From: Nicolas Pitre <nico@fluxnic.net>
To: Kim Phillips <kim.phillips@freescale.com>
Cc: "Woodhouse, David" <david.woodhouse@intel.com>,
Russell King - ARM Linux <linux@arm.linux.org.uk>,
Borislav Petkov <bp@alien8.de>,
Andrew Morton <akpm@linux-foundation.org>,
Daniel Santos <daniel.santos@pobox.com>,
David Rientjes <rientjes@google.com>,
Rusty Russell <rusty@rustcorp.com.au>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Rob Herring <robherring2@gmail.com>
Subject: Re: [RFC] arm: use built-in byte swap function
Date: Fri, 08 Feb 2013 22:16:47 -0500 (EST) [thread overview]
Message-ID: <alpine.LFD.2.02.1302082102380.6300@xanadu.home> (raw)
In-Reply-To: <20130208191208.2ef3d78bda71aa7b44d00d7b@freescale.com>
On Fri, 8 Feb 2013, Kim Phillips wrote:
> On Fri, 8 Feb 2013 17:47:33 -0500
> Nicolas Pitre <nico@fluxnic.net> wrote:
>
> > On Fri, 8 Feb 2013, Woodhouse, David wrote:
> >
> > > On Fri, 2013-02-08 at 15:04 -0500, Nicolas Pitre wrote:
> > > > On Fri, 8 Feb 2013, Woodhouse, David wrote:
> > > >
> > > > > On Thu, 2013-02-07 at 18:13 +0000, Russell King - ARM Linux wrote:
> > > > > >
> > > > > > However, the biggest reason not to use libgcc is that we want to control
> > > > > > what gets used in the kernel - for example, no floating point, and no
> > > > > > use of 64 x 64bit division.
> > > > >
> > > > > Which is all very sensible. But there's no particular reason we couldn't
> > > > > add a __bswap[sd]i2 to the kernel's version of libgcc if we wanted to.
> > > >
> > > > Absolutely.
> > >
> > > And then ARM can just set ARCH_USE_BUILTIN_BSWAP like other
> > > architectures do, right?
> >
> > If that turns out to be beneficial over what we have now, then yes.
> > I didn't read back the whole thread to form an opinion though.
>
> The diff below implements __bswap[sd]i2 in arch/arm/lib, and
> results in the following savings in vmlinux size:
>
> column 1: name of defconfig
> column 2: text+data+bss, linux-next-20130207 vanilla, gcc 4.6.3
> column 3: text+data+bss, linux-next-20130207+below diff, gcc 4.6.3
> column 4: col. 3 - col. 2 (ie., -ve numbers represent savings)
>
[...]
> imx_v6_v7_defconfig: 7672373 7667089 -5284
> lart_defconfig: 2941150 2941054 -96
> mxs_defconfig: 11091983 11095679 3696
The savings are good, with some impressive cases. However the
mxs_defconfig is completely the opposite and by far. Any idea?
> gcc 4.7.3 runs haven't been as good across the board as gcc 4.6.3,
> however:
Not only that, but in many cases the results are wildly different given
the same config:
> imx_v6_v7_defconfig: 7637605 7636935 -670
> lart_defconfig: 2922550 2926600 4050
> mxs_defconfig: 11071139 11070893 -246
The mxs_defconfig became much better while lart_defconfig regressed a
lot.
> Haven't looked at why.
Would be a good idea since this is rather weird and gcc could benefit
from your findings.
> In any case, some questions I have are:
>
> (a) are the __bswap[sd]i2 implementations acceptable written in C,
> as in the diff? I don't speak ARM asm (yet at least). The
> generated code looks pretty optimal in both armv5 and 6+.
It looks pretty nice indeed:
__bswapsi2:
eor r2, r0, r0, ror #16
mov r1, r2, lsr #8
bic r3, r1, #65280
eor r0, r3, r0, ror #8
bx lr
There is no way to do better than that. But that's true only if -Os is
_not_ used. With -Os we get the following output:
__bswapsi2:
mov r3, r0, asl #24
and r2, r0, #65280
orr r3, r3, r0, lsr #24
orr r3, r3, r2, asl #8
and r0, r0, #16711680
orr r0, r3, r0, lsr #8
bx lr
I really don't get why gcc thinks the above is shorter. I'm saving you
from pasting the __bswapdi2 result which is also way way worse.
That was with Linaro gcc v4.6.2.
With Sourcery gcc v4.5.1 we get:
__bswapsi2:
stmfd sp!, {r3, lr}
bl __bswapsi2
ldmfd sp!, {r3, pc}
This is indeed shorter, but much less useful. So you better enforce -O2
for this file. And the nice thing with C code is that it is fully
optimized with the rev instruction when compiling for ARMv6+ if it is
ever used in that case.
