[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User hjs changed the following: What|Old value |New value Status|VERIFIED |CLOSED --- Additional comments from [EMAIL PROTECTED] Tue Jan 23 06:34:14 -0800 2007 --- seen in MWS - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User hjs changed the following: What|Old value |New value Status|RESOLVED |VERIFIED --- Additional comments from [EMAIL PROTECTED] Tue Jan 16 03:08:08 -0800 2007 --- looked at cygwin (testcase of cause all successfull :)) regarding process handling: dmake spawn version with latest of cygwin and dmake47 beside the warning dmake: Warning: -- Internal Warning: finished pid 1548 is not in pq!? (see comment from Dec 20 08:55:54 -0800 2006), i couldn't force dmake into anything bad. even breaking in real live (sc/source/core/date and sc/util) didn't take loneger than 5 seconds. so this seems to be the version to go with. one minor issue to care for in dmake 4.8 is that the output gets quite unreadable with -Pn as it is mixed on character base ;) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Jan 15 18:37:32 -0800 2007 --- Oops! Patch committed to dmake47. Thanks! - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=61856
--- Additional comments from [EMAIL PROTECTED] Mon Jan 15 09:37:02 -0800
2007 ---
i verified the initial testcase makefiles - and it looked like expected. also
checked windows native, which seems to need a small fix:
--- msdos/runargv.c 12 Nov 2006 05:06:39 - 1.3.2.1
+++ msdos/runargv.c 15 Jan 2007 16:27:34 -
@@ -142,7 +142,7 @@
int abort_flg;
int pid;
{
- return(1);
+ return(-1);
}
will have a look at the cygwin specifics later...
this means to me, it's almost verified ;)
-
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Tue Jan 2 14:41:29 -0800 2007 --- A small update with respect to spawn and parallel execution. The cygwin1-20070101.dll snapshot fixes the problem that spawnvp() behaves more like system(), i.e. waits for the completion of the spawned process instead of continuing, for native W32 programs, but the child is still "unbreakable" if CYGWIN=tty is set or from rxvt. I'll follow up on this in the cygwin ml. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Thu Dec 21 21:37:42 -0800 2006 --- Oops, committed. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=61856
--- Additional comments from [EMAIL PROTECTED] Thu Dec 21 10:21:45 -0800
2006 ---
unix/runargv.c is still too verbose to survive testcases.
--- runargv.c 21 Dec 2006 03:16:17 - 1.10.2.9
+++ runargv.c 21 Dec 2006 18:20:59 -
@@ -661,8 +661,7 @@
if( i == Max_proc ) {
Warning("Internal Warning: finished pid %d is not in pq!?", pid);
return;
- } else
-fprintf(stderr, "finishing pid %d\n", pid);
+ }
}
/* Not a running process anymore, the next runargv() will not use
-
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 20 19:17:13 -0800 2006 --- Committed a not-so verbose version of the last patch. Please verify. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 20 09:16:17 -0800 2006 --- fine with me. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 20 08:55:54 -0800 2006 --- Thanks for testing. I wouldn't be worried about the: dmake: Warning: -- Internal Warning: finished pid 1548 is not in pq!? warning. This has been like this forever but it was silently ignored before. It points to a potential problem that we wait for all children but we should probably only wait for childs of the current process group. (Something for later) This dmake: Error: -- Internal Error: Child is missing but still listed in _procs[x] 10: No child processes Temporary or .ERRREMOVE targets might not have been removed! we should slightly worry about but I guess it's a cygwin problem when native childs get killed (Especially from a cygwin.bat shell). The processes shouldn't vanish without a wait. *UNLESS* this was a recursive dmake and the first mentioned warning occured and we were treating a "rogue" process. How about I clean the patch from all the verbosity and commit it. If it also passes a few tinderbox builds from several different OSs we can finally close this CWS. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 20 01:14:05 -0800 2006 --- did an update of my cygwin to 1.5.23 and started another round of torturing dmake (spawn) containing the latest patch (iz61856_i.diff): the bad news is that i still got a couple of messages like those when breaking: -- dmake: Warning: -- Internal Warning: finished pid 1548 is not in pq!? --- dmake: Error: -- Internal Error: Child is missing but still listed in _procs[x] 10: No child processes --- dmake: Error: -- Internal Error: Child is missing but still listed in _procs[x] 10: No child processes Temporary or .ERRREMOVE targets might not have been removed! Quit(0) was called. Trying to quit ... -- ok, i really tried to force them with varying -Pn (2<=n<=30), but at least the last one makes me feel uncomfortable... the good new is that the bash error about signal handling (my comment from Wed Dec 6 03:28:43) and my privat hang in odk are gone. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Sat Dec 16 11:57:19 -0800 2006 --- @ause: I tested the previous patch with OOo and the short testcases. It should work but it includes a lot of debugging output. Give it a try, for me it significantly reduces the "lost pid" problem. I don't know how to solve the "unbreakability" problem. Native W32 programms are always unbreakable when the spawn dmake is used. I'll report this as a bug to the cygwin ml. The unbreakability in cmd.exe when tty is not set also looks like a cygwin problem but using rxvt (or ssh'ing into the machine) or setting CYGWIN=tty fixes this (It doesn't fully fix it, but it speeds up the processing for me.). This actually doesn't look like a new feature to me, I just tried a -P6 (fork) build with dmake 4.6 in sc/source/core/data and it really took a long time until my CTRL-c got processed. I guess we just didn't realize this problem before. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41468) No kill() on SIGINT | |/SIGQUIT - now working ;) --- Additional comments from [EMAIL PROTECTED] Sat Dec 16 11:42:10 -0800 2006 --- Created an attachment (id=41468) No kill() on SIGINT/SIGQUIT - now working ;) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41333) No kill() on SIGINT | |/SIGQUIT. --- Additional comments from [EMAIL PROTECTED] Mon Dec 11 06:37:11 -0800 2006 --- Created an attachment (id=41333) No kill() on SIGINT/SIGQUIT. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Dec 11 06:36:12 -0800 2006 --- Hi Ause, no idea (and no possible chance to fix) the unbreakability but I think I found the problem with the kill errors/closing shells, etc. CTRL-c sends a SIGINT to all processes in the process group and dmake (currently) sends a kill to all it's childs in its internal process queue. That is rather racy because a process (especially a native windows one) might be caught with a second kill while still closing. This is (should be) save on all sane *NIXs but on cygwin ... ;) The second problem is that after killing a process the PID might be used again and then get killed. Shouldn't happen, that's what wait() is for but ... ;) Can you try the following patch? - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Dec 11 01:19:22 -0800 2006 --- same settings, except that i needed -P4 today to get the "unbreakablility": - started from cygwin.bat takes several seconds to "see" the signal - starded from cygwin.bat with CYGWIN=tty or rxvt stops almost immediately sometimes breaking the build in a "cygwin.bat" window (only seen with CYGWIN=tty) also kills the rxvt window (not even started from that shell) ;) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Fri Dec 8 18:00:55 -0800 2006 --- Small correction: I can produce the unbreakability now. If I start cygwin.bat and use that shell CTRL-c is rather, aehh, unresponsive. It just takes a while but at some point the CTRL-c will arrive at dmake and then it will start terminating its childs. (This is with the CYGWIN environment variable set to "", i.e. empty. BUT ;) if you set CYGWIN=tty *before* starting bash in cygwin.bat it is breakable again. Somehow working in cygwin.bat produces a lot more missing/extra child messages and zombies than working from rxvt or an ssh login. (I never use that batch file, that might explain why I usually get less problems/no hangs.) On a positive side I got an Idea where the "famous" hangs might come from. If cygwin manages to leave a process behind that doesn't close or maybe even "zombifies" (You sometimes see this leftover processes) and no other process in the queue is left that finishes the wait() will just wait forever. (I managed to get into this state once.) I'll try replacing the wait() with a waitpid(-1, .. , NOHANG); usleep(10); plus some diagnostics to complain if we loose a process while waiting. ... @ause: Can you confirm that CYGWIN=tty (or using rxvt) cures the unbreakability? - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41284) Like