[ANNOUNCEMENT] New: steghide 0.5.1-1 -- A steganography hiding tool

[Jari Aalto]

To: cygwin-annou...@cygwin.com
Subject: New: package 0.5.1-1 -- A steganography hiding tool

PACKAGE DESCRIPTION
===

Homepage: http://freshmeat.net/projects/steghide
License : GPL

A steganography program which hides bits of a data file in some of the
least significant bits of another file in such a way that the
existence of the data file is not visible and cannot be proven.
Steghide is designed to be portable and configurable and features
hiding data in bmp, wav and au files, blowfish encryption, MD5 hashing
of passphrases to blowfish keys, and pseudo-random distribution of
hidden bits in the container data.

CHANGES SINCE LAST RELEASE
==

See homepage.

INSTALL OR UPGRADE NOTES


Standard install.

CYGWIN INSTALLATION INFORMATION
===

To install this package, click on the "Install Cygwin now" link on the
 web page. This downloads setup.exe to your
system. Then, run setup and answer all of the questions. You'll find
the package listed in the "All" category. After installation, read the
documentation at directories:

/usr/share/doc//*
/usr/share/doc/Cygwin/.README

If you have questions or comments, please send them to the Cygwin
mailing list at .

CYGWIN-ANNOUNCE UNSUBSCRIBE INFO


This message has been sent to cygwin-announce list.

If you want to unsubscribe from the 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:

cygwin-announce-unsubscribe-you=yourdomain@cygwin.com

More information on unsubscribing can be found:

http://sources.redhat.com/lists.html#unsubscribe-simple

Please read *all* of the information on unsubscribing that is available
starting at the above URL.

--
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_conv_path() documentation RFE

[Corinna Vinschen]

On Dec 17 19:29, Warren Young wrote:
> On this page
> 
>   http://cygwin.com/1.7/cygwin-api/func-cygwin-conv-path.html
> 
> the size parameter should say whether it's measured in bytes or
> characters.
> 
> For instance, if I pass CCP_POSIX_TO_WIN_W for the first parameter
> and a pure ASCII file name for the 'from' parameter, it's obvious
> the 'to' buffer needs to be at least 2x the size of 'from'.  The
> question the docs don't answer is whether I should pass >=
> 2*(strlen(from)+1) for size or >= strlen(from)+1?

I really thought that would be clear from the description of what
happens if size is 0:

  "If size is 0, cygwin_conv_path just returns the required buffer size
   in bytes."

and the subsequent example code which uses the size returned by
cygwin_conv_path in a call to malloc and then, unchanged in the next
cygwin_conv_path call.  I suggest to implement it as in the example.
Otherwise you will potentially need a buffer of 64K to return the
path in it.  Alternatively just use the cygwin_create_path function
which will malloc the buffer for you.

I'll change the description so that the value of "size" should be clear
from the beginning.


Corinna

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader  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: SO_REUSEADDR and windows enhanced socket security

[Corinna Vinschen]

On Dec 18 09:58, yesin wrote:
> Corinna! Thanks!
> Can you shedule a build for cygwin.dll with changes included in CVS?
> I promise to test it!

Not yet.  We're looking into getting 1.7.1 out of the door and we
still have a bit of trouble with the setup.exe installer to fix.
The change will not be in 1.7.1, but only in 1.7.2.  We produce
ocassional developer snapshots, just wait for the next one showing
up on http://cygwin.com/snapshots/
Alternatively just build it yourself from CVS.


Corinna

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader  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: perl: XML-Parser missing in 5.10.1-1

[Reini Urban]

2009/12/16 Yaakov (Cygwin/X):
> XML-Parser appears to have gone missing in perl-5.10.1-1.  This module is
> essential for a number of purposes, including intltool (a GNOME build tool).
>  Please let me know ASAP if you are planning on restoring this in the VERY
> near future, or if I should go ahead and ITP this.

Oops, my build missed that without exit. Sorry.

I wanted to fix the other perl issues (Cwd, whohas crash,
Win32::TieRegistry) before the next update.

