[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#465028: libqt3-mt: Missing weak symbols for stat64 functions



On Sunday 10 February 2008, Guillem Jover wrote:
> Package: libqt3-mt
> Version: 3:3.3.8b-1
> Severity: serious
>
> Hi,
>
> The Qt shared libraries were providing weak symbols for stat64,
> lstat64 and fstat64 until version 3:3.3.7-9, but not anymore, other
> Qt/KDE apps depend on those symbols, so they get terminated when the
> symbols cannot be resolved during runtime (as the linking is done
> lazyly). Affected applications I've seen terminate during runtime or
> just when starting: gwenview and k3b.

urgh.

I wonder what caused this.  Rebuilding 3:3.3.7-9 in a current sid environment 
also don't have these symbols.

3.3.7 from the archive:
 objdump -T /usr/lib/libqt-mt.so | grep stat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __fxstat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __lxstat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __xstat64
00616a72  w   DF .text  00000032  Base        lstat64
00326a32  w   DF .text  00000032  Base        fstat64
0050abac  w   DF .text  00000032  Base        stat64


3.3.8 from the archive:
$ objdump -T libqt-mt.so.3 | grep stat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __fxstat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __lxstat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __xstat64

locally built 3.3.7:
$ objdump -T libqt-mt.so.3 | grep stat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __fxstat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __lxstat64
00000000      DF *UND*  0000003f  GLIBC_2.2   __xstat64

I wonder what happpened.
Any advice appreciated.

/Sune
-- 
Man, how might I remove a utility to a Fast TCP/IP forward from the control 
file within LinuxPPC 6000?

First of all from Internet Explorer 99 you either should never ping to a 
computer of the hard disk to the Internet site, or cannot boot from a mouse 
in order to link the memory.

Attachment: signature.asc
Description: This is a digitally signed message part.


Reply to: