From: Catalin Marinas <catalin.marinas@arm.com>
To: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Arnd Bergmann <arnd@arndb.de>,
"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
Marc Zyngier <Marc.Zyngier@arm.com>,
Will Deacon <Will.Deacon@arm.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH v2 20/31] arm64: User access library function
Date: Thu, 6 Sep 2012 09:36:07 +0100 [thread overview]
Message-ID: <20120906083607.GA13975@arm.com> (raw)
In-Reply-To: <20120905210534.GI24085@n2100.arm.linux.org.uk>
On Wed, Sep 05, 2012 at 10:05:34PM +0100, Russell King - ARM Linux wrote:
> On Wed, Sep 05, 2012 at 10:01:37PM +0100, Catalin Marinas wrote:
> > There are indeed a few KB gain in code size but that's probably coming
> > from the exception table since otherwise you just replace a bl with
> > ldrt. It depends on what the compiler does as well, the arm code has
> > some carefully chosen registers when calling the __get_user_x function.
>
> It's more than that - it's not just the ldr but also a zeroing of a
> temporary register to hold the error code should the instruction fault.
> So it's not only the exception tables but also an increase in the
> main path - and that's where you benefit from having it out of line and
> thereby a hotter i-cache.
On 32-bit we have __get_user() inline and get_user() out of line. What
was the history behind this?
> > If you do the access_ok inline and the __get_user_x separately, the size
> > increase is even greater (at least in the arm64 case it can get to over
> > 20KB). I think x86 does the access_ok check out of line.
>
> Please talk to Will about get_user() and put_user(). Afterwards you
> will definitely want to keep them out of line on 64-bit ARM.
As I said, I already made the change to always inline get_user/put_user
with some penalty in the Image size but it makes the code cleaner. I'm
not entirely convinced of the performance gain/loss especially on ARMv8
cores with physically tagged caches. There is room for optimisation when
I get real silicon.
--
Catalin
next prev parent reply other threads:[~2012-09-06 8:37 UTC|newest]
Thread overview: 170+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-14 17:52 [PATCH v2 00/31] AArch64 Linux kernel port Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 01/31] arm64: Assembly macros and definitions Catalin Marinas
2012-08-15 12:57 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 02/31] arm64: Kernel booting and initialisation Catalin Marinas
2012-08-14 23:06 ` Olof Johansson
2012-08-15 17:37 ` Catalin Marinas
2012-08-15 19:03 ` Olof Johansson
2012-08-15 19:53 ` Catalin Marinas
2012-08-15 13:20 ` Arnd Bergmann
2012-08-15 17:06 ` Olof Johansson
2012-08-16 12:53 ` Catalin Marinas
2012-08-16 18:59 ` Nicolas Pitre
2012-08-17 11:20 ` Arnd Bergmann
2012-08-17 13:45 ` Catalin Marinas
2012-08-17 18:21 ` Nicolas Pitre
2012-08-17 8:56 ` Tony Lindgren
2012-08-17 9:41 ` Santosh Shilimkar
2012-08-17 10:05 ` Catalin Marinas
2012-08-17 10:10 ` Shilimkar, Santosh
2012-08-17 13:13 ` Tony Lindgren
2012-08-17 13:48 ` Catalin Marinas
2012-08-24 9:50 ` Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 03/31] arm64: Exception handling Catalin Marinas
2012-08-14 23:29 ` Olof Johansson
2012-08-14 23:47 ` Thomas Gleixner
2012-08-15 13:03 ` Arnd Bergmann
2012-08-16 10:05 ` Will Deacon
