-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Package: libc6-sparc64
Version: 2.3.2.ds1-13
Package libc6-sparc64 has no reason to depend on lib64gcc1. None of
libraries provided by libc6-sparc64 are linked agains this library.
With the current situation, I'm getting checken-and-egg problem whil
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> At Sun, 11 Jul 2004 00:51:14 +0400,
>
> Nikita V. Youshchenko <[EMAIL PROTECTED]> wrote:
> > Package: libc6-sparc64
> > Version: 2.3.2.ds1-13
> >
> > Package libc6-sparc64 has no reason to depend on lib64
> so we can start converting the packages you
> mentioned to biarch.
Oh, please do!
This at last will make it possible to build biarch gcc cross compiler
package for x86 target in the debian way (i.e. using current
dpkg-cross-based approach).
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a
> BTW, Nikita, dpkg-shlibdeps in dpkg-cross should have additional elf64
> entries for ppc64 and amd64 like:
>
> @crosslib64formats = ("elf64-sparc", "elf64-s390", "elf64-x86-64",
> "elf64-powerpc");
Thank you for mentioning that.
I've committed this change.
However, before new upload of dpk
> objdump isn't a solution either, while it sometimes can read the other
> shared library, it doesn't provide the linker search patch which is of
> critical importance to get this stuff right.
Hmm...
But what for is that search path?
As far as I understand, dpkg-shlibdeps should just get NEEDED so
Package: libc6
Version: 2.3.5-6
Severity: normal
[EMAIL PROTECTED]:~> echo -n | fakeroot awk '{print}'
awk: relocation error: awk: symbol _dl_catch_error, version GLIBC_PRIVATE not
defined in file ld-linux.so.2 with link time reference
This is fixed in post-sarge gawk 3.1.4-2.0.1.
This may affe
Hello people.
> > > objdump isn't a solution either, while it sometimes can read the
> > > other shared library, it doesn't provide the linker search patch
> > > which is of critical importance to get this stuff right.
> >
> > Hmm...
> > But what for is that search path?
> > As far as I understand
> On Sun, Jan 22, 2006 at 10:14:16PM +0100, Frank Lichtenheld wrote:
> > I've implemented this option. Patch and new script (since the patch is
> > garbled with a little code clean-up I did while going through the
> > script) are attached.
> >
> > Comments and/or testing welcome.
>
> Comment to my
> On Tue, Jan 24, 2006 at 06:43:02PM +0300, Nikita V. Youshchenko wrote:
> > dpkg-shlibdeps calls ldd, which will just fail if LD_LIBRARY_PATH
> > won't point to directories with local libraries.
>
> That's not true. ldd will just happily print "libfoo.so.1 =&g
> On Tue, Jan 24, 2006 at 06:05:43PM +0300, Nikita V. Youshchenko wrote:
> > > On Sun, Jan 22, 2006 at 10:14:16PM +0100, Frank Lichtenheld wrote:
> > > Comment to myself: The current patch probably breaks dh_shlibdeps
> > > -l option because it doesn't
Package: libc6-dev
Version: 2.3.5-12.1
Severity: important
After upgrade libc6 to 2.3.5-12.1 our locat projects failed to build.
Minimal test that reproduces the problem:
[EMAIL PROTECTED]:~/adm/libc> cat t.c
#include
void at() {}
int f(void)
{
atexit(&at);
return 0;
}
[EMAIL
> Removing the databases in /var/db/nscd/ resolved the problem.
So why not just remove old files on nscd package upgrade?..
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
found 351629 2.3.6-7
thanks
Again this bug. Have to downgrade libc6 from snapshots.debian.net ... :(
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: linux-kernel-headers
Version: 2.6.18-3
Severity: normal
/usr/include/asm/system.h file used to provide access to useful
low-level x86 facilities, such as cmpxchg CPU instruction.
Only add #include to program and then use it.
However, this stopped working after recent upgrade of
linux-k
Package: libc6
Version: 2.3.2-8
Severity: normal
UTF-8 encoding is specified in RFC2279 as follows:
UCS-4 range (hex.) UTF-8 octet sequence (binary)
- 007F 0xxx
0080- 07FF 110x 10xx
0800- 1110 10xx 10xx
00
Package: libc6
Version: 2.3.2-8
Severity: normal
Several time ago, what I was migrating my systems to LDAP, I noticed
some nss-related trouble (e.g. finger was unable to show information
about LDAP users, although /etc/nsswitch was set correctly, and other
tools worked).
I even reported several bu
First, this bug is still there in current libc6 from unstable:
[EMAIL PROTECTED]:~> cat t.c
#include
[EMAIL PROTECTED]:~> gcc-2.95 -c -Wall t.c
In file included from /usr/include/asm/sigcontext.h:4,
from /usr/include/bits/sigcontext.h:28,
from /usr/include/signa
Package: libc6
Version 2.1.2-5
Timezones that are provided with libc6 do not have a symlink
/usr/share/zoneinfo/localtime -> /etc/localtime
But this symlink is required for libc5 compatability. Without it, e.g.
libc5 netscape shows incorrect date and time in mail windows.
