From: Eric Blake <eblake@redhat.com>
To: "Denis V. Lunev" <den@openvz.org>,
qemu-devel@nongnu.org, qemu-block@nongnu.org
Cc: vsementsov@virtuozzo.com, Stefan Hajnoczi <stefanha@redhat.com>,
Fam Zheng <famz@redhat.com>, Kevin Wolf <kwolf@redhat.com>,
Max Reitz <mreitz@redhat.com>, Jeff Cody <jcody@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 3/9] mirror: optimize dirty bitmap filling in mirror_run a bit
Date: Wed, 15 Jun 2016 06:25:25 -0600 [thread overview]
Message-ID: <57614935.2030207@redhat.com> (raw)
In-Reply-To: <576114A7.4000403@openvz.org>
[-- Attachment #1: Type: text/plain, Size: 2071 bytes --]
On 06/15/2016 02:41 AM, Denis V. Lunev wrote:
> On 06/15/2016 05:36 AM, Eric Blake wrote:
>> On 06/14/2016 09:25 AM, Denis V. Lunev wrote:
>>> There is no need to scan allocation tables if we have mark_all_dirty
>>> flag
>>> set. Just mark it all dirty.
>>>
>>> int ret, n;
>>> end = s->bdev_length / BDRV_SECTOR_SIZE;
>>> + if (base == NULL && !bdrv_has_zero_init(target_bs)) {
>>> + bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, end);
>> Won't work as written. 'end' is 64 bits, but bdrv_set_dirty_bitmap() is
>> limited to a 32-bit sector count. Might be first worth updating
>> bdrv_set_dirty_bitmap() and friends to be byte-based rather than
>> sector-based (but still tracking a sector per bit, obviously), as well
>> as expand it to operate on 64-bit sizes rather than 32-bit.
> very nice catch! thank you
>
[meta-comment - your reply style is a bit hard to read. It's best to
include a blank line both before and after any text you write when
replying in context, as the extra spacing calls visual attention to your
reply rather than being part of a wall of text]
>> I'm also worried slightly that the existing code repeated things in a
>> loop, and therefore had pause points every iteration and could thus
>> remain responsive to an early cancel. But doing the entire operation in
>> one chunk (assuming you fix bitmap code to handle a 64-bit size) may end
>> up running for so long without interruption that you lose the benefits
>> of an early interruption that you have by virtue of a 32-bit limit.
>>
> I do not think that this should be worried actually. We just perform memset
> inside for not that big area (1 Tb disk will have 2 Mb dirty area
> bitmap) under
> default parameters.
>
Okay, so the real slowdown in the loop was the rest of the code
(checking backing status in bdrv_is_allocated_above() and not in
actually writing to the bitmap (bdrv_set_dirty_bitmap()).
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]
next prev parent reply other threads:[~2016-06-15 12:25 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-14 15:25 [Qemu-devel] [PATCH 0/9] major rework of drive-mirror Denis V. Lunev
2016-06-14 15:25 ` [Qemu-devel] [PATCH 1/9] mirror: fix calling of blk_aio_pwritev/blk_aio_preadv Denis V. Lunev
2016-06-14 22:48 ` Eric Blake
2016-06-14 15:25 ` [Qemu-devel] [PATCH 2/9] mirror: create mirror_dirty_init helper for mirror_run Denis V. Lunev
2016-06-15 2:29 ` Eric Blake
2016-06-14 15:25 ` [Qemu-devel] [PATCH 3/9] mirror: optimize dirty bitmap filling in mirror_run a bit Denis V. Lunev
2016-06-15 2:36 ` Eric Blake
2016-06-15 8:41 ` Denis V. Lunev
2016-06-15 12:25 ` Eric Blake [this message]
2016-06-14 15:25 ` [Qemu-devel] [PATCH 4/9] mirror: efficiently zero out target Denis V. Lunev
2016-06-15 3:00 ` Eric Blake
2016-06-15 8:46 ` Denis V. Lunev
2016-06-15 12:34 ` Eric Blake
2016-06-15 13:18 ` Denis V. Lunev
2016-07-06 14:33 ` Denis V. Lunev
2016-06-14 15:25 ` [Qemu-devel] [PATCH 5/9] mirror: improve performance of mirroring of empty disk Denis V. Lunev
2016-06-15 3:20 ` Eric Blake
2016-06-15 9:19 ` Stefan Hajnoczi
2016-06-15 10:37 ` Denis V. Lunev
2016-06-16 10:10 ` Stefan Hajnoczi
2016-06-17 2:53 ` Eric Blake
2016-06-17 13:56 ` Stefan Hajnoczi
2016-06-14 15:25 ` [Qemu-devel] [PATCH 6/9] block: pass qiov into before_write notifier Denis V. Lunev
2016-06-15 4:07 ` Eric Blake
2016-06-15 9:21 ` Stefan Hajnoczi
2016-06-15 9:24 ` Denis V. Lunev
2016-06-15 9:22 ` Stefan Hajnoczi
2016-06-14 15:25 ` [Qemu-devel] [PATCH 7/9] mirror: allow to save buffer for QEMUIOVector in MirrorOp Denis V. Lunev
2016-06-15 4:11 ` Eric Blake
2016-06-14 15:25 ` [Qemu-devel] [PATCH 8/9] mirror: use synch scheme for drive mirror Denis V. Lunev
2016-06-15 4:18 ` Eric Blake
2016-06-15 8:52 ` Denis V. Lunev
2016-06-15 9:48 ` Stefan Hajnoczi
2016-06-14 15:25 ` [Qemu-devel] [PATCH 9/9] mirror: replace bdrv_dirty_bitmap with plain hbitmap Denis V. Lunev
2016-06-15 9:06 ` [Qemu-devel] [PATCH 0/9] major rework of drive-mirror Kevin Wolf
2016-06-15 9:34 ` Denis V. Lunev
2016-06-15 10:25 ` Kevin Wolf
2016-06-15 10:44 ` Denis V. Lunev
2016-06-15 9:50 ` Stefan Hajnoczi
2016-06-15 11:09 ` Dr. David Alan Gilbert
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=57614935.2030207@redhat.com \
--to=eblake@redhat.com \
--cc=den@openvz.org \
--cc=famz@redhat.com \
--cc=jcody@redhat.com \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=vsementsov@virtuozzo.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).