetic keys, no problem -- just not zero. I'm afraid I can't reproduce this problem; xev and urxvt both respond to ctrl-shift-0. snip Is this a bug, or a designed behavior, in XWin? This may be designed behaviour /in windows/. Just out of curiosity, can you check your language bar options. If I remember correctly, you can set Ctrl+Shift global hotkeys there to shift between keyboard layouts and input languages. If it is the case that Windows is treating ctrl-shift-0 as a special keypress for some reason and processing it without offering it to applications, starting the X server with the -keyhook option should help. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Problem with new xinit - console window doesn't open (but bash starts)
On 22/11/2009 20:24, Jon TURNEY wrote: On 20/11/2009 15:05, Jim Reisert AD1C wrote: I tried adding the missing fonts. I tried SET LANG=C Starting the X server is still unpredictable. Does this mean it sometimes works and sometimes doesn't? Or does it mean it fails in different ways each time you try? Now my log file (attached) shows: It also claims another window manage is running (and therefore shuts down), which is untrue (there is no other Xwin process running). Since I have touched 'check for another window manager' code recently, I was kind of expecting to see that I'd managed to break it in some way, but.. I'm not quite sure what to make of that log file. This is very strange. If you look for (II) xorg.conf is not supported, you'll see that it looks like that server starts again about half-way through. I did wonder if you were managing to somehow start two copies of XWin near-simultaneously, but I don't think that would produce a logfile like that (since it's not opened in append mode, you probably get the logfile from the last process to close it or something like that) An alternative explanation is that there is a really inconveniently timed server regeneration happening: just after the internal client threads have been created, but before they can connect, so they are still hanging around to be accepted by the next server generation. If we did have twice as many threads as we are supposed to, that would explain the final failure due to another WM running (as there's another WM thread, note that there are two 'winMultiWindowXMsgProc - Hello' lines) No idea why the server is regenerating though, needs more meditation :-) After some meditation, I think that the server regeneration may be being caused by a race condition introduced by the use of checkX in the startxwin.bat script: If checkX connects to and then disconnects from the Xserver before the internal client threads get a chance to connect, the Xserver will restart when checkX disconnects, causing these duplicate threads. Fortunately, this should have a simple workaround of removing the checkX line from startxwin.bat (or replacing it with a delay if you care about getting an xterm started by it) The correct fix for this behaviour is not immediately obvious, and I don't really have time to work on it at the moment, so I don't have a patch for this. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: [ANNOUNCEMENT] [1.7] Updated: cygwin-1.7.0-68
On 05/12/2009 09:03, Frédéric Bron wrote: I just uploaded a new Cygwin 1.7 test release, 1.7.0-68. Installed this version and cannot start Xwin anymore. Here is the content of /var/log/XWin.0.log: This looks to be the same problem as in [1]. You might try the workaround of removing the checkX line from startxwin.bat [1] http://cygwin.com/ml/cygwin-xfree/2009-12/msg00074.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: timed out waiting for reply from selection owner
On 16/12/2009 07:45, Hemal Pandya wrote: This is rather strange and I don't even know if I have a correct repro. The symptoms I see is that I am able to copy in X (XEmacs or xterm) and paste in Windows but not vice-versa. Immediately after I start X copy/paste works both ways. I suspect this has something to do with Microsoft Visual Studio. It seems to hijack the clipboard and copy/paste stops working altogether. I found this using a utility GetOpenClipboardWindow I found on the net. When copy/paste across windows Apps stops working, this utility always reports that VSS owns the clipboard. I often have to kill VS to get other programs to interact with the clipboard. But even after I kill VS Studio, Cygwin-X is not able to get data from the clipboard. Any time I paste in X I see the above message in /var/log/XWin.0.log Any suggestions what might be wrong and how I can work-around this issue? Thanks for the problem report. Unfortunately it is lacking information to be useful. You might consider reading the problem reporting guidelines at [1] and trying again. In this particular case, attaching the /var/log/XWin.0.log to show the context for the error message is would be sensible. [1] http://cygwin.com/problems.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: extensions (shared mem, security)
On 15/12/2009 19:00, Linda Walsh wrote: looking at the startup log, I don't see the X-Security extension being initialized. Is it not supported? Also discussed in the Cyginw/X FAQ at [1] Also, perhaps more important for me is the message: 2009-12-15 10:39:27 XFree86-Bigfont extension local-client optimization disabled due to lack of shared memory support in the kernel NT supports shared memory objects and I thought cygwin uses shared memory for cross-process communication. So what's this about X not knowing about the shared memory support? I don't know how this can be made clearer than [2] [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#AEN569 [2] http://x.cygwin.com/docs/ug/using-shared-memory.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cut/paste in an xterm takes extraordinarily long time
On 10/12/2009 18:18, Jim Reisert AD1C wrote: I'm having problems with copying text from a Windows application (Firefox) and pasting into Cygwin Emacs (23.1.1). I'm running Cygwin 1.7 (latest) and Xorg (latest). When pasting into Emacs, my mouse pointer changes to an hourglass, and after what seems like an eternity (10-30 seconds), the paste will finish, often times with the wrong text, rarely with the correct test. When the text is wrong, how is it wrong? Scrambled? Previous paste? Backwards? Random letters? Translated to Urdu? I am running RealVNC also. I remember several/many months back, there seems to be a cut/paste issue with Cygwin that was exacerbated by running VNC. Could that problem have returned? You could test this by closing RealVNC and seeing if the problem doesn't occur? How can I help debug this? I've attached my cygcheck output. If you could follow the steps I outlined in my email replying to the last time you mentioned this [1], that would be helpful? http://sourceware.org/ml/cygwin-xfree/2009-10/msg00129.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: [PATCH] Repair '-nolock'
