Skip to contentSkip to conversation

MSG_ZEROCOPY + multifd

PATCH Superseded Leonardo Bras 4 participants 17 messages Apr 25, 2022 – May 4, 2022

v9 of 13 Superseded Latest is v13 →

lore
Share & tools
lore
#0 lore raw

v9 cover MSG_ZEROCOPY + multifd

This patch series intends to enable MSG_ZEROCOPY in QIOChannel, and make use of it for multifd migration performance improvement, by reducing cpu usage.

Patch #1 creates new callbacks for QIOChannel, allowing the implementation of zero copy writing.

Patch #2 implements io_writev flags and io_flush() on QIOChannelSocket, making use of MSG_ZEROCOPY on Linux.

Patch #3 adds a "zero_copy_send" migration property, only available with CONFIG_LINUX, and compiled-out in any other architectures. This migration property has to be enabled before multifd migration starts.

Patch #4 adds a helper function that allows to see if TLS is going to be used. This helper will be later used in patch #5.

Patch #5 changes multifd_send_sync_main() so it returns int instead of void. The return value is used to understand if any error happened in the function, allowing migration to possible fail earlier.

Patch #6 implements an workaround: The behavior introduced in d48c3a0445 is hard to deal with in zerocopy, so a workaround is introduced to send the header in a different syscall, without MSG_ZEROCOPY.

Patch #7 Makes use of QIOChannelSocket zero_copy implementation on nocomp multifd migration.

Results:
In preliminary tests, the resource usage of __sys_sendmsg() reduced 15 times, and the overall migration took 13-22% less time, based in synthetic cpu workload.

In further tests, it was noted that, on multifd migration with 8 channels:
- On idle hosts, migration time reduced in 10% to 21%.
- On hosts busy with heavy cpu stress (1 stress thread per cpu, but not cpu-pinned) migration time reduced in ~25% by enabling zero-copy.
- On hosts with heavy cpu-pinned workloads (1 stress thread per cpu, cpu-pinned), migration time reducted in ~66% by enabling zero-copy.

Above tests setup:
- Sending and Receiving hosts:
- CPU : Intel(R) Xeon(R) Platinum 8276L CPU @ 2.20GHz (448 CPUS)
- Network card: E810-C (100Gbps)
- >1TB RAM
- QEMU: Upstream master branch + This patchset
- Linux: Upstream v5.15
- VM configuration:
- 28 VCPUs
- 512GB RAM

---
Changes since v8:
- Inserted two new patches #5 & #6, previous patch #5 is now #7.
- Workaround an optimization introduced in d48c3a0445
- Removed unnecessary assert in qio_channel_writev_full_all

Changes since v7:
- Migration property renamed from zero-copy to zero-copy-send
- A few early tests added to help misconfigurations to fail earlier
- qio_channel_full*_flags() renamed back to qio_channel_full*()
- multifd_send_sync_main() reverted back to not receiving a flag, so it always sync zero-copy when enabled.
- Improve code quality on a few points

Changes since v6:
- Remove io_writev_zero_copy(), and makes use of io_writev() new flags to achieve the same results.
- Rename io_flush_zero_copy() to io_flush()
- Previous patch #2 became too small, so it was squashed in previous patch #3 (now patch #2)

Changes since v5:
- flush_zero_copy now returns -1 on fail, 0 on success, and 1 when all processed writes were not able to use zerocopy in kernel.
- qio_channel_socket_poll() removed, using qio_channel_wait() instead
- ENOBUFS is now processed inside qio_channel_socket_writev_flags()
- Most zerocopy parameter validation moved to migrate_params_check(), leaving only feature test to socket_outgoing_migration() callback
- Naming went from *zerocopy to *zero_copy or *zero-copy, due to QAPI/QMP preferences
- Improved docs

Changes since v4:
- 3 patches got splitted in 6
- Flush is used for syncing after each iteration, instead of only at the end
- If zerocopy is not available, fail in connect instead of failing on write
- 'multifd-zerocopy' property renamed to 'zerocopy'
- Fail migrations that don't support zerocopy, if it's enabled.
- Instead of checking for zerocopy at each write, save the flags in MultiFDSendParams->write_flags and use them on write
- Reorganized flag usage in QIOChannelSocket
- A lot of typos fixed
- More doc on buffer restrictions

Changes since v3:
- QIOChannel interface names changed from io_async_{writev,flush} to io_{writev,flush}_zerocopy
- Instead of falling back in case zerocopy is not implemented, return error and abort operation.
- Flush now waits as long as needed, or return error in case anything goes wrong, aborting the operation.
- Zerocopy is now conditional in multifd, being set by parameter multifd-zerocopy
- Moves zerocopy_flush to multifd_send_sync_main() from multifd_save_cleanup so migration can abort if flush goes wrong.
- Several other small improvements

Changes since v2:
- Patch #1: One more fallback
- Patch #2: Fall back to sync if fails to lock buffer memory in MSG_ZEROCOPY send.

Changes since v1:
- Reimplemented the patchset using async_write + async_flush approach.
- Implemented a flush to be able to tell whenever all data was written.

Leonardo Bras (7):
QIOChannel: Add flags on io_writev and introduce io_flush callback
QIOChannelSocket: Implement io_writev zero copy flag & io_flush for
CONFIG_LINUX
migration: Add zero-copy-send parameter for QMP/HMP for Linux
migration: Add migrate_use_tls() helper
multifd: multifd_send_sync_main now returns negative on error
multifd: Send header packet without flags if zero-copy-send is enabled
multifd: Implement zero copy write in multifd migration
(multifd-zero-copy)

 qapi/migration.json                 |  24 ++++++
 include/io/channel-socket.h         |   2 +
 include/io/channel.h                |  38 +++++++++-
 migration/migration.h               |   6 ++
 migration/multifd.h                 |   4 +-
 chardev/char-io.c                   |   2 +-
 hw/remote/mpqemu-link.c             |   2 +-
 io/channel-buffer.c                 |   1 +
 io/channel-command.c                |   1 +
 io/channel-file.c                   |   1 +
 io/channel-socket.c                 | 110 +++++++++++++++++++++++++++-
 io/channel-tls.c                    |   1 +
 io/channel-websock.c                |   1 +
 io/channel.c                        |  49 ++++++++++---
 migration/channel.c                 |   3 +-
 migration/migration.c               |  52 ++++++++++++-
 migration/multifd.c                 |  72 +++++++++++++++---
 migration/ram.c                     |  29 ++++++--
 migration/rdma.c                    |   1 +
 migration/socket.c                  |  12 ++-
 monitor/hmp-cmds.c                  |   6 ++
 scsi/pr-manager-helper.c            |   2 +-
 tests/unit/test-io-channel-socket.c |   1 +
 23 files changed, 377 insertions(+), 43 deletions(-)
Signature · 1 line
2.36.0
#1 lore raw

PATCH v9 1/7 QIOChannel: Add flags on io_writev and introduce io_flush callback

Add flags to io_writev and introduce io_flush as optional callback to QIOChannelClass, allowing the implementation of zero copy writes by subclasses.

How to use them:
- Write data using qio_channel_writev*(...,QIO_CHANNEL_WRITE_FLAG_ZERO_COPY),
- Wait write completion with qio_channel_flush().

Notes:
As some zero copy write implementations work asynchronously, it's recommended to keep the write buffer untouched until the return of qio_channel_flush(), to avoid the risk of sending an updated buffer instead of the buffer state during write.

As io_flush callback is optional, if a subclass does not implement it, then:
- io_flush will return 0 without changing anything.

Also, some functions like qio_channel_writev_full_all() were adapted to receive a flag parameter. That allows shared code between zero copy and non-zero copy writev, and also an easier implementation on new flags.