I have reported this
Hello
Is there any way to make jdk 1.3.1 to work with the new libc? Maybe some
hack?
Currently, it fails with the following message:
Error occurred during initialization of VM
Unable to load native library:
/usr/opt/sun-jdk-1.3.1/jre/lib/i386/libjava.so: symbol __libc_wait, version
GLIBC_2.0
> On Tue, Oct 29, 2002 at 06:14:29PM +0300, Nikita V. Youshchenko wrote:
> > Is there any way to make jdk 1.3.1 to work with the new libc? Maybe
> > some hack?
>
> Not in any sort of supported way. If you need 1.3.1, perhaps try the
> blackdown jvm - It still works.
Thank you. It really works.
>
> When libc6 installs it forgets to restart XDM. On my server, I had to
> restart it manually because none of my users who access the system via
> XDM where able to login
It is not safe to restart xdm. It may cause running sessions to fail
immidiatly.
In general, seems that libc 2.2 -> glibc
Package: libc6-dev
Version: 2.3.1-5
Severity: normal
Libc documentation states that trunc() function is declared in math.h.
However, math.h provided by current libc6-dev does not declared in
math.h or any header included from math.h
> cat x.cpp
#include
int main()
{
double d = trunc(3.5)
Hello
Is there any way to make jdk 1.3.1 to work with the new libc? Maybe some
hack?
Currently, it fails with the following message:
Error occurred during initialization of VM
Unable to load native library:
/usr/opt/sun-jdk-1.3.1/jre/lib/i386/libjava.so: symbol __libc_wait, version
GLIBC_2.0
> On Tue, Oct 29, 2002 at 06:14:29PM +0300, Nikita V. Youshchenko wrote:
> > Is there any way to make jdk 1.3.1 to work with the new libc? Maybe
> > some hack?
>
> Not in any sort of supported way. If you need 1.3.1, perhaps try the
> blackdown jvm - It still works.
Th
>
> When libc6 installs it forgets to restart XDM. On my server, I had to
> restart it manually because none of my users who access the system via
> XDM where able to login
It is not safe to restart xdm. It may cause running sessions to fail
immidiatly.
In general, seems that libc 2.2 -> glibc
Package: libc6-dev
Version: 2.3.1-5
Severity: normal
Libc documentation states that trunc() function is declared in math.h.
However, math.h provided by current libc6-dev does not declared in
math.h or any header included from math.h
> cat x.cpp
#include
int main()
{
double d = trunc(3.5)
Package: libc6
Version: 2.3.2-8
Severity: normal
UTF-8 encoding is specified in RFC2279 as follows:
UCS-4 range (hex.) UTF-8 octet sequence (binary)
- 007F 0xxx
0080- 07FF 110x 10xx
0800- 1110 10xx 10xx
00
Package: libc6
Version: 2.3.2-8
Severity: normal
Several time ago, what I was migrating my systems to LDAP, I noticed
some nss-related trouble (e.g. finger was unable to show information
about LDAP users, although /etc/nsswitch was set correctly, and other
tools worked).
I even reported several bu
First, this bug is still there in current libc6 from unstable:
[EMAIL PROTECTED]:~> cat t.c
#include
[EMAIL PROTECTED]:~> gcc-2.95 -c -Wall t.c
In file included from /usr/include/asm/sigcontext.h:4,
from /usr/include/bits/sigcontext.h:28,
from /usr/include/signa
Hello.
I have libc6-686 package installed, and most programs work correctly with
optimized libraries.
However, there are at least two examples of program crashes when optimized
libraries are used (one is mencoder, another is crossopver office running
word).
It is possible to make individual p
Hello.
I have libc6-686 package installed, and most programs work correctly with
optimized libraries.
However, there are at least two examples of program crashes when optimized
libraries are used (one is mencoder, another is crossopver office running
word).
It is possible to make individual p
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Package: libc6-sparc64
Version: 2.3.2.ds1-13
Package libc6-sparc64 has no reason to depend on lib64gcc1. None of
libraries provided by libc6-sparc64 are linked agains this library.
With the current situation, I'm getting checken-and-egg problem whil
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
> At Sun, 11 Jul 2004 00:51:14 +0400,
>
> Nikita V. Youshchenko <[EMAIL PROTECTED]> wrote:
> > Package: libc6-sparc64
> > Version: 2.3.2.ds1-13
> >
> > Package libc6-sparc64 has no reason to depend on lib64
> My libnss-ldap.conf isn't world readable, to protect my LDAP password.
Does 'finger user' work under non-root with such configuration?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
34 matches
Mail list logo
