From: Anthony Liguori <anthony@codemonkey.ws>
To: quintela@redhat.com
Cc: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 25/36] migration: Our release callback was just free
Date: Mon, 17 Oct 2011 10:20:37 -0500 [thread overview]
Message-ID: <4E9C47C5.7030006@codemonkey.ws> (raw)
In-Reply-To: <m3mxczslc9.fsf@trasno.mitica>
On 10/17/2011 10:12 AM, Juan Quintela wrote:
> Anthony Liguori<anthony@codemonkey.ws> wrote:
>> On 10/11/2011 05:00 AM, Juan Quintela wrote:
>>> We called it from a single place, and always with state !=
>>> MIG_STATE_ACTIVE. Just remove the whole callback. For users of the
>>> notifier, notice that this is exactly the case where they don't care,
>>> we are just freeing the state from previous failed migration (it can't
>>> be a sucessful one, otherwise we would not be running on that machine
>>> in the first place).
>>>
>>> Signed-off-by: Juan Quintela<quintela@redhat.com>
>>> ---
>>> migration.c | 19 +------------------
>>> migration.h | 1 -
>>> 2 files changed, 1 insertions(+), 19 deletions(-)
>>>
>>> diff --git a/migration.c b/migration.c
>>> index a8e936e..689464d 100644
>>> --- a/migration.c
>>> +++ b/migration.c
>>> @@ -123,10 +123,7 @@ int do_migrate(Monitor *mon, const QDict *qdict, QObject **ret_data)
>>> goto free_migrate_state;
>>> }
>>>
>>> - if (current_migration) {
>>> - current_migration->release(current_migration);
>>> - }
>>> -
>>> + g_free(current_migration);
>>
>> migrate_fd_cleanup() is no longer being called.
>
> It was never called.
>
>> Regards,
>>
>> Anthony Liguori
>>
>>> current_migration = s;
>>> notifier_list_notify(&migration_state_notifiers, NULL);
>>> return 0;
>>> @@ -416,19 +413,6 @@ static void migrate_fd_cancel(MigrationState *s)
>>> migrate_fd_cleanup(s);
>>> }
>>>
>>> -static void migrate_fd_release(MigrationState *s)
>>> -{
>>> -
>>> - DPRINTF("releasing state\n");
>>> -
>>> - if (s->state == MIG_STATE_ACTIVE) {
>
> The first thing that we look is that MIG_STATE_ACTIVE, and we call
> migrate_fd_cleanup() on that case.
It was meant to cancel existing migration but there was a later patch that
didn't allow the implicit cancel anymore.
So ignore my comments here.
Regards,
Anthony Liguori
>
>
> See the beggining of do_migrate(). If we are in MIG_STATE_ACTIVE, we
> just don't continue. the function. Notice that you complained about
> comments being bad on the other patches, and in this very case, it was
> explicitely stated on the comment O:-)
>
> int do_migrate(Monitor *mon, const QDict *qdict, QObject **ret_data)
> {
> MigrationState *s = NULL;
> const char *p;
> int detach = qdict_get_try_bool(qdict, "detach", 0);
> int blk = qdict_get_try_bool(qdict, "blk", 0);
> int inc = qdict_get_try_bool(qdict, "inc", 0);
> const char *uri = qdict_get_str(qdict, "uri");
>
> if (current_migration&&
> current_migration->get_status(current_migration) == MIG_STATE_ACTIVE) {
> monitor_printf(mon, "migration already in progress\n");
> return -1;
>
> ************ We stop here ****
> call to ->release() is below this.