> (b) would adding __bswap[sd]i2 to the kernel build need to be
> disabled on armv6+? AFAICT, gcc doesn't emit calls - even for the
> 8-byte swap, even with -Os, on armv6+.
I wouldn't bother. That would save only 6 instructions total. And who
knows if some gcc flavor start calling them for some reason eventually.
> (c) testing allyesconfigs is proving to be a pain - lots of
> breakeage - other than defconfig testing, is there any more I can do?
The defconfigs provide wildly different results and that is a good
thing for further investigation. You may concentrate on a small
interesting sample such as those I kept above.
With allyesconfig the good configs would cancel out the bad ones making
the bad ones invisible.
Nicolas
next prev parent reply other threads:[~2013-02-09 3:16 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-29 1:30 [RFC] arm: use built-in byte swap function Kim Phillips
2013-01-29 8:35 ` Borislav Petkov
2013-01-29 16:46 ` Woodhouse, David
2013-01-29 17:42 ` Borislav Petkov
2013-01-29 17:55 ` Woodhouse, David
2013-01-29 18:10 ` Borislav Petkov
2013-01-30 10:22 ` Woodhouse, David
2013-01-31 2:09 ` Kim Phillips
2013-01-31 6:44 ` Borislav Petkov
2013-01-31 9:28 ` Russell King - ARM Linux
2013-01-31 20:59 ` Kim Phillips
2013-01-31 21:33 ` Borislav Petkov
2013-01-31 22:11 ` Woodhouse, David
2013-02-01 0:37 ` [PATCH v4] " Kim Phillips
2013-02-01 10:46 ` Russell King - ARM Linux
2013-02-01 1:17 ` [RFC] " Russell King - ARM Linux
2013-02-01 7:33 ` Woodhouse, David
2013-02-06 3:04 ` Kim Phillips
2013-02-06 9:02 ` Woodhouse, David
2013-02-07 1:19 ` Kim Phillips
2013-02-07 10:19 ` Will Newton
2013-02-07 10:43 ` Catalin Marinas
2013-02-07 18:13 ` Russell King - ARM Linux
2013-02-08 17:25 ` Woodhouse, David
2013-02-08 20:04 ` Nicolas Pitre
2013-02-08 22:40 ` Woodhouse, David
2013-02-08 22:47 ` Nicolas Pitre
2013-02-09 1:12 ` Kim Phillips
2013-02-09 3:16 ` Nicolas Pitre [this message]
2013-02-20 2:31 ` Kim Phillips
2013-02-20 2:38 ` Stephen Boyd
2013-02-20 3:17 ` Nicolas Pitre
2013-02-20 10:38 ` Woodhouse, David
2013-02-20 13:36 ` Nicolas Pitre
2013-02-20 13:44 ` Woodhouse, David
2013-02-20 14:06 ` Nicolas Pitre
2013-02-20 14:53 ` Woodhouse, David
2013-02-20 15:43 ` Nicolas Pitre
2013-02-21 3:49 ` Kim Phillips
2013-02-21 4:29 ` Nicolas Pitre
2013-02-21 6:52 ` Kim Phillips
2013-02-21 16:40 ` Nicolas Pitre
2013-02-22 2:33 ` Kim Phillips
2013-02-22 3:40 ` Nicolas Pitre
2013-02-23 1:40 ` [PATCH v6] " Kim Phillips
2013-02-23 2:40 ` Nicolas Pitre
2013-02-23 23:20 ` Woodhouse, David
2013-05-23 16:46 ` [PATCH v7] " Kim Phillips
2013-05-23 20:09 ` Nicolas Pitre
2013-05-23 23:13 ` Russell King - ARM Linux
2013-06-06 22:12 ` Russell King - ARM Linux
2013-06-06 22:23 ` Borislav Petkov
2013-06-07 0:03 ` Stephen Rothwell
2013-10-27 2:41 ` Nicolas Pitre
2013-11-05 21:45 ` Kim Phillips
2013-02-21 16:37 ` [RFC] " Woodhouse, David
2013-02-21 17:27 ` Nicolas Pitre
2013-03-13 13:35 ` Woodhouse, David
2013-01-29 14:13 ` Russell King - ARM Linux
2013-01-29 14:43 ` Woodhouse, David
2013-01-29 14:53 ` Rob Herring
2013-01-29 15:10 ` Woodhouse, David
2013-01-31 11:44 Woodhouse, David
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=alpine.LFD.2.02.1302082102380.6300@xanadu.home \
--to=nico@fluxnic.net \
--cc=akpm@linux-foundation.org \
--cc=bp@alien8.de \
--cc=daniel.santos@pobox.com \
--cc=david.woodhouse@intel.com \
--cc=kim.phillips@freescale.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@arm.linux.org.uk \
--cc=rientjes@google.com \
--cc=robherring2@gmail.com \
--cc=rusty@rustcorp.com.au \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).