Re: patch cygrunsrv for running native win32 applications
On Fri, Oct 14, 2005 at 12:24:44PM +0100, Henry S. Thompson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Christopher Faylor writes: LOL. Propose a gold star for Antony Baxter - I believe he is the first person on record as getting a laugh out of CGF! It's been known to happen from time to time: On Thu, Apr 14, 2005 at 10:05:58AM -0400, Christopher Faylor wrote: On Thu, Apr 14, 2005 at 06:22:06AM -0700, Yitzchak Scott-Thoennes wrote: I recently ran across this (not work safe) similar request: http://www.scs.cs.nyu.edu/~dm/remove.pdf I very nearly died while reading that link. I started laughing so hard that I aspirated some tea and then I had to cough for a minute. That was very funny. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SETUP: In-use files have been replaced
On Wed, Oct 19, 2005 at 08:15:14PM +0100, Chris Taylor wrote: Herb Martin wrote: Eric Blake wrote: Herb Martin wrote: So what is the method to teach Setup that the file has been updated. Have you tried simply uninstalling the Cygwin package? If you installed the new one into another location, you presumably don't need or want the other one. For most packages at least, SETUP doesn't automatically try to update it if you haven't installed it. I didn't install Exim 4.54 into another location; someone else mentioned an alternate locationa and I (perhaps incorrectly) mentioned that I had downloaded and compiled it FROM another location. The make install was run normally and the specially compiled (make options) is in the default (/usr/bin) location. All I wish to do is make Setup aware of this if it is possible. For now, I must (carefully) ensure that setup doesn't overwrite my good version with the default. If you reinstalled all of exim, you don't really need the cygwin version.. So you want to edit the /etc/setup/installed.db and give it an artificially high number, say 99.999, as the installed version of exim. This will stop cygwin from ever overwriting your installation of exim (unless the version ever gets higher than that.. unlikely in our lifetimes to be honest) Are you sure? I didn't think setup actually compared versions at all. There's been discussion on cygwin-apps of updating packages to a lower version number that is actually a neweer version, and people didn't seem to think there would be any problem with that... -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: When are Windows Short Path Names required
On Sun, Oct 23, 2005 at 05:04:23PM +0100, zzapper wrote: On Sat, 22 Oct 2005 20:46:04 +0100, zzapper wrote: Hi $ cygpath -d c:/Program\ Files/Internet\ Explorer c:\PROGRA~1\INTERN~1 I've just had another case where the I had to use the short form, is the short form the real name? No, it's an alternate name, and may not even exist, as Brian said. I think I've may be mislead you, I think my problem was more that when theres an interaction between *nix scripts environment. Sometimes even quoting doesn't seem to carry the correct path info thru to the final WinApp , so you have to move the app to non-space path location or use the short name above. Now I don't know what your actual question was. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: please test: coreutils-5.90-3
On Wed, Oct 19, 2005 at 10:37:08PM +0200, Angelo Graziosi wrote: With the snap 20051019 13:12:47 the command 'cp -p' of coreutils 5.90-3 works fine. A curiosit. The recent snapshots contain also mingw-runtime and w32api: is there a special reason for this? Um, they changed? snapshots don't just have the cygwin package; they often have others. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: SETUP: In-use files have been replaced
On Mon, Oct 24, 2005 at 12:34:30AM +0100, Chris Taylor wrote: Yitzchak Scott-Thoennes wrote: On Wed, Oct 19, 2005 at 08:15:14PM +0100, Chris Taylor wrote: If you reinstalled all of exim, you don't really need the cygwin version.. So you want to edit the /etc/setup/installed.db and give it an artificially high number, say 99.999, as the installed version of exim. This will stop cygwin from ever overwriting your installation of exim (unless the version ever gets higher than that.. unlikely in our lifetimes to be honest) Are you sure? I didn't think setup actually compared versions at all. There's been discussion on cygwin-apps of updating packages to a lower version number that is actually a neweer version, and people didn't seem to think there would be any problem with that... Well, I find that Eric tends to be pretty reliable, and I have to say that I cannot think of a logical reason why setup *wouldn't* use this method. It makes no sense not to - you'd have to add yet another set of information to setup for it to know if a version was newer. I sincerely doubt that the original setup developers obfuscated it *that* much. The version number is the best way of doing things.. If things change, generally the package name changes slightly as well, and you just make the new one replace the old... Ah, but versions are strings, not numbers, and how to compare them isn't always obvious. (NB: I'm not 100% sure w/o referring to earlier messages that it was Eric that originally posted this (in this thread), but I'm pretty sure it was.) Yes, it was: On Wed, Oct 19, 2005 at 06:38:21AM -0600, Eric Blake wrote: According to Herb Martin on 10/18/2005 5:53 PM: So what is the method to teach Setup that the file has been updated. Why does setup need to be taught? However, you may be looking for /etc/setup/installed.db; edit that for the package in question to tell setup.exe that the installed version has the same version number (or greater) than what setup.exe can offer from the mirrors. But I don't think this will work; as an example, there was recently a libIDL package that had a version 0.8.3 but a new version 0.5.17 was released to replace it; setup correctly removed the 0.8.3 and installed the 0.5.17. Also, note that setup in curr mode will always try to upgrade exp packages back to the curr level. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: textmount/binmount issue
On Mon, Oct 24, 2005 at 11:05:13AM -0400, Igor Pechtchanski wrote: On Fri, 21 Oct 2005, Lemke, Michael IZ/HZA-IOM wrote: This is my current mount table: pc00626 mount -m mount -f -s -b c:/MyStuff/cygwin/bin /usr/bin mount -f -s -b c:/MyStuff/cygwin/lib /usr/lib mount -f -s -b c:/MyStuff/cygwin / mount -f -s -t -E u: /u mount -s -b --change-cygdrive-prefix / Note that /u is mounted textmode while the rest is binary. However, if I access files on /u they are binmode as well. But when I change the mode of the drive prefix to text then /u is also text. Seems like prefix is overwriting the explicit setting for /u. If the drive prefix is something else but /u is still there then /u is not affected by the prefix settings. Is this intended behavior? Looks like the cygdrive prefix takes precedence over explicit mounts. This is arguably a bug. http://cygwin.com/acronyms/#PTC. Corinna or Chris, care to comment? My name doesn't start with C, but I will comment that if the precedence went the other way, a mount like: mount c:/foo /cygdrive/d would make a windows path d:\bar have no posix translation. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: textmount/binmount issue
On Mon, Oct 24, 2005 at 05:56:36PM -0400, Igor Pechtchanski wrote: On Mon, 24 Oct 2005, Yitzchak Scott-Thoennes wrote: On Mon, Oct 24, 2005 at 11:05:13AM -0400, Igor Pechtchanski wrote: Looks like the cygdrive prefix takes precedence over explicit mounts. This is arguably a bug. http://cygwin.com/acronyms/#PTC. Corinna or Chris, care to comment? My name doesn't start with C, but I will comment that if the precedence went the other way, a mount like: mount c:/foo /cygdrive/d would make a windows path d:\bar have no posix translation. How is this different from mount c:/baz /somemount mount c:/foo /somemount/d (except for the fact that /cygdrive/d is an automount)? I'm actually not sure what's meant by posix translation in this context... For example, assume a windows PATH=d:\bar. What does a cygwin program convert PATH to at startup? There would be no equivlent, since /cygdrive/d/bar would mean c:/foo/bar, not d:/bar. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Accessing signal context from a handler
On Wed, Oct 26, 2005 at 05:56:15PM -0400, Christopher Faylor wrote: On Wed, Oct 26, 2005 at 09:36:53PM +, Adam Irawan wrote: Igor Pechtchanski pechtcha at cs.nyu.edu writes: On Mon, 15 Jul 2002, Christopher Faylor wrote: On Mon, Jul 15, 2002 at 01:00:23PM -0400, Igor Pechtchanski wrote: Does Cygwin allow accessing the context (register state, etc) of the faulted instruction from a signal handler installed by sigaction(), similar to ucontext/mcontext in Linux? [snip] It's not available. cgf [snip snip] Basically, I'm trying to port an application that intercepts SIGSEGV and makes decisions based on the IP of the faulted instruction. The application currently uses the sigaction/ucontext mechanism in Linux. Any suggestions on how to do this in Cygwin (preferably similar to the Linux code, to maximize reuse) would be appreciated. [snipped the code bit] Wow, dug up an ancient post today that seems to relate to my current problem more than anything else I've searched for. Just wondering, has there been changes regarding this topic at all? Thanks alot, It's not available. cgf You left out the bit where ucontext used to not be passed at all, but in current snapshots it's passed as a null pointer... -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mksh (shell)
On Fri, Oct 28, 2005 at 07:01:59PM +, Thorsten Glaser wrote: Sorry for the people, please do XXX but my ressorces are limited, my responsibilities as BSD developer high, and I only have one Windows box which is already occupied with Interix. Having no Interix myself, I can't see what problem that might cause; putting cygwin's /bin first in your path should allow you to build under cygwin. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygcheck improvements
On Wed, Nov 02, 2005 at 01:23:39PM -0500, Christopher Faylor wrote: Here's a scary one: How about a cygcheck --report option which queries for a from email address and sends the report (as an attachment) to the cygwin mailing list? Sounds like a separate program to me. As flea is to mutt, so hippo is to cygwin? Then we could say, cygwin doesn't have bugs, it's got mammals. Oh, and: On Thu, Jul 07, 2005 at 08:29:31PM -0700, Yitzchak Scott-Thoennes wrote: On Wed, Jul 06, 2005 at 11:27:15AM -0400, Larry Hall wrote: I said virus program. That would be something like Norton AV, McAfee VirusScan, etc. Hmm, is there any way cygcheck could report on intrusive software like virus scanners and firewalls? Though my knowlege on the subject is close to nil I think the latter may be doable with WSCEnumProtocols. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Missing directories in @INC with perl 5.8.7 and cygwin 1.5.18-1
On Fri, Nov 04, 2005 at 08:30:01AM +0100, Frank_Ehlert wrote: If PERL5LIB is set, perl searches for according version directories and architecture specific directories. If they are present they will be included in the @INC array. Now I am faced with the problem that with perl 5.8.7 this does not work anymore! export PERL5LIB=//swece/sweperl/module/CPAN/lib/perl5/site_perl As a workaround, do (untested): export PERL5LIB=//swece/sweperl/module/CPAN/lib/perl5/site_perl PERL5LIB=$PERL5LIB:$PERL5LIB/5.8/cygwin:$PERL5LIB/5.8:$PERL5LIB/cygwin perl -V ... @INC: //swece/sweperl/module/CPAN/lib/perl5/site_perl /usr/lib/perl5/5.8/cygwin /usr/lib/perl5/5.8 ... With perl 5.8.2 the output looks like perl -V ... @INC: //swece/sweperl/module/CPAN/lib/perl5/site_perl/5.8.2 /cygwin-thread-multi-64int //swece/sweperl/module/CPAN/lib/perl5/site_perl/5.8.2 //swece/sweperl/module/CPAN/lib/perl5/site_perl /usr/lib/perl5/5.8.2/cygwin-thread-multi-64int /usr/lib/perl5/5.8.2 ... The perl code to add directories from Easiest workaround to make this work in general would be to build perl with inc_version_list=5.8/cygwin 5.8 cygwin in Policy.sh. Gerrit, do you want to release a new perl package with this set? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Missing directories in @INC with perl 5.8.7 and cygwin 1.5.18-1
On Sat, Nov 05, 2005 at 04:43:03AM +0100, Gerrit P. Haase wrote:
[EMAIL PROTECTED] wrote:
Hi,
If PERL5LIB is set, perl searches for according version directories and
architecture specific directories.
If they are present they will be included in the @INC array.
Now I am faced with the problem that with perl 5.8.7 this does not work
anymore!
There is a new feature in perl, if there is a the script
$Config{siteperl}/sitecustomize.pl present it will be executed. See
http://search.cpan.org/~nwclark/perl-5.8.7/pod/perlrun.pod#%2Df how to
disable it or how it may help you until an update of perl with a
workaround is released, from perlrun.pod:
-f
Disable executing $Config{siteperl}/sitecustomize.pl at startup.
Perl can be built so that it by default will try to execute
$Config{siteperl}/sitecustomize.pl at startup. This is a hook that
allows the sysadmin to customize how perl behaves. It can for instance
be used to add entries to the @INC array to make perl find modules in
non-standard locations.
Maybe this feature is also the reason that it behaves different from
previous releases.
No, it's just a problem with the cygwin release. S_incpush in perl.c
only looks for the full archname (cygwin-thread-multi-64int) and
revision.patchlevel.subversion (5.8.7) as subdirectories, but we are
setting things to make makemaker install modules using just 5.8 and
cygwin, even when a PREFIX is specified (not sure if this applies to
INSTALL_BASE as well).
Some changes to Configure are needed to make this cleaner.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
futimes fails on file open for read
The following gives:
futimes returned: -1 (Permission denied)
I'm told it succeeds on linux; is it even possible for it to work on
cygwin or am I SOL?