On 30/11/2009 07:57, Fergus wrote: (Sorry to send twice: originally sent yesterday Sunday 0728 GMT but never arrived at cygwin-xfree at cygwin dot com; 2nd attempt today Monday 0752 GMT and hope for better progress.) Message was: The reason why your first message did not arrive at the mailing list was because you only sent it to me, personally. Such mails get ignored for reasons laid out at [1]. Please be more careful in future :-) [1] http://cygwin.com/problems.html Thank you. 1 Will this amendment eventually be incorporated into an updated .bz2 file under /release? Assuming it works and doesn't break anything else, yes. 2 If not then (I am sorry) I do not know how to introduce a patch into an existing executable. Something like: unwrap src, change os/utils.c, rebuild? If this is something everybody in the world except me knows, can you direct me to a howto? http://en.wikipedia.org/wiki/Patch_%28Unix%29 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: extensions (shared mem, security)
On 16/12/2009 20:31, Jon TURNEY wrote: On 15/12/2009 19:00, Linda Walsh wrote: looking at the startup log, I don't see the X-Security extension being initialized. Is it not supported? Also discussed in the Cyginw/X FAQ at [1] [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#AEN569 Hmm... Fixed an error in the docbook source so that anchor has a proper name which doesn't change every time the documentation is rebuilt. http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-trusted-untrusted-x11-forwarding -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Unrecognized option: -nolock
On 28/11/2009 06:53, Fergus wrote: With the latest release 1.7.0-67 Cygwin [1.7] now works glitch-free on FAT32 filesystems; but to run X on FAT32 we still have to revert to 1.7.0-60. The recommended fix (there may be more than one but this seems potentially the easiest and most effective) is to run X -nolock. But this option is unrecognised in the current XWin. Please could -nolock be reintroduced so that FAT32 users can take full advantage of the current .dll? Thank you. Fergus Thanks for reporting this bug. This is an upstream change in 1.7.0 and later, and looks like it's been disabled by accident during some code clean-up. Patch to follow. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: cygwinx 1.7 beta - crashes runing virt-manager
On 20/11/2009 04:04, Michael Breuer wrote: Connecting to F12 running Gnome, running python based applications causes Xwin to segfault. For example, running virt-manager kills Xwin immediately. Most of the F12 admin tools exhibit this behavior. I tried -swcursor based on some old discussions on this list regarding Java crashing Xwin (from 2007). Doesn't help now :( Anyone have suggestions? I can't even use XDMCP to log-in to Fedora 12 (Constantine), due to random segfaults :-( These seem to be due to a bad interaction between the damage tracking for the Composite extension and damage tracking for the shadow framebuffer. So can you try disabling the composite extension with '-extension Composite'? (note the captialization) -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: checkX problems
On 20/11/2009 19:43, Ken Brown wrote: On 11/20/2009 12:14 PM, Charles Wilson wrote: I've integrated Lothar's patch into run2/checkX (along with some other internal changes), and published a test release. Please try run-0.3.1-1 and let me know if it fixes your problems with checkX. But the new behavior of the timeout option (my problem #1) works fine, with one caveat: If I start the X server with startxwin.bat, it immediately exits and claims that a server is already running. This was also reported earlier today by Jim Reisert: http://cygwin.com/ml/cygwin-xfree/2009-11/msg00158.html Not quite: It seems that report says that the internal multiwindow-mode window manager claims another window manager is running (although it's difficult to be absolutely sure as the exact error message isn't reported :-)) Could it be that checkX is tricking XWin into thinking that a different X server is running? (I have no idea how XWin decides whether an X server is running, so this may or may not be plausible.) Strangely, the problem doesn't occur if I use startxwin.sh instead of startxwin.bat. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Problem with new xinit - console window doesn't open (but bash starts)
On 20/11/2009 04:07, Jim Reisert AD1C wrote: More info: If I run startxwin.bat from a CMD.exe command line, the bash console xterm opens just fine. I only have problems starting it from the shortcut. This is typical of the current issue we have where 'run xterm' blocks when xterm tries to output 'Warning: Missing charsets in String to FontSet conversion' Any one of: - installing the CJK fonts - having 'tty' in the CYGWIN environment variable - having the LANG environment set to a non-UTF-8 locale should work around this problem Note that the environment variable will have to be set via the system applet in the Windows control panel, as only that controls the environment for the startxwin.bat started from the start menu... I do see some interesting things in the log file, however: MIT-SHM extension disabled due to lack of kernel support XFree86-Bigfont extension local-client optimization disabled due to lack of shared memory support in the kernel I hope these are explained by http://x.cygwin.com/docs/ug/using-shared-memory.html [dix] Could not init font path element /usr/share/fonts/TTF/, removing from list! [dix] Could not init font path element /usr/share/fonts/OTF/, removing from list! [dix] Could not init font path element /usr/share/fonts/Type1/, removing from list! These are expected when these font directories are empty, which is quite normal if you haven't installed all possible fonts Warning: Missing charsets in String to FontSet conversion This is xterm warning that the fontset has no fonts for CJK locales (as no CJK fonts are installed) -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Problem with new xinit - console window doesn't open (but bash starts)
