Re: subversion authentication broken in version 1.8, works with 1.7.10
(Sorry, not subscribed so replying to my own message instead) Yes, the problem is likely within svn itself, or the new serf http/s thingy. Discussion on #svn mentioned issues:: http://svn.haxx.se/users/archive-2013-06/0313.shtml http://svn.haxx.se/dev/archive-2013-06/0413.shtml These should be fixed in upcoming svn 1.8.1 and serf version 1.3.0. There does not seem to be debug options in svn to get https details of the failures. But 1.7.10 works so I'll stick to that. -Mikko -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: setup ChangeLog win32.h
On 2 July 2013 18:10, wrote:
>
> CVSROOT:/cvs/cygwin-apps
> Module name:setup
> Changes by: 2013-07-02 16:10:48
>
> Modified files:
> . : ChangeLog win32.h
>
> Log message:
> * win32.h (struct acl_t): Make sure struct is 4 byte aligned.
>
> Patches:
> http://sourceware.org/cgi-bin/cvsweb.cgi/setup/ChangeLog.diff?cvsroot=cygwin-apps&r1=2.807&r2=2.808
> http://sourceware.org/cgi-bin/cvsweb.cgi/setup/win32.h.diff?cvsroot=cygwin-apps&r1=2.29&r2=2.30
>
Why not using the union trick?
struct acl_t {
union {
LONG __align; /* Make sure &acl is 4-byte aligned. */
ACL acl;
};
char aclbuf[TOKEN_ACL_SIZE (7)];
};
Same effect, AFAIK, and no size overhead.
--
VZ
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Updated Git build
On Wed, Jun 19, 2013 at 4:05 AM, Steven Penny wrote: > Git 1.7.10 is now over one year old Until this is released on primary mirrors Cygwinports can work wget downloads.sf.net/cygwin-ports/git-1.8.3.1-1.tar.bz2 tar xf git-1.8.3.1-1.tar.bz2 -C/ -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: [ANNOUNCEMENT] Updated: screen 4.1.0-20130513-1
Andrew Schulman writes: > Hm. Thanks for pointing this out. It isn't anything I meant to do. > Looking now in the screen source, I see that there are two sample screenrc > files: etcscreenrc, and screenrc. etcscreenrc is supposed to be a sample > /etc/screenrc, and screenrc is supposed to be a sample ~/.screenrc. The > diff is at http://www.diffnow.com/?report=iz5ex . > > Neither of those files has exactly the termcapinfo line you cited, but > screenrc seems to have most of the capabilities in it, in xterm lines. This is not what I see in the distro. Both screen-4.0.3-7.tar.bz2 and screen-4.1.0-20130513.tar.bz2 have only etc/default/etc/screenrc and install that in /etc and the diff between these two is as I sent it: diff -bw [4.0.3/]etc/defaults/etc/screenrc [4.1.0/]etc/defaults/etc/screenrc 12d11 < defflow off # leave this off, so we can save in *emacs 68,70d66 < < # Set the hardstatus prop on gui terms to set the titlebar/icon title < termcapinfo xterm*|rxvt*|kterm*|Eterm*|cygwin hs:ts=\E]0;:fs=\007:ds=\E]0;\007 > I don't mind installing screenrc instead of etcscreenrc as the global > /etc/screenrc, if it's got better functionality in it OOTB. I'd appreciate > it if you'd test it first though, especially with mintty, and let me know > what you find. I'll do the same, but I override most of that stuff in my > ~/.screenrc, so it may not make much difference for me. I don't think the .screenrc sample I see in the diff you pointed to should be the /etc/screenrc. But it's not in the distro in any case, and the relevant line termcapinfo * '' 'hs:ts=\E_:fs=\E\\:ds=\E_\E\\' doesn't work for me if I try it in place of the one I show above from 4.0.3 So, it's up to you whether you add the sample .screenrc to the distro, but I still think the right thing to do is just to revert the deletion of the termcapinfo line from /etc/default/etc/screenrc Any other users out there care to contribute their experience? ht -- Henry S. Thompson, School of Informatics, University of Edinburgh 10 Crichton Street, Edinburgh EH8 9AB, SCOTLAND -- (44) 131 650-4440 Fax: (44) 131 650-4587, e-mail: h...@inf.ed.ac.uk URL: http://www.ltg.ed.ac.uk/~ht/ [mail from me _always_ has a .sig like this -- mail without it is forged spam] -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: setup ChangeLog win32.h
On Jul 3 10:42, Václav Zeman wrote:
> On 2 July 2013 18:10, wrote:
> >
> > CVSROOT:/cvs/cygwin-apps
> > Module name:setup
> > Changes by: 2013-07-02 16:10:48
> >
> > Modified files:
> > . : ChangeLog win32.h
> >
> > Log message:
> > * win32.h (struct acl_t): Make sure struct is 4 byte aligned.
> >
> > Patches:
> > http://sourceware.org/cgi-bin/cvsweb.cgi/setup/ChangeLog.diff?cvsroot=cygwin-apps&r1=2.807&r2=2.808
> > http://sourceware.org/cgi-bin/cvsweb.cgi/setup/win32.h.diff?cvsroot=cygwin-apps&r1=2.29&r2=2.30
> >
> Why not using the union trick?