2012-08-16 11:54 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 04/31] arm64: MMU definitions Catalin Marinas
2012-08-15 13:30 ` Arnd Bergmann
2012-08-15 13:39 ` Catalin Marinas
2012-08-15 16:34 ` Geert Uytterhoeven
2012-08-15 16:45 ` Catalin Marinas
2012-08-17 9:04 ` Tony Lindgren
2012-08-17 9:21 ` Catalin Marinas
2012-08-17 9:38 ` Tony Lindgren
2012-08-14 17:52 ` [PATCH v2 05/31] arm64: MMU initialisation Catalin Marinas
2012-08-15 13:45 ` Arnd Bergmann
2012-08-17 10:06 ` Santosh Shilimkar
2012-08-17 10:15 ` Catalin Marinas
2012-08-17 10:25 ` Shilimkar, Santosh
2012-08-14 17:52 ` [PATCH v2 06/31] arm64: MMU fault handling and page table management Catalin Marinas
2012-08-15 13:47 ` Arnd Bergmann
2012-08-17 16:07 ` Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 07/31] arm64: Process management Catalin Marinas
2012-08-14 23:50 ` Olof Johansson
2012-09-14 17:33 ` Catalin Marinas
2012-09-16 0:29 ` Olof Johansson
2012-08-15 13:53 ` Arnd Bergmann
2012-08-17 16:15 ` Catalin Marinas
2012-08-16 15:09 ` Tobias Klauser
2012-08-14 17:52 ` [PATCH v2 08/31] arm64: CPU support Catalin Marinas
2012-08-15 0:10 ` Olof Johansson
2012-08-20 15:57 ` Catalin Marinas
2012-08-20 20:47 ` Arnd Bergmann
2012-08-21 9:50 ` Catalin Marinas
2012-09-14 17:38 ` Catalin Marinas
2012-08-15 13:56 ` Arnd Bergmann
2012-08-20 16:00 ` Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 09/31] arm64: Cache maintenance routines Catalin Marinas
2012-08-17 9:57 ` Santosh Shilimkar
2012-08-17 10:07 ` Catalin Marinas
2012-08-17 10:12 ` Shilimkar, Santosh
2012-08-14 17:52 ` [PATCH v2 10/31] arm64: TLB maintenance functionality Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 11/31] arm64: IRQ handling Catalin Marinas
2012-08-14 23:22 ` Aaro Koskinen
2012-08-14 17:52 ` [PATCH v2 12/31] arm64: Atomic operations Catalin Marinas
2012-08-15 0:21 ` Olof Johansson
2012-08-14 17:52 ` [PATCH v2 13/31] arm64: Device specific operations Catalin Marinas
2012-08-15 0:33 ` Olof Johansson
2012-09-14 17:29 ` Catalin Marinas
2012-09-14 17:31 ` Arnd Bergmann
2012-09-14 17:39 ` Catalin Marinas
2012-09-16 0:28 ` Olof Johansson
2012-08-15 16:13 ` Arnd Bergmann
2012-08-17 9:19 ` Tony Lindgren
2012-08-14 17:52 ` [PATCH v2 14/31] arm64: DMA mapping API Catalin Marinas
2012-08-15 0:40 ` Olof Johansson
2012-08-21 13:05 ` Catalin Marinas
2012-08-15 16:16 ` Arnd Bergmann
2012-08-21 12:59 ` Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 15/31] arm64: SMP support Catalin Marinas
2012-08-15 0:49 ` Olof Johansson
2012-08-15 13:04 ` Arnd Bergmann
2012-08-17 9:21 ` Tony Lindgren
2012-08-17 9:32 ` Catalin Marinas
2012-08-17 9:39 ` Tony Lindgren
2012-08-14 17:52 ` [PATCH v2 16/31] arm64: ELF definitions Catalin Marinas
2012-08-15 14:15 ` Arnd Bergmann
2012-08-16 10:23 ` Will Deacon
2012-08-16 12:37 ` Arnd Bergmann
2012-08-21 16:06 ` Catalin Marinas
2012-08-21 18:17 ` Geert Uytterhoeven
2012-08-21 18:27 ` Catalin Marinas
2012-08-21 18:53 ` Mike Frysinger
2012-08-21 20:17 ` Arnd Bergmann
2012-09-05 19:56 ` Chris Metcalf
2012-08-14 17:52 ` [PATCH v2 17/31] arm64: System calls handling Catalin Marinas
2012-08-15 14:22 ` Arnd Bergmann
2012-08-21 17:51 ` Catalin Marinas
2012-08-21 20:14 ` Arnd Bergmann
2012-08-21 22:01 ` Catalin Marinas
2012-08-22 7:56 ` Arnd Bergmann
2012-08-22 10:29 ` Catalin Marinas
2012-08-22 12:27 ` Arnd Bergmann
2012-08-22 17:13 ` Catalin Marinas
2012-09-03 11:48 ` Catalin Marinas
2012-09-03 12:39 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 18/31] arm64: VDSO support Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 19/31] arm64: Signal handling support Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 20/31] arm64: User access library functions Catalin Marinas
2012-08-15 14:49 ` [PATCH v2 20/31] arm64: User access library function Arnd Bergmann
2012-09-03 12:58 ` Catalin Marinas
2012-09-05 19:13 ` Russell King - ARM Linux
2012-09-05 21:01 ` Catalin Marinas
2012-09-05 21:05 ` Russell King - ARM Linux
2012-09-06 8:36 ` Catalin Marinas [this message]
2012-08-14 17:52 ` [PATCH v2 21/31] arm64: 32-bit (compat) applications support Catalin Marinas
2012-08-15 14:34 ` Arnd Bergmann
2012-08-16 10:28 ` Will Deacon
2012-08-16 12:39 ` Arnd Bergmann
2012-08-23 6:46 ` PER_LINUX32, Was: " Arnd Bergmann
2012-08-23 10:42 ` Catalin Marinas
2012-08-28 18:28 ` Jiri Kosina
2012-08-24 10:43 ` Catalin Marinas
2012-08-26 4:49 ` Arnd Bergmann
2012-08-20 10:53 ` Pavel Machek
2012-08-20 20:34 ` Arnd Bergmann
2012-08-21 10:28 ` Pavel Machek
2012-08-14 17:52 ` [PATCH v2 22/31] arm64: Floating point and SIMD Catalin Marinas
2012-08-15 14:35 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 23/31] arm64: Debugging support Catalin Marinas
2012-08-15 15:07 ` Arnd Bergmann
2012-08-16 10:47 ` Will Deacon
2012-08-16 12:49 ` Arnd Bergmann
2012-08-17 7:06 ` Arnd Bergmann
2012-08-20 9:07 ` Will Deacon
2012-08-20 9:27 ` Will Deacon
2012-08-20 20:10 ` Arnd Bergmann
2012-08-21 8:58 ` Will Deacon
2012-08-14 17:52 ` [PATCH v2 24/31] arm64: Add support for /proc/sys/debug/exception-trace Catalin Marinas
2012-08-15 15:08 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 25/31] arm64: Performance counters support Catalin Marinas
2012-08-15 15:11 ` Arnd Bergmann
2012-08-16 10:51 ` Will Deacon
2012-08-14 17:52 ` [PATCH v2 26/31] arm64: Miscellaneous library functions Catalin Marinas
2012-08-15 15:21 ` Arnd Bergmann
2012-08-16 10:57 ` Will Deacon
2012-08-16 13:00 ` Arnd Bergmann
2012-08-16 14:11 ` Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 27/31] arm64: Loadable modules Catalin Marinas
2012-08-15 15:23 ` Arnd Bergmann
2012-08-15 15:35 ` Catalin Marinas
2012-08-15 16:16 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 28/31] arm64: Generic timers support Catalin Marinas
2012-08-15 15:52 ` Arnd Bergmann
2012-08-16 12:40 ` Linus Walleij
2012-08-17 9:29 ` Tony Lindgren
2012-08-17 10:21 ` Santosh Shilimkar
2012-08-21 19:20 ` Christopher Covington
2012-08-14 17:52 ` [PATCH v2 29/31] arm64: Miscellaneous header files Catalin Marinas
2012-08-15 15:56 ` Arnd Bergmann
2012-08-14 17:52 ` [PATCH v2 30/31] arm64: Build infrastructure Catalin Marinas
2012-08-14 21:01 ` Sam Ravnborg
2012-08-15 16:07 ` Arnd Bergmann
2012-08-17 9:32 ` Tony Lindgren
2012-08-17 9:46 ` Catalin Marinas
2012-08-14 17:52 ` [PATCH v2 31/31] arm64: MAINTAINERS update Catalin Marinas
2012-08-15 15:57 ` Arnd Bergmann
2012-08-17 9:36 ` [PATCH v2 00/31] AArch64 Linux kernel port Tony Lindgren
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=20120906083607.GA13975@arm.com \
--to=catalin.marinas@arm.com \
--cc=Marc.Zyngier@arm.com \
--cc=Will.Deacon@arm.com \
--cc=arnd@arndb.de \
--cc=linux-arch@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@arm.linux.org.uk \
/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).