role of executable_checksum & LTO?

[Andrew Pinski]

On Jun 29, 2010, at 9:35 AM, Basile Starynkevitch <basile@starynkevitch.net 
 > wrote:

> On Tue, 2010-06-29 at 07:02 -0700, Ian Lance Taylor wrote:
>> Basile Starynkevitch <basile@starynkevitch.net> writes:
>> I think it would be fairly difficult to construct a case where a  
>> plugin
>> cared about the exact compiler, rather than the exact version and
>> configuration.  The only reason that the PCH support cares about the
>> exact compiler is that it more or less takes a memory dump of the
>> allocated GC memory.
>
>
> I agree, but a plugin could also do likewise, e.g. write memory  
> contents
> in some kind of persistent storage.
>
> Beside, the hack is quite easy to implement (just add
> executable_checksum to lto1 binary, or perhaps generate an md5sum of
> most GCC files and store it).
>
> While I can understand the reason, I still find strange and
> counter-intuitive that lto1 lacks executable_checksum that cc1 has.

How so f951 lacks it too so does gnat1 and gcj1. And any other  
frontend besides cc1, cc1plus, cc1obj and cc1plusobj. It is like  
saying lto1 is missing c_build* too.

>
> Cheers.
> -- 
> Basile STARYNKEVITCH         http://starynkevitch.net/Basile/
> email: basile<at>starynkevitch<dot>net mobile: +33 6 8501 2359
> 8, rue de la Faiencerie, 92340 Bourg La Reine, France
> *** opinions {are only mines, sont seulement les miennes} ***
>
>