---
 include/io/channel.h                | 38 +++++++++++++++++++++-
 chardev/char-io.c                   |  2 +-
 hw/remote/mpqemu-link.c             |  2 +-
 io/channel-buffer.c                 |  1 +
 io/channel-command.c                |  1 +
 io/channel-file.c                   |  1 +
 io/channel-socket.c                 |  2 ++
 io/channel-tls.c                    |  1 +
 io/channel-websock.c                |  1 +
 io/channel.c                        | 49 +++++++++++++++++++++++------
 migration/rdma.c                    |  1 +
 scsi/pr-manager-helper.c            |  2 +-
 tests/unit/test-io-channel-socket.c |  1 +
 13 files changed, 88 insertions(+), 14 deletions(-)
include/io/channel.h +37 -1
index 88988979f8..c680ee7480 100644
@@ -32,12 +32,15 @@ OBJECT_DECLARE_TYPE(QIOChannel, QIOChannelClass,
 
 #define QIO_CHANNEL_ERR_BLOCK -2
 
+#define QIO_CHANNEL_WRITE_FLAG_ZERO_COPY 0x1
+
 typedef enum QIOChannelFeature QIOChannelFeature;
 
 enum QIOChannelFeature {
     QIO_CHANNEL_FEATURE_FD_PASS,
     QIO_CHANNEL_FEATURE_SHUTDOWN,
     QIO_CHANNEL_FEATURE_LISTEN,
+    QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY,
 };
 
 
@@ -104,6 +107,7 @@ struct QIOChannelClass {
                          size_t niov,
                          int *fds,
                          size_t nfds,
+                         int flags,
                          Error **errp);
     ssize_t (*io_readv)(QIOChannel *ioc,
                         const struct iovec *iov,
@@ -136,6 +140,8 @@ struct QIOChannelClass {
                                   IOHandler *io_read,
                                   IOHandler *io_write,
                                   void *opaque);
+    int (*io_flush)(QIOChannel *ioc,
+                    Error **errp);
 };
 
 /* General I/O handling functions */
@@ -228,6 +234,7 @@ ssize_t qio_channel_readv_full(QIOChannel *ioc,
  * @niov: the length of the @iov array
  * @fds: an array of file handles to send
  * @nfds: number of file handles in @fds
+ * @flags: write flags (QIO_CHANNEL_WRITE_FLAG_*)
  * @errp: pointer to a NULL-initialized error object
  *
  * Write data to the IO channel, reading it from the
@@ -260,6 +267,7 @@ ssize_t qio_channel_writev_full(QIOChannel *ioc,
                                 size_t niov,
                                 int *fds,
                                 size_t nfds,
+                                int flags,
                                 Error **errp);
 
 /**
@@ -837,6 +845,7 @@ int qio_channel_readv_full_all(QIOChannel *ioc,
  * @niov: the length of the @iov array
  * @fds: an array of file handles to send
  * @nfds: number of file handles in @fds
+ * @flags: write flags (QIO_CHANNEL_WRITE_FLAG_*)
  * @errp: pointer to a NULL-initialized error object
  *
  *
@@ -846,6 +855,14 @@ int qio_channel_readv_full_all(QIOChannel *ioc,
  * to be written, yielding from the current coroutine
  * if required.
  *
+ * If QIO_CHANNEL_WRITE_FLAG_ZERO_COPY is passed in flags,
+ * instead of waiting for all requested data to be written,
+ * this function will wait until it's all queued for writing.
+ * In this case, if the buffer gets changed between queueing and
+ * sending, the updated buffer will be sent. If this is not a
+ * desired behavior, it's suggested to call qio_channel_flush()
+ * before reusing the buffer.
+ *
  * Returns: 0 if all bytes were written, or -1 on error
  */
 
@@ -853,6 +870,25 @@ int qio_channel_writev_full_all(QIOChannel *ioc,
                                 const struct iovec *iov,
                                 size_t niov,
                                 int *fds, size_t nfds,
-                                Error **errp);
+                                int flags, Error **errp);
+
+/**
+ * qio_channel_flush:
+ * @ioc: the channel object
+ * @errp: pointer to a NULL-initialized error object
+ *
+ * Will block until every packet queued with
+ * qio_channel_writev_full() + QIO_CHANNEL_WRITE_FLAG_ZERO_COPY
+ * is sent, or return in case of any error.
+ *
+ * If not implemented, acts as a no-op, and returns 0.
+ *
+ * Returns -1 if any error is found,
+ *          1 if every send failed to use zero copy.
+ *          0 otherwise.
+ */
+
+int qio_channel_flush(QIOChannel *ioc,
+                      Error **errp);
 
 #endif /* QIO_CHANNEL_H */
chardev/char-io.c +1 -1
index 8ced184160..4451128cba 100644
@@ -122,7 +122,7 @@ int io_channel_send_full(QIOChannel *ioc,
 
         ret = qio_channel_writev_full(
             ioc, &iov, 1,
-            fds, nfds, NULL);
+            fds, nfds, 0, NULL);
         if (ret == QIO_CHANNEL_ERR_BLOCK) {
             if (offset) {
                 return offset;
hw/remote/mpqemu-link.c +1 -1
index 2a4aa651ca..9bd98e8219 100644
@@ -68,7 +68,7 @@ bool mpqemu_msg_send(MPQemuMsg *msg, QIOChannel *ioc, Error **errp)
     }
 
     if (!qio_channel_writev_full_all(ioc, send, G_N_ELEMENTS(send),
-                                    fds, nfds, errp)) {
+                                    fds, nfds, 0, errp)) {
         ret = true;
     } else {
         trace_mpqemu_send_io_error(msg->cmd, msg->size, nfds);
io/channel-buffer.c +1 -0
index baa4e2b089..bf52011be2 100644
@@ -81,6 +81,7 @@ static ssize_t qio_channel_buffer_writev(QIOChannel *ioc,
                                          size_t niov,
                                          int *fds,
                                          size_t nfds,
+                                         int flags,
                                          Error **errp)
 {
     QIOChannelBuffer *bioc = QIO_CHANNEL_BUFFER(ioc);
io/channel-command.c +1 -0
index 338da73ade..54560464ae 100644
@@ -258,6 +258,7 @@ static ssize_t qio_channel_command_writev(QIOChannel *ioc,
                                           size_t niov,
                                           int *fds,
                                           size_t nfds,
+                                          int flags,
                                           Error **errp)
 {
     QIOChannelCommand *cioc = QIO_CHANNEL_COMMAND(ioc);
io/channel-file.c +1 -0
index d7cf6d278f..ef6807a6be 100644
@@ -114,6 +114,7 @@ static ssize_t qio_channel_file_writev(QIOChannel *ioc,
                                        size_t niov,
                                        int *fds,
                                        size_t nfds,
+                                       int flags,
                                        Error **errp)
 {
     QIOChannelFile *fioc = QIO_CHANNEL_FILE(ioc);
io/channel-socket.c +2 -0
index 9f5ddf68b6..696a04dc9c 100644
@@ -524,6 +524,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
                                          size_t niov,
                                          int *fds,
                                          size_t nfds,
+                                         int flags,
                                          Error **errp)
 {
     QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
@@ -619,6 +620,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
                                          size_t niov,
                                          int *fds,
                                          size_t nfds,
+                                         int flags,
                                          Error **errp)
 {
     QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
io/channel-tls.c +1 -0
index 2ae1b92fc0..4ce890a538 100644
@@ -301,6 +301,7 @@ static ssize_t qio_channel_tls_writev(QIOChannel *ioc,
                                       size_t niov,
                                       int *fds,
                                       size_t nfds,
+                                      int flags,
                                       Error **errp)
 {
     QIOChannelTLS *tioc = QIO_CHANNEL_TLS(ioc);
io/channel-websock.c +1 -0
index 55145a6a8c..9619906ac3 100644
@@ -1127,6 +1127,7 @@ static ssize_t qio_channel_websock_writev(QIOChannel *ioc,
                                           size_t niov,
                                           int *fds,
                                           size_t nfds,
+                                          int flags,
                                           Error **errp)
 {
     QIOChannelWebsock *wioc = QIO_CHANNEL_WEBSOCK(ioc);
io/channel.c +39 -10

