PERFORCE change 132434 for review

[Peter Wemm]

On Jan 3, 2008 5:25 PM, Robert Watson <rwatson at freebsd.org> wrote:

> On Thu, 3 Jan 2008, Peter Wemm wrote:
>
> > http://perforce.freebsd.org/chv.cgi?CH=132434
> >
> > Change 132434 by peter at peter_daintree on 2008/01/03 22:24:56
> >
> >       Export the build uuid to the dump header.  Nothing interesting is
> done
> >       with it at this point.
> >
> >       While here, move all the mkdumpheader() copies into a MI place, as
> >       the (duplicated) comment suggested.  Allow magic numbers to be
> passed
> >       through so textdumps can share the code too.
>
> Currently, I duplicate the panic string and kernel version string from the
> dump header in the textdump itself as data files.  Any chance you want to
> do
> the same with the uuid?
>
> (Perhaps a better model would be to have savecore append another file to
> the
> textdump containing the header information, but the above seems to work
> fine
> for now, and has some advantages).
>

I was originally planning to put the uuid in the minidump header.  But, I
believe that savecore will store the uuid in the info.xxxx file.  I have not
actually tested this.

Still, more copies is probably better than less.  Granted, the version
string is probably more than good enough given that it stores a build
timestamp, but it wasn't really meant for that.  The build uuid is meant to
guarantee the ability to conclusively match between a kernel dump and the
actual source kernel file.

-- 
Peter Wemm - peter at wemm.org; peter at FreeBSD.org; peter at yahoo-inc.com
"All of this is for nothing if we don't go to the stars" - JMS/B5