* [Qemu-devel] [PATCH 0/4] Migration Deciphering aid
@ 2014-12-22 23:03 Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 1/4] QJSON: Add JSON writer Alexander Graf
` (3 more replies)
0 siblings, 4 replies; 5+ messages in thread
From: Alexander Graf @ 2014-12-22 23:03 UTC (permalink / raw)
To: qemu-devel; +Cc: Amit Shah, Paolo Bonzini, qemu-ppc, afaerber, Juan Quintela
Migration is a black hole to most people. One of the biggest reasons for
this is that its protocol is a secret, undocumented sauce of code rolling
around random parts of the QEMU code base.
But what if we simply exposed the description of how the format looks like
alongside the actual migration stream? This is what this patch set does.
It adds a new section that comes after the end of stream marker (so that it
doesn't slow down migration) that contains a JSON description of the device
state description.
Along with this patch set also comes a python script that can read said JSON
from a migration dump and decipher the device state contents of the migration
dump using it.
With this, you can now fully examine all glorious details that go over the
wire when virtual machine state gets dumped, such as during live migration.
We discussed the approach taken here during KVM Forum 2013. Originally, my idea
was to include a special device that contains the JSON data which can be enabled
on demand. Anthony suggested however to just always include the description data
after the end marker which I think is a great idea.
Example decoded migration: http://csgraf.de/mig/mig.txt
Example migration description: http://csgraf.de/mig/mig.desc.txt
Presentation: https://www.youtube.com/watch?v=iq1x40Qsrew
Slides: https://www.dropbox.com/s/otp2pk2n3g087zp/Live%20Migration.pdf
(If you received this email twice, I'm sorry about this. I forgot to send it
to qemu-devel in the first round. Sigh. Merry Christmas! )
Alexander Graf (4):
QJSON: Add JSON writer
qemu-file: Add fast ftell code path
migration: Append JSON description of migration stream
Add migration stream analyzation script
Makefile.objs | 1 +
hw/pci/pci.c | 2 +-
hw/scsi/spapr_vscsi.c | 2 +-
hw/virtio/virtio.c | 2 +-
include/migration/migration.h | 1 +
include/migration/qemu-file.h | 1 +
include/migration/vmstate.h | 3 +-
include/qjson.h | 28 +++
migration/qemu-file.c | 5 +
migration/vmstate.c | 181 ++++++++++++++-
qjson.c | 97 ++++++++
savevm.c | 53 ++++-
scripts/analyze-migration.py | 529 ++++++++++++++++++++++++++++++++++++++++++
tests/test-vmstate.c | 6 +-
14 files changed, 892 insertions(+), 19 deletions(-)
create mode 100644 include/qjson.h
create mode 100644 qjson.c
create mode 100755 scripts/analyze-migration.py
--
1.7.12.4
^ permalink raw reply [flat|nested] 5+ messages in thread
* [Qemu-devel] [PATCH 1/4] QJSON: Add JSON writer
2014-12-22 23:03 [Qemu-devel] [PATCH 0/4] Migration Deciphering aid Alexander Graf
@ 2014-12-22 23:03 ` Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 2/4] qemu-file: Add fast ftell code path Alexander Graf
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Alexander Graf @ 2014-12-22 23:03 UTC (permalink / raw)
To: qemu-devel; +Cc: Amit Shah, Paolo Bonzini, qemu-ppc, afaerber, Juan Quintela
To support programmatic JSON assembly while keeping the code that generates it
readable, this patch introduces a simple JSON writer. It emits JSON serially
into a buffer in memory.
The nice thing about this writer is its simplicity and low memory overhead.
Unlike the QMP JSON writer, this one does not need to spawn QObjects for every
element it wants to represent.
This is a prerequisite for the migration stream format description generator.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
Makefile.objs | 1 +
include/qjson.h | 28 +++++++++++++++++
qjson.c | 97 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 126 insertions(+)
create mode 100644 include/qjson.h
create mode 100644 qjson.c
diff --git a/Makefile.objs b/Makefile.objs
index abeb902..28999d3 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -51,6 +51,7 @@ common-obj-$(CONFIG_LINUX) += fsdev/
common-obj-y += migration/
common-obj-y += qemu-char.o #aio.o
common-obj-y += page_cache.o
+common-obj-y += qjson.o
common-obj-$(CONFIG_SPICE) += spice-qemu-char.o
diff --git a/include/qjson.h b/include/qjson.h
new file mode 100644
index 0000000..8f8c145
--- /dev/null
+++ b/include/qjson.h
@@ -0,0 +1,28 @@
+/*
+ * QEMU JSON writer
+ *
+ * Copyright Alexander Graf
+ *
+ * Authors:
+ * Alexander Graf <agraf@suse.de
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.1 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ *
+ */
+#ifndef QEMU_QJSON_H
+#define QEMU_QJSON_H
+
+typedef struct QJSON QJSON;
+
+QJSON *qjson_new(void);
+void json_prop_str(QJSON *json, const char *name, const char *str);
+void json_prop_int(QJSON *json, const char *name, int64_t val);
+void json_end_array(QJSON *json);
+void json_start_array(QJSON *json, const char *name);
+void json_end_object(QJSON *json);
+void json_start_object(QJSON *json, const char *name);
+const char *qjson_get_str(QJSON *json);
+void qjson_finish(QJSON *json);
+
+#endif /* QEMU_QJSON_H */
diff --git a/qjson.c b/qjson.c
new file mode 100644
index 0000000..7a7cd72
--- /dev/null
+++ b/qjson.c
@@ -0,0 +1,97 @@
+/*
+ * QEMU JSON writer
+ *
+ * Copyright Alexander Graf
+ *
+ * Authors:
+ * Alexander Graf <agraf@suse.de
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.1 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ *
+ */
+
+#include <qapi/qmp/qstring.h>
+#include <stdbool.h>
+#include <glib.h>
+#include <qjson.h>
+
+struct QJSON {
+ QString *str;
+ bool omit_comma;
+ unsigned long self_size_offset;
+};
+
+static void json_emit_element(QJSON *json, const char *name)
+{
+ /* Check whether we need to print a , before an element */
+ if (json->omit_comma) {
+ json->omit_comma = false;
+ } else {
+ qstring_append(json->str, ", ");
+ }
+
+ if (name) {
+ qstring_append(json->str, "\"");
+ qstring_append(json->str, name);
+ qstring_append(json->str, "\" : ");
+ }
+}
+
+void json_start_object(QJSON *json, const char *name)
+{
+ json_emit_element(json, name);
+ qstring_append(json->str, "{ ");
+ json->omit_comma = true;
+}
+
+void json_end_object(QJSON *json)
+{
+ qstring_append(json->str, " }");
+ json->omit_comma = false;
+}
+
+void json_start_array(QJSON *json, const char *name)
+{
+ json_emit_element(json, name);
+ qstring_append(json->str, "[ ");
+ json->omit_comma = true;
+}
+
+void json_end_array(QJSON *json)
+{
+ qstring_append(json->str, " ]");
+ json->omit_comma = false;
+}
+
+void json_prop_int(QJSON *json, const char *name, int64_t val)
+{
+ json_emit_element(json, name);
+ qstring_append_int(json->str, val);
+}
+
+void json_prop_str(QJSON *json, const char *name, const char *str)
+{
+ json_emit_element(json, name);
+ qstring_append_chr(json->str, '"');
+ qstring_append(json->str, str);
+ qstring_append_chr(json->str, '"');
+}
+
+const char *qjson_get_str(QJSON *json)
+{
+ return qstring_get_str(json->str);
+}
+
+QJSON *qjson_new(void)
+{
+ QJSON *json = g_new(QJSON, 1);
+ json->str = qstring_from_str("{ ");
+ json->omit_comma = true;
+ return json;
+}
+
+void qjson_finish(QJSON *json)
+{
+ json_end_object(json);
+}
--
1.7.12.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [Qemu-devel] [PATCH 2/4] qemu-file: Add fast ftell code path
2014-12-22 23:03 [Qemu-devel] [PATCH 0/4] Migration Deciphering aid Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 1/4] QJSON: Add JSON writer Alexander Graf
@ 2014-12-22 23:03 ` Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 3/4] migration: Append JSON description of migration stream Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 4/4] Add migration stream analyzation script Alexander Graf
3 siblings, 0 replies; 5+ messages in thread
From: Alexander Graf @ 2014-12-22 23:03 UTC (permalink / raw)
To: qemu-devel; +Cc: Amit Shah, Paolo Bonzini, qemu-ppc, afaerber, Juan Quintela
For ftell we flush the output buffer to ensure that we don't have anything
lingering in our internal buffers. This is a very safe thing to do.
However, with the dynamic size measurement that the dynamic vmstate
description will bring this would turn out quite slow.
Instead, we can fast path this specific measurement and just take the
internal buffers into account when telling the kernel our position.
I'm sure I overlooked some corner cases where this doesn't work, so
instead of tuning the safe, existing version, this patch adds a fast
variant of ftell that gets used by the dynamic vmstate description code
which isn't critical when it fails.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
include/migration/qemu-file.h | 1 +
migration/qemu-file.c | 5 +++++
2 files changed, 6 insertions(+)
diff --git a/include/migration/qemu-file.h b/include/migration/qemu-file.h
index 401676b..6b6772b 100644
--- a/include/migration/qemu-file.h
+++ b/include/migration/qemu-file.h
@@ -112,6 +112,7 @@ QEMUFile *qemu_bufopen(const char *mode, QEMUSizedBuffer *input);
int qemu_get_fd(QEMUFile *f);
int qemu_fclose(QEMUFile *f);
int64_t qemu_ftell(QEMUFile *f);
+int64_t qemu_ftell_fast(QEMUFile *f);
void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, int size);
void qemu_put_byte(QEMUFile *f, int v);
/*
diff --git a/migration/qemu-file.c b/migration/qemu-file.c
index d2d4007..a1e2357 100644
--- a/migration/qemu-file.c
+++ b/migration/qemu-file.c
@@ -440,6 +440,11 @@ int qemu_get_byte(QEMUFile *f)
return result;
}
+int64_t qemu_ftell_fast(QEMUFile *f)
+{
+ return f->pos + f->buf_index;
+}
+
int64_t qemu_ftell(QEMUFile *f)
{
qemu_fflush(f);
--
1.7.12.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [Qemu-devel] [PATCH 3/4] migration: Append JSON description of migration stream
2014-12-22 23:03 [Qemu-devel] [PATCH 0/4] Migration Deciphering aid Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 1/4] QJSON: Add JSON writer Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 2/4] qemu-file: Add fast ftell code path Alexander Graf
@ 2014-12-22 23:03 ` Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 4/4] Add migration stream analyzation script Alexander Graf
3 siblings, 0 replies; 5+ messages in thread
From: Alexander Graf @ 2014-12-22 23:03 UTC (permalink / raw)
To: qemu-devel; +Cc: Amit Shah, Paolo Bonzini, qemu-ppc, afaerber, Juan Quintela
One of the annoyances of the current migration format is the fact that
it's not self-describing. In fact, it's not properly describing at all.
Some code randomly scattered throughout QEMU elaborates roughly how to
read and write a stream of bytes.
We discussed an idea during KVM Forum 2013 to add a JSON description of
the migration protocol itself to the migration stream. This patch
adds a section after the VM_END migration end marker that contains
description data on what the device sections of the stream are composed of.
