Discussion:
[PATCH 10/79] block: Do not unref bs->file on error in BD's open
(too old to reply)
Michael Roth
2017-08-29 00:13:45 UTC
Permalink
From: Max Reitz <***@redhat.com>

The block layer takes care of removing the bs->file child if the block
driver's bdrv_open()/bdrv_file_open() implementation fails. The block
driver therefore does not need to do so, and indeed should not unless it
sets bs->file to NULL afterwards -- because if this is not done, the
bdrv_unref_child() in bdrv_open_inherit() will dereference the freed
memory block at bs->file afterwards, which is not good.

We can now decide whether to add a "bs->file = NULL;" after each of the
offending bdrv_unref_child() invocations, or just drop them altogether.
The latter is simpler, so let's do that.

Cc: qemu-stable <qemu-***@nongnu.org>
Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit de234897b60e034ba94b307fc289e2dc692c9251)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 4 +---
block/blkreplay.c | 3 ---
block/blkverify.c | 3 ---
3 files changed, 1 insertion(+), 9 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index 67e8024..cc4a146 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -389,14 +389,12 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
} else if (align) {
error_setg(errp, "Invalid alignment");
ret = -EINVAL;
- goto fail_unref;
+ goto out;
}

ret = 0;
goto out;

-fail_unref:
- bdrv_unref_child(bs, bs->file);
out:
if (ret < 0) {
g_free(s->config_file);
diff --git a/block/blkreplay.c b/block/blkreplay.c
index e110211..6aa5fd4 100755
--- a/block/blkreplay.c
+++ b/block/blkreplay.c
@@ -37,9 +37,6 @@ static int blkreplay_open(BlockDriverState *bs, QDict *options, int flags,

ret = 0;
fail:
- if (ret < 0) {
- bdrv_unref_child(bs, bs->file);
- }
return ret;
}

diff --git a/block/blkverify.c b/block/blkverify.c
index 9a1e21c..af23281 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -142,9 +142,6 @@ static int blkverify_open(BlockDriverState *bs, QDict *options, int flags,

ret = 0;
fail:
- if (ret < 0) {
- bdrv_unref_child(bs, bs->file);
- }
qemu_opts_del(opts);
return ret;
}
--
2.7.4
Michael Roth
2017-08-29 00:13:52 UTC
Permalink
From: Eric Blake <***@redhat.com>

An upcoming Coccinelle cleanup script wanted to reformat the casts
present in this file - but on closer look, we don't need the casts
at all because C automatically converts void* to any other pointer.

Signed-off-by: Eric Blake <***@redhat.com>
Message-Id: <20170405194741.18956-4-***@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <***@amsat.org>
Signed-off-by: Cornelia Huck <***@de.ibm.com>
(cherry picked from commit cb55c19a26d6928b445f1bd9394843c9eb60f8ab)
* prereq for 46f5ac2
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
target/s390x/cpu_models.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/target/s390x/cpu_models.c b/target/s390x/cpu_models.c
index 1434d15..ce461cc 100644
--- a/target/s390x/cpu_models.c
+++ b/target/s390x/cpu_models.c
@@ -376,12 +376,12 @@ static void cpu_model_from_info(S390CPUModel *model, const CpuModelInfo *info,

static void qdict_add_disabled_feat(const char *name, void *opaque)
{
- qdict_put((QDict *) opaque, name, qbool_from_bool(false));
+ qdict_put(opaque, name, qbool_from_bool(false));
}

static void qdict_add_enabled_feat(const char *name, void *opaque)
{
- qdict_put((QDict *) opaque, name, qbool_from_bool(true));
+ qdict_put(opaque, name, qbool_from_bool(true));
}

/* convert S390CPUDef into a static CpuModelInfo */
--
2.7.4
Michael Roth
2017-08-29 00:13:46 UTC
Permalink
From: Bruce Rogers <***@suse.com>

Commit f0c9d64a exposed the issue that with a xenfv machine using
pci passthrough, acpi pci hotplug code was being executed by mistake.
Guard calls to acpi_pcihp_device_plug_cb (and corresponding
acpi_pcihp_device_unplug_cb) with a check for xen_enabled(). Without
this check I am seeing an error that the bus doesn't have the
acpi-pcihp-bsel property set.

Signed-off-by: Bruce Rogers <***@suse.com>
Reviewed-by: Igor Mammedov <***@redhat.com>
Reviewed-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Michael S. Tsirkin <***@redhat.com>
(cherry picked from commit 153eba4726dfa1bdfc31d1fe973b2a61b9035492)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/acpi/piix4.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index a553a7e..c409374 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -385,7 +385,10 @@ static void piix4_device_plug_cb(HotplugHandler *hotplug_dev,
dev, errp);
}
} else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
- acpi_pcihp_device_plug_cb(hotplug_dev, &s->acpi_pci_hotplug, dev, errp);
+ if (!xen_enabled()) {
+ acpi_pcihp_device_plug_cb(hotplug_dev, &s->acpi_pci_hotplug, dev,
+ errp);
+ }
} else if (object_dynamic_cast(OBJECT(dev), TYPE_CPU)) {
if (s->cpu_hotplug_legacy) {
legacy_acpi_cpu_plug_cb(hotplug_dev, &s->gpe_cpu, dev, errp);
@@ -408,8 +411,10 @@ static void piix4_device_unplug_request_cb(HotplugHandler *hotplug_dev,
acpi_memory_unplug_request_cb(hotplug_dev, &s->acpi_memory_hotplug,
dev, errp);
} else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
- acpi_pcihp_device_unplug_cb(hotplug_dev, &s->acpi_pci_hotplug, dev,
- errp);
+ if (!xen_enabled()) {
+ acpi_pcihp_device_unplug_cb(hotplug_dev, &s->acpi_pci_hotplug, dev,
+ errp);
+ }
} else if (object_dynamic_cast(OBJECT(dev), TYPE_CPU) &&
!s->cpu_hotplug_legacy) {
acpi_cpu_unplug_request_cb(hotplug_dev, &s->cpuhp_state, dev, errp);
--
2.7.4
Michael Roth
2017-08-29 00:13:44 UTC
Permalink
From: "Herongguang (Stephen)" <***@huawei.com>

If a pci device is not reset by VM (by writing into config space)
and unplugged by VM, after that when VM reboots, qemu may assert:
pcibus_reset: Assertion `bus->irq_count[i] == 0' failed

Cc: qemu-***@nongnu.org
Signed-off-by: herongguang <***@huawei.com>
Reviewed-by: Marcel Apfelbaum <***@redhat.com>
Reviewed-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Michael S. Tsirkin <***@redhat.com>
(cherry picked from commit 3936161f1fd72d1dfa577aaba910819c5e873260)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/pci/pci.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/hw/pci/pci.c b/hw/pci/pci.c
index 259483b..98ccc27 100644
--- a/hw/pci/pci.c
+++ b/hw/pci/pci.c
@@ -1083,6 +1083,7 @@ static void pci_qdev_unrealize(DeviceState *dev, Error **errp)
pc->exit(pci_dev);
}

+ pci_device_deassert_intx(pci_dev);
do_pci_unregister_device(pci_dev);
}
--
2.7.4
Michael Roth
2017-08-29 00:13:58 UTC
Permalink
From: Max Reitz <***@redhat.com>

This patch makes vhdx_create() always set errp in case of an error. It
also adds errp parameters to vhdx_create_bat() and
vhdx_create_new_region_table() so we can pass on the error object
generated by blk_truncate() as of a future commit.

Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Kevin Wolf <***@redhat.com>
Message-id: 20170328205129.15138-2-***@redhat.com
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit 55b9392b98e500399f2da1edc1d110bbfd40fb05)
* prereq for 698bdfa
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/vhdx.c | 23 +++++++++++++++++++----
1 file changed, 19 insertions(+), 4 deletions(-)

diff --git a/block/vhdx.c b/block/vhdx.c
index 052a753..d25bcd9 100644
--- a/block/vhdx.c
+++ b/block/vhdx.c
@@ -1586,7 +1586,7 @@ exit:
static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s,
uint64_t image_size, VHDXImageType type,
bool use_zero_blocks, uint64_t file_offset,
- uint32_t length)
+ uint32_t length, Error **errp)
{
int ret = 0;
uint64_t data_file_offset;
@@ -1609,14 +1609,19 @@ static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s,
* is the furthest thing we have written yet */
ret = blk_truncate(blk, data_file_offset);
if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "Failed to resize the underlying file");
goto exit;
}
} else if (type == VHDX_TYPE_FIXED) {
ret = blk_truncate(blk, data_file_offset + image_size);
if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "Failed to resize the underlying file");
goto exit;
}
} else {
+ error_setg(errp, "Unsupported image type");
ret = -ENOTSUP;
goto exit;
}
@@ -1627,6 +1632,7 @@ static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s,
/* for a fixed file, the default BAT entry is not zero */
s->bat = g_try_malloc0(length);
if (length && s->bat == NULL) {
+ error_setg(errp, "Failed to allocate memory for the BAT");
ret = -ENOMEM;
goto exit;
}
@@ -1646,6 +1652,7 @@ static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s,
}
ret = blk_pwrite(blk, file_offset, s->bat, length, 0);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to write the BAT");
goto exit;
}
}
@@ -1671,7 +1678,8 @@ static int vhdx_create_new_region_table(BlockBackend *blk,
uint32_t log_size,
bool use_zero_blocks,
VHDXImageType type,
- uint64_t *metadata_offset)
+ uint64_t *metadata_offset,
+ Error **errp)
{
int ret = 0;
uint32_t offset = 0;
@@ -1740,7 +1748,7 @@ static int vhdx_create_new_region_table(BlockBackend *blk,
/* The region table gives us the data we need to create the BAT,
* so do that now */
ret = vhdx_create_bat(blk, s, image_size, type, use_zero_blocks,
- bat_file_offset, bat_length);
+ bat_file_offset, bat_length, errp);
if (ret < 0) {
goto exit;
}
@@ -1749,12 +1757,14 @@ static int vhdx_create_new_region_table(BlockBackend *blk,
ret = blk_pwrite(blk, VHDX_REGION_TABLE_OFFSET, buffer,
VHDX_HEADER_BLOCK_SIZE, 0);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to write first region table");
goto exit;
}

ret = blk_pwrite(blk, VHDX_REGION_TABLE2_OFFSET, buffer,
VHDX_HEADER_BLOCK_SIZE, 0);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to write second region table");
goto exit;
}

@@ -1825,6 +1835,7 @@ static int vhdx_create(const char *filename, QemuOpts *opts, Error **errp)
ret = -ENOTSUP;
goto exit;
} else {
+ error_setg(errp, "Invalid subformat '%s'", type);
ret = -EINVAL;
goto exit;
}
@@ -1879,12 +1890,14 @@ static int vhdx_create(const char *filename, QemuOpts *opts, Error **errp)
ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET, &signature, sizeof(signature),
0);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to write file signature");
goto delete_and_exit;
}
if (creator) {
ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET + sizeof(signature),
creator, creator_items * sizeof(gunichar2), 0);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to write creator field");
goto delete_and_exit;
}
}
@@ -1893,13 +1906,14 @@ static int vhdx_create(const char *filename, QemuOpts *opts, Error **errp)
/* Creates (B),(C) */
ret = vhdx_create_new_headers(blk, image_size, log_size);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to write image headers");
goto delete_and_exit;
}

/* Creates (D),(E),(G) explicitly. (F) created as by-product */
ret = vhdx_create_new_region_table(blk, image_size, block_size, 512,
log_size, use_zero_blocks, image_type,
- &metadata_offset);
+ &metadata_offset, errp);
if (ret < 0) {
goto delete_and_exit;
}
@@ -1908,6 +1922,7 @@ static int vhdx_create(const char *filename, QemuOpts *opts, Error **errp)
ret = vhdx_create_new_metadata(blk, image_size, block_size, 512,
metadata_offset, image_type);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Failed to initialize metadata");
goto delete_and_exit;
}
--
2.7.4
Michael Roth
2017-08-29 00:13:49 UTC
Permalink
From: Eric Blake <***@redhat.com>

We have macros in place to make it less verbose to add a subtype
of QObject to both QDict and QList. While we have made cleanups
like this in the past (see commit fcfcd8ffc, for example), having
it be automated by Coccinelle makes it easier to maintain.

The script is separate from the cleanups, for ease of review and
backporting. A later patch will then add further possible cleanups.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Markus Armbruster <***@redhat.com>
Message-Id: <20170427215821.19397-4-***@redhat.com>
Signed-off-by: Markus Armbruster <***@redhat.com>
(cherry picked from commit a2f3453ebc64b1ebe094aeaf83f9e67896f90ac3)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
MAINTAINERS | 1 +
scripts/coccinelle/qobject.cocci | 13 +++++++++++++
2 files changed, 14 insertions(+)
create mode 100644 scripts/coccinelle/qobject.cocci

diff --git a/MAINTAINERS b/MAINTAINERS
index c60235e..430efb0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1395,6 +1395,7 @@ S: Supported
F: qobject/
F: include/qapi/qmp/
X: include/qapi/qmp/dispatch.h
+F: scripts/coccinelle/qobject.cocci
F: tests/check-qdict.c
F: tests/check-qfloat.c
F: tests/check-qint.c
diff --git a/scripts/coccinelle/qobject.cocci b/scripts/coccinelle/qobject.cocci
new file mode 100644
index 0000000..aa899e2
--- /dev/null
+++ b/scripts/coccinelle/qobject.cocci
@@ -0,0 +1,13 @@
+// Use QDict macros where they make sense
+@@
+expression Obj, Key, E;
+@@
+- qdict_put_obj(Obj, Key, QOBJECT(E));
++ qdict_put(Obj, Key, E);
+
+// Use QList macros where they make sense
+@@
+expression Obj, E;
+@@
+- qlist_append_obj(Obj, QOBJECT(E));
++ qlist_append(Obj, E);
--
2.7.4
Michael Roth
2017-08-29 00:13:47 UTC
Permalink
From: Markus Armbruster <***@redhat.com>

Broken in commit daa33c5.

Cc: qemu-***@nongnu.org
Signed-off-by: Markus Armbruster <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Zhang Chen <***@cn.fujitsu.com>
Message-id: 1493298053-17140-1-git-send-email-***@redhat.com
Signed-off-by: Stefan Hajnoczi <***@redhat.com>
(cherry picked from commit 38bb54f323bf7c83496b6a044cfd28896e997a00)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
migration/colo.c | 12 ++++++++++++
monitor.c | 5 +++++
2 files changed, 17 insertions(+)

diff --git a/migration/colo.c b/migration/colo.c
index c19eb3f..963c802 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -153,6 +153,7 @@ void qmp_xen_set_replication(bool enable, bool primary,
bool has_failover, bool failover,
Error **errp)
{
+#ifdef CONFIG_REPLICATION
ReplicationMode mode = primary ?
REPLICATION_MODE_PRIMARY :
REPLICATION_MODE_SECONDARY;
@@ -171,10 +172,14 @@ void qmp_xen_set_replication(bool enable, bool primary,
}
replication_stop_all(failover, failover ? NULL : errp);
}
+#else
+ abort();
+#endif
}

ReplicationStatus *qmp_query_xen_replication_status(Error **errp)
{
+#ifdef CONFIG_REPLICATION
Error *err = NULL;
ReplicationStatus *s = g_new0(ReplicationStatus, 1);

@@ -189,11 +194,18 @@ ReplicationStatus *qmp_query_xen_replication_status(Error **errp)

error_free(err);
return s;
+#else
+ abort();
+#endif
}

void qmp_xen_colo_do_checkpoint(Error **errp)
{
+#ifdef CONFIG_REPLICATION
replication_do_checkpoint_all(errp);
+#else
+ abort();
+#endif
}

static void colo_send_message(QEMUFile *f, COLOMessage msg,
diff --git a/monitor.c b/monitor.c
index be282ec..2799a37 100644
--- a/monitor.c
+++ b/monitor.c
@@ -974,6 +974,11 @@ static void qmp_unregister_commands_hack(void)
#ifndef CONFIG_SPICE
qmp_unregister_command(&qmp_commands, "query-spice");
#endif
+#ifndef CONFIG_REPLICATION
+ qmp_unregister_command(&qmp_commands, "xen-set-replication");
+ qmp_unregister_command(&qmp_commands, "query-xen-replication-status");
+ qmp_unregister_command(&qmp_commands, "xen-colo-do-checkpoint");
+#endif
#ifndef TARGET_I386
qmp_unregister_command(&qmp_commands, "rtc-reset-reinjection");
#endif
--
2.7.4
Michael Roth
2017-08-29 00:13:51 UTC
Permalink
From: Eric Blake <***@redhat.com>

Rather than making lots of callers wrap a scalar in a QInt, QString,
or QBool, provide helper macros that do the wrapping automatically.

Update the Coccinelle script to make mass conversions easy, although
the conversion itself will be done as a separate patches to ease
review and backport efforts.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Markus Armbruster <***@redhat.com>
Message-Id: <20170427215821.19397-6-***@redhat.com>
Signed-off-by: Markus Armbruster <***@redhat.com>
(cherry picked from commit a92c21591b5bb9543996538f14854ca6b528318b)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
include/qapi/qmp/qdict.h | 8 ++++++++
include/qapi/qmp/qlist.h | 8 ++++++++
scripts/coccinelle/qobject.cocci | 22 ++++++++++++++++++++++
3 files changed, 38 insertions(+)

diff --git a/include/qapi/qmp/qdict.h b/include/qapi/qmp/qdict.h
index fe9a4c5..188440a 100644
--- a/include/qapi/qmp/qdict.h
+++ b/include/qapi/qmp/qdict.h
@@ -52,6 +52,14 @@ void qdict_destroy_obj(QObject *obj);
#define qdict_put(qdict, key, obj) \
qdict_put_obj(qdict, key, QOBJECT(obj))

+/* Helpers for int, bool, and string */
+#define qdict_put_int(qdict, key, value) \
+ qdict_put(qdict, key, qint_from_int(value))
+#define qdict_put_bool(qdict, key, value) \
+ qdict_put(qdict, key, qbool_from_bool(value))
+#define qdict_put_str(qdict, key, value) \
+ qdict_put(qdict, key, qstring_from_str(value))
+
/* High level helpers */
double qdict_get_double(const QDict *qdict, const char *key);
int64_t qdict_get_int(const QDict *qdict, const char *key);
diff --git a/include/qapi/qmp/qlist.h b/include/qapi/qmp/qlist.h
index a84117e..5dc4ed9 100644
--- a/include/qapi/qmp/qlist.h
+++ b/include/qapi/qmp/qlist.h
@@ -29,6 +29,14 @@ typedef struct QList {
#define qlist_append(qlist, obj) \
qlist_append_obj(qlist, QOBJECT(obj))

+/* Helpers for int, bool, and string */
+#define qlist_append_int(qlist, value) \
+ qlist_append(qlist, qint_from_int(value))
+#define qlist_append_bool(qlist, value) \
+ qlist_append(qlist, qbool_from_bool(value))
+#define qlist_append_str(qlist, value) \
+ qlist_append(qlist, qstring_from_str(value))
+
#define QLIST_FOREACH_ENTRY(qlist, var) \
for ((var) = ((qlist)->head.tqh_first); \
(var); \
diff --git a/scripts/coccinelle/qobject.cocci b/scripts/coccinelle/qobject.cocci
index aa899e2..97703a4 100644
--- a/scripts/coccinelle/qobject.cocci
+++ b/scripts/coccinelle/qobject.cocci
@@ -2,12 +2,34 @@
@@
expression Obj, Key, E;
@@
+(
- qdict_put_obj(Obj, Key, QOBJECT(E));
+ qdict_put(Obj, Key, E);
+|
+- qdict_put(Obj, Key, qint_from_int(E));
++ qdict_put_int(Obj, Key, E);
+|
+- qdict_put(Obj, Key, qbool_from_bool(E));
++ qdict_put_bool(Obj, Key, E);
+|
+- qdict_put(Obj, Key, qstring_from_str(E));
++ qdict_put_str(Obj, Key, E);
+)

// Use QList macros where they make sense
@@
expression Obj, E;
@@
+(
- qlist_append_obj(Obj, QOBJECT(E));
+ qlist_append(Obj, E);
+|
+- qlist_append(Obj, qint_from_int(E));
++ qlist_append_int(Obj, E);
+|
+- qlist_append(Obj, qbool_from_bool(E));
++ qlist_append_bool(Obj, E);
+|
+- qlist_append(Obj, qstring_from_str(E));
++ qlist_append_str(Obj, E);
+)
--
2.7.4
Michael Roth
2017-08-29 00:13:54 UTC
Permalink
From: Fam Zheng <***@redhat.com>

Opening the backing image for the second time is bad, especially here
when it is also in use as the active image as the source. The
drive-backup job itself doesn't read from target->backing for COW,
instead it gets data from the write notifier, so it's not a big problem.
However, exporting the target to NBD etc. won't work, because of the
likely stale metadata cache.

Use BDRV_O_NO_BACKING in this case and manually set up the backing
BdrvChild.

Cc: qemu-***@nongnu.org
Signed-off-by: Fam Zheng <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit fc0932fdcfc3e5cafa3641e361b681c07f639812)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
blockdev.c | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/blockdev.c b/blockdev.c
index e2f9c1e..841200e 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -3170,6 +3170,7 @@ static BlockJob *do_drive_backup(DriveBackup *backup, BlockJobTxn *txn,
Error *local_err = NULL;
int flags;
int64_t size;
+ bool set_backing_hd = false;

if (!backup->has_speed) {
backup->speed = 0;
@@ -3220,6 +3221,8 @@ static BlockJob *do_drive_backup(DriveBackup *backup, BlockJobTxn *txn,
}
if (backup->sync == MIRROR_SYNC_MODE_NONE) {
source = bs;
+ flags |= BDRV_O_NO_BACKING;
+ set_backing_hd = true;
}

size = bdrv_getlength(bs);
@@ -3246,7 +3249,9 @@ static BlockJob *do_drive_backup(DriveBackup *backup, BlockJobTxn *txn,
}

if (backup->format) {
- options = qdict_new();
+ if (!options) {
+ options = qdict_new();
+ }
qdict_put_str(options, "driver", backup->format);
}

@@ -3257,6 +3262,14 @@ static BlockJob *do_drive_backup(DriveBackup *backup, BlockJobTxn *txn,

bdrv_set_aio_context(target_bs, aio_context);

+ if (set_backing_hd) {
+ bdrv_set_backing_hd(target_bs, source, &local_err);
+ if (local_err) {
+ bdrv_unref(target_bs);
+ goto out;
+ }
+ }
+
if (backup->has_bitmap) {
bmap = bdrv_find_dirty_bitmap(bs, backup->bitmap);
if (!bmap) {
--
2.7.4
Michael Roth
2017-08-29 00:13:48 UTC
Permalink
From: Greg Kurz <***@kaod.org>

When trying to remove a file from a directory, both created in non-mapped
mode, the file remains and EBADF is returned to the guest.

This is a regression introduced by commit "df4938a6651b 9pfs: local:
unlinkat: don't follow symlinks" when fixing CVE-2016-9602. It changed the
way we unlink the metadata file from

ret = remove("$dir/.virtfs_metadata/$name");
if (ret < 0 && errno != ENOENT) {
/* Error out */
}
/* Ignore absence of metadata */

to

fd = openat("$dir/.virtfs_metadata")
unlinkat(fd, "$name")
if (ret < 0 && errno != ENOENT) {
/* Error out */
}
/* Ignore absence of metadata */

If $dir was created in non-mapped mode, openat() fails with ENOENT and
we pass -1 to unlinkat(), which fails in turn with EBADF.

We just need to check the return of openat() and ignore ENOENT, in order
to restore the behaviour we had with remove().

