All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: willy@infradead.org, wei.w.wang@intel.com,
	virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
	qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
	kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
	akpm@linux-foundation.org, mawilcox@microsoft.com,
	david@redhat.com, cornelia.huck@de.ibm.com,
	mgorman@techsingularity.net, aarcange@redhat.com,
	amit.shah@redhat.com, pbonzini@redhat.com,
	liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
	quan.xu@aliyun.com, nilal@redhat.com, riel@redhat.com
Subject: [virtio-dev] Re: [PATCH v19 3/7] xbitmap: add more operations
Date: Fri, 15 Dec 2017 20:26:22 +0200	[thread overview]
Message-ID: <20171215202238-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <201712160121.BEJ26052.HOFFOOQFMLtSVJ@I-love.SAKURA.ne.jp>

On Sat, Dec 16, 2017 at 01:21:52AM +0900, Tetsuo Handa wrote:
> My understanding is that virtio-balloon wants to handle sparsely spreaded
> unsigned long values (which is PATCH 4/7) and wants to find all chunks of
> consecutive "1" bits efficiently. Therefore, I guess that holding the values
> in ascending order at store time is faster than sorting the values at read
> time.

Are you asking why is a bitmap used here, as opposed to a tree?  It's
not just store versus read. There's also the issue that memory can get
highly fragmented, if it is, the number of 1s is potentially very high.
A bitmap can use as little as 1 bit per value, it is hard to beat in
this respect.

-- 
MST

---------------------------------------------------------------------
To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org


WARNING: multiple messages have this Message-ID (diff)
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: willy@infradead.org, wei.w.wang@intel.com,
	virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
	qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
	kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
	akpm@linux-foundation.org, mawilcox@microsoft.com,
	david@redhat.com, cornelia.huck@de.ibm.com,
	mgorman@techsingularity.net, aarcange@redhat.com,
	amit.shah@redhat.com, pbonzini@redhat.com,
	liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
	quan.xu@aliyun.com, nilal@redhat.com, riel@redhat.com
Subject: Re: [PATCH v19 3/7] xbitmap: add more operations
Date: Fri, 15 Dec 2017 20:26:22 +0200	[thread overview]
Message-ID: <20171215202238-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <201712160121.BEJ26052.HOFFOOQFMLtSVJ@I-love.SAKURA.ne.jp>

On Sat, Dec 16, 2017 at 01:21:52AM +0900, Tetsuo Handa wrote:
> My understanding is that virtio-balloon wants to handle sparsely spreaded
> unsigned long values (which is PATCH 4/7) and wants to find all chunks of
> consecutive "1" bits efficiently. Therefore, I guess that holding the values
> in ascending order at store time is faster than sorting the values at read
> time.

Are you asking why is a bitmap used here, as opposed to a tree?  It's
not just store versus read. There's also the issue that memory can get
highly fragmented, if it is, the number of 1s is potentially very high.
A bitmap can use as little as 1 bit per value, it is hard to beat in
this respect.

-- 
MST

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: willy@infradead.org, wei.w.wang@intel.com,
	virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
	qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
	kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
	akpm@linux-foundation.org, mawilcox@microsoft.com,
	david@redhat.com, cornelia.huck@de.ibm.com,
	mgorman@techsingularity.net, aarcange@redhat.com,
	amit.shah@redhat.com, pbonzini@redhat.com,
	liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
	quan.xu@aliyun.com, nilal@redhat.com, riel@redhat.com
Subject: Re: [PATCH v19 3/7] xbitmap: add more operations
Date: Fri, 15 Dec 2017 20:26:22 +0200	[thread overview]
Message-ID: <20171215202238-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <201712160121.BEJ26052.HOFFOOQFMLtSVJ@I-love.SAKURA.ne.jp>

On Sat, Dec 16, 2017 at 01:21:52AM +0900, Tetsuo Handa wrote:
> My understanding is that virtio-balloon wants to handle sparsely spreaded
> unsigned long values (which is PATCH 4/7) and wants to find all chunks of
> consecutive "1" bits efficiently. Therefore, I guess that holding the values
> in ascending order at store time is faster than sorting the values at read
> time.

Are you asking why is a bitmap used here, as opposed to a tree?  It's
not just store versus read. There's also the issue that memory can get
highly fragmented, if it is, the number of 1s is potentially very high.
A bitmap can use as little as 1 bit per value, it is hard to beat in
this respect.

