Re: [PATCH/RFC] m68k: Add full preempt support

[Geert Uytterhoeven]

Hi Adrian,

On Mon, Oct 9, 2023 at 1:45 PM John Paul Adrian Glaubitz
<glaubitz@xxxxxxxxxxxxxxxxxxx> wrote:
> On Mon, 2023-10-09 at 13:29 +0200, Geert Uytterhoeven wrote:
> > On Fri, Oct 6, 2023 at 4:49 PM Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> > > Currently, a Preemptible Kernel is only supported on the Coldfire CPU
> > > family.  Extend preempt support to the full Classic M68K CPU family
> > > (68020+ with MMU, and 68000-derivatives without MMU).
> > >
> > > Make sure preemption is disabled in loops involving cache and TLB
> > > flushing.
> > >
> > > Signed-off-by: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
> > > ---
> > > Tested on ARAnyM, using none/voluntary/full preemption.
> >
> > bloat-o-meter summaries for atari_defconfig:
> >
> > PREEMPT_NONE -> PREEMPT_VOLUNTARY:
> >
> >     add/remove: 0/0 grow/shrink: 238/30 up/down: 2804/-748 (2056)
> >     Total: Before=4478149, After=4480205, chg +0.05%
> >
> > PREEMPT_VOLUNTARY -> PREEMPT:
> >
> >     add/remove: 735/277 grow/shrink: 3050/347 up/down: 244359/-11440 (232919)
> >     Total: Before=4480205, After=4713124, chg +5.20%
> >
> > Ouch.
>
> What exactly does the change document? More scheduling events?

No, kernel image size.  Hence enabling full preemption increases kernel
size by ca. 229 KiB (which is more or less the equivalent size increase of
upgrading the kernel by 10 releases).

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds