Re: [PATCH 1/2 RFC] rpm: Add workaround for debugedit-segv

[Mark Hatle <mark.hatle@windriver.com>]

On 3/25/13 12:45 PM, Phil Blundell wrote:
> On Mon, 2013-03-25 at 12:10 -0500, Mark Hatle wrote:
>> I'm going to look into identifying if the section is a NOBITS and skipping the
>> whole operation if it is.
>
> That would mean that a change in the size of the .bss wouldn't have any
> impact on the hash.  Maybe that's fine for your application though, I
> dunno.

I'm not completely familiar with how the buildid is calculated, other then an 
md5 hash over the sections themselves.  I can't think of a case where the 
contents of the sections wouldn't change along with the .bss and .plt size.

(buildid is supposed to be used by gdb to find/verify that the debuginfo matches 
the binary actually being debugged...)

Through my google search, I never found a full spec of what the buildid was 
supposed to contain, only a high-level description from an old Fedora 8 work 
item.  (Note, as long the buildid is consistent it really doesn't appear to 
matter how it was generated, just that both the app and debuginfo contain it.)

--Mark


> p.
>
>