From: Peter Zijlstra <a.p.zijlstra@chello.nl> To: Andrea Arcangeli <aarcange@redhat.com>, Avi Kivity <avi@redhat.com>, Thomas Gleixner <tglx@linutronix.de>, Rik van Riel <riel@redhat.com>, Ingo Molnar <mingo@elte.hu>, akpm@linux-fou Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, Benjamin Herrenschmidt <benh@kernel.crashing.org>, David Miller <davem@davemloft.net>, Hugh Dickins <hugh.dickins@tiscali.co.uk>, Mel Gorman <mel@csn.ul.ie>, Nick Piggin <npiggin@suse.de>, Peter Zijlstra <a.p.zijlstra@chello.nl>, Paul McKenney <paulmck@linux.vnet.ibm.com>, Yanmin Zhang <yanmin_zhang@linux.intel.com>, Stephen Rothwell <sfr@canb.auug.org.au> Subject: [PATCH 20/20] mm: Optimize page_lock_anon_vma() fast-path Date: Sat, 28 Aug 2010 16:16:57 +0200 [thread overview] Message-ID: <20100828142456.804579956@chello.nl> (raw) In-Reply-To: 20100828141637.421594670@chello.nl [-- Attachment #1: mm-opt-page_lock_anon_vma.patch --] [-- Type: text/plain, Size: 2722 bytes --] Optimize the page_lock_anon_vma() fast path to be one LOCKed op, instead of two. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> --- mm/rmap.c | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 61 insertions(+), 4 deletions(-) Index: linux-2.6/mm/rmap.c =================================================================== --- linux-2.6.orig/mm/rmap.c +++ linux-2.6/mm/rmap.c @@ -353,20 +353,69 @@ out: return anon_vma; } +/* + * Similar to page_get_anon_vma() except it locks the anon_vma. + * + * Its a little more complex as it tries to keep the fast path to a single + * atomic op -- the trylock. If we fail the trylock, we fall back to getting a + * reference like with page_get_anon_vma() and then block on the mutex. + */ struct anon_vma *page_lock_anon_vma(struct page *page) { - struct anon_vma *anon_vma = page_get_anon_vma(page); + struct anon_vma *anon_vma = NULL; + unsigned long anon_mapping; + + rcu_read_lock(); + anon_mapping = (unsigned long) ACCESS_ONCE(page->mapping); + if ((anon_mapping & PAGE_MAPPING_FLAGS) != PAGE_MAPPING_ANON) + goto out; + if (!page_mapped(page)) + goto out; + + anon_vma = (struct anon_vma *) (anon_mapping - PAGE_MAPPING_ANON); + if (mutex_trylock(&anon_vma->root->lock)) { + /* + * If we observe a !0 refcount, then holding the lock ensures + * the anon_vma will not go away, see __put_anon_vma(). + */ + if (!atomic_read(&anon_vma->refcount)) { + anon_vma_unlock(anon_vma); + anon_vma = NULL; + } + goto out; + } + + /* trylock failed, we got to sleep */ + if (!atomic_inc_not_zero(&anon_vma->refcount)) { + anon_vma = NULL; + goto out; + } + + /* we pinned the anon_vma, its safe to sleep */ + rcu_read_unlock(); + anon_vma_lock(anon_vma); + + if (atomic_dec_and_test(&anon_vma->refcount)) { + /* + * Oops, we held the last refcount, release the lock + * and bail -- can't simply use put_anon_vma() because + * we'll deadlock on the anon_vma_lock() recursion. + */ + anon_vma_unlock(anon_vma); + __put_anon_vma(anon_vma); + anon_vma = NULL; + } - if (anon_vma) - anon_vma_lock(anon_vma); + return anon_vma; +out: + rcu_read_unlock(); return anon_vma; } void page_unlock_anon_vma(struct anon_vma *anon_vma) { anon_vma_unlock(anon_vma); - put_anon_vma(anon_vma); } /* @@ -1458,6 +1507,14 @@ int try_to_munlock(struct page *page) void __put_anon_vma(struct anon_vma *anon_vma) { + /* + * Synchronize against page_lock_anon_vma() such that + * we can safely hold the lock without the anon_vma getting + * freed. + */ + anon_vma_lock(anon_vma); + anon_vma_unlock(anon_vma); + if (anon_vma->root != anon_vma) put_anon_vma(anon_vma->root); anon_vma_free(anon_vma);
