[Qemu-devel] [PULL 0/5] migration queue

[Qemu-devel] [PULL 0/5] migration queue

[Juan Quintela]

Hi

This includes xbzrle fixes.  They are just:
- refactor xbzrle into its own file
- tests for xbzrle !!!!
- fix documentanion
- bugfix
- fix error message

please pull

The following changes since commit 8a55ebf01507ab73cc458cfcd5b9cb856aba0b9e:

  Merge remote-tracking branch 'afaerber/qom-cpu' into staging (2013-01-31 19:37:33 -0600)

are available in the git repository at:


  git://repo.or.cz/qemu/quintela.git migration.next

for you to fetch changes up to a31ca017aaf9074c1bb636b3ddaceb40f994375f:

  Fix error message in migrate_set_capability HMP command (2013-02-01 08:32:21 +0100)

----------------------------------------------------------------
Orit Wasserman (5):
      Move XBZRLE encoding code to a separate file to allow testing
      Add XBZRLE testing
      Fix example for query-migrate-capabilities
      Allow XBZRLE decoding without enabling the capability
      Fix error message in migrate_set_capability HMP command

 Makefile.objs       |   2 +-
 arch_init.c         |   3 -
 hmp.c               |   2 +-
 qmp-commands.hx     |   6 +-
 savevm.c            | 159 ------------------------------------------
 tests/Makefile      |   3 +
 tests/test-xbzrle.c | 196 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 xbzrle.c            | 173 ++++++++++++++++++++++++++++++++++++++++++++++
 8 files changed, 376 insertions(+), 168 deletions(-)
 create mode 100644 tests/test-xbzrle.c
 create mode 100644 xbzrle.c

[Qemu-devel] [PULL 0/5] migration queue

[Juan Quintela]

Hi

Please pull

- Fix missmerge of fwrite patch (armbru)
- FIX XBZRLE crash: Gonglei
- Add more traces for migration (Alexey)

Especial thanks to Amit for getting the patches together.

Later, Juan.


The following changes since commit 6fc0303b95c873d9e384d7fb51e412ac2e53b9c1:

  Merge remote-tracking branch 'remotes/kraxel/tags/pull-input-4' into staging (2014-03-07 18:29:33 +0000)

are available in the git repository at:


  git://github.com/juanquintela/qemu.git tags/migration/20140308

for you to fetch changes up to 849defaa967cd9e20f4a3d426c8c23887b0e2305:

  migration: add more traces (2014-03-08 02:05:53 +0100)

----------------------------------------------------------------
migration/next for 20140308

----------------------------------------------------------------
Alexey Kardashevskiy (3):
      vl: add system_wakeup_request tracepoint
      migration: extend section_start/end traces
      migration: add more traces

Gonglei (1):
      XBZRLE: Fix qemu crash when resize the xbzrle cache

Markus Armbruster (1):
      qemu_file: Fix mismerge of "use fwrite() correctly"

 arch_init.c  | 52 +++++++++++++++++++++++++++++++++++++++++++++++++---
 migration.c  | 30 ++++++------------------------
 qemu-file.c  |  4 +++-
 savevm.c     | 22 ++++++++++++++++------
 trace-events | 21 +++++++++++++++++++--
 vl.c         |  2 ++
 vmstate.c    |  2 ++
 7 files changed, 97 insertions(+), 36 deletions(-)

[Qemu-devel] [PATCH 1/5] XBZRLE: Fix qemu crash when resize the xbzrle cache

[Juan Quintela]

From: Gonglei <arei.gonglei@huawei.com>

Resizing the xbzrle cache during migration causes qemu-crash,
because the main-thread and migration-thread modify the xbzrle
cache size concurrently without lock-protection.