But since I'm leaving for holidays on Monday I'll update perl this
weekend with the XML.:Parser fix only.
We certainly need that, yes.
-- 
Reini Urban
http://phpwiki.org/   http://murbreak.at/

--
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 1.7 GNU screen freezing when disconnected

[Jason Tishler]

Karim,

On Wed, Dec 16, 2009 at 07:05:49PM -0700, Karim Ali wrote:
> Sometimes, I get disconnected from my host computer. When this
> happens, I cannot reattach to the screen session when I can reconnect
> via ssh, and my screen session on the host computer is completely
> locked up. ctrl-q, ctrl-a K, doesn't do anything and I have to kill
> screen and the program I was running in it.

If you only need the reattach feature of screen, then I recommend trying
dtach:

http://dtach.sourceforge.net/

dtach builds OOTB and installs easily (it's just one file).  I've been
using it daily under Cygwin 1.5 for about 6 months without any problems.
I tested it under Cygwin 1.7 and it seems to work well there too.

I know the above doesn't help solve the screen reattach problem, but it
does offer a very viable workaround.  Hopefully, you and/or others will
find dtach as useful as I have.

Jason

-- 
PGP/GPG Key: http://www.tishler.net/jason/pubkey.asc or key servers
Fingerprint: 7A73 1405 7F2B E669 C19D  8784 1AFD E4CC ECF4 8EF6

--
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

/de/conout (was: Looking for the 'stdout' of a Windows application)

[Marc Girod]

defaria wrote:
> 
> the problem is that cleartool is written for Windows 
> thus uses /dev/conout
> 
I can see /dev/conout, but what can I do with it?

I found one hit in the docs:
http://www.cygwin.com/1.7/cygwin-ug-net/using-specialnames.html

Thanks,
Marc
-- 
View this message in context: 
http://old.nabble.com/Looking-for-the-%27stdout%27-of-a-Windows-application-tp26769438p26847257.html
Sent from the Cygwin list mailing list archive at Nabble.com.


--
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: /dev/conout

[Andrew DeFaria]

On 12/18/2009 11:00 AM, Marc Girod wrote:

defaria wrote:
the problem is that cleartool is written for Windows thus uses 
/dev/conout

I can see /dev/conout, but what can I do with it?

I don't know. I've never tried to do anything with it.

I found one hit in the docs:
http://www.cygwin.com/1.7/cygwin-ug-net/using-specialnames.html
As I understand it /dev/conout is the same as the Windows CONOUT$ 
device. As such it's Windows stdout, as per se. As a Windows stdout, it 
doesn't know anything about Cygwin's POSIX implementation of ptys. So, 
in a Cygwin Windows Window (i.e. the same window that you get when you 
do a cmd window in Windows) output to /dev/conout should show up just 
find in Cygwin's stdout. Therefore a cleartool ci  should prompt 
you for a checkin comment.


However, if you used ptys then the output to /dev/conout, and thus to 
Cygwin's stdout, will not show up because, again, Windows CONOUT$ does 
not understand ptys. How/Why do you use ptys in Cygwin? Various ways. 
For example, if you use xterm or rxvt, you'll be using ptys. This sucks 
because I like rxvt much better than the cruddy Windows window and rxvt 
doesn't incur the overhead of an X server - it can work without it.


How else do you start using ptys? Well if you ssh to a machine, again, 
you'll be using ptys.


So I've always just used something like "cleartool ci -c 'my comment' 
" or "cleartool ci -nc " and if I ever do a cleartool 
command and it seems to hang the first thing I think of is "Oh I bet 
it's that pty thing".


There is another terminal emulator software called "terminal" I think, 
that runs on Windows and interacts well with Cygwin. It also allows you 
to do gnome-terminal like things like transparent backgrounds. IIRC it 
seems to run a cmd window in the background and handles this issue with 
cleartool (and, BTW cqperl's debugger) properly - well sorta properly.


I'm sorry I don't have a better reference for that terminal emulator. I 
used it at my last client. I have IM'ed a friend to ask him where he 
found it. I'll report back if I get a good reference.

--
Andrew DeFaria 
Attempt to get a new car for your spouse-it'll be a great trade!