>
> struct acl_t {
> union {
> LONG __align; /* Make sure &acl is 4-byte aligned. */
> ACL acl;
> };
> char aclbuf[TOKEN_ACL_SIZE (7)];
> };
>
> Same effect, AFAIK, and no size overhead.
Yep, done.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Maintainer cygwin AT cygwin DOT com
Red Hat
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Bug with Cygwin's 'quilt' is actually in 'patch'
On Jul 3 05:03, Peter B. wrote: > On Jun 20 09:46, Corinna Vinschen wrote: > > I'll build a new patch 2.7.1 for 32 bit today. I hope that fixes it > > for 32 bit as well. > > The new 2.7.1 (32bit) brakes CRLF handling for me completely. The old > 2.6.1 works properly. 2.7.1 with text files converted from CRLF to LF > works properly as well. > > Can you take a look at the download link below? It includes an example > file (with CRLF), the corresponding patch (LF) and the two binaries > (patch v2.6.1 and v2.7.1). > > http://www60.zippyshare.com/v/13352963/file.html Please provide a direct link which doesn't require to enable scripting in a web browser to download. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Bug with Cygwin's 'quilt' is actually in 'patch'
On 03.07.2013 14:33 Corinna Vinschen wrote: > On Jul 3 05:03, Peter B. wrote: >> On Jun 20 09:46, Corinna Vinschen wrote: >>> I'll build a new patch 2.7.1 for 32 bit today. I hope that fixes it >>> for 32 bit as well. >> The new 2.7.1 (32bit) brakes CRLF handling for me completely. The old >> 2.6.1 works properly. 2.7.1 with text files converted from CRLF to LF >> works properly as well. >> >> Can you take a look at the download link below? It includes an example >> file (with CRLF), the corresponding patch (LF) and the two binaries >> (patch v2.6.1 and v2.7.1). >> >> http://www60.zippyshare.com/v/13352963/file.html > Please provide a direct link which doesn't require to enable scripting > in a web browser to download. > > > Thanks, > Corinna > Please download this picture (jpg) and rename the file suffix to 7z or open with corresponding archiver directly. http://abload.de/img/patchfus75.jpg (99KB) -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Bug with Cygwin's 'quilt' is actually in 'patch'
On Jul 3 15:04, Peter B. wrote: > > On 03.07.2013 14:33 Corinna Vinschen wrote: > > On Jul 3 05:03, Peter B. wrote: > >> On Jun 20 09:46, Corinna Vinschen wrote: > >>> I'll build a new patch 2.7.1 for 32 bit today. I hope that fixes it > >>> for 32 bit as well. > >> The new 2.7.1 (32bit) brakes CRLF handling for me completely. The old > >> 2.6.1 works properly. 2.7.1 with text files converted from CRLF to LF > >> works properly as well. > >> > >> Can you take a look at the download link below? It includes an example > >> file (with CRLF), the corresponding patch (LF) and the two binaries > >> (patch v2.6.1 and v2.7.1). > >> > >> http://www60.zippyshare.com/v/13352963/file.html > > Please provide a direct link which doesn't require to enable scripting > > in a web browser to download. > > > > > > Thanks, > > Corinna > > > Please download this picture (jpg) and rename the file suffix to 7z or > open with corresponding archiver directly. > > http://abload.de/img/patchfus75.jpg (99KB) Thanks. The behaviour here is a deliberate upstream choice to apply the patch only if the line endings are the same as in the file to patch. If you really must apply non-CRLF patches to CRLF files, consider using a temporary text mount: http://cygwin.com/cygwin-ug-net/using-utils.html#mount Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Bug with Cygwin's 'quilt' is actually in 'patch'
On 03.07.2013 15:30 Corinna Vinschen wrote: > On Jul 3 15:04, Peter B. wrote: >> On 03.07.2013 14:33 Corinna Vinschen wrote: >>> On Jul 3 05:03, Peter B. wrote: On Jun 20 09:46, Corinna Vinschen wrote: > I'll build a new patch 2.7.1 for 32 bit today. I hope that fixes it > for 32 bit as well. The new 2.7.1 (32bit) brakes CRLF handling for me completely. The old 2.6.1 works properly. 2.7.1 with text files converted from CRLF to LF works properly as well. Can you take a look at the download link below? It includes an example file (with CRLF), the corresponding patch (LF) and the two binaries (patch v2.6.1 and v2.7.1). http://www60.zippyshare.com/v/13352963/file.html >>> Please provide a direct link which doesn't require to enable scripting >>> in a web browser to download. >>> >>> >>> Thanks, >>> Corinna >>> >> Please download this picture (jpg) and rename the file suffix to 7z or >> open with corresponding archiver directly. >> >> http://abload.de/img/patchfus75.jpg (99KB) > Thanks. The behaviour here is a deliberate upstream choice to apply the > patch only if the line endings are the same as in the file to patch. > > If you really must apply non-CRLF patches to CRLF files, consider using > a temporary text mount: > http://cygwin.com/cygwin-ug-net/using-utils.html#mount > > > Corinna > Sounds like intended behavior. I will try to tackle the problem otherwise then. The culprit here is the JavaVM that produces different EOLs on different platforms (running a binary decompiler -> txt). Thanks. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
octave-3.4.0.exe: error while loading shared libraries