Signed-off-by: ChenLiang <chenliang88@huawei.com>
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 arch_init.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 49 insertions(+), 3 deletions(-)

diff --git a/arch_init.c b/arch_init.c
index fe17279..60c975d 100644
--- a/arch_init.c
+++ b/arch_init.c
@@ -164,8 +164,9 @@ static struct {
     uint8_t *encoded_buf;
     /* buffer for storing page content */
     uint8_t *current_buf;
-    /* Cache for XBZRLE */
+    /* Cache for XBZRLE, Protected by lock. */
     PageCache *cache;
+    QemuMutex lock;
 } XBZRLE = {
     .encoded_buf = NULL,
     .current_buf = NULL,
@@ -174,16 +175,52 @@ static struct {
 /* buffer used for XBZRLE decoding */
 static uint8_t *xbzrle_decoded_buf;

+static void XBZRLE_cache_lock(void)
+{
+    if (migrate_use_xbzrle())
+        qemu_mutex_lock(&XBZRLE.lock);
+}
+
+static void XBZRLE_cache_unlock(void)
+{
+    if (migrate_use_xbzrle())
+        qemu_mutex_unlock(&XBZRLE.lock);
+}
+
 int64_t xbzrle_cache_resize(int64_t new_size)
 {
+    PageCache *new_cache, *cache_to_free;
+
     if (new_size < TARGET_PAGE_SIZE) {
         return -1;
     }

+    /* no need to lock, the current thread holds qemu big lock */
     if (XBZRLE.cache != NULL) {
-        return cache_resize(XBZRLE.cache, new_size / TARGET_PAGE_SIZE) *
-            TARGET_PAGE_SIZE;
+        /* check XBZRLE.cache again later */
+        if (pow2floor(new_size) == migrate_xbzrle_cache_size()) {
+            return pow2floor(new_size);
+        }
+        new_cache = cache_init(new_size / TARGET_PAGE_SIZE,
+                                        TARGET_PAGE_SIZE);
+        if (!new_cache) {
+            DPRINTF("Error creating cache\n");
+            return -1;
+        }
+
+        XBZRLE_cache_lock();
+        /* the XBZRLE.cache may have be destroyed, check it again */
+        if (XBZRLE.cache != NULL) {
+            cache_to_free = XBZRLE.cache;
+            XBZRLE.cache = new_cache;
+        } else {
+            cache_to_free = new_cache;
+        }
+        XBZRLE_cache_unlock();
+
+        cache_fini(cache_to_free);
     }
+
     return pow2floor(new_size);
 }

@@ -539,6 +576,8 @@ static int ram_save_block(QEMUFile *f, bool last_stage)
             ret = ram_control_save_page(f, block->offset,
                                offset, TARGET_PAGE_SIZE, &bytes_sent);

+            XBZRLE_cache_lock();
+
             current_addr = block->offset + offset;
             if (ret != RAM_SAVE_CONTROL_NOT_SUPP) {
                 if (ret != RAM_SAVE_CONTROL_DELAYED) {
@@ -587,6 +626,7 @@ static int ram_save_block(QEMUFile *f, bool last_stage)
                 acct_info.norm_pages++;
             }

+            XBZRLE_cache_unlock();
             /* if page is unmodified, continue to the next */
             if (bytes_sent > 0) {
                 last_sent_block = block;
@@ -654,6 +694,7 @@ static void migration_end(void)
         migration_bitmap = NULL;
     }

+    XBZRLE_cache_lock();
     if (XBZRLE.cache) {
         cache_fini(XBZRLE.cache);
         g_free(XBZRLE.cache);
@@ -663,6 +704,7 @@ static void migration_end(void)
         XBZRLE.encoded_buf = NULL;
         XBZRLE.current_buf = NULL;
     }
+    XBZRLE_cache_unlock();
 }

 static void ram_migration_cancel(void *opaque)
@@ -693,13 +735,17 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
     dirty_rate_high_cnt = 0;

     if (migrate_use_xbzrle()) {
+        qemu_mutex_lock_iothread();
         XBZRLE.cache = cache_init(migrate_xbzrle_cache_size() /
                                   TARGET_PAGE_SIZE,
                                   TARGET_PAGE_SIZE);
         if (!XBZRLE.cache) {
+            qemu_mutex_unlock_iothread();
             DPRINTF("Error creating cache\n");
             return -1;
         }
+        qemu_mutex_init(&XBZRLE.lock);
+        qemu_mutex_unlock_iothread();

         /* We prefer not to abort if there is no memory */
         XBZRLE.encoded_buf = g_try_malloc0(TARGET_PAGE_SIZE);
-- 
1.8.5.3

[Qemu-devel] [PATCH 2/5] qemu_file: Fix mismerge of "use fwrite() correctly"

[Juan Quintela]

From: Markus Armbruster <armbru@redhat.com>

Reviewers accepted v2 of the patch, but what got committed was v1,
with the R-bys for v2.  This is the v1->v2 followup fix.

[Amit:
 This fixes commit aded6539d983280212e08d09f14157b1cb4d58cc
]

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 qemu-file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/qemu-file.c b/qemu-file.c
index f074af1..e5ec798 100644
--- a/qemu-file.c
+++ b/qemu-file.c
@@ -105,7 +105,7 @@ static int stdio_put_buffer(void *opaque, const uint8_t *buf, int64_t pos,
     res = fwrite(buf, 1, size, s->stdio_file);

     if (res != size) {
-        return -EIO;	/* fake errno value */
+        return -errno;
     }
     return res;
 }
-- 
1.8.5.3

[Qemu-devel] [PATCH 3/5] vl: add system_wakeup_request tracepoint

[Juan Quintela]

From: Alexey Kardashevskiy <aik@ozlabs.ru>

It might be useful for tracing migration.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 trace-events | 1 +
 vl.c         | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/trace-events b/trace-events
index aec4202..466c27e 100644
--- a/trace-events
+++ b/trace-events
@@ -486,6 +486,7 @@ runstate_set(int new_state) "new state %d"
 g_malloc(size_t size, void *ptr) "size %zu ptr %p"
 g_realloc(void *ptr, size_t size, void *newptr) "ptr %p size %zu newptr %p"
 g_free(void *ptr) "ptr %p"
+system_wakeup_request(int reason) "reason=%d"

 # block/qcow2.c
 qcow2_writev_start_req(void *co, int64_t sector, int nb_sectors) "co %p sector %" PRIx64 " nb_sectors %d"
diff --git a/vl.c b/vl.c
index 41581c1..50693e6 100644
--- a/vl.c
+++ b/vl.c
@@ -1837,6 +1837,8 @@ void qemu_register_suspend_notifier(Notifier *notifier)

 void qemu_system_wakeup_request(WakeupReason reason)
 {
+    trace_system_wakeup_request(reason);
+
     if (!runstate_check(RUN_STATE_SUSPENDED)) {
         return;
     }
-- 
1.8.5.3

[Qemu-devel] [PATCH 4/5] migration: extend section_start/end traces

[Juan Quintela]

From: Alexey Kardashevskiy <aik@ozlabs.ru>

This adds @idstr to savevm_section_start and savevm_section_end
tracepoints.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 savevm.c     | 12 ++++++------
 trace-events |  4 ++--
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/savevm.c b/savevm.c
index 7329fc5..d094fbb 100644
--- a/savevm.c
+++ b/savevm.c
@@ -527,13 +527,13 @@ int qemu_savevm_state_iterate(QEMUFile *f)
         if (qemu_file_rate_limit(f)) {
             return 0;
         }
-        trace_savevm_section_start();
+        trace_savevm_section_start(se->idstr, se->section_id);
         /* Section type */
         qemu_put_byte(f, QEMU_VM_SECTION_PART);
         qemu_put_be32(f, se->section_id);

         ret = se->ops->save_live_iterate(f, se->opaque);
-        trace_savevm_section_end(se->section_id);
+        trace_savevm_section_end(se->idstr, se->section_id);

         if (ret < 0) {
             qemu_file_set_error(f, ret);
@@ -565,13 +565,13 @@ void qemu_savevm_state_complete(QEMUFile *f)
                 continue;
             }
         }
-        trace_savevm_section_start();
+        trace_savevm_section_start(se->idstr, se->section_id);
         /* Section type */
         qemu_put_byte(f, QEMU_VM_SECTION_END);
         qemu_put_be32(f, se->section_id);

         ret = se->ops->save_live_complete(f, se->opaque);
-        trace_savevm_section_end(se->section_id);
+        trace_savevm_section_end(se->idstr, se->section_id);
         if (ret < 0) {
             qemu_file_set_error(f, ret);
             return;
@@ -584,7 +584,7 @@ void qemu_savevm_state_complete(QEMUFile *f)
         if ((!se->ops || !se->ops->save_state) && !se->vmsd) {
             continue;
         }
-        trace_savevm_section_start();
+        trace_savevm_section_start(se->idstr, se->section_id);
         /* Section type */
         qemu_put_byte(f, QEMU_VM_SECTION_FULL);
         qemu_put_be32(f, se->section_id);
@@ -598,7 +598,7 @@ void qemu_savevm_state_complete(QEMUFile *f)
         qemu_put_be32(f, se->version_id);

         vmstate_save(f, se);
-        trace_savevm_section_end(se->section_id);
+        trace_savevm_section_end(se->idstr, se->section_id);
     }

     qemu_put_byte(f, QEMU_VM_EOF);
diff --git a/trace-events b/trace-events
index 466c27e..002c260 100644
--- a/trace-events
+++ b/trace-events
@@ -1040,8 +1040,8 @@ vmware_scratch_write(uint32_t index, uint32_t value) "index %d, value 0x%x"
 vmware_setmode(uint32_t w, uint32_t h, uint32_t bpp) "%dx%d @ %d bpp"

 # savevm.c
-savevm_section_start(void) ""
-savevm_section_end(unsigned int section_id) "section_id %u"
+savevm_section_start(const char *id, unsigned int section_id) "%s, section_id %u"
+savevm_section_end(const char *id, unsigned int section_id) "%s, section_id %u"

 # arch_init.c
 migration_bitmap_sync_start(void) ""
-- 
1.8.5.3

[Qemu-devel] [PATCH 5/5] migration: add more traces

[Juan Quintela]

From: Alexey Kardashevskiy <aik@ozlabs.ru>

This replaces DPRINTF macro with tracepoints.

This moves some messages from migration.c to savevm.c.

This adds tracepoint to signal about fields failed to migrate.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 migration.c  | 30 ++++++------------------------
 qemu-file.c  |  2 ++
 savevm.c     | 10 ++++++++++
 trace-events | 16 ++++++++++++++++
 vmstate.c    |  2 ++
 5 files changed, 36 insertions(+), 24 deletions(-)

diff --git a/migration.c b/migration.c
index 14235b2..2497c5d 100644
--- a/migration.c
+++ b/migration.c
@@ -26,16 +26,6 @@
 #include "qmp-commands.h"
 #include "trace.h"

-//#define DEBUG_MIGRATION
-
-#ifdef DEBUG_MIGRATION
-#define DPRINTF(fmt, ...) \
-    do { printf("migration: " fmt, ## __VA_ARGS__); } while (0)
-#else
-#define DPRINTF(fmt, ...) \
-    do { } while (0)
-#endif
-
 enum {
     MIG_STATE_ERROR = -1,
     MIG_STATE_NONE,
@@ -111,7 +101,6 @@ static void process_incoming_migration_co(void *opaque)
         exit(EXIT_FAILURE);
     }
     qemu_announce_self();
-    DPRINTF("successfully loaded vm state\n");

     bdrv_clear_incoming_migration_all();
     /* Make sure all file formats flush their mutable metadata */
@@ -300,7 +289,7 @@ static void migrate_fd_cleanup(void *opaque)
     s->cleanup_bh = NULL;

     if (s->file) {
-        DPRINTF("closing file\n");
+        trace_migrate_fd_cleanup();
         qemu_mutex_unlock_iothread();
         qemu_thread_join(&s->thread);
         qemu_mutex_lock_iothread();
@@ -323,7 +312,7 @@ static void migrate_fd_cleanup(void *opaque)

