qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Juan Quintela <quintela@redhat.com>
Cc: qemu-devel@nongnu.org, lvivier@redhat.com, peterx@redhat.com
Subject: Re: [Qemu-devel] [PATCH v12 12/21] migration: Add multifd traces for start/end thread
Date: Wed, 2 May 2018 19:35:34 +0100	[thread overview]
Message-ID: <20180502183533.GN2679@work-vm> (raw)
In-Reply-To: <20180425112723.1111-13-quintela@redhat.com>

* Juan Quintela (quintela@redhat.com) wrote:
> We want to know how many pages/packets each channel has sent.  Add
> counters for those.
> 
> Signed-off-by: Juan Quintela <quintela@redhat.com>
> ---
>  migration/ram.c        | 20 ++++++++++++++++++++
>  migration/trace-events |  4 ++++
>  2 files changed, 24 insertions(+)
> 
> diff --git a/migration/ram.c b/migration/ram.c
> index 804c83ed89..0f1340b4e3 100644
> --- a/migration/ram.c
> +++ b/migration/ram.c
> @@ -465,6 +465,10 @@ typedef struct {
>      uint32_t flags;
>      /* global number of generated multifd packets */
>      uint32_t seq;
> +    /* packets sent through this channel */
> +    uint32_t num_packets;
> +    /* pages sent through this channel */
> +    uint32_t num_pages;
>  }  MultiFDSendParams;
>  
>  typedef struct {
> @@ -495,6 +499,10 @@ typedef struct {
>      uint32_t flags;
>      /* global number of generated multifd packets */
>      uint32_t seq;
> +    /* packets sent through this channel */
> +    uint32_t num_packets;
> +    /* pages sent through this channel */
> +    uint32_t num_pages;

Doesn't the comment for 'mutex' say 'protects the following parameters'
(added in 09/21) ?  In which case....

>  } MultiFDRecvParams;
>  
>  static int multifd_send_initial_packet(MultiFDSendParams *p, Error **errp)
> @@ -737,9 +745,13 @@ static void *multifd_send_thread(void *opaque)
>      MultiFDSendParams *p = opaque;
>      Error *local_err = NULL;
>  
> +    trace_multifd_send_thread_start(p->id);
> +
>      if (multifd_send_initial_packet(p, &local_err) < 0) {
>          goto out;
>      }
> +    /* initial packet */
> +    p->num_packets = 1;

That's writing to num_packets before the mutex is taken.

>  
>      while (true) {
>          qemu_mutex_lock(&p->mutex);
> @@ -761,6 +773,8 @@ out:
>      p->running = false;
>      qemu_mutex_unlock(&p->mutex);
>  
> +    trace_multifd_send_thread_end(p->id, p->num_packets, p->num_pages);
> +

and should this happen before they're released (OK, it's only a trace)

>      return NULL;
>  }
>  
> @@ -888,6 +902,8 @@ static void *multifd_recv_thread(void *opaque)
>      Error *local_err = NULL;
>      int ret;
>  
> +    trace_multifd_recv_thread_start(p->id);
> +
>      while (true) {
>          qemu_mutex_lock(&p->mutex);
>          if (false)  {
> @@ -910,6 +926,8 @@ static void *multifd_recv_thread(void *opaque)
>      p->running = false;
>      qemu_mutex_unlock(&p->mutex);
>  
> +    trace_multifd_recv_thread_end(p->id, p->num_packets, p->num_pages);
> +
>      return NULL;
>  }
>  
> @@ -975,6 +993,8 @@ void multifd_recv_new_channel(QIOChannel *ioc)
>      }
>      p->c = ioc;
>      object_ref(OBJECT(ioc));
> +    /* initial packet */
> +    p->num_packets = 1;

OK, the receive side here is different.

>      p->running = true;
>      qemu_thread_create(&p->thread, p->name, multifd_recv_thread, p,
> diff --git a/migration/trace-events b/migration/trace-events
> index a180d7b008..e480eb050e 100644
> --- a/migration/trace-events
> +++ b/migration/trace-events
> @@ -77,6 +77,10 @@ ram_load_postcopy_loop(uint64_t addr, int flags) "@%" PRIx64 " %x"
>  ram_postcopy_send_discard_bitmap(void) ""
>  ram_save_page(const char *rbname, uint64_t offset, void *host) "%s: offset: 0x%" PRIx64 " host: %p"
>  ram_save_queue_pages(const char *rbname, size_t start, size_t len) "%s: start: 0x%zx len: 0x%zx"
> +multifd_send_thread_start(uint8_t id) "%d"
> +multifd_send_thread_end(uint8_t id, uint32_t packets, uint32_t pages) "channel %d packets %d pages %d"
> +multifd_recv_thread_start(uint8_t id) "%d"
> +multifd_recv_thread_end(uint8_t id, uint32_t packets, uint32_t pages) "channel %d packets %d pages %d"

