RE[2]: Exporting const variables from DLLs (GCC bug?)
--- Danny Smith <[EMAIL PROTECTED]> wrote: > Jon Foster wrote:
>
> > Given this source code:
> > extern const int meaning_of_life __declspec(dllexport);
> > const int meaning_of_life __declspec(dllexport) = 42;
> >
> >
> > GCC complains:
> > $ c++ -g -O2 -c test.cxx -o test.o
> > test.cxx:2: error: external linkage required for symbol 'const int
> >meaning_of_life' because of 'dllexport' attribute
> >
>
>
> It is a bug in gcc. The above code compiles okay with C, but
> strangely, not C++. For some reason, g++ does not immediately
> mark the the definition of global constants as public when they
> are defined after a prior declaration. I have a fix that I will
> submit to gcc-patches after reg-testing
>
On second thought, I don't think I will.
In c++, const variables at namespace scope have internal linkage unless
declared with explicit extern (as you have done). But I've just run this
testcase on gcc and VC and I get essentially the same error reports:
/* const.cpp */
/* 1: TREE_PUBLIC is set when dllexport attribute is handled. */
__declspec(dllexport) int bar = 42; // OK
/* 2: This is OK too. TREE_PUBLIC is set, because of 'extern' */
extern __declspec(dllexport) const int baz;
const int baz = 42; //OK
/* 3: TREE_PUBLIC _not_ set. Nor should it be. consts are local
by default in c++ */
__declspec(dllexport) const int foo = 42; // ERROR
/* 4: TREE_PUBLIC _not_ set. */
extern __declspec(dllexport) const int faz;
__declspec (dllexport) const int faz = 42; // ERROR
int fun()
{ return foo + faz; }
I thought that case 4 should be okay because of the 'extern' on declaration,
but if GCC wants to maintain consistency with native compiler, 4 should report
an error as well. The error mesage that VC emits is: error C2201: 'faz' : must
have external linkage in order to be exported/imported
Danny
http://personals.yahoo.com.au - Yahoo! Personals
New people, new possibilities. FREE for a limited time.
--
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 CPAN module help
Thank you for your help. I understand what you are saying here. I will try
installing under /usr/local. My question, though, is what do I do about all
the modules I've installed under /usr? How do I deal with them? Do I need to
re-install them under /usr/local? Do I then need to somehow remove them from
/usr? Or can I have modules under both directories, leaving the ones that
work under /usr and place new ones under /usr/local?
-Original Message-
From: Peter J. Acklam [mailto:[EMAIL PROTECTED]
Sent: Monday, November 17, 2003 1:40 AM
To: Gary Nielson
Cc: Peter J. Acklam; [EMAIL PROTECTED]
Subject: Re: Perl CPAN module help
"Gary Nielson" <[EMAIL PROTECTED]> wrote:
> I am getting somewhere. I used setup and installed needed
> executables such as gcc. Did a force install in cpan for LWP
> modules and it seemed to be go great. All tests were successful
> in make test. But when running make install I got the error:
You shouldn't use "force install" unless you really know what
you're doing. If your module fails a regular "install" you should
investigate the problem and find the solution rather than do a
"force install". With a "force install" you are likely to install
modules which fail some way on your system and hence shouldn't
have been installed.
> "Cannot forceunlink /usr/bin/HEAD: No such file or directory at
> /usr/lib/perl5/5.8.0/File/Find.pm line 873.
> make: *** [pure_site_install] Error 255
> /usr/bin/make install -- NOT OK.
>
> The Find.pm line in question is: { $wanted_callback->() }; #
> protect against wild "next"
When installing LWP you are asked whether you want to install the
GET, HEAD, and POST programs. You have chosed "yes" or the "force
install" did it for you. Either way, it was discovered that HEAD
exists (as /usr/bin/head.exe) and Perl is trying to remove it, but
although "which head" says "/usr/bin/HEAD", there really is no
"/usr/bin/HEAD.exe", it's "/usr/bin/head.exe". The problem is
that "which" matches case insensitively, but "rm" doesn't.
The solution is: Don't install CPAN modules under "/usr", use
"/usr/local"! When you install modules without CPAN, use the
steps
perl Makefile.PL PREFIX=/usr/local
make
make test
make install
this is done with the CPAN shell by setting
cpan> o conf makepl_arg PREFIX=/usr/local
cpan> o conf commit
> Any idea what is going wrong, or is the question better posed to
> a perl forum?
The "head" vs "HEAD" is a Cygwin thing, so I think it belongs
equally well here.
Peter
--
Peter J. Acklam - [EMAIL PROTECTED] - http://home.online.no/~pjacklam
--
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 CPAN module help
"Gary Nielson" <[EMAIL PROTECTED]> wrote: > Thank you for your help. I understand what you are saying > here. I will try installing under /usr/local. I forgot to mention that Perl will not, by default, search for modules in /usr/local. This is a disadvantage, but it's worth it, in my opinion. The simplest way to make Perl look for modules there is to add PERL5LIB=/usr/local/lib/perl5:/usr/local/lib/perl5/site_perl to your personal startup file (~/.bash_profile or whatever). > My question, though, is what do I do about all the modules I've > installed under /usr? How do I deal with them? Do I need to > re-install them under /usr/local? Do I then need to somehow > remove them from /usr? Or can I have modules under both > directories, leaving the ones that work under /usr and place new > ones under /usr/local? If they're working properly, you might as well leave them alone. If they're broken, re-install them with PREFIX=/usr/local. Note that perl will search the directories specified in PERL5LIB before the other directories, so if you have a working module under /usr/local and a broken one under /usr, then the working one under /usr/local will be used. The default search path: $ unset PERL5LIB $ perl -wle 'print for @INC' /usr/lib/perl5/5.8.0/cygwin-multi-64int /usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl . Now, note how /usr/local/... comes first: $ export PERL5LIB=/usr/local/lib/perl5:/usr/local/lib/perl5/site_perl $ perl -wle 'print for @INC' /usr/local/lib/perl5/5.8.0/cygwin-multi-64int /usr/local/lib/perl5/5.8.0 /usr/local/lib/perl5 /usr/local/lib/perl5/site_perl/5.8.0/cygwin-multi-64int /usr/local/lib/perl5/site_perl/5.8.0 /usr/local/lib/perl5/site_perl /usr/lib/perl5/5.8.0/cygwin-multi-64int /usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl . You may try to remove modules under /usr, but be careful so you don't remove things you need. If you really want to remove them, I'd rather do that by uninstalling Perl, removing everything under /usr/lib/perl5 and start over with a clean Perl installation. Peter -- Peter J. Acklam - [EMAIL PROTECTED] - http://home.online.no/~pjacklam -- 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 CPAN module help
Hi, I just read what Peter said about 'install' and 'force install'. I had to use 'force install' as well for installing the BerkeleyDB module, as 9 out of 20 tests were failing. So far it is running fine, but I haven't done any heavy usage tests yet, just some simple storages and retrievals. I'm using Perl v5.8.0-3, BerkeleyDB v0.25 and db3.1 v3.1.17-2, but saw this behaviour ever since I started with BerkeleyDB v0.21. I'm wondering whether anybody else is seeing this behaviour and managed to do a plain install. I've attached the output I get from 'test BerkeleyDB' below. I also do get warnings when doing a normal install for other modules that they are not allowed to install into /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int/, so they install into /usr/lib/perl5/site_perl/5.8.0/. Checking the ACL for above directory yields $ getfacl /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int/ # file: /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int/ # owner: eisenacher # group: Benutzer user::--- group::--- group:Administratoren:rwx mask:rwx other:r-x default:group:Administratoren:rwx default:mask:rwx default:other:r-x And checking my /etc/passwd shows: $ cat /etc/passwd SYSTEM:*:18:544:,S-1-5-18:: Administratoren:*:544:544:,S-1-5-32-544:: Administrator:unused_by_nt/2000/xp:500:513:U-PC-EISENACHER\Administrator,S-1-5-21-44 8539723-1606980848-1708537768-500:/home/Administrator:/bin/bash Gast:unused_by_nt/2000/xp:501:513:U-PC-EISENACHER\Gast,S-1-5-21-448539723-1606980848 -1708537768-501:/home/Gast:/bin/bash eisenacher:unused_by_nt/2000/xp:1000:513:Patrick Eisenacher,U-PC-EISENACHER\eisenacher,S-1-5-21-448539723-1606980848-1708537768-1000:/home/eisenacher:/bin/bash VUSR_PC-EISENACHER:unused_by_nt/2000/xp:1001:513:VSA Server Account,U-PC-EISENACHER\VUSR_PC-EISENACHER,S-1-5-21-448539723-1606980848-1708537768-1001:/home/VUSR_PC-EISENACHER:/bin/bash and /etc/group: $ cat '/etc/group' SYSTEM:S-1-5-18:18: Kein:S-1-5-21-448539723-1606980848-1708537768-513:513: Administratoren:S-1-5-32-544:544: Benutzer:S-1-5-32-545:545: Gäste:S-1-5-32-546:546: Hauptbenutzer:S-1-5-32-547:547: Replikations-Operator:S-1-5-32-552:552: Sicherungs-Operatoren:S-1-5-32-551:551: Which is weird in 2 ways. First I don't belong to any group, but the owning group is 'Benutzer' and perl sets up the directory in a way that only members of the group 'Administratoren' can write into it. Clearly, there are no members of any of those groups. Anybody any tips? Thanks in advance, Patrick Running make test /usr/bin/perl.exe "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/btreeok 177/244Can't call method "txn_begin" on an undefined value at t/btree.t line 638. t/btreedubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 28, 178-244 Failed 68/244 tests, 72.13% okay t/db-3.0...ok 1/14Can't call method "set_mutexlocks" on an undefined value at t/db-3.0.t line 39. t/db-3.0...dubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 2-14 Failed 13/14 tests, 7.14% okay t/db-3.1...ok t/db-3.2...skipped all skipped: this needs Berkeley DB 3.2.x or better t/db-3.3...skipped all skipped: this needs Berkeley DB 3.3.x or better t/destroy..ok 1/15Can't call method "txn_begin" on an undefined value at t/destroy.t line 33. t/destroy..dubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 2-15 Failed 14/15 tests, 6.67% okay t/encrypt..skipped all skipped: this needs Berkeley DB 4.1.x or better t/env..ok 18/50Can't call method "txn_begin" on an undefined value at t/env.t line 104. t/env..dubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 12, 14, 21-50 Failed 32/50 tests, 36.00% okay t/examples.ok t/examples3ok t/filter...ok t/hash.ok 146/212Can't call method "txn_begin" on an undefined value at t/hash.t line 444. t/hash.dubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 29, 147-212 Failed 67/212 tests, 68.40% okay t/join.Can't call method "txn_begin" on an undefined value at t/join.t line 86. t/join.dubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 8-41 Failed 34/41 tests, 17.07% okay t/mldbmok t/queueskipped all skipped: Queue needs Berkeley DB 3.3.x or better t/recnook 166/226Can't call method "txn_begin" on an undefined value at t/recno.t line 471. t/recnodubious Test returned status 255 (wstat 65280, 0xff00) DIED. FAILED tests 28, 168-226 Failed 60/226 tests, 73.45% okay t/strict...NOK 2Use of uninitialized value in numeric eq (==) at t/strict.t line 40. t/strict...NOK 8Can't call method "txn_begin" on an undefined value at t/str t/strict...ok 9/44 t/strict...dubious Test returne
cygutils 1.2.2-1: small error in ipck
The ipck script in the current cygutils (1.2.2-1) has an extraneous tick-mark on line 17 immediately after the GPL header, resulting in the following error when run: $ ipck /usr/bin/ipck: 18: Syntax error: end of file unexpected (expecting ";;") Simply deleting that line fixes the script. _ ** SMS(mobile text messages) around the world, Get 100 messages for $7.95, plus sign up now and get 2 free tryout messages http://sms.aemail4u.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/
Re: For masochists: the leap o faith
On Sun, Nov 16, 2003 at 08:10:08AM +1100, Robert Collins wrote: > Chris has noted that posixly correct behaviour and common practice may > diverge. I think for this scenario, that posix behaviour allows the most > accurate representation of the variety programs may encounter on cygwin > at runtime. Therefore we'll get the best results (and perhaps uncover a > few portability bugs) going that way. > > We have two choices (no particular order of preference): > a) make MAX_PATH and posix friends the maximum length path cygwin will > accept/return. Return ENAMETOOLONG on path calls on win9x, or winnt > FAT[32] calls. Update pathconf to return appropriate values. > b) blow away MAX_PATH and MAXPATHLEN so that programs using cygwin > fallback to pathconf, or 'good enough for me' static arrays. Update > pathconf to return appropriate values. Well, I guess you meant PATH_MAX here. Just to reiterate, MAX_PATH is a Windows specific constant, giving a maximum size for path names when using ASCII mode. Even this MAX_PATH constant doesn't actually reflect the maximum path length possible in all cases: CreateDirectory for instance can only create directories with a path length up to 248 characters. So even on native Windows, the constant is used as an upper limit, e. g. usable for static buffer sizes in applications. Having said that, however, you must not change MAX_PATH. I would prefer to change PATH_MAX and MAXPATHLEN to an arbitrary big value as, e. g. the same as on Linux, 4096, or even the biggest possible plus one: 32768. The latter is probably the better value. So my choice is a) Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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/
[wayne@cs.toronto.edu: Cygwin-1.5.5 sscanf on floats: 20 times slower than 2 years ago]
Since scanf and the floating point arithmetic is implemented in newlib, I've redirected this message there. Does anybody have an idea, what could slow down float scanning in sscanf by a factor of 20? Corinna - Forwarded message from Wayne Hayes <[EMAIL PROTECTED]> - > Date: Sun, 16 Nov 2003 09:54:31 -0500 > From: Wayne Hayes <[EMAIL PROTECTED]> > Subject: Cygwin-1.5.5 sscanf on floats: 20 times slower than 2 years ago > To: [EMAIL PROTECTED] > > Hello. Until recently I was running W2k with an ancient version of > cygwin; I don't know what version it was, but if anybody cares I've > put a copy of cygwin1.dll (dated 2001-May-20) at > > http://www.cs.toronto.edu/~wayne/tmp/cygwin1-old.dll.gz > > It worked fine on W2k for the past 2.5 years. > > I recently upgraded to Windows XP and this old cygwin stopped working. > No problem, I say, it's also time to upgrade cygwin, so I go install the > most recent version. > > Everything works fine, except one of my I/O intensive simulations starts > running about 20 times slower! > > After copious mucking about, I finally narrowed it down to sscanf: sscanf > on double precisions numbers, such as > > double a; > char line [80]; > fgets(line...) > sscanf(line, "%lf", &a); > > runs about 20 times slower than in the old cygwin. Replacing the sscanf > with a call to atof gets back the old speed. It's only noticable, of > course, if you're scanning a huge file. In my case, it's a 60MB file > that contains tens of thousands of lines of ASCII floating point numbers. > > So, consider this a bug report. A slowdown of a factor of 20 is not really > a good thing. > > The output of "cygcheck -s -v -r > cygcheck.out" can be found at > > http://www.cs.toronto.edu/~wayne/tmp/cygcheck-XP-1.5.5.out.gz > > Thanks! > > - Wayne - End forwarded message - -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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/
Building bison?
Hi, This may well verge on FAQ-like issues, but a quick search didn't find it. I've been investigating a possible bug in bison, and talking with the folk on the bison-bug mail-list. Fairly reasonably, their first request was that I build their latest version and see if the bug still occurs. An they gave me a version numbered 1.875c (the current cygwin version is 1.975b, so they are pretty similar). 1.875c behaves nicely to a ./configure and make but it then crashes immediately when I give it my input script. I was wondering whether I should have expected that, or whether packages generally get patched when converted to cygwin? I ask because I seem re recall some posts on this list a few months ago, which were about how much needed patching, every time new sources where brought over from GNU? So, are any special actions required when getting source from other places than via the cygwin setup program? Ian B Free transport into the future, while you wait. -- 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: Building bison?
Ian Badcoe wrote on Monday, November 17, 2003 12:49 PM: [snip] > 1.875c behaves nicely to a ./configure and make but it > then crashes > immediately when I give it my input script. Just an idea: Since Cygwin compiles as Unix flavour, the original source may not be prepared to read files with CRLF. Try to run it from a binary mounted folder providing a file with LF endings only. Regards, Jörg -- 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/
encrypt on unix and decrypt on windows
Hello All, I need to use public key encryption to encrypt information using perl on unix and decrypt it on windows. I appreciate any help or pointers in the right direction that anyone can provide me. Thanks in advance! Artem Korneev. -- 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: messed up user permissions from w2k terminal session
Hi,
Le mer 08 oct 2003 19:45:46 GMT, Pierre A. Humblet a tapoté sur son clavier :
> On Fri, Oct 03, [EMAIL PROTECTED]:48:43PM -0400, James D Below wrote:
> > HI everyone,
> >
> > I'm not sure how I did it but I messed up my user permissions or local
> > policy settings. Now whenever I run any cygwin app (bash.exe, wc.exe,
> > rxvt.exe) from a w2k terminal session and logged in as a user, I see the
> > following error:
> >
> > CreateFileMapping, Win32 error 5. Terminating.
> >
> > I'm running Windows 2000 SP4 and CYGWIN = binmode tty ntsec
>
> This problem has now been explained with James' help and this message
> is to close the thread.
> It turns out that on some recent Windows systems a special privilege,
> "create global objects", is required to run Cygwin 1.5.X from
> a terminal session. It can be given to users with the "editrights.exe"
> utility.
Hmmm. I can't find such right. editrights gives me this list:
editrights version 1.01: a cygwin application to edit user rights
on a Windows NT system.
Copyright Chris Rodgers , Sep, 2003.
All rights reserved. See LICENCE for further details.
Usage: editrights -u USER {-a|-l|-r|-t} [options]
-a Se... Add right to the specified user.
-h Show this help message. -hv lists available user rights.
-l List user rights. May be combined with -a or
-r to list final state.
-m MACHINE Make all changes on the specified MACHINE.
-r Se... Remove right from the specified user.
-t Se... Test if the specified right is held by user. Returns
0 for YES and 2 for NO.
-u USER/GROUPMake changes to the specified USER or GROUP.
-v Verbose mode.
Return values:
0 Success or YES.
1 Error.
2 NO.
Available user rights include:
SeAssignPrimaryTokenPrivilege
SeAuditPrivilege
SeBackupPrivilege
SeBatchLogonRight
SeChangeNotifyPrivilege
SeCreatePagefilePrivilege
SeCreatePermanentPrivilege
SeCreateTokenPrivilege
SeDebugPrivilege
SeDenyBatchLogonRight
SeDenyInteractiveLogonRight
SeDenyNetworkLogonRight
SeDenyRemoteInteractiveLogonRight
SeDenyServiceLogonRight
SeEnableDelegationPrivilege
SeIncreaseBasePriorityPrivilege
SeIncreaseQuotaPrivilege
SeInteractiveLogonRight
SeLoadDriverPrivilege
SeLockMemoryPrivilege
SeMachineAccountPrivilege
SeManageVolumePrivilege
SeNetworkLogonRight
SeProfileSingleProcessPrivilege
SeRemoteInteractiveLogonRight
SeRemoteShutdownPrivilege
SeRestorePrivilege
SeSecurityPrivilege
SeServiceLogonRight
SeShutdownPrivilege
SeSyncAgentPrivilege
SeSystemEnvironmentPrivilege
SeSystemProfilePrivilege
SeSystemtimePrivilege
SeTakeOwnershipPrivilege
SeTcbPrivilege
SeUndockPrivilege
SeUnsolicitedInputPrivilege
Which one should be set ?
