qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: mrhines@linux.vnet.ibm.com
To: qemu-devel@nongnu.org
Cc: aliguori@us.ibm.com, quintela@redhat.com, owasserm@redhat.com,
	abali@us.ibm.com, mrhines@us.ibm.com, gokul@us.ibm.com,
	pbonzini@redhat.com
Subject: [Qemu-devel] [PATCH v5 04/12] rdma: introduce qemu_get_max_size()
Date: Sun, 21 Apr 2013 17:17:53 -0400	[thread overview]
Message-ID: <1366579081-6857-5-git-send-email-mrhines@linux.vnet.ibm.com> (raw)
In-Reply-To: <1366579081-6857-1-git-send-email-mrhines@linux.vnet.ibm.com>

From: "Michael R. Hines" <mrhines@us.ibm.com>

This functions allows you to perform your own per-QEMUFileOps
calculation for the value of 'max_size'.

For RDMA, this calculation artificially limits migration throughput
and needs to be done differently for high-throughput links.

Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>
---
 include/migration/qemu-file.h |   12 ++++++++++++
 migration.c                   |    3 ++-
 savevm.c                      |   19 +++++++++++++++++++
 3 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/include/migration/qemu-file.h b/include/migration/qemu-file.h
index 70eb9bd..1803aeb 100644
--- a/include/migration/qemu-file.h
+++ b/include/migration/qemu-file.h
@@ -57,12 +57,22 @@ typedef int (QEMUFileGetFD)(void *opaque);
 typedef ssize_t (QEMUFileWritevBufferFunc)(void *opaque, struct iovec *iov,
                                            int iovcnt, int64_t pos);
 
+/*
+ * This function allows override of how to calculate max_size
+ * after each iteration.
+ */
+typedef size_t (QEMUFileMaxSizeFunc)(QEMUFile *f, void *opaque,
+                               uint64_t transferred_bytes,
+                               uint64_t time_spent,
+                               uint64_t max_downtime);
+
 typedef struct QEMUFileOps {
     QEMUFilePutBufferFunc *put_buffer;
     QEMUFileGetBufferFunc *get_buffer;
     QEMUFileCloseFunc *close;
     QEMUFileGetFD *get_fd;
     QEMUFileWritevBufferFunc *writev_buffer;
+    QEMUFileMaxSizeFunc *get_max_size;
 } QEMUFileOps;
 
 QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops);
@@ -81,6 +91,8 @@ void qemu_put_byte(QEMUFile *f, int v);
  */
 void qemu_put_buffer_async(QEMUFile *f, const uint8_t *buf, int size);
 double qemu_get_mbps(void);
+size_t qemu_get_max_size(QEMUFile *f, uint64_t transferred_bytes,
+                         uint64_t time_spent, uint64_t max_downtime);
 
 
 static inline void qemu_put_ubyte(QEMUFile *f, unsigned int v)
diff --git a/migration.c b/migration.c
index e2d73c4..48b5174 100644
--- a/migration.c
+++ b/migration.c
@@ -545,7 +545,8 @@ static void *migration_thread(void *opaque)
             uint64_t transferred_bytes = qemu_ftell(s->file) - initial_bytes;
             uint64_t time_spent = current_time - initial_time - sleep_time;
             double bandwidth = transferred_bytes / time_spent;
-            max_size = bandwidth * migrate_max_downtime() / 1000000;
+            max_size = qemu_get_max_size(s->file, transferred_bytes, 
+                    time_spent, migrate_max_downtime());
 
             DPRINTF("transferred %" PRIu64 " time_spent %" PRIu64
                     " bandwidth %g max_size %" PRId64 "\n",
diff --git a/savevm.c b/savevm.c
index 6fcad19..60776e5 100644
--- a/savevm.c
+++ b/savevm.c
@@ -647,6 +647,25 @@ int qemu_get_fd(QEMUFile *f)
     return -1;
 }
 
+size_t qemu_get_max_size(QEMUFile *f, uint64_t transferred_bytes,
+                         uint64_t time_spent, uint64_t max_downtime)
+{
+    if (time_spent) {
+        mbps = (((double) transferred_bytes * 8.0) / 
+            ((double) time_spent / 1000.0)) / 1000.0 / 1000.0;
+    } else {
+        mbps = -1.0;
+    }
+
+    if (f->ops->get_max_size) {
+        return f->ops->get_max_size(f, f->opaque, 
+            transferred_bytes, time_spent, max_downtime);
+    }
+
+    return ((double) (transferred_bytes / time_spent)) * 
+                    max_downtime / 1000000;
+}
+
 /** Closes the file
  *
  * Returns negative error value if any error happened on previous operations or
-- 
1.7.10.4

  parent reply	other threads:[~2013-04-21 21:18 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-21 21:17 [Qemu-devel] [PATCH v5 00/12] rdma: migration support mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 01/12] rdma: add documentation mrhines
2013-04-22 20:16   ` Eric Blake
2013-04-22 20:58     ` Michael R. Hines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 02/12] rdma: export yield_until_fd_readable() mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 03/12] rdma: export throughput w/ MigrationStats QMP mrhines
2013-04-21 21:17 ` mrhines [this message]
2013-04-22 20:20   ` [Qemu-devel] [PATCH v5 04/12] rdma: introduce qemu_get_max_size() Eric Blake
2013-04-22 20:59     ` Michael R. Hines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 05/12] rdma: introduce qemu_file_mode_is_not_valid() mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 06/12] rdma: export qemu_fflush() mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 07/12] rdma: introduce ram_handle_compressed() mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 08/12] rdma: introduce qemu_ram_foreach_block() mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 09/12] rdma: new QEMUFileOps hooks mrhines
2013-04-21 21:17 ` [Qemu-devel] [PATCH v5 10/12] rdma: introduce capability x-rdma-pin-all mrhines
2013-04-22 20:24   ` Eric Blake
2013-04-22 20:59     ` Michael R. Hines
2013-04-21 21:18 ` [Qemu-devel] [PATCH v5 11/12] rdma: core logic mrhines
2013-04-21 21:18 ` [Qemu-devel] [PATCH v5 12/12] rdma: send pc.ram mrhines
  -- strict thread matches above, loose matches on Subject: below --
2013-04-23  1:55 [Qemu-devel] [PATCH v5 00/12] rdma: migration support mrhines
2013-04-23  1:55 ` [Qemu-devel] [PATCH v5 04/12] rdma: introduce qemu_get_max_size() mrhines

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=1366579081-6857-5-git-send-email-mrhines@linux.vnet.ibm.com \
    --to=mrhines@linux.vnet.ibm.com \
    --cc=abali@us.ibm.com \
    --cc=aliguori@us.ibm.com \
    --cc=gokul@us.ibm.com \
    --cc=mrhines@us.ibm.com \
    --cc=owasserm@redhat.com \
    --cc=pbonzini@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).