linux-kernel - Re: [PATCH] mm: don't invoke __alloc_pages_direct

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 for Android: free password hash cracker in your pocket

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

Date:	Wed, 11 Jul 2012 00:24:41 +0900
From:	JoonSoo Kim <js1304@...il.com>
To:	Mel Gorman <mgorman@...e.de>
Cc:	David Rientjes <rientjes@...gle.com>, akpm@...ux-foundation.org,
	Pekka Enberg <penberg@...nel.org>,
	Christoph Lameter <cl@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH] mm: don't invoke __alloc_pages_direct_compact when order 0

2012/7/10 Mel Gorman <mgorman@...e.de>:
> You say that invoking the function is very costly. I agree that a function
> call with that many parameters is hefty but it is also in the slow path of
> the allocator. For order-0 allocations we are about to enter direct reclaim
> where I would expect the cost far exceeds the cost of a function call.

Yes, I agree.

> If the cost is indeed high and you have seen this in profiles then I
> suggest you create a forced inline function alloc_pages_direct_compact
> that does this;
>
> if (order != 0)
>         __alloc_pages_direct_compact(...)
>
> and then call alloc_pages_direct_compact instead of
> __alloc_pages_direct_compact. After that, recheck the profiles (although I
> expect the difference to be marginal) and the size of vmlinux (if it gets
> bigger, it's probably not worth it).
> That would be functionally similar to your patch but it will preserve git
> blame, churn less code and be harder to make mistakes with in the unlikely
> event a third call to alloc_pages_direct_compact is ever added.

Your suggestion looks good.
But, the size of page_alloc.o is more than before.

I test 3 approaches, vanilla, always_inline and
wrapping(alloc_page_direct_compact which is your suggestion).
In my environment (v3.5-rc5, gcc 4.6.3, x86_64), page_alloc.o shows
below number.

                                         total, .text section, .text.unlikely
page_alloc_vanilla.o:     93432,   0x510a,        0x243
page_alloc_inline.o:       93336,   0x52ca,          0xa4
page_alloc_wrapping.o: 93528,   0x515a,        0x238

Andrew said that inlining add only 26 bytes to .text of page_alloc.o,
but in my system, need more bytes.
Currently, I think this patch doesn't have obvious benefit, so I want
to drop it.
Any objections?

Thanks for good comments.
--
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/