99 lines not on this page · raw message

migration/rdma.c +1 -0
index ef1e65ec36..672d1958a9 100644
@@ -2840,6 +2840,7 @@ static ssize_t qio_channel_rdma_writev(QIOChannel *ioc,
                                        size_t niov,
                                        int *fds,
                                        size_t nfds,
+                                       int flags,
                                        Error **errp)
 {
     QIOChannelRDMA *rioc = QIO_CHANNEL_RDMA(ioc);
scsi/pr-manager-helper.c +1 -1
index 451c7631b7..3be52a98d5 100644
@@ -77,7 +77,7 @@ static int pr_manager_helper_write(PRManagerHelper *pr_mgr,
         iov.iov_base = (void *)buf;
         iov.iov_len = sz;
         n_written = qio_channel_writev_full(QIO_CHANNEL(pr_mgr->ioc), &iov, 1,
-                                            nfds ? &fd : NULL, nfds, errp);
+                                            nfds ? &fd : NULL, nfds, 0, errp);
 
         if (n_written <= 0) {
             assert(n_written != QIO_CHANNEL_ERR_BLOCK);
tests/unit/test-io-channel-socket.c +1 -0
index c49eec1f03..6713886d02 100644
@@ -444,6 +444,7 @@ static void test_io_channel_unix_fd_pass(void)
                             G_N_ELEMENTS(iosend),
                             fdsend,
                             G_N_ELEMENTS(fdsend),
+                            0,
                             &error_abort);
 
     qio_channel_readv_full(dst,
Signature · 1 line
2.36.0
#2 lore raw

PATCH v9 2/7 QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX

In reply to Leonardo Bras

For CONFIG_LINUX, implement the new zero copy flag and the optional callback io_flush on QIOChannelSocket, but enables it only when MSG_ZEROCOPY feature is available in the host kernel, which is checked on qio_channel_socket_connect_sync()

qio_channel_socket_flush() was implemented by counting how many times sendmsg(...,MSG_ZEROCOPY) was successfully called, and then reading the socket's error queue, in order to find how many of them finished sending. Flush will loop until those counters are the same, or until some error occurs.

Notes on using writev() with QIO_CHANNEL_WRITE_FLAG_ZERO_COPY:
1: Buffer
- As MSG_ZEROCOPY tells the kernel to use the same user buffer to avoid copying, some caution is necessary to avoid overwriting any buffer before it's sent. If something like this happen, a newer version of the buffer may be sent instead.
- If this is a problem, it's recommended to call qio_channel_flush() before freeing or re-using the buffer.

2: Locked memory
- When using MSG_ZERCOCOPY, the buffer memory will be locked after queued, and unlocked after it's sent.
- Depending on the size of each buffer, and how often it's sent, it may require a larger amount of locked memory than usually available to non-root user.
- If the required amount of locked memory is not available, writev_zero_copy will return an error, which can abort an operation like migration,
- Because of this, when an user code wants to add zero copy as a feature, it requires a mechanism to disable it, so it can still be accessible to less privileged users.

---
 include/io/channel-socket.h |   2 +
 io/channel-socket.c         | 108 ++++++++++++++++++++++++++++++++++--
 2 files changed, 106 insertions(+), 4 deletions(-)
include/io/channel-socket.h +2 -0
index e747e63514..513c428fe4 100644
@@ -47,6 +47,8 @@ struct QIOChannelSocket {
     socklen_t localAddrLen;
     struct sockaddr_storage remoteAddr;
     socklen_t remoteAddrLen;
+    ssize_t zero_copy_queued;
+    ssize_t zero_copy_sent;
 };
 
 
io/channel-socket.c +104 -4

155 lines not on this page · raw message

Signature · 1 line
2.36.0
#3 lore raw

PATCH v9 3/7 migration: Add zero-copy-send parameter for QMP/HMP for Linux

In reply to Leonardo Bras

Add property that allows zero-copy migration of memory pages on the sending side, and also includes a helper function migrate_use_zero_copy_send() to check if it's enabled.

No code is introduced to actually do the migration, but it allow future implementations to enable/disable this feature.

On non-Linux builds this parameter is compiled-out.

---
 qapi/migration.json   | 24 ++++++++++++++++++++++++
 migration/migration.h |  5 +++++
 migration/migration.c | 32 ++++++++++++++++++++++++++++++++
 migration/socket.c    | 11 +++++++++--
 monitor/hmp-cmds.c    |  6 ++++++
 5 files changed, 76 insertions(+), 2 deletions(-)
qapi/migration.json +24 -0
index 409eb086a2..04246481ce 100644
@@ -741,6 +741,13 @@
 #                      will consume more CPU.
 #                      Defaults to 1. (Since 5.0)
 #
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+#                  When true, enables a zero-copy mechanism for sending memory
+#                  pages, if host supports it.
+#                  Requires that QEMU be permitted to use locked memory for guest
+#                  RAM pages.
+#                  Defaults to false. (Since 7.1)
+#
 # @block-bitmap-mapping: Maps block nodes and bitmaps on them to
 #                        aliases for the purpose of dirty bitmap migration.  Such
 #                        aliases may for example be the corresponding names on the
@@ -780,6 +787,7 @@
            'xbzrle-cache-size', 'max-postcopy-bandwidth',
            'max-cpu-throttle', 'multifd-compression',
            'multifd-zlib-level' ,'multifd-zstd-level',
+           { 'name': 'zero-copy-send', 'if' : 'CONFIG_LINUX'},
            'block-bitmap-mapping' ] }
 
 ##
@@ -906,6 +914,13 @@
 #                      will consume more CPU.
 #                      Defaults to 1. (Since 5.0)
 #
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+#                  When true, enables a zero-copy mechanism for sending memory
+#                  pages, if host supports it.
+#                  Requires that QEMU be permitted to use locked memory for guest
+#                  RAM pages.
+#                  Defaults to false. (Since 7.1)
+#
 # @block-bitmap-mapping: Maps block nodes and bitmaps on them to
 #                        aliases for the purpose of dirty bitmap migration.  Such
 #                        aliases may for example be the corresponding names on the
@@ -960,6 +975,7 @@
             '*multifd-compression': 'MultiFDCompression',
             '*multifd-zlib-level': 'uint8',
             '*multifd-zstd-level': 'uint8',
+            '*zero-copy-send': { 'type': 'bool', 'if': 'CONFIG_LINUX' },
             '*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ] } }
 
 ##
@@ -1106,6 +1122,13 @@
 #                      will consume more CPU.
 #                      Defaults to 1. (Since 5.0)
 #
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+#                  When true, enables a zero-copy mechanism for sending memory
+#                  pages, if host supports it.
+#                  Requires that QEMU be permitted to use locked memory for guest
+#                  RAM pages.
+#                  Defaults to false. (Since 7.1)
+#
 # @block-bitmap-mapping: Maps block nodes and bitmaps on them to
 #                        aliases for the purpose of dirty bitmap migration.  Such
 #                        aliases may for example be the corresponding names on the
@@ -1158,6 +1181,7 @@
             '*multifd-compression': 'MultiFDCompression',
             '*multifd-zlib-level': 'uint8',
             '*multifd-zstd-level': 'uint8',
+            '*zero-copy-send': { 'type': 'bool', 'if': 'CONFIG_LINUX' },
             '*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ] } }
 