Hello there, I'm sorry to disturb but as an absolute beginner with Cygwin I couldn't find the correct answer to this problem, even if I found some interesting clues from previous answer in mailing list (http: //cygwin.com/ml/cygwin/2011-04/msg00254.html). I've sent this email at cygwin-apps at first, but I'm afraid it's the wrong list, please excuse me and ignore it. This is my very first installation (default + math package; Windows 7 machine), I did not performed any change or customization but I tried to re-install the whole a couple of times to recover, unsuccessfully. As the former reporter said in aforementioned question, I get this error: gennarif@NIDORAN ~ $ octave /usr/bin/octave- 3.6.4.exe: error while loading shared libraries: ?: cannot open shared object file: No such file or directory Then I performed the suggested check but got different output gennarif@NIDORAN ~ $ cygcheck octave- 3.4.0.exe cygcheck: could not find 'octave-3.4.0.exe' gennarif@NIDORAN ~ $ cygcheck -l liblapack0 /etc/profile.d/lapack0.csh /etc/profile. d/lapack0.sh /usr/lib/lapack/cygblas-0.dll /usr/lib/lapack/cyglapack-0. dll That's really strange because octave DO exist, but in this path (which has not the /usr/ parent folder): ...\cygwin\bin\octave-3.6.4. exe As so the the dll: ...\cygwin\lib\lapack\cygblas-0.dll ... \cygwin\lib\lapack\cyglapack-0.dll Really, I can't figure out what's on earth is going on, what I'm doing wrong? (well, actually I really did do nothing at all... ) Am I missing some package, setting or else? As already said, I just did the automatic installation from a directory after having downloaded the files via the setup.exe dialog GUI. thanks in advance, Federico Invita i tuoi amici e Tiscali ti premia! Il consiglio di un amico vale più di uno spot in TV. Per ogni nuovo abbonato 30 € di premio per te e per lui! Un amico al mese e parli e navighi sempre gratis: http://freelosophy.tiscali.it/ -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
MSYS mode (continue)
I want to continue our discussion about integrating MSYS mode into Cygwin sources. Right now we need to determine where you can create hooks for msys plugin I think. MSYS plugin need to has ability to change next logic: - generating osname - reading /etc/fstab - changing environment variables and filtering them, that passed to non-cygwin applications - changing program arguments that passed to non-cygwin applications Regards, Alexey. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: octave-3.4.0.exe: error while loading shared libraries
Il 7/3/2013 5:41 PM, gennari.federico ha scritto: Hello there, wrong mailing list. Please use cygwin(at)cygwin.com for this type of issue, for follow ups I'm sorry to disturb but as an absolute beginner with Cygwin I couldn't find the correct answer to this problem, even if I found some interesting clues from previous answer in mailing list (http: //cygwin.com/ml/cygwin/2011-04/msg00254.html) This is my very first installation (default + math package; Windows 7 machine), I did not performed any change or customization but I tried to re-install the whole a couple of times to recover, unsuccessfully. As the former reporter said in aforementioned question, I get this error: gennarif@NIDORAN ~ $ octave /usr/bin/octave-3.6.4.exe: error while loading shared libraries: ?: cannot open shared object file: No such file or directory Then I performed the suggested check but got different output gennarif@NIDORAN ~ $ cygcheck octave-3.4.0.exe cygcheck: could not find 'octave-3.4.0.exe' gennarif@NIDORAN ~ $ cygcheck -l liblapack0 /etc/profile.d/lapack0.csh /etc/profile.d/lapack0.sh /usr/lib/lapack/cygblas-0.dll /usr/lib/lapack/cyglapack-0. dll That's really strange because Octave package DO exist in this path (which has not the /usr/ parent folder): ...\cygwin\bin\octave- 3.6.4.exe /usr/bin is mounted on /bin (same content) As so the the dll: ...\cygwin\lib\lapack\cygblas-0.dll ... \cygwin\lib\lapack\cyglapack-0.dll I will bet "/usr/lib/lapack" is missing from your PATH see "/etc/profile.d/lapack0.sh" content thanks in advance, Federico If you have still issue, could you as reported on http://cygwin.com/problems.html "Run cygcheck -s -v -r > cygcheck.out and include that file as an attachment in your report. Please do not compress or otherwise encode the output. Just attach it as a straight text file so that it can be easily viewed. " Regards Marco -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
R: Re: octave-3.4.0.exe: error while loading shared libraries
Thanks Marco for both your your patience and prompt answer, here's some
additional info.
The directory list under \cygwin\usr is:
i686-pc-
cygwin
include
local
sbin
share
src
ssl
tmp
And is not including /bin.
The only one I found is /cygwin/bin, as it has been created right under
the root installation directory and not under /usr.
Reading the
lapack0.sh content I found that:
LA_PREFIX=/usr
LA_LIBDIR=${LA_PREFIX}
/lib
LA_BINDIR=${LA_LIBDIR}/lapack
So "/usr/lib/lapack" is actually
NOT missing from PATH, but correct path should not have /usr as prefix
in my particular case, isn't it?
So strange since it's default
directory structure created by setup.exe (during installation I chose C:
\Users\gennarif\Documents\programmi\cygwin\ as root directory).