 void migrate_fd_error(MigrationState *s)
 {
-    DPRINTF("setting error state\n");
+    trace_migrate_fd_error();
     assert(s->file == NULL);
     s->state = MIG_STATE_ERROR;
     trace_migrate_set_state(MIG_STATE_ERROR);
@@ -333,7 +322,7 @@ void migrate_fd_error(MigrationState *s)
 static void migrate_fd_cancel(MigrationState *s)
 {
     int old_state ;
-    DPRINTF("cancelling migration\n");
+    trace_migrate_fd_cancel();

     do {
         old_state = s->state;
@@ -583,29 +572,23 @@ static void *migration_thread(void *opaque)
     int64_t start_time = initial_time;
     bool old_vm_running = false;

-    DPRINTF("beginning savevm\n");
     qemu_savevm_state_begin(s->file, &s->params);

     s->setup_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) - setup_start;
     migrate_set_state(s, MIG_STATE_SETUP, MIG_STATE_ACTIVE);

-    DPRINTF("setup complete\n");
-
     while (s->state == MIG_STATE_ACTIVE) {
         int64_t current_time;
         uint64_t pending_size;

         if (!qemu_file_rate_limit(s->file)) {
-            DPRINTF("iterate\n");
             pending_size = qemu_savevm_state_pending(s->file, max_size);
-            DPRINTF("pending size %" PRIu64 " max %" PRIu64 "\n",
-                    pending_size, max_size);
+            trace_migrate_pending(pending_size, max_size);
             if (pending_size && pending_size >= max_size) {
                 qemu_savevm_state_iterate(s->file);
             } else {
                 int ret;

-                DPRINTF("done iterating\n");
                 qemu_mutex_lock_iothread();
                 start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
                 qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER);
@@ -644,9 +627,8 @@ static void *migration_thread(void *opaque)
             s->mbps = time_spent ? (((double) transferred_bytes * 8.0) /
                     ((double) time_spent / 1000.0)) / 1000.0 / 1000.0 : -1;

-            DPRINTF("transferred %" PRIu64 " time_spent %" PRIu64
-                    " bandwidth %g max_size %" PRId64 "\n",
-                    transferred_bytes, time_spent, bandwidth, max_size);
+            trace_migrate_transferred(transferred_bytes, time_spent,
+                                      bandwidth, max_size);
             /* if we haven't sent anything, we don't want to recalculate
                10000 is a small enough number for our purposes */
             if (s->dirty_bytes_rate && transferred_bytes > 10000) {
diff --git a/qemu-file.c b/qemu-file.c
index e5ec798..8d5f45d 100644
--- a/qemu-file.c
+++ b/qemu-file.c
@@ -4,6 +4,7 @@
 #include "block/coroutine.h"
 #include "migration/migration.h"
 #include "migration/qemu-file.h"
+#include "trace.h"

 #define IO_BUF_SIZE 32768
 #define MAX_IOV_SIZE MIN(IOV_MAX, 64)
@@ -595,6 +596,7 @@ int qemu_fclose(QEMUFile *f)
         ret = f->last_error;
     }
     g_free(f);
+    trace_qemu_file_fclose();
     return ret;
 }

diff --git a/savevm.c b/savevm.c
index d094fbb..ef7d9ce 100644
--- a/savevm.c
+++ b/savevm.c
@@ -41,6 +41,7 @@
 #include "qemu/iov.h"
 #include "block/snapshot.h"
 #include "block/qapi.h"
+#include <netinet/ether.h>

 #define SELF_ANNOUNCE_ROUNDS 5

@@ -81,6 +82,8 @@ static void qemu_announce_self_iter(NICState *nic, void *opaque)
     uint8_t buf[60];
     int len;

