From: David Hildenbrand <david@redhat.com>
To: Dave Hansen <dave.hansen@intel.com>,
Nitesh Narayan Lal <nitesh@redhat.com>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, pbonzini@redhat.com, lcapitulino@redhat.com,
pagupta@redhat.com, wei.w.wang@intel.com,
yang.zhang.wz@gmail.com, riel@surriel.com, mst@redhat.com,
dodgen@google.com, konrad.wilk@oracle.com, dhildenb@redhat.com,
aarcange@redhat.com, alexander.duyck@gmail.com,
john.starks@microsoft.com, mhocko@suse.com
Subject: Re: [RFC][Patch v11 1/2] mm: page_hinting: core infrastructure
Date: Mon, 15 Jul 2019 11:33:01 +0200 [thread overview]
Message-ID: <46336efb-3243-0083-1d20-7e8578131679@redhat.com> (raw)
In-Reply-To: <f9bca947-f88e-51a7-fdaf-4403fda1b783@intel.com>
On 11.07.19 20:21, Dave Hansen wrote:
> On 7/10/19 12:51 PM, Nitesh Narayan Lal wrote:
>> +static void bm_set_pfn(struct page *page)
>> +{
>> + struct zone *zone = page_zone(page);
>> + int zone_idx = page_zonenum(page);
>> + unsigned long bitnr = 0;
>> +
>> + lockdep_assert_held(&zone->lock);
>> + bitnr = pfn_to_bit(page, zone_idx);
>> + /*
>> + * TODO: fix possible underflows.
>> + */
>> + if (free_area[zone_idx].bitmap &&
>> + bitnr < free_area[zone_idx].nbits &&
>> + !test_and_set_bit(bitnr, free_area[zone_idx].bitmap))
>> + atomic_inc(&free_area[zone_idx].free_pages);
>> +}
>
> Let's say I have two NUMA nodes, each with ZONE_NORMAL and ZONE_MOVABLE
> and each zone with 1GB of memory:
>
> Node: 0 1
> NORMAL 0->1GB 2->3GB
> MOVABLE 1->2GB 3->4GB
>
> This code will allocate two bitmaps. The ZONE_NORMAL bitmap will
> represent data from 0->3GB and the ZONE_MOVABLE bitmap will represent
> data from 1->4GB. That's the result of this code:
>
>> + if (free_area[zone_idx].base_pfn) {
>> + free_area[zone_idx].base_pfn =
>> + min(free_area[zone_idx].base_pfn,
>> + zone->zone_start_pfn);
>> + free_area[zone_idx].end_pfn =
>> + max(free_area[zone_idx].end_pfn,
>> + zone->zone_start_pfn +
>> + zone->spanned_pages);
>
> But that means that both bitmaps will have space for PFNs in the other
> zone type, which is completely bogus. This is fundamental because the
> data structures are incorrectly built per zone *type* instead of per zone.
>
I don't think it's incorrect, it's just not optimal in all scenarios.
E.g., in you example, this approach would "waste" 2 * 1GB of tracking
data for the wholes (2* 64bytes when using 1 bit for 2MB).
FWIW, this is not a numa-specific thingy. We can have sparse zones
easily on single-numa systems.
Node: 0
NORMAL 0->1GB, 2->3GB
MOVABLE 1->2GB, 3->4GB
So tracking it per zones instead instead of zone type is only one part
of the story.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2019-07-15 9:33 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-10 19:51 [RFC][PATCH v11 0/2] mm: Support for page hinting Nitesh Narayan Lal
2019-07-10 19:51 ` [RFC][Patch v11 1/2] mm: page_hinting: core infrastructure Nitesh Narayan Lal
2019-07-10 20:45 ` Dave Hansen
2019-07-11 11:48 ` Nitesh Narayan Lal
2019-07-11 15:25 ` Nitesh Narayan Lal
2019-07-11 15:50 ` Nitesh Narayan Lal
2019-07-11 16:22 ` Dave Hansen
2019-07-11 16:36 ` Nitesh Narayan Lal
2019-07-11 16:45 ` Dave Hansen
2019-07-11 16:52 ` Nitesh Narayan Lal
2019-07-15 9:26 ` David Hildenbrand
2019-07-10 21:56 ` Alexander Duyck
2019-07-11 17:58 ` Nitesh Narayan Lal
2019-07-11 23:20 ` Alexander Duyck
2019-07-12 1:12 ` Nitesh Narayan Lal
2019-07-12 16:22 ` Alexander Duyck
2019-07-12 16:25 ` Nitesh Narayan Lal
2019-08-08 11:41 ` Nitesh Narayan Lal
2019-07-11 18:21 ` Dave Hansen
2019-07-15 9:33 ` David Hildenbrand [this message]
2019-07-15 14:40 ` David Hildenbrand
2019-07-10 19:51 ` [RFC][Patch v11 2/2] virtio-balloon: page_hinting: reporting to the host Nitesh Narayan Lal
2019-07-24 19:47 ` Michael S. Tsirkin
2019-07-24 19:56 ` David Hildenbrand
2019-07-24 20:10 ` Nitesh Narayan Lal
2019-07-24 20:06 ` Nitesh Narayan Lal
2019-07-10 19:53 ` [QEMU Patch] virtio-baloon: Support for page hinting Nitesh Narayan Lal
2019-07-10 20:17 ` Alexander Duyck
2019-07-11 12:03 ` Nitesh Narayan Lal
2019-07-11 8:49 ` Cornelia Huck
2019-07-11 11:13 ` Nitesh Narayan Lal
2019-07-11 18:55 ` Michael S. Tsirkin
2019-07-11 19:06 ` Nitesh Narayan Lal
2019-07-11 22:36 ` Alexander Duyck
2019-07-10 20:19 ` [RFC][PATCH v11 0/2] mm: " Dave Hansen
2019-07-11 11:37 ` Nitesh Narayan Lal
2019-07-10 23:40 ` Alexander Duyck
2019-07-11 11:30 ` Nitesh Narayan Lal
2019-07-11 14:58 ` Alexander Duyck
2019-07-11 15:03 ` Nitesh Narayan Lal
2019-07-11 15:08 ` Alexander Duyck
2019-07-11 15:19 ` Nitesh Narayan Lal
2019-07-11 17:01 ` 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=46336efb-3243-0083-1d20-7e8578131679@redhat.com \
--to=david@redhat.com \
--cc=aarcange@redhat.com \
--cc=alexander.duyck@gmail.com \
--cc=dave.hansen@intel.com \
--cc=dhildenb@redhat.com \
--cc=dodgen@google.com \
--cc=john.starks@microsoft.com \
--cc=konrad.wilk@oracle.com \
--cc=kvm@vger.kernel.org \
--cc=lcapitulino@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=mst@redhat.com \
--cc=nitesh@redhat.com \
--cc=pagupta@redhat.com \
--cc=pbonzini@redhat.com \
--cc=riel@surriel.com \
--cc=wei.w.wang@intel.com \
--cc=yang.zhang.wz@gmail.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).