Add to
this, trying to produce the requested file, I got this error:
cygcheck
-s -v -r > cygcheck.out
cygcheck: Wrong architecture. Only ix86
executables supported.
Dis some mis-installation happen?
Thanks &
regards,
Federico
Invita i tuoi amici e Tiscali ti premia! Il consiglio di un amico vale più di
uno spot in TV. Per ogni nuovo abbonato 30 € di premio per te e per lui! Un
amico al mese e parli e navighi sempre gratis: http://freelosophy.tiscali.it/
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: R: Re: octave-3.4.0.exe: error while loading shared libraries
Il 7/3/2013 6:38 PM, gennari.feder...@tiscali.it ha scritto:
Thanks Marco for both your your patience and prompt answer, here's some
additional info.
The directory list under \cygwin\usr is:
i686-pc-cygwin
include
local
sbin
share
src
ssl
tmp
You misunderstood; try
ls /usr/bin
ls /bin
from within cygwin terminal or xterm
They should provide the same "long" file list
And is not including /bin. The only one I found is /cygwin/bin, as it
has been created right under the root installation directory and not
under /usr.
Reading the lapack0.sh content I found that:
LA_PREFIX=/usr
LA_LIBDIR=${LA_PREFIX}/lib
LA_BINDIR=${LA_LIBDIR}/lapack
So "/usr/lib/lapack" is actually NOT missing from PATH, but correct path
should not have /usr as prefix in my particular case, isn't it?
So strange since it's default directory structure created by setup.exe
(during installation I chose
C:\Users\gennarif\Documents\programmi\cygwin\ as root directory).
what is the output of
echo $PATH
Add to this, trying to produce the requested file, I got this error:
cygcheck -s -v -r > cygcheck.out
cygcheck: Wrong architecture. Only ix86 executables supported.
> Dis some mis-installation happen?
no. Someone else also reported that warning, but I ignore the root cause.
Could you attach the cygcheck.out file that was created ?
Thanks & regards,
Federico
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: [ANNOUNCEMENT] Updated: screen 4.1.0-20130513-1
> This is not what I see in the distro. Both >screen-4.0.3-7.tar.bz2 > and >screen-4.1.0-20130513.tar.bz2 > > have only etc/default/etc/screenrc > > and install that in /etc > > and the diff between these two is as I sent it: > > diff -bw [4.0.3/]etc/defaults/etc/screenrc [4.1.0/]etc/defaults/etc/screenrc > 12d11 > < defflow off # leave this off, so we can save in *emacs > 68,70d66 > < > < # Set the hardstatus prop on gui terms to set the titlebar/icon > title > < termcapinfo xterm*|rxvt*|kterm*|Eterm*|cygwin > hs:ts=\E]0;:fs=\007:ds=\E]0;\007 OK, I see now. The line you cited was present in /etc/screenrc in 4.0.3, because I had added a patch way back when to put it there. That patch got left out by mistake in the latest release. So, no problem, I'll issue a new release with the missing line restored. Andrew -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
On Wed, Jul 03, 2013 at 08:03:25PM +0400, Alexey Pavlov wrote: >I want to continue our discussion about integrating MSYS mode into >Cygwin sources. >Right now we need to determine where you can create hooks for msys >plugin I think. >MSYS plugin need to has ability to change next logic: > > - generating osname > - reading /etc/fstab > - changing environment variables and filtering them, that passed to >non-cygwin applications > - changing program arguments that passed to non-cygwin applications That all sounds doable but could you refresh my memory on why you need an /etc/fstab hook? Was it to map /cygdrive/c -> /c or something? If so, as long as you're going to have an fstab why not just do that in the fstab itself? cgf -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
2013/7/3 Christopher Faylor: > On Wed, Jul 03, 2013 at 08:03:25PM +0400, Alexey Pavlov wrote: >>I want to continue our discussion about integrating MSYS mode into >>Cygwin sources. >>Right now we need to determine where you can create hooks for msys >>plugin I think. >>MSYS plugin need to has ability to change next logic: >> >> - generating osname >> - reading /etc/fstab >> - changing environment variables and filtering them, that passed to >>non-cygwin applications >> - changing program arguments that passed to non-cygwin applications > > That all sounds doable but could you refresh my memory on why you need > an /etc/fstab hook? Was it to map /cygdrive/c -> /c or something? > I need it to be able read short mount points from fstab: . All other options are set by default. With my changes I can have in /etc/fstab both types of mount points - cygwin-like and msys-like. > If so, as long as you're going to have an fstab why not just do that > in the fstab itself? I already add line to stab that drop cygdrive prefix. > > cgf > > -- > Problem reports: http://cygwin.com/problems.html > FAQ: http://cygwin.com/faq/ > Documentation: http://cygwin.com/docs.html > Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple > -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
CYGWIN_NT-6.1-WOW64: no `fc-list'?
usr/bin/bash: fc-list: command not found Is fc-list not available with Cygwin? I used to have an older (2007?), 32-bit version of Cygwin, and I was able to invoke `fc-list' without problem. What am I missing? SHELLis /bin/bash HOME is d:\usr\me. CYGWIN_HOME is c:\cygwin. CYGWINBINDIR is c:\cygwin\bin. These dirs are in my PATH, in this order: c:\cygwin\bin c:\cygwin\usr\sbin c:\cygwin\usr\local\bin Please cc me, as I am not subscribed to the list. Thx. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: emacs very slow; zombie processes eventually cleared
Ken Brown writes: > On 7/2/2013 11:41 AM, Rockefeller, Harry wrote: >> When I start emacs I get this message: >> (emacs:192): Gtk-CRITICAL **: gtk_distribute_natural_allocation: assertion >> `extra_space >= 0' failed > > Are you saying that this happens every time you start emacs? Does it > even happen when you start with 'emacs -Q'? If not, maybe you could > do some testing to figure out what in your initialization (including > X11 initialization) triggers it. This might help to pin down the bug. > > > I think I might have mentioned once before when you wrote about zombie > processes that there is a known problem in emacs-24.3 that can cause > problems with subprocesses. This is caused by race conditions between > emacs and glib. The problem has recently been fixed on the emacs > development trunk. If you'd like, I could build emacs from a snapshot > of the trunk and let you test it to see if some of your problems go > away. > > Ken Yes, please, I think I have been bouncing off of this myself lately... Dave -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: CYGWIN_NT-6.1-WOW64: no `fc-list'?
Il 7/3/2013 8:36 PM, Drew Adams ha scritto: usr/bin/bash: fc-list: command not found Is fc-list not available with Cygwin? I used to have an older (2007?), 32-bit version of Cygwin, and I was able to invoke `fc-list' without problem. What am I missing? SHELLis /bin/bash HOME is d:\usr\me. CYGWIN_HOME is c:\cygwin. CYGWINBINDIR is c:\cygwin\bin. These dirs are in my PATH, in this order: c:\cygwin\bin c:\cygwin\usr\sbin c:\cygwin\usr\local\bin Please cc me, as I am not subscribed to the list. Thx. see http://cygwin.com/packages/fontconfig/fontconfig-2.10.2-1 -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: setup ChangeLog win32.h
On Wed, Jul 03, 2013 at 02:30:10PM +0200, Corinna Vinschen wrote:
>On Jul 3 10:42, V??clav Zeman wrote:
>> On 2 July 2013 18:10, wrote:
>> >
>> > CVSROOT:/cvs/cygwin-apps
>> > Module name:setup
>> > Changes by: 2013-07-02 16:10:48
>> >
>> > Modified files:
>> > . : ChangeLog win32.h
>> >
>> > Log message:
>> > * win32.h (struct acl_t): Make sure struct is 4 byte aligned.
>> >
>> > Patches:
>> > http://sourceware.org/cgi-bin/cvsweb.cgi/setup/ChangeLog.diff?cvsroot=cygwin-apps&r1=2.807&r2=2.808
>> > http://sourceware.org/cgi-bin/cvsweb.cgi/setup/win32.h.diff?cvsroot=cygwin-apps&r1=2.29&r2=2.30
>> >
>> Why not using the union trick?
>>
>> struct acl_t {
>> union {
>> LONG __align; /* Make sure &acl is 4-byte aligned. */
>> ACL acl;
>> };
>> char aclbuf[TOKEN_ACL_SIZE (7)];
>> };
>>
>> Same effect, AFAIK, and no size overhead.
>
>Yep, done.
Shouldn't we really explicitly specify the alignment rather than just
assuming that LONG is aligned?
The below works.
cgf
RCS file: /cvs/cygwin-apps/setup/win32.h,v
retrieving revision 2.31
diff -d -u -p -r2.31 win32.h
--- win32.h 3 Jul 2013 12:30:04 - 2.31
+++ win32.h 3 Jul 2013 19:14:46 -
@@ -75,10 +75,7 @@
(cnt) * (sizeof (ACCESS_ALLOWED_ACE) +
MAX_SID_LEN))
struct acl_t {
- union {
-LONG __align; /* Make sure &acl is 4-byte aligned. */
-ACL acl;
- };
+ ACL acl __attribute__ ((aligned (4)));
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: setup ChangeLog win32.h
On Jul 3 15:15, Christopher Faylor wrote:
> On Wed, Jul 03, 2013 at 02:30:10PM +0200, Corinna Vinschen wrote:
> >On Jul 3 10:42, V??clav Zeman wrote:
> >> On 2 July 2013 18:10, wrote:
> >> >
> >> > CVSROOT:/cvs/cygwin-apps
> >> > Module name:setup
> >> > Changes by: 2013-07-02 16:10:48
> >> >
> >> > Modified files:
> >> > . : ChangeLog win32.h
> >> >
> >> > Log message:
> >> > * win32.h (struct acl_t): Make sure struct is 4 byte aligned.
> >> >
> >> > Patches:
> >> > http://sourceware.org/cgi-bin/cvsweb.cgi/setup/ChangeLog.diff?cvsroot=cygwin-apps&r1=2.807&r2=2.808
> >> > http://sourceware.org/cgi-bin/cvsweb.cgi/setup/win32.h.diff?cvsroot=cygwin-apps&r1=2.29&r2=2.30
> >> >
> >> Why not using the union trick?
> >>
> >> struct acl_t {
> >> union {
> >> LONG __align; /* Make sure &acl is 4-byte aligned. */
> >> ACL acl;
> >> };
> >> char aclbuf[TOKEN_ACL_SIZE (7)];
> >> };
> >>
> >> Same effect, AFAIK, and no size overhead.
