linux-kernel - Re: [PATCH 0/6] PM / Hibernate: Memory bitmap scalability improvements

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20140719113502.GJ30979@8bytes.org>
Date:	Sat, 19 Jul 2014 13:35:02 +0200
From:	Joerg Roedel <joro@...tes.org>
To:	Pavel Machek <pavel@....cz>
Cc:	"Rafael J. Wysocki" <rjw@...ysocki.net>,
	Len Brown <len.brown@...el.com>, linux-pm@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/6] PM / Hibernate: Memory bitmap scalability
 improvements

Hi,

On Sat, Jul 19, 2014 at 12:26:23PM +0200, Pavel Machek wrote:
> > A test on a 12TB machine showed an improvement in resume
> > time from 76s with the old implementation to 2.4s with the
> > radix tree and the improved swsusp_free function. See below
> > for details of this test.
> 
> Ok, nice.
> 
> How did space requirements change?
> 
> In particular, do we need to reserve a bit more pages for hibernation
> now?

This depends on the amount of memory in the system. On one side the
struct bm_block (32 bytes) was replaced by struct rtree_node (16 bytes).

On the other side more pages are required to build the radix tree.
For systems with up to 64GB of RAM on amd64 this is one additional page.
On a 12TB system this sums up to 193 additional pages (again, on amd64)
required for the radix tree inner nodes.

These 193 pages take around 800kb while we save ca. 1.5MB because of the
smaller struct rtree_node.

On a 4GB system the situation is different, there we need to reserve one
additional page while saving only 512 bytes with the smaller struct. The
break-even is somewhere around 32GB where the savings of struct
rtree_node outweight the additional page required for the radix tree.

Thanks,

	Joerg

--
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/