qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: "Daniel P. Berrangé" <berrange@redhat.com>
Cc: qemu-devel@nongnu.org, Paolo Bonzini <pbonzini@redhat.com>,
	Laurent Vivier <lvivier@redhat.com>,
	Juan Quintela <quintela@redhat.com>,
	Thomas Huth <thuth@redhat.com>, Cornelia Huck <cohuck@redhat.com>,
	qemu-s390x@nongnu.org
Subject: Re: [PATCH 4/5] tests: use consistent bandwidth/downtime limits in migration tests
Date: Tue, 28 Jun 2022 15:16:38 +0100	[thread overview]
Message-ID: <YrsNRtFh9LVwlHuq@work-vm> (raw)
In-Reply-To: <20220628105434.295905-5-berrange@redhat.com>

* Daniel P. Berrangé (berrange@redhat.com) wrote:
> The different migration test cases are using a variety of settings
> to ensure convergance/non-convergance. Introduce two helpers to
> extra the common functionality and ensure consistency.
> 
>   * Non-convergance: 1ms downtime, 30mbs bandwidth
>   * Convergance: 30s downtime, 1gbs bandwidth
> 
> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

> ---
>  tests/qtest/migration-test.c | 54 +++++++++++++-----------------------
>  1 file changed, 20 insertions(+), 34 deletions(-)
> 
> diff --git a/tests/qtest/migration-test.c b/tests/qtest/migration-test.c
> index a54eff6d56..9e64125f02 100644
> --- a/tests/qtest/migration-test.c
> +++ b/tests/qtest/migration-test.c
> @@ -46,9 +46,6 @@ unsigned start_address;
>  unsigned end_address;
>  static bool uffd_feature_thread_id;
>  
> -/* A downtime where the test really should converge */
> -#define CONVERGE_DOWNTIME (1000 * 30)
> -
>  #if defined(__linux__)
>  #include <sys/syscall.h>
>  #include <sys/vfs.h>
> @@ -402,6 +399,20 @@ static void migrate_set_parameter_str(QTestState *who, const char *parameter,
>      migrate_check_parameter_str(who, parameter, value);
>  }
>  
> +static void migrate_ensure_non_converge(QTestState *who)
> +{
> +    /* Can't converge with 1ms downtime + 30 mbs bandwidth limit */
> +    migrate_set_parameter_int(who, "max-bandwidth", 30 * 1000 * 1000);
> +    migrate_set_parameter_int(who, "downtime-limit", 1);
> +}
> +
> +static void migrate_ensure_converge(QTestState *who)
> +{
> +    /* Should converge with 30s downtime + 1 gbs bandwidth limit */
> +    migrate_set_parameter_int(who, "max-bandwidth", 1 * 1000 * 1000 * 1000);
> +    migrate_set_parameter_int(who, "downtime-limit", 30 * 1000);
> +}
> +
>  static void migrate_pause(QTestState *who)
>  {
>      QDict *rsp;
> @@ -984,12 +995,7 @@ static int migrate_postcopy_prepare(QTestState **from_ptr,
>      migrate_set_capability(to, "postcopy-ram", true);
>      migrate_set_capability(to, "postcopy-blocktime", true);
>  
> -    /* 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.
> -     */
> -    migrate_set_parameter_int(from, "max-bandwidth", 30000000);
> -    migrate_set_parameter_int(from, "downtime-limit", 1);
> +    migrate_ensure_non_converge(from);
>  
>      /* Wait for the first serial output from the source */
>      wait_for_serial("src_serial");
> @@ -1188,15 +1194,7 @@ static void test_precopy_common(MigrateCommon *args)
>          return;
>      }
>  
> -    /*
> -     * 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.
> -     */
> -    /* 1 ms should make it not converge*/
> -    migrate_set_parameter_int(from, "downtime-limit", 1);
> -    /* 1GB/s */
> -    migrate_set_parameter_int(from, "max-bandwidth", 1000000000);
> +    migrate_ensure_non_converge(from);
>  
>      if (args->start_hook) {
>          data_hook = args->start_hook(from, to);
> @@ -1230,7 +1228,7 @@ static void test_precopy_common(MigrateCommon *args)
>              wait_for_migration_pass(from);
>          }
>  
> -        migrate_set_parameter_int(from, "downtime-limit", CONVERGE_DOWNTIME);
> +        migrate_ensure_converge(from);
>  
>          /* We do this first, as it has a timeout to stop us
>           * hanging forever if migration didn't converge */
> @@ -1694,8 +1692,7 @@ static void test_migrate_auto_converge(void)
>       * Set the initial parameters so that the migration could not converge
>       * without throttling.
>       */
> -    migrate_set_parameter_int(from, "downtime-limit", 1);
> -    migrate_set_parameter_int(from, "max-bandwidth", 100000000); /* ~100Mb/s */
> +    migrate_ensure_non_converge(from);
>  
>      /* To check remaining size after precopy */
>      migrate_set_capability(from, "pause-before-switchover", true);
> @@ -2000,15 +1997,7 @@ static void test_multifd_tcp_cancel(void)
>          return;
>      }
>  
> -    /*
> -     * 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.
> -     */
> -    /* 1 ms should make it not converge*/
> -    migrate_set_parameter_int(from, "downtime-limit", 1);
> -    /* 300MB/s */
> -    migrate_set_parameter_int(from, "max-bandwidth", 30000000);
> +    migrate_ensure_non_converge(from);
>  
>      migrate_set_parameter_int(from, "multifd-channels", 16);
>      migrate_set_parameter_int(to, "multifd-channels", 16);
> @@ -2054,10 +2043,7 @@ static void test_multifd_tcp_cancel(void)
>  
>      wait_for_migration_status(from, "cancelled", NULL);
>  
> -    /* 300ms it should converge */
> -    migrate_set_parameter_int(from, "downtime-limit", 300);
> -    /* 1GB/s */
> -    migrate_set_parameter_int(from, "max-bandwidth", 1000000000);
> +    migrate_ensure_converge(from);
>  
>      migrate_qmp(from, uri, "{}");
>  
> -- 
> 2.36.1
> 
-- 
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK



  reply	other threads:[~2022-06-28 14:17 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-28 10:54 [PATCH 0/5] tests: improve reliability of migration test Daniel P. Berrangé
2022-06-28 10:54 ` [PATCH 1/5] tests: wait max 120 seconds for migration test status changes Daniel P. Berrangé
2022-06-28 12:47   ` Laurent Vivier
2022-06-28 12:49   ` Thomas Huth
2022-06-28 10:54 ` [PATCH 2/5] tests: wait for migration completion before looking for STOP event Daniel P. Berrangé
2022-06-28 12:47   ` Laurent Vivier
2022-06-28 14:08   ` Dr. David Alan Gilbert
2022-06-28 14:10     ` Daniel P. Berrangé
2022-06-28 10:54 ` [PATCH 3/5] tests: increase migration test converge downtime to 30 seconds Daniel P. Berrangé
2022-06-28 12:47   ` Laurent Vivier
2022-06-28 10:54 ` [PATCH 4/5] tests: use consistent bandwidth/downtime limits in migration tests Daniel P. Berrangé
2022-06-28 14:16   ` Dr. David Alan Gilbert [this message]
2022-06-28 10:54 ` [RFC PATCH 5/5] tests: stop skipping migration test on s390x/ppc64 Daniel P. Berrangé
2022-06-28 13:18   ` Thomas Huth
2022-07-05  8:06   ` Thomas Huth
2022-07-05  8:09     ` Daniel P. Berrangé
2022-07-05  8:38       ` Dr. David Alan Gilbert
2022-06-28 13:19 ` [PATCH 0/5] tests: improve reliability of migration test Thomas Huth

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=YrsNRtFh9LVwlHuq@work-vm \
    --to=dgilbert@redhat.com \
    --cc=berrange@redhat.com \
    --cc=cohuck@redhat.com \
    --cc=lvivier@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-s390x@nongnu.org \
    --cc=quintela@redhat.com \
    --cc=thuth@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).