From: Peter Zijlstra <peterz@infradead.org>
To: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Cc: paulmck@linux.vnet.ibm.com, akpm@linux-foundation.org,
kirill@shutemov.name, ak@linux.intel.com, mhocko@kernel.org,
dave@stgolabs.net, jack@suse.cz,
Matthew Wilcox <willy@infradead.org>,
benh@kernel.crashing.org, mpe@ellerman.id.au, paulus@samba.org,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>,
hpa@zytor.com, Will Deacon <will.deacon@arm.com>,
Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
Andrea Arcangeli <aarcange@redhat.com>,
Alexei Starovoitov <alexei.starovoitov@gmail.com>,
kemi.wang@intel.com, sergey.senozhatsky.work@gmail.com,
Daniel Jordan <daniel.m.jordan@oracle.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
haren@linux.vnet.ibm.com, khandual@linux.vnet.ibm.com,
npiggin@gmail.com, bsingharora@gmail.com,
Tim Chen <tim.c.chen@linux.intel.com>,
linuxppc-dev@lists.ozlabs.org, x86@kernel.org
Subject: Re: [PATCH v9 17/24] mm: Protect mm_rb tree with a rwlock
Date: Wed, 14 Mar 2018 09:48:44 +0100 [thread overview]
Message-ID: <20180314084844.GP4043@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <1520963994-28477-18-git-send-email-ldufour@linux.vnet.ibm.com>
On Tue, Mar 13, 2018 at 06:59:47PM +0100, Laurent Dufour wrote:
> This change is inspired by the Peter's proposal patch [1] which was
> protecting the VMA using SRCU. Unfortunately, SRCU is not scaling well in
> that particular case, and it is introducing major performance degradation
> due to excessive scheduling operations.
Do you happen to have a little more detail on that?
> diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
> index 34fde7111e88..28c763ea1036 100644
> --- a/include/linux/mm_types.h
> +++ b/include/linux/mm_types.h
> @@ -335,6 +335,7 @@ struct vm_area_struct {
> struct vm_userfaultfd_ctx vm_userfaultfd_ctx;
> #ifdef CONFIG_SPECULATIVE_PAGE_FAULT
> seqcount_t vm_sequence;
> + atomic_t vm_ref_count; /* see vma_get(), vma_put() */
> #endif
> } __randomize_layout;
>
> @@ -353,6 +354,9 @@ struct kioctx_table;
> struct mm_struct {
> struct vm_area_struct *mmap; /* list of VMAs */
> struct rb_root mm_rb;
> +#ifdef CONFIG_SPECULATIVE_PAGE_FAULT
> + rwlock_t mm_rb_lock;
> +#endif
> u32 vmacache_seqnum; /* per-thread vmacache */
> #ifdef CONFIG_MMU
> unsigned long (*get_unmapped_area) (struct file *filp,
When I tried this, it simply traded contention on mmap_sem for
contention on these two cachelines.
This was for the concurrent fault benchmark, where mmap_sem is only ever
acquired for reading (so no blocking ever happens) and the bottle-neck
was really pure cacheline access.
Only by using RCU can you avoid that thrashing.
Also note that if your database allocates the one giant mapping, it'll
be _one_ VMA and that vm_ref_count gets _very_ hot indeed.
next prev parent reply other threads:[~2018-03-14 8:52 UTC|newest]
Thread overview: 98+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-13 17:59 [PATCH v9 00/24] Speculative page faults Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 01/24] mm: Introduce CONFIG_SPECULATIVE_PAGE_FAULT Laurent Dufour
2018-03-25 21:50 ` David Rientjes
2018-03-28 7:49 ` Laurent Dufour
2018-03-28 10:16 ` David Rientjes
2018-03-28 11:15 ` Laurent Dufour
2018-03-28 21:18 ` David Rientjes
2018-03-13 17:59 ` [PATCH v9 02/24] x86/mm: Define CONFIG_SPECULATIVE_PAGE_FAULT Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 03/24] powerpc/mm: " Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 05/24] mm: Introduce pte_spinlock for FAULT_FLAG_SPECULATIVE Laurent Dufour
2018-03-25 21:50 ` David Rientjes
2018-03-28 8:15 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 06/24] mm: make pte_unmap_same compatible with SPF Laurent Dufour
2018-03-27 21:18 ` David Rientjes
2018-03-28 8:27 ` Laurent Dufour
2018-03-28 10:20 ` David Rientjes
2018-03-28 10:43 ` Laurent Dufour
2018-04-03 19:10 ` Jerome Glisse
2018-04-03 20:40 ` David Rientjes
2018-04-03 21:04 ` Jerome Glisse
2018-04-04 9:53 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 08/24] mm: Protect VMA modifications using VMA sequence count Laurent Dufour
2018-03-27 21:45 ` David Rientjes
2018-03-28 16:57 ` Laurent Dufour
2018-03-27 21:57 ` David Rientjes
2018-03-28 17:10 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 09/24] mm: protect mremap() against SPF hanlder Laurent Dufour
2018-03-27 22:12 ` David Rientjes
2018-03-28 18:11 ` Laurent Dufour
2018-03-28 21:21 ` David Rientjes
2018-04-04 8:24 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 10/24] mm: Protect SPF handler against anon_vma changes Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 11/24] mm: Cache some VMA fields in the vm_fault structure Laurent Dufour
