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

Re: Why does gdb now give lots of warnings?



On Jan 4, 2008 10:27 AM, James Antill <james antill redhat com> wrote:
>
> On Fri, 2008-01-04 at 17:15 +0000, Andrew Haley wrote:
> > Jan Kratochvil writes:
> >  > On Fri, 04 Jan 2008 17:53:09 +0100, Andrew Haley wrote:
> >  > > Jan Kratochvil writes:
> >  > ...
> >  > > > OK, you are right Fedora GDB could; the build-id support messages should be
> >  > > > cross-OS ones, this loading feature is still not imported into upstream GDB and
> >  > > > it is heading there.
> >  > >
> >  > > Sure, but we could have a simple local message like
> >  > >
> >  > > Try "yum install /usr/lib/debug/.build-id/72/67a2ecd318b0f87a0747a6986d0d6dc01c6d8d.debug"
> >  > >
> >  > > I learnt only today that would work...
> >  >
> >  > Such as this one - just for the first such message printed?  I guess changing
> >  > the `Missing ...' message itself would be already too OS-specific.
> >
> > Perfection would be:
> >
> > Missing debuginfo for /lib64/libnss_files-2.7.so
> > Try "yum install /usr/lib/debug/.build-id/72/67a2ecd318b0f87a0747a6986d0d6dc01c6d8d.debug"
> >
> > If we can't do that upstream, then the closer we get, the better.  Or
> > we have a local patch.
>
>  Well what we really want here is:
>
>  Try "debuginfo-install glibc" (or "debuginfo-install python" if they
> are running gdb on that), but that requires some kind of integration
> with rpm or yum.
>
Interesting.....

I tried 'debuginfo-install rhythmbox' and got a list of 43 packages..... cool.

However, it looks like it wants to install 3 'real' packages:

Installing:
 GConf2-debuginfo        i386       2.20.1-4.fc9
development-debuginfo  553 k
 ORBit2-debuginfo        i386       2.14.10-2.fc8
development-debuginfo  838 k
<<<<<SNIP>>>>>
Installing for dependencies:
 avahi-gobject           i386       0.6.22-4.fc9     development        26 k
 avahi-qt3               i386       0.6.22-4.fc9     development        19 k
 minizip                 i386       1.2.3-16.fc9     development        23 k

That seems strange to me.  Is that to be expected?

tom
-- 
Tom London


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