With an additional external program this allows us to decipher the
contents of any migration stream and hopefully make migration bugs easier
to track down.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
hw/pci/pci.c | 2 +-
hw/scsi/spapr_vscsi.c | 2 +-
hw/virtio/virtio.c | 2 +-
include/migration/migration.h | 1 +
include/migration/vmstate.h | 3 +-
migration/vmstate.c | 181 ++++++++++++++++++++++++++++++++++++++++--
savevm.c | 53 +++++++++++--
tests/test-vmstate.c | 6 +-
8 files changed, 231 insertions(+), 19 deletions(-)
diff --git a/hw/pci/pci.c b/hw/pci/pci.c
index 371699c..38d0b7b 100644
--- a/hw/pci/pci.c
+++ b/hw/pci/pci.c
@@ -512,7 +512,7 @@ void pci_device_save(PCIDevice *s, QEMUFile *f)
* This makes us compatible with old devices
* which never set or clear this bit. */
s->config[PCI_STATUS] &= ~PCI_STATUS_INTERRUPT;
- vmstate_save_state(f, pci_get_vmstate(s), s);
+ vmstate_save_state(f, pci_get_vmstate(s), s, NULL);
/* Restore the interrupt status bit. */
pci_update_irq_status(s);
}
diff --git a/hw/scsi/spapr_vscsi.c b/hw/scsi/spapr_vscsi.c
index 20b20f0..3639235 100644
--- a/hw/scsi/spapr_vscsi.c
+++ b/hw/scsi/spapr_vscsi.c
@@ -630,7 +630,7 @@ static void vscsi_save_request(QEMUFile *f, SCSIRequest *sreq)
vscsi_req *req = sreq->hba_private;
assert(req->active);
- vmstate_save_state(f, &vmstate_spapr_vscsi_req, req);
+ vmstate_save_state(f, &vmstate_spapr_vscsi_req, req, NULL);
DPRINTF("VSCSI: saving tag=%u, current desc#%d, offset=%x\n",
req->qtag, req->cur_desc_num, req->cur_desc_offset);
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 013979a..d735343 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -955,7 +955,7 @@ void virtio_save(VirtIODevice *vdev, QEMUFile *f)
}
/* Subsections */
- vmstate_save_state(f, &vmstate_virtio, vdev);
+ vmstate_save_state(f, &vmstate_virtio, vdev, NULL);
}
int virtio_set_features(VirtIODevice *vdev, uint32_t val)
diff --git a/include/migration/migration.h b/include/migration/migration.h
index 3cb5ba8..f37348b 100644
--- a/include/migration/migration.h
+++ b/include/migration/migration.h
@@ -33,6 +33,7 @@
#define QEMU_VM_SECTION_END 0x03
#define QEMU_VM_SECTION_FULL 0x04
#define QEMU_VM_SUBSECTION 0x05
+#define QEMU_VM_VMDESCRIPTION 0x06
struct MigrationParams {
bool blk;
diff --git a/include/migration/vmstate.h b/include/migration/vmstate.h
index e45fc49..1ed5311 100644
--- a/include/migration/vmstate.h
+++ b/include/migration/vmstate.h
@@ -29,6 +29,7 @@
#ifndef CONFIG_USER_ONLY
#include <migration/qemu-file.h>
#endif
+#include <qjson.h>
typedef void SaveStateHandler(QEMUFile *f, void *opaque);
typedef int LoadStateHandler(QEMUFile *f, void *opaque, int version_id);
@@ -779,7 +780,7 @@ extern const VMStateInfo vmstate_info_bitmap;
int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,
void *opaque, int version_id);
void vmstate_save_state(QEMUFile *f, const VMStateDescription *vmsd,
- void *opaque);
+ void *opaque, QJSON *vmdesc);
int vmstate_register_with_alias_id(DeviceState *dev, int instance_id,
const VMStateDescription *vmsd,
diff --git a/migration/vmstate.c b/migration/vmstate.c
index 3dde574..486f470 100644
--- a/migration/vmstate.c
+++ b/migration/vmstate.c
@@ -4,9 +4,10 @@
#include "migration/vmstate.h"
#include "qemu/bitops.h"
#include "trace.h"
+#include "qjson.h"
static void vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
- void *opaque);
+ void *opaque, QJSON *vmdesc);
static int vmstate_subsection_load(QEMUFile *f, const VMStateDescription *vmsd,
void *opaque);
@@ -138,32 +139,176 @@ int vmstate_load_state(QEMUFile *f, const VMStateDescription *vmsd,
return 0;
}
+static int vmfield_name_num(VMStateField *start, VMStateField *search)
+{
+ VMStateField *field;
+ int found = 0;
+
+ for (field = start; field->name; field++) {
+ if (!strcmp(field->name, search->name)) {
+ if (field == search) {
+ return found;
+ }
+ found++;
+ }
+ }
+
+ return -1;
+}
+
+static bool vmfield_name_is_unique(VMStateField *start, VMStateField *search)
+{
+ VMStateField *field;
+ int found = 0;
+
+ for (field = start; field->name; field++) {
+ if (!strcmp(field->name, search->name)) {
+ found++;
+ /* name found more than once, so it's not unique */
+ if (found > 1) {
+ return false;
+ }
+ }
+ }
+
+ return true;
+}
+
+static const char *vmfield_get_type_name(VMStateField *field)
+{
+ const char *type = "unknown";
+
+ if (field->flags & VMS_STRUCT) {
+ type = "struct";
+ } else if (field->info->name) {
+ type = field->info->name;
+ }
+
+ return type;
+}
+
+static bool vmsd_can_compress(VMStateField *field)
+{
+ if (field->field_exists) {
+ /* Dynamically existing fields mess up compression */
+ return false;
+ }
+
+ if (field->flags & VMS_STRUCT) {
+ VMStateField *sfield = field->vmsd->fields;
+ while (sfield->name) {
+ if (!vmsd_can_compress(sfield)) {
+ /* Child elements can't compress, so can't we */
+ return false;
+ }
+ sfield++;
+ }
