From: David Hildenbrand <david@redhat.com>
To: Alexander Duyck <alexander.duyck@gmail.com>,
virtio-dev@lists.oasis-open.org, kvm@vger.kernel.org,
mst@redhat.com, catalin.marinas@arm.com, dave.hansen@intel.com,
linux-kernel@vger.kernel.org, willy@infradead.org,
mhocko@kernel.org, linux-mm@kvack.org, akpm@linux-foundation.org,
will@kernel.org, linux-arm-kernel@lists.infradead.org,
osalvador@suse.de
Cc: yang.zhang.wz@gmail.com, pagupta@redhat.com, riel@surriel.com,
konrad.wilk@oracle.com, ying.huang@intel.com,
lcapitulino@redhat.com, wei.w.wang@intel.com,
aarcange@redhat.com, nitesh@redhat.com, pbonzini@redhat.com,
dan.j.williams@intel.com, fengguang.wu@intel.com,
alexander.h.duyck@linux.intel.com,
kirill.shutemov@linux.intel.com
Subject: Re: [PATCH v9 1/8] mm: Add per-cpu logic to page shuffling
Date: Mon, 9 Sep 2019 10:14:24 +0200 [thread overview]
Message-ID: <0df2e5d0-af92-04b4-aa7d-891387874039@redhat.com> (raw)
In-Reply-To: <20190907172512.10910.74435.stgit@localhost.localdomain>
On 07.09.19 19:25, Alexander Duyck wrote:
> From: Alexander Duyck <alexander.h.duyck@linux.intel.com>
>
> Change the logic used to generate randomness in the suffle path so that we
> can avoid cache line bouncing. The previous logic was sharing the offset
> and entropy word between all CPUs. As such this can result in cache line
> bouncing and will ultimately hurt performance when enabled.
So, usually we perform such changes if there is real evidence. Do you
have any such performance numbers to back your claims?
>
> To resolve this I have moved to a per-cpu logic for maintaining a unsigned
> long containing some amount of bits, and an offset value for which bit we
> can use for entropy with each call.
>
> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
> Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
> ---
> mm/shuffle.c | 33 +++++++++++++++++++++++----------
> 1 file changed, 23 insertions(+), 10 deletions(-)
>
> diff --git a/mm/shuffle.c b/mm/shuffle.c
> index 3ce12481b1dc..9ba542ecf335 100644
> --- a/mm/shuffle.c
> +++ b/mm/shuffle.c
> @@ -183,25 +183,38 @@ void __meminit __shuffle_free_memory(pg_data_t *pgdat)
> shuffle_zone(z);
> }
>
> +struct batched_bit_entropy {
> + unsigned long entropy_bool;
> + int position;
> +};
> +
> +static DEFINE_PER_CPU(struct batched_bit_entropy, batched_entropy_bool);
> +
> void add_to_free_area_random(struct page *page, struct free_area *area,
> int migratetype)
> {
> - static u64 rand;
> - static u8 rand_bits;
> + struct batched_bit_entropy *batch;
> + unsigned long entropy;
> + int position;
>
> /*
> - * The lack of locking is deliberate. If 2 threads race to
> - * update the rand state it just adds to the entropy.
> + * We shouldn't need to disable IRQs as the only caller is
> + * __free_one_page and it should only be called with the zone lock
> + * held and either from IRQ context or with local IRQs disabled.
> */
> - if (rand_bits == 0) {
> - rand_bits = 64;
> - rand = get_random_u64();
> + batch = raw_cpu_ptr(&batched_entropy_bool);
> + position = batch->position;
> +
> + if (--position < 0) {
> + batch->entropy_bool = get_random_long();
> + position = BITS_PER_LONG - 1;
> }
>
> - if (rand & 1)
> + batch->position = position;
> + entropy = batch->entropy_bool;
> +
> + if (1ul & (entropy >> position))
> add_to_free_area(page, area, migratetype);
> else
> add_to_free_area_tail(page, area, migratetype);
> - rand_bits--;
> - rand >>= 1;
> }
>
>
--
Thanks,
David / dhildenb
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2019-09-09 8:14 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-07 17:25 [PATCH v9 0/8] stg mail -e --version=v9 \ Alexander Duyck
2019-09-07 17:25 ` [PATCH v9 1/8] mm: Add per-cpu logic to page shuffling Alexander Duyck
2019-09-09 8:14 ` David Hildenbrand [this message]
2019-09-09 15:11 ` Alexander Duyck
2019-09-10 12:11 ` Michal Hocko
2019-09-10 22:14 ` Alexander Duyck
2019-09-10 22:11 ` Alexander Duyck
2019-09-09 9:07 ` Kirill A. Shutemov
2019-09-09 15:12 ` Alexander Duyck
2019-09-07 17:25 ` [PATCH v9 2/8] mm: Adjust shuffle code to allow for future coalescing Alexander Duyck