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <a.p.zijlstra@chello.nl> To: Andrea Arcangeli <aarcange@redhat.com>, Avi Kivity <avi@redhat.com>, Thomas Gleixner <tglx@linutronix.de>, Rik van Riel <riel@redhat.com>, Ingo Molnar <mingo@elte.hu>, akpm@linux-foundation.org, Linus Torvalds <torvalds@linux-foundation.org> Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, Benjamin Herrenschmidt <benh@kernel.crashing.org>, David Miller <davem@davemloft.net>, Hugh Dickins <hugh.dickins@tiscali.co.uk>, Mel Gorman <mel@csn.ul.ie>, Nick Piggin <npiggin@suse.de>, Peter Zijlstra <a.p.zijlstra@chello.nl>, Paul McKenney <paulmck@linux.vnet.ibm.com>, Yanmin Zhang <yanmin_zhang@linux.intel.com>, Stephen Rothwell <sfr@canb.auug.org.au> Subject: [PATCH 20/20] mm: Optimize page_lock_anon_vma() fast-path Date: Sat, 28 Aug 2010 16:16:57 +0200 [thread overview] Message-ID: <20100828142456.804579956@chello.nl> (raw) Message-ID: <20100828141657.s_ttov8_1Y0D6NY4NwnCn17g6Sr61gUC_Y4pOmYxgZs@z> (raw) In-Reply-To: 20100828141637.421594670@chello.nl [-- Attachment #1: mm-opt-page_lock_anon_vma.patch --] [-- Type: text/plain, Size: 2724 bytes --] Optimize the page_lock_anon_vma() fast path to be one LOCKed op, instead of two. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> --- mm/rmap.c | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 61 insertions(+), 4 deletions(-) Index: linux-2.6/mm/rmap.c =================================================================== --- linux-2.6.orig/mm/rmap.c +++ linux-2.6/mm/rmap.c @@ -353,20 +353,69 @@ out: return anon_vma; } +/* + * Similar to page_get_anon_vma() except it locks the anon_vma. + * + * Its a little more complex as it tries to keep the fast path to a single + * atomic op -- the trylock. If we fail the trylock, we fall back to getting a + * reference like with page_get_anon_vma() and then block on the mutex. + */ struct anon_vma *page_lock_anon_vma(struct page *page) { - struct anon_vma *anon_vma = page_get_anon_vma(page); + struct anon_vma *anon_vma = NULL; + unsigned long anon_mapping; + + rcu_read_lock(); + anon_mapping = (unsigned long) ACCESS_ONCE(page->mapping); + if ((anon_mapping & PAGE_MAPPING_FLAGS) != PAGE_MAPPING_ANON) + goto out; + if (!page_mapped(page)) + goto out; + + anon_vma = (struct anon_vma *) (anon_mapping - PAGE_MAPPING_ANON); + if (mutex_trylock(&anon_vma->root->lock)) { + /* + * If we observe a !0 refcount, then holding the lock ensures + * the anon_vma will not go away, see __put_anon_vma(). + */ + if (!atomic_read(&anon_vma->refcount)) { + anon_vma_unlock(anon_vma); + anon_vma = NULL; + } + goto out; + } + + /* trylock failed, we got to sleep */ + if (!atomic_inc_not_zero(&anon_vma->refcount)) { + anon_vma = NULL; + goto out; + } + + /* we pinned the anon_vma, its safe to sleep */ + rcu_read_unlock(); + anon_vma_lock(anon_vma); + + if (atomic_dec_and_test(&anon_vma->refcount)) { + /* + * Oops, we held the last refcount, release the lock + * and bail -- can't simply use put_anon_vma() because + * we'll deadlock on the anon_vma_lock() recursion. + */ + anon_vma_unlock(anon_vma); + __put_anon_vma(anon_vma); + anon_vma = NULL; + } - if (anon_vma) - anon_vma_lock(anon_vma); + return anon_vma; +out: + rcu_read_unlock(); return anon_vma; } void page_unlock_anon_vma(struct anon_vma *anon_vma) { anon_vma_unlock(anon_vma); - put_anon_vma(anon_vma); } /* @@ -1458,6 +1507,14 @@ int try_to_munlock(struct page *page) void __put_anon_vma(struct anon_vma *anon_vma) { + /* + * Synchronize against page_lock_anon_vma() such that + * we can safely hold the lock without the anon_vma getting + * freed. + */ + anon_vma_lock(anon_vma); + anon_vma_unlock(anon_vma); + if (anon_vma->root != anon_vma) put_anon_vma(anon_vma->root); anon_vma_free(anon_vma);
next prev parent reply other threads:[~2010-08-28 14:16 UTC|newest] Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top 2010-08-28 14:16 [PATCH 00/20] mm: Preemptibility -v4 Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 01/20] powerpc: Use call_rcu_sched() for pagetables Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-31 6:10 ` Benjamin Herrenschmidt 2010-08-28 14:16 ` [PATCH 02/20] mm: Improve page_lock_anon_vma() comment Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 03/20] mm: Rename drop_anon_vma to put_anon_vma Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 15:08 ` Pekka Enberg 2010-08-28 14:16 ` [PATCH 04/20] mm: Move anon_vma ref out from under