Re: [patch 7/7] Make "inline" no longer mandatory for gcc 4.x

From: Kurt Wall
Date: Sat Jan 07 2006 - 22:15:15 EST


[CCs trimmed because Ingo's mail server says relay denied]

On Sat, Jan 07, 2006 at 08:44:48PM +0100, Arjan van de Ven took 25 lines to write:
> On Sat, 2006-01-07 at 14:05 -0500, Kurt Wall wrote:
>
> >
> > This patch was applied on top of the previous 6 in the series from
> > Arjan. NB that it _did_ build with 3.4.4 and -Os enabled. I'm
> > rechecking, but this is the second time I've encountered this failure.
>
>
> Does this fix it?

Quite. The results are interesting. For my vanilla desktop config with
all patches applied, including the fixmap.h patch for x86_64,
differences are pretty dramatic:

text data bss dec hex filename
2577982 462352 479920 3520254 35b6fe vmlinux.344.NO_OPT
2620255 462336 479984 3562575 365c4f vmlinux.442.NO_OPT
2326785 462352 479920 3269057 31e1c1 vmlinux.344.OPT
2227294 502680 479984 3209958 30fae6 vmlinux.442.OPT

344 : gcc 3.4.4
402 : gcc 4.0.2
NO_OPT: without -Os
OPT : with -Os

Based on .text size alone, if I was going to build and run a kernel
with GCC 4.x, I'd definitely enable -Os.

Kurt
--
The plot was designed in a light vein that somehow became varicose.
-- David Lardner
-
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/