From: Peter Xu <peterx@redhat.com>
To: Fabiano Rosas <farosas@suse.de>
Cc: qemu-devel@nongnu.org,
"Maciej S . Szmigiero" <mail@maciej.szmigiero.name>,
"Cédric Le Goater" <clg@redhat.com>,
"Avihai Horon" <avihaih@nvidia.com>,
"Alex Williamson" <alex.williamson@redhat.com>,
"Prasad Pandit" <ppandit@redhat.com>
Subject: Re: [PATCH v2 2/7] migration/multifd: Allow to sync with sender threads only
Date: Fri, 6 Dec 2024 09:50:33 -0500 [thread overview]
Message-ID: <Z1MPOQ6Tg_FVSiF9@x1n> (raw)
In-Reply-To: <877c8dc58x.fsf@suse.de>
On Fri, Dec 06, 2024 at 10:26:06AM -0300, Fabiano Rosas wrote:
> Peter Xu <peterx@redhat.com> writes:
>
> > Teach multifd_send_sync_main() to sync with threads only.
> >
> > We already have such requests, which is when mapped-ram is enabled with
> > multifd. In that case, no SYNC messages will be pushed to the stream when
> > multifd syncs the sender threads because there's no destination threads
> > waiting for that. The whole point of the sync is to make sure all threads
> > flushed their jobs.
>
> s/flushed/finished/ otherwise we risk confusing people.
done.
>
> >
> > So fundamentally we have a request to do the sync in different ways:
> >
> > - Either to sync the threads only,
> > - Or to sync the threads but also with the destination side.
> >
> > Mapped-ram did it already because of the use_packet check in the sync
> > handler of the sender thread. It works.
> >
> > However it may stop working when e.g. VFIO may start to reuse multifd
> > channels to push device states. In that case VFIO has similar request on
> > "thread-only sync" however we can't check a flag because such sync request
> > can still come from RAM which needs the on-wire notifications.
> >
> > Paving way for that by allowing the multifd_send_sync_main() to specify
> > what kind of sync the caller needs. We can use it for mapped-ram already.
> >
> > No functional change intended.
> >
> > Signed-off-by: Peter Xu <peterx@redhat.com>
> > ---
> > migration/multifd.h | 19 ++++++++++++++++---
> > migration/multifd-nocomp.c | 7 ++++++-
> > migration/multifd.c | 15 +++++++++------
> > 3 files changed, 31 insertions(+), 10 deletions(-)
> >
> > diff --git a/migration/multifd.h b/migration/multifd.h
> > index 50d58c0c9c..bd337631ec 100644
> > --- a/migration/multifd.h
> > +++ b/migration/multifd.h
> > @@ -19,6 +19,18 @@
> > typedef struct MultiFDRecvData MultiFDRecvData;
> > typedef struct MultiFDSendData MultiFDSendData;
> >
> > +typedef enum {
> > + /* No sync request */
> > + MULTIFD_SYNC_NONE = 0,
> > + /* Sync locally on the sender threads without pushing messages */
> > + MULTIFD_SYNC_LOCAL,
> > + /*
> > + * Sync not only on the sender threads, but also push "SYNC" message to
> > + * the wire (which is for a remote sync).
>
> s/SYNC/MULTIFD_FLAG_SYNC/
>
> Do we need to also mention that this needs to be paired with a
> multifd_recv_sync_main() via the emission of the
> RAM_SAVE_FLAG_MULTIFD_FLUSH flag on the stream?
If we want to mention something, IMO it would be better about what happens
on the src, not dest. It can be too hard to follow if we connect that
directly to the dest behavior.
Does this look good to you?
/*
* Sync not only on the sender threads, but also push MULTIFD_FLAG_SYNC
* message to the wire for each iochannel (which is for a remote sync).
*
* When remote sync is used, need to be paired with a follow up
* RAM_SAVE_FLAG_EOS / RAM_SAVE_FLAG_MULTIFD_FLUSH message on the main
* channel.
*/
>
> > + */
> > + MULTIFD_SYNC_ALL,
> > +} MultiFDSyncReq;
> > +
> > bool multifd_send_setup(void);
> > void multifd_send_shutdown(void);
> > void multifd_send_channel_created(void);
> > @@ -28,7 +40,7 @@ void multifd_recv_shutdown(void);
> > bool multifd_recv_all_channels_created(void);
> > void multifd_recv_new_channel(QIOChannel *ioc, Error **errp);
> > void multifd_recv_sync_main(void);
> > -int multifd_send_sync_main(void);
> > +int multifd_send_sync_main(MultiFDSyncReq req);
> > bool multifd_queue_page(RAMBlock *block, ram_addr_t offset);
> > bool multifd_recv(void);
> > MultiFDRecvData *multifd_get_recv_data(void);
> > @@ -143,7 +155,7 @@ typedef struct {
> > /* multifd flags for each packet */
> > uint32_t flags;
> > /*
> > - * The sender thread has work to do if either of below boolean is set.
> > + * The sender thread has work to do if either of below field is set.
> > *
> > * @pending_job: a job is pending
> > * @pending_sync: a sync request is pending
> > @@ -152,7 +164,8 @@ typedef struct {
> > * cleared by the multifd sender threads.