#include stdio.h
#include stdlib.h
#include sys/time.h
#include fcntl.h
#include errno.h
#include string.h
int main(int argc,char **argv)
{
int fd, ret;
struct timeval tv[2] = {{5, 0}, {5, 0}};
if (system(touch foo) == -1) return 1;
if ((fd = open(foo, O_RDONLY)) == -1) return 1;
errno = 0;
ret = futimes(fd, tv);
printf(futimes returned: %d (%s)\n, ret, strerror(errno));
return 0;
}
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: futimes fails on file open for read
On Mon, Nov 07, 2005 at 06:08:40PM +0100, Corinna Vinschen wrote: On Nov 6 23:34, Yitzchak Scott-Thoennes wrote: The following gives: futimes returned: -1 (Permission denied) I'm told it succeeds on linux; is it even possible for it to work on cygwin or am I SOL? I've created a patch for this. It will show up in the next snapshot. All is well with 20051108, thanks. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
unexpected failure of cygwin32_winpid_to_pid
Using 20051108, I just saw failures twice in tests of the perl binding for cygwin32_winpid_to_pid, but no failure on every try since. The test looks like: my $catpid = open my $cat, |cat or die Couldn't cat: $!; open my $ps, ps| or die Couldn't do ps: $!; my ($catwinpid) = map /^.\s+$catpid\s+\d+\s+\d+\s+(\d+)/, $ps; close($ps); is(Cygwin::winpid_to_pid($catwinpid), $catpid, winpid to pid); is(Cygwin::pid_to_winpid($catpid), $catwinpid, pid to winpid); close($cat); in perl, where Cygwin::winpid_to_pid internally just calls cygwin32_winpid_to_pid. I'm not able to think of any way in which system load or a race could cause this failure. Any suggestions? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: unexpected failure of cygwin32_winpid_to_pid
On Tue, Nov 08, 2005 at 04:17:41PM -0800, Yitzchak Scott-Thoennes wrote: Using 20051108, I just saw failures twice in tests of the perl binding for cygwin32_winpid_to_pid, but no failure on every try since. The test looks like: my $catpid = open my $cat, |cat or die Couldn't cat: $!; open my $ps, ps| or die Couldn't do ps: $!; my ($catwinpid) = map /^.\s+$catpid\s+\d+\s+\d+\s+(\d+)/, $ps; close($ps); is(Cygwin::winpid_to_pid($catwinpid), $catpid, winpid to pid); is(Cygwin::pid_to_winpid($catpid), $catwinpid, pid to winpid); close($cat); in perl, where Cygwin::winpid_to_pid internally just calls cygwin32_winpid_to_pid. I'm not able to think of any way in which system load or a race could cause this failure. Any suggestions? Hmm, I thought I had ruled out exec() of cat failing, but now I don't think so; probably that's what it is, so please disregard this message. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: [ANNOUNCEMENT] Updated: apr(-util) 0.9.7-1
On Tue, Nov 15, 2005 at 03:51:10PM -0500, Christopher Faylor wrote: On Tue, Nov 15, 2005 at 07:37:53PM +, zzapper wrote: On Mon, 14 Nov 2005 21:01:43 +, wrote: zzapper wrote: On Sun, 13 Nov 2005 16:16:58 +, wrote: APR and APR-util, portability and miscellaneous utility libraries, have been updated in the Cygwin net distribution to version 0.9.7-1. Gosh this was hard to Google I think it's sdesc: Additional utility library for use with the Apache Portable Runtime (development/documentation package) ldesc: The mission of the Apache Portable Runtime (APR) project is to create and maintain software libraries that provide a predictable and consistent interface to underlying platform-specific implementations. I have no idea what you are trying to say, zzapper. I was just trying to find some doc on APR, as I didn't even know what it was. Please don't bother the mailing list with your lack of knowledge about things like this. To clarify: zzapper has often responded to announcements that don't describe the package with a short description s/he has found by searching. Are you asking that s/he no longer do this? Or not do it when the only description s/he finds is from the setup hint? Or other? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: upgrade/add a single package
On Wed, Nov 16, 2005 at 02:40:27PM -0800, Brian Dessent wrote: beau wrote: I think I've got my /etc/setup properly in place Huh? Normally there is no need to touch anything under /etc/setup at all. , but when I try to add a new package I end up getting everything. The only way I can see to stop this currently is to manually set each package to keep in the package choosing window. Put the radio button at the top to 'keep', then select whichever individual package you want. BTW, this doesn't work (last I tried) if you want an experimental version of a package. Should it? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
open() giving ENOENT when trying to create files with control chars
I'm trying to create a file (on NTFS) with a CR in the name and getting
ENOENT; is it possible for this to work (without a managed mount)?
Trying all possible characters:
#include stdio.h
#include fcntl.h
#include unistd.h
#include string.h
#include errno.h
#include ctype.h
int main(int argc, char **argv)
{
int fd, c, err;
char filename[4] = foo;
char buf[12];
for (c = 1; c = 255; ++c) {
filename[1] = c;
fd = open (filename, O_WRONLY|O_CREAT|O_EXCL, 0640);
if (fd 0) {
err = errno;
printf (%02x%s: %s\n, c,
(iscntrl (c) ? : (sprintf (buf, (%c), c), buf)),
strerror (err));
} else {
close (fd);
if (unlink (filename))
fprintf (stderr, %x: unlink failed: %s\n, c, strerror (errno));
}
}
return 0;
}
shows all control characters and just a few others not working (obviously
/ and : are out):
01: No such file or directory
02: No such file or directory
03: No such file or directory
04: No such file or directory
05: No such file or directory
06: No such file or directory
07: No such file or directory
08: No such file or directory
09: No such file or directory
0a: No such file or directory
0b: No such file or directory
0c: No such file or directory
0d: No such file or directory
0e: No such file or directory
0f: No such file or directory
10: No such file or directory
11: No such file or directory
12: No such file or directory
13: No such file or directory
14: No such file or directory
15: No such file or directory
16: No such file or directory
17: No such file or directory
18: No such file or directory
19: No such file or directory
1a: No such file or directory
1b: No such file or directory
1c: No such file or directory
1d: No such file or directory
1e: No such file or directory
1f: No such file or directory
22 (): No such file or directory
2a (*): No such file or directory
2f (/): No such file or directory
3a (:): No such file or directory
3c (): No such file or directory
3e (): No such file or directory
3f (?): No such file or directory
5c (\): No such file or directory
7c (|): No such file or directory
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: ECANCELED
On Fri, Dec 02, 2005 at 05:09:03PM -0500, Christopher Faylor wrote: On Fri, Dec 02, 2005 at 10:38:27PM +0100, Samuel Thibault wrote: There is no ECANCELED value in cygwin's /usr/include/sys/errno.h. Is that on purpose? Of course it is. It's about time you noticed. We refused to implement it just to irritate you. OK, then you wouldn't be interested in a patch implementing /usr/include/aio.h and all the functions therein? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: open() giving ENOENT when trying to create files with control chars
On Fri, Dec 02, 2005 at 11:09:05PM +0100, Corinna Vinschen wrote: On Dec 2 20:38, Eric Blake wrote: I'm trying to create a file (on NTFS) with a CR in the name and getting ENOENT; is it possible for this to work (without a managed mount)? Welcome to Windows. None of these non-portable characters are supported in filenames except in managed mounts. Just to push the point a bit, note the words non-portable. I did do a bit of looking around before posting, and saw the *?/\|: characters mentioned in several places, but not the 0-0x1f characters. Now, I've looked more and see those officially verboten as well. (Though I also saw a page that documents how 8.3 filenames are formed and seems to say that the *?/\|: characters *are* possible in NTFS filenames.) http://support.microsoft.com/default.aspx?scid=kb;en-us;117258 is interesting... -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: command-line archiver that understands Windows ACLs?
On Tue, Dec 06, 2005 at 01:21:54PM -0800, Mark McWiggins wrote: I'm working with a company that deploys tar files from Unix to Windows where the Windows administrators set their directories up with various ACLs. Cygwin's default TAR program destroys these ACLs when files are unpacked into these directories. What do you mean destroys? tar changes the ACLs on the directories? Or just creates files that don't inherit their directory's ACLs? If the latter, does setting CYGWIN=nontsec help? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygcheck bug when listing services (Was Re: Parallel writes to a single FIFO do not queue, and deadlock cygwin)
On Mon, Dec 05, 2005 at 09:40:01AM -0500, Igor Pechtchanski wrote:
Running cygcheck under strace shows that after listing all the available
services, it invokes cygrunsrv --query grunsrv.exe --list, which results
in the above message. I think this may be because the output of
cygrunsrv --list doesn't contain a trailing '\n', and so strtok gets
confused. The following patch (against a slightly older cygcheck) fixes
it for me, but I haven't had the time to test it extensively:
Index: cygcheck.cc
===
RCS file: /cvs/src/src/winsup/utils/cygcheck.cc,v
retrieving revision 1.77
diff -u -p -r1.77 cygcheck.cc
--- cygcheck.cc 17 Aug 2005 00:52:43 - 1.77
+++ cygcheck.cc 5 Dec 2005 14:37:58 -
@@ -950,8 +950,8 @@ dump_sysinfo_services ()
else
{
/* read the output of --list, and then run --query for each service */
- size_t nchars = fread ((void *) buf, 1, sizeof (buf) - 1, f);
- buf[nchars] = 0;
+ size_t nchars = fread ((void *) buf, 1, sizeof (buf) - 2, f);
+ buf[nchars] = buf[nchars+1] = 0;
That needs a comment in the code.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: open() giving ENOENT when trying to create files with control chars
On Fri, Dec 02, 2005 at 11:09:05PM +0100, Corinna Vinschen wrote:
On Dec 2 20:38, Eric Blake wrote:
I'm trying to create a file (on NTFS) with a CR in the name and getting
ENOENT; is it possible for this to work (without a managed mount)?
Welcome to Windows. None of these non-portable characters are
supported in filenames except in managed mounts.
Just to push the point a bit, note the words non-portable.
Moving on to another non-portable problem, I want to create a file
with a space at the end of the name, but cygwin is stripping spaces.
Despite the comment in the code, this does seem to be allowed (though
I suspect it may be via NtCreateFile only, since windows commands
don't seem to handle filenames with spaces at the end well). I tried
this:
--- path.cc.orig2005-10-24 03:33:48.532065000 -0700
+++ path.cc 2005-12-08 11:39:23.237267200 -0800
@@ -482,7 +482,7 @@ path_conv::set_normalized_path (const ch
if (strip_tail)
{
- while (*--p == '.' || *p == ' ')
+ while (*--p == '.') // || *p == ' ')
continue;
*++p = '\0';
}
@@ -957,7 +957,7 @@ out:
/* Windows ignores trailing dots and spaces */
char *tail = NULL;
for (char *p = path; *p; p++)
- if (*p != '.' *p != ' ')
+ if (*p != '.') // *p != ' ')
tail = NULL;
else if (p[1] == '\\')
{
__END__
and with limited testing on XP Pro SP1 didn't find any problems. Is
it possible this space-stripping predates use of NtCreateFile? Are
there problems likely to arise with this (e.g. cygwin calling other
winapi functions that are passed filenames not correctly handling
spaces at the end)?
