qemu-kvm/kvm-multifd-Implement-zero-copy-write-in-multifd-migrati.patch

From c1a2866d158ac67179fa0d17f1710302eb9a3866 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Fri, 13 May 2022 03:28:37 -0300
Subject: [PATCH 14/18] multifd: Implement zero copy write in multifd migration
 (multifd-zero-copy)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 95: MSG_ZEROCOPY + Multifd
RH-Commit: [8/11] b93009cc94b2cc4b464b4f68ebfb37b870dd6f7d (LeoBras/centos-qemu-kvm)
RH-Bugzilla: 1968509
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>

Implement zero copy send on nocomp_send_write(), by making use of QIOChannel
writev + flags & flush interface.

Change multifd_send_sync_main() so flush_zero_copy() can be called
after each iteration in order to make sure all dirty pages are sent before
a new iteration is started. It will also flush at the beginning and at the
end of migration.

Also make it return -1 if flush_zero_copy() fails, in order to cancel
the migration process, and avoid resuming the guest in the target host
without receiving all current RAM.

This will work fine on RAM migration because the RAM pages are not usually freed,
and there is no problem on changing the pages content between writev_zero_copy() and
the actual sending of the buffer, because this change will dirty the page and
cause it to be re-sent on a next iteration anyway.

A lot of locked memory may be needed in order to use multifd migration
with zero-copy enabled, so disabling the feature should be necessary for
low-privileged users trying to perform multifd migrations.

Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220513062836.965425-9-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 5b1d9bab2da4fca3a3caee97c430e5709cb32b7b)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
 migration/migration.c | 11 ++++++++++-
 migration/multifd.c   | 37 +++++++++++++++++++++++++++++++++++--
 migration/multifd.h   |  2 ++
 migration/socket.c    |  5 +++--
 4 files changed, 50 insertions(+), 5 deletions(-)

diff --git a/migration/migration.c b/migration/migration.c
index d91efb66fe..102236fba0 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1485,7 +1485,16 @@ static bool migrate_params_check(MigrationParameters *params, Error **errp)
         error_prepend(errp, "Invalid mapping given for block-bitmap-mapping: ");
         return false;
     }
-
+#ifdef CONFIG_LINUX
+    if (params->zero_copy_send &&
+        (!migrate_use_multifd() ||
+         params->multifd_compression != MULTIFD_COMPRESSION_NONE ||
+         (params->tls_creds && *params->tls_creds))) {
+        error_setg(errp,
+                   "Zero copy only available for non-compressed non-TLS multifd migration");
+        return false;
+    }
+#endif
     return true;
 }
 
diff --git a/migration/multifd.c b/migration/multifd.c
index 8fca6c970e..0b5b41c53f 100644
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -571,6 +571,7 @@ void multifd_save_cleanup(void)
 int multifd_send_sync_main(QEMUFile *f)
 {
     int i;
+    bool flush_zero_copy;
 
     if (!migrate_use_multifd()) {
         return 0;
@@ -581,6 +582,20 @@ int multifd_send_sync_main(QEMUFile *f)
             return -1;
         }
     }
+
+    /*
+     * When using zero-copy, it's necessary to flush the pages before any of
+     * the pages can be sent again, so we'll make sure the new version of the
+     * pages will always arrive _later_ than the old pages.
+     *
+     * Currently we achieve this by flushing the zero-page requested writes
+     * per ram iteration, but in the future we could potentially optimize it
+     * to be less frequent, e.g. only after we finished one whole scanning of
+     * all the dirty bitmaps.
+     */
+
+    flush_zero_copy = migrate_use_zero_copy_send();
+
     for (i = 0; i < migrate_multifd_channels(); i++) {
         MultiFDSendParams *p = &multifd_send_state->params[i];
 
@@ -602,6 +617,17 @@ int multifd_send_sync_main(QEMUFile *f)
         ram_counters.transferred += p->packet_len;
         qemu_mutex_unlock(&p->mutex);
         qemu_sem_post(&p->sem);
+
+        if (flush_zero_copy && p->c) {
+            int ret;
+            Error *err = NULL;
+
+            ret = qio_channel_flush(p->c, &err);
+            if (ret < 0) {
+                error_report_err(err);
+                return -1;
+            }
+        }
     }
     for (i = 0; i < migrate_multifd_channels(); i++) {
         MultiFDSendParams *p = &multifd_send_state->params[i];
@@ -686,8 +712,8 @@ static void *multifd_send_thread(void *opaque)
                 p->iov[0].iov_base = p->packet;
             }
 
-            ret = qio_channel_writev_all(p->c, p->iov, p->iovs_num,
-                                         &local_err);
+            ret = qio_channel_writev_full_all(p->c, p->iov, p->iovs_num, NULL,
+                                              0, p->write_flags, &local_err);
             if (ret != 0) {
                 break;
             }
@@ -928,6 +954,13 @@ int multifd_save_setup(Error **errp)
         /* We need one extra place for the packet header */
         p->iov = g_new0(struct iovec, page_count + 1);
         p->normal = g_new0(ram_addr_t, page_count);
+
+        if (migrate_use_zero_copy_send()) {
+            p->write_flags = QIO_CHANNEL_WRITE_FLAG_ZERO_COPY;
+        } else {
+            p->write_flags = 0;
+        }
+
         socket_send_channel_create(multifd_new_send_channel_async, p);
     }
 
diff --git a/migration/multifd.h b/migration/multifd.h
index cd495195ce..7ec688fb4f 100644
--- a/migration/multifd.h
+++ b/migration/multifd.h
@@ -96,6 +96,8 @@ typedef struct {
     uint32_t packet_len;
     /* pointer to the packet */
     MultiFDPacket_t *packet;
+    /* multifd flags for sending ram */
+    int write_flags;
     /* multifd flags for each packet */
     uint32_t flags;
     /* size of the next packet that contains pages */
diff --git a/migration/socket.c b/migration/socket.c
index 3754d8f72c..4fd5e85f50 100644
--- a/migration/socket.c
+++ b/migration/socket.c
@@ -79,8 +79,9 @@ static void socket_outgoing_migration(QIOTask *task,
 
     trace_migration_socket_outgoing_connected(data->hostname);
 
-    if (migrate_use_zero_copy_send()) {
-        error_setg(&err, "Zero copy send not available in migration");
+    if (migrate_use_zero_copy_send() &&
+        !qio_channel_has_feature(sioc, QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY)) {
+        error_setg(&err, "Zero copy send feature not detected in host kernel");
     }
 
 out:
-- 
2.35.3