From: Alexey Starikovskiy <aystarik@gmail.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: Nick Piggin <npiggin@suse.de>,
Pekka Enberg <penberg@cs.helsinki.fi>,
Linux Memory Management List <linux-mm@kvack.org>,
linux-acpi@vger.kernel.org, lenb@kernel.org
Subject: Re: [patch][rfc] acpi: do not use kmem caches
Date: Mon, 01 Dec 2008 20:32:41 +0300 [thread overview]
Message-ID: <49341FB9.80702@gmail.com> (raw)
In-Reply-To: <20081201171806.GA14074@infradead.org>
Christoph Hellwig wrote:
> On Mon, Dec 01, 2008 at 05:14:36PM +0300, Alexey Starikovskiy wrote:
>
>> Why then you try to delete ACPICA code, which might be just disabled by
>> undefining ACPI_USE_LOCAL_CACHE?
>> If you do want to go that path, you need to create patch against ACPICA, not
>> Linux code.
>>
>
> Sorry dude, but that's not how Linux development works. Please talk to
> some intel OTC folks to get an advice on how it does.
>
>
We are not speaking about Linux code here -- Nick changed ACPICA files.
And he already admits, that his patch is at least half-way wrong.
Sorry dude, ACPICA code is not Linux only, so one needs some care while
dropping some functionality from it.
>>> Ah OK I misread, that's the cache's freelist... ACPI shouldn't be poking
>>> this button inside the slab allocator anyway, honestly. What is it
>>> for?
>>>
>>>
>> And it is not actually used -- you cannot unload ACPI interpreter, and
>> this function is called only from there.
>>
>
> Care to remove all this dead code?
>
>
It is used at least in Windows userspace programs. So, removing these 4
lines only from Linux will
create another headache for Len during his merging of each new ACPICA
release into Linux.
>>> Is there a reasonable performance or memory win by using kmem cache? If
>>> not, then they should not be used
>>>
>> ACPI is still working in machines with several megabytes of RAM and
>> 100mhz Pentium processors. Do you say we should just not consider them
>> any longer?
>> If so, then just delete all ACPICA caches altogether.
>>
>
> As Nick is trying to explain you for a while it's not actually going
> to be a performance benefit for these, quite contrary because of how
> slab caches waste a lot of memory when only used very lightly or not
> at all.
>
>
If you care to do the math, and I helped you in another posting, he may
save about 11k in 32bit mode on thinkpad, and loose 70k in 64bit mode on
similar thinkpad.
Regards,
Alex.
next prev parent reply other threads:[~2008-12-01 17:32 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-01 8:31 [patch][rfc] acpi: do not use kmem caches Nick Piggin
2008-12-01 11:18 ` Pekka Enberg
2008-12-01 12:00 ` Nick Piggin
2008-12-01 13:12 ` Alexey Starikovskiy
2008-12-01 13:36 ` Nick Piggin
2008-12-01 14:14 ` Alexey Starikovskiy
2008-12-01 16:32 ` Nick Piggin
2008-12-01 17:18 ` Christoph Hellwig
2008-12-01 17:32 ` Alexey Starikovskiy [this message]
2008-12-01 13:37 ` Pekka Enberg
2008-12-01 14:02 ` Alexey Starikovskiy
2008-12-01 16:14 ` Nick Piggin
2008-12-01 16:45 ` Alexey Starikovskiy
2008-12-01 16:58 ` Nick Piggin
2008-12-01 17:20 ` Moore, Robert
2008-12-01 17:30 ` Andi Kleen
2008-12-01 17:32 ` Moore, Robert
2008-12-01 17:20 ` Christoph Hellwig
2008-12-01 17:49 ` Alexey Starikovskiy
2008-12-01 17:53 ` Len Brown
2008-12-01 18:10 ` Nick Piggin
2008-12-31 22:04 ` Len Brown
2009-01-05 4:14 ` Nick Piggin
2009-01-05 5:43 ` Skywing
2009-01-05 6:55 ` Nick Piggin
2008-12-01 14:32 ` Christoph Lameter
2008-12-01 14:48 ` Alexey Starikovskiy
2008-12-01 16:20 ` Nick Piggin
2008-12-01 17:04 ` Alexey Starikovskiy
2008-12-01 17:12 ` Nick Piggin
2008-12-01 17:25 ` Pekka Enberg
2008-12-01 17:32 ` Pekka Enberg
2008-12-01 17:36 ` Alexey Starikovskiy
2008-12-01 17:48 ` Pekka Enberg
2008-12-01 18:09 ` Christoph Lameter
2008-12-01 17:43 ` Alexey Starikovskiy
2008-12-01 17:31 ` Len Brown
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=49341FB9.80702@gmail.com \
--to=aystarik@gmail.com \
--cc=hch@infradead.org \
--cc=lenb@kernel.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=npiggin@suse.de \
--cc=penberg@cs.helsinki.fi \
/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