Re: [PATCH 2/2] headers: reduction in interrupt.h

From: Ingo Molnar
Date: Wed Oct 29 2008 - 02:21:25 EST



* Alexey Dobriyan <adobriyan@xxxxxxxxx> wrote:

> >From 2255be0a5f55ffd58b0ebdf456cf77b7258889fe Mon Sep 17 00:00:00 2001
> From: Alexey Dobriyan <adobriyan@xxxxxxxxx>
> Date: Wed, 29 Oct 2008 06:41:45 +0300
> Subject: [PATCH 2/2] headers: reduction in interrupt.h
>
> 1. remove sched.h from interrupt.h, it's not needed for anything
> 2. remove sched.h from smp_lock.h, so (1) won't be a lie due to
>
> linux/interrupt.h => linux/hardirq.h => linux/smp_lock.h => linux/sched.h
>
> After static inline reacquire_kernel_lock() was moved from there, inclusion
> isn't needed.
> 3. remove also kernel.h, linkage.h, preempt.h, smp.h, asm/ptrace.h inclusions,
> I manually checked every token.
> 4. fix fallout
>
> The end result is that, e.g. i386 allyesconfig kernel shrinks (with all those
> bloated debuginfo sections) from 604.1 MB to 593.1 MB (-1.8 %).
>
> Compile time should also improve simply because compiler will parse less.
>
> Cross-compile tested on 469 configs, big and small, across many arches
> and a dozen of randconfigs.

very nice! Regarding my previous mail, we can move the BKL interfaces
away from sched.c in another patch, i dont want this very cool header
cleanup be held up.

Acked-by: Ingo Molnar <mingo@xxxxxxx>

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