From: Rasmus Villemoes <linux@rasmusvillemoes.dk>
To: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Vishnu Pratap Singh <vishnu.ps@samsung.com>,
Pintu Kumar <pintu.k@samsung.com>,
Michal Nazarewicz <mina86@mina86.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/5] mm/page_alloc.c: Pull out init code from build_all_zonelists
Date: Tue, 03 Feb 2015 10:57:23 +0100 [thread overview]
Message-ID: <871tm7jp18.fsf@rasmusvillemoes.dk> (raw)
In-Reply-To: <alpine.DEB.2.10.1502021624090.667@chino.kir.corp.google.com> (David Rientjes's message of "Mon, 2 Feb 2015 16:25:25 -0800 (PST)")
On Tue, Feb 03 2015, David Rientjes <rientjes@google.com> wrote:
> On Tue, 3 Feb 2015, Rasmus Villemoes wrote:
>
>> Pulling the code protected by if (system_state == SYSTEM_BOOTING) into
>> its own helper allows us to shrink .text a little. This relies on
>> build_all_zonelists already having a __ref annotation. Add a comment
>> explaining why so one doesn't have to track it down through git log.
>>
>
> I think we should see the .text savings in the changelog to decide whether
> we want a __ref function (granted, with comment) calling an __init
> function in the source code.
Well, the real saving comes in 3/5, (mm/mm_init.c: Mark
mminit_verify_zonelist as __init), where one saves about 400
bytes. I originally did just that, while still adding a comment to
build_all_zonelists to explain both the old and new cause of __ref.
Then I noticed that cpuset_init_current_mems_allowed is also only called
from build_all_zonelists and could thus also be __init. But then the
__ref would cover two __init functions, both defined elsewhere, so I
thought it would be a little cleaner to make these calls from a single
__init function defined very close to its user. That it also happens to
shave a few bytes from build_all_zonelists is just gravy. A better
commit log would have been something like
Pulling the code protected by if (system_state == SYSTEM_BOOTING) into
its own helper allows us to shrink .text by a few bytes. But more
importantly, this provides a (somewhat) clean way of annotating
mminit_verify_zonelist and cpuset_init_current_mems_allowed with
__init, thus saving around 450 bytes of .text.
This relies on build_all_zonelists already having a __ref
annotation. Add a comment explaining both uses so one doesn't have to
track it down through git log.
Rasmus
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2015-02-03 9:57 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-02-02 23:50 [PATCH 0/5] mm: Some .text savings Rasmus Villemoes
2015-02-02 23:50 ` [PATCH 1/5] mm/internal.h: Don't split printk call in two Rasmus Villemoes
2015-02-02 23:50 ` [PATCH 2/5] mm/page_alloc.c: Pull out init code from build_all_zonelists Rasmus Villemoes
2015-02-03 0:25 ` David Rientjes
2015-02-03 9:57 ` Rasmus Villemoes [this message]
2015-02-02 23:50 ` [PATCH 3/5] mm/mm_init.c: Mark mminit_verify_zonelist as __init Rasmus Villemoes
2015-02-02 23:50 ` [PATCH 4/5] mm/mm_init.c: Mark mminit_loglevel __meminitdata Rasmus Villemoes
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=871tm7jp18.fsf@rasmusvillemoes.dk \
--to=linux@rasmusvillemoes.dk \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mina86@mina86.com \
--cc=pintu.k@samsung.com \
--cc=rientjes@google.com \
--cc=vishnu.ps@samsung.com \
/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