Signed-off-by: Greg Kurz <***@kaod.org>
Reviewed-by: Eric Blake <***@redhat.com>
[groug: rewrote the comments as suggested by Eric]

(cherry picked from commit 6a87e7929f97b86c5823d4616fa1aa7636b2f116)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/9pfs/9p-local.c | 34 +++++++++++++++-------------------
1 file changed, 15 insertions(+), 19 deletions(-)

diff --git a/hw/9pfs/9p-local.c b/hw/9pfs/9p-local.c
index f3ebca4..7a0c383 100644
--- a/hw/9pfs/9p-local.c
+++ b/hw/9pfs/9p-local.c
@@ -957,6 +957,14 @@ static int local_unlinkat_common(FsContext *ctx, int dirfd, const char *name,
if (ctx->export_flags & V9FS_SM_MAPPED_FILE) {
int map_dirfd;

+ /* We need to remove the metadata as well:
+ * - the metadata directory if we're removing a directory
+ * - the metadata file in the parent's metadata directory
+ *
+ * If any of these are missing (ie, ENOENT) then we're probably
+ * trying to remove something that wasn't created in mapped-file
+ * mode. We just ignore the error.
+ */
if (flags == AT_REMOVEDIR) {
int fd;

@@ -964,32 +972,20 @@ static int local_unlinkat_common(FsContext *ctx, int dirfd, const char *name,
if (fd == -1) {
goto err_out;
}
- /*
- * If directory remove .virtfs_metadata contained in the
- * directory
- */
ret = unlinkat(fd, VIRTFS_META_DIR, AT_REMOVEDIR);
close_preserve_errno(fd);
if (ret < 0 && errno != ENOENT) {
- /*
- * We didn't had the .virtfs_metadata file. May be file created
- * in non-mapped mode ?. Ignore ENOENT.
- */
goto err_out;
}
}
- /*
- * Now remove the name from parent directory
- * .virtfs_metadata directory.
- */
map_dirfd = openat_dir(dirfd, VIRTFS_META_DIR);
- ret = unlinkat(map_dirfd, name, 0);
- close_preserve_errno(map_dirfd);
- if (ret < 0 && errno != ENOENT) {
- /*
- * We didn't had the .virtfs_metadata file. May be file created
- * in non-mapped mode ?. Ignore ENOENT.
- */
+ if (map_dirfd != -1) {
+ ret = unlinkat(map_dirfd, name, 0);
+ close_preserve_errno(map_dirfd);
+ if (ret < 0 && errno != ENOENT) {
+ goto err_out;
+ }
+ } else if (errno != ENOENT) {
goto err_out;
}
}
--
2.7.4
Michael Roth
2017-08-29 00:14:08 UTC
Permalink
From: Greg Kurz <***@kaod.org>

According to section 2.1.2 of the virtio-1 specification:

"The device SHOULD set DEVICE_NEEDS_RESET when it enters an error state that
a reset is needed. If DRIVER_OK is set, after it sets DEVICE_NEEDS_RESET,
the device MUST send a device configuration change notification to the
driver."

Commit "f5ed36635d8f virtio: stop virtqueue processing if device is broken"
introduced a virtio_error() call that just does that:

- internally mark the device as broken
- set the DEVICE_NEEDS_RESET bit in the status
- send a configuration change notification

Unfortunately, virtio_notify_vector(), called by virtio_notify_config(),
returns right away when the device is marked as broken and the notification
isn't sent in this case.

The spec doesn't say whether a broken device can send notifications
in other situations or not. But since the driver isn't supposed to do
anything but to reset the device, it makes sense to keep the check in
virtio_notify_config().

Marking the device as broken AFTER the configuration change notification was
sent is enough to fix the issue.

Signed-off-by: Greg Kurz <***@kaod.org>
Reviewed-by: Cornelia Huck <***@de.ibm.com>
Reviewed-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Michael S. Tsirkin <***@redhat.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
(cherry picked from commit 66453cff9e5e75344c601cd7674c8ef5fefee8a6)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/virtio/virtio.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 03592c5..890b4d7 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -2451,12 +2451,12 @@ void GCC_FMT_ATTR(2, 3) virtio_error(VirtIODevice *vdev, const char *fmt, ...)
error_vreport(fmt, ap);
va_end(ap);

- vdev->broken = true;
-
if (virtio_vdev_has_feature(vdev, VIRTIO_F_VERSION_1)) {
virtio_set_status(vdev, vdev->status | VIRTIO_CONFIG_S_NEEDS_RESET);
virtio_notify_config(vdev);
}
+
+ vdev->broken = true;
}

static void virtio_memory_listener_commit(MemoryListener *listener)
--
2.7.4
Michael Roth
2017-08-29 00:13:50 UTC
Permalink
From: Eric Blake <***@redhat.com>

We have macros in place to make it less verbose to add a subtype
of QObject to both QDict and QList. While we have made cleanups
like this in the past (see commit fcfcd8ffc, for example), having
it be automated by Coccinelle makes it easier to maintain.

Patch created mechanically via:
spatch --sp-file scripts/coccinelle/qobject.cocci \
--macro-file scripts/cocci-macro-file.h --dir . --in-place
then I verified that no manual touchups were required.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Reviewed-by: Alberto Garcia <***@igalia.com>
Reviewed-by: Markus Armbruster <***@redhat.com>
Message-Id: <20170427215821.19397-5-***@redhat.com>
Signed-off-by: Markus Armbruster <***@redhat.com>
(cherry picked from commit de6e7951fe66053dfeaac1a237f7aceb9e079619)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block.c | 12 ++++--------
block/blkdebug.c | 4 ++--
block/blkverify.c | 7 +++----
block/file-posix.c | 6 +++---
block/file-win32.c | 4 ++--
block/quorum.c | 16 ++++++----------
tests/check-qdict.c | 20 ++++++++++----------
tests/test-qmp-commands.c | 24 ++++++++++++------------
8 files changed, 42 insertions(+), 51 deletions(-)

diff --git a/block.c b/block.c
index 46da908..f156348 100644
--- a/block.c
+++ b/block.c
@@ -4674,11 +4674,9 @@ void bdrv_refresh_filename(BlockDriverState *bs)
* contain a representation of the filename, therefore the following
* suffices without querying the (exact_)filename of this BDS. */
if (bs->file->bs->full_open_options) {
- qdict_put_obj(opts, "driver",
- QOBJECT(qstring_from_str(drv->format_name)));
+ qdict_put(opts, "driver", qstring_from_str(drv->format_name));
QINCREF(bs->file->bs->full_open_options);
- qdict_put_obj(opts, "file",
- QOBJECT(bs->file->bs->full_open_options));
+ qdict_put(opts, "file", bs->file->bs->full_open_options);

bs->full_open_options = opts;
} else {
@@ -4694,8 +4692,7 @@ void bdrv_refresh_filename(BlockDriverState *bs)

opts = qdict_new();
append_open_options(opts, bs);
- qdict_put_obj(opts, "driver",
- QOBJECT(qstring_from_str(drv->format_name)));
+ qdict_put(opts, "driver", qstring_from_str(drv->format_name));

if (bs->exact_filename[0]) {
/* This may not work for all block protocol drivers (some may
@@ -4705,8 +4702,7 @@ void bdrv_refresh_filename(BlockDriverState *bs)
* needs some special format of the options QDict, it needs to
* implement the driver-specific bdrv_refresh_filename() function.
*/
- qdict_put_obj(opts, "filename",
- QOBJECT(qstring_from_str(bs->exact_filename)));
+ qdict_put(opts, "filename", qstring_from_str(bs->exact_filename));
}

bs->full_open_options = opts;
diff --git a/block/blkdebug.c b/block/blkdebug.c
index cc4a146..42b8d80 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -693,10 +693,10 @@ static void blkdebug_refresh_filename(BlockDriverState *bs, QDict *options)
}

opts = qdict_new();
- qdict_put_obj(opts, "driver", QOBJECT(qstring_from_str("blkdebug")));
+ qdict_put(opts, "driver", qstring_from_str("blkdebug"));

QINCREF(bs->file->bs->full_open_options);
- qdict_put_obj(opts, "image", QOBJECT(bs->file->bs->full_open_options));
+ qdict_put(opts, "image", bs->file->bs->full_open_options);

for (e = qdict_first(options); e; e = qdict_next(options, e)) {
if (strcmp(qdict_entry_key(e), "x-image")) {
diff --git a/block/blkverify.c b/block/blkverify.c
index af23281..cc29cd2 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -288,13 +288,12 @@ static void blkverify_refresh_filename(BlockDriverState *bs, QDict *options)
&& s->test_file->bs->full_open_options)
{
QDict *opts = qdict_new();
- qdict_put_obj(opts, "driver", QOBJECT(qstring_from_str("blkverify")));
+ qdict_put(opts, "driver", qstring_from_str("blkverify"));

QINCREF(bs->file->bs->full_open_options);
- qdict_put_obj(opts, "raw", QOBJECT(bs->file->bs->full_open_options));
+ qdict_put(opts, "raw", bs->file->bs->full_open_options);
QINCREF(s->test_file->bs->full_open_options);
- qdict_put_obj(opts, "test",
- QOBJECT(s->test_file->bs->full_open_options));
+ qdict_put(opts, "test", s->test_file->bs->full_open_options);

bs->full_open_options = opts;
}
diff --git a/block/file-posix.c b/block/file-posix.c
index 0c48968..5370ba0 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -377,7 +377,7 @@ static void raw_parse_filename(const char *filename, QDict *options,
* function call can be ignored. */
strstart(filename, "file:", &filename);

- qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
+ qdict_put(options, "filename", qstring_from_str(filename));
}

static QemuOptsList raw_runtime_opts = {
@@ -2150,7 +2150,7 @@ static void hdev_parse_filename(const char *filename, QDict *options,
/* The prefix is optional, just as for "file". */
strstart(filename, "host_device:", &filename);

- qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
+ qdict_put(options, "filename", qstring_from_str(filename));
}

static bool hdev_is_sg(BlockDriverState *bs)
@@ -2449,7 +2449,7 @@ static void cdrom_parse_filename(const char *filename, QDict *options,
/* The prefix is optional, just as for "file". */
strstart(filename, "host_cdrom:", &filename);

- qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
+ qdict_put(options, "filename", qstring_from_str(filename));
}
#endif

diff --git a/block/file-win32.c b/block/file-win32.c
index 800fabd..57c4a78 100644
--- a/block/file-win32.c
+++ b/block/file-win32.c
@@ -282,7 +282,7 @@ static void raw_parse_filename(const char *filename, QDict *options,
* function call can be ignored. */
strstart(filename, "file:", &filename);

- qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
+ qdict_put(options, "filename", qstring_from_str(filename));
}

static QemuOptsList raw_runtime_opts = {
@@ -669,7 +669,7 @@ static void hdev_parse_filename(const char *filename, QDict *options,
/* The prefix is optional, just as for "file". */
strstart(filename, "host_device:", &filename);

- qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
+ qdict_put(options, "filename", qstring_from_str(filename));
}

static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
diff --git a/block/quorum.c b/block/quorum.c
index 40205fb..f7949e2 100644
--- a/block/quorum.c
+++ b/block/quorum.c
@@ -1096,19 +1096,15 @@ static void quorum_refresh_filename(BlockDriverState *bs, QDict *options)
children = qlist_new();
for (i = 0; i < s->num_children; i++) {
QINCREF(s->children[i]->bs->full_open_options);
- qlist_append_obj(children,
- QOBJECT(s->children[i]->bs->full_open_options));
+ qlist_append(children, s->children[i]->bs->full_open_options);
}

opts = qdict_new();
- qdict_put_obj(opts, "driver", QOBJECT(qstring_from_str("quorum")));
- qdict_put_obj(opts, QUORUM_OPT_VOTE_THRESHOLD,
- QOBJECT(qint_from_int(s->threshold)));
- qdict_put_obj(opts, QUORUM_OPT_BLKVERIFY,
- QOBJECT(qbool_from_bool(s->is_blkverify)));
- qdict_put_obj(opts, QUORUM_OPT_REWRITE,
- QOBJECT(qbool_from_bool(s->rewrite_corrupted)));
- qdict_put_obj(opts, "children", QOBJECT(children));
+ qdict_put(opts, "driver", qstring_from_str("quorum"));
+ qdict_put(opts, QUORUM_OPT_VOTE_THRESHOLD, qint_from_int(s->threshold));
+ qdict_put(opts, QUORUM_OPT_BLKVERIFY, qbool_from_bool(s->is_blkverify));
+ qdict_put(opts, QUORUM_OPT_REWRITE, qbool_from_bool(s->rewrite_corrupted));
+ qdict_put(opts, "children", children);

bs->full_open_options = opts;
}
diff --git a/tests/check-qdict.c b/tests/check-qdict.c
index 81162ee..9cdc985 100644
--- a/tests/check-qdict.c
+++ b/tests/check-qdict.c
@@ -47,7 +47,7 @@ static void qdict_put_obj_test(void)
qdict = qdict_new();

// key "" will have tdb hash 12345
- qdict_put_obj(qdict, "", QOBJECT(qint_from_int(num)));
+ qdict_put(qdict, "", qint_from_int(num));

g_assert(qdict_size(qdict) == 1);
ent = QLIST_FIRST(&qdict->table[12345 % QDICT_BUCKET_MAX]);
@@ -66,8 +66,8 @@ static void qdict_destroy_simple_test(void)
QDict *qdict;

qdict = qdict_new();
- qdict_put_obj(qdict, "num", QOBJECT(qint_from_int(0)));
- qdict_put_obj(qdict, "str", QOBJECT(qstring_from_str("foo")));
+ qdict_put(qdict, "num", qint_from_int(0));
+ qdict_put(qdict, "str", qstring_from_str("foo"));

QDECREF(qdict);
}
@@ -297,16 +297,16 @@ static void qdict_flatten_test(void)
qdict_put(dict1, "a", qint_from_int(0));
qdict_put(dict1, "b", qint_from_int(1));

- qlist_append_obj(list1, QOBJECT(qint_from_int(23)));
- qlist_append_obj(list1, QOBJECT(qint_from_int(66)));
- qlist_append_obj(list1, QOBJECT(dict1));
- qlist_append_obj(list2, QOBJECT(qint_from_int(42)));
- qlist_append_obj(list2, QOBJECT(list1));
+ qlist_append(list1, qint_from_int(23));
+ qlist_append(list1, qint_from_int(66));
+ qlist_append(list1, dict1);
+ qlist_append(list2, qint_from_int(42));
+ qlist_append(list2, list1);

qdict_put(dict2, "c", qint_from_int(2));
qdict_put(dict2, "d", qint_from_int(3));
- qdict_put_obj(dict3, "e", QOBJECT(list2));
- qdict_put_obj(dict3, "f", QOBJECT(dict2));
+ qdict_put(dict3, "e", list2);
+ qdict_put(dict3, "f", dict2);
qdict_put(dict3, "g", qint_from_int(4));

qdict_flatten(dict3);
diff --git a/tests/test-qmp-commands.c b/tests/test-qmp-commands.c
index 0f81a98..ac3fd03 100644
--- a/tests/test-qmp-commands.c
+++ b/tests/test-qmp-commands.c
@@ -94,7 +94,7 @@ static void test_dispatch_cmd(void)
QDict *req = qdict_new();
QObject *resp;

- qdict_put_obj(req, "execute", QOBJECT(qstring_from_str("user_def_cmd")));
+ qdict_put(req, "execute", qstring_from_str("user_def_cmd"));

resp = qmp_dispatch(&qmp_commands, QOBJECT(req));
assert(resp != NULL);
@@ -111,7 +111,7 @@ static void test_dispatch_cmd_failure(void)
QDict *args = qdict_new();
QObject *resp;

- qdict_put_obj(req, "execute", QOBJECT(qstring_from_str("user_def_cmd2")));
+ qdict_put(req, "execute", qstring_from_str("user_def_cmd2"));

resp = qmp_dispatch(&qmp_commands, QOBJECT(req));
assert(resp != NULL);
@@ -125,7 +125,7 @@ static void test_dispatch_cmd_failure(void)
qdict_put(args, "a", qint_from_int(66));
qdict_put(req, "arguments", args);

- qdict_put_obj(req, "execute", QOBJECT(qstring_from_str("user_def_cmd")));
+ qdict_put(req, "execute", qstring_from_str("user_def_cmd"));

resp = qmp_dispatch(&qmp_commands, QOBJECT(req));
assert(resp != NULL);
@@ -164,14 +164,14 @@ static void test_dispatch_cmd_io(void)
QDict *ret_dict_dict2, *ret_dict_dict2_userdef;
QInt *ret3;

- qdict_put_obj(ud1a, "integer", QOBJECT(qint_from_int(42)));
- qdict_put_obj(ud1a, "string", QOBJECT(qstring_from_str("hello")));
- qdict_put_obj(ud1b, "integer", QOBJECT(qint_from_int(422)));
- qdict_put_obj(ud1b, "string", QOBJECT(qstring_from_str("hello2")));
- qdict_put_obj(args, "ud1a", QOBJECT(ud1a));
- qdict_put_obj(args, "ud1b", QOBJECT(ud1b));
- qdict_put_obj(req, "arguments", QOBJECT(args));
- qdict_put_obj(req, "execute", QOBJECT(qstring_from_str("user_def_cmd2")));
+ qdict_put(ud1a, "integer", qint_from_int(42));
+ qdict_put(ud1a, "string", qstring_from_str("hello"));
+ qdict_put(ud1b, "integer", qint_from_int(422));
+ qdict_put(ud1b, "string", qstring_from_str("hello2"));
+ qdict_put(args, "ud1a", ud1a);
+ qdict_put(args, "ud1b", ud1b);
+ qdict_put(req, "arguments", args);
+ qdict_put(req, "execute", qstring_from_str("user_def_cmd2"));

ret = qobject_to_qdict(test_qmp_dispatch(req));

@@ -244,7 +244,7 @@ static void test_dealloc_partial(void)
Visitor *v;

ud2_dict = qdict_new();
- qdict_put_obj(ud2_dict, "string0", QOBJECT(qstring_from_str(text)));
+ qdict_put(ud2_dict, "string0", qstring_from_str(text));

v = qobject_input_visitor_new(QOBJECT(ud2_dict));
visit_type_UserDefTwo(v, NULL, &ud2, &err);
--
2.7.4
Michael Roth
2017-08-29 00:13:55 UTC
Permalink
From: Zhiyong Yang <***@intel.com>

Qemu2.7~2.9 and vhost user for dpdk 17.02 release work together
to cause failures of new connection when negotiating to set MQ.
(one queue pair works well).
Because there exist some bugs in qemu code when introducing
VHOST_USER_PROTOCOL_F_REPLY_ACK to qemu. When vhost_user_set_mem_table
is invoked to deal with the vhost message VHOST_USER_SET_MEM_TABLE
for the second time, qemu indeed doesn't send the messge (The message
needs to be sent only once)but still will be waiting for dpdk's reply
ack, then, qemu is always freezing, while DPDK is always waiting for
next vhost message from qemu.
The patch aims to fix the bug, MQ can work well.
The same bug is found in function vhost_user_net_set_mtu, it is fixed
at the same time.
DPDK related patch is as following:
http://www.dpdk.org/dev/patchwork/patch/23955/

Signed-off-by: Zhiyong Yang <***@intel.com>
Cc: qemu-***@nongnu.org
Fixes: ca525ce5618b ("vhost-user: Introduce a new protocol feature REPLY_ACK.")
Reviewed-by: Maxime Coquelin <***@redhat.com>
Reviewed-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Michael S. Tsirkin <***@redhat.com>
Tested-by: Jens Freimann <***@redhat.com>
Reviewed-by: Marc-André Lureau <***@redhat.com>
(cherry picked from commit 60cd11024f41cc73175e651a2dfe09a3cade56bb)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/virtio/vhost-user.c | 21 +++++++++++++--------
1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index 9334a8a..32a95a8c 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -163,22 +163,26 @@ fail:
}

static int process_message_reply(struct vhost_dev *dev,
- VhostUserRequest request)
+ VhostUserMsg msg)
{
- VhostUserMsg msg;
+ VhostUserMsg msg_reply;

- if (vhost_user_read(dev, &msg) < 0) {
+ if ((msg.flags & VHOST_USER_NEED_REPLY_MASK) == 0) {
+ return 0;
+ }
+
+ if (vhost_user_read(dev, &msg_reply) < 0) {
return -1;
}

- if (msg.request != request) {
+ if (msg_reply.request != msg.request) {
error_report("Received unexpected msg type."
"Expected %d received %d",
- request, msg.request);
+ msg.request, msg_reply.request);
return -1;
}

- return msg.payload.u64 ? -1 : 0;
+ return msg_reply.payload.u64 ? -1 : 0;
}

static bool vhost_user_one_time_request(VhostUserRequest request)
@@ -208,6 +212,7 @@ static int vhost_user_write(struct vhost_dev *dev, VhostUserMsg *msg,
* request, we just ignore it.
*/
if (vhost_user_one_time_request(msg->request) && dev->vq_index != 0) {
+ msg->flags &= ~VHOST_USER_NEED_REPLY_MASK;
return 0;
}

@@ -320,7 +325,7 @@ static int vhost_user_set_mem_table(struct vhost_dev *dev,
}

if (reply_supported) {
- return process_message_reply(dev, msg.request);
+ return process_message_reply(dev, msg);
}

return 0;
@@ -712,7 +717,7 @@ static int vhost_user_net_set_mtu(struct vhost_dev *dev, uint16_t mtu)

/* If reply_ack supported, slave has to ack specified MTU is valid */
if (reply_supported) {
- return process_message_reply(dev, msg.request);
+ return process_message_reply(dev, msg);
}

return 0;
--
2.7.4
Michael Roth
2017-08-29 00:13:37 UTC
Permalink
From: Eric Blake <***@redhat.com>

We've been documenting the value in bytes since its introduction
in commit b9a9b3a4 (v1.3), where it was actually reported in bytes.

Commit e4654d2 (v2.0) then removed things from block/qapi.c, in
preparation for a rewrite to a list of dirty sectors in the next
commit 21b5683 in block.c, but the new code mistakenly started
reporting in sectors.

Fixes: https://bugzilla.redhat.com/1441460

CC: qemu-***@nongnu.org
Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: John Snow <***@redhat.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 6c98c57af3f4fab85bdf5f01616c91322bd4312a)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/dirty-bitmap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c
index 519737c..6d8ce5f 100644
--- a/block/dirty-bitmap.c
+++ b/block/dirty-bitmap.c
@@ -345,7 +345,7 @@ BlockDirtyInfoList *bdrv_query_dirty_bitmaps(BlockDriverState *bs)
QLIST_FOREACH(bm, &bs->dirty_bitmaps, list) {
BlockDirtyInfo *info = g_new0(BlockDirtyInfo, 1);
BlockDirtyInfoList *entry = g_new0(BlockDirtyInfoList, 1);
- info->count = bdrv_get_dirty_count(bm);
+ info->count = bdrv_get_dirty_count(bm) << BDRV_SECTOR_BITS;
info->granularity = bdrv_dirty_bitmap_granularity(bm);
info->has_name = !!bm->name;
info->name = g_strdup(bm->name);
--
2.7.4
Michael Roth
2017-08-29 00:13:57 UTC
Permalink
From: Anton Nefedov <***@virtuozzo.com>

On error path (like i/o error in one of the coroutines), it's required to
- wait for coroutines completion before cleaning the common structures
- reenter dependent coroutines so they ever finish

Introduced in 2d9187bc65.

Cc: qemu-***@nongnu.org
Signed-off-by: Anton Nefedov <***@virtuozzo.com>
Reviewed-by: Peter Lieven <***@kamp.de>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit b91127edd0ff96f27f1e58e47f4e9f9d6a0fed02)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
qemu-img.c | 26 +++++++++++---------------
1 file changed, 11 insertions(+), 15 deletions(-)

diff --git a/qemu-img.c b/qemu-img.c
index 2e61561..26ded22 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -1761,13 +1761,13 @@ static void coroutine_fn convert_co_do_copy(void *opaque)
qemu_co_mutex_lock(&s->lock);
if (s->ret != -EINPROGRESS || s->sector_num >= s->total_sectors) {
qemu_co_mutex_unlock(&s->lock);
- goto out;
+ break;
}
n = convert_iteration_sectors(s, s->sector_num);
if (n < 0) {
qemu_co_mutex_unlock(&s->lock);
s->ret = n;
- goto out;
+ break;
}
/* save current sector and allocation status to local variables */
sector_num = s->sector_num;
@@ -1792,7 +1792,6 @@ static void coroutine_fn convert_co_do_copy(void *opaque)
error_report("error while reading sector %" PRId64
": %s", sector_num, strerror(-ret));
s->ret = ret;
- goto out;
}
} else if (!s->min_sparse && status == BLK_ZERO) {
status = BLK_DATA;
@@ -1801,22 +1800,20 @@ static void coroutine_fn convert_co_do_copy(void *opaque)

if (s->wr_in_order) {
/* keep writes in order */
- while (s->wr_offs != sector_num) {
- if (s->ret != -EINPROGRESS) {
- goto out;
- }
+ while (s->wr_offs != sector_num && s->ret == -EINPROGRESS) {
s->wait_sector_num[index] = sector_num;
qemu_coroutine_yield();
}
s->wait_sector_num[index] = -1;
}

- ret = convert_co_write(s, sector_num, n, buf, status);
- if (ret < 0) {
- error_report("error while writing sector %" PRId64
- ": %s", sector_num, strerror(-ret));
- s->ret = ret;
- goto out;
+ if (s->ret == -EINPROGRESS) {
+ ret = convert_co_write(s, sector_num, n, buf, status);
+ if (ret < 0) {
+ error_report("error while writing sector %" PRId64
+ ": %s", sector_num, strerror(-ret));
+ s->ret = ret;
+ }
}

if (s->wr_in_order) {
@@ -1837,7 +1834,6 @@ static void coroutine_fn convert_co_do_copy(void *opaque)
}
}

-out:
qemu_vfree(buf);
s->co[index] = NULL;
s->running_coroutines--;
@@ -1899,7 +1895,7 @@ static int convert_do_copy(ImgConvertState *s)
qemu_coroutine_enter(s->co[i]);
}

- while (s->ret == -EINPROGRESS) {
+ while (s->running_coroutines) {
main_loop_wait(false);
}
--
2.7.4
Michael Roth
2017-08-29 00:13:36 UTC
Permalink
From: Sameeh Jubran <***@daynix.com>

The QGA schema states:

@can-offline: Whether offlining the VCPU is possible. This member
is always filled in by the guest agent when the structure
is returned, and always ignored on input (hence it can be
omitted then).

