All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@amacapital.net>
To: Minchan Kim <minchan@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>
Cc: Rik van Riel <riel@redhat.com>, Mel Gorman <mgorman@suse.de>,
	Hugh Dickins <hughd@google.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	John Stultz <john.stultz@linaro.org>, Jason Evans <je@fb.com>
Subject: Re: [RFC 0/6] mm: support madvise(MADV_FREE)
Date: Tue, 18 Mar 2014 10:55:24 -0700	[thread overview]
Message-ID: <5328888C.7030402@mit.edu> (raw)
In-Reply-To: <1394779070-8545-1-git-send-email-minchan@kernel.org>

On 03/13/2014 11:37 PM, Minchan Kim wrote:
> This patch is an attempt to support MADV_FREE for Linux.
> 
> Rationale is following as.
> 
> Allocators call munmap(2) when user call free(3) if ptr is
> in mmaped area. But munmap isn't cheap because it have to clean up
> all pte entries, unlinking a vma and returns free pages to buddy
> so overhead would be increased linearly by mmaped area's size.
> So they like madvise_dontneed rather than munmap.
> 
> "dontneed" holds read-side lock of mmap_sem so other threads
> of the process could go with concurrent page faults so it is
> better than munmap if it's not lack of address space.
> But the problem is that most of allocator reuses that address
> space soonish so applications see page fault, page allocation,
> page zeroing if allocator already called madvise_dontneed
> on the address space.
> 
> For avoidng that overheads, other OS have supported MADV_FREE.
> The idea is just mark pages as lazyfree when madvise called
> and purge them if memory pressure happens. Otherwise, VM doesn't
> detach pages on the address space so application could use
> that memory space without above overheads.

I must be missing something.

If the application issues MADV_FREE and then writes to the MADV_FREEd
range, the kernel needs to know that the pages are no longer safe to
lazily free.  This would presumably happen via a page fault on write.
For that to happen reliably, the kernel has to write protect the pages
when MADV_FREE is called, which in turn requires flushing the TLBs.

How does this end up being faster than munmap?

--Andy

WARNING: multiple messages have this Message-ID (diff)
From: Andy Lutomirski <luto@amacapital.net>
To: Minchan Kim <minchan@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>
Cc: Rik van Riel <riel@redhat.com>, Mel Gorman <mgorman@suse.de>,
	Hugh Dickins <hughd@google.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	John Stultz <john.stultz@linaro.org>, Jason Evans <je@fb.com>
Subject: Re: [RFC 0/6] mm: support madvise(MADV_FREE)
Date: Tue, 18 Mar 2014 10:55:24 -0700	[thread overview]
Message-ID: <5328888C.7030402@mit.edu> (raw)
In-Reply-To: <1394779070-8545-1-git-send-email-minchan@kernel.org>

On 03/13/2014 11:37 PM, Minchan Kim wrote:
> This patch is an attempt to support MADV_FREE for Linux.
> 
> Rationale is following as.
> 
> Allocators call munmap(2) when user call free(3) if ptr is
> in mmaped area. But munmap isn't cheap because it have to clean up
> all pte entries, unlinking a vma and returns free pages to buddy
> so overhead would be increased linearly by mmaped area's size.
> So they like madvise_dontneed rather than munmap.
> 
> "dontneed" holds read-side lock of mmap_sem so other threads
> of the process could go with concurrent page faults so it is
> better than munmap if it's not lack of address space.
> But the problem is that most of allocator reuses that address
> space soonish so applications see page fault, page allocation,
> page zeroing if allocator already called madvise_dontneed
> on the address space.
> 
> For avoidng that overheads, other OS have supported MADV_FREE.
> The idea is just mark pages as lazyfree when madvise called
> and purge them if memory pressure happens. Otherwise, VM doesn't
> detach pages on the address space so application could use
> that memory space without above overheads.

I must be missing something.

If the application issues MADV_FREE and then writes to the MADV_FREEd
range, the kernel needs to know that the pages are no longer safe to
lazily free.  This would presumably happen via a page fault on write.
For that to happen reliably, the kernel has to write protect the pages
when MADV_FREE is called, which in turn requires flushing the TLBs.

How does this end up being faster than munmap?

--Andy

--
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	other threads:[~2014-03-18 17:55 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-14  6:37 [RFC 0/6] mm: support madvise(MADV_FREE) Minchan Kim
2014-03-14  6:37 ` Minchan Kim
2014-03-14  6:37 ` [RFC 1/6] mm: clean up PAGE_MAPPING_FLAGS Minchan Kim
2014-03-14  6:37   ` Minchan Kim
2014-03-14  6:37 ` [RFC 2/6] mm: work deactivate_page with anon pages Minchan Kim
2014-03-14  6:37   ` Minchan Kim
2014-03-14  6:37 ` [RFC 3/6] mm: support madvise(MADV_FREE) Minchan Kim
2014-03-14  6:37   ` Minchan Kim
2014-03-14  7:49   ` Minchan Kim
2014-03-14  7:49     ` Minchan Kim
2014-03-14 13:33   ` Kirill A. Shutemov
2014-03-14 13:33     ` Kirill A. Shutemov
2014-03-14 15:24     ` Minchan Kim
2014-03-14 15:24       ` Minchan Kim
2014-03-18 18:26   ` Johannes Weiner
2014-03-18 18:26     ` Johannes Weiner
2014-03-19  1:22     ` Minchan Kim
2014-03-14  6:37 ` [RFC 4/6] mm: add stat about lazyfree pages Minchan Kim
2014-03-14  6:37   ` Minchan Kim
2014-03-14  6:37 ` [RFC 5/6] mm: reclaim lazyfree pages in swapless system Minchan Kim
2014-03-14  6:37   ` Minchan Kim
2014-03-14  6:37 ` [RFC 6/6] mm: ksm: don't merge lazyfree page Minchan Kim
2014-03-14  6:37   ` Minchan Kim
2014-03-14  7:37 ` [RFC 0/6] mm: support madvise(MADV_FREE) Zhang Yanfei
2014-03-14  7:37   ` Zhang Yanfei
2014-03-14  7:56   ` Minchan Kim
2014-03-14  7:56     ` Minchan Kim
2014-03-18 17:55 ` Andy Lutomirski [this message]
2014-03-18 17:55   ` Andy Lutomirski
2014-03-19  0:18   ` Minchan Kim
2014-03-19  0:23     ` Andy Lutomirski
2014-03-19  0:23       ` Andy Lutomirski
2014-03-19  1:02       ` Minchan Kim
2014-03-19  5:15       ` Johannes Weiner
2014-03-19  5:15         ` Johannes Weiner

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=5328888C.7030402@mit.edu \
    --to=luto@amacapital.net \
    --cc=akpm@linux-foundation.org \
    --cc=dave.hansen@intel.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=je@fb.com \
    --cc=john.stultz@linaro.org \
    --cc=kosaki.motohiro@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=minchan@kernel.org \
    --cc=riel@redhat.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.