+    trace_qemu_announce_self_iter(ether_ntoa((struct ether_addr *)
+                                             &nic->conf->macaddr));
     len = announce_self_create(buf, nic->conf->macaddr.a);

     qemu_send_packet_raw(qemu_get_queue(nic), buf, len);
@@ -429,6 +432,7 @@ void vmstate_unregister(DeviceState *dev, const VMStateDescription *vmsd,

 static int vmstate_load(QEMUFile *f, SaveStateEntry *se, int version_id)
 {
+    trace_vmstate_load(se->idstr, se->vmsd ? se->vmsd->name : "(old)");
     if (!se->vmsd) {         /* Old style */
         return se->ops->load_state(f, se->opaque, version_id);
     }
@@ -437,6 +441,7 @@ static int vmstate_load(QEMUFile *f, SaveStateEntry *se, int version_id)

 static void vmstate_save(QEMUFile *f, SaveStateEntry *se)
 {
+    trace_vmstate_save(se->idstr, se->vmsd ? se->vmsd->name : "(old)");
     if (!se->vmsd) {         /* Old style */
         se->ops->save_state(f, se->opaque);
         return;
@@ -463,6 +468,7 @@ void qemu_savevm_state_begin(QEMUFile *f,
     SaveStateEntry *se;
     int ret;

+    trace_savevm_state_begin();
     QTAILQ_FOREACH(se, &savevm_handlers, entry) {
         if (!se->ops || !se->ops->set_params) {
             continue;
@@ -515,6 +521,7 @@ int qemu_savevm_state_iterate(QEMUFile *f)
     SaveStateEntry *se;
     int ret = 1;

+    trace_savevm_state_iterate();
     QTAILQ_FOREACH(se, &savevm_handlers, entry) {
         if (!se->ops || !se->ops->save_live_iterate) {
             continue;
@@ -554,6 +561,8 @@ void qemu_savevm_state_complete(QEMUFile *f)
     SaveStateEntry *se;
     int ret;

+    trace_savevm_state_complete();
+
     cpu_synchronize_all_states();

     QTAILQ_FOREACH(se, &savevm_handlers, entry) {
@@ -628,6 +637,7 @@ void qemu_savevm_state_cancel(void)
 {
     SaveStateEntry *se;

+    trace_savevm_state_cancel();
     QTAILQ_FOREACH(se, &savevm_handlers, entry) {
         if (se->ops && se->ops->cancel) {
             se->ops->cancel(se->opaque);
diff --git a/trace-events b/trace-events
index 002c260..657d476 100644
--- a/trace-events
+++ b/trace-events
@@ -1042,6 +1042,17 @@ vmware_setmode(uint32_t w, uint32_t h, uint32_t bpp) "%dx%d @ %d bpp"
 # savevm.c
 savevm_section_start(const char *id, unsigned int section_id) "%s, section_id %u"
 savevm_section_end(const char *id, unsigned int section_id) "%s, section_id %u"
+savevm_state_begin(void) ""
+savevm_state_iterate(void) ""
+savevm_state_complete(void) ""
+savevm_state_cancel(void) ""
+vmstate_save(const char *idstr, const char *vmsd_name) "%s, %s"
+vmstate_load(const char *idstr, const char *vmsd_name) "%s, %s"
+vmstate_load_field_error(const char *field, int ret) "field \"%s\" load failed, ret = %d"
+qemu_announce_self_iter(const char *mac) "%s"
+
+# qemu-file.c
+qemu_file_fclose(void) ""

 # arch_init.c
 migration_bitmap_sync_start(void) ""
@@ -1181,6 +1192,11 @@ flic_reset_failed(int err) "flic: reset failed %d"

 # migration.c
 migrate_set_state(int new_state) "new state %d"
+migrate_fd_cleanup(void) ""
+migrate_fd_error(void) ""
+migrate_fd_cancel(void) ""
+migrate_pending(uint64_t size, uint64_t max) "pending size %" PRIu64 " max %" PRIu64
+migrate_transferred(uint64_t tranferred, uint64_t time_spent, double bandwidth, uint64_t size) "transferred %" PRIu64 " time_spent %" PRIu64 " bandwidth %g max_size %" PRId64