Currently 'can-offline' is missing entirely from the reply. This causes
errors in libvirt which is expecting the reply to be compliant with the
schema docs.

BZ#1438735: https://bugzilla.redhat.com/show_bug.cgi?id=1438735

Signed-off-by: Sameeh Jubran <***@daynix.com>
Reviewed-by: Eric Blake <***@redhat.com>
Cc: qemu-***@nongnu.org
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
(cherry picked from commit 54858553def1879a3b0781529fb12a028ba36713)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
qga/commands-win32.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/qga/commands-win32.c b/qga/commands-win32.c
index 19d72b2..f0d72a0 100644
--- a/qga/commands-win32.c
+++ b/qga/commands-win32.c
@@ -1344,7 +1344,7 @@ GuestLogicalProcessorList *qmp_guest_get_vcpus(Error **errp)
vcpu = g_malloc0(sizeof *vcpu);
vcpu->logical_id = current++;
vcpu->online = true;
- vcpu->has_can_offline = false;
+ vcpu->has_can_offline = true;

entry = g_malloc0(sizeof *entry);
entry->value = vcpu;
--
2.7.4
Michael Roth
2017-08-29 00:14:14 UTC
Permalink
From: Eric Blake <***@redhat.com>

Rather than repeat the logic at each caller of checking if a Rule
exists that warrants an error injection, fold that logic into
inject_error(); and rename it to rule_check() for legibility.
This will help the next patch, which adds two more callers that
need to check rules for the potential of injecting errors.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170429191419.30051-6-***@redhat.com
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit d157ed5f7235f3d2d5596a514ad7507b18e24b88)
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 74 +++++++++++++++++++++++++-------------------------------
1 file changed, 33 insertions(+), 41 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index b2cee0c..c5d2edb 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -403,11 +403,30 @@ out:
return ret;
}

-static int inject_error(BlockDriverState *bs, BlkdebugRule *rule)
+static int rule_check(BlockDriverState *bs, uint64_t offset, uint64_t bytes)
{
BDRVBlkdebugState *s = bs->opaque;
- int error = rule->options.inject.error;
- bool immediately = rule->options.inject.immediately;
+ BlkdebugRule *rule = NULL;
+ int error;
+ bool immediately;
+
+ QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
+ uint64_t inject_offset = rule->options.inject.offset;
+
+ if (inject_offset == -1 ||
+ (bytes && inject_offset >= offset &&
+ inject_offset < offset + bytes))
+ {
+ break;
+ }
+ }
+
+ if (!rule || !rule->options.inject.error) {
+ return 0;
+ }
+
+ immediately = rule->options.inject.immediately;
+ error = rule->options.inject.error;

if (rule->options.inject.once) {
QSIMPLEQ_REMOVE(&s->active_rules, rule, BlkdebugRule, active_next);
@@ -426,8 +445,7 @@ static int coroutine_fn
blkdebug_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
QEMUIOVector *qiov, int flags)
{
- BDRVBlkdebugState *s = bs->opaque;
- BlkdebugRule *rule = NULL;
+ int err;

/* Sanity check block layer guarantees */
assert(QEMU_IS_ALIGNED(offset, bs->bl.request_alignment));
@@ -436,18 +454,9 @@ blkdebug_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
assert(bytes <= bs->bl.max_transfer);
}

- QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
- uint64_t inject_offset = rule->options.inject.offset;
-
- if (inject_offset == -1 ||
- (inject_offset >= offset && inject_offset < offset + bytes))
- {
- break;
- }
- }
-
- if (rule && rule->options.inject.error) {
- return inject_error(bs, rule);
+ err = rule_check(bs, offset, bytes);
+ if (err) {
+ return err;
}

return bdrv_co_preadv(bs->file, offset, bytes, qiov, flags);
@@ -457,8 +466,7 @@ static int coroutine_fn
blkdebug_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
QEMUIOVector *qiov, int flags)
{
- BDRVBlkdebugState *s = bs->opaque;
- BlkdebugRule *rule = NULL;
+ int err;

/* Sanity check block layer guarantees */
assert(QEMU_IS_ALIGNED(offset, bs->bl.request_alignment));
@@ -467,18 +475,9 @@ blkdebug_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
assert(bytes <= bs->bl.max_transfer);
}

- QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
- uint64_t inject_offset = rule->options.inject.offset;
-
- if (inject_offset == -1 ||
- (inject_offset >= offset && inject_offset < offset + bytes))
- {
- break;
- }
- }
-
- if (rule && rule->options.inject.error) {
- return inject_error(bs, rule);
+ err = rule_check(bs, offset, bytes);
+ if (err) {
+ return err;
}

return bdrv_co_pwritev(bs->file, offset, bytes, qiov, flags);
@@ -486,17 +485,10 @@ blkdebug_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,

static int blkdebug_co_flush(BlockDriverState *bs)
{
- BDRVBlkdebugState *s = bs->opaque;
- BlkdebugRule *rule = NULL;
-
- QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
- if (rule->options.inject.offset == -1) {
- break;
- }
- }
+ int err = rule_check(bs, 0, 0);

- if (rule && rule->options.inject.error) {
- return inject_error(bs, rule);
+ if (err) {
+ return err;
}

return bdrv_co_flush(bs->file->bs);
--
2.7.4
Michael Roth
2017-08-29 00:13:53 UTC
Permalink
From: Eric Blake <***@redhat.com>

We now have macros in place to make it less verbose to add a scalar
to QDict and QList, so use them.

Patch created mechanically via:
spatch --sp-file scripts/coccinelle/qobject.cocci \
--macro-file scripts/cocci-macro-file.h --dir . --in-place
then touched up manually to fix a couple of '?:' back to original
spacing, as well as avoiding a long line in monitor.c.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Markus Armbruster <***@redhat.com>
Message-Id: <20170427215821.19397-7-***@redhat.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Reviewed-by: Alberto Garcia <***@igalia.com>
Signed-off-by: Markus Armbruster <***@redhat.com>
(cherry picked from commit 46f5ac205a9dc5e2c24274c7df371509a286281f)
* prereq for fc0932f
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block.c | 45 +++++-------
block/blkdebug.c | 6 +-
block/blkverify.c | 6 +-
block/curl.c | 2 +-
block/file-posix.c | 8 +--
block/file-win32.c | 4 +-
block/nbd.c | 41 ++++++-----
block/nfs.c | 43 +++++-------
block/null.c | 2 +-
block/qcow2.c | 4 +-
block/quorum.c | 8 +--
block/rbd.c | 16 ++---
block/snapshot.c | 2 +-
block/ssh.c | 16 ++---
block/vvfat.c | 10 +--
blockdev.c | 30 ++++----
hw/block/xen_disk.c | 2 +-
hw/usb/xen-usb.c | 12 ++--
monitor.c | 23 +++----
qapi/qmp-event.c | 2 +-
qemu-img.c | 6 +-
qemu-io.c | 2 +-
qemu-nbd.c | 2 +-
qobject/qdict.c | 2 +-
target/s390x/cpu_models.c | 4 +-
tests/check-qdict.c | 134 ++++++++++++++++++------------------
tests/check-qlist.c | 4 +-
tests/device-introspect-test.c | 4 +-
tests/test-qemu-opts.c | 4 +-
tests/test-qmp-commands.c | 24 +++----
tests/test-qmp-event.c | 30 ++++----
tests/test-qobject-output-visitor.c | 6 +-
util/qemu-option.c | 2 +-
33 files changed, 241 insertions(+), 265 deletions(-)

diff --git a/block.c b/block.c
index f156348..513e6e5 100644
--- a/block.c
+++ b/block.c
@@ -937,16 +937,14 @@ static void update_flags_from_options(int *flags, QemuOpts *opts)
static void update_options_from_flags(QDict *options, int flags)
{
if (!qdict_haskey(options, BDRV_OPT_CACHE_DIRECT)) {
- qdict_put(options, BDRV_OPT_CACHE_DIRECT,
- qbool_from_bool(flags & BDRV_O_NOCACHE));
+ qdict_put_bool(options, BDRV_OPT_CACHE_DIRECT, flags & BDRV_O_NOCACHE);
}
if (!qdict_haskey(options, BDRV_OPT_CACHE_NO_FLUSH)) {
- qdict_put(options, BDRV_OPT_CACHE_NO_FLUSH,
- qbool_from_bool(flags & BDRV_O_NO_FLUSH));
+ qdict_put_bool(options, BDRV_OPT_CACHE_NO_FLUSH,
+ flags & BDRV_O_NO_FLUSH);
}
if (!qdict_haskey(options, BDRV_OPT_READ_ONLY)) {
- qdict_put(options, BDRV_OPT_READ_ONLY,
- qbool_from_bool(!(flags & BDRV_O_RDWR)));
+ qdict_put_bool(options, BDRV_OPT_READ_ONLY, !(flags & BDRV_O_RDWR));
}
}

@@ -1362,7 +1360,7 @@ static int bdrv_fill_options(QDict **options, const char *filename,
/* Fetch the file name from the options QDict if necessary */
if (protocol && filename) {
if (!qdict_haskey(*options, "filename")) {
- qdict_put(*options, "filename", qstring_from_str(filename));
+ qdict_put_str(*options, "filename", filename);
parse_filename = true;
} else {
error_setg(errp, "Can't specify 'file' and 'filename' options at "
@@ -1383,7 +1381,7 @@ static int bdrv_fill_options(QDict **options, const char *filename,
}

drvname = drv->format_name;
- qdict_put(*options, "driver", qstring_from_str(drvname));
+ qdict_put_str(*options, "driver", drvname);
} else {
error_setg(errp, "Must specify either driver or file");
return -EINVAL;
@@ -2038,7 +2036,7 @@ int bdrv_open_backing_file(BlockDriverState *bs, QDict *parent_options,
}

if (bs->backing_format[0] != '\0' && !qdict_haskey(options, "driver")) {
- qdict_put(options, "driver", qstring_from_str(bs->backing_format));
+ qdict_put_str(options, "driver", bs->backing_format);
}

backing_hd = bdrv_open_inherit(*backing_filename ? backing_filename : NULL,
@@ -2193,12 +2191,9 @@ static BlockDriverState *bdrv_append_temp_snapshot(BlockDriverState *bs,
}

/* Prepare options QDict for the temporary file */
- qdict_put(snapshot_options, "file.driver",
- qstring_from_str("file"));
- qdict_put(snapshot_options, "file.filename",
- qstring_from_str(tmp_filename));
- qdict_put(snapshot_options, "driver",
- qstring_from_str("qcow2"));
+ qdict_put_str(snapshot_options, "file.driver", "file");
+ qdict_put_str(snapshot_options, "file.filename", tmp_filename);
+ qdict_put_str(snapshot_options, "driver", "qcow2");

bs_snapshot = bdrv_open(NULL, NULL, snapshot_options, flags, errp);
snapshot_options = NULL;
@@ -2373,8 +2368,7 @@ static BlockDriverState *bdrv_open_inherit(const char *filename,
goto fail;
}

- qdict_put(options, "file",
- qstring_from_str(bdrv_get_node_name(file_bs)));
+ qdict_put_str(options, "file", bdrv_get_node_name(file_bs));
}
}

@@ -2396,8 +2390,8 @@ static BlockDriverState *bdrv_open_inherit(const char *filename,
* sure to update both bs->options (which has the full effective
* options for bs) and options (which has file.* already removed).
*/
- qdict_put(bs->options, "driver", qstring_from_str(drv->format_name));
- qdict_put(options, "driver", qstring_from_str(drv->format_name));
+ qdict_put_str(bs->options, "driver", drv->format_name);
+ qdict_put_str(options, "driver", drv->format_name);
} else if (!drv) {
error_setg(errp, "Must specify either driver or file");
goto fail;
@@ -2772,12 +2766,12 @@ int bdrv_reopen_prepare(BDRVReopenState *reopen_state, BlockReopenQueue *queue,
* that they are checked at the end of this function. */
value = qemu_opt_get(opts, "node-name");
if (value) {
- qdict_put(reopen_state->options, "node-name", qstring_from_str(value));
+ qdict_put_str(reopen_state->options, "node-name", value);
}

value = qemu_opt_get(opts, "driver");
if (value) {
- qdict_put(reopen_state->options, "driver", qstring_from_str(value));
+ qdict_put_str(reopen_state->options, "driver", value);
}

/* if we are to stay read-only, do not allow permission change
@@ -4268,8 +4262,7 @@ void bdrv_img_create(const char *filename, const char *fmt,

if (backing_fmt) {
backing_options = qdict_new();
- qdict_put(backing_options, "driver",
- qstring_from_str(backing_fmt));
+ qdict_put_str(backing_options, "driver", backing_fmt);
}

bs = bdrv_open(full_backing, NULL, backing_options, back_flags,
@@ -4674,7 +4667,7 @@ void bdrv_refresh_filename(BlockDriverState *bs)
* contain a representation of the filename, therefore the following
* suffices without querying the (exact_)filename of this BDS. */
if (bs->file->bs->full_open_options) {
- qdict_put(opts, "driver", qstring_from_str(drv->format_name));
+ qdict_put_str(opts, "driver", drv->format_name);
QINCREF(bs->file->bs->full_open_options);
qdict_put(opts, "file", bs->file->bs->full_open_options);

@@ -4692,7 +4685,7 @@ void bdrv_refresh_filename(BlockDriverState *bs)

opts = qdict_new();
append_open_options(opts, bs);
- qdict_put(opts, "driver", qstring_from_str(drv->format_name));
+ qdict_put_str(opts, "driver", drv->format_name);

if (bs->exact_filename[0]) {
/* This may not work for all block protocol drivers (some may
@@ -4702,7 +4695,7 @@ void bdrv_refresh_filename(BlockDriverState *bs)
* needs some special format of the options QDict, it needs to
* implement the driver-specific bdrv_refresh_filename() function.
*/
- qdict_put(opts, "filename", qstring_from_str(bs->exact_filename));
+ qdict_put_str(opts, "filename", bs->exact_filename);
}

bs->full_open_options = opts;
diff --git a/block/blkdebug.c b/block/blkdebug.c
index 42b8d80..ffc6f1d 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -301,7 +301,7 @@ static void blkdebug_parse_filename(const char *filename, QDict *options,
if (!strstart(filename, "blkdebug:", &filename)) {
/* There was no prefix; therefore, all options have to be already
present in the QDict (except for the filename) */
- qdict_put(options, "x-image", qstring_from_str(filename));
+ qdict_put_str(options, "x-image", filename);
return;
}

@@ -320,7 +320,7 @@ static void blkdebug_parse_filename(const char *filename, QDict *options,

/* TODO Allow multi-level nesting and set file.filename here */
filename = c + 1;
- qdict_put(options, "x-image", qstring_from_str(filename));
+ qdict_put_str(options, "x-image", filename);
}

static QemuOptsList runtime_opts = {
@@ -693,7 +693,7 @@ static void blkdebug_refresh_filename(BlockDriverState *bs, QDict *options)
}

opts = qdict_new();
- qdict_put(opts, "driver", qstring_from_str("blkdebug"));
+ qdict_put_str(opts, "driver", "blkdebug");

QINCREF(bs->file->bs->full_open_options);
qdict_put(opts, "image", bs->file->bs->full_open_options);
diff --git a/block/blkverify.c b/block/blkverify.c
index cc29cd2..6b0a603 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -67,7 +67,7 @@ static void blkverify_parse_filename(const char *filename, QDict *options,
if (!strstart(filename, "blkverify:", &filename)) {
/* There was no prefix; therefore, all options have to be already
present in the QDict (except for the filename) */
- qdict_put(options, "x-image", qstring_from_str(filename));
+ qdict_put_str(options, "x-image", filename);
return;
}

@@ -84,7 +84,7 @@ static void blkverify_parse_filename(const char *filename, QDict *options,

/* TODO Allow multi-level nesting and set file.filename here */
filename = c + 1;
- qdict_put(options, "x-image", qstring_from_str(filename));
+ qdict_put_str(options, "x-image", filename);
}

static QemuOptsList runtime_opts = {
@@ -288,7 +288,7 @@ static void blkverify_refresh_filename(BlockDriverState *bs, QDict *options)
&& s->test_file->bs->full_open_options)
{
QDict *opts = qdict_new();
- qdict_put(opts, "driver", qstring_from_str("blkverify"));
+ qdict_put_str(opts, "driver", "blkverify");

QINCREF(bs->file->bs->full_open_options);
qdict_put(opts, "raw", bs->file->bs->full_open_options);
diff --git a/block/curl.c b/block/curl.c
index 2708d57..aa6e8cc 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -548,7 +548,7 @@ static void curl_clean_state(CURLState *s)
static void curl_parse_filename(const char *filename, QDict *options,
Error **errp)
{
- qdict_put(options, CURL_BLOCK_OPT_URL, qstring_from_str(filename));
+ qdict_put_str(options, CURL_BLOCK_OPT_URL, filename);
}

static void curl_detach_aio_context(BlockDriverState *bs)
diff --git a/block/file-posix.c b/block/file-posix.c
index 5370ba0..9431ad1 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -377,7 +377,7 @@ static void raw_parse_filename(const char *filename, QDict *options,
* function call can be ignored. */
strstart(filename, "file:", &filename);

- qdict_put(options, "filename", qstring_from_str(filename));
+ qdict_put_str(options, "filename", filename);
}

static QemuOptsList raw_runtime_opts = {
@@ -2150,7 +2150,7 @@ static void hdev_parse_filename(const char *filename, QDict *options,
/* The prefix is optional, just as for "file". */
strstart(filename, "host_device:", &filename);

- qdict_put(options, "filename", qstring_from_str(filename));
+ qdict_put_str(options, "filename", filename);
}

static bool hdev_is_sg(BlockDriverState *bs)
@@ -2239,7 +2239,7 @@ static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
goto hdev_open_Mac_error;
}

- qdict_put(options, "filename", qstring_from_str(bsd_path));
+ qdict_put_str(options, "filename", bsd_path);

hdev_open_Mac_error:
g_free(mediaType);
@@ -2449,7 +2449,7 @@ static void cdrom_parse_filename(const char *filename, QDict *options,
/* The prefix is optional, just as for "file". */
strstart(filename, "host_cdrom:", &filename);

- qdict_put(options, "filename", qstring_from_str(filename));
+ qdict_put_str(options, "filename", filename);
}
#endif

diff --git a/block/file-win32.c b/block/file-win32.c
index 57c4a78..0d455d1 100644
--- a/block/file-win32.c
+++ b/block/file-win32.c
@@ -282,7 +282,7 @@ static void raw_parse_filename(const char *filename, QDict *options,
* function call can be ignored. */
strstart(filename, "file:", &filename);

- qdict_put(options, "filename", qstring_from_str(filename));
+ qdict_put_str(options, "filename", filename);
}

static QemuOptsList raw_runtime_opts = {
@@ -669,7 +669,7 @@ static void hdev_parse_filename(const char *filename, QDict *options,
/* The prefix is optional, just as for "file". */
strstart(filename, "host_device:", &filename);

- qdict_put(options, "filename", qstring_from_str(filename));
+ qdict_put_str(options, "filename", filename);
}

static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
diff --git a/block/nbd.c b/block/nbd.c
index 814ab26d..b3545f5 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -79,7 +79,7 @@ static int nbd_parse_uri(const char *filename, QDict *options)
p = uri->path ? uri->path : "/";
p += strspn(p, "/");
if (p[0]) {
- qdict_put(options, "export", qstring_from_str(p));
+ qdict_put_str(options, "export", p);
}

qp = query_params_parse(uri->query);
@@ -94,9 +94,8 @@ static int nbd_parse_uri(const char *filename, QDict *options)
ret = -EINVAL;
goto out;
}
- qdict_put(options, "server.type", qstring_from_str("unix"));
- qdict_put(options, "server.path",
- qstring_from_str(qp->p[0].value));
+ qdict_put_str(options, "server.type", "unix");
+ qdict_put_str(options, "server.path", qp->p[0].value);
} else {
QString *host;
char *port_str;
@@ -115,11 +114,11 @@ static int nbd_parse_uri(const char *filename, QDict *options)
host = qstring_from_str(uri->server);
}

- qdict_put(options, "server.type", qstring_from_str("inet"));
+ qdict_put_str(options, "server.type", "inet");
qdict_put(options, "server.host", host);

port_str = g_strdup_printf("%d", uri->port ?: NBD_DEFAULT_PORT);
- qdict_put(options, "server.port", qstring_from_str(port_str));
+ qdict_put_str(options, "server.port", port_str);
g_free(port_str);
}

@@ -181,7 +180,7 @@ static void nbd_parse_filename(const char *filename, QDict *options,
export_name[0] = 0; /* truncate 'file' */
export_name += strlen(EN_OPTSTR);

- qdict_put(options, "export", qstring_from_str(export_name));
+ qdict_put_str(options, "export", export_name);
}

/* extract the host_spec - fail if it's not nbd:... */
@@ -196,8 +195,8 @@ static void nbd_parse_filename(const char *filename, QDict *options,

/* are we a UNIX or TCP socket? */
if (strstart(host_spec, "unix:", &unixpath)) {
- qdict_put(options, "server.type", qstring_from_str("unix"));
- qdict_put(options, "server.path", qstring_from_str(unixpath));
+ qdict_put_str(options, "server.type", "unix");
+ qdict_put_str(options, "server.path", unixpath);
} else {
InetSocketAddress *addr = NULL;

@@ -206,9 +205,9 @@ static void nbd_parse_filename(const char *filename, QDict *options,
goto out;
}

- qdict_put(options, "server.type", qstring_from_str("inet"));
- qdict_put(options, "server.host", qstring_from_str(addr->host));
- qdict_put(options, "server.port", qstring_from_str(addr->port));
+ qdict_put_str(options, "server.type", "inet");
+ qdict_put_str(options, "server.host", addr->host);
+ qdict_put_str(options, "server.port", addr->port);
qapi_free_InetSocketAddress(addr);
}

@@ -247,13 +246,13 @@ static bool nbd_process_legacy_socket_options(QDict *output_options,
return false;
}

- qdict_put(output_options, "server.type", qstring_from_str("unix"));
- qdict_put(output_options, "server.path", qstring_from_str(path));
+ qdict_put_str(output_options, "server.type", "unix");
+ qdict_put_str(output_options, "server.path", path);
} else if (host) {
- qdict_put(output_options, "server.type", qstring_from_str("inet"));
- qdict_put(output_options, "server.host", qstring_from_str(host));
- qdict_put(output_options, "server.port",
- qstring_from_str(port ?: stringify(NBD_DEFAULT_PORT)));
+ qdict_put_str(output_options, "server.type", "inet");
+ qdict_put_str(output_options, "server.host", host);
+ qdict_put_str(output_options, "server.port",
+ port ?: stringify(NBD_DEFAULT_PORT));
}

return true;
@@ -528,7 +527,7 @@ static void nbd_refresh_filename(BlockDriverState *bs, QDict *options)
path = s->saddr->u.q_unix.path;
} /* else can't represent as pseudo-filename */

- qdict_put(opts, "driver", qstring_from_str("nbd"));
+ qdict_put_str(opts, "driver", "nbd");

if (path && s->export) {
snprintf(bs->exact_filename, sizeof(bs->exact_filename),
@@ -551,10 +550,10 @@ static void nbd_refresh_filename(BlockDriverState *bs, QDict *options)
qdict_put_obj(opts, "server", saddr_qdict);

if (s->export) {
- qdict_put(opts, "export", qstring_from_str(s->export));
+ qdict_put_str(opts, "export", s->export);
}
if (s->tlscredsid) {
- qdict_put(opts, "tls-creds", qstring_from_str(s->tlscredsid));
+ qdict_put_str(opts, "tls-creds", s->tlscredsid);
}

qdict_flatten(opts);
diff --git a/block/nfs.c b/block/nfs.c
index 0816678..bfeebc1 100644
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -104,9 +104,9 @@ static int nfs_parse_uri(const char *filename, QDict *options, Error **errp)
goto out;
}

- qdict_put(options, "server.host", qstring_from_str(uri->server));
- qdict_put(options, "server.type", qstring_from_str("inet"));
- qdict_put(options, "path", qstring_from_str(uri->path));
+ qdict_put_str(options, "server.host", uri->server);
+ qdict_put_str(options, "server.type", "inet");
+ qdict_put_str(options, "path", uri->path);

for (i = 0; i < qp->n; i++) {
unsigned long long val;
@@ -121,23 +121,17 @@ static int nfs_parse_uri(const char *filename, QDict *options, Error **errp)
goto out;
}
if (!strcmp(qp->p[i].name, "uid")) {
- qdict_put(options, "user",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "user", qp->p[i].value);
} else if (!strcmp(qp->p[i].name, "gid")) {
- qdict_put(options, "group",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "group", qp->p[i].value);
} else if (!strcmp(qp->p[i].name, "tcp-syncnt")) {
- qdict_put(options, "tcp-syn-count",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "tcp-syn-count", qp->p[i].value);
} else if (!strcmp(qp->p[i].name, "readahead")) {
- qdict_put(options, "readahead-size",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "readahead-size", qp->p[i].value);
} else if (!strcmp(qp->p[i].name, "pagecache")) {
- qdict_put(options, "page-cache-size",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "page-cache-size", qp->p[i].value);
} else if (!strcmp(qp->p[i].name, "debug")) {
- qdict_put(options, "debug",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "debug", qp->p[i].value);
} else {
error_setg(errp, "Unknown NFS parameter name: %s",
qp->p[i].name);
@@ -811,7 +805,7 @@ static void nfs_refresh_filename(BlockDriverState *bs, QDict *options)
QObject *server_qdict;
Visitor *ov;

- qdict_put(opts, "driver", qstring_from_str("nfs"));
+ qdict_put_str(opts, "driver", "nfs");

if (client->uid && !client->gid) {
snprintf(bs->exact_filename, sizeof(bs->exact_filename),
@@ -834,28 +828,25 @@ static void nfs_refresh_filename(BlockDriverState *bs, QDict *options)
visit_type_NFSServer(ov, NULL, &client->server, &error_abort);
visit_complete(ov, &server_qdict);
qdict_put_obj(opts, "server", server_qdict);
- qdict_put(opts, "path", qstring_from_str(client->path));
+ qdict_put_str(opts, "path", client->path);

if (client->uid) {
- qdict_put(opts, "user", qint_from_int(client->uid));
+ qdict_put_int(opts, "user", client->uid);
}
if (client->gid) {
- qdict_put(opts, "group", qint_from_int(client->gid));
+ qdict_put_int(opts, "group", client->gid);
}
if (client->tcp_syncnt) {
- qdict_put(opts, "tcp-syn-cnt",
- qint_from_int(client->tcp_syncnt));
+ qdict_put_int(opts, "tcp-syn-cnt", client->tcp_syncnt);
}
if (client->readahead) {
- qdict_put(opts, "readahead-size",
- qint_from_int(client->readahead));
+ qdict_put_int(opts, "readahead-size", client->readahead);
}
if (client->pagecache) {
- qdict_put(opts, "page-cache-size",
- qint_from_int(client->pagecache));
+ qdict_put_int(opts, "page-cache-size", client->pagecache);
}
if (client->debug) {
- qdict_put(opts, "debug", qint_from_int(client->debug));
+ qdict_put_int(opts, "debug", client->debug);
}

visit_free(ov);
diff --git a/block/null.c b/block/null.c
index b300390..876f909 100644
--- a/block/null.c
+++ b/block/null.c
@@ -232,7 +232,7 @@ static void null_refresh_filename(BlockDriverState *bs, QDict *opts)
bs->drv->format_name);
}

- qdict_put(opts, "driver", qstring_from_str(bs->drv->format_name));
+ qdict_put_str(opts, "driver", bs->drv->format_name);
bs->full_open_options = opts;
}

diff --git a/block/qcow2.c b/block/qcow2.c
index 6a92d2e..49c737f 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2265,7 +2265,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,
* table)
*/
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str("qcow2"));
+ qdict_put_str(options, "driver", "qcow2");
blk = blk_new_open(filename, NULL, options,
BDRV_O_RDWR | BDRV_O_RESIZE | BDRV_O_NO_FLUSH,
&local_err);
@@ -2327,7 +2327,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,

/* Reopen the image without BDRV_O_NO_FLUSH to flush it before returning */
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str("qcow2"));
+ qdict_put_str(options, "driver", "qcow2");
blk = blk_new_open(filename, NULL, options,
BDRV_O_RDWR | BDRV_O_NO_BACKING, &local_err);
if (blk == NULL) {
diff --git a/block/quorum.c b/block/quorum.c
index f7949e2..1b2a8c3 100644
--- a/block/quorum.c
+++ b/block/quorum.c
@@ -1100,10 +1100,10 @@ static void quorum_refresh_filename(BlockDriverState *bs, QDict *options)
}

opts = qdict_new();
- qdict_put(opts, "driver", qstring_from_str("quorum"));
- qdict_put(opts, QUORUM_OPT_VOTE_THRESHOLD, qint_from_int(s->threshold));
- qdict_put(opts, QUORUM_OPT_BLKVERIFY, qbool_from_bool(s->is_blkverify));
- qdict_put(opts, QUORUM_OPT_REWRITE, qbool_from_bool(s->rewrite_corrupted));
+ qdict_put_str(opts, "driver", "quorum");
+ qdict_put_int(opts, QUORUM_OPT_VOTE_THRESHOLD, s->threshold);
+ qdict_put_bool(opts, QUORUM_OPT_BLKVERIFY, s->is_blkverify);
+ qdict_put_bool(opts, QUORUM_OPT_REWRITE, s->rewrite_corrupted);
qdict_put(opts, "children", children);

bs->full_open_options = opts;
diff --git a/block/rbd.c b/block/rbd.c
index 1ceeeb5..2354ffc 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -154,20 +154,20 @@ static void qemu_rbd_parse_filename(const char *filename, QDict *options,
goto done;
}
qemu_rbd_unescape(found_str);
- qdict_put(options, "pool", qstring_from_str(found_str));
+ qdict_put_str(options, "pool", found_str);

if (strchr(p, '@')) {
found_str = qemu_rbd_next_tok(p, '@', &p);
qemu_rbd_unescape(found_str);
- qdict_put(options, "image", qstring_from_str(found_str));
+ qdict_put_str(options, "image", found_str);

found_str = qemu_rbd_next_tok(p, ':', &p);
qemu_rbd_unescape(found_str);
- qdict_put(options, "snapshot", qstring_from_str(found_str));
+ qdict_put_str(options, "snapshot", found_str);
} else {
found_str = qemu_rbd_next_tok(p, ':', &p);
qemu_rbd_unescape(found_str);
- qdict_put(options, "image", qstring_from_str(found_str));
+ qdict_put_str(options, "image", found_str);
}
if (!p) {
goto done;
@@ -189,9 +189,9 @@ static void qemu_rbd_parse_filename(const char *filename, QDict *options,
qemu_rbd_unescape(value);

if (!strcmp(name, "conf")) {
- qdict_put(options, "conf", qstring_from_str(value));
+ qdict_put_str(options, "conf", value);
} else if (!strcmp(name, "id")) {
- qdict_put(options, "user" , qstring_from_str(value));
+ qdict_put_str(options, "user", value);
} else {
/*
* We pass these internally to qemu_rbd_set_keypairs(), so
@@ -204,8 +204,8 @@ static void qemu_rbd_parse_filename(const char *filename, QDict *options,
if (!keypairs) {
keypairs = qlist_new();
}
- qlist_append(keypairs, qstring_from_str(name));
- qlist_append(keypairs, qstring_from_str(value));
+ qlist_append_str(keypairs, name);
+ qlist_append_str(keypairs, value);
}
}

diff --git a/block/snapshot.c b/block/snapshot.c
index 06b1185..a46564e 100644
--- a/block/snapshot.c
+++ b/block/snapshot.c
@@ -200,7 +200,7 @@ int bdrv_snapshot_goto(BlockDriverState *bs,

qdict_extract_subqdict(options, &file_options, "file.");
QDECREF(file_options);
- qdict_put(options, "file", qstring_from_str(bdrv_get_node_name(file)));
+ qdict_put_str(options, "file", bdrv_get_node_name(file));

drv->bdrv_close(bs);
bdrv_unref_child(bs, bs->file);
diff --git a/block/ssh.c b/block/ssh.c
index 471ba8a..34a2f79 100644
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -227,24 +227,23 @@ static int parse_uri(const char *filename, QDict *options, Error **errp)
}

if(uri->user && strcmp(uri->user, "") != 0) {
- qdict_put(options, "user", qstring_from_str(uri->user));
+ qdict_put_str(options, "user", uri->user);
}

- qdict_put(options, "server.host", qstring_from_str(uri->server));
+ qdict_put_str(options, "server.host", uri->server);

port_str = g_strdup_printf("%d", uri->port ?: 22);
- qdict_put(options, "server.port", qstring_from_str(port_str));
+ qdict_put_str(options, "server.port", port_str);
g_free(port_str);

- qdict_put(options, "path", qstring_from_str(uri->path));
+ qdict_put_str(options, "path", uri->path);

/* Pick out any query parameters that we understand, and ignore
* the rest.
*/
for (i = 0; i < qp->n; ++i) {
if (strcmp(qp->p[i].name, "host_key_check") == 0) {
- qdict_put(options, "host_key_check",
- qstring_from_str(qp->p[i].value));
+ qdict_put_str(options, "host_key_check", qp->p[i].value);
}
}

@@ -574,9 +573,8 @@ static bool ssh_process_legacy_socket_options(QDict *output_opts,
}

if (host) {
- qdict_put(output_opts, "server.host", qstring_from_str(host));
- qdict_put(output_opts, "server.port",
- qstring_from_str(port ?: stringify(22)));
+ qdict_put_str(output_opts, "server.host", host);
+ qdict_put_str(output_opts, "server.port", port ?: stringify(22));
}

return true;
diff --git a/block/vvfat.c b/block/vvfat.c
index af5153d..8b4e4eb 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -1057,10 +1057,10 @@ static void vvfat_parse_filename(const char *filename, QDict *options,
}

/* Fill in the options QDict */
- qdict_put(options, "dir", qstring_from_str(filename));
- qdict_put(options, "fat-type", qint_from_int(fat_type));
- qdict_put(options, "floppy", qbool_from_bool(floppy));
- qdict_put(options, "rw", qbool_from_bool(rw));
+ qdict_put_str(options, "dir", filename);
+ qdict_put_int(options, "fat-type", fat_type);
+ qdict_put_bool(options, "floppy", floppy);
+ qdict_put_bool(options, "rw", rw);
}

static int vvfat_open(BlockDriverState *bs, QDict *options, int flags,
@@ -3040,7 +3040,7 @@ static int enable_write_target(BlockDriverState *bs, Error **errp)
}

options = qdict_new();
- qdict_put(options, "write-target.driver", qstring_from_str("qcow"));
+ qdict_put_str(options, "write-target.driver", "qcow");
s->qcow = bdrv_open_child(s->qcow_filename, options, "write-target", bs,
&child_vvfat_qcow, false, errp);
QDECREF(options);
diff --git a/blockdev.c b/blockdev.c
index 4927914..e2f9c1e 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -527,7 +527,7 @@ static BlockBackend *blockdev_init(const char *file, QDict *bs_opts,
error_setg(errp, "Cannot specify both 'driver' and 'format'");
goto early_err;
}
- qdict_put(bs_opts, "driver", qstring_from_str(buf));
+ qdict_put_str(bs_opts, "driver", buf);
}

on_write_error = BLOCKDEV_ON_ERROR_ENOSPC;
@@ -903,10 +903,8 @@ DriveInfo *drive_new(QemuOpts *all_opts, BlockInterfaceType block_default_type)
copy_on_read = false;
}

- qdict_put(bs_opts, BDRV_OPT_READ_ONLY,
- qstring_from_str(read_only ? "on" : "off"));
- qdict_put(bs_opts, "copy-on-read",
- qstring_from_str(copy_on_read ? "on" :"off"));
+ qdict_put_str(bs_opts, BDRV_OPT_READ_ONLY, read_only ? "on" : "off");
+ qdict_put_str(bs_opts, "copy-on-read", copy_on_read ? "on" : "off");

/* Controller type */
value = qemu_opt_get(legacy_opts, "if");
@@ -1030,7 +1028,7 @@ DriveInfo *drive_new(QemuOpts *all_opts, BlockInterfaceType block_default_type)
new_id = g_strdup_printf("%s%s%i", if_name[type],
mediastr, unit_id);
}
- qdict_put(bs_opts, "id", qstring_from_str(new_id));
+ qdict_put_str(bs_opts, "id", new_id);
g_free(new_id);
}

@@ -1067,7 +1065,7 @@ DriveInfo *drive_new(QemuOpts *all_opts, BlockInterfaceType block_default_type)
error_report("werror is not supported by this bus type");
goto fail;
}
- qdict_put(bs_opts, "werror", qstring_from_str(werror));
+ qdict_put_str(bs_opts, "werror", werror);
}

rerror = qemu_opt_get(legacy_opts, "rerror");
@@ -1077,7 +1075,7 @@ DriveInfo *drive_new(QemuOpts *all_opts, BlockInterfaceType block_default_type)
error_report("rerror is not supported by this bus type");
goto fail;
}
- qdict_put(bs_opts, "rerror", qstring_from_str(rerror));
+ qdict_put_str(bs_opts, "rerror", rerror);
}

/* Actual block device init: Functionality shared with blockdev-add */
@@ -1737,10 +1735,9 @@ static void external_snapshot_prepare(BlkActionState *common,

options = qdict_new();
if (s->has_snapshot_node_name) {
- qdict_put(options, "node-name",
- qstring_from_str(snapshot_node_name));
+ qdict_put_str(options, "node-name", snapshot_node_name);
}
- qdict_put(options, "driver", qstring_from_str(format));
+ qdict_put_str(options, "driver", format);

flags |= BDRV_O_NO_BACKING;
}
@@ -2579,11 +2576,10 @@ void qmp_blockdev_change_medium(bool has_device, const char *device,

options = qdict_new();
detect_zeroes = blk_get_detect_zeroes_from_root_state(blk);
- qdict_put(options, "detect-zeroes",
- qstring_from_str(detect_zeroes ? "on" : "off"));
+ qdict_put_str(options, "detect-zeroes", detect_zeroes ? "on" : "off");

if (has_format) {
- qdict_put(options, "driver", qstring_from_str(format));
+ qdict_put_str(options, "driver", format);
}

medium_bs = bdrv_open(filename, NULL, options, bdrv_flags, errp);
@@ -3251,7 +3247,7 @@ static BlockJob *do_drive_backup(DriveBackup *backup, BlockJobTxn *txn,

if (backup->format) {
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str(backup->format));
+ qdict_put_str(options, "driver", backup->format);
}

target_bs = bdrv_open(backup->target, NULL, options, flags, errp);
@@ -3555,10 +3551,10 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)

options = qdict_new();
if (arg->has_node_name) {
- qdict_put(options, "node-name", qstring_from_str(arg->node_name));
+ qdict_put_str(options, "node-name", arg->node_name);
}
if (format) {
- qdict_put(options, "driver", qstring_from_str(format));
+ qdict_put_str(options, "driver", format);
}

/* Mirroring takes care of copy-on-write using the source's backing
diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c
index 456a2d5..47b2ca1 100644
--- a/hw/block/xen_disk.c
+++ b/hw/block/xen_disk.c
@@ -1082,7 +1082,7 @@ static int blk_connect(struct XenDevice *xendev)

if (strcmp(blkdev->fileproto, "<unset>")) {
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str(blkdev->fileproto));
+ qdict_put_str(options, "driver", blkdev->fileproto);
}

/* setup via xenbus -> create new block driver instance */
diff --git a/hw/usb/xen-usb.c b/hw/usb/xen-usb.c
index 8e676e6..6659415 100644
--- a/hw/usb/xen-usb.c
+++ b/hw/usb/xen-usb.c
@@ -746,16 +746,16 @@ static void usbback_portid_add(struct usbback_info *usbif, unsigned port,
portname++;

qdict = qdict_new();
- qdict_put(qdict, "driver", qstring_from_str("usb-host"));
+ qdict_put_str(qdict, "driver", "usb-host");
tmp = g_strdup_printf("%s.0", usbif->xendev.qdev.id);
- qdict_put(qdict, "bus", qstring_from_str(tmp));
+ qdict_put_str(qdict, "bus", tmp);
g_free(tmp);
tmp = g_strdup_printf("%s-%u", usbif->xendev.qdev.id, port);
- qdict_put(qdict, "id", qstring_from_str(tmp));
+ qdict_put_str(qdict, "id", tmp);
g_free(tmp);
- qdict_put(qdict, "port", qint_from_int(port));
- qdict_put(qdict, "hostbus", qint_from_int(atoi(busid)));
- qdict_put(qdict, "hostport", qstring_from_str(portname));
+ qdict_put_int(qdict, "port", port);
+ qdict_put_int(qdict, "hostbus", atoi(busid));
+ qdict_put_str(qdict, "hostport", portname);
opts = qemu_opts_from_qdict(qemu_find_opts("device"), qdict, &local_err);
if (local_err) {
goto err;
diff --git a/monitor.c b/monitor.c
index 2799a37..957361c 100644
--- a/monitor.c
+++ b/monitor.c
@@ -2676,7 +2676,7 @@ static QDict *monitor_parse_arguments(Monitor *mon,
}
goto fail;
}
- qdict_put(qdict, key, qstring_from_str(buf));
+ qdict_put_str(qdict, key, buf);
}
break;
case 'O':
@@ -2778,9 +2778,9 @@ static QDict *monitor_parse_arguments(Monitor *mon,
size = -1;
}
}
- qdict_put(qdict, "count", qint_from_int(count));
- qdict_put(qdict, "format", qint_from_int(format));
- qdict_put(qdict, "size", qint_from_int(size));
+ qdict_put_int(qdict, "count", count);
+ qdict_put_int(qdict, "format", format);
+ qdict_put_int(qdict, "size", size);
}
break;
case 'i':
@@ -2823,7 +2823,7 @@ static QDict *monitor_parse_arguments(Monitor *mon,
}
val <<= 20;
}
- qdict_put(qdict, key, qint_from_int(val));
+ qdict_put_int(qdict, key, val);
}
break;
case 'o':
@@ -2846,7 +2846,7 @@ static QDict *monitor_parse_arguments(Monitor *mon,
monitor_printf(mon, "invalid size\n");
goto fail;
}
- qdict_put(qdict, key, qint_from_int(val));
+ qdict_put_int(qdict, key, val);
p = end;
}
break;
@@ -2902,7 +2902,7 @@ static QDict *monitor_parse_arguments(Monitor *mon,
monitor_printf(mon, "Expected 'on' or 'off'\n");
goto fail;
}
- qdict_put(qdict, key, qbool_from_bool(val));
+ qdict_put_bool(qdict, key, val);
}
break;
case '-':
@@ -2933,7 +2933,7 @@ static QDict *monitor_parse_arguments(Monitor *mon,
} else {
/* has option */
p++;
- qdict_put(qdict, key, qbool_from_bool(true));
+ qdict_put_bool(qdict, key, true);
}
}
}
@@ -2959,7 +2959,7 @@ static QDict *monitor_parse_arguments(Monitor *mon,
cmd->name);
goto fail;
}
- qdict_put(qdict, key, qstring_from_str(p));
+ qdict_put_str(qdict, key, p);
p += len;
}
break;
@@ -3738,9 +3738,8 @@ static void handle_qmp_command(JSONMessageParser *parser, GQueue *tokens)
QapiErrorClass_lookup[ERROR_CLASS_COMMAND_NOT_FOUND])) {
/* Provide a more useful error message */
qdict_del(qdict, "desc");
- qdict_put(qdict, "desc",
- qstring_from_str("Expecting capabilities negotiation"
- " with 'qmp_capabilities'"));
+ qdict_put_str(qdict, "desc", "Expecting capabilities negotiation"
+ " with 'qmp_capabilities'");
}
}

diff --git a/qapi/qmp-event.c b/qapi/qmp-event.c
index 802ede4..ba3029c 100644
--- a/qapi/qmp-event.c
+++ b/qapi/qmp-event.c
@@ -51,7 +51,7 @@ static void timestamp_put(QDict *qdict)
QDict *qmp_event_build_dict(const char *event_name)
{
QDict *dict = qdict_new();
- qdict_put(dict, "event", qstring_from_str(event_name));
+ qdict_put_str(dict, "event", event_name);
timestamp_put(dict);
return dict;
}
diff --git a/qemu-img.c b/qemu-img.c
index 4b2d59a..2e61561 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -313,7 +313,7 @@ static BlockBackend *img_open_file(const char *filename,

if (fmt) {
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str(fmt));
+ qdict_put_str(options, "driver", fmt);
}

blk = blk_new_open(filename, NULL, options, flags, &local_err);
@@ -3162,7 +3162,7 @@ static int img_rebase(int argc, char **argv)

if (bs->backing_format[0] != '\0') {
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str(bs->backing_format));
+ qdict_put_str(options, "driver", bs->backing_format);
}

bdrv_get_backing_filename(bs, backing_name, sizeof(backing_name));
@@ -3179,7 +3179,7 @@ static int img_rebase(int argc, char **argv)
if (out_baseimg[0]) {
if (out_basefmt) {
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str(out_basefmt));
+ qdict_put_str(options, "driver", out_basefmt);
} else {
options = NULL;
}
diff --git a/qemu-io.c b/qemu-io.c
index 427cbae..ed0e2dc 100644
--- a/qemu-io.c
+++ b/qemu-io.c
@@ -601,7 +601,7 @@ int main(int argc, char **argv)
} else {
if (format) {
opts = qdict_new();
- qdict_put(opts, "driver", qstring_from_str(format));
+ qdict_put_str(opts, "driver", format);
}
if (openfile(argv[optind], flags, writethrough, opts)) {
exit(1);
diff --git a/qemu-nbd.c b/qemu-nbd.c
index e080fb7..e4f00e2 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -959,7 +959,7 @@ int main(int argc, char **argv)
} else {
if (fmt) {
options = qdict_new();
- qdict_put(options, "driver", qstring_from_str(fmt));
+ qdict_put_str(options, "driver", fmt);
}
blk = blk_new_open(srcpath, NULL, options, flags, &local_err);
}
diff --git a/qobject/qdict.c b/qobject/qdict.c
index 291eef1..88e2ecd 100644
--- a/qobject/qdict.c
+++ b/qobject/qdict.c
@@ -463,7 +463,7 @@ void qdict_set_default_str(QDict *dst, const char *key, const char *val)
return;
}

- qdict_put(dst, key, qstring_from_str(val));
+ qdict_put_str(dst, key, val);
}

static void qdict_flatten_qdict(QDict *qdict, QDict *target,
diff --git a/target/s390x/cpu_models.c b/target/s390x/cpu_models.c
index ce461cc..8d27363 100644
--- a/target/s390x/cpu_models.c
+++ b/target/s390x/cpu_models.c
@@ -376,12 +376,12 @@ static void cpu_model_from_info(S390CPUModel *model, const CpuModelInfo *info,

static void qdict_add_disabled_feat(const char *name, void *opaque)
{
- qdict_put(opaque, name, qbool_from_bool(false));
+ qdict_put_bool(opaque, name, false);
}

static void qdict_add_enabled_feat(const char *name, void *opaque)
{
- qdict_put(opaque, name, qbool_from_bool(true));
+ qdict_put_bool(opaque, name, true);
}

/* convert S390CPUDef into a static CpuModelInfo */
diff --git a/tests/check-qdict.c b/tests/check-qdict.c
index 9cdc985..3effca5 100644
--- a/tests/check-qdict.c
+++ b/tests/check-qdict.c
@@ -47,7 +47,7 @@ static void qdict_put_obj_test(void)
qdict = qdict_new();

// key "" will have tdb hash 12345
- qdict_put(qdict, "", qint_from_int(num));
+ qdict_put_int(qdict, "", num);

g_assert(qdict_size(qdict) == 1);
ent = QLIST_FIRST(&qdict->table[12345 % QDICT_BUCKET_MAX]);
@@ -66,8 +66,8 @@ static void qdict_destroy_simple_test(void)
QDict *qdict;

qdict = qdict_new();
- qdict_put(qdict, "num", qint_from_int(0));
- qdict_put(qdict, "str", qstring_from_str("foo"));
+ qdict_put_int(qdict, "num", 0);
+ qdict_put_str(qdict, "str", "foo");