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Hang with 20051205 snapshot while building OOo
On Wed, Dec 07, 2005 at 01:15:50PM -0500, Christopher Faylor wrote: On Wed, Dec 07, 2005 at 05:16:17PM +0100, Volker Quetschke wrote: We got some new features in the hangs with the 20051205 snapshot when building OOo. This trace looks a little bit different from the old versions, see thread around http://sourceware.org/ml/cygwin/2005-10/msg00600.html, but it still hangs. It looks different because I took out all of the debugging strace output that I added to try to track this down. I'm convinced that the hang is coming from timeGetTime. That's a Windows routine and there's not much I can do if a windows routine decides to take a long time to return. So, in other words, I'm admitting defeat in this case, especially since this isn't a regression from 1.5.18. This is likely to enable a fix or workaround in cygwin, but is it possible that this is restricted to certain brands, models, or revisions of processor? Am I correct in remembering that it happens only on some machines, Volker? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Call for testing Cygwin snapshot
On Wed, Nov 30, 2005 at 06:11:37PM +0100, Corinna Vinschen wrote: 1.5.19 has also some new functionality over 1.5.18. If you have sources which take advantage of that stuff if present, we're also interested in getting feeedback about those: - clock_getres, clock_setres. - fts(3) functions (BSD). - futimes. - getline, getdelim. - memmem. - mlock,munlock. - mmap(..., MAP_NORESERVE) for anonymous maps. - pread, pwrite. - readdir_r. - strptime's 'c and 'Z' formats. - timelocal, timegm. perl (bleadperl) now finds and uses clock_getres, futimes, and readdir_r and passes the relevant tests. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Hang with 20051205 snapshot while building OOo
On Thu, Dec 08, 2005 at 03:23:01PM -0800, Yitzchak Scott-Thoennes wrote: On Wed, Dec 07, 2005 at 01:15:50PM -0500, Christopher Faylor wrote: On Wed, Dec 07, 2005 at 05:16:17PM +0100, Volker Quetschke wrote: We got some new features in the hangs with the 20051205 snapshot when building OOo. This trace looks a little bit different from the old versions, see thread around http://sourceware.org/ml/cygwin/2005-10/msg00600.html, but it still hangs. It looks different because I took out all of the debugging strace output that I added to try to track this down. I'm convinced that the hang is coming from timeGetTime. That's a Windows routine and there's not much I can do if a windows routine decides to take a long time to return. So, in other words, I'm admitting defeat in this case, especially since this isn't a regression from 1.5.18. This is likely to enable a fix or workaround in cygwin, but is it ^un ! possible that this is restricted to certain brands, models, or revisions of processor? Am I correct in remembering that it happens only on some machines, Volker? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: open() giving ENOENT when trying to create files with control chars
On Fri, Dec 09, 2005 at 12:31:32AM +, Eric Blake wrote: From: Yitzchak Scott-Thoennes [EMAIL PROTECTED] Moving on to another non-portable problem, I want to create a file with a space at the end of the name, but cygwin is stripping spaces. Despite the comment in the code, this does seem to be allowed (though I suspect it may be via NtCreateFile only, since windows commands don't seem to handle filenames with spaces at the end well). I tried this: Windows strips trailing spaces and dots (unless the file name consists only of spaces). You need a managed mount to preserve those; otherwise foo , foo., foo. . . . , foo, and a bunch of other spellings all refer to the same file. I attempted to indicate in the message above that I tried it and succeeded in using filenames with spaces on the end (and *different* files named the same except without the spaces). It seems this is *not* an across-the-board Windows limitation. Unlike in the other case (non-portable characters, which POSIX allows an implementation to reject), the stripping of trailing '.' from filenames is a violation of POSIX, but since Windows is the culprit, cygwin cannot avoid it (except with the overhead of managed mounts). In the case of ., I'm not sure we would want it allowed it even if Windows made it possible; too backward incompatible with those cases where a filename is specified with a . to indicate no default extension be added (e.g. gcc foo.c -o foo.). -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
status of bash-3.0-12 (was: Re: How do I make /bin/sh=sh)
On Mon, Aug 22, 2005 at 08:40:06PM +, Eric Blake wrote: Actually, I'm playing with a change to bash, soon to be bash-3.0-12, where the postinstall script will leave /bin/sh alone if its timestamp is newer than /bin/bash. For one release. What happens after the next upgrade to bash? My plan for bash-3.0-12 and beyond is to only upgrade /bin/sh to the newest bash version if /bin/sh has an older timestamp than /bin/bash, and is not ksh or zsh. So, using 'touch -d +2 years /bin/sh.exe' would exempt /bin/sh from updates for the next two years, no matter how often bash upgrades occur in the meantime, and no matter if /bin/sh is ash because you wanted it that way (at the expense of having a file modified 2 years in the future! Isn't time travel fun? :) Eric, I see about a week after the above, you put out an experimental bash-3.0-12. I don't see any other announcement of it; is the above the only difference in it? Should it still be experimental? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Patch and Cygwin
On Fri, Dec 16, 2005 at 01:07:30AM +0100, Lennart Borgman wrote: This is a summary of my tests with Cygwin patch and different line ending styles. (Only LF and CRLF are tested here.) I have downloaded patch 2.5.9 from ftp://alpha.gnu.org/gnu/diffutils/patch-2.5.9.tar.gz and compiled it using Cygwin. I have used this patch and the patch that currently comes with Cygwin to do some tests of patch and diff when the files to compare and patch have different line ending styles. I have also made this test using MSYS and GnuWin32 utilities. The result is that the only option that seems to be able to handle the mix of line endings is Cygwin using DOS line endings with patch 2.5.9. (The only thing that did not work was preservation of line endings in the patched file, but that seems to be a small problem here.) I'm not sure the line you are pursuing is going to be a productive one. While automatically working despite CR-mismatches would be good, I'm concerned that you'll break other things (patches on files with varying line endings, patches that are intended to change line endings). Ideally, patch should exactly follow what diff does, and diff does treat CRLF vs LF differences as significant, even in a textmount. The exception is when specifically requested by the --strip-trailing-cr option, which makes diff entirely CRLF/LF-blind. Implementing this option for patch would be much better than trying to do anything automatic. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Perl/TK cygz.dll mapping error
On Sun, Dec 18, 2005 at 05:03:40PM -0800, Brian Dessent wrote: Brett Serkez wrote: From what I can tell by searching for similar issues, it *looks* like the new library may need to be rebased during setup? A packaging change? No, it means you need to install the rebase package and run rebaseall yourself. Or the package containing cygz.dll needs to be rebuilt (by the maintainer) with --enable-auto-image-base. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Linker error after clean reinstallation of Cygwin
On Fri, Dec 16, 2005 at 07:03:12PM -0500, Igor Pechtchanski wrote: On Fri, 16 Dec 2005, Angelo Graziosi wrote: After I reinstalled Cygwin there is the following linker error in build some applications: -- g77 ... foo.F -lcygipc /bin///ld: cannot find -lcygipc ... -- The cygipc package has not been reinstalled, it, now, belongs to obsolte section. Should I reinstall it or is sufficient to delete '-lcygipc' from the build command? Actually, cygipc *is* installed, but the new version is empty (as are all obsolete packages) Doesn't seem to be true in this case: setup.ini: setup-timestamp: 1134667219 ... @ cygipc sdesc: IPC support for cygwin ldesc: CygIPC provides a library, headers, and a daemon that can be used to provide IPC services for cygwin. It is, however, slated to become obsolete once the cygserver is complete (cygserver is a similar library and daemon that will be a built-in part of the main cygwin DLL). category: _obsolete requires: cygwin libpopt0 version: 2.03-2 install: release/_obsolete/cygipc/cygipc-2.03-2.tar.bz2 374271 38e9b514fd9c0946a8d019d8a4c9b031 source: release/_obsolete/cygipc/cygipc-2.03-2-src.tar.bz2 81505 71ff16929a3128a1c9a63c657deb380b [prev] version: 2.02-1 install: release/_obsolete/cygipc/cygipc-2.02-1.tar.bz2 358773 6340582cd657fb3beabef20ee47c5f78 source: release/_obsolete/cygipc/cygipc-2.02-1-src.tar.bz2 75792 d08c0fa695264c881fd6c7fe16a10226 $ cygcheck -c cygipc Cygwin Package Information Package VersionSta cygipc 2.03-2 OK $ cygcheck -l cygipc /etc/postinstall/cygipc.sh /etc/preremove/cygipc.sh /usr/bin/cygcygipc-2.dll /usr/bin/ipc-daemon2.exe /usr/bin/ipcrmcygipc.exe /usr/bin/ipcscygipc.exe /usr/bin/ipctestcygipc.exe /usr/bin/msgtoolcygipc.exe /usr/bin/semstatcygipc.exe /usr/bin/semtoolcygipc.exe /usr/bin/shmtoolcygipc.exe /usr/include/cygipc/sys/ipc.h /usr/include/cygipc/sys/ipctrace.h /usr/include/cygipc/sys/msg.h /usr/include/cygipc/sys/sem.h /usr/include/cygipc/sys/shm.h /usr/lib/libcygipc.a /usr/lib/libcygipc.dll.a /usr/share/doc/cygipc-2.03/BRANCHES /usr/share/doc/cygipc-2.03/contents_motif.gif /usr/share/doc/cygipc-2.03/COPYING /usr/share/doc/cygipc-2.03/index.html /usr/share/doc/cygipc-2.03/NEWS /usr/share/doc/cygipc-2.03/next_motif.gif /usr/share/doc/cygipc-2.03/node21.html /usr/share/doc/cygipc-2.03/node22.html /usr/share/doc/cygipc-2.03/node23.html /usr/share/doc/cygipc-2.03/node24.html /usr/share/doc/cygipc-2.03/node25.html /usr/share/doc/cygipc-2.03/node26.html /usr/share/doc/cygipc-2.03/node27.html /usr/share/doc/cygipc-2.03/node28.html /usr/share/doc/cygipc-2.03/node29.html /usr/share/doc/cygipc-2.03/node30.html /usr/share/doc/cygipc-2.03/node31.html /usr/share/doc/cygipc-2.03/node32.html /usr/share/doc/cygipc-2.03/node33.html /usr/share/doc/cygipc-2.03/node34.html /usr/share/doc/cygipc-2.03/node35.html /usr/share/doc/cygipc-2.03/node36.html /usr/share/doc/cygipc-2.03/node37.html /usr/share/doc/cygipc-2.03/node38.html /usr/share/doc/cygipc-2.03/node39.html /usr/share/doc/cygipc-2.03/node40.html /usr/share/doc/cygipc-2.03/node41.html /usr/share/doc/cygipc-2.03/node42.html /usr/share/doc/cygipc-2.03/node43.html /usr/share/doc/cygipc-2.03/node44.html /usr/share/doc/cygipc-2.03/node45.html /usr/share/doc/cygipc-2.03/node46.html /usr/share/doc/cygipc-2.03/node47.html /usr/share/doc/cygipc-2.03/node48.html /usr/share/doc/cygipc-2.03/node49.html /usr/share/doc/cygipc-2.03/node50.html /usr/share/doc/cygipc-2.03/node51.html /usr/share/doc/cygipc-2.03/node52.html /usr/share/doc/cygipc-2.03/node53.html /usr/share/doc/cygipc-2.03/node54.html /usr/share/doc/cygipc-2.03/node55.html /usr/share/doc/cygipc-2.03/node56.html /usr/share/doc/cygipc-2.03/node57.html /usr/share/doc/cygipc-2.03/node58.html /usr/share/doc/cygipc-2.03/node59.html /usr/share/doc/cygipc-2.03/node60.html /usr/share/doc/cygipc-2.03/node61.html /usr/share/doc/cygipc-2.03/node62.html /usr/share/doc/cygipc-2.03/node63.html /usr/share/doc/cygipc-2.03/node64.html /usr/share/doc/cygipc-2.03/node65.html /usr/share/doc/cygipc-2.03/node66.html /usr/share/doc/cygipc-2.03/node67.html /usr/share/doc/cygipc-2.03/node68.html /usr/share/doc/cygipc-2.03/node69.html /usr/share/doc/cygipc-2.03/node70.html /usr/share/doc/cygipc-2.03/node71.html /usr/share/doc/cygipc-2.03/node72.html /usr/share/doc/cygipc-2.03/node73.html /usr/share/doc/cygipc-2.03/node74.html /usr/share/doc/cygipc-2.03/node75.html /usr/share/doc/cygipc-2.03/node76.html /usr/share/doc/cygipc-2.03/node77.html /usr/share/doc/cygipc-2.03/node78.html /usr/share/doc/cygipc-2.03/node79.html /usr/share/doc/cygipc-2.03/node80.html /usr/share/doc/cygipc-2.03/node81.html /usr/share/doc/cygipc-2.03/previous_motif.gif /usr/share/doc/cygipc-2.03/README /usr/share/doc/cygipc-2.03/README.old /usr/share/doc/cygipc-2.03/TODO /usr/share/doc/cygipc-2.03/up_motif.gif /usr/share/doc/cygipc-2.03/USAGE
Re: error on installing new packages for the third-party cygwin
On Mon, Dec 19, 2005 at 07:55:04PM +0800, Liu Haibin wrote: Hi, I'm using a product which integrates cygwin. I wanted to install a new package into this cygwin but after installation, I got an error message when I ran the newly installed package. The procedure entry point __getreent could not be located in the dynamic link library cygwin1.dll I think the package you installed is expecting a newer version of the cygwin1.dll than your 3PP provided. Run setup and update only the cygwin package (making sure you can restore back to the way things were in case the third-party stuff you are using is for some reason forward-incompatible (which it shouldn't be)). -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Inconvenient ghostscript and transfig dependences
On Tue, Dec 20, 2005 at 07:24:56PM -0500, Larry Hall (Cygwin) wrote: Sorry but there is currently no way to represent either/or dependencies via setup.exe (PTC). That said, there's no reason that you can't override setup.exe and not install gs-no-X11. If you know that's what you want, then you should feel free to do so. If your gripe is that setup.exe will try to install gs-no-X11 each time you run it, you should feel free to manually edit /etc/setup/install.db to include the package name and an impossibly high version number to fool setup.exe into thinking you already have a version installed that is more current than what it has available to it. This keeps getting advocated, but in my testing it does *not* work. If you have found otherwise, or see somewhere in the source for setup where this is supposedly implemented, please give some detailed information. But AFAICT, setup seems to have no concept of higher/lower version numbers, nor do I see how setup *could* understand the variety of version numbers out there. Which is higher, 20030307 or 1.875? 2.5.4a or 2.5.4? 2.8.0 or 2.10.1? These all involve assumptions that may or may not be true. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
perl Bundle::Cygwin / perl-bundle-cygwin package
I'm working on creating a bundle of common Perl modules that build and pass all significant tests on cygwin. I hope to have it accepted as a cygwin package. Any particular module requests or comment on my proposed names for the CPAN bundle or cygwin package? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: perl Bundle::Cygwin / perl-bundle-cygwin package
On Thu, Dec 22, 2005 at 01:14:06PM -0600, Yaakov S (Cygwin Ports) wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Yitzchak Scott-Thoennes wrote: I'm working on creating a bundle of common Perl modules that build and pass all significant tests on cygwin. I hope to have it accepted as a cygwin package. I think it's preferable to make separate packages for each module. My reasoning: 1) this is the precedent set by Linux distributions; 2) bumping one module doesn't require rolling a whole bundle; 3) separate modules minimizes unnecessary dependencies; 4) I'm sure there's something else I'm forgetting. IOW, I do NOT like this idea. If, OTOH, I do believe that more perl modules should go into the distro, without packaging the entire CPAN, certainly: 1) modules which don't build OOTB (e.g. Tk, gtk2-perl bindings, etc.); 2) modules which are prerequisites for other packages (e.g. ExtUtils::PkgConfig, necessary for building gtk2-perl bindings). The same would apply, of course, to python and ruby. You'll see I already have a large selection on Cygwin Ports, although not all of those are candidates for the distro. Large distributions like POE or the DateTime:: modules should have packages of their own. I was thinking of smaller modules that it really would make no sense to have one package per CPAN distribution for, particularly common dependencies of other modules. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Please try a snapshot - final push for 1.5.19
