All of lore.kernel.org
 help / color / mirror / Atom feed
From: Juan Quintela <quintela@redhat.com>
To: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: qemu-devel@nongnu.org, lvivier@redhat.com, peterx@redhat.com
Subject: Re: [Qemu-devel] [PATCH 4/6] tests: Add migration xbzrle test
Date: Wed, 18 Oct 2017 13:56:31 +0200	[thread overview]
Message-ID: <8760bcmzyo.fsf@secure.laptop> (raw)
In-Reply-To: <20171016160735.GF2252@work-vm> (David Alan Gilbert's message of "Mon, 16 Oct 2017 17:07:35 +0100")

"Dr. David Alan Gilbert" <dgilbert@redhat.com> wrote:
> * Juan Quintela (quintela@redhat.com) wrote:
>> Signed-off-by: Juan Quintela <quintela@redhat.com>

> +static void test_xbzrle(const char *uri)
>> +{
>> +    QTestState *from, *to;
>> +
>> +    test_migrate_start(&from, &to, uri);
>> +
>> +    /* We want to pick a speed slow enough that the test completes
>> +     * quickly, but that it doesn't complete precopy even on a slow
>> +     * machine, so also set the downtime.
>> +     */
>> +    /* 100 ms */
>> +    migrate_set_parameter(from, "downtime-limit", "100");
>> +    /* 1MB/s slow*/
>> +    migrate_set_parameter(from, "max-bandwidth", "100000000");
>> +
>> +    migrate_set_cache_size(from, "50000000");
>
> Why 50MB?  I guess this is an interesting test; the program
> dirties 100MB of RAM repeatedly, one byte/page.  So 50MB is probably
> not much use becuase it'll thrash.

No good reason.
What I want (tm):
- if you have fast hardware, test is as fast as it can
- if you don't have fast hardware (or very overloaded), it don't fail

My current idea is:
- using 1GB/s for speed
- using downtime = 1ms when I want to wait for migration not converge
- using downtime = 300ms when I want it to finish

I am open to other numbers.

>> +    migrate_set_capability(from, "xbzrle", "true");
>> +    migrate_set_capability(to, "xbzrle", "true");
>> +    /* Wait for the first serial output from the source */
>> +    wait_for_serial("src_serial");
>> +
>> +    migrate(from, uri);
>> +
>> +    wait_for_migration_pass(from);
>> +
>> +    /* 1GB/s now it should converge */
>> +    migrate_set_parameter(from, "max-bandwidth", "1000000000");
>
> I wonder about doing a set_cache_size 150MB here, with a delay before
> the max-bandwidht?  If we're lucky at 150MB it'll start actually using
> the XBZRLE compression.
>
>> +    if (!got_stop) {
>> +        qtest_qmp_eventwait(from, "STOP");
>> +    }
>> +    qtest_qmp_eventwait(to, "RESUME");
>> +
>> +    wait_for_serial("dest_serial");
>> +    wait_for_migration_complete(from);
>> +
>> +    test_migrate_end(from, to);
>
> Extract xbzrle stats?

Thinking about that.

I actually would want to test that when I set something, read it back
that it is what I setted.  I gues I would use a similar trick to what
you do to get the pass number.

Later, Juan.

  reply	other threads:[~2017-10-18 11:56 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-04 10:39 [Qemu-devel] [PATCH 0/6] Add make check tests for Migration Juan Quintela
2017-10-04 10:39 ` [Qemu-devel] [PATCH 1/6] tests: Add basic migration precopy test Juan Quintela
2017-10-04 12:33   ` Thomas Huth
2017-10-04 12:46     ` Juan Quintela
2017-10-04 13:07       ` Thomas Huth
2017-10-04 15:52   ` Eric Blake
2017-10-09  8:19   ` Peter Xu
2017-10-18 11:49     ` Juan Quintela
2017-10-16 11:59   ` Dr. David Alan Gilbert
2017-10-18  4:26   ` Peter Xu
2017-10-18 11:36     ` Juan Quintela
2017-10-18 12:10       ` Daniel P. Berrange
2017-10-19  3:18         ` Peter Xu
2017-10-19  9:09           ` Daniel P. Berrange
2017-10-19  9:16             ` Dr. David Alan Gilbert
2017-10-19  9:19               ` Daniel P. Berrange
2017-10-19  9:23                 ` Dr. David Alan Gilbert
2017-10-19  9:09           ` Dr. David Alan Gilbert
2017-10-18 12:03   ` Daniel P. Berrange
2017-10-18 12:49     ` Dr. David Alan Gilbert
2017-10-18 13:08       ` Daniel P. Berrange
2017-10-04 10:39 ` [Qemu-devel] [PATCH 2/6] tests: Add basic migration precopy tcp test Juan Quintela
2017-10-16 12:09   ` Dr. David Alan Gilbert
2017-10-18 11:51     ` Juan Quintela
2017-10-04 10:39 ` [Qemu-devel] [PATCH 3/6] tests: Add precopy test using deprecated commands Juan Quintela
2017-10-16 15:40   ` Dr. David Alan Gilbert
2017-10-18 11:52     ` Juan Quintela
2017-10-04 10:39 ` [Qemu-devel] [PATCH 4/6] tests: Add migration xbzrle test Juan Quintela
2017-10-16 16:07   ` Dr. David Alan Gilbert
2017-10-18 11:56     ` Juan Quintela [this message]
2017-10-19  9:06       ` Dr. David Alan Gilbert
2017-10-04 10:39 ` [Qemu-devel] [PATCH 5/6] tests: Add migration compress threads tests Juan Quintela
2017-10-09  8:28   ` Peter Xu
2017-10-18 11:59     ` Juan Quintela
2017-10-16 16:29   ` Dr. David Alan Gilbert
2017-10-18 11:57     ` Juan Quintela
2017-10-04 10:39 ` [Qemu-devel] [PATCH 6/6] tests: Move postcopy migration test to migrate-set-parameters Juan Quintela
2017-10-16 16:31   ` Dr. David Alan Gilbert
2017-10-04 13:25 ` [Qemu-devel] [PATCH 0/6] Add make check tests for Migration no-reply

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=8760bcmzyo.fsf@secure.laptop \
    --to=quintela@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=lvivier@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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.