QDECREF(qdict);
}
@@ -80,7 +80,7 @@ static void qdict_get_test(void)
const char *key = "test";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qint_from_int(value));
+ qdict_put_int(tests_dict, key, value);

obj = qdict_get(tests_dict, key);
g_assert(obj != NULL);
@@ -98,7 +98,7 @@ static void qdict_get_int_test(void)
const char *key = "int";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qint_from_int(value));
+ qdict_put_int(tests_dict, key, value);

ret = qdict_get_int(tests_dict, key);
g_assert(ret == value);
@@ -113,7 +113,7 @@ static void qdict_get_try_int_test(void)
const char *key = "int";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qint_from_int(value));
+ qdict_put_int(tests_dict, key, value);

ret = qdict_get_try_int(tests_dict, key, 0);
g_assert(ret == value);
@@ -128,7 +128,7 @@ static void qdict_get_str_test(void)
const char *str = "string";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qstring_from_str(str));
+ qdict_put_str(tests_dict, key, str);

p = qdict_get_str(tests_dict, key);
g_assert(p != NULL);
@@ -144,7 +144,7 @@ static void qdict_get_try_str_test(void)
const char *str = "string";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qstring_from_str(str));
+ qdict_put_str(tests_dict, key, str);

p = qdict_get_try_str(tests_dict, key);
g_assert(p != NULL);
@@ -188,7 +188,7 @@ static void qdict_haskey_test(void)
const char *key = "test";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qint_from_int(0));
+ qdict_put_int(tests_dict, key, 0);
g_assert(qdict_haskey(tests_dict, key) == 1);

QDECREF(tests_dict);
@@ -199,7 +199,7 @@ static void qdict_del_test(void)
const char *key = "key test";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qstring_from_str("foo"));
+ qdict_put_str(tests_dict, key, "foo");
g_assert(qdict_size(tests_dict) == 1);

qdict_del(tests_dict, key);
@@ -226,9 +226,9 @@ static void qdict_iterapi_test(void)

g_assert(qdict_first(tests_dict) == NULL);

- qdict_put(tests_dict, "key1", qint_from_int(1));
- qdict_put(tests_dict, "key2", qint_from_int(2));
- qdict_put(tests_dict, "key3", qint_from_int(3));
+ qdict_put_int(tests_dict, "key1", 1);
+ qdict_put_int(tests_dict, "key2", 2);
+ qdict_put_int(tests_dict, "key3", 3);

count = 0;
for (ent = qdict_first(tests_dict); ent; ent = qdict_next(tests_dict, ent)){
@@ -294,20 +294,20 @@ static void qdict_flatten_test(void)
* }
*/

- qdict_put(dict1, "a", qint_from_int(0));
- qdict_put(dict1, "b", qint_from_int(1));
+ qdict_put_int(dict1, "a", 0);
+ qdict_put_int(dict1, "b", 1);

- qlist_append(list1, qint_from_int(23));
- qlist_append(list1, qint_from_int(66));
+ qlist_append_int(list1, 23);
+ qlist_append_int(list1, 66);
qlist_append(list1, dict1);
- qlist_append(list2, qint_from_int(42));
+ qlist_append_int(list2, 42);
qlist_append(list2, list1);

- qdict_put(dict2, "c", qint_from_int(2));
- qdict_put(dict2, "d", qint_from_int(3));
+ qdict_put_int(dict2, "c", 2);
+ qdict_put_int(dict2, "d", 3);
qdict_put(dict3, "e", list2);
qdict_put(dict3, "f", dict2);
- qdict_put(dict3, "g", qint_from_int(4));
+ qdict_put_int(dict3, "g", 4);

qdict_flatten(dict3);

@@ -369,12 +369,12 @@ static void qdict_array_split_test(void)
* This example is given in the comment of qdict_array_split().
*/

- qdict_put(test_dict, "1.x", qint_from_int(0));
- qdict_put(test_dict, "4.y", qint_from_int(1));
- qdict_put(test_dict, "0.a", qint_from_int(42));
- qdict_put(test_dict, "o.o", qint_from_int(7));
- qdict_put(test_dict, "0.b", qint_from_int(23));
- qdict_put(test_dict, "2", qint_from_int(66));
+ qdict_put_int(test_dict, "1.x", 0);
+ qdict_put_int(test_dict, "4.y", 1);
+ qdict_put_int(test_dict, "0.a", 42);
+ qdict_put_int(test_dict, "o.o", 7);
+ qdict_put_int(test_dict, "0.b", 23);
+ qdict_put_int(test_dict, "2", 66);

qdict_array_split(test_dict, &test_list);

@@ -441,9 +441,9 @@ static void qdict_array_split_test(void)

test_dict = qdict_new();

- qdict_put(test_dict, "0", qint_from_int(42));
- qdict_put(test_dict, "1", qint_from_int(23));
- qdict_put(test_dict, "1.x", qint_from_int(84));
+ qdict_put_int(test_dict, "0", 42);
+ qdict_put_int(test_dict, "1", 23);
+ qdict_put_int(test_dict, "1.x", 84);

qdict_array_split(test_dict, &test_list);

@@ -472,38 +472,38 @@ static void qdict_array_entries_test(void)

g_assert_cmpint(qdict_array_entries(dict, "foo."), ==, 0);

- qdict_put(dict, "bar", qint_from_int(0));
- qdict_put(dict, "baz.0", qint_from_int(0));
+ qdict_put_int(dict, "bar", 0);
+ qdict_put_int(dict, "baz.0", 0);
g_assert_cmpint(qdict_array_entries(dict, "foo."), ==, 0);

- qdict_put(dict, "foo.1", qint_from_int(0));
+ qdict_put_int(dict, "foo.1", 0);
g_assert_cmpint(qdict_array_entries(dict, "foo."), ==, -EINVAL);
- qdict_put(dict, "foo.0", qint_from_int(0));
+ qdict_put_int(dict, "foo.0", 0);
g_assert_cmpint(qdict_array_entries(dict, "foo."), ==, 2);
- qdict_put(dict, "foo.bar", qint_from_int(0));
+ qdict_put_int(dict, "foo.bar", 0);
g_assert_cmpint(qdict_array_entries(dict, "foo."), ==, -EINVAL);
qdict_del(dict, "foo.bar");

- qdict_put(dict, "foo.2.a", qint_from_int(0));
- qdict_put(dict, "foo.2.b", qint_from_int(0));
- qdict_put(dict, "foo.2.c", qint_from_int(0));
+ qdict_put_int(dict, "foo.2.a", 0);
+ qdict_put_int(dict, "foo.2.b", 0);
+ qdict_put_int(dict, "foo.2.c", 0);
g_assert_cmpint(qdict_array_entries(dict, "foo."), ==, 3);
g_assert_cmpint(qdict_array_entries(dict, ""), ==, -EINVAL);

QDECREF(dict);

dict = qdict_new();
- qdict_put(dict, "1", qint_from_int(0));
+ qdict_put_int(dict, "1", 0);
g_assert_cmpint(qdict_array_entries(dict, ""), ==, -EINVAL);
- qdict_put(dict, "0", qint_from_int(0));
+ qdict_put_int(dict, "0", 0);
g_assert_cmpint(qdict_array_entries(dict, ""), ==, 2);
- qdict_put(dict, "bar", qint_from_int(0));
+ qdict_put_int(dict, "bar", 0);
g_assert_cmpint(qdict_array_entries(dict, ""), ==, -EINVAL);
qdict_del(dict, "bar");

- qdict_put(dict, "2.a", qint_from_int(0));
- qdict_put(dict, "2.b", qint_from_int(0));
- qdict_put(dict, "2.c", qint_from_int(0));
+ qdict_put_int(dict, "2.a", 0);
+ qdict_put_int(dict, "2.b", 0);
+ qdict_put_int(dict, "2.c", 0);
g_assert_cmpint(qdict_array_entries(dict, ""), ==, 3);

QDECREF(dict);
@@ -529,7 +529,7 @@ static void qdict_join_test(void)

/* First iteration: Test movement */
/* Second iteration: Test empty source and non-empty destination */
- qdict_put(dict2, "foo", qint_from_int(42));
+ qdict_put_int(dict2, "foo", 42);

for (i = 0; i < 2; i++) {
qdict_join(dict1, dict2, overwrite);
@@ -541,7 +541,7 @@ static void qdict_join_test(void)
}

/* Test non-empty source and destination without conflict */
- qdict_put(dict2, "bar", qint_from_int(23));
+ qdict_put_int(dict2, "bar", 23);

qdict_join(dict1, dict2, overwrite);

@@ -552,7 +552,7 @@ static void qdict_join_test(void)
g_assert(qdict_get_int(dict1, "bar") == 23);

/* Test conflict */
- qdict_put(dict2, "foo", qint_from_int(84));
+ qdict_put_int(dict2, "foo", 84);

qdict_join(dict1, dict2, overwrite);

@@ -594,15 +594,15 @@ static void qdict_crumple_test_recursive(void)
QList *rules;

src = qdict_new();
- qdict_put(src, "vnc.listen.addr", qstring_from_str("127.0.0.1"));
- qdict_put(src, "vnc.listen.port", qstring_from_str("5901"));
- qdict_put(src, "vnc.acl.rules.0.match", qstring_from_str("fred"));
- qdict_put(src, "vnc.acl.rules.0.policy", qstring_from_str("allow"));
- qdict_put(src, "vnc.acl.rules.1.match", qstring_from_str("bob"));
- qdict_put(src, "vnc.acl.rules.1.policy", qstring_from_str("deny"));
- qdict_put(src, "vnc.acl.default", qstring_from_str("deny"));
- qdict_put(src, "vnc.acl..name", qstring_from_str("acl0"));
- qdict_put(src, "vnc.acl.rule..name", qstring_from_str("acl0"));
+ qdict_put_str(src, "vnc.listen.addr", "127.0.0.1");
+ qdict_put_str(src, "vnc.listen.port", "5901");
+ qdict_put_str(src, "vnc.acl.rules.0.match", "fred");
+ qdict_put_str(src, "vnc.acl.rules.0.policy", "allow");
+ qdict_put_str(src, "vnc.acl.rules.1.match", "bob");
+ qdict_put_str(src, "vnc.acl.rules.1.policy", "deny");
+ qdict_put_str(src, "vnc.acl.default", "deny");
+ qdict_put_str(src, "vnc.acl..name", "acl0");
+ qdict_put_str(src, "vnc.acl.rule..name", "acl0");

dst = qobject_to_qdict(qdict_crumple(src, &error_abort));
g_assert(dst);
@@ -669,8 +669,8 @@ static void qdict_crumple_test_bad_inputs(void)

src = qdict_new();
/* rule.0 can't be both a string and a dict */
- qdict_put(src, "rule.0", qstring_from_str("fred"));
- qdict_put(src, "rule.0.policy", qstring_from_str("allow"));
+ qdict_put_str(src, "rule.0", "fred");
+ qdict_put_str(src, "rule.0.policy", "allow");

g_assert(qdict_crumple(src, &error) == NULL);
g_assert(error != NULL);
@@ -680,8 +680,8 @@ static void qdict_crumple_test_bad_inputs(void)

src = qdict_new();
/* rule can't be both a list and a dict */
- qdict_put(src, "rule.0", qstring_from_str("fred"));
- qdict_put(src, "rule.a", qstring_from_str("allow"));
+ qdict_put_str(src, "rule.0", "fred");
+ qdict_put_str(src, "rule.a", "allow");

g_assert(qdict_crumple(src, &error) == NULL);
g_assert(error != NULL);
@@ -692,7 +692,7 @@ static void qdict_crumple_test_bad_inputs(void)
src = qdict_new();
/* The input should be flat, ie no dicts or lists */
qdict_put(src, "rule.a", qdict_new());
- qdict_put(src, "rule.b", qstring_from_str("allow"));
+ qdict_put_str(src, "rule.b", "allow");

g_assert(qdict_crumple(src, &error) == NULL);
g_assert(error != NULL);
@@ -702,8 +702,8 @@ static void qdict_crumple_test_bad_inputs(void)

src = qdict_new();
/* List indexes must not have gaps */
- qdict_put(src, "rule.0", qstring_from_str("deny"));
- qdict_put(src, "rule.3", qstring_from_str("allow"));
+ qdict_put_str(src, "rule.0", "deny");
+ qdict_put_str(src, "rule.3", "allow");

g_assert(qdict_crumple(src, &error) == NULL);
g_assert(error != NULL);
@@ -713,8 +713,8 @@ static void qdict_crumple_test_bad_inputs(void)

src = qdict_new();
/* List indexes must be in %zu format */
- qdict_put(src, "rule.0", qstring_from_str("deny"));
- qdict_put(src, "rule.+1", qstring_from_str("allow"));
+ qdict_put_str(src, "rule.0", "deny");
+ qdict_put_str(src, "rule.+1", "allow");

g_assert(qdict_crumple(src, &error) == NULL);
g_assert(error != NULL);
@@ -733,8 +733,8 @@ static void qdict_put_exists_test(void)
const char *key = "exists";
QDict *tests_dict = qdict_new();

- qdict_put(tests_dict, key, qint_from_int(1));
- qdict_put(tests_dict, key, qint_from_int(2));
+ qdict_put_int(tests_dict, key, 1);
+ qdict_put_int(tests_dict, key, 2);

value = qdict_get_int(tests_dict, key);
g_assert(value == 2);
diff --git a/tests/check-qlist.c b/tests/check-qlist.c
index e16da5e..4983867 100644
--- a/tests/check-qlist.c
+++ b/tests/check-qlist.c
@@ -74,7 +74,7 @@ static void qlist_destroy_test(void)
qlist = qlist_new();

for (i = 0; i < 42; i++)
- qlist_append(qlist, qint_from_int(i));
+ qlist_append_int(qlist, i);

QDECREF(qlist);
}
@@ -103,7 +103,7 @@ static void qlist_iter_test(void)
qlist = qlist_new();

for (i = 0; i < iter_max; i++)
- qlist_append(qlist, qint_from_int(i));
+ qlist_append_int(qlist, i);

iter_called = 0;
qlist_iter(qlist, iter_func, NULL);
diff --git a/tests/device-introspect-test.c b/tests/device-introspect-test.c
index c5637cc..b1abb2a 100644
--- a/tests/device-introspect-test.c
+++ b/tests/device-introspect-test.c
@@ -32,9 +32,9 @@ static QList *qom_list_types(const char *implements, bool abstract)
QList *ret;
QDict *args = qdict_new();

- qdict_put(args, "abstract", qbool_from_bool(abstract));
+ qdict_put_bool(args, "abstract", abstract);
if (implements) {
- qdict_put(args, "implements", qstring_from_str(implements));
+ qdict_put_str(args, "implements", implements);
}
resp = qmp("{'execute': 'qom-list-types',"
" 'arguments': %p }", args);
diff --git a/tests/test-qemu-opts.c b/tests/test-qemu-opts.c
index 0ad74b4..cc1bb1a 100644
--- a/tests/test-qemu-opts.c
+++ b/tests/test-qemu-opts.c
@@ -299,7 +299,7 @@ static void test_qemu_opt_get_size(void)
dict = qdict_new();
g_assert(dict != NULL);

- qdict_put(dict, "size1", qstring_from_str("10"));
+ qdict_put_str(dict, "size1", "10");

qemu_opts_absorb_qdict(opts, dict, &error_abort);
g_assert(error_abort == NULL);
@@ -309,7 +309,7 @@ static void test_qemu_opt_get_size(void)
g_assert(opt == 10);

/* reset value */
- qdict_put(dict, "size1", qstring_from_str("15"));
+ qdict_put_str(dict, "size1", "15");

qemu_opts_absorb_qdict(opts, dict, &error_abort);
g_assert(error_abort == NULL);
diff --git a/tests/test-qmp-commands.c b/tests/test-qmp-commands.c
index ac3fd03..acdded4 100644
--- a/tests/test-qmp-commands.c
+++ b/tests/test-qmp-commands.c
@@ -94,7 +94,7 @@ static void test_dispatch_cmd(void)
QDict *req = qdict_new();
QObject *resp;

- qdict_put(req, "execute", qstring_from_str("user_def_cmd"));
+ qdict_put_str(req, "execute", "user_def_cmd");

resp = qmp_dispatch(&qmp_commands, QOBJECT(req));
assert(resp != NULL);
@@ -111,7 +111,7 @@ static void test_dispatch_cmd_failure(void)
QDict *args = qdict_new();
QObject *resp;

- qdict_put(req, "execute", qstring_from_str("user_def_cmd2"));
+ qdict_put_str(req, "execute", "user_def_cmd2");

resp = qmp_dispatch(&qmp_commands, QOBJECT(req));
assert(resp != NULL);
@@ -122,10 +122,10 @@ static void test_dispatch_cmd_failure(void)

/* check that with extra arguments it throws an error */
req = qdict_new();
- qdict_put(args, "a", qint_from_int(66));
+ qdict_put_int(args, "a", 66);
qdict_put(req, "arguments", args);

- qdict_put(req, "execute", qstring_from_str("user_def_cmd"));
+ qdict_put_str(req, "execute", "user_def_cmd");

resp = qmp_dispatch(&qmp_commands, QOBJECT(req));
assert(resp != NULL);
@@ -164,14 +164,14 @@ static void test_dispatch_cmd_io(void)
QDict *ret_dict_dict2, *ret_dict_dict2_userdef;
QInt *ret3;

- qdict_put(ud1a, "integer", qint_from_int(42));
- qdict_put(ud1a, "string", qstring_from_str("hello"));
- qdict_put(ud1b, "integer", qint_from_int(422));
- qdict_put(ud1b, "string", qstring_from_str("hello2"));
+ qdict_put_int(ud1a, "integer", 42);
+ qdict_put_str(ud1a, "string", "hello");
+ qdict_put_int(ud1b, "integer", 422);
+ qdict_put_str(ud1b, "string", "hello2");
qdict_put(args, "ud1a", ud1a);
qdict_put(args, "ud1b", ud1b);
qdict_put(req, "arguments", args);
- qdict_put(req, "execute", qstring_from_str("user_def_cmd2"));
+ qdict_put_str(req, "execute", "user_def_cmd2");

ret = qobject_to_qdict(test_qmp_dispatch(req));

@@ -190,9 +190,9 @@ static void test_dispatch_cmd_io(void)
assert(!strcmp(qdict_get_str(ret_dict_dict2, "string"), "blah4"));
QDECREF(ret);

- qdict_put(args3, "a", qint_from_int(66));
+ qdict_put_int(args3, "a", 66);
qdict_put(req, "arguments", args3);
- qdict_put(req, "execute", qstring_from_str("guest-get-time"));
+ qdict_put_str(req, "execute", "guest-get-time");

ret3 = qobject_to_qint(test_qmp_dispatch(req));
assert(qint_get_int(ret3) == 66);
@@ -244,7 +244,7 @@ static void test_dealloc_partial(void)
Visitor *v;

ud2_dict = qdict_new();
- qdict_put(ud2_dict, "string0", qstring_from_str(text));
+ qdict_put_str(ud2_dict, "string0", text);

v = qobject_input_visitor_new(QOBJECT(ud2_dict));
visit_type_UserDefTwo(v, NULL, &ud2, &err);
diff --git a/tests/test-qmp-event.c b/tests/test-qmp-event.c
index 7bb621b..4c0f096 100644
--- a/tests/test-qmp-event.c
+++ b/tests/test-qmp-event.c
@@ -153,7 +153,7 @@ static void test_event_a(TestEventData *data,
{
QDict *d;
d = data->expect;
- qdict_put(d, "event", qstring_from_str("EVENT_A"));
+ qdict_put_str(d, "event", "EVENT_A");
qapi_event_send_event_a(&error_abort);
}

@@ -162,7 +162,7 @@ static void test_event_b(TestEventData *data,
{
QDict *d;
d = data->expect;
- qdict_put(d, "event", qstring_from_str("EVENT_B"));
+ qdict_put_str(d, "event", "EVENT_B");
qapi_event_send_event_b(&error_abort);
}

@@ -177,16 +177,16 @@ static void test_event_c(TestEventData *data,
b.has_enum1 = false;

d_b = qdict_new();
- qdict_put(d_b, "integer", qint_from_int(2));
- qdict_put(d_b, "string", qstring_from_str("test1"));
+ qdict_put_int(d_b, "integer", 2);
+ qdict_put_str(d_b, "string", "test1");

d_data = qdict_new();
- qdict_put(d_data, "a", qint_from_int(1));
+ qdict_put_int(d_data, "a", 1);
qdict_put(d_data, "b", d_b);
- qdict_put(d_data, "c", qstring_from_str("test2"));
+ qdict_put_str(d_data, "c", "test2");

d = data->expect;
- qdict_put(d, "event", qstring_from_str("EVENT_C"));
+ qdict_put_str(d, "event", "EVENT_C");
qdict_put(d, "data", d_data);

qapi_event_send_event_c(true, 1, true, &b, "test2", &error_abort);
@@ -213,22 +213,22 @@ static void test_event_d(TestEventData *data,
a.enum2 = ENUM_ONE_VALUE2;

d_struct1 = qdict_new();
- qdict_put(d_struct1, "integer", qint_from_int(2));
- qdict_put(d_struct1, "string", qstring_from_str("test1"));
- qdict_put(d_struct1, "enum1", qstring_from_str("value1"));
+ qdict_put_int(d_struct1, "integer", 2);
+ qdict_put_str(d_struct1, "string", "test1");
+ qdict_put_str(d_struct1, "enum1", "value1");

d_a = qdict_new();
qdict_put(d_a, "struct1", d_struct1);
- qdict_put(d_a, "string", qstring_from_str("test2"));
- qdict_put(d_a, "enum2", qstring_from_str("value2"));
+ qdict_put_str(d_a, "string", "test2");
+ qdict_put_str(d_a, "enum2", "value2");

d_data = qdict_new();
qdict_put(d_data, "a", d_a);
- qdict_put(d_data, "b", qstring_from_str("test3"));
- qdict_put(d_data, "enum3", qstring_from_str("value3"));
+ qdict_put_str(d_data, "b", "test3");
+ qdict_put_str(d_data, "enum3", "value3");

d = data->expect;
- qdict_put(d, "event", qstring_from_str("EVENT_D"));
+ qdict_put_str(d, "event", "EVENT_D");
qdict_put(d, "data", d_data);

qapi_event_send_event_d(&a, "test3", false, NULL, true, ENUM_ONE_VALUE3,
diff --git a/tests/test-qobject-output-visitor.c b/tests/test-qobject-output-visitor.c
index c213fce..94b9518 100644
--- a/tests/test-qobject-output-visitor.c
+++ b/tests/test-qobject-output-visitor.c
@@ -343,9 +343,9 @@ static void test_visitor_out_any(TestOutputVisitorData *data,

visitor_reset(data);
qdict = qdict_new();
- qdict_put(qdict, "integer", qint_from_int(-42));
- qdict_put(qdict, "boolean", qbool_from_bool(true));
- qdict_put(qdict, "string", qstring_from_str("foo"));
+ qdict_put_int(qdict, "integer", -42);
+ qdict_put_bool(qdict, "boolean", true);
+ qdict_put_str(qdict, "string", "foo");
qobj = QOBJECT(qdict);
visit_type_any(data->ov, NULL, &qobj, &error_abort);
qobject_decref(qobj);
diff --git a/util/qemu-option.c b/util/qemu-option.c
index 5ce1b5c..a36cafa 100644
--- a/util/qemu-option.c
+++ b/util/qemu-option.c
@@ -1060,7 +1060,7 @@ QDict *qemu_opts_to_qdict(QemuOpts *opts, QDict *qdict)
qdict = qdict_new();
}
if (opts->id) {
- qdict_put(qdict, "id", qstring_from_str(opts->id));
+ qdict_put_str(qdict, "id", opts->id);
}
QTAILQ_FOREACH(opt, &opts->head, next) {
val = QOBJECT(qstring_from_str(opt->str));
--
2.7.4
Michael Roth
2017-08-29 00:14:01 UTC
Permalink
From: Max Filippov <***@gmail.com>

Read and write simcalls map physical memory to access I/O buffers, but
'read' simcall need to map it for writing and 'write' simcall need to
map it for reading, i.e. the opposite of what they do now. Fix that.

Cc: qemu-***@nongnu.org
Signed-off-by: Max Filippov <***@gmail.com>
(cherry picked from commit 30c2afd151cbc38c012f7b441088980807183da6)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
target/xtensa/xtensa-semi.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/target/xtensa/xtensa-semi.c b/target/xtensa/xtensa-semi.c
index a888a9d..98ae28c 100644
--- a/target/xtensa/xtensa-semi.c
+++ b/target/xtensa/xtensa-semi.c
@@ -173,7 +173,7 @@ void HELPER(simcall)(CPUXtensaState *env)
TARGET_PAGE_SIZE - (vaddr & (TARGET_PAGE_SIZE - 1));
uint32_t io_sz = page_left < len ? page_left : len;
hwaddr sz = io_sz;
- void *buf = cpu_physical_memory_map(paddr, &sz, is_write);
+ void *buf = cpu_physical_memory_map(paddr, &sz, !is_write);

if (buf) {
vaddr += io_sz;
@@ -182,7 +182,7 @@ void HELPER(simcall)(CPUXtensaState *env)
write(fd, buf, io_sz) :
read(fd, buf, io_sz);
regs[3] = errno_h2g(errno);
- cpu_physical_memory_unmap(buf, sz, is_write, sz);
+ cpu_physical_memory_unmap(buf, sz, !is_write, sz);
if (regs[2] == -1) {
break;
}
--
2.7.4
Michael Roth
2017-08-29 00:13:56 UTC
Permalink
From: Stefan Hajnoczi <***@redhat.com>

The main loop uses aio_disable_external()/aio_enable_external() to
temporarily disable processing of external AioContext clients like
device emulation.

