linux-kernel - Re: [PATCH 1/3] ARM: EXYNOS: pmu: Constify data tables

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 10 Dec 2013 00:11:40 +0800
From:	Daniel Kurtz <djkurtz@...omium.org>
To:	Tomasz Figa <t.figa@...sung.com>
Cc:	kgene.kim@...sung.com, Heiko Stuebner <heiko@...ech.de>,
	Amit Daniel Kachhap <amit.daniel@...sung.com>,
	linux-arm-kernel@...ts.infradead.org,
	linux-samsung-soc@...r.kernel.org,
	Doug Andersen <dianders@...omium.org>,
	Abhilash Kesavan <a.kesavan@...sung.com>,
	Ben Dooks <ben-linux@...ff.org>,
	Russell King <linux@....linux.org.uk>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] ARM: EXYNOS: pmu: Constify data tables

Hi Tomasz,

Thank you for the reviews.

On Dec 9, 2013 5:15 AM, "Tomasz Figa" <t.figa@...sung.com> wrote:
>
> Hi Daniel,
>
> On Thursday 21 of November 2013 02:21:24 Daniel Kurtz wrote:
> > These tables are all immutable, make them const to save 4416 bytes of RAM.
> >
> > size arch/arm/mach-exynos/pmu.o
> >    text          data     bss
> >     848          4420       4         // before
> >    5264             4       4         // after
>
> I'm not sure where the mentioned saving of RAM is. Moving data between
> sections is not supposed to make it use less memory, I believe.

You are correct.  This was my misunderstanding from doing too much
work with microcontrollers, where .text sections are accessed in place
from FLASH for code and const data, but .data memory is copied from a
FLASH section to a second RAM section at init for access at runtime.
Most modern Linux systems copy/decompress their code and data sections
from external storage to RAM anyway, so there is no actual memory
savings (except potentially the compiler may be able to optimize a bit
more with the const hint).

>
> Anyway, it's a good practice to mark constant data as const, to disallow
> changing them at runtime by mistake, so the patch is fine. Except some
> issues I commented on inline.

Were there supposed to be inline comments?  I don't see any.

Best Regards,
-djk

>
> Reviewed-by: Tomasz Figa <t.figa@...sung.com>
>
> Best regards,
> Tomasz
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/