Regards,
Samuel Thibault
--
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/
Problem installing sshd
Dear group, I try to install openssh. Therefore I selected: - Admin - cygrunsrv - Net - mod_ssl, openssh, openssl, openssl096 First problem: obviously there is a dependency to Apache webserver. Although it got selected implicitely when selecting mod_ssl it didn't get installed (i.e. when doing a "cygcheck -c -v" I got tons of missing files). After Reinstalling them this was fixed. However, second problem: I still get the warning: Missing file: /usr/lib/apache/new/libssl.dll from package mod_ssl mod_ssl 2.8.8-1.3.24-1 Incomplete which I am not able to fix. When trying to start sshd I get the error message: Application popup: sshd.exe - Entry Point Not Found : The procedure entry point __getreent could not be located in the dynamic link library cygwin1.dll. and I assume that the missing library could be the cause. Any help or pointer to documentation is highly appreciated. Friedrich -- 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: Problem installing sshd
At 11:29 AM 11/17/2003, [EMAIL PROTECTED] you wrote: >Dear group, > >I try to install openssh. Therefore I selected: >- Admin - cygrunsrv >- Net - mod_ssl, openssh, openssl, openssl096 > >First problem: >obviously there is a dependency to Apache webserver. >Although it got selected implicitely when selecting mod_ssl it didn't get >installed (i.e. when doing a "cygcheck -c -v" I got tons of missing files). >After Reinstalling them this was fixed. > >However, second problem: >I still get the warning: > > Missing file: /usr/lib/apache/new/libssl.dll from package mod_ssl > mod_ssl 2.8.8-1.3.24-1 Incomplete > >which I am not able to fix. When trying to start sshd I get the error >message: > > Application popup: sshd.exe - Entry Point Not Found : The procedure >entry point __getreent could not be located in the dynamic link library >cygwin1.dll. > >and I assume that the missing library could be the cause. > >Any help or pointer to documentation is highly appreciated. Didn't setup complain at the end that it couldn't install everything you asked without a reboot? If so, did you reboot? This looks like the problem to me. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 838 Washington Street (508) 893-9889 - FAX Holliston, MA 01746 -- 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: Problem installing sshd
On Mon, Nov 17, 2003 at 11:37:36AM -0500, Larry Hall wrote: >At 11:29 AM 11/17/2003, [EMAIL PROTECTED] you wrote: >>Dear group, >> >>I try to install openssh. Therefore I selected: >>- Admin - cygrunsrv >>- Net - mod_ssl, openssh, openssl, openssl096 >> >>First problem: >>obviously there is a dependency to Apache webserver. >>Although it got selected implicitely when selecting mod_ssl it didn't get >>installed (i.e. when doing a "cygcheck -c -v" I got tons of missing files). >>After Reinstalling them this was fixed. >> >>However, second problem: >>I still get the warning: >> >> Missing file: /usr/lib/apache/new/libssl.dll from package mod_ssl >> mod_ssl 2.8.8-1.3.24-1 Incomplete >> >>which I am not able to fix. When trying to start sshd I get the error >>message: >> >> Application popup: sshd.exe - Entry Point Not Found : The procedure >>entry point __getreent could not be located in the dynamic link library >>cygwin1.dll. >> >>and I assume that the missing library could be the cause. >> >>Any help or pointer to documentation is highly appreciated. > >Didn't setup complain at the end that it couldn't install everything you >asked without a reboot? If so, did you reboot? This looks like the problem >to me. And/or there's more than one version of cygwin1.dll on the system. -- 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/
Problems with Make, VPATH and MS-DOS paths
Hi,
I'm having a problem getting MS-DOS paths to work properly with VPATH under
GNU Make 3.80. The problem is that the VPATH processing tacks on a Unix path
separator ('/') to the end of the VPATH giving me a source file name something
similar to code\src\fw/foo.c. This is a significant problem for the compiler
I'm using (NOT gcc) since it spits out map and list files based on the stem of
the input source name which it decides is 'fw/foo.c'. Because of this same
compiler, I cannot use Unix path names under a cygwin shell due to the fact that
it screws up the internal processing of this compiler. So, I would like to know
if it's possible to get the VPATH processing to use an MS-DOS separator when
generating source file names. I've tried using the --win32 switch with no
success.
Thanks in advance
Nate Bohlmann ([EMAIL PROTECTED])
--
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: messed up user permissions from w2k terminal session
On Mon, Nov 17, 2003 at 04:37:48PM +0100, Samuel Thibault wrote: > Le mer 08 oct 2003 19:45:46 GMT, Pierre A. Humblet a tapot? sur son clavier : > > It turns out that on some recent Windows systems a special privilege, > > "create global objects", is required to run Cygwin 1.5.X from > > a terminal session. It can be given to users with the "editrights.exe" > > utility. > > Hmmm. I can't find such right. editrights gives me this list: The right only exists beginning with 2003 Server. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developermailto:[EMAIL PROTECTED] Red Hat, Inc. -- 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: 1.5.5: sshd problem
On Mon, 17 Nov 2003, John Pye wrote:
> Thanks for the extra tips, Igor. Do any of these results look strange to
> you?
>
> Igor Pechtchanski wrote:
>
> >That's the default mode Windows gives it. This should work, but somehow
> >doesn't... Can sshd get to all the necessary files and directories? Look
> >at the permissions on /etc and the files in it, as well as /bin.
>
> I tried an ls -l /etc and found
>
> -rwxr-xr-x1 SYSTEM None 1159 Nov 13 19:02 ssh_config
> -rw---1 SYSTEM None 668 Nov 13 19:02 ssh_host_dsa_key
> -rw-r--r--1 SYSTEM None 599 Nov 13 19:02 ssh_host_dsa_key.pub
> -rw---1 SYSTEM None 524 Nov 13 19:02 ssh_host_key
> -rw-r--r--1 SYSTEM None 328 Nov 13 19:02 ssh_host_key.pub
> -rw---1 SYSTEM None 887 Nov 13 19:02 ssh_host_rsa_key
> -rw-r--r--1 SYSTEM None 219 Nov 13 19:02 ssh_host_rsa_key.pub
> -rw-r--r--1 SYSTEM None 2427 Nov 13 19:03 sshd_config
>
> That looks OK I thought... or is it? Does 'SYSTEM' need to be able to
> read those files, or does 'sshd'?
Whatever account the ssh daemon is running as (if you use cygrunsrv or the
ssh-host-config script, it's most likely "SYSTEM"). How about 'ls -ld
/etc'?
> I also had a look at /etc/bin and it's all owned by john.Users, for example
Huh? /etc/bin? I assume you mean /usr/bin or /bin.
> -rwxrwxrwx1 john Users 19456 Feb 20 2002 split.exe
> -rwxrwxrwx1 john Users 68608 Nov 6 02:47 ssh-add.exe
> -rwxrwxrwx1 john Users 57856 Nov 6 02:47 ssh-agent.exe
> -rwxrwxrwx1 john Users 17333 Nov 6 02:47 ssh-host-config
> -rwxrwxrwx1 john Users 75776 Nov 6 02:47 ssh-keygen.exe
> -rwxrwxrwx1 john Users 130048 Nov 6 02:47 ssh-keyscan.exe
> -rwxrwxrwx1 john Users6266 Nov 6 02:47 ssh-user-config
> -rwxrwxrwx1 john Users 223232 Nov 6 02:47 ssh.exe
> -rwxrwxrwx1 john Users 18944 Sep 21 06:32 ssp.exe
> -rwxrwxrwx1 john Users 87552 Aug 26 2002 states.exe
> -rwxrwxrwx1 john Users 23552 Sep 21 06:32 strace.exe
>
> That could be a problem perhaps? Should the 'Users' group contain 'sshd'
> or is access for sshd to the /bin executables handled somehow else?
No, the access on the files themselves is ok. How about 'ls -ld /bin'?
> >Ok, looks like all your mounts are system mounts, unless you simply don't
> >have the permission to read the registry keys for the SYSTEM user...