 ##
migration/migration.h +5 -0
index a863032b71..e8f2941a55 100644
@@ -375,6 +375,11 @@ MultiFDCompression migrate_multifd_compression(void);
 int migrate_multifd_zlib_level(void);
 int migrate_multifd_zstd_level(void);
 
+#ifdef CONFIG_LINUX
+bool migrate_use_zero_copy_send(void);
+#else
+#define migrate_use_zero_copy_send() (false)
+#endif
 int migrate_use_xbzrle(void);
 uint64_t migrate_xbzrle_cache_size(void);
 bool migrate_colo_enabled(void);
migration/migration.c +32 -0
index 5a31b23bd6..3e91f4b5e2 100644
@@ -910,6 +910,10 @@ MigrationParameters *qmp_query_migrate_parameters(Error **errp)
     params->multifd_zlib_level = s->parameters.multifd_zlib_level;
     params->has_multifd_zstd_level = true;
     params->multifd_zstd_level = s->parameters.multifd_zstd_level;
+#ifdef CONFIG_LINUX
+    params->has_zero_copy_send = true;
+    params->zero_copy_send = s->parameters.zero_copy_send;
+#endif
     params->has_xbzrle_cache_size = true;
     params->xbzrle_cache_size = s->parameters.xbzrle_cache_size;
     params->has_max_postcopy_bandwidth = true;
@@ -1567,6 +1571,11 @@ static void migrate_params_test_apply(MigrateSetParameters *params,
     if (params->has_multifd_compression) {
         dest->multifd_compression = params->multifd_compression;
     }
+#ifdef CONFIG_LINUX
+    if (params->has_zero_copy_send) {
+        dest->zero_copy_send = params->zero_copy_send;
+    }
+#endif
     if (params->has_xbzrle_cache_size) {
         dest->xbzrle_cache_size = params->xbzrle_cache_size;
     }
@@ -1679,6 +1688,11 @@ static void migrate_params_apply(MigrateSetParameters *params, Error **errp)
     if (params->has_multifd_compression) {
         s->parameters.multifd_compression = params->multifd_compression;
     }
+#ifdef CONFIG_LINUX
+    if (params->has_zero_copy_send) {
+        s->parameters.zero_copy_send = params->zero_copy_send;
+    }
+#endif
     if (params->has_xbzrle_cache_size) {
         s->parameters.xbzrle_cache_size = params->xbzrle_cache_size;
         xbzrle_cache_resize(params->xbzrle_cache_size, errp);
@@ -2563,6 +2577,17 @@ int migrate_multifd_zstd_level(void)
     return s->parameters.multifd_zstd_level;
 }
 
+#ifdef CONFIG_LINUX
+bool migrate_use_zero_copy_send(void)
+{
+    MigrationState *s;
+
+    s = migrate_get_current();
+
+    return s->parameters.zero_copy_send;
+}
+#endif
+
 int migrate_use_xbzrle(void)
 {
     MigrationState *s;
@@ -4206,6 +4231,10 @@ static Property migration_properties[] = {
     DEFINE_PROP_UINT8("multifd-zstd-level", MigrationState,
                       parameters.multifd_zstd_level,
                       DEFAULT_MIGRATE_MULTIFD_ZSTD_LEVEL),
+#ifdef CONFIG_LINUX
+    DEFINE_PROP_BOOL("zero_copy_send", MigrationState,
+                      parameters.zero_copy_send, false),
+#endif
     DEFINE_PROP_SIZE("xbzrle-cache-size", MigrationState,
                       parameters.xbzrle_cache_size,
                       DEFAULT_MIGRATE_XBZRLE_CACHE_SIZE),
@@ -4303,6 +4332,9 @@ static void migration_instance_init(Object *obj)
     params->has_multifd_compression = true;
     params->has_multifd_zlib_level = true;
     params->has_multifd_zstd_level = true;
+#ifdef CONFIG_LINUX
+    params->has_zero_copy_send = true;
+#endif
     params->has_xbzrle_cache_size = true;
     params->has_max_postcopy_bandwidth = true;
     params->has_max_cpu_throttle = true;
migration/socket.c +9 -2
index 05705a32d8..3754d8f72c 100644
@@ -74,9 +74,16 @@ static void socket_outgoing_migration(QIOTask *task,
 
     if (qio_task_propagate_error(task, &err)) {
         trace_migration_socket_outgoing_error(error_get_pretty(err));
-    } else {
-        trace_migration_socket_outgoing_connected(data->hostname);
+           goto out;
     }
+
+    trace_migration_socket_outgoing_connected(data->hostname);
+
+    if (migrate_use_zero_copy_send()) {
+        error_setg(&err, "Zero copy send not available in migration");
+    }
+
+out:
     migration_channel_connect(data->s, sioc, data->hostname, err);
     object_unref(OBJECT(sioc));
 }
monitor/hmp-cmds.c +6 -0
index 634968498b..55b48d3733 100644
@@ -1309,6 +1309,12 @@ void hmp_migrate_set_parameter(Monitor *mon, const QDict *qdict)
         p->has_multifd_zstd_level = true;
         visit_type_uint8(v, param, &p->multifd_zstd_level, &err);
         break;
+#ifdef CONFIG_LINUX
+    case MIGRATION_PARAMETER_ZERO_COPY_SEND:
+        p->has_zero_copy_send = true;
+        visit_type_bool(v, param, &p->zero_copy_send, &err);
+        break;
+#endif
     case MIGRATION_PARAMETER_XBZRLE_CACHE_SIZE:
         p->has_xbzrle_cache_size = true;
         if (!visit_type_size(v, param, &cache_size, &err)) {
Signature · 1 line
2.36.0
Acked-by on v9 3/7
#4 lore raw

Re: [PATCH v9 3/7] migration: Add zero-copy-send parameter for QMP/HMP for Linux

Quoted 26 lines · Leonardo Bras

Add property that allows zero-copy migration of memory pages on the sending side, and also includes a helper function migrate_use_zero_copy_send() to check if it's enabled.

No code is introduced to actually do the migration, but it allow future implementations to enable/disable this feature.

On non-Linux builds this parameter is compiled-out.

---
 qapi/migration.json   | 24 ++++++++++++++++++++++++
 migration/migration.h |  5 +++++
 migration/migration.c | 32 ++++++++++++++++++++++++++++++++
 migration/socket.c    | 11 +++++++++--
 monitor/hmp-cmds.c    |  6 ++++++
 5 files changed, 76 insertions(+), 2 deletions(-)
qapi/migration.json +0 -0
index 409eb086a2..04246481ce 100644

qapi/migration.json line 748

Quoted 8 lines
qapi/migration.json +5 -0
@@ -741,6 +741,13 @@
 #                      will consume more CPU.
 #                      Defaults to 1. (Since 5.0)
 #
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+#                  When true, enables a zero-copy mechanism for sending memory
+#                  pages, if host supports it.
+#                  Requires that QEMU be permitted to use locked memory for guest
+#                  RAM pages.

Please wrap lines around column 75. More of the same below.

qapi/migration.json line 753

Quoted 5 lines
qapi/migration.json +2 -0
+#                  Defaults to false. (Since 7.1)
+#
 # @block-bitmap-mapping: Maps block nodes and bitmaps on them to
 #                        aliases for the purpose of dirty bitmap migration.  Such
 #                        aliases may for example be the corresponding names on the

With that, QAPI schema

[...]