On 20/11/2009 15:05, Jim Reisert AD1C wrote: I tried adding the missing fonts. I tried SET LANG=C Starting the X server is still unpredictable. Does this mean it sometimes works and sometimes doesn't? Or does it mean it fails in different ways each time you try? Now my log file (attached) shows: It also claims another window manage is running (and therefore shuts down), which is untrue (there is no other Xwin process running). Since I have touched 'check for another window manager' code recently, I was kind of expecting to see that I'd managed to break it in some way, but.. I'm not quite sure what to make of that log file. This is very strange. If you look for (II) xorg.conf is not supported, you'll see that it looks like that server starts again about half-way through. I did wonder if you were managing to somehow start two copies of XWin near-simultaneously, but I don't think that would produce a logfile like that (since it's not opened in append mode, you probably get the logfile from the last process to close it or something like that) An alternative explanation is that there is a really inconveniently timed server regeneration happening: just after the internal client threads have been created, but before they can connect, so they are still hanging around to be accepted by the next server generation. If we did have twice as many threads as we are supposed to, that would explain the final failure due to another WM running (as there's another WM thread, note that there are two 'winMultiWindowXMsgProc - Hello' lines) No idea why the server is regenerating though, needs more meditation :-) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: XWin stopped working with xorg-server 1.7.1 - no log-file created
On 19/11/2009 16:25, Michael Breuer wrote: For what it's worth, I tracked that down yesterday by looking at the error log on the server - there was a complaint about clipboard - don't recall what... seemed innocuous but figured I'd try turning off clipboard. Viola, it worked. Could you tell me what the clipboard related error message is? Then I could perhaps try to fix it rather than tell everyone to use -noclipboard :-) One other point - as older versions of CygwinX have major issues (100% cpu) with the newer version of Gnome (now in F12), more people are going to be trying the beta and hitting this. Not sure what code is involved, but including this on the beta FAQ might be worthwhile. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: 'run xterm' fails to open a window
On 13/11/2009 03:14, Ken Brown wrote: On 11/12/2009 3:17 PM, Jon TURNEY wrote: On 10/11/2009 23:44, Mike Ayers wrote:] Apparently, run.exe is not providing stdout/stderr to dump to. The workaround: [SNIP] C:\mikeC:\cygwin\bin\run.exe -p /usr/bin sh -c gvim/dev/null 21 C:\mike [/SNIP] Wait, are you saying that the process that run starts is blocked if it tries to output anything? Ah, so this explains my xterm problem that started this thread. When I start xterm from a shell, I always get the message Warning: Missing charsets in String to FontSet conversion When I start xterm via run, this message apparently has nowhere to go and the xterm process is blocked. Mike's workaround (with 'gvim' replaced by 'xterm') solves the problem, as he said. Is this a bug in run? It does kind of look like some problem with run, but not a universal one, 'run xterm' has been working as expected for me, even when it has to output something. Experimenting a bit, this seems to be related to having CYWIN=tty set in the windows environment. If I remove that, I can reproduce the problem you see. I really don't understand the invisible console stuff well enough to attempt a fix, though... -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: XWin stopped working with xorg-server 1.7.1 - no log-file created
On 20/11/2009 15:46, Michael Breuer wrote: I've attached the log. Thanks. Can you also please tell me what OS the XDMCP host at 10.0.0.1 is running? Jon TURNEY wrote: On 19/11/2009 16:25, Michael Breuer wrote: For what it's worth, I tracked that down yesterday by looking at the error log on the server - there was a complaint about clipboard - don't recall what... seemed innocuous but figured I'd try turning off clipboard. Viola, it worked. Could you tell me what the clipboard related error message is? 2009-11-20 10:46:08 winClipboardIOErrorHandler! 2009-11-20 10:46:08 winClipboardProc - setjmp returned for IO Error Handler. 2009-11-20 10:46:08 winClipboardShutdown - Clipboard thread has exited. Hmmm... The problem here is that the internal clipboard integration client is terminating for some reason. (This is not entirely unexpected in XDMCP use, as login managers often try to kill other clients to ensure they don't watch you type your password, but there is code in place to try to prevent this happening to the clipboard integration client, in case you want to paste your password...) Since Xserver 1.7.1-1, terminating the clipboard integration client causes the Xserver to terminate as well (although this change, which also applies to the -multiwindow internal window manager, doesn't seem to be mentioned in any of the announce emails) My reasoning for this change was that we shouldn't accept the options turning these internal clients on and then silently allow them to fail. This was particularly bad with the -multiwindow internal window manager, as if that client exits all new windows have no native frames, which is particularly useless. I guess this also interacts unfortunately with -clipboard being the default now, since 1.6.1.902-1 So, perhaps we need to do something smarter here, either improve the way we try to avoid the internal clipboard client getting killed, or perhaps restart it after an interval ... -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: XWin stopped working with xorg-server 1.7.1 - no log-file created
On 19/11/2009 08:33, Gertjan van Noord wrote: after upgrading to xorg-server 1.7.1-3 starting X does no longer work. Previous version 1.6.? worked. However, I can't seem to go back to that version using the cygwin setup programme. the startxwin.bat script does not complain, but no X logo appears, and indeed no X programs can be started. moreover no log-file is created in /var/log, which seems to suggest something very basic is going wrong? starting XWin from the command line, also with -logverbose 10 does not produce any output. rebooting does not change a thing I've attached my cygcheck.out You should perhaps try 'cygcheck /usr/bin/XWin.exe' to see if it identifies a missing DLL. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin error
On 18/11/2009 17:20, baseball07 wrote: Thank you for the suggestion. Could you please type out the syntax. I'm a newbie to Linux code. Thank you very much. http://x.cygwin.com/docs/ug/using-remote-apps.html -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Some Eclipse dialogs have become non-resizable with X server 1.7.1
On 13/11/2009 13:06, Csaba Raduly wrote: In the process of switching over to 1.7, I started running its X server (instead of the X server of 1.5). A few days later I noticed that some Eclipse windows (e.g. the Find dialog, the Project-Properties, the Window-Preferences, etc.) can not be resized anymore (the Size entry in the system menu is grayed out). This annoyed me when the Find dialog came up too small to see (or use) one of the checkboxes, and I couldn't make it bigger. I use ssh -Y lab1 to log in to the remote machine. The environment has DISPLAY=localhost:14.0 (or some other number). Then I start Eclipse on the remote machine, with X tunneled to my PC. This appears to be related to 1.7, because after stopping the X server from 1.7 and starting the one from 1.5, the dialogs could be resized. This also affects e.g. the File Open and File Save dialogs of nedit (resizable under 1.5, fixed size under 1.7) Has anybody else experienced this ? Thanks for reporting this. This is actually a problem which has been reported and fixed before (see [1]), but it seems the patch to address this has gone AWOL in the update to Xserver 1.7 Fortunately, this is easier to fix the second time, patch to follow :-) [1] http://sourceware.org/ml/cygwin-xfree/2009-01/msg9.html -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
[PATCH] Cygwin/X: Make transient windows resizable again
Reverts the change which makes parented windows non-resizeable
Because this was trying to do something which we should be doing, as an
alternative we
try checking WM_NORMAL_HINTS for windows which shouldn't be resizable
If a window has a maximum size specified, no maximize box
If a window has a fixed size (max size == min size, per EWMH 1.3 Implementation
Notes), no resizing frame
Signed-off-by: Jon TURNEY jon.tur...@dronecode.org.uk
---
hw/xwin/winmultiwindowwm.c | 41 -
1 files changed, 36 insertions(+), 5 deletions(-)
diff --git a/hw/xwin/winmultiwindowwm.c b/hw/xwin/winmultiwindowwm.c
index ef94335..06fc599 100644
--- a/hw/xwin/winmultiwindowwm.c
+++ b/hw/xwin/winmultiwindowwm.c
@@ -1551,6 +1551,7 @@ winDeinitMultiWindowWM (void)
#define HINT_BORDER(1L1)
#define HINT_SIZEBOX (1l2)
#define HINT_CAPTION (1l3)
+#define HINT_NOMAXIMIZE (1L4)
/* These two are used on their own */
#define HINT_MAX (1L0)
#define HINT_MIN (1L1)
@@ -1632,7 +1633,30 @@ winApplyHints (Display *pDisplay, Window iWindow, HWND
hWnd, HWND *zstyle)
if (pAtom) XFree(pAtom);
}
- /* Apply Styles, overriding hint settings from above */
+ XSizeHints *normal_hint = XAllocSizeHints();
+ long supplied;
+ if (normal_hint (XGetWMNormalHints(pDisplay, iWindow, normal_hint,
supplied) == Success))
+{
+ if (normal_hint-flags PMaxSize)
+ {
+ /* Not maximizable if a maximum size is specified */
+ hint |= HINT_NOMAXIMIZE;
+
+ if (normal_hint-flags PMinSize)
+ {
+ /*
+If both minimum size and maximum size are specified and are
the same,
+don't bother with a resizing frame
+ */
+ if ((normal_hint-min_width == normal_hint-max_width)
+ (normal_hint-min_height == normal_hint-max_height))
+hint = (hint ~HINT_SIZEBOX);
+ }
+ }
+}
+ XFree(normal_hint);
+
+ /* Override hint settings from above with settings from config file */
style = winOverrideStyle((unsigned long)pWin);
if (style STYLE_TOPMOST) *zstyle = HWND_TOPMOST;
else if (style STYLE_MAXIMIZE) maxmin = (hint ~HINT_MIN) | HINT_MAX;
@@ -1649,14 +1673,21 @@ winApplyHints (Display *pDisplay, Window iWindow, HWND
hWnd, HWND *zstyle)
else if (style STYLE_NOFRAME)
hint = (hint ~HINT_BORDER ~HINT_CAPTION ~HINT_SIZEBOX) |
HINT_NOFRAME;
+ /* Now apply styles to window */
style = GetWindowLongPtr(hWnd, GWL_STYLE) ~WS_CAPTION ~WS_SIZEBOX; /*
Just in case */
if (!style) return;
- if (!hint) /* All on, but no resize of children is allowed */
-style = style | WS_CAPTION | (GetParent(hWnd) ? 0 : WS_SIZEBOX);
- else if (hint HINT_NOFRAME); /* All off, so do nothing */
+
+ if (!hint) /* All on */
+style = style | WS_CAPTION | WS_SIZEBOX;
+ else if (hint HINT_NOFRAME) /* All off */
+style = style ~WS_CAPTION ~WS_SIZEBOX;
else style = style | ((hint HINT_BORDER) ? WS_BORDER : 0) |
- ((hint HINT_SIZEBOX) ? (GetParent(hWnd) ? 0 : WS_SIZEBOX) :
0) |
+ ((hint HINT_SIZEBOX) ? WS_SIZEBOX : 0) |
((hint HINT_CAPTION) ? WS_CAPTION : 0);
+
+ if (hint HINT_NOMAXIMIZE)
+style = style ~WS_MAXIMIZEBOX;
+
SetWindowLongPtr (hWnd, GWL_STYLE, style);
}
--
1.6.4.2
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://x.cygwin.com/docs/
FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin 1.7 beta XWin Crashs from Ubuntun 9.10 after Firefox sessions
On 03/11/2009 23:34, ERIC HO wrote: Hi Jon, the crash is not caused by doing some specific action or looking at specific pages in firefox. Sometimes I sit on the Firefox screen and the cash happens. I ssh to the the remote Ubuntu Box and start up a gnome-session there back to my display. Over ssh, I can't even get gnome-session to start up properly, so I'm bit stymied trying to reproduce this. I then bring up the firefox on the Ubuntu box. After a few clicks, the cygwin Xserver died. I tried starting the X server with -multiwindow, but could not even get a complete GNOME screen back. You won't be able to get gnome-session to work with -multiwindow mode, as you are already running the integrated window manager. What I meant was, you should try starting the X server in -multiwindow mode, then ssh'ing to your Ubuntu host and just running firefox Are you sure that it is firefox that causes the X server crash? i.e. does the crash not happen if you use other applications or just idle in your gnome session? I've tried -query but did not get a screen back. It could be only ssh is allowed into the Ubuntu box. Yes, it seems that XDMCP is disabled by default in Ubuntu 9.10 and they seem to have neglected to provide a GUI to turn it on. A quick google should reveal how to do that, e.g. [1] [1] http://www.peppertop.com/blog/?p=690 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Running Java application with drag and drop support in cygwin