Please put the m's before the r's

Dave

>  # migration/migration.c
>  await_return_path_close_on_source_close(void) ""
> -- 
> 2.17.0
> 
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

  reply	other threads:[~2018-05-02 18:35 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-04-25 11:27 [Qemu-devel] [PATCH v12 00/21] Multifd Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 01/21] migration: Set error state in case of error Juan Quintela
2018-05-02 15:53   ` Dr. David Alan Gilbert
2018-05-09  8:15     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 02/21] migration: Introduce multifd_recv_new_channel() Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 03/21] migration: terminate_* can be called for other threads Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 04/21] migration: Be sure all recv channels are created Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 05/21] migration: Export functions to create send channels Juan Quintela
2018-04-26  7:28   ` Peter Xu
2018-05-09  8:05     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 06/21] migration: Create multifd channels Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 07/21] migration: Delay start of migration main routines Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 08/21] migration: Transmit initial package through the multifd channels Juan Quintela
2018-05-02 17:19   ` Dr. David Alan Gilbert
2018-05-09  8:34     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 09/21] migration: Define MultifdRecvParams sooner Juan Quintela
2018-05-02 17:32   ` Dr. David Alan Gilbert
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 10/21] migration: Create multipage support Juan Quintela
2018-04-26  7:15   ` Peter Xu
2018-05-09 10:52     ` Juan Quintela
2018-05-02 17:52   ` Dr. David Alan Gilbert
2018-05-09 10:53     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 11/21] migration: Create multifd packet Juan Quintela
2018-05-02 18:04   ` Dr. David Alan Gilbert
2018-05-09 11:09     ` Juan Quintela
2018-05-09 11:12       ` Dr. David Alan Gilbert
2018-05-09 19:46         ` Juan Quintela
2018-05-11 16:36           ` Dr. David Alan Gilbert
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 12/21] migration: Add multifd traces for start/end thread Juan Quintela
2018-05-02 18:35   ` Dr. David Alan Gilbert [this message]
2018-05-09 11:11     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 13/21] migration: Calculate transferred ram correctly Juan Quintela
2018-05-02 18:59   ` Dr. David Alan Gilbert
2018-05-09 11:14     ` Juan Quintela
2018-05-09 19:46     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 14/21] migration: Multifd channels always wait on the sem Juan Quintela
2018-05-03  9:36   ` Dr. David Alan Gilbert
2018-05-23 10:59     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 15/21] migration: Add block where to send/receive packets Juan Quintela
2018-05-03 10:03   ` Dr. David Alan Gilbert
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 16/21] migration: Synchronize multifd threads with main thread Juan Quintela
2018-05-03 10:44   ` Dr. David Alan Gilbert
2018-05-09 19:45     ` Juan Quintela
2018-05-11 16:32       ` Dr. David Alan Gilbert
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 17/21] migration: Create ram_multifd_page Juan Quintela
2018-04-26  7:43   ` Peter Xu
2018-04-26  8:18   ` Peter Xu
2018-05-03 11:30     ` Dr. David Alan Gilbert
2018-05-23 11:13     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 18/21] migration: Start sending messages Juan Quintela
2018-05-03 14:55   ` Dr. David Alan Gilbert
2018-05-23 10:51     ` Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 19/21] migration: Wait for blocking IO Juan Quintela
2018-05-03 15:04   ` Dr. David Alan Gilbert
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 20/21] migration: Remove not needed semaphore and quit Juan Quintela
2018-04-25 11:27 ` [Qemu-devel] [PATCH v12 21/21] migration: Stop sending whole pages through main channel Juan Quintela
2018-05-03 15:24   ` Dr. David Alan Gilbert
2018-04-25 11:44 ` [Qemu-devel] [PATCH v12 00/21] Multifd Juan Quintela
2018-05-03 15:32   ` Dr. David Alan Gilbert
2018-04-26  8:28 ` Peter Xu

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=20180502183533.GN2679@work-vm \
    --to=dgilbert@redhat.com \
    --cc=lvivier@redhat.com \
    --cc=peterx@redhat.com \
    --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).