Linux-mm Archive on lore.kernel.org
 help / color / Atom feed
From: Minchan Kim <minchan@kernel.org>
To: Dan Streetman <ddstreet@ieee.org>
Cc: Chulmin Kim <cmlaika.kim@samsung.com>,
	Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linux-MM <linux-mm@kvack.org>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Subject: Re: [PATCH v7 11/12] zsmalloc: page migration support
Date: Thu, 2 Feb 2017 17:48:18 +0900
Message-ID: <20170202084818.GA20335@bbox> (raw)
In-Reply-To: <CALZtONDQ5yQKV6-jpJGBg7gakV8-7XXmmAATSQq346Tby4WLsg@mail.gmail.com>

Hi Dan,

On Tue, Jan 31, 2017 at 08:09:53AM -0500, Dan Streetman wrote:
> On Mon, Jan 30, 2017 at 7:10 PM, Minchan Kim <minchan@kernel.org> wrote:
> > Hi Dan,
> >
> > On Thu, Jan 26, 2017 at 12:04:03PM -0500, Dan Streetman wrote:
> >> On Wed, Jan 25, 2017 at 12:26 AM, Minchan Kim <minchan@kernel.org> wrote:
> >> > On Tue, Jan 24, 2017 at 11:06:51PM -0500, Chulmin Kim wrote:
> >> >> On 01/23/2017 12:40 AM, Minchan Kim wrote:
> >> >> >On Mon, Jan 23, 2017 at 02:30:56PM +0900, Sergey Senozhatsky wrote:
> >> >> >>On (01/23/17 14:22), Minchan Kim wrote:
> >> >> >>[..]
> >> >> >>>>Anyway, I will let you know the situation when it gets more clear.
> >> >> >>>
> >> >> >>>Yeb, Thanks.
> >> >> >>>
> >> >> >>>Perhaps, did you tried flush page before the writing?
> >> >> >>>I think arm64 have no d-cache alising problem but worth to try it.
> >> >> >>>Who knows :)
> >> >> >>
> >> >> >>I thought that flush_dcache_page() is only for cases when we write
> >> >> >>to page (store that makes pages dirty), isn't it?
> >> >> >
> >> >> >I think we need both because to see recent stores done by the user.
> >> >> >I'm not sure it should be done by block device driver rather than
> >> >> >page cache. Anyway, brd added it so worth to try it, I thought. :)
> >> >> >
> >> >>
> >> >> Thanks for the suggestion!
> >> >> It might be helpful
> >> >> though proving it is not easy as the problem appears rarely.
> >> >>
> >> >> Have you thought about
> >> >> zram swap or zswap dealing with self modifying code pages (ex. JIT)?
> >> >> (arm64 may have i-cache aliasing problem)
> >> >
> >> > It can happen, I think, although I don't know how arm64 handles it.
> >> >
> >> >>
> >> >> If it is problematic,
> >> >> especiallly zswap (without flush_dcache_page in zswap_frontswap_load()) may
> >> >> provide the corrupted data
> >> >> and even swap out (compressing) may see the corrupted data sooner or later,
> >> >> i guess.
> >> >
> >> > try_to_unmap_one calls flush_cache_page which I hope to handle swap-out side
> >> > but for swap-in, I think zswap need flushing logic because it's first
> >> > touch of the user buffer so it's his resposibility.
> >>
> >> Hmm, I don't think zswap needs to, because all the cache aliases were
> >> flushed when the page was written out.  After that, any access to the
> >> page will cause a fault, and the fault will cause the page to be read
> >> back in (via zswap).  I don't see how the page could be cached at any
> >> time between the swap write-out and swap read-in, so there should be
> >> no need to flush any caches when it's read back in; am I missing
> >> something?
> >
> > Documentation/cachetlb.txt says
> >
> >   void flush_dcache_page(struct page *page)
> >
> >         Any time the kernel writes to a page cache page, _OR_
> >         the kernel is about to read from a page cache page and
> >         user space shared/writable mappings of this page potentially
> >         exist, this routine is called.
> >
> > For swap-in side, I don't see any logic to prevent the aliasing
> > problem. Let's consider other examples like cow_user_page->
> > copy_user_highpage. For architectures which can make aliasing,
> > it has arch specific functions which has flushing function.
> 
> COW works with a page that has a physical backing.  swap-in does not.
> COW pages can be accessed normally; swapped out pages cannot.
> 
> >
> > IOW, if a kernel makes store operation to the page which will
> > be mapped to user space address, kernel should call flush function.
> > Otherwise, user space will miss recent update from kernel side.
> 
> as I said before, when it's swapped out caches are flushed, and the
> page mapping invalidated, so it will cause a fault on any access, and
> thus cause swap to re-load the page from disk (or zswap).  So how
> would a cache of the page be created after swap-out, but before
> swap-in?  It's not possible for user space to have any caches to the
> page, unless (as I said) I'm missing something?
> 

Let's assume VIVT architecture which get index and tag from virtual
address.