previous patch | | plus debug info before an | |d after wait() --- Additional comments from [EMAIL PROTECTED] Fri Dec 8 16:45:31 -0800 2006 --- Created an attachment (id=41284) Like previous patch plus debug info before and after wait() - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Fri Dec 8 16:44:10 -0800 2006 --- Just for the record, the "unbreakability" of the dmake / spawn version seems to come from spawn being unbreakable for native w32 processes. (This is not new but already exists in dmake 4.6 and seems to be cygwin bug. I tried to reproduce the "unbreakability" with the dmake / fork/exec version but had no luck so far. It just breaks immediately, maybe 1sec delayed, but that might just be the OS. :( Here is what I did, please tell me what you're doing differently: Take the m196 milestone, just recently build. 1. build dmake (fork version) with patch iz61856_e.diff (follows) and copy it in the OOo build environment (overwrite solenv/wntmsci10/bin/dmake.exe) 2. Enter sc/source/core/data $ cd sc/source/core/data 3. Delete old output tree and build current dir with new dmake: $ rm -rf ../../../wntmsci10.pro; dmake -P2 4. Now press CTRL-c and watch the Caught SIGINT. Trying to quit ... appear inbetween the current output of the running processes. Repeat 1 to 4 with different wait times before pressing CTRL-c. ... I wasn't able to detect a problem with this. Note, this was done with an ssh connection (cygwin sshd) into the build machine. I will now repeat this experiment locally on that machine with a shell started with cygwin.bat and also with a rxvt terminal. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 6 10:26:44 -0800 2006 --- The last patch adds: @@ -527,10 +539,14 @@ } while( Wait_for_child(TRUE, -1) != -1 ); + + fprintf(stderr, "Finished cleaning old targets. Doing epilog now\n"); } } - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41174) One more line of de | |bugging output --- Additional comments from [EMAIL PROTECTED] Wed Dec 6 10:21:55 -0800 2006 --- Created an attachment (id=41174) One more line of debugging output - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41167) Same as before plus | | process debugging info --- Additional comments from [EMAIL PROTECTED] Wed Dec 6 07:34:17 -0800 2006 --- Created an attachment (id=41167) Same as before plus process debugging info - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 6 03:28:43 -0800 2006 --- correction: breaking is no problem with the sample makefile. in "real life" (sc/source/core/data) both dmake versions are hard to break. a single ctrl-c is sufficient to break but it looks to me (tried whith -P2 to get less output) that even after breaking a arbitrary number of new processes is still started. somtimes breaking ends up in this error after waiting a while for dmake to finish: 39 [main] bash 3192 sig_send: wait for sig_complete event failed, signal -41, rc 258, Win32 error 0 try to break early to reproduce the spawn problem with windows native binaries shouldn't hurt the build as we still have guw.exe ;) btw.: it would be handy to have -V tell you if your dmake version uses spawn - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Wed Dec 6 02:12:18 -0800 2006 --- my findings on a different machine (dual core) - will check on the previuos later: used latest patch... breaking is no problem at all, both spawn/nospawn, sleep/wsleep?! with the sample makefile, the spawn version waits for a1:wsleep to finish before a2:wsleep is started. note that the echo command in a2 is executed in parallel. nospawn version executes wsleep in parallel. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Tue Dec 5 15:52:56 -0800 2006 --- $%#$* Grmpf! I just tested with a cygwin dmake *without* --enable-spawn (using fork/exec) and guess what, it honors CTRL-c and it also nicely working with parallel native W32 processes. Can you confirm this? I'm going to prepare a short standalone testcase for the cygwin guys ... - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=61856
--- Additional comments from [EMAIL PROTECTED] Tue Dec 5 15:18:27 -0800
2006 ---
Well, it looks like we found a nice bug in cygwin.
I tried the following makefile first with SLEEP=sleep 10
--- makefile.mk ---
SHELL*:=/bin/sh
SHELLFLAGS*:=-ce
#SLEEP=wsleep.exe
SLEEP=sleep 10
all : a1 a2
@+echo all
a1 :
@+echo a1
@$(SLEEP)
a2 :
@+echo a2
@$(SLEEP)
--- makefile.mk ---
$ ./dmake/dmake.exe -P4 -r
and CTRL-c stopps it immediately.