-- 
MST

WARNING: multiple messages have this Message-ID (diff)
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: willy@infradead.org, wei.w.wang@intel.com,
	virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
	qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
	kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
	akpm@linux-foundation.org, mawilcox@microsoft.com,
	david@redhat.com, cornelia.huck@de.ibm.com,
	mgorman@techsingularity.net, aarcange@redhat.com,
	amit.shah@redhat.com, pbonzini@redhat.com,
	liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
	quan.xu@aliyun.com, nilal@redhat.com, riel@redhat.com
Subject: Re: [Qemu-devel] [PATCH v19 3/7] xbitmap: add more operations
Date: Fri, 15 Dec 2017 20:26:22 +0200	[thread overview]
Message-ID: <20171215202238-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <201712160121.BEJ26052.HOFFOOQFMLtSVJ@I-love.SAKURA.ne.jp>

On Sat, Dec 16, 2017 at 01:21:52AM +0900, Tetsuo Handa wrote:
> My understanding is that virtio-balloon wants to handle sparsely spreaded
> unsigned long values (which is PATCH 4/7) and wants to find all chunks of
> consecutive "1" bits efficiently. Therefore, I guess that holding the values
> in ascending order at store time is faster than sorting the values at read
> time.

Are you asking why is a bitmap used here, as opposed to a tree?  It's
not just store versus read. There's also the issue that memory can get
highly fragmented, if it is, the number of 1s is potentially very high.
A bitmap can use as little as 1 bit per value, it is hard to beat in
this respect.

-- 
MST

  reply	other threads:[~2017-12-15 18:26 UTC|newest]