In zswap_frontswap_load, let's assume dst is kernel virtual address,
0xc0002000 so cacheline for 0xc0002000 has recent uptodate data.
Now, VM mapped the page into user address space, 0x80003000, for
example. In that case, userland application try to read data from
0x80003000 but it is associated with another cacheline so it cannot
see recent uptodate data written by zswap.

flush_dcache_page will handle both kernel side and user side.
I'm not sure how I explain well. :-(

Thanks.

--
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>

  parent reply index

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-31 23:21 [PATCH v7 00/12] Support non-lru page migration Minchan Kim
2016-05-31 23:21 ` [PATCH v7 01/12] mm: use put_page to free page instead of putback_lru_page Minchan Kim
2016-05-31 23:21 ` [PATCH v7 02/12] mm: migrate: support non-lru movable page migration Minchan Kim
2016-05-31 23:21 ` [PATCH v7 03/12] mm: balloon: use general non-lru movable page feature Minchan Kim
2016-05-31 23:21 ` [PATCH v7 04/12] zsmalloc: keep max_object in size_class Minchan Kim
2016-05-31 23:21 ` [PATCH v7 05/12] zsmalloc: use bit_spin_lock Minchan Kim
2016-05-31 23:21 ` [PATCH v7 06/12] zsmalloc: use accessor Minchan Kim
2016-05-31 23:21 ` [PATCH v7 07/12] zsmalloc: factor page chain functionality out Minchan Kim
2016-05-31 23:21 ` [PATCH v7 08/12] zsmalloc: introduce zspage structure Minchan Kim
2016-05-31 23:21 ` [PATCH v7 09/12] zsmalloc: separate free_zspage from putback_zspage Minchan Kim
2016-05-31 23:21 ` [PATCH v7 10/12] zsmalloc: use freeobj for index Minchan Kim
2016-05-31 23:21 ` [PATCH v7 11/12] zsmalloc: page migration support Minchan Kim
2016-06-01 14:09   ` Vlastimil Babka
2016-06-02  0:25     ` Minchan Kim
2016-06-02 11:44       ` Vlastimil Babka
2016-06-01 21:39   ` Andrew Morton
2016-06-02  0:15     ` Minchan Kim
     [not found]   ` <CGME20170119001317epcas1p188357c77e1f4ff08b6d3dcb76dedca06@epcas1p1.samsung.com>
2017-01-19  0:13     ` Chulmin Kim
2017-01-19  2:44       ` Minchan Kim
2017-01-19  3:39         ` Chulmin Kim
2017-01-19  6:21           ` Minchan Kim
2017-01-19  8:16             ` Chulmin Kim
2017-01-23  5:22               ` Minchan Kim
2017-01-23  5:30                 ` Sergey Senozhatsky
2017-01-23  5:40                   ` Minchan Kim
2017-01-25  4:06                     ` Chulmin Kim
2017-01-25  4:25                       ` Sergey Senozhatsky
2017-01-25  5:26                       ` Minchan Kim
2017-01-26 17:04                         ` Dan Streetman
2017-01-31  0:10                           ` Minchan Kim
2017-01-31 13:09                             ` Dan Streetman
2017-02-01  6:51                               ` Minchan Kim
2017-02-01 19:38                                 ` Dan Streetman
2017-02-02  8:48                               ` Minchan Kim [this message]
2016-05-31 23:21 ` [PATCH v7 12/12] zram: use __GFP_MOVABLE for memory allocation Minchan Kim
2016-06-01 21:41 ` [PATCH v7 00/12] Support non-lru page migration Andrew Morton
2016-06-01 22:40   ` Daniel Vetter
2016-06-02  0:36   ` Minchan Kim
2016-06-15  7:59 ` Sergey Senozhatsky
2016-06-15 23:12   ` Minchan Kim
2016-06-16  2:48     ` Sergey Senozhatsky
2016-06-16  2:58       ` Minchan Kim
2016-06-16  4:23         ` Sergey Senozhatsky
2016-06-16  4:47           ` Minchan Kim
2016-06-16  5:22             ` Sergey Senozhatsky
2016-06-16  6:47               ` Minchan Kim
2016-06-16  8:42                 ` Sergey Senozhatsky
2016-06-16 10:09                   ` Minchan Kim
2016-06-17  7:28                     ` Joonsoo Kim

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=20170202084818.GA20335@bbox \
    --to=minchan@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=cmlaika.kim@samsung.com \
    --cc=ddstreet@ieee.org \
    --cc=linux-mm@kvack.org \
    --cc=sergey.senozhatsky.work@gmail.com \
    --cc=sergey.senozhatsky@gmail.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

Linux-mm Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-mm/0 linux-mm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-mm linux-mm/ https://lore.kernel.org/linux-mm \
		linux-mm@kvack.org
	public-inbox-index linux-mm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kvack.linux-mm


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git