-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Carl Lowenstein wrote:
> On Nov 15, 2007 10:59 AM, Darren New <[EMAIL PROTECTED]> wrote:
>> Carl Lowenstein wrote:
> I do believe that you have forgotten the initial goal -- to embed ID
> strings in separately compiled program pieces with a minimum of extra
> effort.

I've never needed this, and I was trying to think of why. I haven't
worked on embedded systems in ages, so that might be part of it. When I
think about it though, the way I really verify what is on disk is I
check file sizes and look at checksums. Indeed, by having this as part
of the install process I never run in to situations where I have
deployments that are half of one release and half of another (a scenario
that was alluded to earlier). Is doing a checksum too much work for the
CPU (and if so, is pattern matching really lighter weight)?

> In fact, it seems to me that the aggressive optimizer would remove
> definitions of external functions if they are never called.

Even the most aggressive optimizers would have difficulty being certain
that an external function is never called... particularly if it is in
fact called.

- --Chris
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHPeGcOagjPOywMBARAk8JAJ9QKoNShS1FCgAcEDOCSQY84FZR2gCgtIe1
gMXbnzxyfPGzgU3BeTVQJMA=
=Xh1q
-----END PGP SIGNATURE-----

-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to