> >
> >Yep, done.
>
> Shouldn't we really explicitly specify the alignment rather than just
> assuming that LONG is aligned?
>
> The below works.
>
> cgf
>
> RCS file: /cvs/cygwin-apps/setup/win32.h,v
> retrieving revision 2.31
> diff -d -u -p -r2.31 win32.h
> --- win32.h 3 Jul 2013 12:30:04 - 2.31
> +++ win32.h 3 Jul 2013 19:14:46 -
> @@ -75,10 +75,7 @@
>(cnt) * (sizeof (ACCESS_ALLOWED_ACE) +
> MAX_SID_LEN))
>
> struct acl_t {
> - union {
> -LONG __align;/* Make sure &acl is 4-byte aligned. */
> -ACL acl;
> - };
> + ACL acl __attribute__ ((aligned (4)));
That's even better. Applied.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Maintainer cygwin AT cygwin DOT com
Red Hat
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
On Jul 3 22:24, Alexey Pavlov wrote: > 2013/7/3 Christopher Faylor: > > On Wed, Jul 03, 2013 at 08:03:25PM +0400, Alexey Pavlov wrote: > >>I want to continue our discussion about integrating MSYS mode into > >>Cygwin sources. > >>Right now we need to determine where you can create hooks for msys > >>plugin I think. > >>MSYS plugin need to has ability to change next logic: > >> > >> - generating osname > >> - reading /etc/fstab > >> - changing environment variables and filtering them, that passed to > >>non-cygwin applications > >> - changing program arguments that passed to non-cygwin applications > > > > That all sounds doable but could you refresh my memory on why you need > > an /etc/fstab hook? Was it to map /cygdrive/c -> /c or something? > > > I need it to be able read short mount points from fstab: > . > All other options are set by default. With my changes I can have in > /etc/fstab both types of mount points - cygwin-like and msys-like. And why is that necessary? There are always default options set anyway, but a layout change of the fstab file simply makes no sense. There's no win at all. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: pty issue causes 'screen' to hang when run from mintty as detached
On Mon, Jun 17, 2013 at 09:58:35AM -0400, Christopher Faylor wrote: >On Mon, Jun 17, 2013 at 06:08:27AM -0400, Andrew Schulman wrote: >>> > Do we have any idea whether the problem is with mintty or with screen? >>> >>> I observed the same behaviour as the OP, but also reproduced it in >>> rxvt and xterm, so not a mintty issue. Since it works in a console, >>> the Cygwin pty layer has to be the main suspect, unless screen has >>> specfic code for Cygwin consoles or ptys? >> >>I don't think so, but there's one big, crufty old patch that I had to use to >>get >>screen to run a few years ago. >> >>The version of screen that we're using now in Cygwin, 4.0.3, is about 8 years >>old. There haven't been any official new releases since then, but development >>has continued and there have been a lot of improvements. I'm in the process >>of >>building an updated package from the latest git, and that version appears not >>to >>need any of the patches I used to have to use to get screen to compile and run >>in Cygwin. >> >>So, I suggest that we wait until the new screen test release comes out, >>within a >>few days, and see if that fixes the problem. > >I looked at this briefly over the weekend and saw a problem in Cygwin's >pty handling code. I don't know exactly what the problem is yet but there >is something not quite right there. Something is closing a windows handle >and that's causing a SIGHUP to be sent to the pty owner. That's not right. > >I'll fix this problem eventually but it will be interesting to see if a >newer screen behaves any better. This problem should be fixed in the upcoming snapshot. It was a longstanding bug in the pty code. Thanks for providing the succinct steps required to duplicate the problem. cgf -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
On 7/3/2013 3:50 PM, Corinna Vinschen wrote: On Jul 3 22:24, Alexey Pavlov wrote: 2013/7/3 Christopher Faylor: I need it to be able read short mount points from fstab: . All other options are set by default. With my changes I can have in /etc/fstab both types of mount points - cygwin-like and msys-like. And why is that necessary? There are always default options set anyway, but a layout change of the fstab file simply makes no sense. There's no win at all. I think Alexey is trying to replicate MSYS's behavior, so it becomes more of a drop-in replacement. Here's my MSYS /etc/fstab (MSYS doesn't yet support per-user fstab.d/username settings): C:/MinGW /mingw C:/Users/me/Desktop/desktop C:/Users/me/Documents /docs C:/msys-src/usr/src I agree with Corinna that there's no need to make cygwin-in-msys-mode EXACTLY like MSYS in all of its configuration settings (such as the syntax of /etc/fstab). Your MSYS-like cygwin installation would have its own /etc/fstab (in regular cygwin format), but perhaps your /home area is mapped differently in that fstab, than in the /etc/fstab in your "regular" cygwin installation. Or, like me, I have a separate msys-src and cygwin-src directory mapped onto MSYS's /usr/src and cygwin's /usr/src, respectively. Or is the proposal on the table for the same installation to be usable in both modes, depending on the nature of the launched application? -- Chuck -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: emacs very slow; zombie processes eventually cleared