This allows monitor commands to quiesce I/O and prevent the guest from
submitting new requests while a monitor command is in progress.

The aio_enable_external() API is currently broken when an IOThread is in
aio_poll() waiting for fd activity when the main loop re-enables
external clients. Incrementing ctx->external_disable_cnt does not wake
the IOThread from ppoll(2) so fd processing remains suspended and leads
to unresponsive emulated devices.

This patch adds an aio_notify() call to aio_enable_external() so the
IOThread is kicked out of ppoll(2) and will re-arm the file descriptors.

The bug can be reproduced as follows:

$ qemu -M accel=kvm -m 1024 \
-object iothread,id=iothread0 \
-device virtio-scsi-pci,iothread=iothread0,id=virtio-scsi-pci0 \
-drive if=none,id=drive0,aio=native,cache=none,format=raw,file=test.img \
-device scsi-hd,id=scsi-hd0,drive=drive0 \
-qmp tcp::5555,server,nowait

$ scripts/qmp/qmp-shell localhost:5555
(qemu) blockdev-snapshot-sync device=drive0 snapshot-file=sn1.qcow2
mode=absolute-paths format=qcow2

After blockdev-snapshot-sync completes the SCSI disk will be
unresponsive. This leads to request timeouts inside the guest.

Reported-by: Qianqian Zhu <***@redhat.com>
Reviewed-by: Fam Zheng <***@redhat.com>
Signed-off-by: Stefan Hajnoczi <***@redhat.com>
Message-id: 20170508180705.20609-1-***@redhat.com
Suggested-by: Fam Zheng <***@redhat.com>
Signed-off-by: Stefan Hajnoczi <***@redhat.com>
(cherry picked from commit 321d1dba8bef9676a77e9399484e3cd8bf2cf16a)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
include/block/aio.h | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/include/block/aio.h b/include/block/aio.h
index 406e323..e9aeeae 100644
--- a/include/block/aio.h
+++ b/include/block/aio.h
@@ -454,8 +454,14 @@ static inline void aio_disable_external(AioContext *ctx)
*/
static inline void aio_enable_external(AioContext *ctx)
{
- assert(ctx->external_disable_cnt > 0);
- atomic_dec(&ctx->external_disable_cnt);
+ int old;
+
+ old = atomic_fetch_dec(&ctx->external_disable_cnt);
+ assert(old > 0);
+ if (old == 1) {
+ /* Kick event loop so it re-arms file descriptors */
+ aio_notify(ctx);
+ }
}

/**
--
2.7.4
Michael Roth
2017-08-29 00:14:12 UTC
Permalink
From: Yunjian Wang <***@huawei.com>

The tx_bh or tx_timer will free in virtio_net_del_queue() function, when
removing virtio-net queues if the guest doesn't support multiqueue. But
it might be still referenced by virtio_net_set_status(), which needs to
be set NULL. And also the tx_waiting needs to be set zero to prevent
virtio_net_set_status() accessing tx_bh or tx_timer.

Cc: qemu-***@nongnu.org
Signed-off-by: Yunjian Wang <***@huawei.com>
Signed-off-by: Jason Wang <***@redhat.com>
(cherry picked from commit f989c30cf834ba8625e98b808eac30e4e7ec5008)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/net/virtio-net.c | 3 +++
1 file changed, 3 insertions(+)

diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 7d091c9..98bd683 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -1522,9 +1522,12 @@ static void virtio_net_del_queue(VirtIONet *n, int index)
if (q->tx_timer) {
timer_del(q->tx_timer);
timer_free(q->tx_timer);
+ q->tx_timer = NULL;
} else {
qemu_bh_delete(q->tx_bh);
+ q->tx_bh = NULL;
}
+ q->tx_waiting = 0;
virtio_del_queue(vdev, index * 2 + 1);
}
--
2.7.4
Michael Roth
2017-08-29 00:13:59 UTC
Permalink
From: Max Reitz <***@redhat.com>

For one thing, this allows us to drop the error message generation from
qemu-img.c and blockdev.c and instead have it unified in
bdrv_truncate().

Signed-off-by: Max Reitz <***@redhat.com>
Message-id: 20170328205129.15138-3-***@redhat.com
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit ed3d2ec98a33fbdeabc471b11ff807075f07e996)
* prereq for 698bdfa
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block.c | 16 ++++++++++++----
block/blkdebug.c | 2 +-
block/block-backend.c | 5 +++--
block/commit.c | 5 +++--
block/crypto.c | 2 +-
block/mirror.c | 2 +-
block/parallels.c | 13 ++++++++-----
block/qcow.c | 6 +++---
block/qcow2-refcount.c | 5 ++++-
block/qcow2.c | 14 +++++++++-----
block/qed.c | 2 +-
block/raw-format.c | 2 +-
block/vdi.c | 4 ++--
block/vhdx-log.c | 2 +-
block/vhdx.c | 6 +++---
block/vmdk.c | 13 +++----------
block/vpc.c | 13 +++++++------
blockdev.c | 21 +--------------------
include/block/block.h | 2 +-
include/sysemu/block-backend.h | 2 +-
qemu-img.c | 17 ++++-------------
qemu-io-cmds.c | 5 +++--
22 files changed, 73 insertions(+), 86 deletions(-)

diff --git a/block.c b/block.c
index 513e6e5..21cb65e 100644
--- a/block.c
+++ b/block.c
@@ -3262,7 +3262,7 @@ exit:
/**
* Truncate file to 'offset' bytes (needed only for file protocols)
*/
-int bdrv_truncate(BdrvChild *child, int64_t offset)
+int bdrv_truncate(BdrvChild *child, int64_t offset, Error **errp)
{
BlockDriverState *bs = child->bs;
BlockDriver *drv = bs->drv;
@@ -3274,12 +3274,18 @@ int bdrv_truncate(BdrvChild *child, int64_t offset)
* cannot assert this permission in that case. */
// assert(child->perm & BLK_PERM_RESIZE);

- if (!drv)
+ if (!drv) {
+ error_setg(errp, "No medium inserted");
return -ENOMEDIUM;
- if (!drv->bdrv_truncate)
+ }
+ if (!drv->bdrv_truncate) {
+ error_setg(errp, "Image format driver does not support resize");
return -ENOTSUP;
- if (bs->read_only)
+ }
+ if (bs->read_only) {
+ error_setg(errp, "Image is read-only");
return -EACCES;
+ }

ret = drv->bdrv_truncate(bs, offset);
if (ret == 0) {
@@ -3287,6 +3293,8 @@ int bdrv_truncate(BdrvChild *child, int64_t offset)
bdrv_dirty_bitmap_truncate(bs);
bdrv_parent_cb_resize(bs);
++bs->write_gen;
+ } else {
+ error_setg_errno(errp, -ret, "Failed to resize image");
}
return ret;
}
diff --git a/block/blkdebug.c b/block/blkdebug.c
index ffc6f1d..46bd08a 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -661,7 +661,7 @@ static int64_t blkdebug_getlength(BlockDriverState *bs)

static int blkdebug_truncate(BlockDriverState *bs, int64_t offset)
{
- return bdrv_truncate(bs->file, offset);
+ return bdrv_truncate(bs->file, offset, NULL);
}

static void blkdebug_refresh_filename(BlockDriverState *bs, QDict *options)
diff --git a/block/block-backend.c b/block/block-backend.c
index 7405024..1e70c4d 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -1746,13 +1746,14 @@ int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf,
BDRV_REQ_WRITE_COMPRESSED);
}

-int blk_truncate(BlockBackend *blk, int64_t offset)
+int blk_truncate(BlockBackend *blk, int64_t offset, Error **errp)
{
if (!blk_is_available(blk)) {
+ error_setg(errp, "No medium inserted");
return -ENOMEDIUM;
}

- return bdrv_truncate(blk->root, offset);
+ return bdrv_truncate(blk->root, offset, errp);
}

static void blk_pdiscard_entry(void *opaque)
diff --git a/block/commit.c b/block/commit.c
index 91d2c34..76a0d98 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -151,7 +151,7 @@ static void coroutine_fn commit_run(void *opaque)
}

if (base_len < s->common.len) {
- ret = blk_truncate(s->base, s->common.len);
+ ret = blk_truncate(s->base, s->common.len, NULL);
if (ret) {
goto out;
}
@@ -511,8 +511,9 @@ int bdrv_commit(BlockDriverState *bs)
* grow the backing file image if possible. If not possible,
* we must return an error */
if (length > backing_length) {
- ret = blk_truncate(backing, length);
+ ret = blk_truncate(backing, length, &local_err);
if (ret < 0) {
+ error_report_err(local_err);
goto ro_cleanup;
}
}
diff --git a/block/crypto.c b/block/crypto.c
index 4a20388..52e4f2b 100644
--- a/block/crypto.c
+++ b/block/crypto.c
@@ -389,7 +389,7 @@ static int block_crypto_truncate(BlockDriverState *bs, int64_t offset)

offset += payload_offset;

- return bdrv_truncate(bs->file, offset);
+ return bdrv_truncate(bs->file, offset, NULL);
}

static void block_crypto_close(BlockDriverState *bs)
diff --git a/block/mirror.c b/block/mirror.c
index 164438f..2173a2f 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -724,7 +724,7 @@ static void coroutine_fn mirror_run(void *opaque)
}

if (s->bdev_length > base_length) {
- ret = blk_truncate(s->target, s->bdev_length);
+ ret = blk_truncate(s->target, s->bdev_length, NULL);
if (ret < 0) {
goto immediate_exit;
}
diff --git a/block/parallels.c b/block/parallels.c
index 90acf79..8be46a7 100644
--- a/block/parallels.c
+++ b/block/parallels.c
@@ -223,7 +223,8 @@ static int64_t allocate_clusters(BlockDriverState *bs, int64_t sector_num,
space << BDRV_SECTOR_BITS, 0);
} else {
ret = bdrv_truncate(bs->file,
- (s->data_end + space) << BDRV_SECTOR_BITS);
+ (s->data_end + space) << BDRV_SECTOR_BITS,
+ NULL);
}
if (ret < 0) {
return ret;
@@ -456,8 +457,10 @@ static int parallels_check(BlockDriverState *bs, BdrvCheckResult *res,
size - res->image_end_offset);
res->leaks += count;
if (fix & BDRV_FIX_LEAKS) {
- ret = bdrv_truncate(bs->file, res->image_end_offset);
+ Error *local_err = NULL;
+ ret = bdrv_truncate(bs->file, res->image_end_offset, &local_err);
if (ret < 0) {
+ error_report_err(local_err);
res->check_errors++;
return ret;
}
@@ -504,7 +507,7 @@ static int parallels_create(const char *filename, QemuOpts *opts, Error **errp)

blk_set_allow_write_beyond_eof(file, true);

- ret = blk_truncate(file, 0);
+ ret = blk_truncate(file, 0, errp);
if (ret < 0) {
goto exit;
}
@@ -696,7 +699,7 @@ static int parallels_open(BlockDriverState *bs, QDict *options, int flags,
}

if (!(flags & BDRV_O_RESIZE) || !bdrv_has_zero_init(bs->file->bs) ||
- bdrv_truncate(bs->file, bdrv_getlength(bs->file->bs)) != 0) {
+ bdrv_truncate(bs->file, bdrv_getlength(bs->file->bs), NULL) != 0) {
s->prealloc_mode = PRL_PREALLOC_MODE_FALLOCATE;
}

@@ -739,7 +742,7 @@ static void parallels_close(BlockDriverState *bs)
}

if (bs->open_flags & BDRV_O_RDWR) {
- bdrv_truncate(bs->file, s->data_end << BDRV_SECTOR_BITS);
+ bdrv_truncate(bs->file, s->data_end << BDRV_SECTOR_BITS, NULL);
}

g_free(s->bat_dirty_bmap);
diff --git a/block/qcow.c b/block/qcow.c
index 9d6ac83..5d147b9 100644
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -473,7 +473,7 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
/* round to cluster size */
cluster_offset = (cluster_offset + s->cluster_size - 1) &
~(s->cluster_size - 1);
- bdrv_truncate(bs->file, cluster_offset + s->cluster_size);
+ bdrv_truncate(bs->file, cluster_offset + s->cluster_size, NULL);
/* if encrypted, we must initialize the cluster
content which won't be written */
if (bs->encrypted &&
@@ -833,7 +833,7 @@ static int qcow_create(const char *filename, QemuOpts *opts, Error **errp)

blk_set_allow_write_beyond_eof(qcow_blk, true);

- ret = blk_truncate(qcow_blk, 0);
+ ret = blk_truncate(qcow_blk, 0, errp);
if (ret < 0) {
goto exit;
}
@@ -916,7 +916,7 @@ static int qcow_make_empty(BlockDriverState *bs)
if (bdrv_pwrite_sync(bs->file, s->l1_table_offset, s->l1_table,
l1_length) < 0)
return -1;
- ret = bdrv_truncate(bs->file, s->l1_table_offset + l1_length);
+ ret = bdrv_truncate(bs->file, s->l1_table_offset + l1_length, NULL);
if (ret < 0)
return ret;

diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 9e96f64..4efca7e 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1728,14 +1728,17 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,

if (fix & BDRV_FIX_ERRORS) {
int64_t new_nb_clusters;
+ Error *local_err = NULL;

if (offset > INT64_MAX - s->cluster_size) {
ret = -EINVAL;
goto resize_fail;
}

- ret = bdrv_truncate(bs->file, offset + s->cluster_size);
+ ret = bdrv_truncate(bs->file, offset + s->cluster_size,
+ &local_err);
if (ret < 0) {
+ error_report_err(local_err);
goto resize_fail;
}
size = bdrv_getlength(bs->file->bs);
diff --git a/block/qcow2.c b/block/qcow2.c
index 49c737f..0d7e9e4 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2294,9 +2294,9 @@ static int qcow2_create2(const char *filename, int64_t total_size,
}

/* Okay, now that we have a valid image, let's give it the right size */
- ret = blk_truncate(blk, total_size);
+ ret = blk_truncate(blk, total_size, errp);
if (ret < 0) {
- error_setg_errno(errp, -ret, "Could not resize image");
+ error_prepend(errp, "Could not resize image: ");
goto out;
}

@@ -2584,7 +2584,7 @@ qcow2_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
/* align end of file to a sector boundary to ease reading with
sector based I/Os */
cluster_offset = bdrv_getlength(bs->file->bs);
- return bdrv_truncate(bs->file, cluster_offset);
+ return bdrv_truncate(bs->file, cluster_offset, NULL);
}

buf = qemu_blockalign(bs, s->cluster_size);
@@ -2674,6 +2674,7 @@ fail:
static int make_completely_empty(BlockDriverState *bs)
{
BDRVQcow2State *s = bs->opaque;
+ Error *local_err = NULL;
int ret, l1_clusters;
int64_t offset;
uint64_t *new_reftable = NULL;
@@ -2798,8 +2799,10 @@ static int make_completely_empty(BlockDriverState *bs)
goto fail;
}

- ret = bdrv_truncate(bs->file, (3 + l1_clusters) * s->cluster_size);
+ ret = bdrv_truncate(bs->file, (3 + l1_clusters) * s->cluster_size,
+ &local_err);
if (ret < 0) {
+ error_report_err(local_err);
goto fail;
}

@@ -3273,9 +3276,10 @@ static int qcow2_amend_options(BlockDriverState *bs, QemuOpts *opts,
return ret;
}

- ret = blk_truncate(blk, new_size);
+ ret = blk_truncate(blk, new_size, &local_err);
blk_unref(blk);
if (ret < 0) {
+ error_report_err(local_err);
return ret;
}
}
diff --git a/block/qed.c b/block/qed.c
index 5ec7fd8..53199ac 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -635,7 +635,7 @@ static int qed_create(const char *filename, uint32_t cluster_size,
blk_set_allow_write_beyond_eof(blk, true);

/* File must start empty and grow, check truncate is supported */
- ret = blk_truncate(blk, 0);
+ ret = blk_truncate(blk, 0, errp);
if (ret < 0) {
goto out;
}
diff --git a/block/raw-format.c b/block/raw-format.c
index 86fbc65..a800733 100644
--- a/block/raw-format.c
+++ b/block/raw-format.c
@@ -341,7 +341,7 @@ static int raw_truncate(BlockDriverState *bs, int64_t offset)

s->size = offset;
offset += s->offset;
- return bdrv_truncate(bs->file, offset);
+ return bdrv_truncate(bs->file, offset, NULL);
}

static int raw_media_changed(BlockDriverState *bs)
diff --git a/block/vdi.c b/block/vdi.c
index 9b4f70e..d12d9cd 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -832,9 +832,9 @@ static int vdi_create(const char *filename, QemuOpts *opts, Error **errp)
}

if (image_type == VDI_TYPE_STATIC) {
- ret = blk_truncate(blk, offset + blocks * block_size);
+ ret = blk_truncate(blk, offset + blocks * block_size, errp);
if (ret < 0) {
- error_setg(errp, "Failed to statically allocate %s", filename);
+ error_prepend(errp, "Failed to statically allocate %s", filename);
goto exit;
}
}
diff --git a/block/vhdx-log.c b/block/vhdx-log.c
index 67a91c0..3f4c2aa 100644
--- a/block/vhdx-log.c
+++ b/block/vhdx-log.c
@@ -548,7 +548,7 @@ static int vhdx_log_flush(BlockDriverState *bs, BDRVVHDXState *s,
if (new_file_size % (1024*1024)) {
/* round up to nearest 1MB boundary */
new_file_size = ((new_file_size >> 20) + 1) << 20;
- bdrv_truncate(bs->file, new_file_size);
+ bdrv_truncate(bs->file, new_file_size, NULL);
}
}
qemu_vfree(desc_entries);
diff --git a/block/vhdx.c b/block/vhdx.c
index d25bcd9..0a38f77 100644
--- a/block/vhdx.c
+++ b/block/vhdx.c
@@ -1171,7 +1171,7 @@ static int vhdx_allocate_block(BlockDriverState *bs, BDRVVHDXState *s,
/* per the spec, the address for a block is in units of 1MB */
*new_offset = ROUND_UP(*new_offset, 1024 * 1024);

- return bdrv_truncate(bs->file, *new_offset + s->block_size);
+ return bdrv_truncate(bs->file, *new_offset + s->block_size, NULL);
}

/*
@@ -1607,14 +1607,14 @@ static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s,
if (type == VHDX_TYPE_DYNAMIC) {
/* All zeroes, so we can just extend the file - the end of the BAT
* is the furthest thing we have written yet */
- ret = blk_truncate(blk, data_file_offset);
+ ret = blk_truncate(blk, data_file_offset, errp);
if (ret < 0) {
error_setg_errno(errp, -ret,
"Failed to resize the underlying file");
goto exit;
}
} else if (type == VHDX_TYPE_FIXED) {
- ret = blk_truncate(blk, data_file_offset + image_size);
+ ret = blk_truncate(blk, data_file_offset + image_size, errp);
if (ret < 0) {
error_setg_errno(errp, -ret,
"Failed to resize the underlying file");
diff --git a/block/vmdk.c b/block/vmdk.c
index a9bd22b..c61b9cc 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -1714,10 +1714,7 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
blk_set_allow_write_beyond_eof(blk, true);

if (flat) {
- ret = blk_truncate(blk, filesize);
- if (ret < 0) {
- error_setg_errno(errp, -ret, "Could not truncate file");
- }
+ ret = blk_truncate(blk, filesize, errp);
goto exit;
}
magic = cpu_to_be32(VMDK4_MAGIC);
@@ -1780,9 +1777,8 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
goto exit;
}

- ret = blk_truncate(blk, le64_to_cpu(header.grain_offset) << 9);
+ ret = blk_truncate(blk, le64_to_cpu(header.grain_offset) << 9, errp);
if (ret < 0) {
- error_setg_errno(errp, -ret, "Could not truncate file");
goto exit;
}

@@ -2090,10 +2086,7 @@ static int vmdk_create(const char *filename, QemuOpts *opts, Error **errp)
/* bdrv_pwrite write padding zeros to align to sector, we don't need that
* for description file */
if (desc_offset == 0) {
- ret = blk_truncate(new_blk, desc_len);
- if (ret < 0) {
- error_setg_errno(errp, -ret, "Could not truncate file");
- }
+ ret = blk_truncate(new_blk, desc_len, errp);
}
exit:
if (new_blk) {
diff --git a/block/vpc.c b/block/vpc.c
index f591d4b..ecfee77 100644
--- a/block/vpc.c
+++ b/block/vpc.c
@@ -851,20 +851,21 @@ static int create_dynamic_disk(BlockBackend *blk, uint8_t *buf,
}

static int create_fixed_disk(BlockBackend *blk, uint8_t *buf,
- int64_t total_size)
+ int64_t total_size, Error **errp)
{
int ret;

/* Add footer to total size */
total_size += HEADER_SIZE;

- ret = blk_truncate(blk, total_size);
+ ret = blk_truncate(blk, total_size, errp);
if (ret < 0) {
return ret;
}

ret = blk_pwrite(blk, total_size - HEADER_SIZE, buf, HEADER_SIZE, 0);
if (ret < 0) {
+ error_setg_errno(errp, -ret, "Unable to write VHD header");
return ret;
}

@@ -996,11 +997,11 @@ static int vpc_create(const char *filename, QemuOpts *opts, Error **errp)

if (disk_type == VHD_DYNAMIC) {
ret = create_dynamic_disk(blk, buf, total_sectors);
+ if (ret < 0) {
+ error_setg(errp, "Unable to create or write VHD header");
+ }
} else {
- ret = create_fixed_disk(blk, buf, total_size);
- }
- if (ret < 0) {
- error_setg(errp, "Unable to create or write VHD header");
+ ret = create_fixed_disk(blk, buf, total_size, errp);
}

out:
diff --git a/blockdev.c b/blockdev.c
index 841200e..78e1204 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2926,26 +2926,7 @@ void qmp_block_resize(bool has_device, const char *device,
/* complete all in-flight operations before resizing the device */
bdrv_drain_all();

- ret = blk_truncate(blk, size);
- switch (ret) {
- case 0:
- break;
- case -ENOMEDIUM:
- error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
- break;
- case -ENOTSUP:
- error_setg(errp, QERR_UNSUPPORTED);
- break;
- case -EACCES:
- error_setg(errp, "Device '%s' is read only", device);
- break;
- case -EBUSY:
- error_setg(errp, QERR_DEVICE_IN_USE, device);
- break;
- default:
- error_setg_errno(errp, -ret, "Could not resize");
- break;
- }
+ ret = blk_truncate(blk, size, errp);

out:
blk_unref(blk);
diff --git a/include/block/block.h b/include/block/block.h
index 5ddc0cf..4bf4843 100644
--- a/include/block/block.h
+++ b/include/block/block.h
@@ -294,7 +294,7 @@ BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
const char *backing_file);
int bdrv_get_backing_file_depth(BlockDriverState *bs);
void bdrv_refresh_filename(BlockDriverState *bs);
-int bdrv_truncate(BdrvChild *child, int64_t offset);
+int bdrv_truncate(BdrvChild *child, int64_t offset, Error **errp);
int64_t bdrv_nb_sectors(BlockDriverState *bs);
int64_t bdrv_getlength(BlockDriverState *bs);
int64_t bdrv_get_allocated_file_size(BlockDriverState *bs);
diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h
index 7462228..0ba4e27 100644
--- a/include/sysemu/block-backend.h
+++ b/include/sysemu/block-backend.h
@@ -225,7 +225,7 @@ int coroutine_fn blk_co_pwrite_zeroes(BlockBackend *blk, int64_t offset,
int count, BdrvRequestFlags flags);
int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf,
int count);
-int blk_truncate(BlockBackend *blk, int64_t offset);
+int blk_truncate(BlockBackend *blk, int64_t offset, Error **errp);
int blk_pdiscard(BlockBackend *blk, int64_t offset, int count);
int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,
int64_t pos, int size);
diff --git a/qemu-img.c b/qemu-img.c
index 26ded22..e4a2686 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -3500,20 +3500,11 @@ static int img_resize(int argc, char **argv)
goto out;
}

- ret = blk_truncate(blk, total_size);
- switch (ret) {
- case 0:
+ ret = blk_truncate(blk, total_size, &err);
+ if (!ret) {
qprintf(quiet, "Image resized.\n");
- break;
- case -ENOTSUP:
- error_report("This image does not support resize");
- break;
- case -EACCES:
- error_report("Image is read-only");
- break;
- default:
- error_report("Error resizing image: %s", strerror(-ret));
- break;
+ } else {
+ error_report_err(err);
}
out:
blk_unref(blk);
diff --git a/qemu-io-cmds.c b/qemu-io-cmds.c
index 312fc6d..21af9e6 100644
--- a/qemu-io-cmds.c
+++ b/qemu-io-cmds.c
@@ -1567,6 +1567,7 @@ static const cmdinfo_t flush_cmd = {

static int truncate_f(BlockBackend *blk, int argc, char **argv)
{
+ Error *local_err = NULL;
int64_t offset;
int ret;

@@ -1576,9 +1577,9 @@ static int truncate_f(BlockBackend *blk, int argc, char **argv)
return 0;
}

- ret = blk_truncate(blk, offset);
+ ret = blk_truncate(blk, offset, &local_err);
if (ret < 0) {
- printf("truncate: %s\n", strerror(-ret));
+ error_report_err(local_err);
return 0;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:00 UTC
Permalink
From: John Snow <***@redhat.com>

Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1447551

If one tries to issue a block_resize while a guest is busy
accessing the disk, it is possible that qemu may deadlock
when invoking aio_poll from both the main loop and the iothread.