+ }
+
+ return true;
+}
+
+static void vmsd_desc_field_start(const VMStateDescription *vmsd, QJSON *vmdesc,
+ VMStateField *field, int i, int max)
+{
+ char *name, *old_name;
+ bool is_array = max > 1;
+ bool can_compress = vmsd_can_compress(field);
+
+ if (!vmdesc) {
+ return;
+ }
+
+ name = g_strdup(field->name);
+
+ /* Field name is not unique, need to make it unique */
+ if (!vmfield_name_is_unique(vmsd->fields, field)) {
+ int num = vmfield_name_num(vmsd->fields, field);
+ old_name = name;
+ name = g_strdup_printf("%s[%d]", name, num);
+ g_free(old_name);
+ }
+
+ json_start_object(vmdesc, NULL);
+ json_prop_str(vmdesc, "name", name);
+ if (is_array) {
+ if (can_compress) {
+ json_prop_int(vmdesc, "array_len", max);
+ } else {
+ json_prop_int(vmdesc, "index", i);
+ }
+ }
+ json_prop_str(vmdesc, "type", vmfield_get_type_name(field));
+
+ if (field->flags & VMS_STRUCT) {
+ json_start_object(vmdesc, "struct");
+ }
+
+ g_free(name);
+}
+
+static void vmsd_desc_field_end(const VMStateDescription *vmsd, QJSON *vmdesc,
+ VMStateField *field, size_t size, int i)
+{
+ if (!vmdesc) {
+ return;
+ }
+
+ if (field->flags & VMS_STRUCT) {
+ /* We printed a struct in between, close its child object */
+ json_end_object(vmdesc);
+ }
+
+ json_prop_int(vmdesc, "size", size);
+ json_end_object(vmdesc);
+}
+
void vmstate_save_state(QEMUFile *f, const VMStateDescription *vmsd,
- void *opaque)
+ void *opaque, QJSON *vmdesc)
{
VMStateField *field = vmsd->fields;
if (vmsd->pre_save) {
vmsd->pre_save(opaque);
}
+
+ if (vmdesc) {
+ json_prop_str(vmdesc, "vmsd_name", vmsd->name);
+ json_prop_int(vmdesc, "version", vmsd->version_id);
+ json_start_array(vmdesc, "fields");
+ }
+
while (field->name) {
if (!field->field_exists ||
field->field_exists(opaque, vmsd->version_id)) {
void *base_addr = vmstate_base_addr(opaque, field, false);
int i, n_elems = vmstate_n_elems(opaque, field);
int size = vmstate_size(opaque, field);
+ int64_t old_offset, written_bytes;
+ QJSON *vmdesc_loop = vmdesc;
for (i = 0; i < n_elems; i++) {
void *addr = base_addr + size * i;
+ vmsd_desc_field_start(vmsd, vmdesc_loop, field, i, n_elems);
+ old_offset = qemu_ftell_fast(f);
+
if (field->flags & VMS_ARRAY_OF_POINTER) {
addr = *(void **)addr;
}
if (field->flags & VMS_STRUCT) {
- vmstate_save_state(f, field->vmsd, addr);
+ vmstate_save_state(f, field->vmsd, addr, vmdesc_loop);
} else {
field->info->put(f, addr, size);
}
+
+ written_bytes = qemu_ftell_fast(f) - old_offset;
+ vmsd_desc_field_end(vmsd, vmdesc_loop, field, written_bytes, i);
+
+ /* Compressed arrays only care about the first element */
+ if (vmdesc_loop && vmsd_can_compress(field)) {
+ vmdesc_loop = NULL;
+ }
}
} else {
if (field->flags & VMS_MUST_EXIST) {
@@ -174,7 +319,12 @@ void vmstate_save_state(QEMUFile *f, const VMStateDescription *vmsd,
}
field++;
}
- vmstate_subsection_save(f, vmsd, opaque);
+
+ if (vmdesc) {
+ json_end_array(vmdesc);
+ }
+
+ vmstate_subsection_save(f, vmsd, opaque, vmdesc);
}
static const VMStateDescription *
@@ -231,24 +381,43 @@ static int vmstate_subsection_load(QEMUFile *f, const VMStateDescription *vmsd,
}
static void vmstate_subsection_save(QEMUFile *f, const VMStateDescription *vmsd,
- void *opaque)
+ void *opaque, QJSON *vmdesc)
{
const VMStateSubsection *sub = vmsd->subsections;
+ bool subsection_found = false;
while (sub && sub->needed) {
if (sub->needed(opaque)) {
const VMStateDescription *vmsd = sub->vmsd;
uint8_t len;
+ if (vmdesc) {
+ /* Only create subsection array when we have any */
+ if (!subsection_found) {
+ json_start_array(vmdesc, "subsections");
+ subsection_found = true;
+ }
+
+ json_start_object(vmdesc, NULL);
+ }
+
qemu_put_byte(f, QEMU_VM_SUBSECTION);
len = strlen(vmsd->name);
qemu_put_byte(f, len);
qemu_put_buffer(f, (uint8_t *)vmsd->name, len);
qemu_put_be32(f, vmsd->version_id);
- vmstate_save_state(f, vmsd, opaque);
+ vmstate_save_state(f, vmsd, opaque, vmdesc);
+
+ if (vmdesc) {
+ json_end_object(vmdesc);
+ }
}
sub++;
}
+
+ if (vmdesc && subsection_found) {
+ json_end_array(vmdesc);
+ }
}
/* bool */
diff --git a/savevm.c b/savevm.c
index 08ec678..d1ee7c9 100644
--- a/savevm.c
+++ b/savevm.c
@@ -572,14 +572,34 @@ static int vmstate_load(QEMUFile *f, SaveStateEntry *se, int version_id)
return vmstate_load_state(f, se->vmsd, se->opaque, version_id);
}
-static void vmstate_save(QEMUFile *f, SaveStateEntry *se)
+static void vmstate_save_old_style(QEMUFile *f, SaveStateEntry *se, QJSON *vmdesc)
+{
+ int64_t old_offset, size;
+
+ old_offset = qemu_ftell_fast(f);
+ se->ops->save_state(f, se->opaque);
+ size = qemu_ftell_fast(f) - old_offset;
+
+ if (vmdesc) {
+ json_prop_int(vmdesc, "size", size);
+ json_start_array(vmdesc, "fields");
+ json_start_object(vmdesc, NULL);
+ json_prop_str(vmdesc, "name", "data");
+ json_prop_int(vmdesc, "size", size);
+ json_prop_str(vmdesc, "type", "buffer");
+ json_end_object(vmdesc);
+ json_end_array(vmdesc);
+ }
+}
+
+static void vmstate_save(QEMUFile *f, SaveStateEntry *se, QJSON *vmdesc)
{
trace_vmstate_save(se->idstr, se->vmsd ? se->vmsd->name : "(old)");
- if (!se->vmsd) { /* Old style */
- se->ops->save_state(f, se->opaque);
+ if (!se->vmsd) {
+ vmstate_save_old_style(f, se, vmdesc);
return;
}
- vmstate_save_state(f, se->vmsd, se->opaque);
+ vmstate_save_state(f, se->vmsd, se->opaque, vmdesc);
}
bool qemu_savevm_state_blocked(Error **errp)
@@ -692,6 +712,8 @@ int qemu_savevm_state_iterate(QEMUFile *f)
void qemu_savevm_state_complete(QEMUFile *f)
{
+ QJSON *vmdesc;
+ int vmdesc_len;
SaveStateEntry *se;
int ret;
@@ -721,6 +743,9 @@ void qemu_savevm_state_complete(QEMUFile *f)
}
}
+ vmdesc = qjson_new();
+ json_prop_int(vmdesc, "page_size", TARGET_PAGE_SIZE);
+ json_start_array(vmdesc, "devices");
QTAILQ_FOREACH(se, &savevm_handlers, entry) {
int len;
@@ -728,6 +753,11 @@ void qemu_savevm_state_complete(QEMUFile *f)
continue;
}
trace_savevm_section_start(se->idstr, se->section_id);
+
+ json_start_object(vmdesc, NULL);
+ json_prop_str(vmdesc, "name", se->idstr);
+ json_prop_int(vmdesc, "instance_id", se->instance_id);
+
/* Section type */
qemu_put_byte(f, QEMU_VM_SECTION_FULL);
qemu_put_be32(f, se->section_id);
@@ -740,11 +770,22 @@ void qemu_savevm_state_complete(QEMUFile *f)
qemu_put_be32(f, se->instance_id);
qemu_put_be32(f, se->version_id);
- vmstate_save(f, se);
+ vmstate_save(f, se, vmdesc);
+
+ json_end_object(vmdesc);
trace_savevm_section_end(se->idstr, se->section_id);
}
qemu_put_byte(f, QEMU_VM_EOF);
+
+ json_end_array(vmdesc);
+ qjson_finish(vmdesc);
+ vmdesc_len = strlen(qjson_get_str(vmdesc));
+
+ qemu_put_byte(f, QEMU_VM_VMDESCRIPTION);
+ qemu_put_be32(f, vmdesc_len);
+ qemu_put_buffer(f, (uint8_t *)qjson_get_str(vmdesc), vmdesc_len);
+
qemu_fflush(f);
}
@@ -843,7 +884,7 @@ static int qemu_save_device_state(QEMUFile *f)
qemu_put_be32(f, se->instance_id);
qemu_put_be32(f, se->version_id);
- vmstate_save(f, se);
+ vmstate_save(f, se, NULL);
}
qemu_put_byte(f, QEMU_VM_EOF);
diff --git a/tests/test-vmstate.c b/tests/test-vmstate.c
index 5e0fd13..55cb93b 100644
--- a/tests/test-vmstate.c
+++ b/tests/test-vmstate.c
@@ -95,7 +95,7 @@ static void save_vmstate(const VMStateDescription *desc, void *obj)
QEMUFile *f = open_test_file(true);
/* Save file with vmstate */
- vmstate_save_state(f, desc, obj);
+ vmstate_save_state(f, desc, obj, NULL);
qemu_put_byte(f, QEMU_VM_EOF);
g_assert(!qemu_file_get_error(f));
qemu_fclose(f);
@@ -404,7 +404,7 @@ static void test_save_noskip(void)
QEMUFile *fsave = qemu_bufopen("w", NULL);
TestStruct obj = { .a = 1, .b = 2, .c = 3, .d = 4, .e = 5, .f = 6,
.skip_c_e = false };
- vmstate_save_state(fsave, &vmstate_skipping, &obj);
+ vmstate_save_state(fsave, &vmstate_skipping, &obj, NULL);
g_assert(!qemu_file_get_error(fsave));
uint8_t expected[] = {
@@ -424,7 +424,7 @@ static void test_save_skip(void)
QEMUFile *fsave = qemu_bufopen("w", NULL);
TestStruct obj = { .a = 1, .b = 2, .c = 3, .d = 4, .e = 5, .f = 6,
.skip_c_e = true };
- vmstate_save_state(fsave, &vmstate_skipping, &obj);
+ vmstate_save_state(fsave, &vmstate_skipping, &obj, NULL);
g_assert(!qemu_file_get_error(fsave));
uint8_t expected[] = {
--
1.7.12.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [Qemu-devel] [PATCH 4/4] Add migration stream analyzation script
2014-12-22 23:03 [Qemu-devel] [PATCH 0/4] Migration Deciphering aid Alexander Graf
` (2 preceding siblings ...)
2014-12-22 23:03 ` [Qemu-devel] [PATCH 3/4] migration: Append JSON description of migration stream Alexander Graf
@ 2014-12-22 23:03 ` Alexander Graf
3 siblings, 0 replies; 5+ messages in thread
From: Alexander Graf @ 2014-12-22 23:03 UTC (permalink / raw)
To: qemu-devel; +Cc: Amit Shah, Paolo Bonzini, qemu-ppc, afaerber, Juan Quintela
This patch adds a python tool to the scripts directory that can read
a dumped migration stream if it contains the JSON description of the
device states. I constructs a human readable JSON stream out of it.
It's very simple to use:
$ qemu-system-x86_64
(qemu) migrate "exec:cat > mig"
$ ./scripts/analyze_migration.py -f mig
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Remove support for multiple vmsd versions
- Add support for HTAB
- Add support for array_len
- Move to new naming schema for fields
- Use dynamic page size
---
scripts/analyze-migration.py | 529 +++++++++++++++++++++++++++++++++++++++++++
1 file changed, 529 insertions(+)
create mode 100755 scripts/analyze-migration.py
diff --git a/scripts/analyze-migration.py b/scripts/analyze-migration.py
new file mode 100755
index 0000000..2878011
--- /dev/null
+++ b/scripts/analyze-migration.py
@@ -0,0 +1,529 @@
+#!/usr/bin/env python
+#
+# Migration Stream Analyzer
+#
+# Copyright (c) 2013 Alexander Graf <agraf@suse.de>
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2 of the License, or (at your option) any later version.
