Skip to contentSkip to conversation

Zero copy improvements (QIOChannel + multifd)

PATCH In review Leonardo Bras 3 participants 9 messages Jul 11, 2022 – Jul 19, 2022

lore
Share & tools
lore
#0 lore raw

v4 cover Zero copy improvements (QIOChannel + multifd)

The first patch avoid spuriously returning 1 [*] when zero-copy flush is attempted before any buffer was sent using MSG_ZEROCOPY.

[*] zero-copy not being used, even though it's enabled and supported by kernel

The second patch introduces a new migration stat (dirty-sync-missed-zero-copy) that will be used to keep track of [*].

The third patch keeps track of how many zero-copy flushes retured 1 [*]

Changes since v3:
- Patch#1: Also checks if no packet was queued after the last flush
- Patch#2: Improve dirty-sync-missed-zero-copy doc and hmp print message

Changes since v2:
- Documentation release number changed from 7.2 to 7.1
- migration stat renamed from zero-copy-copied to dirty-sync-missed-zero-copy
- Updated documentation to make it more user-friendly

Changes since v1:
- Idea of using a warning replaced by using a migration stat counter

Leonardo Bras (3):
QIOChannelSocket: Fix zero-copy flush returning code 1 when nothing
sent
Add dirty-sync-missed-zero-copy migration stat migration/multifd: Report to user when zerocopy not working

 qapi/migration.json   | 7 ++++++-
 migration/ram.h       | 2 ++
 io/channel-socket.c   | 8 +++++++-
 migration/migration.c | 2 ++
 migration/multifd.c   | 2 ++
 migration/ram.c       | 5 +++++
 monitor/hmp-cmds.c    | 5 +++++
 7 files changed, 29 insertions(+), 2 deletions(-)
Signature · 1 line
2.37.0
#1 lore raw

PATCH v4 1/3 QIOChannelSocket: Fix zero-copy flush returning code 1 when nothing sent

If flush is called when no buffer was sent with MSG_ZEROCOPY, it currently returns 1. This return code should be used only when Linux fails to use MSG_ZEROCOPY on a lot of sendmsg().

Fix this by returning early from flush if no sendmsg(...,MSG_ZEROCOPY) was attempted.

Fixes 2bc58ffc2926 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
---
 io/channel-socket.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
io/channel-socket.c +7 -1
index 4466bb1cd4..74a936cc1f 100644
@@ -716,12 +716,18 @@ static int qio_channel_socket_flush(QIOChannel *ioc,
     struct cmsghdr *cm;
     char control[CMSG_SPACE(sizeof(*serr))];
     int received;
-    int ret = 1;
+    int ret;
+
+    if (sioc->zero_copy_queued == sioc->zero_copy_sent) {
+        return 0;
+    }
 
     msg.msg_control = control;
     msg.msg_controllen = sizeof(control);
     memset(control, 0, sizeof(control));
 
+    ret = 1;
+
     while (sioc->zero_copy_sent < sioc->zero_copy_queued) {
         received = recvmsg(sioc->fd, &msg, MSG_ERRQUEUE);
         if (received < 0) {
Signature · 1 line
2.37.0
Reviewed-by on v4 1/3
#2 lore raw

Re: [PATCH v4 1/3] QIOChannelSocket: Fix zero-copy flush returning code 1 when nothing sent

Quoted 4 lines · Leonardo Bras

If flush is called when no buffer was sent with MSG_ZEROCOPY, it currently returns 1. This return code should be used only when Linux fails to use MSG_ZEROCOPY on a lot of sendmsg().

Quoted 8 lines

Fix this by returning early from flush if no sendmsg(...,MSG_ZEROCOPY) was attempted.

Fixes 2bc58ffc2926 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Signature · 1 line
Peter Xu
#3 lore raw

PATCH v4 2/3 Add dirty-sync-missed-zero-copy migration stat

In reply to Leonardo Bras
---
 qapi/migration.json   | 7 ++++++-
 migration/migration.c | 2 ++
 monitor/hmp-cmds.c    | 5 +++++
 3 files changed, 13 insertions(+), 1 deletion(-)
qapi/migration.json +6 -1
index 7102e474a6..4a03e8f173 100644
@@ -55,6 +55,10 @@
 # @postcopy-bytes: The number of bytes sent during the post-copy phase
 #                  (since 7.0).
 #
+# @dirty-sync-missed-zero-copy: Number of times dirty RAM synchronization could
+#                               not avoid copying dirty pages. This is between
+#                               0 and @dirty-sync-count * @multifd-channels.
+#                               (since 7.1)
 # Since: 0.14
 ##
 { 'struct': 'MigrationStats',
@@ -65,7 +69,8 @@
            'postcopy-requests' : 'int', 'page-size' : 'int',
            'multifd-bytes' : 'uint64', 'pages-per-second' : 'uint64',
            'precopy-bytes' : 'uint64', 'downtime-bytes' : 'uint64',
-           'postcopy-bytes' : 'uint64' } }
+           'postcopy-bytes' : 'uint64',
+           'dirty-sync-missed-zero-copy' : 'uint64' } }
 