> > */
> > bool pending_job;
> > - bool pending_sync;
> > + MultiFDSyncReq pending_sync;
> > +
> > MultiFDSendData *data;
> >
> > /* thread local variables. No locking required */
> > diff --git a/migration/multifd-nocomp.c b/migration/multifd-nocomp.c
> > index 55191152f9..219f9e58ef 100644
> > --- a/migration/multifd-nocomp.c
> > +++ b/migration/multifd-nocomp.c
> > @@ -345,6 +345,8 @@ retry:
> >
> > int multifd_ram_flush_and_sync(void)
> > {
> > + MultiFDSyncReq req;
> > +
> > if (!migrate_multifd()) {
> > return 0;
> > }
> > @@ -356,7 +358,10 @@ int multifd_ram_flush_and_sync(void)
> > }
> > }
> >
> > - return multifd_send_sync_main();
> > + /* File migrations only need to sync with threads */
> > + req = migrate_mapped_ram() ? MULTIFD_SYNC_LOCAL : MULTIFD_SYNC_ALL;
> > +
> > + return multifd_send_sync_main(req);
> > }
> >
> > bool multifd_send_prepare_common(MultiFDSendParams *p)
> > diff --git a/migration/multifd.c b/migration/multifd.c
> > index 498e71fd10..2248bd2d46 100644
> > --- a/migration/multifd.c
> > +++ b/migration/multifd.c
> > @@ -523,7 +523,7 @@ static int multifd_zero_copy_flush(QIOChannel *c)
> > return ret;
> > }
> >
> > -int multifd_send_sync_main(void)
> > +int multifd_send_sync_main(MultiFDSyncReq req)
> > {
> > int i;
> > bool flush_zero_copy;
>
> assert(req != MULTIFD_SYNC_NONE) ?
Sure.
>
> > @@ -543,8 +543,8 @@ int multifd_send_sync_main(void)
> > * We should be the only user so far, so not possible to be set by
> > * others concurrently.
> > */
> > - assert(qatomic_read(&p->pending_sync) == false);
> > - qatomic_set(&p->pending_sync, true);
> > + assert(qatomic_read(&p->pending_sync) == MULTIFD_SYNC_NONE);
> > + qatomic_set(&p->pending_sync, req);
> > qemu_sem_post(&p->sem);
> > }
> > for (i = 0; i < migrate_multifd_channels(); i++) {
> > @@ -635,14 +635,17 @@ static void *multifd_send_thread(void *opaque)
> > */
> > qatomic_store_release(&p->pending_job, false);
> > } else {
> > + MultiFDSyncReq req = qatomic_read(&p->pending_sync);
> > +
> > /*
> > * If not a normal job, must be a sync request. Note that
> > * pending_sync is a standalone flag (unlike pending_job), so
> > * it doesn't require explicit memory barriers.
> > */
> > - assert(qatomic_read(&p->pending_sync));
> > + assert(req != MULTIFD_SYNC_NONE);
> >
> > - if (use_packets) {
> > + /* Only push the SYNC message if it involves a remote sync */
> > + if (req == MULTIFD_SYNC_ALL) {
> > p->flags = MULTIFD_FLAG_SYNC;
> > multifd_send_fill_packet(p);
> > ret = qio_channel_write_all(p->c, (void *)p->packet,
> > @@ -654,7 +657,7 @@ static void *multifd_send_thread(void *opaque)
> > stat64_add(&mig_stats.multifd_bytes, p->packet_len);
> > }
> >
> > - qatomic_set(&p->pending_sync, false);
> > + qatomic_set(&p->pending_sync, MULTIFD_SYNC_NONE);
> > qemu_sem_post(&p->sem_sync);
> > }
> > }
>
--
Peter Xu
next prev parent reply other threads:[~2024-12-06 14:51 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-06 0:58 [PATCH v2 0/7] migration/multifd: Some VFIO / postcopy preparations on flush Peter Xu
2024-12-06 0:58 ` [PATCH v2 1/7] migration/multifd: Further remove the SYNC on complete Peter Xu
2024-12-06 13:17 ` Fabiano Rosas
2024-12-06 14:40 ` Peter Xu
2024-12-06 0:58 ` [PATCH v2 2/7] migration/multifd: Allow to sync with sender threads only Peter Xu
2024-12-06 13:26 ` Fabiano Rosas
2024-12-06 14:50 ` Peter Xu [this message]
2024-12-06 15:00 ` Fabiano Rosas
2024-12-06 0:58 ` [PATCH v2 3/7] migration/ram: Move RAM_SAVE_FLAG* into ram.h Peter Xu
2024-12-06 13:43 ` Fabiano Rosas
2024-12-06 15:03 ` Peter Xu
2024-12-06 15:10 ` Fabiano Rosas
2024-12-06 15:46 ` Peter Xu
2024-12-06 16:58 ` Fabiano Rosas
2024-12-06 0:58 ` [PATCH v2 4/7] migration/multifd: Unify RAM_SAVE_FLAG_MULTIFD_FLUSH messages Peter Xu
2024-12-06 14:12 ` Fabiano Rosas
2024-12-06 0:58 ` [PATCH v2 5/7] migration/multifd: Remove sync processing on postcopy Peter Xu
2024-12-06 14:19 ` Fabiano Rosas
2024-12-06 0:58 ` [PATCH v2 6/7] migration/multifd: Cleanup src flushes on condition check Peter Xu
2024-12-06 14:18 ` Fabiano Rosas
2024-12-06 15:13 ` Peter Xu
2024-12-06 0:58 ` [PATCH v2 7/7] migration/multifd: Document the reason to sync for save_setup() Peter Xu
2024-12-06 14:40 ` Fabiano Rosas
2024-12-06 15:36 ` Peter Xu
2024-12-06 17:01 ` Fabiano Rosas
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=Z1MPOQ6Tg_FVSiF9@x1n \
--to=peterx@redhat.com \
--cc=alex.williamson@redhat.com \
--cc=avihaih@nvidia.com \
--cc=clg@redhat.com \
--cc=farosas@suse.de \
--cc=mail@maciej.szmigiero.name \
--cc=ppandit@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 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).