2018-04-02 22:24 ` David Rientjes
2018-04-04 15:48 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 12/24] mm/migrate: Pass vm_fault pointer to migrate_misplaced_page() Laurent Dufour
2018-04-02 23:00 ` David Rientjes
2018-03-13 17:59 ` [PATCH v9 13/24] mm: Introduce __lru_cache_add_active_or_unevictable Laurent Dufour
2018-04-02 23:11 ` David Rientjes
2018-03-13 17:59 ` [PATCH v9 14/24] mm: Introduce __maybe_mkwrite() Laurent Dufour
2018-04-02 23:12 ` David Rientjes
2018-04-04 15:56 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 15/24] mm: Introduce __vm_normal_page() Laurent Dufour
2018-04-02 23:18 ` David Rientjes
2018-04-04 16:04 ` Laurent Dufour
2018-04-03 19:39 ` Jerome Glisse
2018-04-03 20:45 ` David Rientjes
2018-04-04 16:26 ` Laurent Dufour
2018-04-04 21:59 ` Jerome Glisse
2018-04-05 12:53 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 16/24] mm: Introduce __page_add_new_anon_rmap() Laurent Dufour
2018-04-02 23:57 ` David Rientjes
2018-04-10 16:30 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 17/24] mm: Protect mm_rb tree with a rwlock Laurent Dufour
2018-03-14 8:48 ` Peter Zijlstra [this message]
2018-03-14 16:25 ` Laurent Dufour
2018-03-16 10:23 ` [mm] b33ddf50eb: INFO:trying_to_register_non-static_key kernel test robot
2018-03-16 16:38 ` Laurent Dufour
2018-04-03 0:11 ` [PATCH v9 17/24] mm: Protect mm_rb tree with a rwlock David Rientjes
2018-04-06 14:23 ` Laurent Dufour
2018-04-10 16:20 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 18/24] mm: Provide speculative fault infrastructure Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 19/24] mm: Adding speculative page fault failure trace events Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 20/24] perf: Add a speculative page fault sw event Laurent Dufour
2018-03-26 21:43 ` David Rientjes
2018-03-13 17:59 ` [PATCH v9 21/24] perf tools: Add support for the SPF perf event Laurent Dufour
2018-03-26 21:44 ` David Rientjes
2018-03-27 3:49 ` Andi Kleen
2018-04-10 6:47 ` David Rientjes
2018-04-12 13:44 ` Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 22/24] mm: Speculative page fault handler return VMA Laurent Dufour
2018-03-13 17:59 ` [PATCH v9 23/24] x86/mm: Add speculative pagefault handling Laurent Dufour
2018-03-26 21:41 ` David Rientjes
2018-03-13 17:59 ` [PATCH v9 24/24] powerpc/mm: Add speculative page fault Laurent Dufour
2018-03-26 21:39 ` David Rientjes
2018-03-14 13:11 ` [PATCH v9 00/24] Speculative page faults Michal Hocko
2018-03-14 13:33 ` Laurent Dufour
2018-04-13 13:34 ` Laurent Dufour
2018-03-22 1:21 ` Ganesh Mahendran
2018-03-29 12:49 ` Laurent Dufour
[not found] ` <1520963994-28477-5-git-send-email-ldufour@linux.vnet.ibm.com>
2018-03-25 21:50 ` [PATCH v9 04/24] mm: Prepare for FAULT_FLAG_SPECULATIVE David Rientjes
2018-03-28 10:27 ` Laurent Dufour
2018-04-03 21:57 ` David Rientjes
2018-04-04 9:23 ` Laurent Dufour
[not found] ` <1520963994-28477-8-git-send-email-ldufour@linux.vnet.ibm.com>
2018-03-17 7:51 ` [mm] b1f0502d04: INFO:trying_to_register_non-static_key kernel test robot
2018-03-21 12:21 ` Laurent Dufour
2018-03-25 22:10 ` David Rientjes
2018-03-28 13:30 ` Laurent Dufour
2018-04-04 0:48 ` David Rientjes
2018-04-04 1:03 ` David Rientjes
2018-04-04 10:28 ` Laurent Dufour
2018-04-04 10:19 ` Laurent Dufour
2018-04-04 21:53 ` David Rientjes
2018-04-05 16:55 ` Laurent Dufour
2018-03-27 21:30 ` [PATCH v9 07/24] mm: VMA sequence count David Rientjes
2018-03-28 17:58 ` Laurent Dufour
2018-04-03 20:37 ` [PATCH v9 00/24] Speculative page faults Jerome Glisse
2018-04-04 7:59 ` Laurent Dufour
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=20180314084844.GP4043@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=aarcange@redhat.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=alexei.starovoitov@gmail.com \
--cc=benh@kernel.crashing.org \
--cc=bsingharora@gmail.com \
--cc=daniel.m.jordan@oracle.com \
--cc=dave@stgolabs.net \
--cc=haren@linux.vnet.ibm.com \
--cc=hpa@zytor.com \
--cc=jack@suse.cz \
--cc=kemi.wang@intel.com \
--cc=khandual@linux.vnet.ibm.com \
--cc=kirill@shutemov.name \
--cc=ldufour@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mhocko@kernel.org \
--cc=mingo@redhat.com \
--cc=mpe@ellerman.id.au \
--cc=npiggin@gmail.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=paulus@samba.org \
--cc=sergey.senozhatsky.work@gmail.com \
--cc=sergey.senozhatsky@gmail.com \
--cc=tglx@linutronix.de \
--cc=tim.c.chen@linux.intel.com \
--cc=will.deacon@arm.com \
--cc=willy@infradead.org \
--cc=x86@kernel.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 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).