All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cyrill Gorcunov <gorcunov@gmail.com>
To: Jan Beulich <JBeulich@suse.com>
Cc: David Vrabel <david.vrabel@citrix.com>,
	Andy Lutomirski <luto@amacapital.net>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Xen-devel@lists.xen.org,
	Boris Ostrovsky <boris.ostrovsky@oracle.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Pavel Emelyanov <xemul@parallels.com>,
	Ingo Molnar <mingo@redhat.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>
Subject: Re: Regression: x86/mm: new _PTE_SWP_SOFT_DIRTY bit conflicts with existing use
Date: Wed, 21 Aug 2013 19:42:38 +0400	[thread overview]
Message-ID: <20130821154238.GV18673@moon> (raw)
In-Reply-To: <5214F09002000078000ED5C3@nat28.tlf.novell.com>

On Wed, Aug 21, 2013 at 03:53:36PM +0100, Jan Beulich wrote:
> >>> On 21.08.13 at 16:12, Cyrill Gorcunov <gorcunov@gmail.com> wrote:
> > On Wed, Aug 21, 2013 at 02:48:20PM +0100, David Vrabel wrote:
> >> All,
> >> 
> >> 179ef71c (mm: save soft-dirty bits on swapped pages) introduces a new
> >> PTE bit on x86 _PTE_SWP_SOFT_DIRTY which has the same value as _PTE_PSE
> >> and _PTE_PAT.
> >> 
> >> With a Xen PV guest, the use of the _PTE_PAT will result in the page
> >> having unexpected cachability which will introduce a range of subtle
> >> performance and correctness issues.  Xen programs the entry 4 in the PAT
> >> table with WC so a page that was previously WB will end up as WC.
> >> 
> > 
> > David, could you please explain, Xen keeps and analyze _PTE_PAT bit
> > for ptes which are not present?
> 
> No, the problem isn't with not-present PTEs (i.e. swap entries),
> but with present ones - the same bit (7) is being used for both,
> according to this comment:
> 
> /*
>  * Tracking soft dirty bit when a page goes to a swap is tricky.
>  * We need a bit which can be stored in pte _and_ not conflict
>  * with swap entry format. On x86 bits 6 and 7 are *not* involved
>  * into swap entry computation, but bit 6 is used for nonlinear
>  * file mapping, so we borrow bit 7 for soft dirty tracking.
>  */
> 
> Or are you telling me that the comment is misleading (at least me),
> and this applies only to not-present PTEs? And even then - where
> would the value of the original PAT bit be stored while swapped
> out (or is it impossible - now and forever - for WC pages to get
> swapped)?

Only to non-present ptes, as far as I know.

do_swap_page
	...
	pte = mk_pte(page, vma->vm_page_prot);

	/* new pte from vm_page_prot generated */
	...
	set_pte_at(mm, address, page_table, pte);
	/* and assigned to old place */

with soft dirty in swap it is somehow more weirdy

	pte = mk_pte(page, vma->vm_page_prot);
	...
	if (pte_swp_soft_dirty(orig_pte))
		pte = pte_mksoft_dirty(pte);
	set_pte_at(mm, address, page_table, pte);

orig_pte has pse bit set if page has been soft dirty
when it reached swap.

  parent reply	other threads:[~2013-08-21 15:42 UTC|newest]