--
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: /de/conout (was: Looking for the 'stdout' of a Windows application)

[Andy Koppe]

2009/12/18 Marc Girod:
>> the problem is that cleartool is written for Windows
>> thus uses /dev/conout
>>
> I can see /dev/conout, but what can I do with it?

It won't literally be using /dev/conout, because that's a Cygwin
wrapper round a Windows concept.

Cleartool is doing either of two things:
- Use CreateFile("CONOUT$", ...) and then call WriteFile or
WriteConsole to write stuff to it. CONOUT$ is a special Windows
filename that provides access to the output side of the console that a
process is attached to. That console may be an invisible one created
by the 'run' tool or Cygwin DLL itself. No matter, the output
certainly won't appear in xterm or emacs.
-  Call WriteConsole on its stdout handle. When run in xterm or emacs,
the stdout handle will be a pipe (as part of Cygwin's pty
implementation), and a WriteConsole call on that will fail. Again, no
output.

There's nothing you can do about this apart from running cleartool in
an actual console or a console wrapper such as ttyfier (if it worked)
or http://sourceforge.net/projects/console/.

Andy

--
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: /de/conout

[Andrew DeFaria]

On 12/18/2009 11:37 AM, Andy Koppe wrote:

Yeah, what he said!
There's nothing you can do about this apart from running cleartool in 
an actual console or a console wrapper such as ttyfier (if it worked) 
or http://sourceforge.net/projects/console/.
Yes that's the one - http://sourceforge.net/projects/console. I don't 
remember if it handles cleartool's output properly but I do recall that 
when in cqperl's debugger the arrow keys suddenly started working again 
to bring up previously entered debug commands and somehow I saw the 
hidden console windows (perhaps during a shutdown). I don't know if it 
solved other pty oriented problems.

--
Andrew DeFaria 
I'm desperately trying to figure out why Kamikaze pilots wore helmets.


--
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 1.7 GNU screen freezing when disconnected

[Karim Ali]

On Fri, Dec 18, 2009 at 6:23 AM, Jason Tishler wrote:
> If you only need the reattach feature of screen, then I recommend trying
> dtach:
>
>    http://dtach.sourceforge.net/

Thanks for this, Jason. It's working very well for what I use screen
for, with no weird ssh problems either. I am still curious as what is
causing the screen weirdness though, so I will still try to help
figure out what's going on.

Thanks,
Karim

--
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: undefined reference to `___real__Znwj'

[Dave "What part of PCYMTNQREAIYR isn't obvious? ;-)" Korn]

jojelino wrote:

>> make -j 10 because of speed gain.
> and it complains.. which i reported it.
> so i got
>> cd i686-pc-cygwin/winsup
> and >make  again.
> and it complains when it comes to cygserver.exe
> 
> this could be answer for your question?

  I was just curious how you got as far as building cygserver.exe without
running into earlier problems compiling the cygwin1.dll; it doesn't build
using 4.5.0 without a lot of patching.

  Anyway, thanks for the bug report; you reminded me that I stumbled across
this bug earlier in the summer but put it to one side because 4.5.0 was still
a long way from release.  Time to fix it, particularly now that the libstdc++
changes have gone in.

  The problem is a bug in the linker, so you'll need to check out binutils
from sourceware.org cvs and apply the attached
weaksyms-vs-undefs-order-of-ref-fix-take-2.diff patch.  You'll also need the
other attached patch for the winsup/cygwin repository before it will all
compile correctly with 4.5.0; there are a number of aliasing problems to which
4.5.0 is more sensitive than earlier GCCs.