 # kvm-all.c
 kvm_ioctl(int type, void *arg) "type 0x%x, arg %p"
diff --git a/vmstate.c b/vmstate.c
index d1f5eb0..b689f2f 100644
--- a/vmstate.c
+++ b/vmstate.c
@@ -3,6 +3,7 @@
 #include "migration/qemu-file.h"
 #include "migration/vmstate.h"
 #include "qemu/bitops.h"
+#include "trace.h"

 static void vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
                                     void *opaque);
@@ -73,6 +74,7 @@ int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,

                 }
                 if (ret < 0) {
+                    trace_vmstate_load_field_error(field->name, ret);
                     return ret;
                 }
             }
-- 
1.8.5.3

Re: [Qemu-devel] [PULL 0/5] migration queue

[Peter Maydell]

On 8 March 2014 01:23, Juan Quintela <quintela@redhat.com> wrote:
> Hi
>
> Please pull
>
> - Fix missmerge of fwrite patch (armbru)
> - FIX XBZRLE crash: Gonglei
> - Add more traces for migration (Alexey)
>
> Especial thanks to Amit for getting the patches together.

I'm afraid this doesn't build for Windows:

/home/petmay01/linaro/qemu-for-merges/savevm.c:44:27: warning:
netinet/ether.h: No such file or directory
/home/petmay01/linaro/qemu-for-merges/savevm.c: In function
‘qemu_announce_self_iter’:
/home/petmay01/linaro/qemu-for-merges/savevm.c:85: warning: implicit
declaration of function ‘ether_ntoa’
/home/petmay01/linaro/qemu-for-merges/savevm.c:85: warning: nested
extern declaration of ‘ether_ntoa’
/home/petmay01/linaro/qemu-for-merges/savevm.c:86: warning: passing
argument 1 of ‘trace_qemu_announce_self_iter’ makes pointer from
integer without a cast
  CC    alpha-softmmu/savevm.o
  LINK  alpha-softmmu/qemu-system-alpha.exe
savevm.o: In function `qemu_announce_self_iter':
/home/petmay01/linaro/qemu-for-merges/savevm.c:85: undefined reference
to `_ether_ntoa'
collect2: ld returned 1 exit status

thanks
-- PMM

[Qemu-devel] [PULL 0/5] migration queue

[Dr. David Alan Gilbert (git)]

From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>

The following changes since commit 4c2c1015905fa1d616750dfe024b4c0b35875950:

  Merge remote-tracking branch 'remotes/borntraeger/tags/s390x-20180323' into staging (2018-03-23 10:20:54 +0000)

are available in the Git repository at:

  git://github.com/dagrh/qemu.git tags/pull-migration-20180323a

for you to fetch changes up to 09576e74dbe697c5f0a7bc2ad7b59601457a2ff4:

  migration: Fix block migration flag case (2018-03-23 18:24:11 +0000)

----------------------------------------------------------------
Migration fixes for 2.12

All small fixes.  Dan's is a missing piece
of a cleanup that finally completes something,
and between Paolo, Dan and myself we recon it's
still on the edge of being a bug fix.

----------------------------------------------------------------
Daniel P. Berrangé (1):
      migration: convert socket server to QIONetListener

Dr. David Alan Gilbert (1):
      migration: Fix block migration flag case

Lidong Chen (1):
      migration: Fix rate limiting issue on RDMA migration

Peter Lieven (2):
      migration/block: limit the number of parallel I/O requests
      migration/block: compare only read blocks against the rate limiter

 migration/block.c     |  5 +++--
 migration/migration.c |  1 +
 migration/qemu-file.c |  2 +-
 migration/socket.c    | 48 ++++++++++++++++--------------------------------
 4 files changed, 21 insertions(+), 35 deletions(-)

Re: [Qemu-devel] [PULL 0/5] migration queue

[Peter Maydell]

