This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
Re: separated debuginfo patch
Hi Jim,
>> > I need to know how GDB guys want I deal with the gdb part, for now
>> > gdb.diff just remove (#if 0) all duplicated code from bfd and use
>> > bfd_follow_gnu_debuglink() to retrieve the debug info file. Is it
>> > ok to remove this code or must I update the duplicated code according
>> > to the change in bfd ?
>>
>> Well this is up to the gdb maintainers to decide, but it certainly
>> seems like a good idea to avoid the code duplication.
>
> Sure, the plan has long been for GDB to just use the function in BFD.
> The code was added to GDB before BFD; that's the only reason it's
> there at all.
>
> Just to be sure --- under this arrangement, the old-style debug links
> will continue to work, right?
Yes.
> One could use something like '(date; ps auxww; vmstat) | md5sum | cut
> -b 1-33' to generate nice unique ID strings.
yikes! Well that ought to be a reasonably randon number. The only
problem would be making sure that the same number was added to both
the debuginfo file and the stripped binary. I guess you would need to
make sure your makefile only computed the value once, before it
creates either file.
Cheers
Nick