Thread overview: 89+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-21 13:48 Regression: x86/mm: new _PTE_SWP_SOFT_DIRTY bit conflicts with existing use David Vrabel
2013-08-21 13:53 ` konrad wilk
2013-08-21 13:53 ` konrad wilk
2013-08-21 14:11 ` H. Peter Anvin
2013-08-21 14:11 ` H. Peter Anvin
2013-08-21 14:19   ` Cyrill Gorcunov
2013-08-21 14:19   ` Cyrill Gorcunov
2013-08-21 14:22     ` H. Peter Anvin
2013-08-21 14:22     ` H. Peter Anvin
2013-08-21 14:29       ` Cyrill Gorcunov
2013-08-21 14:29       ` Cyrill Gorcunov
2013-08-21 16:30     ` Linus Torvalds
2013-08-21 16:42       ` Cyrill Gorcunov
2013-08-21 16:42       ` Cyrill Gorcunov
2013-08-21 23:05       ` Cyrill Gorcunov
2013-08-21 23:05       ` Cyrill Gorcunov
2013-08-21 23:42         ` Andi Kleen
2013-08-22  5:49           ` Cyrill Gorcunov
2013-08-22  6:37             ` Minchan Kim
2013-08-22  6:37             ` Minchan Kim
2013-08-22 13:12               ` Cyrill Gorcunov
2013-08-22 13:12               ` Cyrill Gorcunov
2013-08-22  5:49           ` Cyrill Gorcunov
2013-08-21 23:42         ` Andi Kleen
2013-08-27 22:04       ` Benjamin Herrenschmidt
2013-08-27 22:04       ` Benjamin Herrenschmidt
2013-08-21 16:30     ` Linus Torvalds
2013-08-21 14:12 ` Cyrill Gorcunov
2013-08-21 14:12 ` Cyrill Gorcunov
2013-08-21 14:22   ` H. Peter Anvin
2013-08-21 14:22   ` H. Peter Anvin
2013-08-21 14:53   ` Jan Beulich
2013-08-21 14:53   ` Jan Beulich
2013-08-21 14:58     ` H. Peter Anvin
2013-08-21 14:58     ` H. Peter Anvin
2013-08-21 15:42     ` Cyrill Gorcunov
2013-08-21 15:42     ` Cyrill Gorcunov [this message]
2013-08-21 16:03       ` Jan Beulich
2013-08-21 16:03       ` Jan Beulich
2013-08-21 16:19         ` Cyrill Gorcunov
2013-08-21 16:19         ` Cyrill Gorcunov
2013-08-21 16:56           ` David Vrabel
2013-08-21 17:25             ` Cyrill Gorcunov
2013-08-21 17:25             ` Cyrill Gorcunov
2013-08-21 18:17               ` Cyrill Gorcunov
2013-08-21 18:50                 ` H. Peter Anvin
2013-08-21 19:03                   ` Cyrill Gorcunov
2013-08-21 19:07                     ` Andy Lutomirski
2013-08-21 19:07                     ` Andy Lutomirski
2013-08-21 19:20                       ` Cyrill Gorcunov
2013-08-21 19:20                       ` Cyrill Gorcunov
2013-08-21 19:21                       ` Pavel Emelyanov
2013-08-21 19:21                       ` Pavel Emelyanov
2013-08-21 23:04                     ` Linus Torvalds
2013-08-21 23:04                     ` Linus Torvalds
2013-08-22  0:51                       ` Dave Jones
2013-08-22  5:44                         ` Cyrill Gorcunov
2013-08-22  5:44                         ` Cyrill Gorcunov
2013-08-22  6:41                         ` Pavel Emelyanov
2013-08-22  6:41                         ` Pavel Emelyanov
2013-08-22  0:51                       ` Dave Jones
2013-08-22  7:47                       ` Jan Beulich
2013-08-22  7:47                       ` Jan Beulich
2013-08-22  9:32                       ` David Vrabel
2013-08-22 10:16                         ` Pavel Emelyanov
2013-08-22 10:16                         ` Pavel Emelyanov
2013-08-22  9:32                       ` David Vrabel
2013-08-21 19:03                   ` Cyrill Gorcunov
2013-08-21 18:50                 ` H. Peter Anvin
2013-08-21 18:17               ` Cyrill Gorcunov
2013-08-21 16:56           ` David Vrabel
2013-08-22  6:56           ` Jan Beulich
2013-08-22  7:03             ` Cyrill Gorcunov
2013-08-22  7:27               ` Jan Beulich
2013-08-22  7:27               ` Jan Beulich
2013-08-22 11:27                 ` Cyrill Gorcunov
2013-08-22 11:33                   ` Jan Beulich
2013-08-22 11:33                   ` Jan Beulich
2013-08-22 12:18                     ` Pavel Emelyanov
2013-08-22 12:18                     ` Pavel Emelyanov
2013-08-22 11:27                 ` Cyrill Gorcunov
2013-08-22  7:03             ` Cyrill Gorcunov
2013-08-22  6:56           ` Jan Beulich
2013-08-21 17:28 ` Andy Lutomirski
2013-08-22  7:54   ` Jan Beulich
2013-08-22  9:06     ` Stefan Bader
2013-08-22  7:54   ` Jan Beulich
2013-08-21 17:28 ` Andy Lutomirski
  -- strict thread matches above, loose matches on Subject: below --
2013-08-21 13:48 David Vrabel

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=20130821154238.GV18673@moon \
    --to=gorcunov@gmail.com \
    --cc=JBeulich@suse.com \
    --cc=Xen-devel@lists.xen.org \
    --cc=akpm@linux-foundation.org \
    --cc=boris.ostrovsky@oracle.com \
    --cc=david.vrabel@citrix.com \
    --cc=hpa@zytor.com \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@amacapital.net \
    --cc=mingo@redhat.com \
    --cc=torvalds@linux-foundation.org \
    --cc=xemul@parallels.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.