#5 lore raw

Re: [PATCH v9 3/7] migration: Add zero-copy-send parameter for QMP/HMP for Linux

Quoted 40 lines · Markus Armbruster
Quoted 26 lines · Leonardo Bras

Add property that allows zero-copy migration of memory pages on the sending side, and also includes a helper function migrate_use_zero_copy_send() to check if it's enabled.

No code is introduced to actually do the migration, but it allow future implementations to enable/disable this feature.

On non-Linux builds this parameter is compiled-out.

---
 qapi/migration.json   | 24 ++++++++++++++++++++++++
 migration/migration.h |  5 +++++
 migration/migration.c | 32 ++++++++++++++++++++++++++++++++
 migration/socket.c    | 11 +++++++++--
 monitor/hmp-cmds.c    |  6 ++++++
 5 files changed, 76 insertions(+), 2 deletions(-)
qapi/migration.json +0 -0
index 409eb086a2..04246481ce 100644

qapi/migration.json line 748

Quoted 8 lines
qapi/migration.json +5 -0
@@ -741,6 +741,13 @@
 #                      will consume more CPU.
 #                      Defaults to 1. (Since 5.0)
 #
+# @zero-copy-send: Controls behavior on sending memory pages on migration.
+#                  When true, enables a zero-copy mechanism for sending memory
+#                  pages, if host supports it.
+#                  Requires that QEMU be permitted to use locked memory for guest
+#                  RAM pages.

Please wrap lines around column 75. More of the same below.

Quoted 8 lines

qapi/migration.json line 753

Quoted 5 lines
qapi/migration.json +2 -0
+#                  Defaults to false. (Since 7.1)
+#
 # @block-bitmap-mapping: Maps block nodes and bitmaps on them to
 #                        aliases for the purpose of dirty bitmap migration.  Such
 #                        aliases may for example be the corresponding names on the

With that, QAPI schema

Thanks Markus!

Best regards,
Leo

Quoted 3 lines

[...]

#6 lore raw

PATCH v9 4/7 migration: Add migrate_use_tls() helper

In reply to Leonardo Bras

A lot of places check parameters.tls_creds in order to evaluate if TLS is in use, and sometimes call migrate_get_current() just for that test.

Add new helper function migrate_use_tls() in order to simplify testing for TLS usage.

---
 migration/migration.h | 1 +
 migration/channel.c   | 3 +--
 migration/migration.c | 9 +++++++++
 migration/multifd.c   | 5 +----
 4 files changed, 12 insertions(+), 6 deletions(-)
migration/migration.h +1 -0
index e8f2941a55..485d58b95f 100644
@@ -380,6 +380,7 @@ bool migrate_use_zero_copy_send(void);
 #else
 #define migrate_use_zero_copy_send() (false)
 #endif
+int migrate_use_tls(void);
 int migrate_use_xbzrle(void);
 uint64_t migrate_xbzrle_cache_size(void);
 bool migrate_colo_enabled(void);
migration/channel.c +1 -2
index c6a8dcf1d7..a162d00fea 100644
@@ -38,8 +38,7 @@ void migration_channel_process_incoming(QIOChannel *ioc)
     trace_migration_set_incoming_channel(
         ioc, object_get_typename(OBJECT(ioc)));
 
-    if (s->parameters.tls_creds &&
-        *s->parameters.tls_creds &&
+    if (migrate_use_tls() &&
         !object_dynamic_cast(OBJECT(ioc),
                              TYPE_QIO_CHANNEL_TLS)) {
         migration_tls_channel_process_incoming(s, ioc, &local_err);
migration/migration.c +9 -0
index 3e91f4b5e2..4b6df2eb5e 100644
@@ -2588,6 +2588,15 @@ bool migrate_use_zero_copy_send(void)
 }
 #endif
 
+int migrate_use_tls(void)
+{
+    MigrationState *s;
+
+    s = migrate_get_current();
+
+    return s->parameters.tls_creds && *s->parameters.tls_creds;
+}
+
 int migrate_use_xbzrle(void)
 {
     MigrationState *s;
migration/multifd.c +1 -4
index 9ea4f581e2..2a8c8570c3 100644
@@ -782,15 +782,12 @@ static bool multifd_channel_connect(MultiFDSendParams *p,
                                     QIOChannel *ioc,
                                     Error *error)
 {
-    MigrationState *s = migrate_get_current();
-
     trace_multifd_set_outgoing_channel(
         ioc, object_get_typename(OBJECT(ioc)),
         migrate_get_current()->hostname, error);
 
     if (!error) {
-        if (s->parameters.tls_creds &&
-            *s->parameters.tls_creds &&
+        if (migrate_use_tls() &&
             !object_dynamic_cast(OBJECT(ioc),
                                  TYPE_QIO_CHANNEL_TLS)) {
             multifd_tls_channel_connect(p, ioc, &error);
Signature · 1 line
2.36.0
#7 lore raw

PATCH v9 5/7 multifd: multifd_send_sync_main now returns negative on error

In reply to Leonardo Bras

Even though multifd_send_sync_main() currently emits error_reports, it's callers don't really check it before continuing.

Change multifd_send_sync_main() to return -1 on error and 0 on success. Also change all it's callers to make use of this change and possibly fail earlier.

(This change is important to next patch on multifd zero copy implementation, to make it sure an error in zero-copy flush does not go unnoticed.

---
 migration/multifd.h |  2 +-
 migration/multifd.c | 10 ++++++----
 migration/ram.c     | 29 ++++++++++++++++++++++-------
 3 files changed, 29 insertions(+), 12 deletions(-)
migration/multifd.h +1 -1
index 7d0effcb03..bcf5992945 100644
@@ -20,7 +20,7 @@ int multifd_load_cleanup(Error **errp);
 bool multifd_recv_all_channels_created(void);
 bool multifd_recv_new_channel(QIOChannel *ioc, Error **errp);
 void multifd_recv_sync_main(void);
-void multifd_send_sync_main(QEMUFile *f);
+int multifd_send_sync_main(QEMUFile *f);
 int multifd_queue_page(QEMUFile *f, RAMBlock *block, ram_addr_t offset);
 
 /* Multifd Compression flags */
migration/multifd.c +6 -4
index 2a8c8570c3..15fb668e64 100644
@@ -566,17 +566,17 @@ void multifd_save_cleanup(void)
     multifd_send_state = NULL;
 }
 
-void multifd_send_sync_main(QEMUFile *f)
+int multifd_send_sync_main(QEMUFile *f)
 {
     int i;
 
     if (!migrate_use_multifd()) {
-        return;
+        return 0;
     }
     if (multifd_send_state->pages->num) {
         if (multifd_send_pages(f) < 0) {
             error_report("%s: multifd_send_pages fail", __func__);
-            return;
+            return -1;
         }
     }
     for (i = 0; i < migrate_multifd_channels(); i++) {
@@ -589,7 +589,7 @@ void multifd_send_sync_main(QEMUFile *f)
         if (p->quit) {
             error_report("%s: channel %d has already quit", __func__, i);
             qemu_mutex_unlock(&p->mutex);
-            return;
+            return -1;
         }
 
         p->packet_num = multifd_send_state->packet_num++;
@@ -608,6 +608,8 @@ void multifd_send_sync_main(QEMUFile *f)
         qemu_sem_wait(&p->sem_sync);
     }
     trace_multifd_send_sync_main(multifd_send_state->packet_num);
+
+    return 0;
 }
 
 static void *multifd_send_thread(void *opaque)
migration/ram.c +22 -7
index a2489a2699..5f5e37f64d 100644
@@ -2909,6 +2909,7 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
 {
     RAMState **rsp = opaque;
     RAMBlock *block;
+    int ret;
 
     if (compress_threads_save_setup()) {
         return -1;
@@ -2943,7 +2944,11 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
     ram_control_before_iterate(f, RAM_CONTROL_SETUP);
     ram_control_after_iterate(f, RAM_CONTROL_SETUP);
 
