[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 4/4] qemu-iotests: Add 080 for IO throttling
From: |
Stefan Hajnoczi |
Subject: |
Re: [Qemu-devel] [PATCH 4/4] qemu-iotests: Add 080 for IO throttling |
Date: |
Wed, 29 Jan 2014 15:45:08 +0100 |
User-agent: |
Mutt/1.5.21 (2010-09-15) |
On Wed, Jan 29, 2014 at 04:40:43PM +0800, Fam Zheng wrote:
> + def do_test_throttle(self, seconds=10, **limits):
> + def check_limit(limit, num):
> + # IO throttling algorithm is discrete, allow 10% error so the
> test
> + # is more deterministic
> + return limit == 0 or num < seconds * limit * 1.1
> +
> + nsec_per_sec = 1000000000
> +
> + limits['bps_max'] = 1
> + limits['iops_max'] = 1
> +
> + # Enqueue many requests to throttling queue
> + result = self.vm.qmp("block_set_io_throttle", conv_keys=False,
> **limits)
> + self.assert_qmp(result, 'return', {})
> +
> + # Set vm clock to a known value
> + ns = nsec_per_sec
> + self.vm.qtest_cmd("clock_step %d" % ns)
> +
> + # Append many requests into the throttle queue
> + # They drain bps_max and iops_max
> + # The rest requests won't get executed until qtest clock is driven
> + for i in range(1000):
> + self.vm.hmp_qemu_io("drive0", "aio_read -a -q 0 512")
> + self.vm.hmp_qemu_io("drive0", "aio_write -a -q 0 512")
> +
> + start_rd_bytes, start_rd_iops, start_wr_bytes, start_wr_iops =
> self.blockstats('drive0')
> +
> + ns += seconds * nsec_per_sec
> + self.vm.qtest_cmd("clock_step %d" % ns)
> + # wait for a while to let requests take off
> + time.sleep(1)
> + 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
> +
> + assert check_limit(limits['bps'], rd_bytes)
> + assert check_limit(limits['bps_rd'], rd_bytes)
> + assert check_limit(limits['bps'], wr_bytes)
> + assert check_limit(limits['bps_wr'], wr_bytes)
> + assert check_limit(limits['iops'], rd_iops)
> + assert check_limit(limits['iops_rd'], rd_iops)
> + assert check_limit(limits['iops'], wr_iops)
> + assert check_limit(limits['iops_wr'], wr_iops)
> +
> + def test_bps(self):
> + self.do_test_throttle(**{
> + 'device': 'drive0',
> + 'bps': 1000,
> + 'bps_rd': 0,
> + 'bps_wr': 0,
> + 'iops': 0,
> + 'iops_rd': 0,
> + 'iops_wr': 0,
> + })
> +
> + def test_bps_rd(self):
> + self.do_test_throttle(**{
> + 'device': 'drive0',
> + 'bps': 0,
> + 'bps_rd': 1000,
> + 'bps_wr': 0,
> + 'iops': 0,
> + 'iops_rd': 0,
> + 'iops_wr': 0,
> + })
I guess you'll get more stable behavior with 1024 instead of 1000. The
request themselves are 512 bytes.
As a side note, I think we'll wait for all 1000 pending requests each
time QEMU shuts down. Probably not a huge time waste but even 100
requests exceed the I/O limits you've specified. Perhaps using just 100
will make the test slightly faster.
- [Qemu-devel] [PATCH 1/4] qemu-io: New option "-a" to aio_read and aio_write, (continued)