From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: Andi Kleen <ak@linux.intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Peter Anvin <hpa@zytor.com>, Ingo Molnar <mingo@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
Peter Zijlstra <peterz@infradead.org>,
the arch/x86 maintainers <x86@kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Ning Qu <quning@google.com>,
Dave Hansen <dave.hansen@linux.intel.com>,
Matthew Wilcox <willy@linux.intel.com>
Subject: Re: [RFC, PATCH] mm: map few pages around fault address if they are in page cache
Date: Fri, 7 Feb 2014 19:56:22 +0200 [thread overview]
Message-ID: <20140207175622.GA14953@node.dhcp.inet.fi> (raw)
In-Reply-To: <20140207173200.GC12219@tassilo.jf.intel.com>
On Fri, Feb 07, 2014 at 09:32:00AM -0800, Andi Kleen wrote:
> > I haven't tested it much, but my kvm boots. There're few places where code
> > should be fixed. __do_fault() and filemap_fault() are too ugly and need to
> > be cleaned.
> >
> > I don't have any performance data yet.
> >
> > Any thoughts?
>
> It seems very drastic to do it unconditionally. How about at least a simple
> stream detection heuristic and perhaps also madvise?
We already have readahead here it can be reused here. But see below.
> There are some extreme cases where workloads could use a lot more memory
> than before, if they access their memory sparsely in the right pattern.
Have you noticied that we don't actually allocate any memory: only reuse
what's already there. Sure, it will increase VmSize, but do we care?
--
Kirill A. Shutemov
next prev parent reply other threads:[~2014-02-07 17:57 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-26 22:28 [RFC] de-asmify the x86-64 system call slowpath Linus Torvalds
2014-01-27 0:22 ` Al Viro
2014-01-27 4:32 ` Linus Torvalds
2014-01-27 4:48 ` H. Peter Anvin
2014-01-27 7:42 ` Al Viro
2014-01-27 22:06 ` Andy Lutomirski
2014-01-27 22:17 ` Linus Torvalds
2014-01-27 22:32 ` Al Viro
2014-01-27 22:43 ` Linus Torvalds
2014-01-27 22:46 ` Andy Lutomirski
2014-01-28 0:22 ` H. Peter Anvin
2014-01-28 0:44 ` Andy Lutomirski
2014-01-27 23:07 ` Al Viro
2014-01-27 10:27 ` Peter Zijlstra
2014-01-27 11:36 ` Al Viro
2014-01-27 17:39 ` Oleg Nesterov
2014-01-28 1:18 ` Al Viro
2014-01-28 16:38 ` Oleg Nesterov
2014-01-28 16:48 ` Al Viro
2014-01-28 17:19 ` Oleg Nesterov
2014-02-06 0:32 ` Linus Torvalds
2014-02-06 0:55 ` Kirill A. Shutemov
2014-02-06 2:32 ` Linus Torvalds
2014-02-06 4:33 ` Linus Torvalds
2014-02-06 21:29 ` Linus Torvalds
2014-02-06 22:24 ` Kirill A. Shutemov
2014-02-07 1:31 ` Linus Torvalds
2014-02-07 15:42 ` [RFC, PATCH] mm: map few pages around fault address if they are in page cache Kirill A. Shutemov
2014-02-07 17:32 ` Andi Kleen
2014-02-07 17:56 ` Kirill A. Shutemov [this message]
2014-02-07 18:11 ` Andi Kleen
2014-02-06 5:42 ` [RFC] de-asmify the x86-64 system call slowpath Ingo Molnar
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=20140207175622.GA14953@node.dhcp.inet.fi \
--to=kirill@shutemov.name \
--cc=ak@linux.intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=quning@google.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=willy@linux.intel.com \
--cc=x86@kernel.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;
as well as URLs for NNTP newsgroup(s).