All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: linux-mml@vger.kernel.org, Andy Lutomirski <luto@amacapital.net>,
	Andrew Morton <akpm@linux-foundation.org>,
	Denys Vlasenko <dvlasenk@redhat.com>,
	Brian Gerst <brgerst@gmail.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Borislav Petkov <bp@alien8.de>, "H. Peter Anvin" <hpa@zytor.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Oleg Nesterov <oleg@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Waiman Long <Waiman.Long@hp.com>
Subject: [RFC PATCH 00/12] x86/mm: Implement lockless pgd_alloc()/pgd_free()
Date: Thu, 11 Jun 2015 16:07:05 +0200	[thread overview]
Message-ID: <1434031637-9091-1-git-send-email-mingo@kernel.org> (raw)

Waiman Long reported 'pgd_lock' contention on high CPU count systems and proposed 
moving pgd_lock on a separate cacheline to eliminate false sharing and to reduce 
some of the lock bouncing overhead.

I think we can do much better: this series eliminates the pgd_list and makes 
pgd_alloc()/pgd_free() lockless.

Now the lockless initialization of the PGD has a few preconditions, which the 
initial part of the series implements:

 - no PGD clearing is allowed, only additions. This makes sense as a single PGD 
   entry covers 512 GB of RAM so the 4K overhead per 0.5TB of RAM mapped is 
   miniscule.

The patches after that convert existing pgd_list users to walk the task list.

PGD locking is kept intact: coherency guarantees between the CPA, vmalloc, 
hotplug, etc. code are unchanged.

The final patches eliminate the pgd_list and thus make pgd_alloc()/pgd_free() 
lockless.

The patches have been boot tested on 64-bit and 32-bit x86 systems.

Architectures not making use of the new facility are unaffected.

Thanks,

	Ingo

=====
Ingo Molnar (12):
  x86/mm/pat: Don't free PGD entries on memory unmap
  x86/mm/hotplug: Remove pgd_list use from the memory hotplug code
  x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()
  x86/mm/hotplug: Simplify sync_global_pgds()
  mm: Introduce arch_pgd_init_late()
  x86/mm: Enable and use the arch_pgd_init_late() method
  x86/virt/guest/xen: Remove use of pgd_list from the Xen guest code
  x86/mm: Remove pgd_list use from vmalloc_sync_all()
  x86/mm/pat/32: Remove pgd_list use from the PAT code
  x86/mm: Make pgd_alloc()/pgd_free() lockless
  x86/mm: Remove pgd_list leftovers
  x86/mm: Simplify pgd_alloc()

 arch/Kconfig                      |   9 ++++
 arch/x86/Kconfig                  |   1 +
 arch/x86/include/asm/pgtable.h    |   3 --
 arch/x86/include/asm/pgtable_64.h |   3 +-
 arch/x86/mm/fault.c               |  24 +++++++----
 arch/x86/mm/init_64.c             |  73 +++++++++++----------------------
 arch/x86/mm/pageattr.c            |  34 ++++++++--------
 arch/x86/mm/pgtable.c             | 129 +++++++++++++++++++++++++++++-----------------------------
 arch/x86/xen/mmu.c                |  34 +++++++++++++---
 fs/exec.c                         |   3 ++
 include/linux/mm.h                |   6 +++
 kernel/fork.c                     |  16 ++++++++
 12 files changed, 183 insertions(+), 152 deletions(-)