-    multifd_send_sync_main(f);
+    ret =  multifd_send_sync_main(f);
+    if (ret < 0) {
+        return ret;
+    }
+
     qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
     qemu_fflush(f);
 
@@ -3052,7 +3057,11 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
 out:
     if (ret >= 0
         && migration_is_setup_or_active(migrate_get_current()->state)) {
-        multifd_send_sync_main(rs->f);
+        ret = multifd_send_sync_main(rs->f);
+        if (ret < 0) {
+            return ret;
+        }
+
         qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
         qemu_fflush(f);
         ram_transferred_add(8);
@@ -3112,13 +3121,19 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
         ram_control_after_iterate(f, RAM_CONTROL_FINISH);
     }
 
-    if (ret >= 0) {
-        multifd_send_sync_main(rs->f);
-        qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
-        qemu_fflush(f);
+    if (ret < 0) {
+        return ret;
     }
 
-    return ret;
+    ret = multifd_send_sync_main(rs->f);
+    if (ret < 0) {
+        return ret;
+    }
+
+    qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
+    qemu_fflush(f);
+
+    return 0;
 }
 
 static void ram_save_pending(QEMUFile *f, void *opaque, uint64_t max_size,
Signature · 1 line
2.36.0
Reviewed-by on v9 5/7
#8 lore raw

Re: [PATCH v9 5/7] multifd: multifd_send_sync_main now returns negative on error

Quoted 11 lines · Leonardo Bras

Even though multifd_send_sync_main() currently emits error_reports, it's callers don't really check it before continuing.

Change multifd_send_sync_main() to return -1 on error and 0 on success. Also change all it's callers to make use of this change and possibly fail earlier.

(This change is important to next patch on multifd zero copy implementation, to make it sure an error in zero-copy flush does not go unnoticed.

Quoted 6 lines
---
 migration/multifd.h |  2 +-
 migration/multifd.c | 10 ++++++----
 migration/ram.c     | 29 ++++++++++++++++++++++-------
 3 files changed, 29 insertions(+), 12 deletions(-)

With regards,
Daniel

Signature · 3 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
Reviewed-by on v9 5/7
#9 lore raw

Re: [PATCH v9 5/7] multifd: multifd_send_sync_main now returns negative on error

In reply to Leonardo Bras
Quoted 7 lines · Leonardo Bras

Even though multifd_send_sync_main() currently emits error_reports, it's callers don't really check it before continuing.

Change multifd_send_sync_main() to return -1 on error and 0 on success. Also change all it's callers to make use of this change and possibly fail earlier.

Quoted 5 lines

(This change is important to next patch on multifd zero copy implementation, to make it sure an error in zero-copy flush does not go unnoticed.

Signature · 1 line
Peter Xu
#10 lore raw

Re: [PATCH v9 5/7] multifd: multifd_send_sync_main now returns negative on error

Quoted 13 lines · Peter Xu
Quoted 11 lines · Leonardo Bras

Even though multifd_send_sync_main() currently emits error_reports, it's callers don't really check it before continuing.

Change multifd_send_sync_main() to return -1 on error and 0 on success. Also change all it's callers to make use of this change and possibly fail earlier.

(This change is important to next patch on multifd zero copy implementation, to make it sure an error in zero-copy flush does not go unnoticed.

Quoted 7 lines
Quoted 1 line

--
Peter Xu

Thanks Peter!

Best regards,
Leo

#11 lore raw

PATCH v9 6/7 multifd: Send header packet without flags if zero-copy-send is enabled

In reply to Leonardo Bras

Since d48c3a0445 ("multifd: Use a single writev on the send side"), sending the header packet and the memory pages happens in the same writev, which can potentially make the migration faster.

Using channel-socket as example, this works well with the default copying mechanism of sendmsg(), but with zero-copy-send=true, it will cause the migration to often break.

This happens because the header packet buffer gets reused quite often, and there is a high chance that by the time the MSG_ZEROCOPY mechanism get to send the buffer, it has already changed, sending the wrong data and causing the migration to abort.

It means that, as it is, the buffer for the header packet is not suitable for sending with MSG_ZEROCOPY.

In order to enable zero copy for multifd, send the header packet on an individual write(), without any flags, and the remanining pages with a writev(), as it was happening before. This only changes how a migration with zero-copy-send=true works, not changing any current behavior for migrations with zero-copy-send=false.

---
 migration/multifd.c | 29 ++++++++++++++++++++++++++---
 1 file changed, 26 insertions(+), 3 deletions(-)
migration/multifd.c +26 -3
index 15fb668e64..6c940aaa98 100644
@@ -639,6 +639,8 @@ static void *multifd_send_thread(void *opaque)
         if (p->pending_job) {
             uint64_t packet_num = p->packet_num;
             uint32_t flags = p->flags;
+            int iov_offset = 0;
+
             p->iovs_num = 1;
             p->normal_num = 0;
 
@@ -665,15 +667,36 @@ static void *multifd_send_thread(void *opaque)
             trace_multifd_send(p->id, packet_num, p->normal_num, flags,
                                p->next_packet_size);
 
-            p->iov[0].iov_len = p->packet_len;
-            p->iov[0].iov_base = p->packet;
+            if (migrate_use_zero_copy_send()) {
+                /* Send header without zerocopy */
+                ret = qio_channel_write_all(p->c, (void *)p->packet,
+                                            p->packet_len, &local_err);
+                if (ret != 0) {
+                    break;
+                }
+
+                if (!p->normal_num) {
+                    /* No pages will be sent */
+                    goto skip_send;
+                }
 
-            ret = qio_channel_writev_all(p->c, p->iov, p->iovs_num,
+                /* Skip first iov : header */
+                iov_offset = 1;
+            } else {
+                /* Send header using the same writev call */
+                p->iov[0].iov_len = p->packet_len;
+                p->iov[0].iov_base = p->packet;
+            }
+
+            ret = qio_channel_writev_all(p->c, p->iov + iov_offset,
+                                         p->iovs_num - iov_offset,
                                          &local_err);
+
             if (ret != 0) {
                 break;
             }
 
+skip_send:
             qemu_mutex_lock(&p->mutex);
             p->pending_job--;
             qemu_mutex_unlock(&p->mutex);
Signature · 1 line
2.36.0
#12 lore raw

Re: [PATCH v9 6/7] multifd: Send header packet without flags if zero-copy-send is enabled

Quoted 32 lines · Leonardo Bras

Since d48c3a0445 ("multifd: Use a single writev on the send side"), sending the header packet and the memory pages happens in the same writev, which can potentially make the migration faster.

Using channel-socket as example, this works well with the default copying mechanism of sendmsg(), but with zero-copy-send=true, it will cause the migration to often break.

This happens because the header packet buffer gets reused quite often, and there is a high chance that by the time the MSG_ZEROCOPY mechanism get to send the buffer, it has already changed, sending the wrong data and causing the migration to abort.

It means that, as it is, the buffer for the header packet is not suitable for sending with MSG_ZEROCOPY.

In order to enable zero copy for multifd, send the header packet on an individual write(), without any flags, and the remanining pages with a writev(), as it was happening before. This only changes how a migration with zero-copy-send=true works, not changing any current behavior for migrations with zero-copy-send=false.

