Build is verified!

Ryan.Oliver at pha.com.au Ryan.Oliver at pha.com.au
Thu Mar 20 18:00:36 PST 2003


Ian Molton wrote:
> Ok, so, having established that it doesnt garauntee the new compiler is
> bug free (see, that wasnt so hard, was it?) can you please do what I
> asked, and explain WHY it builds a better toolchain that the current
> simple-yet-not-100%-efficient method does?

I can answer this one

1) because we now dont have to worry about what glibc our host system has,
therefore we shouldn't see any issues during a glibc migration.

Remember this was the main reason this build method came about... and with
the present build method this can't be guaranteed. This way the build is
effectively divorced from the host early in the piece...

2) better because our second and subsequent gcc build is built against the
glibc we are migrating too, our build env for ch5 is pretty much the same
as our ch6 build.

3) the toolchain is self hosting.

This is a good and fun thing in and of itself, I currently use mods of this
to run chrooted services on systems via backend NFS mounts, nicely
migratable applications which take their libraries with them, and can run
on linux systems with different base c-library versions making it nice and
easy to shunt them around onto the least loaded system...

Its more a matter of the build now being more resilient to changing
packages, you can never guarantee a bug free gcc, what we can do is
minimise the effects of this...

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list