Re: [PATCH] Generic dead function optimisation

From: Andrew Morton (andrewm@uow.edu.au)
Date: Sun Apr 23 2000 - 20:47:50 EST

Previous message: Oliver Xymoron: "Testers wanted! (was Re: "movb" for spin-unlock)"
In reply to: Pavel Machek: "Re: [PATCH] Generic dead function optimisation"
Next in thread: Stefan Monnier: "Re: [PATCH] Generic dead function optimisation"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Pavel Machek wrote:
>
> Hi!
>
> > > This is a followup to an article about dead function optimisation, which I
> > > posted a couple of weeks ago.
> >
> > x86/SMP: Patch applies cleanly against 2.2.14-pre9. vmlinux ends
> > up with ~300 bytes less text, from one meg.
>
> That is not too bad -- main advantage is that with this patch in we no
> longer have to care about putting #ifdef CONFIG_FOO around functions
> which are needed only sometimes.

I agree, Pavel. (The 300 bytes was bogus, BTW: more like 35 kbytes.
Still small).

As far as I can see, Graham's patches have no downside at all, apart
from the 100x slower link. This is caused by a few little hotspots in
ld - 23 million calls to gldelf_i386_place_section()!. Alan is looking
into it.

-- -akpm-

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/

Previous message: Oliver Xymoron: "Testers wanted! (was Re: "movb" for spin-unlock)"
In reply to: Pavel Machek: "Re: [PATCH] Generic dead function optimisation"
Next in thread: Stefan Monnier: "Re: [PATCH] Generic dead function optimisation"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:22 EST