On 7/3/2013 2:51 PM, J. David Boyd wrote: Ken Brown writes: On 7/2/2013 11:41 AM, Rockefeller, Harry wrote: When I start emacs I get this message: (emacs:192): Gtk-CRITICAL **: gtk_distribute_natural_allocation: assertion `extra_space >= 0' failed Are you saying that this happens every time you start emacs? Does it even happen when you start with 'emacs -Q'? If not, maybe you could do some testing to figure out what in your initialization (including X11 initialization) triggers it. This might help to pin down the bug. I think I might have mentioned once before when you wrote about zombie processes that there is a known problem in emacs-24.3 that can cause problems with subprocesses. This is caused by race conditions between emacs and glib. The problem has recently been fixed on the emacs development trunk. If you'd like, I could build emacs from a snapshot of the trunk and let you test it to see if some of your problems go away. Ken Yes, please, I think I have been bouncing off of this myself lately... OK, I've built bzr revision 113271 and uploaded it to my personal Cygwin repository: http://sanibeltranquility.com/cygwin/ There are instructions on that page for installing it. Please let me know how it works. I'm especially interested in knowing whether it solves some of the problems with zombie processes that have been reported. Ken -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
On Wed, Jul 03, 2013 at 05:03:04PM -0400, Charles Wilson wrote: >On 7/3/2013 3:50 PM, Corinna Vinschen wrote: >> On Jul 3 22:24, Alexey Pavlov wrote: >>> 2013/7/3 Christopher Faylor: >>> I need it to be able read short mount points from fstab: >>> . >>> All other options are set by default. With my changes I can have in >>> /etc/fstab both types of mount points - cygwin-like and msys-like. >> >> And why is that necessary? There are always default options set anyway, >> but a layout change of the fstab file simply makes no sense. There's no >> win at all. > >I think Alexey is trying to replicate MSYS's behavior, so it becomes >more of a drop-in replacement. Here's my MSYS /etc/fstab (MSYS doesn't >yet support per-user fstab.d/username settings): > >C:/MinGW /mingw >C:/Users/me/Desktop/desktop >C:/Users/me/Documents /docs >C:/msys-src/usr/src I think Cygwin fstab parsing could be loosened up to allow that syntax. >I agree with Corinna that there's no need to make cygwin-in-msys-mode >EXACTLY like MSYS in all of its configuration settings (such as the >syntax of /etc/fstab). > >Your MSYS-like cygwin installation would have its own /etc/fstab (in >regular cygwin format), but perhaps your /home area is mapped >differently in that fstab, than in the /etc/fstab in your "regular" >cygwin installation. Or, like me, I have a separate msys-src and >cygwin-src directory mapped onto MSYS's /usr/src and cygwin's /usr/src, >respectively. > >Or is the proposal on the table for the same installation to be usable >in both modes, depending on the nature of the launched application? That's an interesting idea but I don't think anyone was proposing that. Is there any reason why we'd need to add a hook to filename parsing? That would be an alternative to hooking fstab. cgf -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: lftp bug: mirror [-R] didn't work with sftp connection
Yes, I have same problem. This is the problem in version 4.4.7 of lftp - see http://lftp.yar.ru/news.html. Problem is solved in 4.4.8. As a workaround downgrade to 4.4.4, until cygwin adopts version 4.4.8. Vlado On 3.7.2013 7:31, Arthur Tu wrote: $ lftp sftp://usr@addr I can mget or mput files with sftp connection. However, when it comes to directory operation, lftp always failed. Both `mirror abc` to download whole directory or 'mirror -R abc' to upload whole directory didn't work. The former one stuck at 'mkdir abc' while the latter one stuck at 'cd abc'. So i guess it's some bugs of operations on local file system. Some more information on another two conditions. 1. `sftp usr@addr` I can use 'get -r' to download whole directory. 2. 'lftp someusr@another_addr'. When i used lftp to connect to normal ftp servers instead of sftp ones, `mirror` worked. Can someone reproduce this problem? -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
On 7/3/2013 5:57 PM, Christopher Faylor wrote: On Wed, Jul 03, 2013 at 05:03:04PM -0400, Charles Wilson wrote: Or is the proposal on the table for the same installation to be usable in both modes, depending on the nature of the launched application? That's an interesting idea but I don't think anyone was proposing that. Is there any reason why we'd need to add a hook to filename parsing? That would be an alternative to hooking fstab. Well, MSYS doesn't mind dos-style filenames, so the big-ugly-warn-once thing shouldn't happen... And if we ever do, really and truly, deprecate dos-style paths for cygwin /proper/, you'd need to hook the parsing in order to restore that functionality for cygwin-/msys/. -- Chuck -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Cygwin Installation Stops in the Middle
On 7/3/2013 8:22 PM, Fitzharris, Margaret A wrote: I was downloading the first Cygwin setup installation located at http://www.cygwin.com/install.html. I tried installing directly from the internet, using all the default settings, such as root directory: C:/Cygwin. However, the installation gets stuck whenever it starts installing something like, "/etc/postinstall/dbus.sh". I have turned off my firewall. Is there anything else that could be causing this problem? It's not installing anything at this point. It's running postinstall scripts, some of which can be time consuming. Maybe you just weren't patient enough. Did you by any chance choose to install more than the packages that are installed by default? (These are the packages in the Base category.) If so, I recommend that you start by just installing the Base packages. You can always re-run setup.exe later and choose additional packages. Ken -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: lftp bug: mirror [-R] didn't work with sftp connection
