linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Nick Piggin <npiggin@suse.de>
To: Hugh Dickins <hugh@veritas.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Linux Memory Management List <linux-mm@kvack.org>
Subject: Re: [patch 0/6] lockless pagecache
Date: Mon, 19 Nov 2007 23:58:24 +0100	[thread overview]
Message-ID: <20071119225824.GB24255@wotan.suse.de> (raw)
In-Reply-To: <Pine.LNX.4.64.0711172001420.9287@blonde.wat.veritas.com>

On Sat, Nov 17, 2007 at 08:16:18PM +0000, Hugh Dickins wrote:
> On Sun, 11 Nov 2007, Nick Piggin wrote:
> > 
> > I wonder what everyone thinks about getting the lockless pagecache patch
> > into -mm? This version uses Hugh's suggestion to avoid a smp_rmb and a load
> > and branch in the lockless lookup side, and avoids some atomic ops in the
> > reclaim path, and avoids using a page flag! The coolest thing about it is
> > that it speeds up single-threaded pagecache lookups...
> 
> I've liked this in the past, with the exception of PageNoNewRefs which
> seemed an unnecessary ugliness.  Now you've eliminated that, thank you,
> I expect I should like it through and through (if I actually found time
> to redigest it).  A moment came up and I thought I'd give it a spin...

Yeah I decided it is actually just as good or better at it's job --
neither really protects against an errant get_page() or put_page(),
however at least this scheme will go bug with CONFIG_DEBUG_VM, as opposed
to the current or old lockless schemes (which I guess will silently allow
it).


> > Patches are against latest git for RFC.
> 
> ... but they're not.  You seem to have descended into sending out
> ?cleanup? patches at intervals, and recursive dependence upon them.
> This set relies on there being something called __set_page_locked()
> in include/linux/pagemap.h, but there isn't in latest git (nor mm).
> Ah, you posted a patch earlier which introduced that, but it relies on
> there being something called set_page_locked() in include/linux/pagemap.h,
> but there isn't in latest git (nor mm).  Ah, you posted a patch earlier
> which introduced that ... I gave up at this point.
> 
> We've all got lots of other things to do, please make it easier.

Sorry, I honestly didn't pay enough attention there because I didn't
think anybody would run it! I'll update it and resend.

--
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>

  reply	other threads:[~2007-11-19 22:58 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-11  8:45 Nick Piggin
2007-11-11  8:47 ` [patch 1/6] mm: readahead scan lockless Nick Piggin
2007-11-11  8:49 ` [patch 2/6] radix-tree: gang_lookup_slot Nick Piggin
2007-11-11  8:50 ` [patch 3/6] mm: speculative get page Nick Piggin
2007-11-12 20:21   ` Christoph Lameter
2007-11-13  0:35     ` Nick Piggin
2007-11-13  0:37       ` Christoph Lameter
2007-11-13  0:51         ` Nick Piggin
2007-11-11  8:51 ` [patch 4/6] mm: lockless pagecache lookups Nick Piggin
2007-11-11  8:51 ` [patch 5/6] mm: spinlock tree_lock Nick Piggin
2007-11-11  8:52 ` [patch 6/6] mm: speculative refcount debug Nick Piggin
2007-11-17  9:48 ` [patch 0/6] lockless pagecache Peter Zijlstra
2007-11-17 20:16 ` Hugh Dickins
2007-11-19 22:58   ` Nick Piggin [this message]
2008-01-09 15:45 ` Peter Zijlstra

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=20071119225824.GB24255@wotan.suse.de \
    --to=npiggin@suse.de \
    --cc=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=hugh@veritas.com \
    --cc=linux-mm@kvack.org \
    --cc=torvalds@linux-foundation.org \
    /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