2019-09-09 8:19 ` David Hildenbrand
2019-09-09 9:47 ` Kirill A. Shutemov
2019-09-09 15:22 ` Alexander Duyck
2019-09-09 15:35 ` Kirill A. Shutemov
2019-09-09 15:37 ` Alexander Duyck
2019-09-09 16:43 ` Alexander Duyck
2019-09-09 17:00 ` Kirill A. Shutemov
2019-09-10 12:20 ` Michal Hocko
2019-09-10 14:48 ` Alexander Duyck
2019-09-07 17:25 ` [PATCH v9 3/8] mm: Move set/get_pcppage_migratetype to mmzone.h Alexander Duyck
2019-09-09 8:22 ` David Hildenbrand
2019-09-09 9:56 ` Kirill A. Shutemov
2019-09-09 18:01 ` Alexander Duyck
2019-09-09 18:12 ` Alexander Duyck
2019-09-10 12:23 ` Michal Hocko
2019-09-10 14:46 ` Alexander Duyck
2019-09-10 17:45 ` Michal Hocko
2019-09-10 20:26 ` Alexander Duyck
2019-09-07 17:25 ` [PATCH v9 4/8] mm: Use zone and order instead of free area in free_list manipulators Alexander Duyck
2019-09-10 12:27 ` Michal Hocko
2019-09-07 17:25 ` [PATCH v9 5/8] arm64: Move hugetlb related definitions out of pgtable.h to page-defs.h Alexander Duyck
2019-09-09 8:52 ` David Hildenbrand
2019-09-09 15:27 ` Alexander Duyck
2019-09-17 17:48 ` Will Deacon
2019-09-17 20:07 ` Alexander Duyck
2019-09-07 17:25 ` [PATCH v9 6/8] mm: Introduce Reported pages Alexander Duyck
2019-09-09 14:42 ` Kirill A. Shutemov
2019-09-09 16:25 ` Alexander Duyck
2019-09-09 16:33 ` Kirill A. Shutemov
2019-09-07 17:26 ` [PATCH v9 7/8] virtio-balloon: Pull page poisoning config out of free page hinting Alexander Duyck
2019-09-09 8:59 ` David Hildenbrand
2019-09-09 15:31 ` Alexander Duyck
2019-09-07 17:26 ` [PATCH v9 8/8] virtio-balloon: Add support for providing unused page reports to host Alexander Duyck
2019-09-07 17:34 ` [PATCH v9 0/8] mm / virtio: Provide support for unused page reporting Alexander Duyck
2019-09-10 12:42 ` [PATCH v9 0/8] stg mail -e --version=v9 \ Michal Hocko
2019-09-10 14:42 ` Alexander Duyck
2019-09-10 14:47 ` Michal Hocko
2019-09-10 16:05 ` Alexander Duyck
2019-09-10 16:18 ` [virtio-dev] " Dr. David Alan Gilbert
2019-09-10 16:22 ` David Hildenbrand
2019-09-11 9:23 ` Michael S. Tsirkin
2019-09-11 9:50 ` David Hildenbrand
2019-09-10 17:52 ` Michal Hocko
2019-09-10 18:00 ` Michal Hocko
2019-09-10 20:37 ` Alexander Duyck
2019-09-10 21:23 ` Alexander Duyck
2019-09-11 11:36 ` Michal Hocko
2019-09-11 11:47 ` David Hildenbrand
2019-09-11 12:08 ` Michael S. Tsirkin
2019-09-11 12:19 ` Michal Hocko
2019-09-11 12:25 ` Michal Hocko
2019-09-11 12:42 ` David Hildenbrand
2019-09-11 12:54 ` Michal Hocko
2019-09-11 13:03 ` David Hildenbrand
2019-09-11 13:20 ` Michal Hocko
2019-09-11 13:51 ` Michal Hocko
2019-09-11 16:09 ` David Hildenbrand
2019-09-12 7:16 ` Michal Hocko
2019-09-12 7:47 ` David Hildenbrand
2019-09-12 9:26 ` Michal Hocko
2019-09-12 12:00 ` Nitesh Narayan Lal
2019-09-11 14:03 ` Nitesh Narayan Lal
2019-09-11 16:02 ` David Hildenbrand
2019-09-11 13:19 ` Nitesh Narayan Lal
2019-09-11 12:55 ` Nitesh Narayan Lal
2019-09-11 15:12 ` Alexander Duyck
2019-09-12 9:19 ` Michal Hocko
2019-09-12 10:24 ` Kirill A. Shutemov
2019-09-12 11:11 ` Michal Hocko
2019-09-12 15:42 ` Alexander Duyck
2019-09-12 16:35 ` Mel Gorman
2019-09-12 17:48 ` Alexander Duyck
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=0df2e5d0-af92-04b4-aa7d-891387874039@redhat.com \
--to=david@redhat.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alexander.duyck@gmail.com \
--cc=alexander.h.duyck@linux.intel.com \
--cc=catalin.marinas@arm.com \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@intel.com \
--cc=fengguang.wu@intel.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=konrad.wilk@oracle.com \
--cc=kvm@vger.kernel.org \
--cc=lcapitulino@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=mst@redhat.com \
--cc=nitesh@redhat.com \
--cc=osalvador@suse.de \
--cc=pagupta@redhat.com \
--cc=pbonzini@redhat.com \
--cc=riel@surriel.com \
--cc=virtio-dev@lists.oasis-open.org \
--cc=wei.w.wang@intel.com \
--cc=will@kernel.org \
--cc=willy@infradead.org \
--cc=yang.zhang.wz@gmail.com \
--cc=ying.huang@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: 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).