  (I'll be feeding these patches back upstream in due course, once I've given
them all some further testing, but they build an apparently fully working
4.5.0-compiled cygwin1.dll so far, anyway.)

cheers,
  DaveK
Index: bfd/cofflink.c
===
RCS file: /cvs/src/src/bfd/cofflink.c,v
retrieving revision 1.73
diff -p -u -r1.73 cofflink.c
--- bfd/cofflink.c	10 Oct 2009 04:58:48 -	1.73
+++ bfd/cofflink.c	18 Dec 2009 18:09:03 -
@@ -466,6 +466,30 @@ coff_link_add_symbols (bfd *abfd,
 		goto error_return;
 	}
 
+	  if (obj_pe (abfd) && (flags & BSF_WEAK) != 0 && addit && sym_hash 
+		&& (*sym_hash)->root.type == bfd_link_hash_undefined)
+	{
+	  /* If we try to add a PE weak external after having already
+	 seen an undefined reference to it in an earlier object,
+		 _bfd_generic_link_add_one_symbol will return the hash
+		 entry for the original symbol, and we'll lose track of
+		 the actual weak external definition and its associated
+		 auxiliary symbol containing the default value, leading
+		 to an undefined reference error at the end of linking.
+		 So here, if we are adding a weak external, and the sym_hash
+		 returned an existing undefined hash entry, we update the
+		 returned entry with the details of the new symbol.  */
+	  if ((*sym_hash)->symbol_class != C_NT_WEAK)
+		{
+		  (*sym_hash)->symbol_class = C_NT_WEAK;
+		  (*sym_hash)->root.type = bfd_link_hash_undefweak;
+		  (*sym_hash)->root.u.undef.abfd = abfd;
+		  (*sym_hash)->root.u.undef.weak = abfd;
+		  /* Now fix the undefs chain.  */
+		  bfd_link_repair_undef_list (info->hash);
+		}
+	}
+
 	  if (obj_pe (abfd) && (flags & BSF_SECTION_SYM) != 0)
 	(*sym_hash)->coff_link_hash_flags |=
 	  COFF_LINK_HASH_PE_SECTION_SYMBOL;
? newlib/libc/libc.info
? newlib/libm/libm.info
? winsup/cygwin/gendef.unwind
Index: winsup/cygwin/fhandler.h
===
RCS file: /cvs/src/src/winsup/cygwin/fhandler.h,v
retrieving revision 1.385
diff -p -u -r1.385 fhandler.h
--- winsup/cygwin/fhandler.h	16 Dec 2009 14:56:10 -	1.385
+++ winsup/cygwin/fhandler.h	18 Dec 2009 13:41:23 -
@@ -291,6 +291,18 @@ class fhandler_base
 			  DWORD nNumberOfLinks,
 			  DWORD dwFileAttributes)
 __attribute__ ((regparm (3)));
+  int __stdcall fstat_helper (struct __stat64 *buf,
+			  LARGE_INTEGER ftChangeTime,
+			  LARGE_INTEGER ftLastAccessTime,
+			  LARGE_INTEGER ftLastWriteTime,
+			  LARGE_INTEGER ftCreationTime,
+			  DWORD dwVolumeSerialNumber,
+			  ULONGLONG nFileSize,
+			  LONGLONG nAllocSize,
+			  ULONGLONG nFileIndex,
+			  DWORD nNumberOfLinks,
+			  DWORD dwFileAttributes)
+__attribute__ ((regparm (3)));
   int __stdcall fstat_by_nfs_ea (struct __stat64 *buf) __attribute__ ((regparm (2)));
   int __stdcall fstat_by_handle (struct __stat64 *buf) __attribute__ ((regparm (2)));
   int __stdcall fstat_by_name (struct __stat64 *buf) __attribute__ ((regparm (2)));
Index: winsup/cygwin/fhandler_disk_file.cc
===
RCS file: /cvs/src/src/winsup/cygwin/fhandler_disk_file.cc,v
retrieving revision 1.318
diff -p -u -r1.318 fhandler_disk_file.cc
--- winsup/cygwin/fhandler_disk_file.cc	27 Nov 2009 14:27:22 -	1.318
+++ winsup/cygwin/fhandler_disk_file.cc	18 Dec 2009 13:41:23 -
@@ -373,11 +373,11 @@ fhandler_base::fstat_by_handle (struct _
   pc.file_attributes (fbi.FileAttributes);
   return fstat_helper (buf,
 		   fbi.ChangeTime.QuadPart
-		   ? *(FILETIME *) (void *) &fbi.ChangeTime
-		   : *(FILETIME *) (void *) &fbi.LastWriteTime,
-		   *(FILETIME *) (void *) &fbi.LastAccessTime,
-		   *(FILETIME *) (void *) &fbi.LastWriteTime,
-		   *(FILETIME *) (void *) &fbi.CreationTime,
+		   ? fbi.ChangeTime
+		   : fbi.LastWr