On Wed, Dec 21, 2005 at 11:22:49AM -0500, Christopher Faylor wrote: Only four more testing days remain if we want to get a cygwin 1.5.19 release out there by Christmas. Please try a snapshot and report success/failure to this thread. When reporting a problem please report the exact steps that you've taken and *please* narrow down the failure to the absolute minimum scenario needed to duplicate it. So, please try the 2005-Dec-21 snapshot: http://cygwin.com/snapshots/ . I've just loaded 20051222, and noticed this (unknown if it is a regression): starting bash in a console window and repeatedly pressing Ctrl-C gets me: 3093118 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093125 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093150 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093157 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093181 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093189 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093214 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093221 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093246 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093254 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093278 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093286 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093311 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093318 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093343 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 3093350 [sig] bash 832 _cygtls::handle_threadlist_exception: called with threadlist_ix -1 (I am using the experimental bash-3.0-12.) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: perl Bundle::Cygwin / perl-bundle-cygwin package
On Thu, Dec 22, 2005 at 01:51:02PM -0800, Yitzchak Scott-Thoennes wrote: On Thu, Dec 22, 2005 at 01:14:06PM -0600, Yaakov S (Cygwin Ports) wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Yitzchak Scott-Thoennes wrote: I'm working on creating a bundle of common Perl modules that build and pass all significant tests on cygwin. I hope to have it accepted as a cygwin package. I think it's preferable to make separate packages for each module. My reasoning: 1) this is the precedent set by Linux distributions; 2) bumping one module doesn't require rolling a whole bundle; 3) separate modules minimizes unnecessary dependencies; 4) I'm sure there's something else I'm forgetting. IOW, I do NOT like this idea. If, OTOH, I do believe that more perl modules should go into the distro, without packaging the entire CPAN, certainly: 1) modules which don't build OOTB (e.g. Tk, gtk2-perl bindings, etc.); 2) modules which are prerequisites for other packages (e.g. ExtUtils::PkgConfig, necessary for building gtk2-perl bindings). The same would apply, of course, to python and ruby. You'll see I already have a large selection on Cygwin Ports, although not all of those are candidates for the distro. Large distributions like POE or the DateTime:: modules should have packages of their own. I was thinking of smaller modules that it really would make no sense to have one package per CPAN distribution for, particularly common dependencies of other modules. To further explain, suppose we had a package of all the DateTime:: modules (which are broken up into many distributions, but I hope you would agree would belong together). They have a number of small general purpose dependencies; putting those in the same package doesn't make sense to me, nor does packaging each individually. I would lump them all together, along with the few modules we don't have packaged yet but that come bundled with ActivePerl (for competitive purposes :) and maybe some of the modules that will be bundled with perl in 5.10. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: base-files-mketc.sh hangs
On Tue, Dec 27, 2005 at 12:01:15AM +0800, Liu Haibin wrote: Hi, I got a problem after I reinstalled cygwin. /etc/postinstall/base-files-mketc.sh hangs there everytime I install packages. And it will produce some log files like setup.log.postinstallXa00156 under /var/log. I tried to run /etc/postinstall/base-files-mketc.sh in the shell and it ran ok. Any idea what went wrong? (Just in case you didn't know this,) if you are happy with things as they are and just want setup to not continue running it, rename it to /etc/postinstall/base-files.mketc.sh.done. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installation without network connection
On Tue, Dec 27, 2005 at 11:43:02AM -0600, Greg Youngdahl wrote: So, can this still be accomplished? Is there some web page or other document that explains how it should be done? If not - can someone help me, and perhaps I can put together such a document? Perhaps it is trivially easy, and I'm making a mountain out of a molehill? Maybe if I could just find a mirror site and download a few things I'd be good to go? If so, what would be a minimum set of files to download (setup.exe, the cygwin.dll and enough packages to be able to have bash run 'find something | cpio -pd...'). Untried, but should work: Run setup.exe in Download without Installing mode, on a system with no cygwin (workarounds exist if not possible), toggle the View to Full, and select findutils and cpio packages (if they aren't by default). (You may need to resize the window to see the package-name column.) Then Next through to the end. Burn your Local Package Directory and everything under it to a CD, along with a copy of setup.exe, and run it on your target machine. Secondly, the system of interest (my WinXP box) already has cygwin installed (it has been there a while, and I no longer recall how I got it there), but it is a 1.3 version of the cygwin.dll, and it doesn't have cpio (it seems to have pretty much everything else I'd need), so can I just upgrade that, or should I uninstall it and start again from scratch? Perhaps all I really need to do is grab a cpio package and install that. But... The current cpio is extremely likely to have been built with 1.5.x and not be usable on 1.3.x. The perplexing thing about this is that there is a setup.exe (somewhere under C:\Windows\system32) that pops up a little window to tell me I should use the control panel to do upgrades when I run it from bash (typing setup.exe to a bash prompt). However there is nothing I can see in the start-control panel for cygwin, nor anything under add-remove programs associated with cygwin. So, I'm not really even sure if that setup.exe is the one associated with cygwin. Perhaps the method I used to install it way back when (potentially 3-4 years ago) did not involve the new modern techniques? Cygwin's setup.exe isn't automatically installed anywhere. You can run it directly by pointing your browser to http://cygwin.com/setup.exe. If you want it saved somewhere, you need to do that manually. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installation without network connection
On Tue, Dec 27, 2005 at 10:51:54AM -0800, Yitzchak Scott-Thoennes wrote: On Tue, Dec 27, 2005 at 11:43:02AM -0600, Greg Youngdahl wrote: So, can this still be accomplished? Is there some web page or other document that explains how it should be done? If not - can someone help me, and perhaps I can put together such a document? Perhaps it is trivially easy, and I'm making a mountain out of a molehill? Maybe if I could just find a mirror site and download a few things I'd be good to go? If so, what would be a minimum set of files to download (setup.exe, the cygwin.dll and enough packages to be able to have bash run 'find something | cpio -pd...'). Untried, but should work: Run setup.exe in Download without Installing mode, on a system with no cygwin (workarounds exist if not possible), toggle the View to Full, and select findutils and cpio packages (if they aren't by default). (You may need to resize the window to see the package-name column.) Then Next through to the end. Burn your Local Package Directory and everything under it to a CD, along with a copy of setup.exe, and run it on your target machine. ... in Install from local directory mode, again selecting findutils and cpio if they aren't by default. Secondly, the system of interest (my WinXP box) already has cygwin installed (it has been there a while, and I no longer recall how I got it there), but it is a 1.3 version of the cygwin.dll, and it doesn't have cpio (it seems to have pretty much everything else I'd need), so can I just upgrade that, or should I uninstall it and start again from scratch? Perhaps all I really need to do is grab a cpio package and install that. But... The current cpio is extremely likely to have been built with 1.5.x and not be usable on 1.3.x. The perplexing thing about this is that there is a setup.exe (somewhere under C:\Windows\system32) that pops up a little window to tell me I should use the control panel to do upgrades when I run it from bash (typing setup.exe to a bash prompt). However there is nothing I can see in the start-control panel for cygwin, nor anything under add-remove programs associated with cygwin. So, I'm not really even sure if that setup.exe is the one associated with cygwin. Perhaps the method I used to install it way back when (potentially 3-4 years ago) did not involve the new modern techniques? Cygwin's setup.exe isn't automatically installed anywhere. You can run it directly by pointing your browser to http://cygwin.com/setup.exe. If you want it saved somewhere, you need to do that manually. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Cygwin Setup Command Line
On Wed, Dec 28, 2005 at 04:23:33PM -0800, Bruce Dobrin wrote: Hmm, I just tried it and it does appear to sort of work, but it only installs the default and I don't see any way to tell it to do otherwise. This has been discussed many times. In short, set up a dummy package in setup.iniwith category Base and listing all non-default packages that you want as prerequisites. ( plus the -q flag still doesn't' suppress the you must reboot window) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Request for fixes to http://cygwin.com/problems.html
On Thu, Dec 29, 2005 at 04:08:07PM -0500, Christopher Faylor wrote: On Thu, Dec 29, 2005 at 12:01:19PM -0800, Rick Moen wrote: Greetings. I'd appreciate it if the http://cygwin.com/ webmaster would make two fixes to the http://cygwin.com/problems.html page, at his/her earliest convenience: 1. Fix your author attribution for smart questions (i.e., the essay How to Ask Questions the Smart Way) to credit _both_ of that essay's co-authors, instead of just Eric. I've removed the attribution entirely. I'll let people figure out who you are by visiting the page. 2. Append a prominent notice saying something like Please do NOT send your Cygwin questions to Eric and Rick. I've added some comments mentioning how stupid such an activity would be. Now it says says: Once you've followed that link and read read the advice... -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Hope this topic isn't taboo yet
On Fri, Dec 30, 2005 at 03:44:08PM -0500, Frier, David wrote: So if I have a two-physical-processor system which HT fools Windoze into thinking is a quad, then turning off HT is not really going to help me? Help with what, the cpu spikes? You could try it and see. You could also see if the spikes exist without cygwin installed at all... -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: [ANNOUNCEMENT] Updated [experimental]: bash-3.1-1
On Sat, Dec 31, 2005 at 07:41:12AM -0700, Eric Blake wrote: To use this release, you MUST install a recent snapshot of cygwin (20051222 or later), as well as use the experimental readline-5.1-1. And presumably the experimental libreadline6-5.1-1? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
snapshot ChangeLog link missing info
http://cygwin.com/snapshots/winsup-changelog-20051229-20060101 is missing the actual winsup/cygwin changes: ChangeLog for 20051229 to 20060101 -- winsup/cygwin/ChangeLog -- 2006-01-01 Christopher Faylor [EMAIL PROTECTED] -- winsup/doc/ChangeLog -- 2005-12-30 Christopher Faylor [EMAIL PROTECTED] * relnotes.texinfo: Remove entry about hyperthreaded processor. 2005-12-30 Christopher Faylor [EMAIL PROTECTED] * faq-problems.html: Remove entry about hyperthreaded processor. -- winsup/testsuite/ChangeLog -- 2006-01-01 Christopher Faylor [EMAIL PROTECTED] * winsup.api/resethand.c: Use SIGSEGV for the signal to test. 2006-01-01 Christopher Faylor [EMAIL PROTECTED] * winsup.api/resethand.c: New file. -- winsup/w32api/ChangeLog -- 2005-12-31 Corinna Vinschen [EMAIL PROTECTED] * include/winsock2.h: Don't define struct sockaddr_storage when building Cygwin. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
services not starting with 20060104 snapshot
I'll try to send a proper bug report in a few hours, but just want to mention now that after loading the 20060104 dll, trying to start cygserver or syslogd fails. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Start sshd server on reinstallation
On Sat, Jan 07, 2006 at 07:09:14PM -0800, Brian Dessent wrote: Ordinary Olson wrote: I recently installed cygwin to my C drive ? however after realizing its size, I decided to move it to the D drive. So I stopped the sshd server that I had running, deleted the registry keys under `HKEY_LOCAL_MACHINE' and `HKEY_CURRENT_USER? and then just deleted the C:\cygwin directory. I then reinstalled cygwin in the D drive, edited my start menu to point to them. However I have problems starting the sshd server. I have already ssh-host-config and ssh-user-config completed. The reason that your sshd service will not start is because you did not remove the service, and the one currently installed still points to the old location. ssh-host-config will not overwrite an existing service, so you need to remove the service and re-run the script. Remove the service with cygrunsrv -U sshd -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Start sshd server on reinstallation
On Sat, Jan 07, 2006 at 07:19:32PM -0800, Yitzchak Scott-Thoennes wrote: On Sat, Jan 07, 2006 at 07:09:14PM -0800, Brian Dessent wrote: Ordinary Olson wrote: I recently installed cygwin to my C drive ? however after realizing its size, I decided to move it to the D drive. So I stopped the sshd server that I had running, deleted the registry keys under `HKEY_LOCAL_MACHINE' and `HKEY_CURRENT_USER? and then just deleted the C:\cygwin directory. I then reinstalled cygwin in the D drive, edited my start menu to point to them. However I have problems starting the sshd server. I have already ssh-host-config and ssh-user-config completed. The reason that your sshd service will not start is because you did not remove the service, and the one currently installed still points to the old location. ssh-host-config will not overwrite an existing service, so you need to remove the service and re-run the script. Remove the service with cygrunsrv -U sshd Ack! I don't know where I got that; -U should be -R. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Cygwin(1.5.18) gcc (3.4.4) problems when compileing esd test prg
On Mon, Jan 09, 2006 at 06:06:41AM -0800, Brian Dessent wrote: Vijay Kiran Kamuju wrote: I am tring to compile this sample program on cygwin 1.5.18 and gcc (gcc 3.4.4) You're writing a program that tries to call esound libraries. These libraries are not in the Cygwin distribution. You'll have to build them yourself or find them elsewhere (Cygwin Ports has packages I believe.) Yes, they are. Packages libesound0 and libesound-devel. But the -lesd should come at the end of the gcc command. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: setup crashes trying to install xorg-x11-fnts
On Tue, Jan 10, 2006 at 11:28:31PM -0500, Igor Peshansky wrote:
If nothing works, however, you can do the following:
1. Download release/X11/xorg-x11-fnts/xorg-x11-fnts-6.8.1.0-3.tar.bz2 from
some mirror (I use mirrors.kernel.org with good results).