>
> That seems strange. My home computer has a couple of mounts when I run
> that command, but this system has none. However, it works fine locally
> (compiling, listing files, /cygdrive/c, etc). I haven't ever tinkered
> with the mounts, so I wonder how that comes to be... Couldn't be related
> to mingw/msys could it?
>
> John
It's related to the mode you installed Cygwin in ("Just me" vs. "All
users"). If you installed for "Just me", you may be missing the necessary
mounts for services like sshd to work.
Looking at your earlier message, however, I start wondering if we aren't
barking up the wrong tree here...
On Thu, 13 Nov 2003, John Pye wrote:
> I still get the error as shown
>
> [EMAIL PROTECTED] ~
> $ ssh localhost
> ssh_exchange_identification: Connection closed by remote host
>
> The verbose output is
>
> $ ssh -vvv localhost
> OpenSSH_3.7.1p2, SSH protocols 1.5/2.0, OpenSSL 0.9.7c 30 Sep 2003
> debug1: Reading configuration data /etc/ssh_config
> debug2: ssh_connect: needpriv 0
> debug1: Connecting to localhost [127.0.0.1] port 22.
> debug1: Connection established.
> debug1: identity file /home/john/.ssh/identity type 0
> debug3: Not a RSA1 key file /home/john/.ssh/id_rsa.
^^^
> debug2: key_type_from_name: unknown key type '-BEGIN'
> debug3: key_read: missing keytype
> debug2: key_type_from_name: unknown key type 'Proc-Type:'
> debug3: key_read: missing keytype
> debug2: key_type_from_name: unknown key type 'DEK-Info:'
> debug3: key_read: missing keytype
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug3: key_read: missing whitespace
> debug2: key_type_from_name: unknown key type '-END'
> debug3: key_read: missing keytype
> debug1: identity file /home/john/.ssh/id_rsa type 1
> debug1: identity file /home/john/.ssh/id_dsa type -1
> ssh_exchange_identification: Connection closed by remote host
> debug1: Calling cleanup 0x41bf10(0x0)
Could you try removing /home/john/.ssh and re-running
/bin/ssh-user-config?
Ig
Re: [wayne@cs.toronto.edu: Cygwin-1.5.5 sscanf on floats: 20 times slower than 2 years ago]
>Since scanf and the floating point arithmetic is implemented in newlib, >I've redirected this message there. Does anybody have an idea, what >could slow down float scanning in sscanf by a factor of 20? Thanks! Just to be pedentic, I realized that it's worse than a factor of 20. My *entire simulation* slows down by a factor of 20; there's significant other computation in it. So the scanf slowdown is probably closer to hundreds of times. *Something* fishy must be going 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/
latex crash
Hello! I've recently found that latex has begun crashing if I execute it from a windows xp (or 2000) cmd prompt. It happens by simply typing "latex" at a cmd prompt. I get (in a windows error box) 16 bit MS-DOS Subsystem C:\WINDOWS\System32\cmd.exe - latex The NTVDM CPU has encountered an illegal instruction. CS:06cc IP:210f OP:63 69 66 69 65 Choose 'Close' to terminate the application. this doesn't happen if I expecute it from a bash prompt. Even more strangely, if I go to a command prompt and into c:\cygwin\bin and type "dir latex.exe" it seems unable to see the file (although it is visable in windows explorer) I can reproduce this on both computers I have (independantly) installed cygwin on. Chris Cygwin Win95/NT Configuration Diagnostics Current System Time: Mon Nov 17 19:19:49 2003 Windows XP Professional Ver 5.1 Build 2600 Service Pack 1 Path: C:\WINDOWS\system32 C:\WINDOWS C:\WINDOWS\System32\Wbem C:\Program Files\Executive Software\Diskeeper\ c:\cygwin\bin c:\cygwin\usr\X11Rb\bin c:\cygwin\usr\local\bin\i686-pc-cygwin Output from C:\cygwin\bin\id.exe (nontsec) UID: 1003(mrjeff) GID: 513(None) 513(None) Output from C:\cygwin\bin\id.exe (ntsec) UID: 1003(mrjeff) GID: 513(None) 513(None)544(Administrators) 545(Users) SysDir: C:\WINDOWS\System32 WinDir: C:\WINDOWS Path = `C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;C:\Program Files\Executive Software\Diskeeper\;c:\cygwin\bin;c:\cygwin\usr\X11Rb\bin;c:\cygwin\usr\local\bin\i686-pc-cygwin' ALLUSERSPROFILE = `C:\Documents and Settings\All Users' APPDATA = `C:\Documents and Settings\mrjeff\Application Data' CLIENTNAME = `Console' CommonProgramFiles = `C:\Program Files\Common Files' COMPUTERNAME = `BUBBLESCOPE' ComSpec = `C:\WINDOWS\system32\cmd.exe' HOMEDRIVE = `C:' HOMEPATH = `\Documents and Settings\mrjeff' LOGONSERVER = `\\BUBBLESCOPE' NUMBER_OF_PROCESSORS = `1' OS = `Windows_NT' PATHEXT = `.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH' PROCESSOR_ARCHITECTURE = `x86' PROCESSOR_IDENTIFIER = `x86 Family 6 Model 7 Stepping 1, AuthenticAMD' PROCESSOR_LEVEL = `6' PROCESSOR_REVISION = `0701' ProgramFiles = `C:\Program Files' PROMPT = `$P$G' SESSIONNAME = `Console' SystemDrive = `C:' SystemRoot = `C:\WINDOWS' TEMP = `C:\DOCUME~1\mrjeff\LOCALS~1\Temp' TMP = `C:\DOCUME~1\mrjeff\LOCALS~1\Temp' USERDOMAIN = `BUBBLESCOPE' USERNAME = `mrjeff' USERPROFILE = `C:\Documents and Settings\mrjeff' windir = `C:\WINDOWS' HKEY_CURRENT_USER\Software\Cygnus Solutions HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin HKEY_CURRENT_USER\Software\Cygnus Solutions\Cygwin\mounts v2 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\/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 10001Mb 60% CP CS UN PA FC d: cd N/AN/A e: hd NTFS 68527Mb 74% CP CS UN PA FC New Volume f: cd CDFS 267Mb 100%CS UN KONAMI_DDRFORPC g: hd NTFS 58635Mb 95% CP CS UN PA FC C:\cygwin / system binmode C:\cygwin/bin /usr/bin system binmode C:\cygwin/lib /usr/lib system binmode C:\cygwin\usr\X11R6\lib\X11\fonts /usr/X11R6/lib/X11/fonts system binmode . /cygdrive system binmode,cygdrive Found: c:\cygwin\bin\awk.exe Found: c:\cygwin\bin\bash.exe Found: c:\cygwin\bin\cat.exe Found: c:\cygwin\bin\cp.exe Found: c:\cygwin\bin\cpp.exe Found: c:\cygwin\bin\find.exe Found: c:\cygwin\bin\gcc.exe Not Found: gdb Found: c:\cygwin\bin\grep.exe Found: c:\cygwin\bin\ld.exe Found: c:\cygwin\bin\ls.exe Found: c:\cygwin\bin\make.exe Found: c:\cygwin\bin\mv.exe Found: c:\cygwin\bin\rm.exe Found: c:\cygwin\bin\sed.exe Found: c:\cygwin\bin\sh.exe Found: c:\cygwin\bin\tar.exe 802k 2003/09/15 c:\cygwin\bin\cygaspell-15.dll - os=4.0 img=1.0 sys=4.0 "cygaspell-15.dll" v0.0 ts=2003/9/15 13:32 61k 2003/08/09 c:\cygwin\bin\cygbz2-1.dll - os=4.0 img=1.0 sys=4.0 "cygbz2-1.dll" v0.0 ts=2003/8/9 7:35 7k 2003/10/19 c:\cygwin\bin\cygcrypt-0.dll - os=4.0 img=1.0 sys=4.0
RE: Problems with Make, VPATH and MS-DOS paths
> From: Nate Bohlmann
> Sent: Monday, November 17, 2003 5:53 PM
> Hi,
> I'm having a problem getting MS-DOS paths to work properly with
> VPATH under
> GNU Make 3.80. The problem is that the VPATH processing tacks on
> a Unix path
> separator ('/') to the end of the VPATH giving me a source file
> name something
> similar to code\src\fw/foo.c. This is a significant problem for
> the compiler
> I'm using (NOT gcc) since it spits out map and list files based
> on the stem of
> the input source name which it decides is 'fw/foo.c'.