-- 
2.1.4


             reply	other threads:[~2015-06-11 14:07 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-11 14:07 Ingo Molnar [this message]
2015-06-11 14:07 ` [PATCH 01/12] x86/mm/pat: Don't free PGD entries on memory unmap Ingo Molnar
2015-06-11 14:07 ` [PATCH 02/12] x86/mm/hotplug: Remove pgd_list use from the memory hotplug code Ingo Molnar
2015-06-12 22:22   ` Oleg Nesterov
2015-06-12 22:48   ` Waiman Long
2015-06-13  7:46     ` Ingo Molnar
2015-06-12 23:15   ` Oleg Nesterov
2015-06-13  7:48     ` Ingo Molnar
2015-06-11 14:07 ` [PATCH 03/12] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable() Ingo Molnar
2015-06-11 14:07 ` [PATCH 04/12] x86/mm/hotplug: Simplify sync_global_pgds() Ingo Molnar
2015-06-12 22:28   ` Oleg Nesterov
2015-06-13  7:51     ` Ingo Molnar
2015-06-11 14:07 ` [PATCH 05/12] mm: Introduce arch_pgd_init_late() Ingo Molnar
2015-06-11 18:23   ` Andrew Morton
2015-06-12  8:16     ` Ingo Molnar
2015-06-12 21:12   ` Oleg Nesterov
2015-06-13  6:54     ` Ingo Molnar
2015-06-11 14:07 ` [PATCH 06/12] x86/mm: Enable and use the arch_pgd_init_late() method Ingo Molnar
2015-06-11 20:04   ` Boris Ostrovsky
2015-06-12  8:19     ` Ingo Molnar
2015-06-12 22:50   ` Oleg Nesterov
2015-06-12 22:57     ` Oleg Nesterov
2015-06-13  6:49       ` Ingo Molnar
2015-06-13  6:47     ` Ingo Molnar
2015-06-13  6:52       ` Ingo Molnar
2015-06-13 17:45         ` Oleg Nesterov
2015-06-14  8:13           ` Ingo Molnar
2015-06-14 20:54             ` Oleg Nesterov
2015-06-14 22:10               ` Oleg Nesterov
2015-06-11 14:07 ` [PATCH 07/12] x86/virt/guest/xen: Remove use of pgd_list from the Xen guest code Ingo Molnar
2015-06-11 20:37   ` Linus Torvalds
2015-06-11 20:41     ` Linus Torvalds
2015-06-12  7:23       ` Ingo Molnar
     [not found]         ` <CA+55aFxXM=DN32JqNmJ=JoMum5OPnsRohCry-=2T=LabX2hzVQ@mail.gmail.com>
2015-06-12  8:04           ` Ingo Molnar
2015-06-12 20:38             ` Oleg Nesterov
2015-06-12 20:53               ` Oleg Nesterov
2015-06-12 22:30                 ` Boris Ostrovsky
2015-06-12 23:36                   ` Oleg Nesterov
2015-06-13  7:26               ` Ingo Molnar
2015-06-13 18:00                 ` Oleg Nesterov
2015-06-11 14:07 ` [PATCH 08/12] x86/mm: Remove pgd_list use from vmalloc_sync_all() Ingo Molnar
2015-06-11 15:04   ` Andy Lutomirski
2015-06-11 15:49     ` Ingo Molnar
2015-06-11 16:10       ` Andy Lutomirski
2015-06-11 14:07 ` [PATCH 09/12] x86/mm/pat/32: Remove pgd_list use from the PAT code Ingo Molnar
2015-06-11 14:07 ` [PATCH 10/12] x86/mm: Make pgd_alloc()/pgd_free() lockless Ingo Molnar
2015-06-11 14:07 ` [PATCH 11/12] x86/mm: Remove pgd_list leftovers Ingo Molnar
2015-06-11 14:07 ` [PATCH 12/12] x86/mm: Simplify pgd_alloc() Ingo Molnar
2015-06-11 14:13 ` [RFC PATCH 00/12] x86/mm: Implement lockless pgd_alloc()/pgd_free() Ingo Molnar
2015-06-11 14:13   ` Ingo Molnar
2015-06-11 16:46 ` H. Peter Anvin

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=1434031637-9091-1-git-send-email-mingo@kernel.org \
    --to=mingo@kernel.org \
    --cc=Waiman.Long@hp.com \
    --cc=akpm@linux-foundation.org \
    --cc=bp@alien8.de \
    --cc=brgerst@gmail.com \
    --cc=dvlasenk@redhat.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mml@vger.kernel.org \
    --cc=luto@amacapital.net \
    --cc=oleg@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --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 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.