2. cd / tar xjf xorg-x11-fnts-6.8.1.0-3.tar.bz2
3. cd /etc/postinstall ./xorg-x11-fnts.sh
mv xorg-x11-fnts.sh{,.done}
4. Edit /etc/setup/installed.db and add this line:
xorg-x11-fnts xorg-x11-fnts-6.8.1.0-3.tar.bz2 0
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Segment fault/Bus error with Singular
On Wed, Jan 11, 2006 at 12:32:37PM -0500, Murray Eisenberg wrote: Sorry, I should not have asked about where to put the downloaded files; of course that would not matter given that the argument to tar includes the path to the downloads. BUT: The How do I install snapshots? FAQ answer is still unclear. It says: Before installing a snapshot, you must first Close all Cygwin applications, including shells and services (e.g. inetd, sshd), before updating cygwin1.dll 1. Download the snapshot, and run: cd / tar ... cd /tmp tar ... 2. After closing all Cygwin apps (see above), use Explorer or the Windows command shell to move C:\cygwin\tmp\usr\bin\cygwin1.dll to C:\cygwin\bin\cygwin1.dll. So once again I ask for clarification. I really do think it's a very simple and perfectly straightforward question: Does one run those 4 commands in Step 1 from a Windows command prompt or from within Cygwin? Not from a Windows command prompt. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: once more unto the breech - please try a snapshot so I can release this thing
On Wed, Jan 11, 2006 at 08:17:08PM -0500, Christopher Faylor wrote: On Wed, Jan 11, 2006 at 10:35:37AM -0800, Yitzchak Scott-Thoennes wrote: On Tue, Jan 10, 2006 at 09:23:41PM -0500, Christopher Faylor wrote: I hope that I've nailed down the last of the problems due to trying to hide the console, aka ssh doesn't work, aka emacs doesn't work, aka rxvt doesn't work, aka setsid something doesn't work. :( emacs doesn't work. Twice now with the 20060110 13:32:19 snapshot, I've had it no longer receive Ctrl-C's (running emacs from mutt from a cygwin.bat window). Doesn't happen consistently, but when it does, it persists in any emacs started in that window. When I finally exit bash, cmd.exe gives a Terminate batch job (Y/N)? prompt. Probably useless information: I had something similar happen once with, I think, the 20060105 snapshot, but then cygwin1.dll showed some error message (TM) about a minute after Ctrl-C was pressed. I notice that you included the part of my message where I say what I think I fixed but didn't include the part where I mentioned that we need to know the steps to duplicate the problem. What *exactly* are you doing? I think it's been mentioned a few times that neither Corinna nor I use emacs so running emacs from mutt doesn't mean a lot to me and it probably doesn't mean much to the only other person who's going to be interested in tracking down this problem. Usually using mutt's g command to reply to an email, which starts e.g. emacs -nw /tmp/mutt-DHX98431-3796-19 to edit an email. And when I've finished editing, I use the Ctrl-X Ctrl-C command to exit, and sometimes the Ctrl-X is registered but the Ctrl-C is ignored. And sometimes it works. What the difference is, I don't know. I've had it happen one more time, and still have the window open. Just now, I tried entering sleep 20 at the bash prompt in that window and right thereafter Ctrl-C and the sleep wasn't interrupted. Sorry I can't tell you what makes this happen or not happen. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: once more unto the breech - please try a snapshot so I can release this thing
On Wed, Jan 11, 2006 at 10:41:20PM -0800, Yitzchak Scott-Thoennes wrote: On Wed, Jan 11, 2006 at 08:17:08PM -0500, Christopher Faylor wrote: On Wed, Jan 11, 2006 at 10:35:37AM -0800, Yitzchak Scott-Thoennes wrote: On Tue, Jan 10, 2006 at 09:23:41PM -0500, Christopher Faylor wrote: I hope that I've nailed down the last of the problems due to trying to hide the console, aka ssh doesn't work, aka emacs doesn't work, aka rxvt doesn't work, aka setsid something doesn't work. :( emacs doesn't work. Twice now with the 20060110 13:32:19 snapshot, I've had it no longer receive Ctrl-C's (running emacs from mutt from a cygwin.bat window). Doesn't happen consistently, but when it does, it persists in any emacs started in that window. When I finally exit bash, cmd.exe gives a Terminate batch job (Y/N)? prompt. Probably useless information: I had something similar happen once with, I think, the 20060105 snapshot, but then cygwin1.dll showed some error message (TM) about a minute after Ctrl-C was pressed. I notice that you included the part of my message where I say what I think I fixed but didn't include the part where I mentioned that we need to know the steps to duplicate the problem. What *exactly* are you doing? I think it's been mentioned a few times that neither Corinna nor I use emacs so running emacs from mutt doesn't mean a lot to me and it probably doesn't mean much to the only other person who's going to be interested in tracking down this problem. Usually using mutt's g command to reply to an email, which starts e.g. emacs -nw /tmp/mutt-DHX98431-3796-19 to edit an email. And when I've finished editing, I use the Ctrl-X Ctrl-C command to exit, and sometimes the Ctrl-X is registered but the Ctrl-C is ignored. And sometimes it works. What the difference is, I don't know. I've had it happen one more time, and still have the window open. Just now, I tried entering sleep 20 at the bash prompt in that window and right thereafter Ctrl-C and the sleep wasn't interrupted. Sorry I can't tell you what makes this happen or not happen. Just in case it's relevant, note that I have experimental bash, readline, libreadline6, findutils, and coreutils. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: [PATCH] Proposed clarification of the snapshot installation FAQ
On Thu, Jan 12, 2006 at 07:57:27PM -0800, Joshua Daniel Franklin wrote: On 1/11/06, Igor Peshansky wrote: As mentioned in http://cygwin.com/ml/cygwin/2006-01/msg00537.html, here's a patch to the FAQ to clarify the section on installing snapshots. I didn't know whether the various *.texinfo files are still used, so I ported the modifications there as well, just in case. Applied to faq-setup.xml (the texinfo files are no longer used... I suppose I should remove them). It would be nice to have a sample batch file that automated the cygwin1.dll replacement, too. I've prettied up a batch file I use for installing cygwin-inst-* snapshots, and added stuff to handle spacey pathnames and inetd (which AIUI isn't always a cygrunsrv service), neither of which I personally use. Just drag and drop the tarball onto the batch file, or run with absolute tarball name as argument via a dos prompt. @echo off cd c:\cygwin path bin;%path% sh -c 'echo Preparing to install `cygpath -ua %1`;echo' echo Please shut down all cygwin applications before proceeding pause echo Shutting down any cygwin services . . . net stop inetd nul 2nul sh -c for serv in `cygrunsrv --list`;do cygrunsrv --stop $serv;done echo Installing . . . move bin\cygwin1.dll sh -c 'tar xfj `cygpath -u %1`' echo Restarting any cygwin services . . . net start inetd 2nul sh -c for serv in `cygrunsrv --list`;do net start $serv;done bin\echo cygstart cygwin.bat bin\echo echo Verify cygwin.bat and any services correctly start before deleting old dll pause del cygwin1.dll -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: once more unto the breech - please try a snapshot so I can release this thing
On Thu, Jan 12, 2006 at 08:35:18PM -0500, Christopher Faylor wrote: On Wed, Jan 11, 2006 at 10:46:24PM -0800, Yitzchak Scott-Thoennes wrote: Just in case it's relevant, note that I have experimental bash, readline, libreadline6, findutils, and coreutils. Does the latest snapshot behave any differently? Neither Corinna nor I can duplicate this problem so neither of us can do anything other than shoot in the dark to try to solve it. I haven't had it happen again using 20060112. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: once more unto the breech - please try a snapshot so I can release this thing
On Sat, Jan 14, 2006 at 05:58:32PM -0800, Yitzchak Scott-Thoennes wrote: On Thu, Jan 12, 2006 at 08:35:18PM -0500, Christopher Faylor wrote: On Wed, Jan 11, 2006 at 10:46:24PM -0800, Yitzchak Scott-Thoennes wrote: Just in case it's relevant, note that I have experimental bash, readline, libreadline6, findutils, and coreutils. Does the latest snapshot behave any differently? Neither Corinna nor I can duplicate this problem so neither of us can do anything other than shoot in the dark to try to solve it. I haven't had it happen again using 20060112. I spoke too soon :(, but I wouldn't think you should hold up the release on this. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installing cygwin by manually copying c:\cygwin to another machine
On Fri, Jan 13, 2006 at 08:59:25PM -0500, Brett Serkez wrote: [snip] Not particularly, assuming you use the same local package cache on each machine, especially if you use the trick for pulling in all the packages you want automatically. Right. I don't know about chere, chere must have settings in the registry, as it adds a 'bash here' menu option in explorer when you right click on a directory. You'd have to run 'chere -i' on the new system. That means that setup, services like sshd, cron, inetd, etcetera will be available for installation. Right, available for installation, you'd have to run the setup scripts. I suppose running ssh-host-config wouldn't be a bad idea, not only would it install the Windows sshd service on the new system, but it would generate unique host keys, otherwise both the new and old systems would have the same key. Sounds like this is doable, but there may be a series of commands that need to be run, depending on the original installation. I'm not sure what this approach buys you over running setup.exe in unattended install mode on a copied local package cache, except fragility. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installing cygwin by manually copying c:\cygwin to another machine
On Mon, Jan 16, 2006 at 10:08:03AM +0200, Adrian Maier wrote: On 1/16/06, Yitzchak Scott-Thoennes [EMAIL PROTECTED] wrote: I'm not sure what this approach buys you over running setup.exe in unattended install mode on a copied local package cache, except fragility. Is there an unattended install mode? I thought that there is no such thing , because according to http://www.cygwin.com/cygwin-ug-net/setup-net.html : Unfortunately setup.exe does not yet support unattended installs. In fact, that's precisely why I asked the question in the first place: i needed a way to install cygwin without the need of choosing individual packages; and with no need of internet connection ... I think the key word there is does not yet *support*. Run setup --help from a command prompt, and setup will output a list of its command line options into setup.log.full. Play with running it using them on a previously prepared local download directory, with non-default packages included either by changing setup.ini to add Base to their category or adding a dummy mirror directory with a setup.ini having a dummy package that has category Base and requires the other packages you want. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
[ANNOUNCEMENT] Updated: fortune-1.99.1-2 (experimental)
This version of fortune contains many tweaks to the fortune data files from the debian 1.99.1-2 and 1.99.1-3 releases, including the new off/disclaimer file, and adds a new, undocumented, -u switch to fortune. It is now built using the cygwin generic build script, and requires a recent snapshot or 1.5.19. Assuming there are no reported problems, it will become the current version shortly after cygwin 1.5.19 is released. = To update your installation, click on the Install Cygwin now link on the http://cygwin.com/ web page. This downloads setup.exe to your system. Then, run setup and answer all of the questions. If you have questions or comments, please send them to the Cygwin mailing list at: cygwin@cygwin.com . *** CYGWIN-ANNOUNCE UNSUBSCRIBE INFO *** If you want to unsubscribe from the cygwin-announce mailing list, look at the List-Unsubscribe: tag in the email header of this message. Send email to the address specified there. It will be in the format: [EMAIL PROTECTED] If you need more information on unsubscribing, start reading here: http://sources.redhat.com/lists.html#unsubscribe-simple Please read *all* of the information on unsubscribing that is available starting at this URL. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installing cygwin by manually copying c:\cygwin to another machine
On Mon, Jan 16, 2006 at 12:23:00PM +0200, Adrian Maier wrote: On 1/16/06, Chris Taylor [EMAIL PROTECTED] wrote: Adrian Maier wrote: On 1/16/06, Yitzchak Scott-Thoennes [EMAIL PROTECTED] wrote: I think the key word there is does not yet *support*. Run setup --help from a command prompt, and setup will output a list of its command line options into setup.log.full. Play with running it using them on a previously prepared local download directory, with non-default packages included either by changing setup.ini to add Base to their category or adding a dummy mirror directory with a setup.ini having a dummy package that has category Base and requires the other packages you want. I see: this functionality is in development and therefore is well hidden for the moment. It's a shame that it's not clearly documented . setup --help displays nothing for me I suspect this is a cmd.exe or win2000 - related problem , because i have seen the same happening with other programs as well (namely qemu). Adrian Maier Please re-read what Yitzchak said - it outputs the list of options into setup.log.full Just for emphasis, I'll say that again: *setup.log.full* Open that for the options. Uh, you are right. Sorry guys, I haven't read the entire phrase written by Yitzchak ... my fault. It looks like I am too used to the idea of programs writing their messages to stdin/stdout . It certainly qualifies as quaint behavior. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
snapshot installation bat file (was: Re: [PATCH] Proposed clarification of the snapshot installation FAQ)
On Wed, Jan 18, 2006 at 06:41:27PM -0800, Joshua Daniel Franklin wrote: On 1/13/06, Yitzchak Scott-Thoennes wrote: It would be nice to have a sample batch file that automated the cygwin1.dll replacement, too. I've prettied up a batch file I use for installing cygwin-inst-* snapshots, and added stuff to handle spacey pathnames and inetd (which AIUI isn't always a cygrunsrv service), neither of which I personally use. Just drag and drop the tarball onto the batch file, or run with absolute tarball name as argument via a dos prompt. This batch file worked great for me on XP. The only issue I see is that if someone set their cygrunsrv or inetd to Manual this would actually start it up. I can put a note to that effect. Any objections to putting this up at http://cygwin.com/faq/faq.setup.html#faq.setup.snapshots ? I don't have any, but... I just tried it again drag-and-dropping a tarball *without* spaces in the path (spaces make Windows include around the parameter), and the \'s weren't properly quoted. Anybody know how to get it to work either way? Also be nice if it could be made to work on just the dll.bz2 file also. If we can come up with a general purpose, no extra packages required, bat file, maybe it could be downloadable directly from the snapshots page. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Cygwin Setup: Fatal Error: Uncaught Exception
On Sun, Jan 22, 2006 at 09:37:12PM -0800, Brian Dessent wrote: Linda Walsh wrote: Is an application crash the expected behavior when package validation fails? Maybe it should mark it unusable or delete it? Yuck. I went through this, too. It appeared to me that there were a large number (a dozen? I no longer remember) of packages uploaded more than once, without bumping the version, and if you were so fortunate as to get the earlier rather than the later copy, setup won't have updated to the later. It might be easier to blow away your local package directory than to run setup over and over, each time identifying one more package to rename/delete. At the moment, yes. It's a consequence of a change in the md5 checking. It just means you need to delete the package, as it has the wrong size or checksum. It's on the todo list to clean this up. Is there a snapshot that has this fixed? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: can I tell setup.exe NOT to upgrade a specific package from now on?
On Sat, Jan 21, 2006 at 02:54:05PM +, Urs Rau wrote: I need to keep a version of postgresql server at level 7.x, but now that 8.x is out setup.exe suggests an upgrade every single time it runs. And one of these days I am going to forget to manually toggle the entry back to 'keep' instead of the suggested 'upgrade'. Is there any mechanism at all where I can either tell setup.exe to ignore future versions of posgresql or else actually be as precise, as to tell it that I want to stay at version 7.x for now? A couple of suggestions: Remove the line for postgresql in /etc/setup/installed.db. Then setup will no longer think you have it installed, and won't upgrade it (unless something else you have has postgresql as a dependency). Always use a .bat file for upgrading that runs setup twice; once to download and once to install, with the postgresql directory renamed during the install phase: cd local-package-dir .\setup -D (now make your package choices and download them) cd *%2f* ren release\postgresql postgresql-noinst cd .. .\setup -L (now install) cd *%2f* ren release\postgresql-noinst postgresql cd .. (AIUI due to a bug in setup.exe, the -D and -L flags won't actually help at the moment, but they don't hurt either.) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: apache 1.3.34 can not compile anymore
On Tue, Jan 24, 2006 at 12:31:11AM -0800, Brian Dessent wrote: [EMAIL PROTECTED] wrote: Tried to roll back to 1.5.18-1 from the setup.exe, but now I get 10s of error messages from packages who do not find getline in the cygwin dll. You would have to also use a previous version of any programs that need getline(). I think this includes findutils and coreutils. Also readline and libreadline6 need 1.5.19 I think. (There's a pattern here...) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Cygwin Setup: Fatal Error: Uncaught Exception
On Mon, Jan 23, 2006 at 02:56:39PM -0500, Igor Peshansky wrote: Moving to cygwin-apps, as this is likely to get technical. On Mon, 23 Jan 2006, Brian Dessent wrote: Igor Peshansky wrote: I've looked at this a bit. Here's the weird part: the error says Uncaught Exception, but all the throws of that exception appear to be properly wrapped in try/catch blocks. So a simple change exception into an mbox kind of solution won't work here. More debugging is needed. The reason for the box is that the md5 checking was changed to run in a different thread than originally designed (now in the main thread instead of the download thread IIRC) and that thread does not have any particular catch handler for that exception, only the TOPLEVEL_CATCH which punts with the generic error. Do you mean packagemeta::ScanDownloadedFiles() calling packageversion::scan(), which calls check_for_cached()? Then yes, this isn't properly wrapped in a try/catch. I'd like to verify that this is the culprit (when someone sends me the corrupt tarball), but I think I see a proper fix for this. Will submit a patch shortly. Just to reemphasize, these are *not* corrupt tarballs. They are tarballs exactly as downloaded, extracted, and installed. It's just that later the versions on the cygwin mirror became different while keeping the same version/filename. I verified in a couple of the cases that the newer version actually had executables rebuilt, with slightly different file sizes and timestamps. I don't think I have any of them around any more, but if you were to pick a current tarball in your local package directory and un-bzip2 it and re-bzip2 it with a different compression level, you should see the problem. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Errors compiling cdrtools under cygwin 1.5.19
On Tue, Jan 24, 2006 at 10:43:30PM +0100, [EMAIL PROTECTED] wrote: I used cygwin happily for very long time to compile apache/php/postgresql and enjoy symlinks, and now I am cut-off from one day to the next. The apache folks do not seem to care. The bug I submitted is still without reply - http://issues.apache.org/bugzilla/show_bug.cgi?id=38364 Apache 2.x/php 5.x do not want to play on cygwin so far. So I am three days in the dark and testing like hell vmware and minigw to save my skin. I don't understand why you haven't just reverted cygwin to the previous version (yes, including packages that depend on 1.5.19 - a brief perusal of the cygwin-announce archives from the release of 1.5.19 onward would show you which packages may fall into this category). The cygwin distribution provides access to previous versions *because* it is known that things break from time to time, whether due to a problem in or out of cygwin's control. Additionally, if cygwin is that mission critical to you, you need to have a testing phase between downloading new versions and putting them into live production, just like for any other mission critical software. Seems this getline() breaks quite a lot and I am not quite sure this is _very_ positive for cygwin. People just get left alone in the dark (no everybody can debug and patch) and the pride of cygwin is somehow self focused. I don't understand the self focused part. Re: alone in the dark, If your concerns aren't addressed in the next few months, I think you can make that claim. I would expect such dramatic moves to be done with more care. The only care that really could be taken to prevent things like this is more users testing pre-release versions. Development snapshots of cygwin with getline() have been available for a long time now. Note that this isn't just addressed to you; if package maintainers heeded the release coming soon, please test a snapshot messages cgf sends out by testing that their packages build and run with the snapshot, there'd be less scope for problems. Otherwise I could call cygwin nice, but not reliable. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: new cygwin dlls
On Thu, Jan 26, 2006 at 11:43:29AM +0900, djh wrote: Has a fix been found for building emacs on the new cygwin versions 5.19 and up yet? There is at least one problem with d_ino as you know. I thought you said you'd worked around that (though you didn't share what your actual changes). An possibly another, since temacs crashes with a stack dump, during bootstrap. I saw you made a partial report of this, and but didn't see any response to the person who tried to help a step further in your investigation via gdb. cygwin distribution has an early emacs in it. Will it still run on the newer 5.19 versions? Has any regression testing been carryied out? Yes. Has anyone been assigned to look into these problems? s/been assigned/volunteered/ you mean? Nobody gets assigned or paid to do anything, unless perhaps through a RedHat support contract. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: [ANNOUNCEMENT] GNU CLISP 2.38 (2006-01-24) released
On Thu, Jan 26, 2006 at 11:07:28AM +0100, Dr. Volker Zell wrote: Sam Steingold writes: o There is NO requires line in setup.ini at least at the following download side: http://mirrors.kernel.org whereas the clisp executables depend on various dlls. setup.hint contains this: requires:crypt cygwin libdb4.3 libiconv2 libintl3 libncurses8 libpcre0 libreadline6 zlib But the GENERATED file on the mirrors doesn't include a requires line. Maybe because of no space between requires: and crypt? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
binutils testsuite failures? (was: Re: Errors compiling cdrtools under cygwin 1.5.19)
On Tue, Jan 24, 2006 at 10:28:50PM -0500, Christopher Faylor wrote:
I should point out that I didn't rebuild my packages under 1.5.19 either
so I'm just as guilty as any other package maintainer in this regard.
Speaking of which, in the course of making a small patch to ld (which
I hope the binutils folk will apply and will find its way into an
updated cygwin package at some point) I tried building binutils and
got 2 unexpected test failures and 1 unexpected pass for ld (see
attached). I didn't revert to 1.5.18 to see if that made any
difference though. I probably just have something not set up
correctly.
make[1]: Entering directory `/usr/src/binutils-20050610-1/ld'
Making check in po
make[2]: Entering directory `/usr/src/binutils-20050610-1/ld/po'
make[2]: Nothing to be done for `check'.
make[2]: Leaving directory `/usr/src/binutils-20050610-1/ld/po'
make[2]: Entering directory `/usr/src/binutils-20050610-1/ld'
make check-DEJAGNU
make[3]: Entering directory `/usr/src/binutils-20050610-1/ld'
Making a new site.exp file...
srcroot=`cd .././ld pwd`; export srcroot; \
r=`pwd`; export r; \
LC_COLLATE=; LC_ALL=; LANG=; export LC_COLLATE LC_ALL LANG; \
EXPECT=expect; export EXPECT; \
if [ -f ./../expect/expect ]; then \
TCL_LIBRARY=`cd .././ld/../tcl/library pwd`; \
export TCL_LIBRARY; \
fi; \
runtest=runtest; \
if /bin/sh -c $runtest --version /dev/null 21; then \
$runtest --tool ld --srcdir ${srcroot}/testsuite \
CC=gcc -L/usr/src/binutils-20050610-1/./ld CFLAGS=-g -O2 \
CXX=c++ -L/usr/src/binutils-20050610-1/./ld CXXFLAGS=-g -O2 \
CC_FOR_HOST=gcc CFLAGS_FOR_HOST=-g -O2 \
OFILES=ldgram.o ldlex.o lexsup.o ldlang.o mri.o ldctor.o ldmain.o
ldwrite.o ldexp.o ldemul.o ldver.o ldmisc.o ldfile.o ldcref.o ei386pe.o
deffilep.o pe-dll.o BFDLIB=../bfd/.libs/libbfd.a \
LIBIBERTY=../libiberty/libiberty.a ./../intl/libintl.a LIBS= \
; \
else echo WARNING: could not find \`runtest' 12; :;\
fi
WARNING: Couldn't find the global config file.
WARNING: Couldn't find tool init file
Test Run By sthoenna on Fri Jan 27 00:23:25 2006
Native configuration is i686-pc-cygwin
=== ld tests ===
Schedule of variations:
unix
Running target unix
Using /usr/share/dejagnu/baseboards/unix.exp as board description file for
target.
Using /usr/share/dejagnu/config/unix.exp as generic interface file for target.
Using /usr/src/binutils-20050610-1/ld/testsuite/config/default.exp as
tool-and-target-specific interface file.
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-alpha/alpha.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-arm/arm-elf.exp ...
Running
/usr/src/binutils-20050610-1/ld/testsuite/ld-auto-import/auto-import.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-bootstrap/bootstrap.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-cdtest/cdtest.exp ...
FAIL: cdtest with -Ur
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-checks/checks.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-cris/cris.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-crx/crx.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-cygwin/exe-export.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-d10v/d10v.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-discard/discard.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elf/elf.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elf/exclude.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elf/frame.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elf/sec64k.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elfcomm/elfcomm.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elfvers/vers.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elfvsb/elfvsb.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-elfweak/elfweak.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-fastcall/fastcall.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-frv/fdpic.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-frv/frv-elf.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-frv/tls.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-h8300/h8300.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-i386/i386.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-ia64/ia64.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-linkonce/linkonce.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-m68hc11/m68hc11.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-maxq/maxq.exp ...
Running
/usr/src/binutils-20050610-1/ld/testsuite/ld-mips-elf/mips-elf-flags.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-mips-elf/mips-elf.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-mmix/mmix.exp ...
Running /usr/src/binutils-20050610-1/ld/testsuite/ld-pe/pe.exp ...
Running
Re: LD_PRELOAD regression on 1.5.19-4 ? no more loaded library in child process
On Fri, Jan 27, 2006 at 06:31:29PM +0100, Louis Lecaroz wrote: from cmd.exe -Start bash --All environment variables from cmd.exe were propaged in bash environment+.profiles one. --Start vi sample.txt from this bash instance A new instance of bash is loaded with also all environment variables. -A VIM.exe instance is loaded but only with PATH, SYSTEMDRIVE, SYSTEMROOT, WINDIR ! I can't help you, but I can see that there's a lot of information missing here. What did you actually *do* to get from one step to the next in the above? Where does your MSVC-linked DLL come into it? What loads that DLL? What is your system like (the information requested in http://cygwin.com/problems.html)? I tried also to verify with the win32 system routine called GetEnvironmentStrings(), but it is exactly the same data retrieved with getenv() ! So it (I feel) seems that this new version of cygwin did not propaged all environment variable to all child process ! Don't know why :( but I Just tested it on the 1.5.18 patched with the snapshot mentioned below, just reinstall the 1.5.19-4 over it (after having removed the previous), my conclusion (but I am not the Cygwin Expert ;) ), is that LD_PRELOAD since 1.5.19-4 previuous does have the same behavior as environment variable appears to be correctly propagated in 1.5.18-x not in 1.5.19-4. I saw (but did not verify) one report that 1.5.19 doesn't keep the windows copy of the environment in some situations where 1.5.18 did, so manually using CreateProcess or the like may work differently. AFAICT, this falls in the category of implementation detail that you shouldn't be counting on. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: /proc/pid/exe points to void
On Tue, Jan 24, 2006 at 06:45:28PM +0100, Corinna Vinschen wrote: On Jan 20 13:50, Sam Steingold wrote: On Mar 10 16:00, Sam Steingold wrote: /proc/pid/exe points to foo, not to foo.exe, so it cannot be opened c. how do I find out which file is running if /proc/pid/exe cannot be opened? access(2) or stat(2) http://www.opengroup.org/onlinepubs/009695399/functions/access.html the above spec of access appears to indicate that if access() succeeds then open() must succeed too. this is not the case in cygwin: /proc/self/exe cannot be open()ed. I've just checked in a patch which tacks on the .exe suffix to /proc/$PID/exe, as well as a patch to realpath which returns the pathname with .exe suffix, even if the original name has no suffix given. We will give this a try. Please test the next snapshot. Mostly for the archives, I note that this makes perl's $^X variable include the .exe suffix. This is probably a good thing, but at the moment triggers a test failure. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: readdir after rewinddir not working in 20060128 snapshot
On Mon, Jan 30, 2006 at 11:39:48AM +0100, Corinna Vinschen wrote:
On Jan 30 00:29, Yitzchak Scott-Thoennes wrote:
After a rewinddir, readdir seems to return as many empty entries as there
were actual entries left to read, followed by . and ..
Thanks for the testcase! Since the underlying NT call NtQueryDirectoryInfo
is able to return more than one directory entry in one call, I thought it
might be a good idea to implement readdir caching. Unfortunately I forgot
to reset the cache in case of rewinddir. I've applied a fix now.
Thanks, but..
With 20060202, that testcase now works, but if you read all the way to
the end, then rewind, you only get . and .. thereafter.
$ cat rewdir.c
#include stdio.h
#include dirent.h
#include errno.h
#include string.h
void readentries(DIR *dh, int limit) {
struct dirent *de;
int count;
for (count = 0; count limit; ++count) {
errno = 0;
if ((de = readdir(dh)) == NULL) {
if (errno != 0) printf (read error: %s\n, strerror (errno));
break;
}
printf (got: %s\n, de-d_name);
}
}
int main(int argc, char **argv)
{
DIR *dh;
int i;
if (argc != 2) return 1;
if ( (dh = opendir (argv[1])) == NULL ) {
fprintf (stderr, couldn't open '%s': %s\n, argv[1], strerror (errno));
return 1;
}
for (i = 0; i 3; ++i) {
readentries (dh, );
printf (rewinding.\n);
errno = 0;
rewinddir (dh);
if (errno != 0) printf (unexpected errno: %d\n, errno);
}
return 0;
}
$ rm -rf foo; mkdir foo; touch foo/{bar,baz,quux,quuux,qx}
$ gcc -Wall rewdir.c -o rewdir ./rewdir foo
got: .
got: ..
got: bar
got: baz
got: qx
got: quuux
got: quux
rewinding.
got: .
got: ..
rewinding.
got: .
got: ..
rewinding.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: strange cygstart bug with current Cygwin versions
On Mon, Feb 06, 2006 at 11:40:22PM +0100, Michael Schaap wrote: What we basically need to do, is copy the Cygwin environment to the Windows environment, taking care of path conversion for all the appropriate variables. Maybe start with: http://www.cygwin.com/ml/cygwin-patches/2005-q4/msg9.html -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: make: rm: command not found
On Sat, Feb 11, 2006 at 03:15:37AM -0800, Tim Prince wrote: At 08:40 AM 2/10/2006, JefV wrote: -- View this message in context: http://www.nabble.com/make%3A-rm%3A-command-not-found-t1100243.html#a2873151 Sent from the Cygwin Users forum at Nabble.com. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ SPAM: -- Spamnix Spam Report -- SPAM: Spamnix identified this message as spam. SPAM: Score: 52 SPAM: Action: Quarantine SPAM: --- End of Spamnix Spam Report -- How many of us do you expect to be viewing your message with spam filtering off? Why should your spam filtering be of any concern to anyone posting to the list? Perhaps if you were to figure out what contributed to that 52 score and make a concrete suggestion, it would be helpful. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: default PATH
On Sun, Feb 12, 2006 at 11:57:00AM -0500, Christopher Faylor wrote: If we don't get a single person indicating that they rely on the current behavior then I'm ok with changing it. We have a patch ready to be checked in, in fact. I assume that: $ PATH=/foo::$PATH cmd /c printenv PATH will continue to show /foo:./:yaddayadda with the patch? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: CPAN: issues with homedirs containing spaces; how to install globally?