On 30/10/2009 09:06, Dees wrote: Your reply is much appreciated Jon. I will try to be more specific about the problem in further mails. On Thu, Oct 29, 2009 at 8:11 PM, Jon TURNEYjon.tur...@dronecode.org.uk wrote: On 28/10/2009 05:57, Dees wrote: I have developed a Java application involving jTree with extensive drag and drop support, which runs correctly in my Linux box. However, when I switch to a windows box and access the same Linux box using cygwin x-server, the drag and drop in jTree stops working. Interestingly, rest of the application still works fine. After analyzing a bit I found that x-server is able to recognize the drag event but fails to recognize a drop event. Details? OS : Suse Linux Enterprise Server 10 (i586) Version : 10 Patch level : 3 Other version information: Java : JDK 5 Cygwin setup-version: 2.573.2.3 Also tried using Xming 6.9.0.31 ssh same Linux setup from Windows, but that also doesn't solve the problem. Is there any setting, which should be done prior to running the Java swing applications? Here is a sample code which behaves in exactly same way. http://www.java2s.com/Code/Java/Swing-JFC/TreeDragandDrop.htm I have no idea how to use that java code to reproduce the problem you are seeing. Using the above java code in Linux: 1. Download and Install Java Development Toolkit on your Linux box (Java sun download site: http://java.sun.com/javase/downloads/index.jsp), if you do not have it already. 2. Save the sample code in the above link with the file name TreeTester.java, say in /home/user/ 3. Navigate to TreeTester.java from shell, and compile the java code: # cd /home/user/ # /usr/java/jdk1.5.0_14/bin/javac TreeTester.java Ignore any warnings of deprecated APIs. 4. This will create a few .class files in /home/user/ directory. Final step is to run the Java code, using: # /usr/java/jdk1.5.0_14/bin/java -classpath . TreeTester This will open up a GUI, with a jTree each on left and right pane. You can drag and drop any of the leaf nodes from one jTree to the root node of the other jTree and this should add a new node in the other jTree. You will get messages on console for the operations being performed. Now ssh the same box using cygwin/xming from any other windows box, and run the application using command in step 4. You should be able to drag (a small icon will come under cursor indicating that something is being dragged) but when you will drop it, the new node would not be added to the tree. Thats where lies my problem!!! Thanks for the test case and instructions, this makes it much easier for me to try it out. However, this testcase and your jar archive both work fine for me (using Xserver 1.7.1-3)! May be my problem is related to some setting. Though, not sure. Has anybody come across something similar? What should be done then? Please let me know. No it's probably a bug in Cygwin/X. But you're going to need to be a lot more specific about the problem before any progress can be made on fixing it. Also, putting some debug messages in the code lets me conclude that it's the drop event which is not being recognized, as the main control never reaches there. There is not really any drop event, as far as the X server is concerned, just mouse click and motion events, which are passed on to you application (which has a framework to interpret them as dragging and dropping an item). Now having a better idea of the problem, it seems less likely it is an Xserver bug at all. The only Xserver cause I can think of would be if it was somehow not sending the correct events to your applications window, which you could test using xev -id your applications window id (you can use xwininfo to find the window id) -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Automatically positioned mouse movements are off target
On 24/08/2009 23:05, Brian Sheppard wrote: I installed Cygwin/X (latest version, according to the Cygwin installer) on my laptop. I use Putty with X11 forwarding to connect to a Red Hat Linux system. I start the X windows server on my laptop using startxwin.bat as installed. After logging in to the Red Hat system, I execute gnome-session, and the gnome desktop shows on my Windows desktop. So far, so good. I am testing a Java application using a tool called Abbot. Abbot launches your Java Swing app within Abbot’s JVM. Abbot reads the coordinates of Swing components from internal Java objects, and then issues mouse and keystroke commands to simulate user actions according to a script. My observation is that the mouse clicks are off target. Specifically, Abbot is aiming to hit the exact center of each component, but it misses either high or low. (Left and right centering seem fine.) The amount of the miss has something to do with my Task Bar. When the Windows Task Bar is locked at top, then the clicks miss below (i.e., lower on the screen) the intended component. When the Task Bar is docked to the right, left, or bottom, or if it is at the top and set to auto-hide then mouse clicks miss above the intended component. Thanks for the problem report. I did some testing against Xserver 1.7.1 with xevent [1] and xdotool [2] to position the mouse pointer, but I wasn't able to reproduce the behaviour you are seeing. I really have no idea how to get started using abbot to reproduce the problem you see, a simple test case would help a lot here. [1] http://www.isv.uu.se/~ziemann/xevent/ [2] http://www.semicomplete.com/projects/xdotool/ -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: building issue
On 07/11/2009 01:10, Cesar Strauss wrote: Jon TURNEY wrote: I think I've finally found out why these Permission denied errors occur. It seems to be caused by the XWin.exe.manifest file we store in the same directory (it's not shipped as it's embedded into the final executable using the resource compiler) Presumably the manifest is incompatible with the libtool wrapper executable in some way and we get this unhelpful error message. Please try setting the execute bit (chmod +x) on the manifest file. See also: Libtool: Generated manifests need execute permission http://article.gmane.org/gmane.os.cygwin/110870 You are absolutely right, setting the execute bit on the manifest file fixes this. Thanks! -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: cygwin 1.7.0-63 problems with X programs