+#
+# This library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, see <http://www.gnu.org/licenses/>.
+
+import numpy as np
+import json
+import os
+import argparse
+import collections
+import pprint
+
+class MigrationFile(object):
+ def __init__(self, filename):
+ self.filename = filename
+ self.file = open(self.filename, "rb")
+
+ def read64(self):
+ return np.asscalar(np.fromfile(self.file, count=1, dtype='>i8')[0])
+
+ def read32(self):
+ return np.asscalar(np.fromfile(self.file, count=1, dtype='>i4')[0])
+
+ def read16(self):
+ return np.asscalar(np.fromfile(self.file, count=1, dtype='>i2')[0])
+
+ def read8(self):
+ return np.asscalar(np.fromfile(self.file, count=1, dtype='>i1')[0])
+
+ def readstr(self, len = None):
+ if len is None:
+ len = self.read8()
+ if len == 0:
+ return ""
+ return np.fromfile(self.file, count=1, dtype=('S%d' % len))[0]
+
+ def readvar(self, size = None):
+ if size is None:
+ size = self.read8()
+ if size == 0:
+ return ""
+ value = self.file.read(size)
+ if len(value) != size:
+ raise Exception("Unexpected end of %s at 0x%x" % (self.filename, self.file.tell()))
+ return value
+
+ # The VMSD description is at the end of the file, after EOF. Look for
+ # the last NULL byte, then for the beginning brace of JSON.
+ def read_migration_debug_json(self):
+ QEMU_VM_VMDESCRIPTION = 0x06
+
+ # Remember the offset in the file when we started
+ entrypos = self.file.tell()
+
+ # Read the last 10MB
+ self.file.seek(0, os.SEEK_END)
+ endpos = self.file.tell()
+ self.file.seek(max(-endpos, -10 * 1024 * 1024), os.SEEK_END)
+ datapos = self.file.tell()
+ data = self.file.read()
+ # The full file read closed the file as well, reopen it
+ self.file = open(self.filename, "rb")
+
+ # Find the last NULL byte, then the first brace after that. This should
+ # be the beginning of our JSON data.
+ nulpos = data.rfind("\0")
+ jsonpos = data.find("{", nulpos)
+
+ # Check backwards from there and see whether we guessed right
+ self.file.seek(datapos + jsonpos - 5, 0)
+ if self.read8() != QEMU_VM_VMDESCRIPTION:
+ raise Exception("No Debug Migration device found")
+
+ jsonlen = self.read32()
+
+ # Seek back to where we were at the beginning
+ self.file.seek(entrypos, 0)
+
+ return data[jsonpos:jsonpos + jsonlen]
+
+ def close(self):
+ self.file.close()
+
+
+class RamSection(object):
+ RAM_SAVE_FLAG_COMPRESS = 0x02
+ RAM_SAVE_FLAG_MEM_SIZE = 0x04
+ RAM_SAVE_FLAG_PAGE = 0x08
+ RAM_SAVE_FLAG_EOS = 0x10
+ RAM_SAVE_FLAG_CONTINUE = 0x20
+ RAM_SAVE_FLAG_XBZRLE = 0x40
+ RAM_SAVE_FLAG_HOOK = 0x80
+ blocks = []
+
+ def __init__(self, file, version_id, device, section_key):
+ if version_id != 4:
+ raise Exception("Unknown RAM version %d" % version_id)
+
+ self.file = file
+ self.section_key = section_key
+ self.TARGET_PAGE_SIZE = device
+
+ def read(self):
+ # Read all RAM sections
+ while True:
+ addr = self.file.read64()
+ flags = addr & (self.TARGET_PAGE_SIZE - 1)
+ addr &= ~(self.TARGET_PAGE_SIZE - 1)
+
+ if flags & self.RAM_SAVE_FLAG_MEM_SIZE:
+ while True:
+ namelen = self.file.read8()
+ # We assume that no RAM chunk is big enough to ever
+ # hit the first byte of the address, so when we see
+ # a zero here we know it has to be an address, not the
+ # length of the next block.
+ if namelen == 0:
+ self.file.file.seek(-1, 1)
+ break
+ name = self.file.readstr(len = namelen)
+ len = self.file.read64()
+ self.blocks.append((name, len))
+ flags &= ~self.RAM_SAVE_FLAG_MEM_SIZE
+
+ if flags & self.RAM_SAVE_FLAG_COMPRESS:
+ if flags & self.RAM_SAVE_FLAG_CONTINUE:
+ flags &= ~self.RAM_SAVE_FLAG_CONTINUE
+ else:
+ name = self.file.readstr()
+ fill_char = self.file.read8()
+ # The page in question would be filled with fill_char now
+ flags &= ~self.RAM_SAVE_FLAG_COMPRESS
+ elif flags & self.RAM_SAVE_FLAG_PAGE:
+ if flags & self.RAM_SAVE_FLAG_CONTINUE:
+ flags &= ~self.RAM_SAVE_FLAG_CONTINUE
+ else:
+ name = self.file.readstr()
+ # Just skip RAM data for now
+ self.file.file.seek(self.TARGET_PAGE_SIZE, 1)
+ flags &= ~self.RAM_SAVE_FLAG_PAGE
+ elif flags & self.RAM_SAVE_FLAG_XBZRLE:
+ raise Exception("XBZRLE RAM compression is not supported yet")
+ elif flags & self.RAM_SAVE_FLAG_HOOK:
+ raise Exception("RAM hooks don't make sense with files")
+
+ # End of RAM section
+ if flags & self.RAM_SAVE_FLAG_EOS:
+ break
+
+ if flags != 0:
+ raise Exception("Unknown RAM flags: %x" % flags)
+
+ def getDict(self):
+ return ""
+
+class HTABSection(object):
+ HASH_PTE_SIZE_64 = 16
+
+ def __init__(self, file, version_id, device, section_key):
+ if version_id != 1:
+ raise Exception("Unknown HTAB version %d" % version_id)
+
+ self.file = file
+ self.section_key = section_key
+
+ def read(self):
+
+ header = self.file.read32()
+
+ if (header > 0):
+ # First section, just the hash shift
+ return
+
+ # Read until end marker
+ while True:
+ index = self.file.read32()
+ n_valid = self.file.read16()
+ n_invalid = self.file.read16()
+
+ if index == 0 and n_valid == 0 and n_invalid == 0:
+ break
+
+ self.file.readvar(n_valid * HASH_PTE_SIZE_64)
+
+ def getDict(self):
+ return ""
+
+class VMSDFieldGeneric(object):
+ def __init__(self, desc, file):
+ self.file = file
+ self.desc = desc
+ self.data = ""
+
+ def __repr__(self):
+ return str(self.__str__())
+
+ def __str__(self):
+ return " ".join("{0:02x}".format(ord(c)) for c in self.data)
+
+ def getDict(self):
+ return self.__str__()
+
+ def read(self):
+ size = int(self.desc['size'])
+ self.data = self.file.readvar(size)
+ return self.data
+
+class VMSDFieldInt(VMSDFieldGeneric):
+ def __init__(self, desc, file):
+ super(VMSDFieldInt, self).__init__(desc, file)
+ self.size = int(desc['size'])
+ self.format = '0x%%0%dx' % (self.size * 2)
+ self.sdtype = '>i%d' % self.size
+ self.udtype = '>u%d' % self.size
+
+ def __repr__(self):
+ if self.data < 0:
+ return ('%s (%d)' % ((self.format % self.udata), self.data))
+ else:
+ return self.format % self.data
+
+ def __str__(self):
+ return self.__repr__()
+
+ def getDict(self):
+ return self.__str__()
+
+ def read(self):
+ super(VMSDFieldInt, self).read()
+ self.sdata = np.fromstring(self.data, count=1, dtype=(self.sdtype))[0]
+ self.udata = np.fromstring(self.data, count=1, dtype=(self.udtype))[0]
+ self.data = self.sdata
+ return self.data
+
+class VMSDFieldUInt(VMSDFieldInt):
+ def __init__(self, desc, file):
+ super(VMSDFieldUInt, self).__init__(desc, file)
+
+ def read(self):
+ super(VMSDFieldUInt, self).read()
+ self.data = self.udata
+ return self.data
+
+class VMSDFieldIntLE(VMSDFieldInt):
+ def __init__(self, desc, file):
+ super(VMSDFieldIntLE, self).__init__(desc, file)
+ self.dtype = '<i%d' % self.size
+
+class VMSDFieldBool(VMSDFieldGeneric):
+ def __init__(self, desc, file):
+ super(VMSDFieldBool, self).__init__(desc, file)
+
+ def __repr__(self):
+ return self.data.__repr__()
+
+ def __str__(self):
+ return self.data.__str__()
+
+ def getDict(self):
+ return self.data
+
+ def read(self):
+ super(VMSDFieldBool, self).read()
+ if self.data[0] == 0:
+ self.data = False
+ else:
+ self.data = True
+ return self.data
+
+class VMSDFieldStruct(VMSDFieldGeneric):
+ QEMU_VM_SUBSECTION = 0x05
+
+ def __init__(self, desc, file):
+ super(VMSDFieldStruct, self).__init__(desc, file)
+ self.data = collections.OrderedDict()
+
+ # When we see compressed array elements, unfold them here
+ new_fields = []
+ for field in self.desc['struct']['fields']:
+ if not 'array_len' in field:
+ new_fields.append(field)
+ continue
+ array_len = field.pop('array_len')
+ field['index'] = 0
+ new_fields.append(field)
+ for i in xrange(1, array_len):
+ c = field.copy()
+ c['index'] = i
+ new_fields.append(c)
+
+ self.desc['struct']['fields'] = new_fields
+
+ def __repr__(self):
+ return self.data.__repr__()
+
+ def __str__(self):
+ return self.data.__str__()
+
+ def read(self):
+ for field in self.desc['struct']['fields']:
+ try:
+ reader = vmsd_field_readers[field['type']]
+ except:
+ reader = VMSDFieldGeneric
+
+ field['data'] = reader(field, self.file)
+ field['data'].read()
+
+ if 'index' in field:
+ if field['name'] not in self.data:
+ self.data[field['name']] = []
+ a = self.data[field['name']]
+ if len(a) != int(field['index']):
+ raise Exception("internal index of data field unmatched (%d/%d)" % (len(a), int(field['index'])))
+ a.append(field['data'])
+ else:
+ self.data[field['name']] = field['data']
+
+ if 'subsections' in self.desc['struct']:
+ for subsection in self.desc['struct']['subsections']:
+ if self.file.read8() != self.QEMU_VM_SUBSECTION:
+ raise Exception("Subsection %s not found" % subsection['vmsd_name'])
+ name = self.file.readstr()
+ version_id = self.file.read32()
+ self.data[name] = VMSDSection(self.file, version_id, subsection, (name, 0))
+ self.data[name].read()
+
+ def getDictItem(self, value):
+ # Strings would fall into the array category, treat
+ # them specially
+ if value.__class__ is ''.__class__:
+ return value
+
+ try:
+ return self.getDictOrderedDict(value)
+ except:
+ try:
+ return self.getDictArray(value)
+ except:
+ try:
+ return value.getDict()
+ except:
+ return value
+
+ def getDictArray(self, array):
+ r = []
+ for value in array:
+ r.append(self.getDictItem(value))
+ return r
+
+ def getDictOrderedDict(self, dict):
+ r = collections.OrderedDict()
+ for (key, value) in dict.items():
+ r[key] = self.getDictItem(value)
+ return r
+
+ def getDict(self):
+ return self.getDictOrderedDict(self.data)
+
+vmsd_field_readers = {
+ "bool" : VMSDFieldBool,
+ "int8" : VMSDFieldInt,
+ "int16" : VMSDFieldInt,
+ "int32" : VMSDFieldInt,
+ "int32 equal" : VMSDFieldInt,
+ "int32 le" : VMSDFieldIntLE,
+ "int64" : VMSDFieldInt,
+ "uint8" : VMSDFieldUInt,
+ "uint16" : VMSDFieldUInt,
+ "uint32" : VMSDFieldUInt,
+ "uint32 equal" : VMSDFieldUInt,
+ "uint64" : VMSDFieldUInt,
+ "int64 equal" : VMSDFieldInt,
+ "uint8 equal" : VMSDFieldInt,
+ "uint16 equal" : VMSDFieldInt,
+ "float64" : VMSDFieldGeneric,
+ "timer" : VMSDFieldGeneric,
+ "buffer" : VMSDFieldGeneric,
+ "unused_buffer" : VMSDFieldGeneric,
+ "bitmap" : VMSDFieldGeneric,
+ "struct" : VMSDFieldStruct,
+ "unknown" : VMSDFieldGeneric,