$ cygpath --help
IMO it should help to solve all your problems. cygpath is part of the base
package so there is no need go looking for it either; if you have cygwin
then you have cygpath too.
/Hannu E K Nevalainen, B.Sc. EE - 59+16.37'N, 17+12.60'E
-- printf("LocalTime: UTC+%02d\n",(DST)? 2:1); --
--END OF 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: RE: Problems with Make, VPATH and MS-DOS paths
11/17/03 12:31:23 PM, Hannu E K Nevalainen <[EMAIL PROTECTED]> wrote:
>> From: Nate Bohlmann
>> Sent: Monday, November 17, 2003 5:53 PM
>
>> Hi,
>> I'm having a problem getting MS-DOS paths to work properly with
>> VPATH under
>> GNU Make 3.80. The problem is that the VPATH processing tacks on
>> a Unix path
>> separator ('/') to the end of the VPATH giving me a source file
>> name something
>> similar to code\src\fw/foo.c. This is a significant problem for
>> the compiler
>> I'm using (NOT gcc) since it spits out map and list files based
>> on the stem of
>> the input source name which it decides is 'fw/foo.c'.
>
>
>$ cygpath --help
>
>IMO it should help to solve all your problems. cygpath is part of the base
>package so there is no need go looking for it either; if you have cygwin
>then you have cygpath too.
How exactly does a command line tool help with VPATH'ing inside of Make?
--
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: cvs complains of no repository (solved)
Well, Skip the below patch. Pete Stieber suggested I check my mounts that they are text mode. Sure enough, the mount point for the filesystem containing my local CVS directory was in bin mode. After mounting in text mode, all works fine. (mount -t E: /e for example). Might be useful to have this in the CVS FAQ if there is one. -- Ken Shaffer On Thu, 13 Nov 2003, Shaffer, Kenneth wrote: > I looked further into this and think the problem is due to the fact > that the 1.10 version I was using was strictly Windows-based and so > whenever a CVS/Root file was created it had the for the end of > line. Then when I used the cygwin version I had trouble. > > The following patch fixed the problem I was having (feel free to make it > better): > > diff -w -urN cvs-1.11.6-3-orig/src/root.c cvs-1.11.6-3-plus/src/root.c > --- cvs-1.11.6-3-orig/src/root.c2003-02-28 18:11:36.0 > -0500 > +++ cvs-1.11.6-3-plus/src/root.c2003-11-13 10:20:32.746353600 > -0500 > @@ -660,6 +660,10 @@ > > /* Hooray! We finally parsed it! */ > free (cvsroot_save); > +if ((p = strrchr (newroot->directory, '\r')) != NULL) > + *p = '\0'; > +if ((p = strrchr (newroot->directory, '\n')) != NULL) > + *p = '\0'; > return newroot; > > error_exit: > > > -- > Ken Shaffer > > - - - - - - - Appended by Scientific-Atlanta, Inc. - - - - - - - This e-mail and any attachments may contain information which is confidential, proprietary, privileged or otherwise protected by law. The information is solely intended for the named addressee (or a person responsible for delivering it to the addressee). If you are not the intended recipient of this message, you are not authorized to read, print, retain, copy or disseminate this message or any part of it. If you have received this e-mail in error, please notify the sender immediately by return e-mail and delete it from your computer. -- 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: latex crash
At 02:20 PM 11/17/2003, chris jefferson you wrote: >Hello! > >I've recently found that latex has begun crashing if I execute it from a windows xp >(or 2000) cmd prompt. > >It happens by simply typing "latex" at a cmd prompt. I get (in a windows error box) > > >16 bit MS-DOS Subsystem > >C:\WINDOWS\System32\cmd.exe - latex >The NTVDM CPU has encountered an illegal instruction. >CS:06cc IP:210f OP:63 69 66 69 65 Choose 'Close' to terminate the application. > > >this doesn't happen if I expecute it from a bash prompt. > >Even more strangely, if I go to a command prompt and into c:\cygwin\bin and type "dir >latex.exe" it seems unable to see the file (although it is visable in windows >explorer) latex.exe is a Cygwin-specific symbolic link. Use tex.exe instead at the cmd prompt. That should work OK. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 838 Washington Street (508) 893-9889 - FAX Holliston, MA 01746 -- 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: dircmp for cygwin?
thanks for the responses! diff -r will do it for me. cheers, leo - Original Message - From: "Yitzchak Scott-Thoennes" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, November 17, 2003 2:02 PM Subject: Re: dircmp for cygwin? > On Sun, Nov 16, 2003 at 08:23:26PM -0600, Bobby McNulty Junior <[EMAIL PROTECTED]> wrote: > > Google it. You might find it, if it's there. > > > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf > > Of leo > > Sent: Sunday, November 16, 2003 7:36 PM > > To: [EMAIL PROTECTED] > > Subject: dircmp for cygwin? > > > > however i couldn't find dircmp in the collection of tools. is there a dircmp > > implementation for cygwin? > > Archaeological notes: > dircmp was listed as legacy in the susv2 standard and removed in susv3 > (in deference to diff -r). Rationale: > >Although a useful concept, the historical output of this directory >comparison program is not suitable for processing in application >programs. Also, the diff -r command gives equivalent functionality. > > Of course, tar was also removed in susv3, in deference to pax. You've > all heard of pax, right? > > -- > 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: Bug in gzip's stdout handling
> -Original Message-
> From: Igor Pechtchanski [mailto:[EMAIL PROTECTED]
> Sent: 14. november 2003 17:22
> To: Thomas Hammer
> Cc: [EMAIL PROTECTED]
> Subject: RE: Bug in gzip's stdout handling
>
<...>
Hi Igor,
This turned out to be a very long mail, so here's the summary:
- I figured it out. I had forgotten to delete the original mount table from
the registry before reinstalling cygwin. All my drives were still mounted in
textmode, and that's why cat and redirection failed.