But then I thought it might have to do this test with a native Windows program
and used this code snipped:
--- wsleep.c ---
// Compile with MinGW
#include
#include
int main(void)
{ printf("Waiting 3 sec ... ");
fflush (stdout);
Sleep(3000);
printf("done\n");
fflush (stdout);
return 0;
}
--- wsleep.c ---
and
$ gcc -mno-cygwin wsleep.c -o wsleep.exe
to create a native W32 executable that takes some time (you have to put it into
your path) and tried that instead.
Big surprise! It couldn't be stopped while wsleep was running!?
But it aborted all remaining processes after wsleep finished. One CTRL-c
is enough to abort dmake but it gets handled after the current W32 process is
finished.
For your real-life example that can feel like "unstoppable" especially if the
current process is a big cl.exe one. Linking?
Anyway, it gets better. ;) I rigged dmake to see where it's "waiting" and found
that the spawnvp() is actually behaving more like an system() for native windows
processes. - In other words when spawn is used ( Default for cygwin builds ;) )
there is not much parallel execution :(
Did your dmake was build with --enable-spawn or not?
-
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Tue Dec 5 07:42:19 -0800 2006 --- with the latest patch dmake -P6 is still almost unstoppable. at the end i get this message "dmake: Warning: -- Internal Warning: finished pid is not in pq!?" - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41128) Maybe this works .. | | --- Additional comments from [EMAIL PROTECTED] Mon Dec 4 18:50:28 -0800 2006 --- Created an attachment (id=41128) Maybe this works .. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Dec 4 11:19:52 -0800 2006 --- > sometimes the last patch causes dmake to be almost "unbreakable", especially > with -Pn (tried in "sc/source/core/data" on windows/bash) Sigh! That's what you get for making dmake continue after a break signal. But you should have gotten something like: Wait was interrupted foo: bar - trying to continue ... I could rig something together that signals dmakes internal process handling to abort (i.e. call Quit() ) when the first CTRL-c is encountered and ignore all breaks after that or alternatively just abort everything once the 2nd break is encountered. Any preferences? - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Dec 4 08:34:22 -0800 2006 --- sometimes the last patch causes dmake to be almost "unbreakable", especially with -Pn (tried in "sc/source/core/data" on windows/bash) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Sun Dec 3 21:25:57 -0800 2006 --- @ause: The previous patch (uncommitted) might improve dmake's process handling when it is aborted with CTRL-c. If you have a few cycles you might want to try it. It passes the testsuite and I'm going to test it on OOo once the pch build is done. (If it helps IMHO it could still go into dmake47.) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41099) Tentative patch to | |improve dmake's behavior w | |hen stopped with CTRL-c --- Additional comments from [EMAIL PROTECTED] Sun Dec 3 21:20:56 -0800 2006 --- Created an attachment (id=41099) Tentative patch to improve dmake's behavior when stopped with CTRL-c - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Status|REOPENED |RESOLVED Resolution| |FIXED --- Additional comments from [EMAIL PROTECTED] Thu Nov 30 17:18:00 -0800 2006 --- Committed. @ause: Please try again ;) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41030) Patch for dmake --- Additional comments from [EMAIL PROTECTED] Thu Nov 30 17:11:19 -0800 2006 --- Created an attachment (id=41030) Patch for dmake - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Status|RESOLVED |REOPENED Resolution|FIXED | --- Additional comments from [EMAIL PROTECTED] Thu Nov 30 10:35:12 -0800 2006 --- The previous patch should fix the "Internal Error: Child is missing .." problem. (untested) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |41023) Fix "Internal Error | |: Child is missing .." pro | |blem --- Additional comments from [EMAIL PROTECTED] Thu Nov 30 10:33:43 -0800 2006 --- Created an attachment (id=41023) Fix "Internal Error: Child is missing .." problem - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Thu Nov 30 09:21:15 -0800 2006 --- latest patch seems to have some issues. easy one: run dmake with no makefile around. gives lot of internal errors about missing childs here. sometimes, of cause unreproduceable, something that sounds like a cygwin exceptionhandler gives a message about stack corruption: 3 [main] dmake 1640 _cygtls::handle_exceptions: Error while dumping state (probably corrupted stack) - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Nov 27 14:25:24 -0800 2006 --- Thanks! Committed after doing a testbuild (MAXPROCESS=2) with this patch on W32-bash. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Mon Nov 27 06:55:44 -0800 2006 --- up to now i just did some (single process) testbuilds on various platforms an ran the testcases. didn't yet start to look at the individual issues. i think it's ok to add this issue to CWS dmake47. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |40931) Patch for dmake/uni | |x/runargv.c --- Additional comments from [EMAIL PROTECTED] Sun Nov 26 20:48:52 -0800 2006 --- Created an attachment (id=40931) Patch for dmake/unix/runargv.c - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Priority|P4|P3 --- Additional comments from [EMAIL PROTECTED] Sun Nov 26 20:47:30 -0800 2006 --- The following patch avoids the (unlikely) event that the dmake process with the pid PID is processed and removed from dmakes (internal) process queue but a new (non dmake) process is created and subsequently caught from the wait() in Wait_for_child(). @ause: I think it's low risk, did you already start QA or is it OK to commit? - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Assigned to|vq|hjs --- Additional comments from [EMAIL PROTECTED] Fri Nov 17 15:11:06 -0800 2006 --- Please verify. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Status|NEW |RESOLVED Resolution| |FIXED --- Additional comments from [EMAIL PROTECTED] Fri Nov 17 15:10:21 -0800 2006 --- Committed the patch and a testcase to dmake47. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |40663) Better "wait for co | |mpletion" implementation --- Additional comments from [EMAIL PROTECTED] Fri Nov 17 13:37:44 -0800 2006 --- Created an attachment (id=40663) Better "wait for completion" implementation - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Thu Nov 16 12:47:25 -0800 2006 --- The following section is mainly meant for myself to explain why the previous patch is not the solution for the problem. First of all the intend of Wait_for_completion=TRUE is not waiting for a specific PID but to wait for all commands of a process array entry to finish. The proposed solution to only keep Wait_for_completion TRUE in Wait_for_child() if we received the PID we were waiting for is bad because the following can happen if other process arrays are active: Wait_for_child() calls _finished_child() (I) without Wait_for_completion=TRUE and _finished_child() calls runargv() with the next process array entry *and* runargv() -> _add_child() starts the command and returns. If now _proc_cnt==Max_proc _finished_child calls Wait_for_child(,-1) and can catch the PID we are waiting for in the first place, it calls _finished_child() (II) -> runargv() -> _add_child() and returns. Now _finished_child() (I) returns to the Wait_for_child() that is still waiting for PID to finish and in the next loop waitpid(PID,...) finds that the PID is no longer active and returns to the caller of Wait_for_child(, PID) (The -1 return code is ignored) even though there are still followup commands in the process array entry. This obviously can lead to inconsistencies. Short summary: Mixing global variables and parallel processing is not good. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 User vq changed the following: What|Old value |New value Attachment is patch| |Created an attachment (id= | |38664) Updated patch --- Additional comments from [EMAIL PROTECTED] Sun Aug 20 21:41:10 -0700 2006 --- Created an attachment (id=38664) Updated patch - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[tools-issues] [Issue 61856] $(shell ...) waits not onl y for its own target
To comment on the following update, log in, then open the issue: http://www.openoffice.org/issues/show_bug.cgi?id=61856 --- Additional comments from [EMAIL PROTECTED] Sun Aug 20 21:40:26 -0700 2006 --- If Wait_for_completion is TRUE (this can only happen for ruleop '!', building .INCLUDED makefiles or $(shell ..) function macros) all parallel execution is effectively disabled, see Wait_for_completion handling in _add_child. Even if one of those targets executes another Wait_for_completion==TRUE command the Wait_for_child() (recursively called) will make sure the recipe is ready before continuing (but finishing previously started childs will be handled) An updated patch follows. - Please do not reply to this automatically generated notification from Issue Tracker. Please log onto the website and enter your comments. http://qa.openoffice.org/issue_handling/project_issues.html#notification - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