---
 migration/multifd.c | 29 ++++++++++++++++++++++++++---
 1 file changed, 26 insertions(+), 3 deletions(-)
migration/multifd.c +0 -0
index 15fb668e64..6c940aaa98 100644

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 643

Quoted 5 lines
migration/multifd.c +2 -0
@@ -639,6 +639,8 @@ static void *multifd_send_thread(void *opaque)
         if (p->pending_job) {
             uint64_t packet_num = p->packet_num;
             uint32_t flags = p->flags;
+            int iov_offset = 0;
+

No need for this if you change:

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 644

Quoted 1 line
migration/multifd.c +0 -0
             p->iovs_num = 1;
   if (!migrate_use_zero_copy_send()) {
      p->iovs_num = 1;
   }
Quoted 10 lines
migration/multifd.c +2 -2
             p->normal_num = 0;
 
@@ -665,15 +667,36 @@ static void *multifd_send_thread(void *opaque)
             trace_multifd_send(p->id, packet_num, p->normal_num, flags,
                                p->next_packet_size);
 
-            p->iov[0].iov_len = p->packet_len;
-            p->iov[0].iov_base = p->packet;
+            if (migrate_use_zero_copy_send()) {
+                /* Send header without zerocopy */

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 681

Quoted 10 lines
migration/multifd.c +10 -0
+                ret = qio_channel_write_all(p->c, (void *)p->packet,
+                                            p->packet_len, &local_err);
+                if (ret != 0) {
+                    break;
+                }
+
+                if (!p->normal_num) {
+                    /* No pages will be sent */
+                    goto skip_send;
+                }

Don't need this AFAIK, because the qio_channel_writev_all call will be a no-op if iovs_num is zero

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 684

Quoted 4 lines
migration/multifd.c +2 -1
-            ret = qio_channel_writev_all(p->c, p->iov, p->iovs_num,
+                /* Skip first iov : header */
+                iov_offset = 1;

Don't need to set this

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 693

Quoted 9 lines
migration/multifd.c +8 -0
+            } else {
+                /* Send header using the same writev call */
+                p->iov[0].iov_len = p->packet_len;
+                p->iov[0].iov_base = p->packet;
+            }
+
+            ret = qio_channel_writev_all(p->c, p->iov + iov_offset,
+                                         p->iovs_num - iov_offset,
                                          &local_err);

This wouldn't need changing if we don't reserve iovs[0] when not required.

Quoted 12 lines
migration/multifd.c +2 -0
+
             if (ret != 0) {
                 break;
             }
 
+skip_send:
             qemu_mutex_lock(&p->mutex);
             p->pending_job--;
             qemu_mutex_unlock(&p->mutex);
Signature · 1 line
2.36.0

With regards,
Daniel

Signature · 3 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
#13 lore raw

Re: [PATCH v9 6/7] multifd: Send header packet without flags if zero-copy-send is enabled

Hello Daniel, thank you for the feedback!

Quoted 42 lines · Daniel P. Berrangé
Quoted 32 lines · Leonardo Bras

Since d48c3a0445 ("multifd: Use a single writev on the send side"), sending the header packet and the memory pages happens in the same writev, which can potentially make the migration faster.

Using channel-socket as example, this works well with the default copying mechanism of sendmsg(), but with zero-copy-send=true, it will cause the migration to often break.

This happens because the header packet buffer gets reused quite often, and there is a high chance that by the time the MSG_ZEROCOPY mechanism get to send the buffer, it has already changed, sending the wrong data and causing the migration to abort.

It means that, as it is, the buffer for the header packet is not suitable for sending with MSG_ZEROCOPY.

In order to enable zero copy for multifd, send the header packet on an individual write(), without any flags, and the remanining pages with a writev(), as it was happening before. This only changes how a migration with zero-copy-send=true works, not changing any current behavior for migrations with zero-copy-send=false.

---
 migration/multifd.c | 29 ++++++++++++++++++++++++++---
 1 file changed, 26 insertions(+), 3 deletions(-)
migration/multifd.c +0 -0
index 15fb668e64..6c940aaa98 100644

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 643

Quoted 5 lines
migration/multifd.c +2 -0
@@ -639,6 +639,8 @@ static void *multifd_send_thread(void *opaque)
         if (p->pending_job) {
             uint64_t packet_num = p->packet_num;
             uint32_t flags = p->flags;
+            int iov_offset = 0;
+

No need for this if you change:

Quoted 6 lines

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 644

Quoted 1 line
migration/multifd.c +0 -0
             p->iovs_num = 1;
   if (!migrate_use_zero_copy_send()) {
      p->iovs_num = 1;
   }

I understand the point now: setting p->iovs_num = 0 before multifd_send_state->ops->send_prepare() causes p->iov[0] to be used for pages instead of the header. I was not aware, so thanks for pointing that out!

But it's also necessary to have an else clause with p->iovs_num = 0, right? It seems like the variable is not set anywhere else, and it would keep growing after the second loop iteration, causing prepare() to access p->iov[] outside bounds.

Am I missing something here?

Quoted 16 lines
Quoted 16 lines
migration/multifd.c +7 -2
             p->normal_num = 0;
 
@@ -665,15 +667,36 @@ static void *multifd_send_thread(void *opaque)
             trace_multifd_send(p->id, packet_num, p->normal_num, flags,
                                p->next_packet_size);
 
-            p->iov[0].iov_len = p->packet_len;
-            p->iov[0].iov_base = p->packet;
+            if (migrate_use_zero_copy_send()) {
+                /* Send header without zerocopy */
+                ret = qio_channel_write_all(p->c, (void *)p->packet,
+                                            p->packet_len, &local_err);
+                if (ret != 0) {
+                    break;
+                }
Quoted 9 lines

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 681

Quoted 5 lines
migration/multifd.c +5 -0
+
+                if (!p->normal_num) {
+                    /* No pages will be sent */
+                    goto skip_send;
+                }

Don't need this AFAIK, because the qio_channel_writev_all call will be a no-op if iovs_num is zero

Oh, I see:
qio_channel_writev_all() will call qio_channel_writev_full_all() where niov == 0 and thus nlocal_iov == 0, avoiding the loop that calls qio_channel_writev_full().

I will remove that in v10

Quoted 6 lines

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 684

Quoted 4 lines
migration/multifd.c +2 -1
-            ret = qio_channel_writev_all(p->c, p->iov, p->iovs_num,
+                /* Skip first iov : header */
+                iov_offset = 1;

Don't need to set this

Agree, that makes sense since the offset part is discontinued.

Quoted 11 lines

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 693

Quoted 10 lines
migration/multifd.c +8 -0
+            } else {
+                /* Send header using the same writev call */
+                p->iov[0].iov_len = p->packet_len;
+                p->iov[0].iov_base = p->packet;
+            }
+
+            ret = qio_channel_writev_all(p->c, p->iov + iov_offset,
+                                         p->iovs_num - iov_offset,
                                          &local_err);
Quoted 2 lines

This wouldn't need changing if we don't reserve iovs[0] when not required.

Agree.

Quoted 17 lines
Quoted 11 lines
migration/multifd.c +2 -1
+
             if (ret != 0) {
                 break;
             }
 
+skip_send:
             qemu_mutex_lock(&p->mutex);
             p->pending_job--;
             qemu_mutex_unlock(&p->mutex);
--
Quoted 2 lines

2.36.0

With regards,
Daniel
--

Quoted 4 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|

I will probably send a v10 shortly.

Thanks for reviewing!

Best regards,
Leo

#14 lore raw

PATCH v9 7/7 multifd: Implement zero copy write in multifd migration (multifd-zero-copy)

In reply to Leonardo Bras

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.