 ##
 # @XBZRLECacheStats:
migration/migration.c +2 -0
index 78f5057373..048f7f8bdb 100644
@@ -1027,6 +1027,8 @@ static void populate_ram_info(MigrationInfo *info, MigrationState *s)
     info->ram->normal_bytes = ram_counters.normal * page_size;
     info->ram->mbps = s->mbps;
     info->ram->dirty_sync_count = ram_counters.dirty_sync_count;
+    info->ram->dirty_sync_missed_zero_copy =
+            ram_counters.dirty_sync_missed_zero_copy;
     info->ram->postcopy_requests = ram_counters.postcopy_requests;
     info->ram->page_size = page_size;
     info->ram->multifd_bytes = ram_counters.multifd_bytes;
monitor/hmp-cmds.c +5 -0
index ca98df0495..a6dc79e0d5 100644
@@ -307,6 +307,11 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
             monitor_printf(mon, "postcopy ram: %" PRIu64 " kbytes\n",
                            info->ram->postcopy_bytes >> 10);
         }
+        if (info->ram->dirty_sync_missed_zero_copy) {
+            monitor_printf(mon,
+                           "Zero-copy-send fallbacks happened: %" PRIu64 " times\n",
+                           info->ram->dirty_sync_missed_zero_copy);
+        }
     }
 
     if (info->has_disk) {
Signature · 1 line
2.37.0
Acked-by on v4 2/3
#4 lore raw

Re: [PATCH v4 2/3] Add dirty-sync-missed-zero-copy migration stat

Quoted 52 lines · Leonardo Bras
---
 qapi/migration.json   | 7 ++++++-
 migration/migration.c | 2 ++
 monitor/hmp-cmds.c    | 5 +++++
 3 files changed, 13 insertions(+), 1 deletion(-)
qapi/migration.json +6 -1
index 7102e474a6..4a03e8f173 100644
@@ -55,6 +55,10 @@
 # @postcopy-bytes: The number of bytes sent during the post-copy phase
 #                  (since 7.0).
 #
+# @dirty-sync-missed-zero-copy: Number of times dirty RAM synchronization could
+#                               not avoid copying dirty pages. This is between
+#                               0 and @dirty-sync-count * @multifd-channels.
+#                               (since 7.1)
 # Since: 0.14
 ##
 { 'struct': 'MigrationStats',
@@ -65,7 +69,8 @@
            'postcopy-requests' : 'int', 'page-size' : 'int',
            'multifd-bytes' : 'uint64', 'pages-per-second' : 'uint64',
            'precopy-bytes' : 'uint64', 'downtime-bytes' : 'uint64',
-           'postcopy-bytes' : 'uint64' } }
+           'postcopy-bytes' : 'uint64',
+           'dirty-sync-missed-zero-copy' : 'uint64' } }
 
 ##
 # @XBZRLECacheStats:
migration/migration.c +2 -0
index 78f5057373..048f7f8bdb 100644
@@ -1027,6 +1027,8 @@ static void populate_ram_info(MigrationInfo *info, MigrationState *s)
     info->ram->normal_bytes = ram_counters.normal * page_size;
     info->ram->mbps = s->mbps;
     info->ram->dirty_sync_count = ram_counters.dirty_sync_count;
+    info->ram->dirty_sync_missed_zero_copy =
+            ram_counters.dirty_sync_missed_zero_copy;
     info->ram->postcopy_requests = ram_counters.postcopy_requests;
     info->ram->page_size = page_size;
     info->ram->multifd_bytes = ram_counters.multifd_bytes;
monitor/hmp-cmds.c +0 -0
index ca98df0495..a6dc79e0d5 100644

monitor/hmp-cmds.c void hmp_info_migrate(Monitor *mon, const QDict *qdict) · line 313

Quoted 7 lines
monitor/hmp-cmds.c +4 -0
@@ -307,6 +307,11 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
             monitor_printf(mon, "postcopy ram: %" PRIu64 " kbytes\n",
                            info->ram->postcopy_bytes >> 10);
         }
+        if (info->ram->dirty_sync_missed_zero_copy) {
+            monitor_printf(mon,
+                           "Zero-copy-send fallbacks happened: %" PRIu64 " times\n",
+                           info->ram->dirty_sync_missed_zero_copy);

Thanks, this looks better. Though I think all the "dirty-sync" wordings are still kept there, assuming flush() is bound to dirty sync even it's not yet. Not a big deal, but let's still keep an eye on the follow up patches..

Signature · 1 line
Peter Xu
#5 lore raw

PATCH v4 3/3 migration/multifd: Report to user when zerocopy not working

In reply to Leonardo Bras

Some errors, like the lack of Scatter-Gather support by the network interface(NETIF_F_SG) may cause sendmsg(...,MSG_ZEROCOPY) to fail on using zero-copy, which causes it to fall back to the default copying mechanism.