On 08/11/2009 14:11, Eliot Moss wrote: Corinna Vinschen wrote: On Nov 7 09:27, Eliot Moss wrote: Like another user, I had difficulty getting X to fire up. After setting LANG=en_US.UTF-8 I got farther, but these issues remain: Each xterm, xemacs, and xclock I start outputs this: Warning: Missing charsets in String to FontSet conversion I also get a series of: twm: warning: font for charset XXX is lacking. Where XXX is replaced by each of these in turn: JISX0208.1983-0 KSC5601.1987-0 GB2312.1980-0 Those three lines repeat 6 times. Now we are in a UTF-8 locale, it seems that Xlib tries rather excessively hard (see [1]) to find some CJK fonts, presumably just in case we try to output some characters they might provide. I think this also causes the noticably slower startup of applications, as they repeatedly search for fonts with these missing CJK encodings. The workaround is too install the font-isas-misc, font-jis-misc and font-daewoo-misc packages to provide fonts with these encodings. I'm not sure if there is a better solution than that. [1] http://cgit.freedesktop.org/xorg/lib/libX11/tree/modules/om/generic/omGeneric.c, line 845 Also: 1) the whole deal takes noticeably longer to start 2) xemacs does not receive its normal geometry from .Xdefaults If you could provide a specific example of a setting you are making in .Xdefaults and how you know it's not being used 3) each xterm started from .xinitrc has a blank line before the initial bash prompt, that wasn't present before Well spotted. In fact this seems to affect all Xterms, but I had completely failed to notice this. Are you absolutely sure it was updating to cygwin 1.7.0-63 which introduces this, and not an Xserver/Xlib/xterm update? I can reproduce that, but I can't make any sense of it. What on earth does that have to do with setting LANG to en_US.UTF-8?!? I'll redirect this to the cygwin-xfree list. Thanks for the redirect, Corinna. I can report a bit more ... - Adding the locale lines for C.UTF-8 into locale.dir and compose.dir do work to get X to start for me. (These are the patches discussed in the X11R7.5 and C.UTF-8 thread on cygwin-xfree.) - The other problems go away if I set LANG=C, so it's definitely something about the locale. Am I unusual in using twm :-) ? Yes, but not in a good way :-) Anyway, for now I am using LANG=C, set in the Windows environment variables, but once there are patches or updates to test, I am happy to try them ... -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Can't read lock file
On 11/11/2009 07:40, Fergus wrote: I have made no alterations/ patches/ amendments and have a completely current [1.7] system. For all of cygwin1.7.61.dll - cygwin1.7.64.dll the command run XWin fails with a reference to /vat/log/XWin.0.log which reads Can't read lock file /tmp/.X0.lock But when reverting to cygwin1.7.60.dll everything goes ahead smoothly and correctly and I can start a X terminal etc. Q1. Is it still the case that this problem is not well understood as in this reference? http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-cant-read-lock-file (My experiments with possible cause/ cure are confounded because on Machine 1 I have NTFS filesystems but lack administrator rights and on Machine 2 where I am Administrator, all filesystems are FAT32.) Perhaps this FAQ needs a bit of clarification. Creating the lockfile will fail if /tmp is on a FAT32 filesystem, for the reasons outlined in Corinna's email. (although I guess this may have behaved differently in some 1.7.x beta version if you haven't seen it there) For reasons which I don't currently understand, creating the lockfile also reportedly fails sometimes for people who have /tmp on a NTFS filesystem. If you can contribute some information which might help in solving this problem, please do so at [1] [1] http://sourceware.org/bugzilla/show_bug.cgi?id=9778 Q2. Is there anything obvious in the progression from cygwin1.7.60.dll to cygwin1.7.61.dll that might explain why the start command run XWin works with 1.7.60 and fails with 1.7.61? Finally Q3. Confession: I am completely confused by all recent posts about LANG, locale and all the rest. Are fiddling with LANG or applying patches such as described at http://cygwin.com/ml/cygwin-xfree/2009-11/msg00071.html fixes intended to address problems starting XWin? If you are not setting LANG (or are setting it to C.UTF-8) and have xorg-server-1.7.1-2, you will probably experience random crashes in multiwindow mode) Until there is a new libX11 release, if you are experiencing such crashes, applying that patch in that email OR setting LANG to something more sensible (e.g. en_GB.UTF-8) should resolve that problem. Thank you. Fergus PS Not quite finally. Can I please ask a 4th question: Q4 Why are questions about X specifically directed to a different mailing list? Apart from occasional high-frequency dialogue as at present, posts about X are (or seem to me to be) no more frequent than posts about grep or ls or chmod or ... . The main Cygwin list has a much higher readership and posts directed there might generate many useful hints, tips, experiences, fixes or even solutions? I am lazy and don't read the main list :-) -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Can't read lock file
On 11/11/2009 10:07, Corinna Vinschen wrote: On Nov 11 07:40, Fergus wrote: Q1. Is it still the case that this problem is not well understood as in this reference? http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-cant-read-lock-file Hardlinks don't work on FAT filesystems since FAT doesn't support them. Up to Cygwin 1.7.0-60, hardlinks on FAT were faked by copying the file. This has obvious downsides (unsecure, potential denial-of-service problems, portability), so we discussed and decided to remove this fake and to return an error instead when trying to create a hardlink on a FAT FS: http://cygwin.com/ml/cygwin/2009-09/msg00208.html (My experiments with possible cause/ cure are confounded because on Machine 1 I have NTFS filesystems but lack administrator rights and on Machine 2 where I am Administrator, all filesystems are FAT32.) Which is bad since FAT32 has no security at all. Any process of any user on the machine can overwrite any file, even in the Windows folder. NTFS is much more secure and has a couple of features you never get with FAT32, and hardlinks are only one minor advantage. You should really update the filesystem to NTFS using the on-board convert.exe tool. As for X, it should have a fallback method if the /tmp filesystem doesn't support hardlinks. Well, the fallback method is to use -nolock :-) The only use of the lock file appears to be to stop people from doing things like starting 'X :0 -nolisten inet -nolisten inet6' and 'X :0 -nolisten unix' at the same time, so I'm kind of tempted just to make '-nolock' the default and let people run with scissors if they want to :-) However, since there does seem to be some code to grovel over the mount options to try to detect if /tmp is textmode and warn about that for some reason, perhaps I'll look at checking if it's FAT and turning off the lockfile automatically in that case. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: 'run xterm' fails to open a window