'setup-1.7.exe -s' anomaly

[Greg Chicares]

If I run
  setup-1.7.exe -s ftp://ftp.heanet.ie
and hit Enter on every screen, I get a messagebox that says:
  Unable to get setup-2.ini from 
Oddly, 'ftp.heanet.ie' is listed twice in "Available Download Sites".
The first 'ftp.heanet.ie' is highlighted, and doesn't work. If I
highlight the second 'ftp.heanet.ie' instead, then it works fine.

Same outcome with:
  setup-1.7.exe -s ftp://ftp.heanet.ie
  setup-1.7.exe -s "ftp://ftp.heanet.ie";
  setup-1.7.exe --site "ftp://ftp.heanet.ie";
  setup-1.7.exe --site="ftp://ftp.heanet.ie";
and also with other mirrors, e.g.:
  setup-1.7.exe -s http://www.mirrorservice.org

This occurs on a brand-new machine with XP SP3 freshly installed, with
no prior installation of any version of Cygwin. I'm running setup-1.7
version 2.661 (current) in an XP "Command Prompt". My ultimate goal is
a '--quiet-mode' installation, but this simplified test case seems to
be the main obstacle.

--
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-1.7.exe -s' anomaly

[Jim Reisert AD1C]

I didn't know about the command-line options.

Is there some way to automate the process so that when I double-click
on my Setup shortcut, it will go right to the "Select Packages"
screen?

Thanks - Jim

On Fri, Dec 18, 2009 at 2:17 PM, Greg Chicares wrote:

> If I run
>  setup-1.7.exe -s ftp://ftp.heanet.ie
> and hit Enter on every screen, I get a messagebox that says:
>  Unable to get setup-2.ini from 
> Oddly, 'ftp.heanet.ie' is listed twice in "Available Download Sites".
> The first 'ftp.heanet.ie' is highlighted, and doesn't work. If I
> highlight the second 'ftp.heanet.ie' instead, then it works fine.
>
> Same outcome with:
>  setup-1.7.exe -s ftp://ftp.heanet.ie
>  setup-1.7.exe -s "ftp://ftp.heanet.ie";
>  setup-1.7.exe --site "ftp://ftp.heanet.ie";
>  setup-1.7.exe --site="ftp://ftp.heanet.ie";
> and also with other mirrors, e.g.:
>  setup-1.7.exe -s http://www.mirrorservice.org
>
> This occurs on a brand-new machine with XP SP3 freshly installed, with
> no prior installation of any version of Cygwin. I'm running setup-1.7
> version 2.661 (current) in an XP "Command Prompt". My ultimate goal is
> a '--quiet-mode' installation, but this simplified test case seems to
> be the main obstacle.

-- 
Jim Reisert AD1C, , http://www.ad1c.us

--
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-1.7.exe -s' anomaly

[Greg Chicares]

On 2009-12-18 21:28Z, Jim Reisert AD1C wrote:
> I didn't know about the command-line options.

To see them all:
  setup-1.7.exe --help

> Is there some way to automate the process so that when I double-click
> on my Setup shortcut, it will go right to the "Select Packages"
> screen?

AIUI, the GUI is either entirely on, or entirely off, controlled by:
  -q --quiet-modeUnattended setup mode
But you may be interested in
  -P --packages  Specify packages to install
which lets you select a list of packages from the command line.

--
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: /dev/conout

[Marc Girod]

defaria wrote:
> 
> So I've always just used something like "cleartool ci -c 'my comment' 
> " or "cleartool ci -nc " and if I ever do a cleartool 
> command and it seems to hang the first thing I think of is "Oh I bet 
> it's that pty thing".
> 
I believe I have better than that with my ClearCase::Wrapper::MGi.
It preempts these interactions.
My problem is twofold:
- the commands in which the interactions are part of the specificity, such
as merge
- the commands which I'd rather not override explicitly
I was thus looking for generically dropping my own treatment to plain
cleartool...
I am still looking forward to tfy...
The console doesn't seem to help me, unless it may be hidden, and I may read
what gets written there.
Marc
-- 
View this message in context: 
http://old.nabble.com/Looking-for-the-%27stdout%27-of-a-Windows-application-tp26769438p26850175.html
Sent from the Cygwin list mailing list archive at Nabble.com.


