linux-kernel - Re: [PATCH v7 1/4] spinlock: A new lockref structure for lockless update of refcount

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: <CA+55aFxvEjdEFK0dXn783eVN0PEXVhb5k8OnrbxSWDpXgyTwWw@mail.gmail.com>
Date:	Fri, 30 Aug 2013 08:34:46 -0700
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Sedat Dilek <sedat.dilek@...il.com>
Cc:	Waiman Long <waiman.long@...com>, Ingo Molnar <mingo@...nel.org>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	Alexander Viro <viro@...iv.linux.org.uk>,
	Jeff Layton <jlayton@...hat.com>,
	Miklos Szeredi <mszeredi@...e.cz>,
	Ingo Molnar <mingo@...hat.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	linux-fsdevel <linux-fsdevel@...r.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Steven Rostedt <rostedt@...dmis.org>,
	Andi Kleen <andi@...stfloor.org>,
	"Chandramouleeswaran, Aswin" <aswin@...com>,
	"Norton, Scott J" <scott.norton@...com>
Subject: Re: [PATCH v7 1/4] spinlock: A new lockref structure for lockless
 update of refcount

On Fri, Aug 30, 2013 at 2:27 AM, Sedat Dilek <sedat.dilek@...il.com> wrote:
>
> Here on Ubuntu/precise v12.04.3 AMD64 I get these numbers for total loops:
>
> lockref:  w/o patch | w/ patch
> ======================
> Run #1: 2.688.094 | 2.643.004
> Run #2: 2.678.884 | 2.652.787
> Run #3: 2.686.450 | 2.650.142
> Run #4: 2.688.435 | 2.648.409
> Run #5: 2.693.770 | 2.651.514

Yes, so this is pretty much expected.

If you don't have a very high core count (you don't mention your
system, but that's pretty - I get ~65 million repetitions in 10
seconds on my i5-670), the cmpxchg will not help - because you don't
actually see the bad "wait on spinlock" behavior in the first place.

And a "cmpxchg" is slightly slower than the very optimized spinlocks,
and has that annoying "read original value" first issue too. So the
patch can make things a bit slower, although it will depends on the
microarchitecture (and as mentioned elsewhere, there are other things
that can make a bigger difference boot-to-boot - dentry allocation
details etc can have "sticky" performance impact).

So we may take a small hit in order to then *not* have horrible
scalability at the high end.

               Linus
--
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/