From: Max Reitz <mreitz@redhat.com>
To: Fam Zheng <famz@redhat.com>, qemu-devel@nongnu.org
Cc: Kevin Wolf <kwolf@redhat.com>, Stefan Hajnoczi <stefanha@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v6 5/5] qemu-iotests: Add 093 for IO throttling
Date: Wed, 28 Jan 2015 11:22:43 -0500 [thread overview]
Message-ID: <54C90CD3.1050805@redhat.com> (raw)
In-Reply-To: <1422412118-13151-6-git-send-email-famz@redhat.com>
On 2015-01-27 at 21:28, Fam Zheng wrote:
> This case utilizes qemu-io command "aio_{read,write} -q" to verify the
> effectiveness of IO throttling options.
>
> It's implemented by driving the vm timer from qtest protocol, so the
> throttling timers are signaled with determinied time duration. Then we
> verify the completed IO requests are within 10% error of bps and iops
> limits.
>
> "null" protocol is used as the disk backend so that no actual disk IO is
> performed on host, this will make the blockstats much more
> deterministic. Both "null-aio" and "null-co" are covered, which is also
> a simple cross validation test for the driver code.
>
> Signed-off-by: Fam Zheng <famz@redhat.com>
> ---
> tests/qemu-iotests/093 | 120 +++++++++++++++++++++++++++++++++++++++++++++
> tests/qemu-iotests/093.out | 5 ++
> tests/qemu-iotests/group | 1 +
> 3 files changed, 126 insertions(+)
> create mode 100755 tests/qemu-iotests/093
> create mode 100644 tests/qemu-iotests/093.out
>
> diff --git a/tests/qemu-iotests/093 b/tests/qemu-iotests/093
> new file mode 100755
> index 0000000..2866536
> --- /dev/null
> +++ b/tests/qemu-iotests/093
> @@ -0,0 +1,120 @@
> +#!/usr/bin/env python
> +#
> +# Tests for IO throttling
> +#
> +# Copyright (C) 2015 Red Hat, Inc.
> +#
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 2 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program. If not, see <http://www.gnu.org/licenses/>.
> +#
> +
> +import iotests
> +
> +class ThrottleTestCase(iotests.QMPTestCase):
> + test_img = "null-aio://"
> +
> + def blockstats(self, device):
> + result = self.vm.qmp("query-blockstats")
> + for r in result['return']:
> + if r['device'] == device:
> + stat = r['stats']
> + return stat['rd_bytes'], stat['rd_operations'], stat['wr_bytes'], stat['wr_operations']
> + raise Exception("Device not found for blockstats: %s" % device)
> +
> + def setUp(self):
> + self.vm = iotests.VM().add_drive(self.test_img)
> + self.vm.launch()
> +
> + def tearDown(self):
> + self.vm.shutdown()
> +
> + def do_test_throttle(self, seconds, params):
> + def check_limit(limit, num):
> + # IO throttling algorithm is discrete, allow 10% error so the test
> + # is more robust
> + return limit == 0 or \
> + (num < seconds * limit * 1.1
> + and num > seconds * limit * 0.9)
> +
> + nsec_per_sec = 1000000000
> +
> + params['device'] = 'drive0'
> +
> + result = self.vm.qmp("block_set_io_throttle", conv_keys=False, **params)
> + self.assert_qmp(result, 'return', {})
> +
> + # Set vm clock to a known value
> + ns = seconds * nsec_per_sec
> + self.vm.qtest("clock_step %d" % ns)
> +
> + # Submit enough requests. They will drain bps_max and iops_max, but the
> + # rest requests won't get executed until we advance the virtual clock
> + # with qtest interface
> + rq_size = 512
> + rd_nr = max(params['bps'] / rq_size / 2,
> + params['bps_rd'] / rq_size,
> + params['iops'] / 2,
> + params['iops_rd']) + \
> + params['bps_max'] / rq_size / 2 + \
> + params['iops_max']
I guess the divisions by two are because those values represent read and
write operations combined. Shouldn't iops_max be divided by two, too, then?