Thread overview: 179+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-12 11:55 [virtio-dev] [PATCH v19 0/7] Virtio-balloon Enhancement Wei Wang
2017-12-12 11:55 ` [Qemu-devel] " Wei Wang
2017-12-12 11:55 ` Wei Wang
2017-12-12 11:55 ` Wei Wang
2017-12-12 11:55 ` [PATCH v19 1/7] xbitmap: Introduce xbitmap Wei Wang
2017-12-12 11:55 ` [virtio-dev] " Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 12:53   ` Philippe Ombredanne
2017-12-12 12:53     ` [Qemu-devel] " Philippe Ombredanne
2017-12-12 12:53     ` Philippe Ombredanne
2017-12-12 12:53   ` Philippe Ombredanne
2017-12-15 11:05   ` kbuild test robot
2017-12-15 11:05     ` [Qemu-devel] " kbuild test robot
2017-12-15 11:05     ` kbuild test robot
2017-12-15 13:24     ` Matthew Wilcox
2017-12-15 13:24     ` Matthew Wilcox
2017-12-15 13:24       ` [Qemu-devel] " Matthew Wilcox
2017-12-15 13:24       ` Matthew Wilcox
2017-12-16 10:10       ` Wei Wang
2017-12-16 10:10       ` [virtio-dev] " Wei Wang
2017-12-16 10:10         ` [Qemu-devel] " Wei Wang
2017-12-16 10:10         ` Wei Wang
2017-12-16 10:10         ` Wei Wang
2017-12-15 11:05   ` kbuild test robot
2017-12-12 11:55 ` [PATCH v19 2/7] xbitmap: potential improvement Wei Wang
2017-12-12 11:55 ` [virtio-dev] " Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-15  3:07   ` kbuild test robot
2017-12-15  3:07     ` [Qemu-devel] " kbuild test robot
2017-12-15  3:07     ` kbuild test robot
2017-12-15  3:07     ` kbuild test robot
2017-12-12 11:55 ` [virtio-dev] [PATCH v19 3/7] xbitmap: add more operations Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 13:20   ` Tetsuo Handa
2017-12-12 13:20     ` [Qemu-devel] " Tetsuo Handa
2017-12-12 13:20     ` Tetsuo Handa
2017-12-13 12:26     ` [virtio-dev] " Wei Wang
2017-12-13 12:26       ` [Qemu-devel] " Wei Wang
2017-12-13 12:26       ` Wei Wang
2017-12-13 12:26       ` Wei Wang
2017-12-13 14:16       ` Tetsuo Handa
2017-12-13 14:16         ` [Qemu-devel] " Tetsuo Handa
2017-12-13 14:16         ` Tetsuo Handa
2017-12-14  3:47         ` [virtio-dev] " Wei Wang
2017-12-14  3:47           ` [Qemu-devel] " Wei Wang
2017-12-14  3:47           ` Wei Wang
2017-12-14  3:47           ` Wei Wang
2017-12-14 11:47           ` [virtio-dev] " Wei Wang
2017-12-14 11:47             ` [Qemu-devel] " Wei Wang
2017-12-14 11:47             ` Wei Wang
2017-12-14 11:47             ` Wei Wang
2017-12-14 11:47           ` Wei Wang
2017-12-14 16:29           ` Tetsuo Handa
2017-12-14 16:29             ` [Qemu-devel] " Tetsuo Handa
2017-12-14 16:29             ` Tetsuo Handa
2017-12-14 18:12             ` Matthew Wilcox
2017-12-14 18:12               ` [Qemu-devel] " Matthew Wilcox
2017-12-14 18:12               ` Matthew Wilcox
2017-12-15 16:21               ` Tetsuo Handa
2017-12-15 16:21                 ` [Qemu-devel] " Tetsuo Handa
2017-12-15 16:21                 ` Tetsuo Handa
2017-12-15 18:26                 ` Michael S. Tsirkin [this message]
2017-12-15 18:26                   ` [Qemu-devel] " Michael S. Tsirkin
2017-12-15 18:26                   ` Michael S. Tsirkin
2017-12-15 18:26                   ` Michael S. Tsirkin
2017-12-16  4:31                   ` Tetsuo Handa
2017-12-16  4:31                     ` [Qemu-devel] " Tetsuo Handa
2017-12-16  4:31                     ` Tetsuo Handa
2017-12-16  5:05                     ` Matthew Wilcox
2017-12-16  5:05                       ` [Qemu-devel] " Matthew Wilcox
2017-12-16  5:05                       ` Matthew Wilcox
2017-12-16  5:57                       ` Tetsuo Handa
2017-12-16  5:57                         ` [Qemu-devel] " Tetsuo Handa
2017-12-16  5:57                         ` Tetsuo Handa
2017-12-16  5:05                     ` Matthew Wilcox
2017-12-15 18:26                 ` Michael S. Tsirkin
2017-12-15 18:49                 ` Matthew Wilcox
2017-12-15 18:49                 ` Matthew Wilcox
2017-12-15 18:49                   ` [Qemu-devel] " Matthew Wilcox
2017-12-15 18:49                   ` Matthew Wilcox
2017-12-15 19:22                   ` Matthew Wilcox
2017-12-15 19:22                   ` Matthew Wilcox
2017-12-15 19:22                     ` [Qemu-devel] " Matthew Wilcox
2017-12-15 19:22                     ` Matthew Wilcox
2017-12-17 13:47                     ` [virtio-dev] " Wang, Wei W
2017-12-17 13:47                       ` [Qemu-devel] " Wang, Wei W
2017-12-17 13:47                       ` Wang, Wei W
2017-12-17 13:47                       ` Wang, Wei W
2017-12-17 13:47                       ` Wang, Wei W
2017-12-17 22:18                       ` Matthew Wilcox
2017-12-17 22:18                       ` Matthew Wilcox
2017-12-17 22:18                         ` [Qemu-devel] " Matthew Wilcox