> Yes, I have same problem. > This is the problem in version 4.4.7 of lftp - see > http://lftp.yar.ru/news.html. > Problem is solved in 4.4.8. > As a workaround downgrade to 4.4.4, until cygwin adopts version 4.4.8. I'll go ahead and upload 4.4.8 then. I already have it built for i686, but I was waiting to see if I could get it built for x86_64 too. I haven't gotten that to work yet though, so I'll not wait for it. Since 4.4.7 is buggy and 4.4.4 is the fallback, I'll leave 4.4.4 as previous and remove 4.4.7. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: emacs very slow; zombie processes eventually cleared
[I'm including your private message to me below to avoid breaking the thread.] On 7/3/2013 5:33 PM, Rockefeller, Harry wrote: FWIW, my corporate email interceptor is broken. This never made it out to the list. I attempted to ask nicely to simply allow all emails from and to domain Cygwin.com, but they don't want to comply. On 7/2/2013 11:41 AM, Rockefeller, Harry wrote: When I start emacs I get this message: (emacs:192): Gtk-CRITICAL **: gtk_distribute_natural_allocation: assertion `extra_space >= 0' failed [snip] Ken asks: maybe you could do some testing to figure out what in your initialization (including X11 initialization) triggers it. This might help to pin down the bug. My first response: It appears to be related to version control 'vc' and maybe 'psvn' packages. I removed "(require 'cl)" and load-library "vc-svn" from my .emacs init file and this message went away. That's not the cause of the above GTk-Critical message. When I start emacs in dired mode on the current directory like this: "emacs -Q ." I don't get the Gtk message. But when I start emacs with my font of choice like this: "emacs --font 6x10 -Q ." I do get the Gtk message. I tried some other fonts: No Gtk message with "fixed", "rk14" and "10x20". Gtk message did happen with "6x9". I can reproduce this, but only when emacs is started in dired mode. And emacs seems to run OK in spite of the message, though I didn't test it much. I don't know if this has anything to do with slowness and/or temporary zombie processes. I don't know either. My build from the emacs trunk still gives that error message. It will be interesting to see whether you still have problems with zombie processes. Ken -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: lftp bug: mirror [-R] didn't work with sftp connection
> > Yes, I have same problem. > > This is the problem in version 4.4.7 of lftp - see > > http://lftp.yar.ru/news.html. > > Problem is solved in 4.4.8. > > As a workaround downgrade to 4.4.4, until cygwin adopts version 4.4.8. > > I'll go ahead and upload 4.4.8 then. I already have it built for i686, but > I was waiting to see if I could get it built for x86_64 too. I haven't > gotten that to work yet though, so I'll not wait for it. > > Since 4.4.7 is buggy and 4.4.4 is the fallback, I'll leave 4.4.4 as > previous and remove 4.4.7. Hm... now I'm not so sure. Someone else is reporting what looks like the same problem with 4.4.8: http://thread.gmane.org/gmane.network.lftp.user/2262 -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: MSYS mode (continue)
My opinion is to extend Cygwin parser to read msys-like mounts. It's
very simple:
--- mount.cc 2013-04-24 20:29:29.0 +0400
+++ Cygwin/winsup/cygwin/mount.cc 2013-06-13 09:35:01.479492100 +0400
@@ -1125,8 +1118,17 @@
if (!*c)
return true;
cend = find_ws (c);
- *cend = '\0';
posix_path = conv_fstab_spaces (c);
+ if (!*cend)
+ {
+ unsigned mount_flags = MOUNT_SYSTEM | MOUNT_BINARY |
MOUNT_NOPOSIX | MOUNT_NOACL;
+
+ int res = mount_table->add_item (native_path, posix_path, mount_flags);
+ if (res && get_errno () == EMFILE)
+ return false;
+ return true;
+ }
+ *cend = '\0';
/* Third field: FS type. */
c = skip_ws (cend + 1);
if (!*c)
Regards,
Alexey.
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Cygwin Installation Stops in Middle
On 07/04/2013 06:31 AM, Fitzharris, Margaret A wrote: > I have been waiting for hours and the installation is stuck at the > same point. I have tried both methods of installing all the files and > just the default files, and the installation process always gets > stuck at the same place. I have turned off the firewall and > anti-virus software on my computer. > > Is there anything else that I could be doing wrong? Are you sure that it is really stuck? When you are installing TexLive, things seemingly stop progressing for a long time because it takes a long time to install it and run all post installation scripts for it. -- VZ signature.asc Description: OpenPGP digital signature
Re: Cygwin Installation Stops in Middle
Il 7/4/2013 6:31 AM, Fitzharris, Margaret A ha scritto: I have been waiting for hours and the installation is stuck at the same point. I have tried both methods of installing all the files and just the default files, and the installation process always gets stuck at the same place. I have turned off the firewall and anti-virus software on my computer. Is there anything else that I could be doing wrong? Thanks, Margaret Fitzharris Margaret please follow up on the same thread. Opening a new one to discuss always the same issue make difficult to support you This script is not part of base /etc/postinstall/dbus.sh so you have residual of previous failed trials. Setup is trying to run all scripts/etc/postinstall/*.sh all completed are renamed/etc/postinstall/*.sh.done Remove the full cygwin install directory, and install only the Base category (that is the default) Regards Marco -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