On Fri, Feb 17, 2006 at 02:32:22AM -0500, [EMAIL PROTECTED] wrote: Hi all, I think there's a bug with cpan and names containing spaces. Please report bugs in CPAN.pm to [EMAIL PROTECTED] This isn't unique to cygwin. I suggest you remove the space. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygheap base mismatch detected
On Sat, Feb 18, 2006 at 09:12:02PM -0600, Dill, Jens (END-CHI) wrote: So, I have a workaround of sorts. I can have my script launch my app by writing the command line to a .bat file and executing it. Definitely not something I can use to convince my management to go with CygWin. Wouldn't just system(cmd /c yourapp) work instead of writing a bat file? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygheap base mismatch detected
On Tue, Feb 21, 2006 at 09:25:45AM -0600, Dill, Jens (END-CHI) wrote: (2) Rebase the CygWin DLL so that it loads by default into a space not used in either memory map (I'd need help in choosing such a space). I've tried both Microsoft's rebase and CygWin's rebase, but the documentation for both is scanty (how is the -b argument to be formatted -- a hex number with 0x in front or without?), and the error messages I get in response are not helpful (error 6?). cygcheck -l rebase | grep README | xargs less -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: cygheap base mismatch detected ( Attn: rebase maintainer )
Jason, are you following this?
On Sat, Feb 18, 2006 at 09:12:02PM -0600, Dill, Jens (END-CHI) wrote:
We are finally zeroing in on the problem.
Mark Geisert writes:
The code at /src/rebase-2.3.1/rebase.c:255 assumes the signature is at
offset 0x80
in the image. This was true in the early Windows days but has long since
been
generalized. The technique nowadays is to obtain the short integer value
e_lfanew
at offset 0x3C in the image, and use that as the offset to check for the
signature.
I dumped out a couple of dlls in hex to see if they supported Mark's
assertion.
Yes, indeed. And I was able to verify (see below) that all the DLLs that
were linked into my app and were not touched by rebase, in fact has
something
other than 0x0080 in that position of the image (see below).
I also did some experimentation that makes me much more certain that
a repaired rebase will fix the problem. The rebase documentation
talks about the fact that rebase exists to fix problems when one
process forks another that has the same DLLs based at an incompatible
address. This appears to be what is happening to my app.
I have verified that if any CygWin shell tries to launch my app (either
by fork/exec or just by exec), or if my my app tries to launch a CygWin
shell, we get the identical cygheap base mismatch problem. If the
launch is indirect (via a Windows .bat file), there is no problem.
So, I have a workaround of sorts. I can have my script launch my app
by writing the command line to a .bat file and executing it. Definitely
not something I can use to convince my management to go with CygWin.
(The latest word was that our VP thinks that there is no enterprise app
out there that uses cygwin so he is skeptical too.)
I finally found where to get the rebase source, and verified that in fact,
what Mark noticed in 2.3.1 is still true in 2.4.2-1. I can easily make the
obvious fix and change the is_rebaseable function to get the
pe_signature_offset
out of position 0x3c in the image rather than assuming it is 0x80.
But that only affects the bit of the code that decides if a DLL is
rebaseable. I would need more time and knowledge to convince myself
that the code that actually does the rebasing is not making the same
mistake.
It seems that there is indeed more to it. I did make the obvious change
and reran rebaseall. The message I got from the first Oracle DLL
it encountered was:
ReBaseImage (/cygdrive/d/oracle/app/oracle/product/9.2.0/bin/orasql9.dll)
failed with last error = 6
I can't do more without learning a lot more than I currently know
about the internals of DLLs and of rebase.
But, let us assume for the moment, that we have found the problem,
that someone can fix rebase, and we can use it to keep out of trouble.
How shall I represent this to my management?
Can someone tell me how long it might take for the fix to get into
a stable CygWin release? My management may be willing to use an
uncertified release for a short while, and even may be willing to
own the responsibility for making the change to rebase, but they'll
want to know how long they have to wait for it to be official.
And if rebase solves the problem, I presume we have to do it
after we've installed Oracle and before we run any of our apps.
What happens if we or one of our customers reinstalls Oracle?
Do we have to make sure that rerunning rebaseall is part of
the drill? The doc for rebase says:
WARNING
Note it is *strongly* recommended that users only use rebaseall unless
they *really* know what they are doing or are instructed by one of the
Cygwin developers.
/WARNING
Not something we want to have to hand off to our customers, or
even to our installation techs if we can avoid it.
-- Jens
below this line is the code I used to peek at our DLLs, and the results
---
$ cat dllpeek.c
#include stdlib.h
#include stdio.h
int main(int argc, char * * argv)
{
while (argc--)
{
int rc = 0;
unsigned char at3C[2];
char signature[4];
char * fname = *++argv;
FILE * fp = fopen(fname,rb);
if (!fp)
{ fprintf (stderr, %s: could not open file\n, fname);
continue;
}
rc = fseek(fp, 0x3c, SEEK_SET);
if (rc)
{ fprintf (stderr, %s: fseek returned %d\n, fname, rc);
continue;
}
rc = fread (at3C, 2, 1, fp);
if (rc != 1)
{ fprintf (stderr, %s: could not read at 0x3c\n, fname);
continue;
}
fprintf (stdout, %s: %02x%02x\n, fname, at3C[1], at3C[0]);
}
return 0;
}
$ ~/dllpeek.exe $(cygpath -u $(cygcheck ./acqjob.exe))
./acqjob.exe: 0080
/cygdrive/d/oracle/app/oracle/product/9.2.0/bin/orasql9.dll: 00e0
Re: cygwin ld --out-implib treats informational message as warning (attn: binutils maintainer)
Can cygwin binutils be updated to include this patch? Thanks. On Fri, Jan 27, 2006 at 05:26:37PM +, Nick Clifton wrote: Hi Yitzchak, 2006-01-27 Yitzchak Scott-Thoennes [EMAIL PROTECTED] * pe-dll.c (pe_dll_generate_implib): Issue Creating library file: as informational message, not warning Approved and applied. Cheers Nick PS. We might as well leave the message in for now. I see no reason for the linker not to be informative if the user has requested it. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: stty under cygwin
On Wed, Apr 21, 2004 at 06:38:09PM -0700, Karl M wrote: I am sure I would have come to the same conclusion if I hadn't recognized the colon separated string. cgf That was a really nice thing to say. I thought so too. LGPL can't be far away. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: stty under cygwin
On Thu, Apr 22, 2004 at 11:03:12AM -0400, Christopher Faylor [EMAIL PROTECTED] wrote: On Thu, Apr 22, 2004 at 04:38:18PM +0200, bertrand marquis wrote: Christopher Faylor a ?crit: On Thu, Apr 22, 2004 at 09:10:46AM +0200, bertrand marquis wrote: First of all stty 1:0:cbd:0:3:1c:7f:15:4:5:1:0:11:13:1a:0:12:f:17:16:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 /dev/ttyS0 is a regular linux command, you can try. In fact i'm trying to porting some stuff from linux to cygwin and it seem that the stty under cygwin doesn't accept to have settings in input. To repeat what I said in a previous message, Cygwin's stty does accept this kind of input. You can't generate the string on linux and use it on Cygwin, however. In fact you are right cygwin stty accept this kind of input but it seems that the 1:..:0 has to be be shorter under cygwin That falls under the You can't generate the string on linux and use it on Cygwin. You are doomed to failure if you try this. Really. Perhaps some kind soul can feed his string to stty on linux (assuming, probably incorrectly, that all linuxen share the same format) and then do stty -a, replicate those settings under cygwin, and report what the cygwin equivalent stty -g would be? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: i want to re-download all packages... how?
On Thu, Apr 22, 2004 at 05:00:50PM +0200, electa wrote: i have some packeges installed, but i have lost their .bz2 files... now i want to retrieve all packages because I would like to put them on a CD-RW! -- setup.exe knows what packages i have alreay installed and don't download them... In download to local directory mode, try toggling the selector next to the All category to Reinstall? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Maildir and Cygwin
On Thu, Apr 22, 2004 at 08:26:00AM -0700, Eduardo Chappa wrote: :) that Cygwin is different (e.g. we could do this in the cygwin :) applications mailing list). :) :) Why do you want to change so many programs when all you need is a :) cygwin managed mount? IMO, the correct solution for mutt to make everyone happy would be a new configuration variable to specify the character (presumably defaulting to non-: for cygwin). I think I've seen both ; and - suggested in different places. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: i want to re-download all packages... how?