After each full dirty-bitmap scan there should be a zero-copy flush happening, which checks for errors each of the previous calls to sendmsg(...,MSG_ZEROCOPY). If all of them failed to use zero-copy, then increment dirty_sync_missed_zero_copy migration stat to let the user know about it.

---
 migration/ram.h     | 2 ++
 migration/multifd.c | 2 ++
 migration/ram.c     | 5 +++++
 3 files changed, 9 insertions(+)
migration/ram.h +2 -0
index ded0a3a086..d3c7eb96f5 100644
@@ -87,4 +87,6 @@ void ram_write_tracking_prepare(void);
 int ram_write_tracking_start(void);
 void ram_write_tracking_stop(void);
 
+void dirty_sync_missed_zero_copy(void);
+
 #endif
migration/multifd.c +2 -0
index 684c014c86..3909b34967 100644
@@ -624,6 +624,8 @@ int multifd_send_sync_main(QEMUFile *f)
             if (ret < 0) {
                 error_report_err(err);
                 return -1;
+            } else if (ret == 1) {
+                dirty_sync_missed_zero_copy();
             }
         }
     }
migration/ram.c +5 -0
index 01f9cc1d72..db948c4787 100644
@@ -407,6 +407,11 @@ static void ram_transferred_add(uint64_t bytes)
     ram_counters.transferred += bytes;
 }
 
+void dirty_sync_missed_zero_copy(void)
+{
+    ram_counters.dirty_sync_missed_zero_copy++;
+}
+
 /* used by the search for pages to send */
 struct PageSearchStatus {
     /* Current block being searched */
Signature · 1 line
2.37.0
Acked-by on v4 3/3
#6 lore raw

Re: [PATCH v4 3/3] migration/multifd: Report to user when zerocopy not working

Quoted 4 lines · Leonardo Bras

Some errors, like the lack of Scatter-Gather support by the network interface(NETIF_F_SG) may cause sendmsg(...,MSG_ZEROCOPY) to fail on using zero-copy, which causes it to fall back to the default copying mechanism.

Quoted 8 lines

After each full dirty-bitmap scan there should be a zero-copy flush happening, which checks for errors each of the previous calls to sendmsg(...,MSG_ZEROCOPY). If all of them failed to use zero-copy, then increment dirty_sync_missed_zero_copy migration stat to let the user know about it.

Signature · 1 line
Peter Xu
#7 lore raw

Re: [PATCH v4 3/3] migration/multifd: Report to user when zerocopy not working

Quoted 12 lines · Peter Xu
Quoted 10 lines · Leonardo Bras

Some errors, like the lack of Scatter-Gather support by the network interface(NETIF_F_SG) may cause sendmsg(...,MSG_ZEROCOPY) to fail on using zero-copy, which causes it to fall back to the default copying mechanism.

After each full dirty-bitmap scan there should be a zero-copy flush happening, which checks for errors each of the previous calls to sendmsg(...,MSG_ZEROCOPY). If all of them failed to use zero-copy, then increment dirty_sync_missed_zero_copy migration stat to let the user know about it.

Quoted 4 lines
Quoted 2 lines

Thanks Peter!

Quoted 3 lines

--
Peter Xu

#8 lore raw
In reply to Leonardo Bras
Quoted 15 lines · * Leonardo Bras ([email protected]) wrote:

The first patch avoid spuriously returning 1 [*] when zero-copy flush is attempted before any buffer was sent using MSG_ZEROCOPY.

[*] zero-copy not being used, even though it's enabled and supported by kernel

The second patch introduces a new migration stat (dirty-sync-missed-zero-copy) that will be used to keep track of [*].

The third patch keeps track of how many zero-copy flushes retured 1 [*]

Changes since v3:
- Patch#1: Also checks if no packet was queued after the last flush
- Patch#2: Improve dirty-sync-missed-zero-copy doc and hmp print message

Quoted 8 lines

Changes since v2:
- Documentation release number changed from 7.2 to 7.1
- migration stat renamed from zero-copy-copied to dirty-sync-missed-zero-copy
- Updated documentation to make it more user-friendly

Changes since v1:
- Idea of using a warning replaced by using a migration stat counter

Queued

Quoted 22 lines

Leonardo Bras (3):
QIOChannelSocket: Fix zero-copy flush returning code 1 when nothing
sent
Add dirty-sync-missed-zero-copy migration stat migration/multifd: Report to user when zerocopy not working

 qapi/migration.json   | 7 ++++++-
 migration/ram.h       | 2 ++
 io/channel-socket.c   | 8 +++++++-
 migration/migration.c | 2 ++
 migration/multifd.c   | 2 ++
 migration/ram.c       | 5 +++++
 monitor/hmp-cmds.c    | 5 +++++
 7 files changed, 29 insertions(+), 2 deletions(-)
Signature · 1 line
2.37.0
Signature · 1 line
Dr. David Alan Gilbert / [email protected] / Manchester, UK