2017-12-17 22:18                         ` Matthew Wilcox
2017-12-17 22:18                         ` Matthew Wilcox
2017-12-18  2:33                         ` Wei Wang
2017-12-18  2:33                         ` [virtio-dev] " Wei Wang
2017-12-18  2:33                           ` [Qemu-devel] " Wei Wang
2017-12-18  2:33                           ` Wei Wang
2017-12-18  2:33                           ` Wei Wang
2017-12-18  2:33                           ` Wei Wang
2017-12-18  2:59                           ` Matthew Wilcox
2017-12-18  2:59                             ` [Qemu-devel] " Matthew Wilcox
2017-12-18  2:59                             ` Matthew Wilcox
2017-12-18  2:59                             ` Matthew Wilcox
2017-12-18  2:59                           ` Matthew Wilcox
2017-12-17 13:47                     ` Wang, Wei W
2017-12-14 18:12             ` Matthew Wilcox
2017-12-16 10:14             ` Wei Wang
2017-12-16 10:14             ` [virtio-dev] " Wei Wang
2017-12-16 10:14               ` [Qemu-devel] " Wei Wang
2017-12-16 10:14               ` Wei Wang
2017-12-16 10:14               ` Wei Wang
2017-12-14  3:47         ` Wei Wang
2017-12-14 12:37       ` Matthew Wilcox
2017-12-14 12:37         ` [Qemu-devel] " Matthew Wilcox
2017-12-14 12:37         ` Matthew Wilcox
2017-12-14 12:37       ` Matthew Wilcox
2017-12-13 12:26     ` Wei Wang
2017-12-15 18:42   ` Matthew Wilcox
2017-12-15 18:42     ` [Qemu-devel] " Matthew Wilcox
2017-12-15 18:42     ` Matthew Wilcox
2017-12-16 10:12     ` Wei Wang
2017-12-16 10:12     ` [virtio-dev] " Wei Wang
2017-12-16 10:12       ` [Qemu-devel] " Wei Wang
2017-12-16 10:12       ` Wei Wang
2017-12-16 10:12       ` Wei Wang
2017-12-16 11:28       ` Tetsuo Handa
2017-12-16 11:28         ` [Qemu-devel] " Tetsuo Handa
2017-12-16 11:28         ` Tetsuo Handa
2017-12-17  5:24         ` [virtio-dev] " Wei Wang
2017-12-17  5:24           ` [Qemu-devel] " Wei Wang
2017-12-17  5:24           ` Wei Wang
2017-12-17  5:24           ` Wei Wang
2017-12-17 10:21           ` Tetsuo Handa
2017-12-17 10:21             ` [Qemu-devel] " Tetsuo Handa
2017-12-17 10:21             ` Tetsuo Handa
2017-12-17 11:50             ` Wang, Wei W
2017-12-17 11:50             ` [virtio-dev] " Wang, Wei W
2017-12-17 11:50               ` [Qemu-devel] " Wang, Wei W
2017-12-17 11:50               ` Wang, Wei W
2017-12-17 11:50               ` Wang, Wei W
2017-12-17 11:50               ` Wang, Wei W
2017-12-17 15:16               ` Tetsuo Handa
2017-12-17 15:16                 ` [Qemu-devel] " Tetsuo Handa
2017-12-17 15:16                 ` Tetsuo Handa
2017-12-18  8:05                 ` [virtio-dev] " Wei Wang
2017-12-18  8:05                   ` [Qemu-devel] " Wei Wang
2017-12-18  8:05                   ` Wei Wang
2017-12-18  8:05                   ` Wei Wang
2017-12-18  8:05                 ` Wei Wang
2017-12-17  5:24         ` Wei Wang
2017-12-15 18:42   ` Matthew Wilcox
2017-12-12 11:55 ` Wei Wang
2017-12-12 11:55 ` [PATCH v19 4/7] virtio-balloon: VIRTIO_BALLOON_F_SG Wei Wang
2017-12-12 11:55 ` [virtio-dev] " Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55 ` [virtio-dev] [PATCH v19 5/7] mm: support reporting free page blocks Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55 ` Wei Wang
2017-12-12 11:55 ` [PATCH v19 6/7] virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_VQ Wei Wang
2017-12-12 11:55 ` [virtio-dev] " Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55 ` [virtio-dev] [PATCH v19 7/7] virtio-balloon: don't report free pages when page poisoning is enabled Wei Wang
2017-12-12 11:55   ` [Qemu-devel] " Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55   ` Wei Wang
2017-12-12 11:55 ` Wei Wang

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=20171215202238-mutt-send-email-mst@kernel.org \
    --to=mst@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=amit.shah@redhat.com \
    --cc=cornelia.huck@de.ibm.com \
    --cc=david@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=liliang.opensource@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mawilcox@microsoft.com \
    --cc=mgorman@techsingularity.net \
    --cc=mhocko@kernel.org \
    --cc=nilal@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=penguin-kernel@I-love.SAKURA.ne.jp \
    --cc=qemu-devel@nongnu.org \
    --cc=quan.xu@aliyun.com \
    --cc=riel@redhat.com \
    --cc=virtio-dev@lists.oasis-open.org \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=wei.w.wang@intel.com \
    --cc=willy@infradead.org \
    --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 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.