So do it once that way to get the installed packages and then run it again to get the uninstalled packages. On Thu, Apr 22, 2004 at 05:43:21PM +0200, electa wrote: it selects only my installed packs, but not the other. i want all for my cygwin-CD! Yitzchak Scott-Thoennes [EMAIL PROTECTED] ha scritto nel messaggio Please feed the spammers my email address; it can only be a drop in the bucket over the 10-30 Mb/day I get now. :) news:[EMAIL PROTECTED] On Thu, Apr 22, 2004 at 05:00:50PM +0200, electa wrote: i have some packeges installed, but i have lost their .bz2 files... now i want to retrieve all packages because I would like to put them on a CD-RW! -- setup.exe knows what packages i have alreay installed and don't download them... In download to local directory mode, try toggling the selector next to the All category to Reinstall? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Maildir and Cygwin
On Thu, Apr 22, 2004 at 11:33:13PM -0500, GARY VANSICKLE wrote: On Thu, Apr 22, 2004 at 08:26:00AM -0700, Eduardo Chappa wrote: :) that Cygwin is different (e.g. we could do this in the cygwin :) applications mailing list). :) :) Why do you want to change so many programs when all you need is a :) cygwin managed mount? IMO, the correct solution for mutt to make everyone happy would be a new configuration variable to specify the character (presumably defaulting to non-: for cygwin). I think I've seen both ; and - suggested in different places. No, all that would do is guarantee that a mutt using ; couldn't read a Maildir written by a fetchmail using - (i.e. the disagreement would simply move from being hardcoded to being defined at compile time). The correct solution is to implement a corrected spec. configuration variables is what mutt calls user settings; settable in .muttrc by the user. Perhaps it would even work to have a folder hook override it so one Maildir folder could use one thing and another use another. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
errors using DB_File perl module under 20040422 snapshot
It seems to be creating files that cygwin cannot thereafter open. $ perl -we'use Fcntl; use DB_File; tie %h, DB_File, bugaboo, O_RDWR()|O_CREAT(), 0666 or die error: $!' Name main::h used only once: possible typo at -e line 1. error: No such file or directory at -e line 1. $ ls *bugaboo* ls: __db.bugaboo.: No such file or directory $ rm *bugaboo* rm: cannot remove `__db.bugaboo.': No such file or directory $ cmd /c ' dir *bugaboo*' Volume in drive C has no label. Volume Serial Number is 7422-45EE Directory of C:\cygwin\home\sthoenna 04/25/2004 12:05 PM 0 __db.bugaboo. 1 File(s) 0 bytes 0 Dir(s) 13,580,779,520 bytes free $ cmd /c 'del *bugaboo*' $ ls *bugaboo* ls: *bugaboo*: No such file or directory Cygwin Configuration Diagnostics Current System Time: Sun Apr 25 12:15:11 2004 Windows XP Professional Ver 5.1 Build 2600 Service Pack 1 Path: C:\cygwin\usr\local\bin C:\cygwin\bin C:\cygwin\bin C:\cygwin\usr\X11R6\bin C:\cygwin\bin c:\WINDOWS\system32 c:\WINDOWS c:\WINDOWS\System32\Wbem c:\Program Files\Common Files\Adaptec Shared\System c:\Program Files\Rational\common Output from C:\cygwin\bin\id.exe (nontsec) UID: 1006(sthoenna) GID: 513(None) 513(None) Output from C:\cygwin\bin\id.exe (ntsec) UID: 1006(sthoenna) GID: 513(None) 513(None)544(Administrators) 545(Users) SysDir: C:\WINDOWS\System32 WinDir: C:\WINDOWS CYGWIN = `tty ntsec title server' HOME = `C:\cygwin\home\sthoenna' MAKE_MODE = `unix' PWD = `/home/sthoenna' USER = `sthoenna' ALLUSERSPROFILE = `C:\Documents and Settings\All Users' APPDATA = `C:\Documents and Settings\sthoenna\Application Data' COMMONPROGRAMFILES = `C:\Program Files\Common Files' COMPUTERNAME = `DHX98431' COMSPEC = `C:\WINDOWS\system32\cmd.exe' CVS_RSH = `/bin/ssh' HOMEDRIVE = `C:' HOMEPATH = `\Documents and Settings\sthoenna' HOSTNAME = `DHX98431' INFOPATH = `/usr/local/info:/usr/info:/usr/share/info:/usr/autotool/devel/info:/usr/autotool/stable/info:' LESS = `-isrR' LOGONSERVER = `\\DHX98431' MANPATH = `/usr/local/man:/usr/man:/usr/share/man:/usr/autotool/devel/man:' NUMBER_OF_PROCESSORS = `1' OLDPWD = `/home/sthoenna/pbed/try' OS = `Windows_NT' PATHEXT = `.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH' PROCESSOR_ARCHITECTURE = `x86' PROCESSOR_IDENTIFIER = `x86 Family 15 Model 2 Stepping 9, GenuineIntel' PROCESSOR_LEVEL = `15' PROCESSOR_REVISION = `0209' PROGRAMFILES = `C:\Program Files' PROMPT = `$P$G' PS1 = `\[\033]0;\w\007 [EMAIL PROTECTED] \[\033[33m\w\033[0m\] $ ' SESSIONNAME = `Console' SHLVL = `1' SYSTEMDRIVE = `C:' SYSTEMROOT = `C:\WINDOWS' TEMP = `c:\DOCUME~1\sthoenna\LOCALS~1\Temp' TERM = `cygwin' TMP = `c:\DOCUME~1\sthoenna\LOCALS~1\Temp' USERDOMAIN = `DHX98431' USERNAME = `sthoenna' USERPROFILE = `C:\Documents and Settings\sthoenna' WINDIR = `C:\WINDOWS' _ = `/usr/bin/cygcheck' POSIXLY_CORRECT = `1' HKEY_CURRENT_USER\Software\Cygnus Solutions HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\mounts v2 (default) = `/cygdrive' cygdrive flags = 0x0022 HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\Program Options HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2 (default) = `/cygdrive' cygdrive flags = 0x0022 HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/ (default) = `C:\cygwin' flags = 0x000a HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/dl/tmp/perl-4.036 (default) = `c:\cygwin\dl\tmp\perl-4.036' flags = 0x080a HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/home/sthoenna/pbed/managed (default) = `c:\cygwin\home\sthoenna\pbed\managed' flags = 0x080a HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/home/sthoenna/pbed/textmount (default) = `c:\cygwin\home\sthoenna\pbed\textmount' flags = 0x0008 HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/usr/bin (default) = `C:\cygwin/bin' flags = 0x000a HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/usr/lib (default) = `C:\cygwin/lib' flags = 0x000a HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\mounts v2\/usr/X11R6/lib/X11/fonts (default) = `C:\cygwin\usr\X11R6\lib\X11\fonts' flags = 0x000a HKEY_LOCAL_MACHINE\SOFTWARE\Cygnus Solutions\Cygwin\Program Options c: hd NTFS 28568Mb 55% CP CS UN PA FC d: cd CDFS 67Mb 100%CS UN STL 05-15-2004 . /cygdrive user binmode,cygdrive C:\cygwin / system binmode c:\cygwin\dl\tmp\perl-4.036 /dl/tmp/perl-4.036 system binmode c:\cygwin\home\sthoenna\pbed\managed/home/sthoenna/pbed/managedsystem binmode
Re: 20040422 snapshot and filenames ending with . (was: Re: errors using DB_File perl module under 20040422 snapshot)
On Tue, Apr 27, 2004 at 02:45:38AM -0700, Yitzchak Scott-Thoennes wrote: you can create files ending with a . which are not stat'able unless another file exists with the same name without the dot. (and even then rm on the dotted name removes the non-dotted file): $ cat foo. bar ^D $ cat foo cat: foo: No such file or directory $ cat foo. bar $ ls -l foo. ls: foo.: No such file or directory $ ls -l foo ls: foo: No such file or directory $ catfoo baz ^D $ ls -l foo. -rw-r--r--1 sthoenna None5 Apr 27 01:43 foo. $ ls -l foo -rw-r--r--1 sthoenna None5 Apr 27 01:44 foo $ cat foo. bar $ cat foo baz $ rm foo. $ ls -l foo* ls: foo.: No such file or directory $ cmd /c 'dir foo*' Volume in drive C has no label. Volume Serial Number is 7422-45EE Directory of C:\cygwin\home\sthoenna 04/27/2004 01:43 AM 5 foo. 1 File(s) 5 bytes 0 Dir(s) 13,487,009,792 bytes free Would it help to try multiple snapshots and see where this broke? I already had the 20040416 and 20040420 ones and they both fail (but 1.5.9 is ok)? I was really hoping for a already fixed in CVS response :), is there anything I can do to help track this down? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: 20040422 snapshot and filenames ending with . (was: Re: errors using DB_File perl module under 20040422 snapshot)
On Tue, Apr 27, 2004 at 12:04:14PM +0200, Gerrit P. Haase wrote: Yitzchak wrote: I have still an older snapshot version of cygwin running here: $ uname -r 1.5.8s(0.110/4/2) And you get the ENOENT error? When issuing the perl command, yes. BTW, this happens also with cygwin-1.5.5 in the same way: $ uname -r 1.5.5(0.94/3/2) [EMAIL PROTECTED] ~ $ perl -we'use Fcntl; use DB_File; tie %h, DB_File, bugaboo,O_RDWR()|O_CREAT(), 0666 or die error: $!' Name main::h used only once: possible typo at -e line 1. error: No such file or directory at -e line 1. $ ls *bugaboo* __db.bugaboo This is on FAT, where DB_File has known problems, right? Works for me on NTFS for any cygwin release dll 1.5.5 through 1.5.9. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: 20040422 snapshot and filenames ending with . (was: Re: errors using DB_File perl module under 20040422 snapshot)
On Wed, Apr 28, 2004 at 12:02:21PM -0700, Yitzchak Scott-Thoennes wrote: On Tue, Apr 27, 2004 at 02:45:38AM -0700, Yitzchak Scott-Thoennes wrote: you can create files ending with a . which are not stat'able unless another file exists with the same name without the dot. (and even then rm on the dotted name removes the non-dotted file): $ cat foo. bar ^D $ ls -l foo. ls: foo.: No such file or directory $ ls -l foo ls: foo: No such file or directory Would it help to try multiple snapshots and see where this broke? I already had the 20040416 and 20040420 ones and they both fail (but 1.5.9 is ok)? I was really hoping for a already fixed in CVS response :), is there anything I can do to help track this down? This is fixed by the 20040430 snapshot; thanks, Corinna (though I wonder if it would be possible to consistently allow filenames to end in a . on ntfs, since it seems you can actually create separate foo and foo.). -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: GPL violation ?
On Thu, May 06, 2004 at 01:50:39PM -0400, Robb, Sam [EMAIL PROTECTED] wrote: You do get a T-shirt for filing the assignment, which is intended to cover all of your subsequent contributions. Uh-oh, Igor... you let the cat out of the bag. The goodwill of the community is only worth so much, but a free T-shirt... now, that's *swag*. cgf let it out a little while ago, also. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: How to force Setup.exe to behave?
On Sun, May 16, 2004 at 04:20:34PM -0700, Lex Ein [EMAIL PROTECTED] wrote: Nine days ago, this was Using Setup.exe to uninstall -bogus. Please help me figure out how to get Setup.exe to STOP TRYING TO INSTALL X. Setup.exe won't stop trying to snag unnecessary components. What file(s) to I have to edit manually to make this happen? What can I manually delete? It's a remote system, so I can't uninstall Cygwin without losing control. Little help? On Fri, 07 May 2004 19:16:54 -0700, Lex Ein [EMAIL PROTECTED] said: If there exists a sensible method for trimming the bulk of an existing Cygwin installation on a running disk-limited REMOTE system accessible only via ssh, I'd love to know it. I've completely lost patience. On a remote system, I've been trying to uninstall large chunks of Cygwin (x, db). Using Setup.exe to do this is like playing Whack-a-Mole, and is extremely frustrating. The Keep | Reinstall | Source | Version1 | Version2...| Uninstall click-to-rotate control forces dependancies both far and near while I'm trying to _reach_ Uninstall. Uninstall LEAVES THOSE DEPENDANCIES ACTIVATED. Without a way to select Uninstall without clocking through all the other dependancy activators, it's impossible to remove features and all their dependancies stickily. In fact, the very process of trying to do this piece by piece has resulted in masses of other crap being installed which I don't need. It's rather disturbing, as I'm running out of space on the remote system. Toggle to the Partial view and turn off the dependancies you don't want? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: write hangs on socket shut down for writing
On Wed, Apr 21, 2004 at 11:50:59AM +0200, Corinna Vinschen [EMAIL PROTECTED] wrote: On Apr 21 02:36, Yitzchak Scott-Thoennes wrote: On Wed, Apr 21, 2004 at 10:46:59AM +0200, Corinna Vinschen [EMAIL PROTECTED] wrote: Did you try it with a recent Cygwin version? I used your above test application and I'm getting a EPIPE (resp. a SIGPIPE) as expected. I think I had heard that others were not seeing this failure. I'm attaching an strace from my test program in case that may be helpful. Is there anything you can think of that might make a difference? Would you mind to debug that by yourself? You are seeing this problem so I guess you're the best person to debug that problem. Setting a breakpoint to 'fhandler_socket::sendmsg(msghdr const*, int, int)' should be fine. I can't see any reason for Cygwin to do that. There's no loop anywhere which could run endlessly. So either WSASendTo itself hangs, which is pretty unlikely for overlapped sockets, or WSASendTo returns SOCKET_ERROR and WSAGetLastError returns WSA_IO_PENDING so that WSAWaitForMultipleEvents in wsock_event::wait hangs. Any chance that a firewall or anti-virus software is interfering? I never had a chance to debug this, but upgrading to the 20040520 snapshot (from the 20040430) fixed this (as well as a hang using perl Net::Ping to ping 127.0.0.1 via udp that I was seeing and others apparently weren't.) Both were probably dependent on firewall (McAfee) or vpn software (Check Point VPN-1 SecureClient NG) I'm running, but seem to be better now. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: g++
On Mon, Jun 07, 2004 at 06:23:57PM -0600, Ian Strascina wrote: I updated my cygwin last Thurs. (6/3/04) to include the Perl and Python packages... Today, I'm trying to compile my c++ programs, and g++ is gone from my /usr/bin ... Anyone know why this would happen...??? Now I want to reinstall it, but it's not on the cygwin packages page... The closest I find is gcc-g++ ... Is this the same and will/should it work correctly with my current programs...??? Please help ASAP... I'm trying to graduate in a month and I need my g++ to work... Yes. gcc was split into multiple packages, and unfortunately you don't automatically get all the functionality you had before when you update. You want gcc-g++. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: nice not setting above/below normal
On Fri, Jun 11, 2004 at 05:42:38AM -0700, Brian Dessent wrote: reserved for real-time processes. The remaining range 1-15 are the regular (dynamic) priorities that most processes run with. In reality you don't set the priority directly this way, rather you choose a priority class (realtime, high, normal, idle; corresponding to 24, 13, 8, 4) and then a modifier (highest, above normal, normal, below normal, lowest; corresponding to +2, +1, 0, -1, -2). Is that correct? Shouldn't idle be 3 to allow the full 1-15 range? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ODBC with PostgreSQL
On Sun, Jun 13, 2004 at 02:04:01AM -0400, Barry wrote: I have ActivePerl installed along with Cygwin's Perl. When I configured ActivePerl, I had .pl files be associated with it, and I'm not sure what Perl is used when I paste perl path in Cygwin, or if there's a conflict, or if my error is caused by Cygwin's version of DBD::ODBC not working for Windows XP. perl -V will show osname=cygwin for cygwin and osname=MSWin32 for ActiveState (among other differences). Depends which is in your path first, and from your cygcheck output it looks like this would be the cygwin perl. No need for them to be pre-built. Get them from CPAN. Have a look at the DBI mail list archives or these mail archives for any further information. I have a feeling that information is out of date. Do I have to build DBD::ODBC? AFAIK, yes. Though it looks as if you already have (if in fact you are running the cygwin perl). I'm a little confused as to how you could have modules but not know you installed them. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: grep problems
On Mon, Jun 14, 2004 at 10:29:13AM +0200, J?rg Schaible wrote: I have upgraded to latest release of Cygwin and this has not changed things. I have tried the above on a Windows 98 and Windows XP system as well - same results. Any ideas how I get around this problem ? Learn the syntax. Backslashes have to be escaped by a backslash ;-) Looks like a cmd.exe shell to me, so no they don't. Also looks like a non-cygwin grep; at least I can't get grep to say GREP:. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: jhead question/problem
On Mon, Jun 14, 2004 at 10:02:15AM -0500, Jason Dufair wrote: If I execute it with the full pathname, however, it seems to have trouble: [EMAIL PROTECTED] /tmp $ jhead /tmp/p660.jpg Error : No such file in file '\tmp\p660.jpg' Just for kicks, try: jhead /cygwin/tmp/p660.jpg :) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Carriage Returns
On Fri, Jun 18, 2004 at 01:17:07PM -0700, Chris Carlson wrote: As defined in C, there is a concept of end-of-line character. I was under the impression that Perl understood this concept, too. It just so happens that M$ doesn't have one end-of-line character, but a pair of them. In some scripting languages, the carriage-return/line-feed pair is considered one end-of-line character and removing one removes both. It would be logical to assume that a function that removes the end-of-line character on a M$ box would remove both the carriage-return and the line-feed. I think this is what Mr. Kramer is trying to say. Perl converts native line endings to \n for text files, based on platform. The problem is that cygwin doesn't have one true line ending. IIRC, Gerrit went through a lot of conniptions getting it to work well everywhere to the degree possible, but there remain cases where you need to use the PERLIO=crlf environment variable, or take other action (there are many ways) to get it working ok. If the original poster really thinks something needs to change, he'd better include cygcheck -s -r -v output as an attachment, and describe explicitly the case that is a problem. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