> + rd_nr *= seconds * 2
> + wr_nr = max(params['bps'] / rq_size / 2,
> + params['bps_wr'] / rq_size,
> + params['iops'] / 2,
> + params['iops_wr']) + \
> + params['bps_max'] / rq_size / 2 + \
> + params['iops_max']
> + wr_nr *= seconds * 2
> + for i in range(rd_nr):
> + self.vm.hmp_qemu_io("drive0", "aio_read %d %d" % (i * rq_size, rq_size))
> + for i in range(wr_nr):
> + self.vm.hmp_qemu_io("drive0", "aio_write %d %d" % (i * rq_size, rq_size))
> +
> + start_rd_bytes, start_rd_iops, start_wr_bytes, start_wr_iops = self.blockstats('drive0')
> +
> + self.vm.qtest("clock_step %d" % ns)
> + end_rd_bytes, end_rd_iops, end_wr_bytes, end_wr_iops = self.blockstats('drive0')
> +
> + rd_bytes = end_rd_bytes - start_rd_bytes
> + rd_iops = end_rd_iops - start_rd_iops
> + wr_bytes = end_wr_bytes - start_wr_bytes
> + wr_iops = end_wr_iops - start_wr_iops
> +
> + self.assertTrue(check_limit(params['bps'], rd_bytes + wr_bytes))
> + self.assertTrue(check_limit(params['bps_rd'], rd_bytes))
> + self.assertTrue(check_limit(params['bps_wr'], wr_bytes))
> + self.assertTrue(check_limit(params['iops'], rd_iops + wr_iops))
> + self.assertTrue(check_limit(params['iops_rd'], rd_iops))
> + self.assertTrue(check_limit(params['iops_wr'], wr_iops))
Hm, you're not checking bps_max and iops_max here. Should you be?
Apart from these two (minor) things: Works for me! :-)
Max
> +
> + def test_all(self):
> + params = {"bps": 4096,
> + "bps_rd": 4096,
> + "bps_wr": 4096,
> + "bps_max": 4096,
> + "iops": 10,
> + "iops_rd": 10,
> + "iops_wr": 10,
> + "iops_max": 10,
> + }
> + # Pick each out of all possible params and test
> + for tk in params:
> + limits = dict([(k, 0) for k in params])
> + limits[tk] = params[tk]
> + self.do_test_throttle(5, limits)
> +
> +class ThrottleTestCoroutine(ThrottleTestCase):
> + test_img = "null-co://"
> +
> +if __name__ == '__main__':
> + iotests.main(supported_fmts=["raw"])
> diff --git a/tests/qemu-iotests/093.out b/tests/qemu-iotests/093.out
> new file mode 100644
> index 0000000..fbc63e6
> --- /dev/null
> +++ b/tests/qemu-iotests/093.out
> @@ -0,0 +1,5 @@
> +..
> +----------------------------------------------------------------------
> +Ran 2 tests
> +
> +OK
> diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group
> index f8bf354..0272c9a 100644
> --- a/tests/qemu-iotests/group
> +++ b/tests/qemu-iotests/group
> @@ -99,6 +99,7 @@
> 090 rw auto quick
> 091 rw auto
> 092 rw auto quick
> +093 auto
> 095 rw auto quick
> 097 rw auto backing
> 098 rw auto backing quick
next prev parent reply other threads:[~2015-01-28 16:22 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-28 2:28 [Qemu-devel] [PATCH v6 0/5] block: Add a qemu-iotests case for IO throttling Fam Zheng
2015-01-28 2:28 ` [Qemu-devel] [PATCH v6 1/5] qemu-io: Account IO by aio_read and aio_write Fam Zheng
2015-01-28 2:28 ` [Qemu-devel] [PATCH v6 2/5] qtest: Add scripts/qtest.py Fam Zheng
2015-01-28 2:28 ` [Qemu-devel] [PATCH v6 3/5] qemu-iotests: Add VM method qtest() to iotests.py Fam Zheng
2015-01-28 2:28 ` [Qemu-devel] [PATCH v6 4/5] qemu-iotests: Allow caller to disable underscore convertion for qmp Fam Zheng
2015-01-28 2:28 ` [Qemu-devel] [PATCH v6 5/5] qemu-iotests: Add 093 for IO throttling Fam Zheng
2015-01-28 16:22 ` Max Reitz [this message]
2015-01-29 0:53 ` Fam Zheng
2015-01-29 2:06 ` Fam Zheng
2015-01-29 14:29 ` Max Reitz
2015-01-30 2:47 ` Fam Zheng
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=54C90CD3.1050805@redhat.com \
--to=mreitz@redhat.com \
--cc=famz@redhat.com \
--cc=kwolf@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.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).