+}
+
+class VMSDSection(VMSDFieldStruct):
+ def __init__(self, file, version_id, device, section_key):
+ self.file = file
+ self.data = ""
+ self.vmsd_name = ""
+ self.section_key = section_key
+ desc = device
+ if 'vmsd_name' in device:
+ self.vmsd_name = device['vmsd_name']
+
+ # A section really is nothing but a FieldStruct :)
+ super(VMSDSection, self).__init__({ 'struct' : desc }, file)
+
+###############################################################################
+
+class MigrationDump(object):
+ QEMU_VM_FILE_MAGIC = 0x5145564d
+ QEMU_VM_FILE_VERSION = 0x00000003
+ QEMU_VM_EOF = 0x00
+ QEMU_VM_SECTION_START = 0x01
+ QEMU_VM_SECTION_PART = 0x02
+ QEMU_VM_SECTION_END = 0x03
+ QEMU_VM_SECTION_FULL = 0x04
+ QEMU_VM_SUBSECTION = 0x05
+ QEMU_VM_VMDESCRIPTION = 0x06
+
+ def __init__(self, filename):
+ self.section_classes = { ( 'ram', 0 ) : [ RamSection, None ],
+ ( 'spapr/htab', 0) : ( HTABSection, None ) }
+ self.filename = filename
+ self.vmsd_desc = None
+
+ def read(self, desc_only = False):
+ # Read in the whole file
+ file = MigrationFile(self.filename)
+
+ # File magic
+ data = file.read32()
+ if data != self.QEMU_VM_FILE_MAGIC:
+ raise Exception("Invalid file magic %x" % data)
+
+ # Version (has to be v3)
+ data = file.read32()
+ if data != self.QEMU_VM_FILE_VERSION:
+ raise Exception("Invalid version number %d" % data)
+
+ self.load_vmsd_json(file)
+
+ # Read sections
+ self.sections = collections.OrderedDict()
+
+ if desc_only:
+ return
+
+ self.section_classes[('ram',0)][1] = self.vmsd_desc['page_size']
+
+ while True:
+ section_type = file.read8()
+ if section_type == self.QEMU_VM_EOF:
+ break
+ elif section_type == self.QEMU_VM_SECTION_START or section_type == self.QEMU_VM_SECTION_FULL:
+ section_id = file.read32()
+ name = file.readstr()
+ instance_id = file.read32()
+ version_id = file.read32()
+ section_key = (name, instance_id)
+ classdesc = self.section_classes[section_key]
+ section = classdesc[0](file, version_id, classdesc[1], section_key)
+ self.sections[section_id] = section
+ section.read()
+ elif section_type == self.QEMU_VM_SECTION_PART or section_type == self.QEMU_VM_SECTION_END:
+ section_id = file.read32()
+ self.sections[section_id].read()
+ else:
+ raise Exception("Unknown section type: %d" % section_type)
+ file.close()
+
+ def load_vmsd_json(self, file):
+ vmsd_json = file.read_migration_debug_json()
+ self.vmsd_desc = json.loads(vmsd_json, object_pairs_hook=collections.OrderedDict)
+ for device in self.vmsd_desc['devices']:
+ key = (device['name'], device['instance_id'])
+ value = ( VMSDSection, device )
+ self.section_classes[key] = value
+
+ def getDict(self):
+ r = collections.OrderedDict()
+ for (key, value) in self.sections.items():
+ key = "%s (%d)" % ( value.section_key[0], key )
+ r[key] = value.getDict()
+ return r
+
+###############################################################################
+
+class JSONEncoder(json.JSONEncoder):
+ def default(self, o):
+ if isinstance(o, VMSDFieldGeneric):
+ return str(o)
+ return json.JSONEncoder.default(self, o)
+
+parser = argparse.ArgumentParser()
+parser.add_argument("-f", "--file", help='migration dump to read from', required=True)
+parser.add_argument("-s", "--descriptionfile", help='migration dump to read vmstate description from')
+parser.add_argument("-d", "--dump", help='what to dump ("state" or "desc")', default='state')
+args = parser.parse_args()
+
+jsonenc = JSONEncoder(indent=4, separators=(',', ': '))
+
+if args.dump == "state":
+ dump = MigrationDump(args.file)
+ if args.descriptionfile:
+ # Fetch the vmstate description from the file passed through -s
+ desc_dump = MigrationDump(args.descriptionfile)
+ desc_dump.read(desc_only = True)
+ # and override all section readers and vmsd description in our
+ # data migration file with the ones from the vmstate description
+ # migration file
+ dump.vmsd_desc = desc_dump.vmsd_desc
+ dump.section_classes = desc_dump.section_classes
+ dump.read()
+ dict = dump.getDict()
+ print jsonenc.encode(dict)
+elif args.dump == "desc":
+ if args.descriptionfile:
+ dump = MigrationDump(args.descriptionfile)
+ else:
+ dump = MigrationDump(args.file)
+ dump.read(desc_only = True)
+ print jsonenc.encode(dump.vmsd_desc)
+else:
+ raise Exception("Unknown dump type \"%s\", available: \"state\", \"desc\"" % args.dump)
--
1.7.12.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
end of thread, other threads:[~2014-12-22 23:04 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-12-22 23:03 [Qemu-devel] [PATCH 0/4] Migration Deciphering aid Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 1/4] QJSON: Add JSON writer Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 2/4] qemu-file: Add fast ftell code path Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 3/4] migration: Append JSON description of migration stream Alexander Graf
2014-12-22 23:03 ` [Qemu-devel] [PATCH 4/4] Add migration stream analyzation script Alexander Graf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).