> }
>
> if (qemu_savevm_state_blocked(mon)) {
> return -1;
> }
>
>
>
>>> - s->state = MIG_STATE_CANCELLED;
>>> - notifier_list_notify(&migration_state_notifiers, NULL);
>>> - migrate_fd_cleanup(s);
>>> - }
>>> - g_free(s);
>>> -}
>>> -
>>> static void migrate_fd_wait_for_unfreeze(void *opaque)
>>> {
>>> MigrationState *s = opaque;
>>> @@ -511,7 +495,6 @@ static MigrationState *migrate_new(Monitor *mon, int64_t bandwidth_limit,
>>>
>>> s->cancel = migrate_fd_cancel;
>>> s->get_status = migrate_fd_get_status;
>>> - s->release = migrate_fd_release;
>>> s->blk = blk;
>>> s->shared = inc;
>>> s->mon = NULL;
>>> diff --git a/migration.h b/migration.h
>>> index 3165140..1cdb539 100644
>>> --- a/migration.h
>>> +++ b/migration.h
>>> @@ -40,7 +40,6 @@ struct MigrationState
>>> int (*write)(MigrationState *s, const void *buff, size_t size);
>>> void (*cancel)(MigrationState *s);
>>> int (*get_status)(MigrationState *s);
>>> - void (*release)(MigrationState *s);
>>> void *opaque;
>>> int blk;
>>> int shared;
>
next prev parent reply other threads:[~2011-10-17 15:20 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-11 10:00 [Qemu-devel] [PATCH v4 00/36] Migration errors & cleanup (the integrated version) Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 01/36] ds1225y: Use stdio instead of QEMUFile Juan Quintela
2011-10-12 8:47 ` Zhi Hui Li
2011-10-17 13:50 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 02/36] migration: simplify state assignmente Juan Quintela
2011-10-17 13:52 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 03/36] migration: Check that migration is active before cancel it Juan Quintela
2011-10-17 13:53 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 04/36] migration: return real error code Juan Quintela
2011-10-17 13:53 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 05/36] migration: add error handling to migrate_fd_put_notify() Juan Quintela
2011-10-17 13:54 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 06/36] migration: If there is one error, it makes no sense to continue Juan Quintela
2011-10-17 13:56 ` Anthony Liguori
2011-10-17 16:58 ` Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 07/36] buffered_file: Use right "opaque" Juan Quintela
2011-10-17 13:56 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 08/36] buffered_file: reuse QEMUFile has_error field Juan Quintela
2011-10-17 13:57 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 09/36] migration: don't "write" when migration is not active Juan Quintela
2011-10-17 13:59 ` Anthony Liguori
2011-10-17 17:04 ` Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 10/36] migration: set error if select return one error Juan Quintela
2011-10-17 13:59 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 11/36] migration: change has_error to contain errno values Juan Quintela
2011-10-17 14:00 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 12/36] migration: rename qemu_file_has_error to qemu_file_get_error Juan Quintela
2011-10-17 14:00 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 13/36] savevm: Rename has_error to last_error field Juan Quintela
2011-10-17 14:00 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 14/36] migration: use qemu_file_get_error() return value when possible Juan Quintela
2011-10-17 14:01 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 15/36] migration: Make *start_outgoing_migration return FdMigrationState Juan Quintela
2011-10-17 14:01 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 16/36] migration: Use FdMigrationState instead of MigrationState when possible Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 17/36] migration: Fold MigrationState into FdMigrationState Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 18/36] migration: Rename FdMigrationState MigrationState Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 19/36] migration: Refactor MigrationState creation Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 20/36] migration: Make all posible migration functions static Juan Quintela
2011-10-17 14:02 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 21/36] migration: move migrate_new to do_migrate Juan Quintela
2011-10-17 14:03 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 22/36] migration: Introduce MIG_STATE_SETUP Juan Quintela
2011-10-17 14:03 ` Anthony Liguori
2011-10-18 1:29 ` Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 23/36] migration: Refactor and simplify error checking in migrate_fd_put_ready Juan Quintela
2011-10-17 14:05 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 24/36] migration: Introduce migrate_fd_completed() for consistency Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 25/36] migration: Our release callback was just free Juan Quintela
2011-10-17 14:06 ` Anthony Liguori
2011-10-17 15:12 ` Juan Quintela
2011-10-17 15:20 ` Anthony Liguori [this message]
2011-10-17 15:18 ` Anthony Liguori
2011-10-11 10:00 ` [Qemu-devel] [PATCH 26/36] migration: Remove get_status() accessor Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 27/36] migration: Remove migration cancel() callback Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 28/36] migration: Move exported functions to the end of the file Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 29/36] migration: create accessor for current_migration Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 30/36] migration: Use bandwidth_limit directly Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 31/36] migration: Pass MigrationState in migration notifiers Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 32/36] migration: Export a function that tells if the migration has finished correctly Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 33/36] migration: Make state definitions local Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 34/36] migration: Don't use callback on file defining it Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 35/36] migration: propagate error correctly Juan Quintela
2011-10-11 10:00 ` [Qemu-devel] [PATCH 36/36] migration: make migration-{tcp, unix} consistent Juan Quintela
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=4E9C47C5.7030006@codemonkey.ws \
--to=anthony@codemonkey.ws \
--cc=qemu-devel@nongnu.org \
--cc=quintela@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).