On 23 March 2018 at 20:23, Dr. David Alan Gilbert (git)
<dgilbert@redhat.com> wrote:
> From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
>
> The following changes since commit 4c2c1015905fa1d616750dfe024b4c0b35875950:
>
>   Merge remote-tracking branch 'remotes/borntraeger/tags/s390x-20180323' into staging (2018-03-23 10:20:54 +0000)
>
> are available in the Git repository at:
>
>   git://github.com/dagrh/qemu.git tags/pull-migration-20180323a
>
> for you to fetch changes up to 09576e74dbe697c5f0a7bc2ad7b59601457a2ff4:
>
>   migration: Fix block migration flag case (2018-03-23 18:24:11 +0000)
>
> ----------------------------------------------------------------
> Migration fixes for 2.12
>
> All small fixes.  Dan's is a missing piece
> of a cleanup that finally completes something,
> and between Paolo, Dan and myself we recon it's
> still on the edge of being a bug fix.
>
> ----------------------------------------------------------------

Applied, thanks.

-- PMM

[Qemu-devel] [PULL 0/5] Migration queue

[Juan Quintela]

The following changes since commit 19b599f7664b2ebfd0f405fb79c14dd241557452:

  Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2018-08-27-v2' into staging (2018-08-27 16:44:20 +0100)

are available in the Git repository at:

  git://github.com/juanquintela/qemu.git tags/migration/20180903

for you to fetch changes up to a28652cd4c138e9ed5d7dd9f7d10e3562f89ccac:

  Add a hint message to loadvm and exits on failure (2018-09-03 19:50:42 +0200)

----------------------------------------------------------------
migration/next for 20180903

Reviewed patches for migration:
- fix clang warning (david)
- Improve loadvm handling (Jose)
- Include migration test for s390 (thomas)
- Improve compression handling (xiao)

Please apply.

----------------------------------------------------------------
Dr. David Alan Gilbert (1):
      migration/rdma: Fix uninitialised rdma_return_path

Jose Ricardo Ziviani (1):
      Add a hint message to loadvm and exits on failure

Thomas Huth (1):
      tests/migration: Enable the migration test on s390x, too

Xiao Guangrong (2):
      migration: fix calculating xbzrle_counters.cache_miss_rate
      migration: handle the error condition properly

 migration/ram.c                      |   36 +-
 migration/rdma.c                     |    2 +-
 migration/savevm.c                   |    4 +-
 tests/Makefile.include               |    1 +
 tests/Makefile.include.orig          | 1025 ++++++++++++++++++++++++++++++++++
 tests/migration-test.c               |   23 +
 tests/migration-test.c.orig          |  780 ++++++++++++++++++++++++++
 tests/migration/s390x-a-b-rebuild.sh |   42 ++
 tests/migration/s390x-a-b.c          |   35 ++
 tests/migration/s390x-a-b.h          |  427 ++++++++++++++
 vl.c                                 |    1 +
 11 files changed, 2362 insertions(+), 14 deletions(-)
 create mode 100644 tests/Makefile.include.orig
 create mode 100644 tests/migration-test.c.orig
 create mode 100755 tests/migration/s390x-a-b-rebuild.sh
 create mode 100644 tests/migration/s390x-a-b.c
 create mode 100644 tests/migration/s390x-a-b.h

Re: [Qemu-devel] [PULL 0/5] Migration queue

[Juan Quintela]

Juan Quintela <quintela@redhat.com> wrote:
> The following changes since commit 19b599f7664b2ebfd0f405fb79c14dd241557452:
>
>   Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2018-08-27-v2' into staging (2018-08-27 16:44:20 +0100)
>
> are available in the Git repository at:
>
>   git://github.com/juanquintela/qemu.git tags/migration/20180903
>
> for you to fetch changes up to a28652cd4c138e9ed5d7dd9f7d10e3562f89ccac:
>
>   Add a hint message to loadvm and exits on failure (2018-09-03 19:50:42 +0200)

Nack myself.  This is included on the following migration pull.

Later, Juan