Replace another instance of bdrv_drain_all that doesn't
quite belong.

Cc: qemu-***@nongnu.org
Suggested-by: Paolo Bonzini <***@redhat.com>
Signed-off-by: John Snow <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Paolo Bonzini <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 698bdfa07d66b5ec218a60229e58eae1dcde00e5)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
blockdev.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/blockdev.c b/blockdev.c
index 78e1204..e8a9a65 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2923,10 +2923,9 @@ void qmp_block_resize(bool has_device, const char *device,
goto out;
}

- /* complete all in-flight operations before resizing the device */
- bdrv_drain_all();
-
+ bdrv_drained_begin(bs);
ret = blk_truncate(blk, size, errp);
+ bdrv_drained_end(bs);

out:
blk_unref(blk);
--
2.7.4
Michael Roth
2017-08-29 00:14:19 UTC
Permalink
From: Eric Blake <***@redhat.com>

Since we are already in coroutine context during the body of
bdrv_co_get_block_status(), we can shave off a few layers of
wrappers when recursing to query the protocol when a format driver
returned BDRV_BLOCK_RAW.

Note that we are already using the correct recursion later on in
the same function, when probing whether the protocol layer is sparse
in order to find out if we can add BDRV_BLOCK_ZERO to an existing
BDRV_BLOCK_DATA|BDRV_BLOCK_OFFSET_VALID.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Reviewed-by: Fam Zheng <***@redhat.com>
Message-id: 20170504173745.27414-1-***@redhat.com
Signed-off-by: Stefan Hajnoczi <***@redhat.com>
(cherry picked from commit ee29d6adefcca7e76abb124183814ed3acc74fac)
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/io.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/block/io.c b/block/io.c
index a7142e0..fe0c867 100644
--- a/block/io.c
+++ b/block/io.c
@@ -1792,8 +1792,8 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,

if (ret & BDRV_BLOCK_RAW) {
assert(ret & BDRV_BLOCK_OFFSET_VALID);
- ret = bdrv_get_block_status(*file, ret >> BDRV_SECTOR_BITS,
- *pnum, pnum, file);
+ ret = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS,
+ *pnum, pnum, file);
goto out;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:17 UTC
Permalink
From: Eric Blake <***@redhat.com>

Make it easier to simulate various unusual hardware setups (for
example, recent commits 3482b9b and b8d0a98 affect the Dell
Equallogic iSCSI with its 15M preferred and maximum unmap and
write zero sizing, or b2f95fe deals with the Linux loopback
block device having a max_transfer of 64k), by allowing blkdebug
to wrap any other device with further restrictions on various
alignments.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170429191419.30051-9-***@redhat.com
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit 430b26a82da61876c4eaf559ae02332582968043)
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 96 +++++++++++++++++++++++++++++++++++++++++++++++++++-
qapi/block-core.json | 33 ++++++++++++++++--
2 files changed, 125 insertions(+), 4 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index 2b934ef..5ccb9ce 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -39,6 +39,11 @@ typedef struct BDRVBlkdebugState {
int state;
int new_state;
uint64_t align;
+ uint64_t max_transfer;
+ uint64_t opt_write_zero;
+ uint64_t max_write_zero;
+ uint64_t opt_discard;
+ uint64_t max_discard;

/* For blkdebug_refresh_filename() */
char *config_file;
@@ -343,6 +348,31 @@ static QemuOptsList runtime_opts = {
.type = QEMU_OPT_SIZE,
.help = "Required alignment in bytes",
},
+ {
+ .name = "max-transfer",
+ .type = QEMU_OPT_SIZE,
+ .help = "Maximum transfer size in bytes",
+ },
+ {
+ .name = "opt-write-zero",
+ .type = QEMU_OPT_SIZE,
+ .help = "Optimum write zero alignment in bytes",
+ },
+ {
+ .name = "max-write-zero",
+ .type = QEMU_OPT_SIZE,
+ .help = "Maximum write zero size in bytes",
+ },
+ {
+ .name = "opt-discard",
+ .type = QEMU_OPT_SIZE,
+ .help = "Optimum discard alignment in bytes",
+ },
+ {
+ .name = "max-discard",
+ .type = QEMU_OPT_SIZE,
+ .help = "Maximum discard size in bytes",
+ },
{ /* end of list */ }
},
};
@@ -354,6 +384,7 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
QemuOpts *opts;
Error *local_err = NULL;
int ret;
+ uint64_t align;

opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
qemu_opts_absorb_qdict(opts, options, &local_err);
@@ -388,13 +419,61 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
bs->file->bs->supported_zero_flags;
ret = -EINVAL;

- /* Set request alignment */
+ /* Set alignment overrides */
s->align = qemu_opt_get_size(opts, "align", 0);
if (s->align && (s->align >= INT_MAX || !is_power_of_2(s->align))) {
error_setg(errp, "Cannot meet constraints with align %" PRIu64,
s->align);
goto out;
}
+ align = MAX(s->align, bs->file->bs->bl.request_alignment);
+
+ s->max_transfer = qemu_opt_get_size(opts, "max-transfer", 0);
+ if (s->max_transfer &&
+ (s->max_transfer >= INT_MAX ||
+ !QEMU_IS_ALIGNED(s->max_transfer, align))) {
+ error_setg(errp, "Cannot meet constraints with max-transfer %" PRIu64,
+ s->max_transfer);
+ goto out;
+ }
+
+ s->opt_write_zero = qemu_opt_get_size(opts, "opt-write-zero", 0);
+ if (s->opt_write_zero &&
+ (s->opt_write_zero >= INT_MAX ||
+ !QEMU_IS_ALIGNED(s->opt_write_zero, align))) {
+ error_setg(errp, "Cannot meet constraints with opt-write-zero %" PRIu64,
+ s->opt_write_zero);
+ goto out;
+ }
+
+ s->max_write_zero = qemu_opt_get_size(opts, "max-write-zero", 0);
+ if (s->max_write_zero &&
+ (s->max_write_zero >= INT_MAX ||
+ !QEMU_IS_ALIGNED(s->max_write_zero,
+ MAX(s->opt_write_zero, align)))) {
+ error_setg(errp, "Cannot meet constraints with max-write-zero %" PRIu64,
+ s->max_write_zero);
+ goto out;
+ }
+
+ s->opt_discard = qemu_opt_get_size(opts, "opt-discard", 0);
+ if (s->opt_discard &&
+ (s->opt_discard >= INT_MAX ||
+ !QEMU_IS_ALIGNED(s->opt_discard, align))) {
+ error_setg(errp, "Cannot meet constraints with opt-discard %" PRIu64,
+ s->opt_discard);
+ goto out;
+ }
+
+ s->max_discard = qemu_opt_get_size(opts, "max-discard", 0);
+ if (s->max_discard &&
+ (s->max_discard >= INT_MAX ||
+ !QEMU_IS_ALIGNED(s->max_discard,
+ MAX(s->opt_discard, align)))) {
+ error_setg(errp, "Cannot meet constraints with max-discard %" PRIu64,
+ s->max_discard);
+ goto out;
+ }

ret = 0;
out:
@@ -789,6 +868,21 @@ static void blkdebug_refresh_limits(BlockDriverState *bs, Error **errp)
if (s->align) {
bs->bl.request_alignment = s->align;
}
+ if (s->max_transfer) {
+ bs->bl.max_transfer = s->max_transfer;
+ }
+ if (s->opt_write_zero) {
+ bs->bl.pwrite_zeroes_alignment = s->opt_write_zero;
+ }
+ if (s->max_write_zero) {
+ bs->bl.max_pwrite_zeroes = s->max_write_zero;
+ }
+ if (s->opt_discard) {
+ bs->bl.pdiscard_alignment = s->opt_discard;
+ }
+ if (s->max_discard) {
+ bs->bl.max_pdiscard = s->max_discard;
+ }
}

static int blkdebug_reopen_prepare(BDRVReopenState *reopen_state,
diff --git a/qapi/block-core.json b/qapi/block-core.json
index 033457c..38edada 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -2428,8 +2428,33 @@
#
# @config: filename of the configuration file
#
-# @align: required alignment for requests in bytes,
-# must be power of 2, or 0 for default
+# @align: required alignment for requests in bytes, must be
+# positive power of 2, or 0 for default
+#
+# @max-transfer: maximum size for I/O transfers in bytes, must be
+# positive multiple of @align and of the underlying
+# file's request alignment (but need not be a power of
+# 2), or 0 for default (since 2.10)
+#
+# @opt-write-zero: preferred alignment for write zero requests in bytes,
+# must be positive multiple of @align and of the
+# underlying file's request alignment (but need not be a
+# power of 2), or 0 for default (since 2.10)
+#
+# @max-write-zero: maximum size for write zero requests in bytes, must be
+# positive multiple of @align, of @opt-write-zero, and of
+# the underlying file's request alignment (but need not
+# be a power of 2), or 0 for default (since 2.10)
+#
+# @opt-discard: preferred alignment for discard requests in bytes, must
+# be positive multiple of @align and of the underlying
+# file's request alignment (but need not be a power of
+# 2), or 0 for default (since 2.10)
+#
+# @max-discard: maximum size for discard requests in bytes, must be
+# positive multiple of @align, of @opt-discard, and of
+# the underlying file's request alignment (but need not
+# be a power of 2), or 0 for default (since 2.10)
#
# @inject-error: array of error injection descriptions
#
@@ -2440,7 +2465,9 @@
{ 'struct': 'BlockdevOptionsBlkdebug',
'data': { 'image': 'BlockdevRef',
'*config': 'str',
- '*align': 'int',
+ '*align': 'int', '*max-transfer': 'int32',
+ '*opt-write-zero': 'int32', '*max-write-zero': 'int32',
+ '*opt-discard': 'int32', '*max-discard': 'int32',
'*inject-error': ['BlkdebugInjectErrorOptions'],
'*set-state': ['BlkdebugSetStateOptions'] } }
--
2.7.4
Michael Roth
2017-08-29 00:14:02 UTC
Permalink
From: Max Filippov <***@gmail.com>

Return value of read/write simcalls is not calculated correctly in case
of operations crossing page boundary and in case of short reads/writes.
Read and write simcalls should return the size of data actually
read/written or -1 in case of error.

Cc: qemu-***@nongnu.org
Signed-off-by: Max Filippov <***@gmail.com>
(cherry picked from commit 347ec03093f9668a379ef6b7fa1feb332fff039c)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
target/xtensa/xtensa-semi.c | 25 ++++++++++++++++++++-----
1 file changed, 20 insertions(+), 5 deletions(-)

diff --git a/target/xtensa/xtensa-semi.c b/target/xtensa/xtensa-semi.c
index 98ae28c..ffcaf8d 100644
--- a/target/xtensa/xtensa-semi.c
+++ b/target/xtensa/xtensa-semi.c
@@ -166,6 +166,7 @@ void HELPER(simcall)(CPUXtensaState *env)
uint32_t fd = regs[3];
uint32_t vaddr = regs[4];
uint32_t len = regs[5];
+ uint32_t len_done = 0;

while (len > 0) {
hwaddr paddr = cpu_get_phys_page_debug(cs, vaddr);
@@ -174,24 +175,38 @@ void HELPER(simcall)(CPUXtensaState *env)
uint32_t io_sz = page_left < len ? page_left : len;
hwaddr sz = io_sz;
void *buf = cpu_physical_memory_map(paddr, &sz, !is_write);
+ uint32_t io_done;
+ bool error = false;

if (buf) {
vaddr += io_sz;
len -= io_sz;
- regs[2] = is_write ?
+ io_done = is_write ?
write(fd, buf, io_sz) :
read(fd, buf, io_sz);
regs[3] = errno_h2g(errno);
- cpu_physical_memory_unmap(buf, sz, !is_write, sz);
- if (regs[2] == -1) {
- break;
+ if (io_done == -1) {
+ error = true;
+ io_done = 0;
}
+ cpu_physical_memory_unmap(buf, sz, !is_write, io_done);
} else {
- regs[2] = -1;
+ error = true;
regs[3] = TARGET_EINVAL;
break;
}
+ if (error) {
+ if (!len_done) {
+ len_done = -1;
+ }
+ break;
+ }
+ len_done += io_done;
+ if (io_done < io_sz) {
+ break;
+ }
}
+ regs[2] = len_done;
}
break;
--
2.7.4
Michael Roth
2017-08-29 00:14:04 UTC
Permalink
From: Paolo Bonzini <***@redhat.com>

All curl callbacks go through curl_multi_do, and hence are called with
s->mutex held. Note that with comments, and make curl_read_cb drop the
lock before invoking the callback.

Likewise for curl_find_buf, where the callback can be invoked by the
caller.

Cc: qemu-***@nongnu.org
Reviewed-by: Jeff Cody <***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170515100059.15795-3-***@redhat.com
Signed-off-by: Jeff Cody <***@redhat.com>
(cherry picked from commit 34db05e7ffe8d61ca7288b9532ad6e8300853318)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/curl.c | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/block/curl.c b/block/curl.c
index c6dc4c0..59e69c6 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -147,6 +147,7 @@ static void curl_multi_do(void *arg);
static void curl_multi_read(void *arg);

#ifdef NEED_CURL_TIMER_CALLBACK
+/* Called from curl_multi_do_locked, with s->mutex held. */
static int curl_timer_cb(CURLM *multi, long timeout_ms, void *opaque)
{
BDRVCURLState *s = opaque;
@@ -163,6 +164,7 @@ static int curl_timer_cb(CURLM *multi, long timeout_ms, void *opaque)
}
#endif

+/* Called from curl_multi_do_locked, with s->mutex held. */
static int curl_sock_cb(CURL *curl, curl_socket_t fd, int action,
void *userp, void *sp)
{
@@ -212,6 +214,7 @@ static int curl_sock_cb(CURL *curl, curl_socket_t fd, int action,
return 0;
}

+/* Called from curl_multi_do_locked, with s->mutex held. */
static size_t curl_header_cb(void *ptr, size_t size, size_t nmemb, void *opaque)
{
BDRVCURLState *s = opaque;
@@ -226,6 +229,7 @@ static size_t curl_header_cb(void *ptr, size_t size, size_t nmemb, void *opaque)
return realsize;
}

+/* Called from curl_multi_do_locked, with s->mutex held. */
static size_t curl_read_cb(void *ptr, size_t size, size_t nmemb, void *opaque)
{
CURLState *s = ((CURLState*)opaque);
@@ -264,7 +268,9 @@ static size_t curl_read_cb(void *ptr, size_t size, size_t nmemb, void *opaque)
request_length - offset);
}

+ qemu_mutex_unlock(&s->s->mutex);
acb->common.cb(acb->common.opaque, 0);
+ qemu_mutex_lock(&s->s->mutex);
qemu_aio_unref(acb);
s->acb[i] = NULL;
}
@@ -305,8 +311,6 @@ static int curl_find_buf(BDRVCURLState *s, size_t start, size_t len,
if (clamped_len < len) {
qemu_iovec_memset(acb->qiov, clamped_len, 0, len - clamped_len);
}
- acb->common.cb(acb->common.opaque, 0);
-
return FIND_RET_OK;
}

@@ -832,8 +836,8 @@ static void curl_readv_bh_cb(void *p)
// we can just call the callback and be done.
switch (curl_find_buf(s, start, acb->nb_sectors * BDRV_SECTOR_SIZE, acb)) {
case FIND_RET_OK:
- qemu_aio_unref(acb);
- // fall through
+ ret = 0;
+ goto out;
case FIND_RET_WAIT:
goto out;
default:
--
2.7.4
Michael Roth
2017-08-29 00:14:22 UTC
Permalink
From: Ladi Prosek <***@redhat.com>

Virtio serial device controls the lifetime of virtio-serial-bus and
virtio-serial-bus links back to the device via its hotplug-handler
property. This extra ref-count prevents the device from getting
finalized, leaving the VirtIODevice memory listener registered and
leading to use-after-free later on.

This patch addresses the same issue as Fam Zheng's
"virtio-scsi: Unset hotplug handler when unrealize"
only for a different virtio device.

Cc: qemu-***@nongnu.org
Signed-off-by: Ladi Prosek <***@redhat.com>
Reviewed-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Michael S. Tsirkin <***@redhat.com>
Reviewed-by: Paolo Bonzini <***@redhat.com>
Reviewed-by: Fam Zheng <***@redhat.com>
(cherry picked from commit f811f97040a48358b456b46ecbc9167f0131034f)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/char/virtio-serial-bus.c | 3 +++
1 file changed, 3 insertions(+)

diff --git a/hw/char/virtio-serial-bus.c b/hw/char/virtio-serial-bus.c
index d797a67..aa9c11a 100644
--- a/hw/char/virtio-serial-bus.c
+++ b/hw/char/virtio-serial-bus.c
@@ -1121,6 +1121,9 @@ static void virtio_serial_device_unrealize(DeviceState *dev, Error **errp)
timer_free(vser->post_load->timer);
g_free(vser->post_load);
}
+
+ qbus_set_hotplug_handler(BUS(&vser->bus), NULL, errp);
+
virtio_cleanup(vdev);
}
--
2.7.4
Michael Roth
2017-08-29 00:14:16 UTC
Permalink
From: Eric Blake <***@redhat.com>

Rather than store into a local variable, then copy to the struct
if the value is valid, then reporting errors otherwise, it is
simpler to just store into the struct and report errors if the
value is invalid. This however requires that the struct store
a 64-bit number, rather than a narrower type. Likewise, setting
a sane errno value in ret prior to the sequence of parsing and
jumping to out: on error makes it easier for the next patch to
add a chain of similar checks.

Signed-off-by: Eric Blake <***@redhat.com>
Message-id: 20170429191419.30051-8-***@redhat.com
Reviewed-by: Max Reitz <***@redhat.com>
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit 3dc834f8795a46f919d90b1e5369308628858601)
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 16 ++++++----------
1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index a3dc5f6..2b934ef 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -38,7 +38,7 @@
typedef struct BDRVBlkdebugState {
int state;
int new_state;
- int align;
+ uint64_t align;

/* For blkdebug_refresh_filename() */
char *config_file;
@@ -353,7 +353,6 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
BDRVBlkdebugState *s = bs->opaque;
QemuOpts *opts;
Error *local_err = NULL;
- uint64_t align;
int ret;

opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
@@ -387,20 +386,17 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
bs->file->bs->supported_write_flags;
bs->supported_zero_flags = (BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP) &
bs->file->bs->supported_zero_flags;
+ ret = -EINVAL;

/* Set request alignment */
- align = qemu_opt_get_size(opts, "align", 0);
- if (align < INT_MAX && is_power_of_2(align)) {
- s->align = align;
- } else if (align) {
- error_setg(errp, "Invalid alignment");
- ret = -EINVAL;
+ s->align = qemu_opt_get_size(opts, "align", 0);
+ if (s->align && (s->align >= INT_MAX || !is_power_of_2(s->align))) {
+ error_setg(errp, "Cannot meet constraints with align %" PRIu64,
+ s->align);
goto out;
}

ret = 0;
- goto out;
-
out:
if (ret < 0) {
g_free(s->config_file);
--
2.7.4
Michael Roth
2017-08-29 00:14:03 UTC
Permalink
From: Paolo Bonzini <***@redhat.com>

curl_clean_state should only be called after all AIOCBs have been
completed. This is not so obvious for the call from curl_detach_aio_context,
so assert that.

Cc: qemu-***@nongnu.org
Reviewed-by: Jeff Cody <***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170515100059.15795-2-***@redhat.com
Signed-off-by: Jeff Cody <***@redhat.com>
(cherry picked from commit 675a775633e68bf8b426a896fea5b93a4f4ff1cc)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/curl.c | 5 +++++
1 file changed, 5 insertions(+)

diff --git a/block/curl.c b/block/curl.c
index aa6e8cc..c6dc4c0 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -532,6 +532,11 @@ static CURLState *curl_init_state(BlockDriverState *bs, BDRVCURLState *s)

static void curl_clean_state(CURLState *s)
{
+ int j;
+ for (j = 0; j < CURL_NUM_ACB; j++) {
+ assert(!s->acb[j]);
+ }
+
if (s->s->multi)
curl_multi_remove_handle(s->s->multi, s->curl);
--
2.7.4
Michael Roth
2017-08-29 00:14:06 UTC
Permalink
From: Alberto Garcia <***@igalia.com>

The code that tries to reopen a BlockDriverState in stream_start()
when the creation of a new block job fails crashes because it attempts
to dereference a pointer that is known to be NULL.

This is a regression introduced in a170a91fd3eab6155da39e740381867e,
likely because the code was copied from stream_complete().

Cc: qemu-***@nongnu.org
Reported-by: Kashyap Chamarthy <***@redhat.com>
Signed-off-by: Alberto Garcia <***@igalia.com>
Tested-by: Kashyap Chamarthy <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 525989a50a70ea0ffa2b1cdf56279765bb2b7de0)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/stream.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block/stream.c b/block/stream.c
index 0113710..52d329f 100644
--- a/block/stream.c
+++ b/block/stream.c
@@ -280,6 +280,6 @@ void stream_start(const char *job_id, BlockDriverState *bs,

fail:
if (orig_bs_flags != bdrv_get_flags(bs)) {
- bdrv_reopen(bs, s->bs_flags, NULL);
+ bdrv_reopen(bs, orig_bs_flags, NULL);
}
}
--
2.7.4
Michael Roth
2017-08-29 00:14:23 UTC
Permalink
From: Paolo Bonzini <***@redhat.com>

The ROM uses the cmovne instruction, which is new in Pentium Pro and does not
work when running QEMU with "-cpu 486". Avoid producing that instruction.