- I don't know why mount and cygcheck -svr gives different reports regarding
mount mode.
- I have one question it would be nice if you answered. It's at the end of
the email.
And here's the rest of the email:
>
> (2) shows that you used to have text mounts. Your cygcheck.out shows
> binary mounts, so that's probably not the problem.
I forgot to mention that I reverted to my original version of cygwin (with
DOS-style newlines) when I discovered that using UNIX-style newlines didn't
fix the problem.
I did the following:
1) Renamed c:\cygwin to c:\cygwin_old
2) Reinstalled cygwin, specifying UNIX-style newlines when asked
3) Test if "cat file.bin | gzip -c > filecopy.gz" produced a valig gz-file.
It didn't
4) Deleted c:\cygwin and renamed c:\cygwin_old to c:\cygwin
I did not delete the registry keys or the local cygwin package directory. I
don't know is this matters or not.
I didn't know about text vs binary mounts until you mentioned it, and did
some reading up. One weird thing I came accross is that if I run mount, the
information seems to conflict with the information from cygcheck -svr.
$ mount
C:\cygwin\bin on /usr/bin type system (textmode)
C:\cygwin\lib on /usr/lib type system (textmode)
C:\cygwin on / type system (textmode)
c: on /cygdrive/c type user (textmode,noumount)
s: on /cygdrive/s type user (textmode,noumount)
The relevant lines from cygcheck -svr:
a: fd N/AN/A
c: hd NTFS 57231Mb 47% CP CS UN PA FC
d: cd N/AN/A
o: net N/AN/A
s: net NTFS 76308Mb 76% CP CS UN PA FC
C:\cygwin / system binmode
C:\cygwin/bin /usr/bin system binmode
C:\cygwin/lib /usr/lib system binmode
. /cygdrive system binmode,cygdrive
It looks to me as if mount claims all mounts are text mounts (which would
explain my problems, I guess). Whereas cygcheck -svr claims all my mounts
are binmode.
I'm beginning to wonder if I did something wrong when reinstalling cygwin
and specifying binary mode. I'm giving it another try now.
1) Renamed c:\cygwin to c:\cygwin_old
2) Renamed the HKCU\Software\Cygnus Solutions registry key
3) Renamed the local cygwin package dir.
4) Reinstalled cygwin, specifying UNIX-style newlines when asked
5) Test if "cat file.bin | gzip -c > filecopy.gz" produced a valig gz-file.
It didn't.
Running mount gives the same result as before (shows textmode on all
mounts).
I tried installing cygwin (downloaded setup.exe from www.cygwin.com) from
scratch on another computer in my office. That computer also runs WindowsXP.
$HOME wasn't defined here, but it was on my primary computer. Don't know if
that mattered.
On that computer, mount shows all mountpoints as being of type binmode, and
"cat file.bin | gzip -c > out.bin" works as it should.
I have no Idea why my primary computer insists on mounting everything in
textmode :-(.
I did try to remove my .bashrc-file on my primary computer. It didn't help -
and I couldn't find anything in there related to mounting.
I wonder where the mount table is stored. Maybe it for some reason survived
a reinstall of cygwin...
Please read on for some more discoveries.
>
> Please try the following: "sed '' binaryfile.bin > acopy.bin" and compare
> the files. Also "echo | sed '' > test.out; od -c test.out" and "perl -e
> 'print q{ }x2560' | gzip -c > test.gz; od -c test.gz".
[EMAIL PROTECTED] /cygdrive/c/temp/temp
$ echo | sed '' >test.out
[EMAIL PROTECTED] /cygdrive/c/temp/temp
$ od -c test.out
000 \r \n
002
[EMAIL PROTECTED] /cygdrive/c/temp/temp
$ perl -e 'print q{ }x2560' | gzip -c > test.gz
[EMAIL PROTECTED] /cygdrive/c/temp/temp
$ od -c test.gz
000 037 213 \b \0 = . 271 ? \0 003 355 301 201 \0 \0 \0
020 \0 303 225 371 S 036 344 U 001 \0 360 d 6 271 357
040 252 \0 \r \n \0 \0
046
Back to the mounting thread of thought.
I tried to mount a directory as binary:
$ mount -b c:\\temp\\temp\\t /mytest
mount: warning - /mytest does not exist.
$ cd /mytest
[EMAIL PROTECTED] /mytest
$ echo | sed '' >test.out; od -c test.out
000 \n
001
Hey, I'm getting somewhere :-).
[EMAIL PROTECTED] /mytest
$ ls -l
total 109
-rwx--1 thammer mkgroup110755 Nov 17 21:51 bin.jpg
[EMAIL PROTECTED] /mytest
$ cat bin.jpg | gzip -c > another.jpg.gz
[EMAIL PROTECTED] /mytest
$ gunzip another.jpg.gz
[EMAIL PROTECTED] /mytest
$ ls -l
total 218
-rw-r--r--1 thammer mkgroup110755 Nov 17 22:08 another.jpg
-rwx--1 tha
RE: Perl CPAN module help
Thank you so much. If I wanted to start fresh with a new perl installation -- replacing the executables and all the modules -- how do you recommend I do this under cygwin. I hadn't installed too many modules and it would be nicer to start clean and set it up to use /usr/local right from the start for everything. I really like setup.exe to install things, it is sweet. Thanks again! -Original Message- From: Peter J. Acklam [mailto:[EMAIL PROTECTED] Sent: Monday, November 17, 2003 5:13 AM To: Gary Nielson Cc: Peter J. Acklam; [EMAIL PROTECTED] Subject: Re: Perl CPAN module help "Gary Nielson" <[EMAIL PROTECTED]> wrote: > Thank you for your help. I understand what you are saying > here. I will try installing under /usr/local. I forgot to mention that Perl will not, by default, search for modules in /usr/local. This is a disadvantage, but it's worth it, in my opinion. The simplest way to make Perl look for modules there is to add PERL5LIB=/usr/local/lib/perl5:/usr/local/lib/perl5/site_perl to your personal startup file (~/.bash_profile or whatever). > My question, though, is what do I do about all the modules I've > installed under /usr? How do I deal with them? Do I need to > re-install them under /usr/local? Do I then need to somehow > remove them from /usr? Or can I have modules under both > directories, leaving the ones that work under /usr and place new > ones under /usr/local? If they're working properly, you might as well leave them alone. If they're broken, re-install them with PREFIX=/usr/local. Note that perl will search the directories specified in PERL5LIB before the other directories, so if you have a working module under /usr/local and a broken one under /usr, then the working one under /usr/local will be used. The default search path: $ unset PERL5LIB $ perl -wle 'print for @INC' /usr/lib/perl5/5.8.0/cygwin-multi-64int /usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl . Now, note how /usr/local/... comes first: $ export PERL5LIB=/usr/local/lib/perl5:/usr/local/lib/perl5/site_perl $ perl -wle 'print for @INC' /usr/local/lib/perl5/5.8.0/cygwin-multi-64int /usr/local/lib/perl5/5.8.0 /usr/local/lib/perl5 /usr/local/lib/perl5/site_perl/5.8.0/cygwin-multi-64int /usr/local/lib/perl5/site_perl/5.8.0 /usr/local/lib/perl5/site_perl /usr/lib/perl5/5.8.0/cygwin-multi-64int /usr/lib/perl5/5.8.0 /usr/lib/perl5/site_perl/5.8.0/cygwin-multi-64int /usr/lib/perl5/site_perl/5.8.0 /usr/lib/perl5/site_perl . You may try to remove modules under /usr, but be careful so you don't remove things you need. If you really want to remove them, I'd rather do that by uninstalling Perl, removing everything under /usr/lib/perl5 and start over with a clean Perl installation. Peter -- Peter J. Acklam - [EMAIL PROTECTED] - http://home.online.no/~pjacklam -- 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/
