From: Adrian Bunk <bunk@kernel.org>
To: Eric Dumazet <dada1@cosmosbay.com>
Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org
Subject: Re: RFC: remove __read_mostly
Date: Fri, 14 Dec 2007 00:00:43 +0100 [thread overview]
Message-ID: <20071213230043.GJ21616@stusta.de> (raw)
In-Reply-To: <4761B6BB.1070504@cosmosbay.com>
On Thu, Dec 13, 2007 at 11:48:27PM +0100, Eric Dumazet wrote:
> Adrian Bunk a écrit :
>> I tried the following patch with a full x86 .config [1]:
>>
>> --- a/include/asm-x86/cache.h
>> +++ b/include/asm-x86/cache.h
>> -#define __read_mostly __attribute__((__section__(".data.read_mostly")))
>> +/* #define __read_mostly __attribute__((__section__(".data.read_mostly"))) */
>>
>> The result [2,3] was:
>>
>> -rwxrwxr-x 1 bunk bunk 46607243 2007-12-13 19:50 vmlinux.old
>> -rwxrwxr-x 1 bunk bunk 46598691 2007-12-13 21:55 vmlinux
>>
>> It's not a surprise that the kernel can become bigger when __read_mostly
>> gets used, especially in cases where __read_mostly prevents gcc
>> optimizations.
>>
>> My question is:
>> Is there anywhere in the kernel a case where __read_mostly brings a
>> measurable improvement or can it be removed?
>
> Yes, there are many cases where read_mostly brings huge improvements.
>
> I did test your idea on a 4 CPUS server, and system time was roughly
> doubled, from 11% to 20%
Thanks, that answered my question.
> Of course, you noticed that puting a __read_mostly attribute force the
> linker to reserve space for the variable. So a null variable previously in
> bss section (no space in vmlinux file) is now in .data.read_mostly. Not a
> big deal.
>
> If you want, you could play some .lds games to create sort of a
> "bss.read_mostly" section to save 10000 bytes in vmlinux.
No, what I was thinking of kernel code like
static int fooxxxvar = 0;
static int __read_mostly fooyyyvar = 0;
void mytest()
{
if (fooxxxvar)
printk("fooxxx");
if (fooyyyvar)
printk("fooyyy");
}
If you run "strings file.c | grep foo" on the file it will only return
"fooyyy" but not "fooxxx".
It's not a big deal, but it's an area where __read_mostly affects gcc
and not ld.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
next prev parent reply other threads:[~2007-12-13 23:00 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-12-13 22:20 RFC: remove __read_mostly Adrian Bunk
2007-12-13 22:29 ` Andi Kleen
2007-12-13 22:41 ` Adrian Bunk
2007-12-14 16:16 ` Arnd Bergmann
2007-12-14 16:31 ` Eric Dumazet
2007-12-13 22:32 ` David Miller
2007-12-13 22:44 ` Harvey Harrison
2007-12-13 23:06 ` Andi Kleen
2007-12-13 22:48 ` Eric Dumazet
2007-12-13 23:00 ` Adrian Bunk [this message]
2007-12-13 23:54 ` Kyle McMartin
2007-12-14 0:33 ` Andi Kleen
2007-12-17 10:33 ` Andrew Morton
2007-12-17 10:53 ` Eric Dumazet
2007-12-17 11:07 ` Andrew Morton
2007-12-17 12:19 ` Andi Kleen
2007-12-17 12:15 ` Andi Kleen
2007-12-17 12:40 ` Adrian Bunk
2007-12-14 15:24 ` Matt Mackall
2007-12-14 15:38 ` Eric Dumazet
2007-12-14 15:42 ` Matt Mackall
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=20071213230043.GJ21616@stusta.de \
--to=bunk@kernel.org \
--cc=dada1@cosmosbay.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.