---
 migration/multifd.h   |  2 ++
 migration/migration.c | 11 ++++++++++-
 migration/multifd.c   | 34 ++++++++++++++++++++++++++++++----
 migration/socket.c    |  5 +++--
 4 files changed, 45 insertions(+), 7 deletions(-)
migration/multifd.h +2 -0
index bcf5992945..4d8d89e5e5 100644
@@ -92,6 +92,8 @@ 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 */
migration/migration.c +10 -1
index 4b6df2eb5e..31739b2af9 100644
@@ -1497,7 +1497,16 @@ 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;
 }
 
migration/multifd.c +30 -4
index 6c940aaa98..e37cc6e0d9 100644
@@ -569,6 +569,7 @@ void multifd_save_cleanup(void)
 int multifd_send_sync_main(QEMUFile *f)
 {
     int i;
+    bool flush_zero_copy;
 
     if (!migrate_use_multifd()) {
         return 0;
@@ -579,6 +580,14 @@ int multifd_send_sync_main(QEMUFile *f)
             return -1;
         }
     }
+
+    /*
+     * When using zero-copy, it's necessary to flush after each iteration to
+     * make sure pages from earlier iterations don't end up replacing newer
+     * pages.
+     */
+    flush_zero_copy = migrate_use_zero_copy_send();
+
     for (i = 0; i < migrate_multifd_channels(); i++) {
         MultiFDSendParams *p = &multifd_send_state->params[i];
 
@@ -600,6 +609,17 @@ 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];
@@ -688,10 +708,9 @@ static void *multifd_send_thread(void *opaque)
                 p->iov[0].iov_base = p->packet;
             }
 
-            ret = qio_channel_writev_all(p->c, p->iov + iov_offset,
-                                         p->iovs_num - iov_offset,
-                                         &local_err);
-
+            ret = qio_channel_writev_full_all(p->c, p->iov + iov_offset,
+                                              p->iovs_num - iov_offset, NULL,
+                                              0, p->write_flags, &local_err);
             if (ret != 0) {
                 break;
             }
@@ -920,6 +939,13 @@ 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);
     }
 
migration/socket.c +3 -2
index 3754d8f72c..4fd5e85f50 100644
@@ -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:
Signature · 1 line
2.36.0
#15 lore raw

Re: [PATCH v9 7/7] multifd: Implement zero copy write in multifd migration (multifd-zero-copy)

Leo,

This patch looks mostly good to me, a few nitpicks below.

Quoted 77 lines · Leonardo Bras

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.

---
 migration/multifd.h   |  2 ++
 migration/migration.c | 11 ++++++++++-
 migration/multifd.c   | 34 ++++++++++++++++++++++++++++++----
 migration/socket.c    |  5 +++--
 4 files changed, 45 insertions(+), 7 deletions(-)
migration/multifd.h +2 -0
index bcf5992945..4d8d89e5e5 100644
@@ -92,6 +92,8 @@ 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 */
migration/migration.c +10 -1
index 4b6df2eb5e..31739b2af9 100644
@@ -1497,7 +1497,16 @@ 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;
 }
 
migration/multifd.c +1 -0
index 6c940aaa98..e37cc6e0d9 100644
@@ -569,6 +569,7 @@ void multifd_save_cleanup(void)
 int multifd_send_sync_main(QEMUFile *f)
 {
     int i;
+    bool flush_zero_copy;
 
     if (!migrate_use_multifd()) {
         return 0;

migration/multifd.c int multifd_send_sync_main(QEMUFile *f) · line 589

Quoted 10 lines
migration/multifd.c +7 -0
@@ -579,6 +580,14 @@ int multifd_send_sync_main(QEMUFile *f)
             return -1;
         }
     }
+
+    /*
+     * When using zero-copy, it's necessary to flush after each iteration to
+     * make sure pages from earlier iterations don't end up replacing newer
+     * pages.
+     */
+    flush_zero_copy = migrate_use_zero_copy_send();

Would you mind inline it if it's only used once?

It's great to have that comment, but IMHO it could be more explicit, even marking a TODO showing that maybe we could do better in the future:

/*
* 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.
*/

Quoted 23 lines
migration/multifd.c +12 -0
+
     for (i = 0; i < migrate_multifd_channels(); i++) {
         MultiFDSendParams *p = &multifd_send_state->params[i];
 
@@ -600,6 +609,17 @@ 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];

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 713

Quoted 10 lines
migration/multifd.c +3 -4
@@ -688,10 +708,9 @@ static void *multifd_send_thread(void *opaque)
                 p->iov[0].iov_base = p->packet;
             }
 
-            ret = qio_channel_writev_all(p->c, p->iov + iov_offset,
-                                         p->iovs_num - iov_offset,
-                                         &local_err);
-
+            ret = qio_channel_writev_full_all(p->c, p->iov + iov_offset,
+                                              p->iovs_num - iov_offset, NULL,
+                                              0, p->write_flags, &local_err);

I kind of agree with Dan in previous patch - this iov_offset is confusing, better drop it.

Quoted 36 lines
migration/multifd.c +7 -0
             if (ret != 0) {
                 break;
             }
@@ -920,6 +939,13 @@ 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);
     }
 
migration/socket.c +3 -2
index 3754d8f72c..4fd5e85f50 100644
@@ -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:
Signature · 1 line
2.36.0
Signature · 1 line
Peter Xu
#16 lore raw

Re: [PATCH v9 7/7] multifd: Implement zero copy write in multifd migration (multifd-zero-copy)

Hello Peter, thanks for helping!

Quoted 6 lines · Peter Xu

Leo,

This patch looks mostly good to me, a few nitpicks below.

On Mon, Apr 25, 2022 at 06:50:56PM -0300, Leonardo Bras wrote:

[...]

Quoted 10 lines

migration/multifd.c int multifd_send_sync_main(QEMUFile *f) · line 589

Quoted 8 lines
     }
migration/multifd.c +7 -0
+
+    /*
+     * When using zero-copy, it's necessary to flush after each iteration to
+     * make sure pages from earlier iterations don't end up replacing newer
+     * pages.
+     */
+    flush_zero_copy = migrate_use_zero_copy_send();

Would you mind inline it if it's only used once?

It's not obvious in the diff, but this is used in a loop bellow, so I inserted the variable to avoid calling migrate_use_zero_copy_send() for each multifd channel.

Quoted 4 lines

It's great to have that comment, but IMHO it could be more explicit, even marking a TODO showing that maybe we could do better in the future:

Quoted 11 lines

/*
* 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.
*/

Thanks! I will insert that in the next version.

The thing here is that I was under the impression an iteration was equivalent to a whole scanning of all the dirty bitmaps. I see now that it may not be the case.

[...]

Quoted 4 lines
Quoted 4 lines
migration/multifd.c +0 -0
@@ -688,10 +708,9 @@ static void *multifd_send_thread(void *opaque)
                 p->iov[0].iov_base = p->packet;
             }
 
Quoted 10 lines

migration/multifd.c static void *multifd_send_thread(void *opaque) · line 713

Quoted 7 lines
migration/multifd.c +3 -4
-            ret = qio_channel_writev_all(p->c, p->iov + iov_offset,
-                                         p->iovs_num - iov_offset,
-                                         &local_err);
-
+            ret = qio_channel_writev_full_all(p->c, p->iov + iov_offset,
+                                              p->iovs_num - iov_offset, NULL,
+                                              0, p->write_flags, &local_err);

I kind of agree with Dan in previous patch - this iov_offset is confusing, better drop it.

Sure, fixed for v10.

Quoted 1 line

[...]

Quoted 3 lines

--
Peter Xu

Best regards,
Leo