mounting
I am having problems mounting files. This is what I do. $mkdir "C:\PalmDev" $mount -tf "C:\PalmDev" /PalmDev $ln -s "C:\Palm OS 5 SDK (68K) R3\Palm OS Support" /PalmDev/sdk-5r3 when I run mount -tf "C:\PalmDev" /PalmDev it does not mount to '/' instead it is mounted on "/cygdrive/c/Documents and Settings/erica". when I run ln -s I get the following error message: ln: creating symbolic link `/PalmDev/sdk-5r3' to `C:\\Palm OS 5 SDK (68K) R3\\Pa lm OS Support': No such file or directory what is going on here? Thanks! -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: mounting
On Mon, Nov 17, 2003 at 05:43:46PM -0600, Erica A Ramsey wrote: >I am having problems mounting files. This is what I do. > >$mkdir "C:\PalmDev" >$mount -tf "C:\PalmDev" /PalmDev >$ln -s "C:\Palm OS 5 SDK (68K) R3\Palm OS Support" /PalmDev/sdk-5r3 > >when I run mount -tf "C:\PalmDev" /PalmDev it does not mount to '/' instead >it is mounted on "/cygdrive/c/Documents and Settings/erica". when I run >ln -s I get the following error message: >ln: creating symbolic link `/PalmDev/sdk-5r3' to `C:\\Palm OS 5 SDK (68K) >R3\\Pa >lm OS Support': No such file or directory > >what is going on here? Please provide the information asked for in: http://cygwin.com/problems.html -- Please use the resources at cygwin.com rather than sending personal email. Special for spam email harvesters: send email to [EMAIL PROTECTED] and be permanently blocked from mailing lists at sources.redhat.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/
Diff for generic readme and generic-build script
All, Based on the 1.8 version of the generic build script I would like to submit this patch, which would: 1/ the option "all" does not call "list". The fix in the patch does. 2/ The patches to both file let the PKG, VER, and REL variables in the README be automatically be filled in by the script - then maintainers won't have to manually do this. 3/ This patch fulfills the wish to have the file names be automatically be placed in the README prior to binary/src build releases. The 1.8 version heads in that direction, but the functionality isn't there. 4/ I have defined a NEWVER variable to handle the part of the original README 5/ Defined a new export variable: 'ThePackageReadMeFile' which defines the Package README file name (saves defining it twice) - both 'list' and 'install' use it 6/ I haven't renamed the routine "list", which it should be since it is a "package" readme editor 7/ Note: the -i option to sed (quoting the man page for sed): -i[suffix], --in-place[=suffix] edit files in place (makes backup if extension supplied) Extended sed command: `e [COMMAND]' This command allows one to pipe input from a shell command into pattern space. Without parameters, the `e' command executes the command that is found in pattern space and replaces the pattern space with the output; a trailing new-line is suppressed. If a parameter is specified, instead, the `e' command interprets it as a command and sends it to the output stream (like `r' does). The command can run across multiple lines, all but the last ending with a back-slash. In both cases, the results are undefined if the command to be executed contains a NUL character. These together constitute the operation of the modifications. _ Alan Miles ICQ#: 171006836 More ways to contact me: http://wwp.icq.com/171006836 _ packaging_templates.diff Description: Binary data -- 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/
df reports negative values on Network Shares
Hi all, this may or may not be a bug in fileutils. Here are the symptoms: $ df Filesystem 1k-blocks Used Available Use% Mounted on C:\cygwin\usr\X11R6\lib\X11\fonts 19542568 7468884 12073684 39% /usr/X11R6/lib/X11/fonts C:\cygwin\bin 19542568 7468884 12073684 39% /usr/bin C:\cygwin\lib 19542568 7468884 12073684 39% /usr/lib C:\cygwin 19542568 7468884 12073684 39% / c:19542568 7468884 12073684 39% /c m: 307200 -73786976294741950464 7308 101% /m g: 512000 -73786976294738780160350176 101% /cygdrive/g j: 512000 -73786976294738780160350176 101% /cygdrive/j n:62693276 53949172 8744104 87% /cygdrive/n p: 512000 -73786976294738780160350176 101% /cygdrive/p z: 512000 -73786976294738780160350176 101% /cygdrive/z [EMAIL PROTECTED]:~ $ df -h FilesystemSize Used Avail Use% Mounted on C:\cygwin\usr\X11R6\lib\X11\fonts 19G 7.2G 11G 39% /usr/X11R6/lib/X11/fonts C:\cygwin\bin 19G 7.2G 11G 39% /usr/bin C:\cygwin\lib 19G 7.2G 11G 39% /usr/lib C:\cygwin 19G 7.2G 11G 39% / c: 19G 7.2G 11G 39% /c m:300M -64Z 7.1M 101% /m g:500M -64Z 341M 101% /cygdrive/g j:500M -64Z 341M 101% /cygdrive/j n: 60G 52G 8.3G 87% /cygdrive/n p:500M -64Z 341M 101% /cygdrive/p z:500M -64Z 341M 101% /cygdrive/z $ df --version df (fileutils) 4.1 Written by Torbjorn Granlund, David MacKenzie, Larry McVoy, and Paul Eggert. Copyright (C) 2001 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. [EMAIL PROTECTED]:~ $ uname -a CYGWIN_NT-5.1 KFIDEMUC110528 1.5.5(0.94/3/2) 2003-09-20 16:31 i686 unknown unknown Cygwin All the drives with the -64Z (zillion?) are on the same server subsystem, n: is on a different server. The explorer show reasonable values under Properties, e.g. 158MB used, 341MB free, and 500MB Capacity. The 500MB for do not represent a physical value, but most likely a quota limit. The values for n: are consistent with what the explorer claims. Any hints where to look further into this? Best regards / Mit freundlichen Grüssen, Thomas Demmer Kraft Foods R&D Inc. Tel.: +49 (0)89 62738-6302 Fax: +49 (0)89 62738-86302 mailto:[EMAIL PROTECTED] Thought of the day A good plan today is better than a perfect plan tomorrow. -- Patton -- 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/