CONFIG_KSM Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 05/20] mm: Simplify anon_vma refcounts Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 15:13 ` Pekka Enberg 2010-08-28 14:16 ` [PATCH 06/20] mm: Use refcounts for page_lock_anon_vma() Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 07/20] mm: Preemptible mmu_gather Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 08/20] powerpc: " Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-31 6:26 ` Benjamin Herrenschmidt 2010-08-31 6:31 ` Benjamin Herrenschmidt 2010-08-31 6:31 ` Benjamin Herrenschmidt 2010-08-31 9:14 ` Peter Zijlstra 2010-08-31 9:14 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 09/20] sparc: " Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 10/20] s390: preemptible mmu_gather Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 11/20] arm: Preemptible mmu_gather Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 12/20] sh: " Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 13/20] um: " Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 14/20] ia64: " Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-30 15:44 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 15/20] mm, powerpc: Move the RCU page-table freeing into generic code Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 16/20] lockdep, mutex: Provide mutex_lock_nest_lock Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 17/20] mutex: Provide mutex_is_contended Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 18/20] mm: Convert i_mmap_lock and anon_vma->lock to mutexes Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` [PATCH 19/20] mm: Extended batches for generic mmu_gather Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra 2010-08-28 14:16 ` Peter Zijlstra [this message] 2010-08-28 14:16 ` [PATCH 20/20] mm: Optimize page_lock_anon_vma() fast-path Peter Zijlstra 2010-08-28 14:32 ` [PATCH 00/20] mm: Preemptibility -v4 Peter Zijlstra 2010-08-28 22:28 ` David Miller 2010-08-28 22:41 ` Peter Zijlstra 2010-08-28 14:56 ` Piotr Hosowicz 2010-08-28 15:10 ` Peter Zijlstra 2010-08-28 15:17 ` Piotr Hosowicz 2010-08-28 15:23 ` Peter Zijlstra 2010-08-28 16:01 ` Piotr Hosowicz 2010-08-29 12:46 ` Piotr Hosowicz 2010-08-29 12:46 ` Piotr Hosowicz 2010-08-29 13:37 ` Peter Zijlstra 2010-08-29 13:43 ` Piotr Hosowicz 2010-08-31 14:02 ` Piotr Hosowicz 2010-08-31 14:14 ` Piotr Hosowicz 2010-09-02 14:53 ` Piotr Hosowicz 2010-08-28 15:19 ` Pekka Enberg 2010-08-28 15:27 ` Peter Zijlstra 2010-08-28 15:27 ` Peter Zijlstra [not found] ` <AANLkTikSm2Mq8hGNac9rpFH-3pvryw2kW57EP45Ny6Vp@mail.gmail.com> 2010-09-14 5:36 ` Alex,Shi 2010-09-14 7:42 ` Peter Zijlstra 2010-09-14 7:42 ` Peter Zijlstra -- strict thread matches above, loose matches on Subject: below -- 2010-10-18 11:24 [PATCH 00/20] mm: Preemptibility -v5 Peter Zijlstra 2010-10-18 11:24 ` [PATCH 20/20] mm: Optimize page_lock_anon_vma() fast-path Peter Zijlstra 2010-10-18 11:24 ` Peter Zijlstra 2011-04-01 12:12 [PATCH 00/20] mm: Preemptibility -v10 Peter Zijlstra 2011-04-01 12:13 ` [PATCH 20/20] mm: Optimize page_lock_anon_vma() fast-path Peter Zijlstra 2011-04-01 12:13 ` Peter Zijlstra 2011-04-19 20:08 ` Andrew Morton 2011-04-19 20:08 ` Andrew Morton 2011-04-20 12:38 ` Peter Zijlstra 2011-04-20 12:38 ` Peter Zijlstra 2011-04-20 15:00 ` Peter Zijlstra 2011-04-20 15:00 ` Peter Zijlstra
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=20100828142456.804579956@chello.nl \ --to=a.p.zijlstra@chello.nl \ --cc=aarcange@redhat.com \ --cc=akpm@linux-fou \ --cc=avi@redhat.com \ --cc=benh@kernel.crashing.org \ --cc=davem@davemloft.net \ --cc=hugh.dickins@tiscali.co.uk \ --cc=linux-arch@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=mel@csn.ul.ie \ --cc=mingo@elte.hu \ --cc=npiggin@suse.de \ --cc=paulmck@linux.vnet.ibm.com \ --cc=riel@redhat.com \ --cc=sfr@canb.auug.org.au \ --cc=tglx@linutronix.de \ --cc=yanmin_zhang@linux.intel.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: linkBe 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).