Re: [2.6 patch] kernel/sched*: optimize inlining

From: Sam Ravnborg
Date: Sun May 11 2008 - 17:44:24 EST


On Mon, May 12, 2008 at 12:19:02AM +0300, Adrian Bunk wrote:
> On Sun, May 11, 2008 at 10:38:27PM +0200, Sam Ravnborg wrote:
> > > >
> > > > You continue to fail to acknowledge that it is valueable information
> > > > that we pass gcc a _hint_ that it is a good idea to inline certain
> > > > functions.
> > > >
> > > > The inline hint is there to tell gcc that it shall inline this function
> > > > in cases where it usual think it should not do so. Which invietably
> > > > will result in a larger codesize in some cases.
> > >
> > > We also give gcc an explicit "Optimize for size.".
> >
> > gcc was asked to optimize for size in general as per the commandline option.
> > But on a much more fine grained level gcc is given a hint that
> > this function would be a good idea to inline.
> >
> > And I really expect gcc to pay most attention to the more specific
> > information provided for a single function than a general commandline option.
>
> Can you try to get from expectations back to reality?

What I wrote is based on common sense.
Let me know if the gcc community does not agree.

> gcc 4.3 even ignores the unlikely() hint in timespec_add_ns()
> (we now have a workaround for this in the kernel).
I do not follow the logic here.
Gcc may fail in a few cases to do what we expect but that
is far from that we shall assume that it always fails.

>
> >...
> > > All the "optimized inlining" does is to allow gcc to no longer inline
> > > functions marked as "inline" if it prefers not to do so.
> > The "optimized inlining" allows gcc (if gcc > 4.0) to make an educated
> > guess if it is worhtwhile to inline a function or not. And when deciding
> > to do so or not gcc may include many different factors inlcuding
> > but not limited to -s.
> > And this is certainly optimized compared to the situation where
> > inline equals to always_inline.
> > Keep in mind that we often perfer to have _less_ inlining than we have
> > today for debugging ease. And this is what we get with optimized inlining
> > compared to farced inlining.
> >
> > >
> > > And what exactly is your problem with my patch if you consider the
> > > general "optimized inlining" approach correct?
> >
> > I've already listed two things:
> > -> It is no longer a simple kconfig change to try with or without.
> > -> It is independent on gcc version
>
> I already asked you previously in this thread:

And you fail to comment why both points are not worth considering.

>
> Do we have any hard data that gcc < 4.0 has a "broken inline algorithm"
> and all gcc versions >= 4.0 have a "working inline algorithm"?

Is it hard data for you that Linus says that gcc < 4.0 is "broken"
so yes. Search the archives.
If you expect me to show you a lot of disassembly then no.

>
> > And for fast path code like sched.c I would much assume a proper analysis
> > when it is acceptable to remove the inline hint is almost mandatory.
> >...
>
> Why didn't you request a proper analysis before the "optimized inlining"
> stuff hit Linus' tree?
Adrian - stop this bullshit.
We are discussing _your_ patch. Not some other patch that you
seems to have some hard feelings about. And yes I saw the reference
in the initial patch which I saw no reason to comment on as this
was purely bullshit then and still is so.

Was the purpose of this patch just a provocation then?
If so - then I just lost 50% of my Linux time tonight on it!

Sam
--
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/