--
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-1.7.exe -s' anomaly

[Dave Korn]

Greg Chicares wrote:
> If I run
>   setup-1.7.exe -s ftp://ftp.heanet.ie
> and hit Enter on every screen, I get a messagebox that says:
>   Unable to get setup-2.ini from 
> Oddly, 'ftp.heanet.ie' is listed twice in "Available Download Sites".
> The first 'ftp.heanet.ie' is highlighted, and doesn't work. If I
> highlight the second 'ftp.heanet.ie' instead, then it works fine.

  Ok, this is a presentational problem.  What setup.exe displays is just the
top level of the site URL: the part from "http://"; up to the next "/", which
is the actual machine name where the mirror is located.

  However that's just an abbreviation to keep the display uncluttered;
generally, the mirror is located in a subdirectory and not at the top-level of
the mirror site.  To find the actual full URL of a mirror, take a look at the
mirrors list:

http://cygwin.com/mirrors.html

  If you hover over the "ftp.heanet.ie" link there, you'll see the full
address of the actual mirror directory itself, which in this case is

ftp://ftp.heanet.ie/pub/cygwin/

  So, that's what you'd have to enter for the -s option.  When you just give
the site name like you did in your example, setup.exe goes and looks for
ftp://ftp.heanet.ie/setup.ini and doesn't find it (because it's actually at
ftp://ftp.heanet.ie/pub/cygwin/setup.ini).  And so those two 'ftp.heanet.ie'
listings that you saw in the "download sites" page *are* actually different:
one of them is the one you entered, without the trailing part of the path, and
one of them is the full URL from the mirrors list including the trailing part
of the path, but setup.exe abbreviates both of them so they look the same way.

cheers,
  DaveK


--
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-1.7.exe -s' anomaly

[Dave Korn]

Jim Reisert AD1C wrote:
> I didn't know about the command-line options.
> 
> Is there some way to automate the process so that when I double-click
> on my Setup shortcut, it will go right to the "Select Packages"
> screen?

  Not yet, but it's on my to-do list to add a "unattended for everything
except the chooser page" mode.

cheers,
  DaveK

--
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: undefined reference to `___real__Znwj'

[jojelino]

thanks for replying. i needed it!
in fact, there is no need to change whole problematic cygwin source code to 
fix aliasing problem..
for me, just getting rid of '-Werror' from makefile.in resolves the problem.
it just complains for some lines of warning :)
"Dave "What part of PCYMTNQREAIYR isn't obvious? ;-)" Korn" 
 wrote in message 
news:4b2bdfab.2020...@googlemail.com...
> jojelino wrote:
>
>>> make -j 10 because of speed gain.
>> and it complains.. which i reported it.
>> so i got
>>> cd i686-pc-cygwin/winsup
>> and >make  again.
>> and it complains when it comes to cygserver.exe
>>
>> this could be answer for your question?
>
>  I was just curious how you got as far as building cygserver.exe without
> running into earlier problems compiling the cygwin1.dll; it doesn't build
> using 4.5.0 without a lot of patching.
>
>  Anyway, thanks for the bug report; you reminded me that I stumbled across
> this bug earlier in the summer but put it to one side because 4.5.0 was 
> still
> a long way from release.  Time to fix it, particularly now that the 
> libstdc++
> changes have gone in.
>
>  The problem is a bug in the linker, so you'll need to check out binutils
> from sourceware.org cvs and apply the attached
> weaksyms-vs-undefs-order-of-ref-fix-take-2.diff patch.  You'll also need 
> the
> other attached patch for the winsup/cygwin repository before it will all
> compile correctly with 4.5.0; there are a number of aliasing problems to 
> which
> 4.5.0 is more sensitive than earlier GCCs.
>
>  (I'll be feeding these patches back upstream in due course, once I've 
> given
> them all some further testing, but they build an apparently fully working
> 4.5.0-compiled cygwin1.dll so far, anyway.)
>
>cheers,
>  DaveK
>





> 




--
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

[ANNOUNCEMENT] Updated: sqlite3-3.6.21-3

[Warren Young]

The previous -2 release had a completely wrong source patch relative to 
the original sources, which caused it to be all but useless.  Thus, 
everyone who got the -2 should upgrade.


--
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 1.7 GNU screen freezing when disconnected

[Karim Ali]

On Fri, Dec 18, 2009 at 2:36 PM, Game Fan wrote:
> Wanted to see if cygwin 1.7 is released or not and seen your letter. Don't
> need another source of spam so I'm not subscribing to cygwin ML.
> Are you sure it's problem with CygWin? I'm seeing such problems quite often
> on Linux. The story goes like this: background screen process wants to stop
> old foreground process when you are attaching new one to it, but old
> foreground process tries to use PTY which lost it's owner! The lock happens
> somewhere in kernel so old foreground process can not respond (yes, blocking
> API are evil and all that). The solution is simple: find and kill old
> foreground process with "kill -9" and connect to screen again.

I got this email today off-list. As suggested, killing the program
that's running inside screen does make it reattachable, but this
doesn't really seem like a real solution to me, since usually I want
to see some state that was in screen.

Also, I do use Linux sometimes, and the steps I ran doesn't cause
screen to freeze there.

Thanks,
Karim

--
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: undefined reference to `___real__Znwj'

[Dave "What part of PCYMTNQREAIYR isn't obvious? ;-)" Korn]

jojelino wrote:
> thanks for replying. i needed it!
> in fact, there is no need to change whole problematic cygwin source code to 
> fix aliasing problem..
> for me, just getting rid of '-Werror' from makefile.in resolves the problem.
> it just complains for some lines of warning :)

  Yeah, and just getting rid of the pain by taking morphine will likewise
"resolve" the problem of a tiger having bitten your leg off, with only some
complaints and a dull throbbing.  Right up until you die of blood-loss, of 
course!

  The warnings are there for a good reason: they identify undefined code that
the compiler is very likely to generate assembly that doesn't do what you were
hoping it would do.  That is why the cygwin makefile deliberately makes these
warnings into fatal errors on purpose: because they are actually quite serious
and could result in a buggy or unreliable DLL.  You would be well advised to
use the patches rather than disabling -Werror.

  (BTW, if you do really want to do that, you can save yourself the trouble of
editing and regenerating the Makefile; removal of -Werror can be specified on
the configure command-line using '--disable-werror'.)

cheers,
  DaveK


--
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

2nd occurrence of signal not being handled?

[Mark Geisert]

I'm trying to get a larger application working but have an issue with signal
handling.  I've boiled the issue down to the STC below.  It causes two access
violations, only the first of which gets handled by the SIGSEGV signal handler
I've registered for the purpose.  The second access violation kills the program,
as if the handler were no longer registered.  Using diagnostic code not shown
I've made sure the handler is still registered, but somehow it's not being
called by Cygwin's internal fault handling.  Or maybe there's some subtle
mistake in my use of the signal functions.  Any advice would be appreciated.

..mark

#include 
#include 
#include 

volatile intstep = 0;
sigjmp_buf  trapoline;

void
segv_handler(int sig)
{
printf("case %d: SIGSEGV handled\n", step++);
siglongjmp(trapoline, step);
}

int
main()
{
struct sigactionsa;

sa.sa_handler = segv_handler;
sa.sa_flags   = 0;
sigemptyset(&sa.sa_mask);

sigaction(SIGSEGV, &sa, NULL);

switch(sigsetjmp(trapoline, 0)) {
case 0:
printf("case %d reached\n", step);
printf("case %d: %08X\n", *(int *) 42);
++step;

case 1:
printf("case %d reached\n", step);
printf("case %d: %08X\n", *(int *) 42);
++step;

default:
printf("case %d reached\n", step);
break;
}

return 0;
}


--
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