On 10/11/2009 23:44, Mike Ayers wrote: On Behalf Of jean-luc malet Sent: Monday, November 09, 2009 2:50 AM no more result... gvim process is still started, appears in ps -a, but nothing is displayed on screen, replacing gvim by xterm still work with this solution... To find out why this is, run gvim from an xterm: [SNIP] m-ayers-lap gvim (gvim:1748): Pango-WARNING **: Error loading GSUB table 0x6EAD (gvim:1748): Pango-WARNING **: Error loading GPOS table 0x6EAD (gvim:1748): Pango-WARNING **: Error loading GSUB table 0x6EAD (gvim:1748): Pango-WARNING **: Error loading GPOS table 0x6EAD m-ayers-lap (gvim:1092): Pango-WARNING **: Error loading GPOS table 0x6EAD [/SNIP] Apparently, run.exe is not providing stdout/stderr to dump to. The workaround: [SNIP] C:\mikeC:\cygwin\bin\run.exe -p /usr/bin sh -c gvim/dev/null 21 C:\mike [/SNIP] Wait, are you saying that the process that run starts is blocked if it tries to output anything? Also, the suggestions that this has something to do with DISPLAY not being set correctly seem a little unlikely. If gvim or any other X application isn't able to determine which display to use, it should exit with an error, not hang around -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: checkX problems
On 30/10/2009 13:48, Ken Brown wrote: I'm having trouble with checkX. I haven't seen other people complain about this, so I assume it's something about my system, but I can't figure out what. There are two symptoms: 1. If I run checkX with a timeout, the timeout seems to be ignored. For example, with the X server *not* running: $ checkX -d 127.0.0.1:0.0 -t 100 --debug checkX.exe DEBUG: displayname : '127.0.0.1:0.0' checkX.exe DEBUG: opt_location: 0 checkX.exe DEBUG: opt_loglevel: 7 checkX.exe DEBUG: opt_nogui : 0 checkX.exe DEBUG: opt_notty : 0 checkX.exe DEBUG: opt_timeout : 100.00 checkX.exe DEBUG: (adjust_path) path is : /usr/local/texlive/2009/bin/i386-cygwin:/usr/local/bin:/usr/bin:/c/Program Files/ThinkPad/Utilities:/c/WINDOWS/system32:/c/WINDOWS:/c/WINDOWS/System32/Wbem:/c/Program Files/Intel/Wireless/Bin/:/c/Program Files/IBM ThinkVantage/Client Security Solution:/c/Program Files/ThinkPad/ConnectUtilities:/c/Program Files/QuickTime/QTSystem/:/c/Program Files/Common Files/Lenovo:/usr/lib/lapack:/usr/X11R6/bin:/usr/bin checkX.exe DEBUG: (find_X11_lib) DLL is /usr/bin/cygX11-6.dll checkX.exe DEBUG: (dlopen_X11_lib) /usr/bin/cygX11-6.dll dlopen'ed successfully. checkX.exe DEBUG: (load_X11_symbols) symbol XOpenDisplay loaded ok checkX.exe DEBUG: (load_X11_symbols) symbol XCloseDisplay loaded ok checkX.exe DEBUG: (try_with_timeout) Using delay of 100 secs, 0 nanosecs (100.00) checkX.exe DEBUG: (try_with_timeout) xserver search was unsuccessful checkX.exe Info: could not open X display '127.0.0.1:0.0' checkX.exe DEBUG: returning with status 1 checkX.exe Info: Exiting with status 1 The problem is that it returns within a second, in spite of the timeout. Or am I misunderstanding what the timeout is supposed to do? I think this is a misunderstanding here. Looking at the source, the timeout is the maximum time checkX will wait for an XOpenDisplay() to complete. If that fails immediately (e.g. due to with ECONNREFUSED), checkX will stop immediately. This is pretty reasonable. If there is nothing listening on the socket for the X server, that is not going to get better if we wait... ... except if the server happens to be starting up when we execute checkX. So, this is not quite what startxwin.bat requires, as the server may still be in the process of starting up. Fortunately, the X server binds it's socket pretty early in the startup, so this probably works pretty well, but in theory at least there is still a possible timing window in startxwin.bat. So it perhaps be useful if checkX retried the XOpenDisplay() periodically until the timeout was up (as xinit does) 2. If I start the X server by using the default startxwin.bat or startxwin.sh (both of which call checkX), the server is very unstable and crashes within a few minutes. This happens consistently, and it never happens if I comment out the line calling checkX. I think I was able to reproduce this problem (it is not how I normally start the X server) However, now I come back to look at this in detail, the problem no longer seems to exist. Are you still able to demonstrate it? I tried strace'ing checkX, but I don't know what to look for in the output. (I'll send it if it would be useful, but I don't want to spam the list otherwise.) I'm attaching cygcheck output. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Cygwin/x window no longer appears
On 12/11/2009 23:04, Mike Ayers wrote: From: cygwin-xfree-ow...@cygwin.com [mailto:cygwin-xfree- ow...@cygwin.com] On Behalf Of Olivia Cheronet Sent: Thursday, November 12, 2009 2:16 PM I have recently started to work with Cygwin/X. Until now, I have been starting Cygwin/X by using startxwin.bat in the Cygwin bash shell. Everything seemed to be working fine. However, it has now stopped working... When I type startxwin.bat in the Cygwin shell, the normal startxwin.bat - starting on Windows NT/2000/XP/2003 appears. Yet, the window which used to appear no longer does. I am really not too sure what to do about this, given that I have not (consciously!) modified anything. I have installed Cygwin (and Cygwin/x) very recently, and have tried to reinstall the latter using Cygwin's setup.exe, but to no effect. Try this: http://www.cygwin.com/ml/cygwin-xfree/2009-11/msg00071.html or set your LOCALE to en_US.UTF-8. You may need to install CJK fonts if things work but are sluggish. If you read the cygcheck output which Olivia kindly provided, you will see that she is running cygwin 1.5, with Xserver 1.5.3, so these recent breakages relating to Xserver 1.7.1 are unlikely to be the cause of her problems. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: checkX problems
On 12/11/2009 23:09, Lothar Brendel wrote: Jon TURNEY wrote: [...] Fortunately, the X server binds it's socket pretty early in the startup, so this probably works pretty well, but in theory at least there is still a possible timing window in startxwin.bat. Yep, and in my setup the X server *always* comes up too late. So it perhaps be useful if checkX retried the XOpenDisplay() periodically until the timeout was up (as xinit does) In principle, the script calling checkX could do that, because ``checkX'' has return status 1 if it couldn't connect. But as I already pointed out (http://www.mail-archive.com/cygwin-xfree@cygwin.com/msg19529.html) ``startxwin.bat'' uses ``run'' as a wrapper for ``checkX''. = Definitely no waiting and no passing on of the status of ``checkX'' to %errorlevel%, as ``run'' immediately goes background (unless called from an xterm, dunno why). But why would you fix the timeout problem by doing some horrible horrible iteration in the DOS batch script (horrible) (which would probably have to busy-wait (also horrible) as there's no sleep command), when you can fix checkX, which has the bonus of fixing other uses of it? Since more people seem to have this problem (cf. also Olivia's post), I repeat my question (essentially already posed by Ken Brown: http://www.mail-archive.com/cygwin-xfree@cygwin.com/msg19402.html): Why using ``run'' at all? If we really need a wrapper (do we?) wouldn't ``sh'' be a better one? I guess we use run for the reason run exists: to hide the console window, which otherwise would be shown? Perhaps if you look how startxwin.bat is started from the start menu shortcut, (as 'run startxwin.bat') you can see why this might be useful? To push this even further: Do we really need two *independent* scripts, ``starxwin.bat'' and ``starxwin.sh''? Why can't the former just delegate to the latter? Indeed. They are useless to me for starting the server and a continual source of problems. I would be quite happy to just delete them completely. :-) Looking forward to reading your patches to address any of these problems. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: [1.7] Cygwin/X crashes with latest server, cygwin
On 09/11/2009 20:48, Jim Reisert AD1C wrote: I updated Cygwin 1.7 with Yakov's fix the LANG problem. It's unstable, the first time I created an Xterm, the server crashed (seg fault): Thanks for testing this. Clearly I hadn't tested what happens if the X server has the change which is supposed to fix this, but the libX11 locale data hadn't been updated to include C.UTF-8. I've written a patch to avoid this crash if the X server is operating in a locale unsupported by libX11, but really the locale needs to be supported if we are going to get useful strings out of the X server for window titles and clipboard integration. Until there is a new libX11 package, you can update your locale data by downloading the locale.patch file from bugzilla [1] and applying it in /usr/share/X11/locale/ 'wget -O- http://sourceware.org/bugzilla/attachment.cgi?id=4340 | patch -d /usr/share/X11/locale/' should do the trick :-) [1] http://sourceware.org/bugzilla/show_bug.cgi?id=10870 2009-11-09 13:43:39 winClipboardProc - Warning: Locale not supported by X. 2009-11-09 13:43:39 winInitMultiWindowWM - XOpenDisplay () returned and successfully opened the display. 2009-11-09 13:43:39 winGetDisplay: DISPLAY=:0.0 2009-11-09 13:43:39 winClipboardProc - DISPLAY=:0.0 2009-11-09 13:43:39 winMultiWindowXMsgProc - XOpenDisplay () returned and successfully opened the display. 2009-11-09 13:43:39 winClipboardProc - XOpenDisplay () returned and successfully opened the display. 2009-11-09 13:43:49 Segmentation fault at address 0xb 2009-11-09 13:43:49 Fatal server error: 2009-11-09 13:43:49 Caught signal 11 (Segmentation fault). Server aborting 2009-11-09 13:43:49 -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: opengl in 1.7 on Win7 again
On 05/11/2009 11:32, Luke J. West wrote: I forgot to mention in my previous email that glxgears works fine - even though my sample opengl program bombs out like this... sh-3.2$ ./sample.bin freeglut (./sample.bin): ERROR: Internal errorVisual with necessary capabilities not found in function fgOpenWindow Segmentation fault (core dumped) Where's the first place you would check if you had this problem? Although you probably will not consider this helpful, the first place to look is: in your sample.c You haven't shown any evidence that the error message reported isn't correct i.e. your sample code is causing freeglut to ask for a visual which the server can't give it. *any* help appreciated. any suggestions as to where to look or what to check will be most helpful. Looking at the source code for fgOpenWindow and fgChooseVisual [1], the specification of glXChooseVisual, the set of visuals available as listed by glxinfo and how your sample code configures freeglut would probably be informative... [1] http://freeglut.svn.sourceforge.net/viewvc/freeglut/trunk/freeglut/freeglut/src/freeglut_window.c?view I've just installed Cygwin 1.7 (Win7 Hmoe Premium) and am trying to get started with OpenGL. I have a sample OpenGL program (sample.c) and I get the following. sh-3.2$ gcc -o sample.bin sample.c -lglut -lglu -lgl sh-3.2$ ./sample.bin freeglut (./sample.bin): ERROR: Internal errorVisual with necessary capabilities not found in function fgOpenWindow Segmentation fault (core dumped) sh-3.2$ glxinfo name of display: :0.0 display: :0 screen: 0 direct rendering: Yes server glx vendor string: Brian Paul ... 64 GLX Visuals visual x bf lv rg d st colorbuffer ax dp st accumbuffer ms cav id dep cl sp sz l ci b ro r g b a bf th cl r g b a ns b eat -- 0x21 24 tc 0 24 0 r y . 8 8 8 0 0 16 8 16 16 16 16 0 0 None 0xc2 24 tc 0 24 0 r y . 8 8 8 0 0 16 8 16 16 16 16 0 0 None 0xc3 24 tc 0 24 0 r y . 8 8 8 0 0 16 8 16 16 16 16 0 0 None 0xc4 24 tc 0 24 0 r y . 8 8 8 0 0 16 8 16 16 16 16 0 0 None -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: X11R7.5 and C.UTF-8
On 29/10/2009 20:20, Andy Koppe wrote: 2009/10/29 Jon TURNEY: I've put a patch in bugzilla [1] which can be applied to /usr/share/X11/locale to temporarily repair this problem. This needs to be looked at more deeply, though, as I'm not sure I've fully understood what that locale data is being used for, or specified C.UTF-8 correctly. [1] http://sourceware.org/bugzilla/show_bug.cgi?id=10870 I think the patch makes plenty of sense in mapping C.UTF-8 to en_US.UTF-8, because most other UTF-8 locales are also mapped to en_US.UTF-8, i.e. from X's perspective they're not actually language-specific. On second look, this patch doesn't seem to be quite right, as it makes the en_US.UTF-8 compose sequences available in C.UTF-8 (which is not the case in the C locale). More generally, there's the issue that Cygwin allows any combination of language and charset, whereas X has a fixed list of permitted combinations. Cygwin also supports many charsets that aren't supported by X (and vice versa). In particular, X only supports a few of the Windows/DOS codepages. But I guess unsupported locales will just have to be a case of don't do that? Yes. Treating XSupportsLocale() returning false as a fatal error as the Xserver currently does is wrong, I would say, unless the application has very specific requirement, though. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
