From: Glauber Costa <glommer@parallels.com>
To: Christoph Lameter <cl@linux.com>
Cc: andi@firstfloor.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, Joonsoo Kim <js1304@gmail.com>,
David Rientjes <rientjes@google.com>,
Pekka Enberg <penberg@kernel.org>
Subject: Re: [PATCH v2 1/2] kmem_cache: include allocators code directly into slab_common
Date: Wed, 24 Oct 2012 18:35:03 +0400 [thread overview]
Message-ID: <5087FC97.6080100@parallels.com> (raw)
In-Reply-To: <0000013a932d456c-8f0cbbce-e3f7-4f2a-b051-7b093a8cfc7e-000000@email.amazonses.com>
On 10/24/2012 06:29 PM, Christoph Lameter wrote:
> On Wed, 24 Oct 2012, Glauber Costa wrote:
>
>> Because of that, we either have to move all the entry points to the
>> mm/slab.h and rely heavily on the pre-processor, or include all .c files
>> in here.
>
> Hmm... That is a bit of a radical solution. The global optimizations now
> possible with the new gcc compiler include the ability to fold functions
> across different linkable objects. Andi, is that usable for kernel builds?
>
In general, it takes quite a lot of time to take all those optimizations
for granted. We still live a lot of time with multiple compiler versions
building distros, etc, for quite some time.
I would expect the end result for anyone not using such a compiler to be
a sudden performance drop when using a new kernel. Not really pleasant.
--
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>
WARNING: multiple messages have this Message-ID (diff)
From: Glauber Costa <glommer@parallels.com>
To: Christoph Lameter <cl@linux.com>
Cc: <andi@firstfloor.org>, <linux-mm@kvack.org>,
<linux-kernel@vger.kernel.org>, Joonsoo Kim <js1304@gmail.com>,
David Rientjes <rientjes@google.com>,
Pekka Enberg <penberg@kernel.org>
Subject: Re: [PATCH v2 1/2] kmem_cache: include allocators code directly into slab_common
Date: Wed, 24 Oct 2012 18:35:03 +0400 [thread overview]
Message-ID: <5087FC97.6080100@parallels.com> (raw)
In-Reply-To: <0000013a932d456c-8f0cbbce-e3f7-4f2a-b051-7b093a8cfc7e-000000@email.amazonses.com>
On 10/24/2012 06:29 PM, Christoph Lameter wrote:
> On Wed, 24 Oct 2012, Glauber Costa wrote:
>
>> Because of that, we either have to move all the entry points to the
>> mm/slab.h and rely heavily on the pre-processor, or include all .c files
>> in here.
>
> Hmm... That is a bit of a radical solution. The global optimizations now
> possible with the new gcc compiler include the ability to fold functions
> across different linkable objects. Andi, is that usable for kernel builds?
>
In general, it takes quite a lot of time to take all those optimizations
for granted. We still live a lot of time with multiple compiler versions
building distros, etc, for quite some time.
I would expect the end result for anyone not using such a compiler to be
a sudden performance drop when using a new kernel. Not really pleasant.
next prev parent reply other threads:[~2012-10-24 14:35 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-24 13:59 [PATCH v2 0/2] common entry point for kmem_cache_free Glauber Costa
2012-10-24 13:59 ` Glauber Costa
2012-10-24 13:59 ` [PATCH v2 1/2] kmem_cache: include allocators code directly into slab_common Glauber Costa
2012-10-24 13:59 ` Glauber Costa
2012-10-24 14:29 ` Christoph Lameter
2012-10-24 14:29 ` Christoph Lameter
2012-10-24 14:35 ` Glauber Costa [this message]
2012-10-24 14:35 ` Glauber Costa
2012-10-24 17:29 ` JoonSoo Kim
2012-10-24 17:29 ` JoonSoo Kim
2012-10-24 15:33 ` Andi Kleen
2012-10-24 15:33 ` Andi Kleen
2012-10-24 18:02 ` Pekka Enberg
2012-10-24 18:02 ` Pekka Enberg
2012-10-24 19:34 ` Christoph Lameter
2012-10-24 19:34 ` Christoph Lameter
2012-10-26 14:15 ` JoonSoo Kim
2012-10-26 14:15 ` JoonSoo Kim
2012-10-30 15:31 ` Christoph Lameter
2012-10-30 15:31 ` Christoph Lameter
2012-10-30 15:35 ` Glauber Costa
2012-10-30 15:35 ` Glauber Costa
2012-10-30 20:27 ` Andi Kleen
2012-10-30 20:27 ` Andi Kleen
2012-10-24 13:59 ` [PATCH v2 2/2] slab: move kmem_cache_free to common code Glauber Costa
2012-10-24 13:59 ` Glauber Costa
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=5087FC97.6080100@parallels.com \
--to=glommer@parallels.com \
--cc=andi@firstfloor.org \
--cc=cl@linux.com \
--cc=js1304@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=penberg@kernel.org \
--cc=rientjes@google.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 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.