MSG_ZEROCOPY + multifd
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
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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
@@ -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
PATCH v9 2/7 QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX
In reply to Leonardo BrasFor 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
@@ -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
PATCH v9 3/7 migration: Add zero-copy-send parameter for QMP/HMP for Linux
In reply to Leonardo BrasAdd 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
@@ -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
@@ -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
@@ -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), +#endifDEFINE_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
@@ -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
@@ -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
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.
Signed-off-by Leonardo Bras <[email protected]>Reviewed-by Peter Xu <[email protected]>Reviewed-by Daniel P. Berrangé <[email protected]>Reviewed-by Juan Quintela <[email protected]>--- 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
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
[...]
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.
Signed-off-by Leonardo Bras <[email protected]>Reviewed-by Peter Xu <[email protected]>Reviewed-by Daniel P. Berrangé <[email protected]>Reviewed-by Juan Quintela <[email protected]>--- 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
-0qapi/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 theWith that, QAPI schema
Thanks Markus!
Best regards,
Leo
Quoted 3 lines
[...]
PATCH v9 4/7 migration: Add migrate_use_tls() helper
In reply to Leonardo BrasA 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
@@ -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
@@ -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
@@ -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
@@ -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
PATCH v9 5/7 multifd: multifd_send_sync_main now returns negative on error
In reply to Leonardo BrasEven 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
@@ -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
@@ -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
@@ -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
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 emitserror_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
Signed-off-by Leonardo Bras <[email protected]>--- 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 :|
Re: [PATCH v9 5/7] multifd: multifd_send_sync_main now returns negative on error
In reply to Leonardo BrasQuoted 7 lines · Leonardo Bras
Even though
multifd_send_sync_main()currently emitserror_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.
Signed-off-by Leonardo Bras <[email protected]>
Signature · 1 line
Peter Xu
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 emitserror_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
Signed-off-by Leonardo Bras <[email protected]>Reviewed-by Peter Xu <[email protected]>--
Peter Xu
Thanks Peter!
Best regards,
Leo
PATCH v9 6/7 multifd: Send header packet without flags if zero-copy-send is enabled
In reply to Leonardo BrasSince 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
@@ -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
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_ZEROCOPYmechanism 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 awritev(), 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.Signed-off-by Leonardo Bras <[email protected]>--- migration/multifd.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-)migration/multifd.c +0
-0
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
-0p->iovs_num =1 ;
if (!migrate_use_zero_copy_send()) {
p->iovs_num = 1;
}Quoted 10 lines
migration/multifd.c +2
-2p->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 :|
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_ZEROCOPYmechanism 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 awritev(), 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.Signed-off-by Leonardo Bras <[email protected]>--- migration/multifd.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-)migration/multifd.c +0
-0migration/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
-0p->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
-2p->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_allcall will be a no-op ifiovs_numis 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
PATCH v9 7/7 multifd: Implement zero copy write in multifd migration (multifd-zero-copy)
In reply to Leonardo BrasImplement 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
@@ -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
@@ -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
@@ -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
@@ -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
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()soflush_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 <[email protected]>--- 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@@ -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@@ -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@@ -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
-0if (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@@ -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
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_offsetis confusing, better drop it.
Sure, fixed for v10.
Quoted 1 line
[...]
Quoted 3 lines
--
Peter Xu
Best regards,
Leo