Suggested-by: Richard W.M. Jones <***@redhat.com>
Suggested-by: Thomas Huth <***@redhat.com>
Reported-by: Rob Landley <***@landley.net>
Cc: qemu-***@nongnu.org
Signed-off-by: Paolo Bonzini <***@redhat.com>
(cherry picked from commit 7e018385103cd7a571b9ea0d6f994af6b1129fe7)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
pc-bios/linuxboot_dma.bin | Bin 1536 -> 1536 bytes
pc-bios/optionrom/Makefile | 1 +
2 files changed, 1 insertion(+)

diff --git a/pc-bios/linuxboot_dma.bin b/pc-bios/linuxboot_dma.bin
index 218d3ab4a29bfb5ab7125ec7a4d29dad1860c673..d176f62797813e4b926dca9dfce7ce554dc1a4d6 100644
GIT binary patch
literal 1536
zcmeHFL2J`s82-{_T52KL%w~}udiW-YBE*9t=s~ClrIm^s9^6H6Pf%onr07*+kWd=L
zpW)d<4G1ZESU1FyUDSCf^YYaxioy<}t?!rmu^#*f?mHxJo;-Qp_kHp#Je^o$|2##;
zYs)n)qh6k7XI07O)9TW>>***@3WSgzkI-MW44`qG_Jxhgq7fAt@+rpuZy7{g_FpG^eo
zuoD<7Fjn!zK~***@Fj1CnQuA}{tXz$^z=7>a?#S!%6CsARNbiO)h=v+Q~***@5O1
ze8y7!akTy&-***@p3Es`0q*km+)Mtxe(3DE2)a5;c%$H0|Yu~mbnR0C7dKHR7Zsr
zmrccq5lm%U=wXc39<t?-(_m*ZnY&YH(d(0lj;0`jj8e`iWgX(jV5j3gg<B9Cgj2^d
zw;W8A$%***@ijhITTF%3U<ui*Zsgx!lc)pgIq-atzea+i1H+3Oj<V?5-VpBVlz4R#!?
zV3N_sX!mjsj*#|9H8ozR&D5ajsurT@(W<h&xG*;8UNvYI;`***@m7;AnvgFCdyy
zI>gV6@%Jl!9^%vQ9V0A-7a<bD(|L-BW|R)`qY*CS(3SQRXOVJqcpcuB8NLnUuosH3
zi{OJ7mufeeG{^49xY<M4U)CrSC7Fp}TjtSE>UkNnpxu8npp2ruY0}SsKn=BiGN9Y;
L*>&K*abWuoMsO(*

literal 1536
zcmeHF&ubGw6rN4Dn#RCxYN|nUh%gpJFME=L9)#^}1F5!ps0f07iy|#(O1<iu!xFNH
ze}?~naY4vZ57wZB1jMR`ST7r__~R&=_04QYJoq2jIn2&}@6GqV_sv^+G|^@MTxG^<
zD>W^nf~VP8b!***@l;?V$SuIwL6uYt5>dGyH&BPBIn1?(K>***@S?eidI1hgSZ^wky
zNA%***@f^krJcQTH!Ptb+b>Z7QOZy_v)9!Gc_A*CUnxyOgP~7<9qN#In@`mGjfw=z$
z*1p4gW?~r|cz_FdqLT)y8y*s0Z-)z#_;!*cE?PD>cN4<HA{;MET(E~F0*Ke4l-T|Q
zN3kAs(B4Nz%(_i>CX-n&AUsLGXj9UuS(2NP%(`OENoGw|EGqnf#7ASHq-k?KrExJ~
z^CHnSr*VvZO%mMQmY+taei4m+{T$|c^(w4)XlpT*^|#=^eK`aF_0%gE_5j9w)aV=c
zk7$UtO_Fkt>S!)b5N****@orPY(na#Hj4<S)0RarUXZ%Tp{&xoU+SGxcuC<|i62e-
zaCiw}LGfe}PpSO8%0DRczx~>MfwIDj65o<I`***@FBlkzycj5td1A}41qQPA(<_yF`A
***@IINro1ha6gE;8HfA4iFwHr(rWNnH$vl3XJyEGJ>hT<|dXkK7E?qHJ9XdoKqMjysh
NXm9N~@Si)N{RVuA8xQ~h

diff --git a/pc-bios/optionrom/Makefile b/pc-bios/optionrom/Makefile
index fa53d9e..a9a9e5e 100644
--- a/pc-bios/optionrom/Makefile
+++ b/pc-bios/optionrom/Makefile
@@ -13,6 +13,7 @@ $(call set-vpath, $(SRC_PATH)/pc-bios/optionrom)
ifeq ($(lastword $(filter -O%, -O0 $(CFLAGS))),-O0)
override CFLAGS += -O2
endif
+override CFLAGS += -march=i486

# Drop -fstack-protector and the like
QEMU_CFLAGS := $(filter -W%, $(QEMU_CFLAGS)) $(CFLAGS_NOPIE) -ffreestanding
--
2.7.4
Michael Roth
2017-08-29 00:14:21 UTC
Permalink
From: Kevin Wolf <***@redhat.com>

This fixes an assertion failure that was triggered by qemu-iotests 129
on some CI host, while the same test case didn't seem to fail on other
hosts.

Essentially the problem is that the blk_unref(s->target) in
mirror_exit() doesn't necessarily mean that the BlockBackend goes away
immediately. It is possible that the job completion was triggered nested
in mirror_drain(), which looks like this:

BlockBackend *target = s->target;
blk_ref(target);
blk_drain(target);
blk_unref(target);

In this case, the write permissions for s->target are retained until
after blk_drain(), which makes removing mirror_top_bs fail for the
active commit case (can't have a writable backing file in the chain
without the filter driver).

Explicitly dropping the permissions first means that the additional
reference doesn't hurt and the job can complete successfully even if
called from the nested blk_drain().

Cc: qemu-***@nongnu.org
Signed-off-by: Kevin Wolf <***@redhat.com>
Acked-by: Paolo Bonzini <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
(cherry picked from commit 63c8ef289087a225d445319d047501d4fe593687)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/mirror.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/block/mirror.c b/block/mirror.c
index 2173a2f..4e8f124 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -514,7 +514,12 @@ static void mirror_exit(BlockJob *job, void *opaque)

/* Remove target parent that still uses BLK_PERM_WRITE/RESIZE before
* inserting target_bs at s->to_replace, where we might not be able to get
- * these permissions. */
+ * these permissions.
+ *
+ * Note that blk_unref() alone doesn't necessarily drop permissions because
+ * we might be running nested inside mirror_drain(), which takes an extra
+ * reference, so use an explicit blk_set_perm() first. */
+ blk_set_perm(s->target, 0, BLK_PERM_ALL, &error_abort);
blk_unref(s->target);
s->target = NULL;
--
2.7.4
Michael Roth
2017-08-29 00:14:05 UTC
Permalink
From: Paolo Bonzini <***@redhat.com>

The curl driver has a ugly hack where, if it cannot find an empty CURLState,
it just uses aio_poll to wait for one to be empty. This is probably
buggy when used together with dataplane, and the simplest way to fix it
is to use coroutines instead.

A more immediate effect of the bug however is that it can cause a
recursive call to curl_readv_bh_cb and recursively taking the
BDRVCURLState mutex. This causes a deadlock.

The fix is to unlock the mutex around aio_poll, but for cleanliness we
should also take the mutex around all calls to curl_init_state, even if
reaching the unlock/lock pair is impossible. The same is true for
curl_clean_state.

Reported-by: Kun Wei <***@redhat.com>
Tested-by: Richard W.M. Jones <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Reviewed-by: Jeff Cody <***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
Message-id: 20170515100059.15795-4-***@redhat.com
Cc: qemu-***@nongnu.org
Cc: Jeff Cody <***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
Signed-off-by: Jeff Cody <***@redhat.com>
(cherry picked from commit 456af346297ebef86aa097b3609534d34f3d2f75)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/curl.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/block/curl.c b/block/curl.c
index 59e69c6..dddf5ef 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -281,6 +281,7 @@ read_end:
return size * nmemb;
}

+/* Called with s->mutex held. */
static int curl_find_buf(BDRVCURLState *s, size_t start, size_t len,
CURLAIOCB *acb)
{
@@ -453,6 +454,7 @@ static void curl_multi_timeout_do(void *arg)
#endif
}

+/* Called with s->mutex held. */
static CURLState *curl_init_state(BlockDriverState *bs, BDRVCURLState *s)
{
CURLState *state = NULL;
@@ -471,7 +473,9 @@ static CURLState *curl_init_state(BlockDriverState *bs, BDRVCURLState *s)
break;
}
if (!state) {
+ qemu_mutex_unlock(&s->mutex);
aio_poll(bdrv_get_aio_context(bs), true);
+ qemu_mutex_lock(&s->mutex);
}
} while(!state);

@@ -534,6 +538,7 @@ static CURLState *curl_init_state(BlockDriverState *bs, BDRVCURLState *s)
return state;
}

+/* Called with s->mutex held. */
static void curl_clean_state(CURLState *s)
{
int j;
@@ -565,6 +570,7 @@ static void curl_detach_aio_context(BlockDriverState *bs)
BDRVCURLState *s = bs->opaque;
int i;

+ qemu_mutex_lock(&s->mutex);
for (i = 0; i < CURL_NUM_STATES; i++) {
if (s->states[i].in_use) {
curl_clean_state(&s->states[i]);
@@ -580,6 +586,7 @@ static void curl_detach_aio_context(BlockDriverState *bs)
curl_multi_cleanup(s->multi);
s->multi = NULL;
}
+ qemu_mutex_unlock(&s->mutex);

timer_del(&s->timer);
}
@@ -677,6 +684,7 @@ static int curl_open(BlockDriverState *bs, QDict *options, int flags,
return -EROFS;
}

+ qemu_mutex_init(&s->mutex);
opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
qemu_opts_absorb_qdict(opts, options, &local_err);
if (local_err) {
@@ -747,7 +755,9 @@ static int curl_open(BlockDriverState *bs, QDict *options, int flags,
DPRINTF("CURL: Opening %s\n", file);
s->aio_context = bdrv_get_aio_context(bs);
s->url = g_strdup(file);
+ qemu_mutex_lock(&s->mutex);
state = curl_init_state(bs, s);
+ qemu_mutex_unlock(&s->mutex);
if (!state)
goto out_noclean;

@@ -791,11 +801,12 @@ static int curl_open(BlockDriverState *bs, QDict *options, int flags,
}
DPRINTF("CURL: Size = %zd\n", s->len);

+ qemu_mutex_lock(&s->mutex);
curl_clean_state(state);
+ qemu_mutex_unlock(&s->mutex);
curl_easy_cleanup(state->curl);
state->curl = NULL;

- qemu_mutex_init(&s->mutex);
curl_attach_aio_context(bs, bdrv_get_aio_context(bs));

qemu_opts_del(opts);
@@ -806,6 +817,7 @@ out:
curl_easy_cleanup(state->curl);
state->curl = NULL;
out_noclean:
+ qemu_mutex_destroy(&s->mutex);
g_free(s->cookie);
g_free(s->url);
qemu_opts_del(opts);
--
2.7.4
Michael Roth
2017-08-29 00:14:24 UTC
Permalink
check-qom-proplist originally added tests for verifying that
object-creation helpers object_new_with_{props,propv} behaved in
similar fashion to the "traditional" method involving setting each
individual property separately after object creation rather than
via a single call.

Another similar "helper" for creating Objects exists in the form of
objects specified via -object command-line parameters. By that
rationale, we extend check-qom-proplist to include similar checks
for command-line-created objects by employing the same
qemu_opts_parse()-based parsing the vl.c employs.

This parser has a side-effect of parsing the object's options into
a QemuOpt structure and registering this in the global QemuOptsList
using the Object's ID. This can conflict with future Object instances
that attempt to use the same ID if we don't ensure this is cleaned
up as part of Object finalization, so we include a FIXME stub to test
for this case, which will then be resolved in a subsequent patch.

Suggested-by: Daniel Berrange <***@redhat.com>
Cc: "Dr. David Alan Gilbert" <***@redhat.com>
Cc: Markus Armbruster <***@redhat.com>
Cc: Eric Blake <***@redhat.com>
Cc: Daniel Berrange <***@redhat.com>
Cc: qemu-***@nongnu.org
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
Reviewed-by: Markus Armbruster <***@redhat.com>
Message-Id: <1496531612-22166-2-git-send-email-***@linux.vnet.ibm.com>
[Comment formatting tidied up]
Signed-off-by: Markus Armbruster <***@redhat.com>

(cherry picked from commit a1af255f065ccf3f47a7bfe88f1dbc9eeca36935)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
tests/check-qom-proplist.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 56 insertions(+)

diff --git a/tests/check-qom-proplist.c b/tests/check-qom-proplist.c
index a16cefc..e3b3ae4 100644
--- a/tests/check-qom-proplist.c
+++ b/tests/check-qom-proplist.c
@@ -23,6 +23,9 @@
#include "qapi/error.h"
#include "qom/object.h"
#include "qemu/module.h"
+#include "qemu/option.h"
+#include "qemu/config-file.h"
+#include "qom/object_interfaces.h"


#define TYPE_DUMMY "qemu-dummy"
@@ -162,6 +165,10 @@ static const TypeInfo dummy_info = {
.instance_finalize = dummy_finalize,
.class_size = sizeof(DummyObjectClass),
.class_init = dummy_class_init,
+ .interfaces = (InterfaceInfo[]) {
+ { TYPE_USER_CREATABLE },
+ { }
+ }
};


@@ -320,6 +327,14 @@ static const TypeInfo dummy_backend_info = {
.class_size = sizeof(DummyBackendClass),
};

+static QemuOptsList qemu_object_opts = {
+ .name = "object",
+ .implied_opt_name = "qom-type",
+ .head = QTAILQ_HEAD_INITIALIZER(qemu_object_opts.head),
+ .desc = {
+ { }
+ },
+};


static void test_dummy_createv(void)
@@ -388,6 +403,46 @@ static void test_dummy_createlist(void)
object_unparent(OBJECT(dobj));
}

+static void test_dummy_createcmdl(void)
+{
+ QemuOpts *opts;
+ DummyObject *dobj;
+ Error *err = NULL;
+ const char *params = TYPE_DUMMY \
+ ",id=dev0," \
+ "bv=yes,sv=Hiss hiss hiss,av=platypus";
+
+ qemu_add_opts(&qemu_object_opts);
+ opts = qemu_opts_parse(&qemu_object_opts, params, true, &err);
+ g_assert(err == NULL);
+ g_assert(opts);
+
+ dobj = DUMMY_OBJECT(user_creatable_add_opts(opts, &err));
+ g_assert(err == NULL);
+ g_assert(dobj);
+ g_assert_cmpstr(dobj->sv, ==, "Hiss hiss hiss");
+ g_assert(dobj->bv == true);
+ g_assert(dobj->av == DUMMY_PLATYPUS);
+
+ user_creatable_del("dev0", &err);
+ g_assert(err == NULL);
+ error_free(err);
+
+ /*
+ * cmdline-parsing via qemu_opts_parse() results in a QemuOpts entry
+ * corresponding to the Object's ID to be added to the QemuOptsList
+ * for objects. To avoid having this entry conflict with future
+ * Objects using the same ID (which can happen in cases where
+ * qemu_opts_parse() is used to parse the object params, such as
+ * with hmp_object_add() at the time of this comment), we need to
+ * check for this in user_creatable_del() and remove the QemuOpts if
+ * it is present.
+ *
+ * FIXME: add an assert to verify that the QemuOpts is cleaned up
+ * once the corresponding cleanup code is added.
+ */
+}
+
static void test_dummy_badenum(void)
{
Error *err = NULL;
@@ -525,6 +580,7 @@ int main(int argc, char **argv)

g_test_add_func("/qom/proplist/createlist", test_dummy_createlist);
g_test_add_func("/qom/proplist/createv", test_dummy_createv);
+ g_test_add_func("/qom/proplist/createcmdline", test_dummy_createcmdl);
g_test_add_func("/qom/proplist/badenum", test_dummy_badenum);
g_test_add_func("/qom/proplist/getenum", test_dummy_getenum);
g_test_add_func("/qom/proplist/iterator", test_dummy_iterator);
--
2.7.4
Michael Roth
2017-08-29 00:14:10 UTC
Permalink
From: Sameeh Jubran <***@daynix.com>

This commit fixes a bug which causes the guest to hang. The bug was
observed upon a "receive overrun" (bit #6 of the ICR register)
interrupt which could be triggered post migration in a heavy traffic
environment. Even though the "receive overrun" bit (#6) is masked out
by the IMS register (refer to the log below) the driver still receives
an interrupt as the "receive overrun" bit (#6) causes the "Other" -
bit #24 of the ICR register - bit to be set as documented below. The
driver handles the interrupt and clears the "Other" bit (#24) but
doesn't clear the "receive overrun" bit (#6) which leads to an
infinite loop. Apparently the Windows driver expects that the "receive
overrun" bit and other ones - documented below - to be cleared when
the "Other" bit (#24) is cleared.

So to sum that up:
1. Bit #6 of the ICR register is set by heavy traffic
2. As a results of setting bit #6, bit #24 is set
3. The driver receives an interrupt for bit 24 (it doesn't receieve an
interrupt for bit #6 as it is masked out by IMS)
4. The driver handles and clears the interrupt of bit #24
5. Bit #6 is still set.
6. 2 happens all over again

The Interrupt Cause Read - ICR register:

The ICR has the "Other" bit - bit #24 - that is set when one or more
of the following ICR register's bits are set:

LSC - bit #2, RXO - bit #6, MDAC - bit #9, SRPD - bit #16, ACK - bit
#17, MNG - bit #18

This bug can occur with any of these bits depending on the driver's
behaviour and the way it configures the device. However, trying to
reproduce it with any bit other than RX0 is challenging and came to
failure as the drivers don't implement most of these bits, trying to
reproduce it with LSC (Link Status Change - bit #2) bit didn't succeed
too as it seems that Windows handles this bit differently.

Log sample of the storm:

***@1494850819.411877:e1000e_irq_pending_interrupts ICR PENDING: 0x1000000 (ICR: 0x815000c2, IMS: 0x1a00004)
***@1494850819.411900:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
***@1494850819.411915:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
***@1494850819.412380:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
***@1494850819.412395:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
***@1494850819.412436:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
***@1494850819.412441:e1000e_irq_pending_interrupts ICR PENDING: 0x0 (ICR: 0x815000c2, IMS: 0xa00004)
***@1494850819.412998:e1000e_irq_pending_interrupts ICR PENDING: 0x1000000 (ICR: 0x815000c2, IMS: 0x1a00004)

* This bug behaviour wasn't observed with the Linux driver.

This commit solves:
https://bugzilla.redhat.com/show_bug.cgi?id=1447935
https://bugzilla.redhat.com/show_bug.cgi?id=1449490

Cc: qemu-***@nongnu.org
Signed-off-by: Sameeh Jubran <***@redhat.com>
Signed-off-by: Jason Wang <***@redhat.com>
(cherry picked from commit 82342e91b60a4a078811df4e1a545e57abffa11d)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/net/e1000e_core.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/hw/net/e1000e_core.c b/hw/net/e1000e_core.c
index 28c5be1..8140564 100644
--- a/hw/net/e1000e_core.c
+++ b/hw/net/e1000e_core.c
@@ -2454,14 +2454,20 @@ e1000e_set_ics(E1000ECore *core, int index, uint32_t val)
static void
e1000e_set_icr(E1000ECore *core, int index, uint32_t val)
{
+ uint32_t icr = 0;
if ((core->mac[ICR] & E1000_ICR_ASSERTED) &&
(core->mac[CTRL_EXT] & E1000_CTRL_EXT_IAME)) {
trace_e1000e_irq_icr_process_iame();
e1000e_clear_ims_bits(core, core->mac[IAM]);
}

- trace_e1000e_irq_icr_write(val, core->mac[ICR], core->mac[ICR] & ~val);
- core->mac[ICR] &= ~val;
+ icr = core->mac[ICR] & ~val;
+ /* Windows driver expects that the "receive overrun" bit and other
+ * ones to be cleared when the "Other" bit (#24) is cleared.
+ */
+ icr = (val & E1000_ICR_OTHER) ? (icr & ~E1000_ICR_OTHER_CAUSES) : icr;
+ trace_e1000e_irq_icr_write(val, core->mac[ICR], icr);
+ core->mac[ICR] = icr;
e1000e_update_interrupt_state(core);
}
--
2.7.4
Michael Roth
2017-08-29 00:14:07 UTC
Permalink
From: Hervé Poussineau <***@reactos.org>

- bs->total_sectors is the number of sectors of the whole disk
- s->sector_count is the number of sectors of the FAT partition

This fixes the following assert in qemu-img map:
qemu-img.c:2641: get_block_status: Assertion `nb_sectors' failed.

This also fixes an infinite loop in qemu-img convert.

Fixes: 4480e0f924a42e1db8b8cfcac4d0634dd1bb27a0
Fixes: https://bugs.launchpad.net/qemu/+bug/1599539
Cc: qemu-***@nongnu.org
Signed-off-by: Hervé Poussineau <***@reactos.org>
Reviewed-by: Eric Blake <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 139921aaa77c435104308ad53b631a00c3b65ae8)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/vvfat.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/block/vvfat.c b/block/vvfat.c
index 8b4e4eb..7d08f65 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -2958,8 +2958,7 @@ vvfat_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
static int64_t coroutine_fn vvfat_co_get_block_status(BlockDriverState *bs,
int64_t sector_num, int nb_sectors, int *n, BlockDriverState **file)
{
- BDRVVVFATState* s = bs->opaque;
- *n = s->sector_count - sector_num;
+ *n = bs->total_sectors - sector_num;
if (*n > nb_sectors) {
*n = nb_sectors;
} else if (*n < 0) {
--
2.7.4
Michael Roth
2017-08-29 00:14:25 UTC
Permalink
Currently objects specified on the command-line are only partially
cleaned up when 'object_del' is issued in either HMP or QMP: the
object itself is fully finalized, but the QemuOpts are not removed.
This results in the following behavior:

x86_64-softmmu/qemu-system-x86_64 -monitor stdio \
-object memory-backend-ram,id=ram1,size=256M

QEMU 2.7.91 monitor - type 'help' for more information
(qemu) object_del ram1
(qemu) object_del ram1
object 'ram1' not found
(qemu) object_add memory-backend-ram,id=ram1,size=256M
Duplicate ID 'ram1' for object
Try "help object_add" for more information

which can be an issue for use-cases like memory hotplug.

This happens on the HMP side because hmp_object_add() attempts to
create a temporary QemuOpts entry with ID 'ram1', which ends up
conflicting with the command-line-created entry, since it was never
cleaned up during the previous hmp_object_del() call.

We address this by adding a check in user_creatable_del(), which
is called by both qmp_object_del() and hmp_object_del() to handle
the actual object cleanup, to determine whether an option group entry
matching the object's ID is present and removing it if it is.

Note that qmp_object_add() never attempts to create a temporary
QemuOpts entry, so it does not encounter the duplicate ID error,
which is why this isn't generally visible in libvirt.

Cc: "Dr. David Alan Gilbert" <***@redhat.com>
Cc: Markus Armbruster <***@redhat.com>
Cc: Eric Blake <***@redhat.com>
Cc: Daniel Berrange <***@redhat.com>
Cc: qemu-***@nongnu.org
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
Reviewed-by: Daniel P. Berrange <***@redhat.com>
Reviewed-by: Markus Armbruster <***@redhat.com>
Message-Id: <1496531612-22166-3-git-send-email-***@linux.vnet.ibm.com>
Signed-off-by: Markus Armbruster <***@redhat.com>
(cherry picked from commit c645d5acee0ae022534cb609184277ec2b4a8577)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
qom/object_interfaces.c | 9 +++++++++
tests/check-qom-proplist.c | 4 ++--
2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/qom/object_interfaces.c b/qom/object_interfaces.c
index d4253a8..ff27e06 100644
--- a/qom/object_interfaces.c
+++ b/qom/object_interfaces.c
@@ -4,6 +4,7 @@
#include "qemu/module.h"
#include "qapi-visit.h"
#include "qapi/opts-visitor.h"
+#include "qemu/config-file.h"

void user_creatable_complete(Object *obj, Error **errp)
{
@@ -181,6 +182,14 @@ void user_creatable_del(const char *id, Error **errp)
error_setg(errp, "object '%s' is in use, can not be deleted", id);
return;
}
+
+ /*
+ * if object was defined on the command-line, remove its corresponding
+ * option group entry
+ */
+ qemu_opts_del(qemu_opts_find(qemu_find_opts_err("object", &error_abort),
+ id));
+
object_unparent(obj);
}

diff --git a/tests/check-qom-proplist.c b/tests/check-qom-proplist.c
index e3b3ae4..8e432e9 100644
--- a/tests/check-qom-proplist.c
+++ b/tests/check-qom-proplist.c
@@ -438,9 +438,9 @@ static void test_dummy_createcmdl(void)
* check for this in user_creatable_del() and remove the QemuOpts if
* it is present.
*
- * FIXME: add an assert to verify that the QemuOpts is cleaned up
- * once the corresponding cleanup code is added.
+ * The below check ensures this works as expected.
*/
+ g_assert_null(qemu_opts_find(&qemu_object_opts, "dev0"));
}

static void test_dummy_badenum(void)
--
2.7.4
Michael Roth
2017-08-29 00:13:39 UTC
Permalink
From: Max Reitz <***@redhat.com>

After storing the creation options for the new image into @opts, we
fetch some things for our own information, like the backing file name,
or whether to use encryption or preallocation.

With the -n parameter, there will not be any creation options; this is
not too bad because this just means that querying a NULL @opts will
always return the default value.

However, we also use @opts for the --object options. Therefore, @opts is
not necessarily NULL if -n was specified; instead, it may contain those
options. In practice, this probably does not cause any problems because
there most likely is no object that supports any of the parameters we
query here, but this is neither something we should rely on nor does
this variable reuse make the code very nice to read.

Therefore, just use an own variable for the --object options.

Cc: qemu-stable <qemu-***@nongnu.org>
Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
* applied patch from v1 of series as suggested by author
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
qemu-img.c | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/qemu-img.c b/qemu-img.c
index 9aa7823..cfc3bc3 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -2065,14 +2065,16 @@ static int img_convert(int argc, char **argv)
case 'W':
wr_in_order = false;
break;
- case OPTION_OBJECT:
- opts = qemu_opts_parse_noisily(&qemu_object_opts,
- optarg, true);
- if (!opts) {
+ case OPTION_OBJECT: {
+ QemuOpts *object_opts;
+ object_opts = qemu_opts_parse_noisily(&qemu_object_opts,
+ optarg, true);
+ if (!object_opts) {
ret = -1;
goto fail_getopt;
}
break;
+ }
case OPTION_IMAGE_OPTS:
image_opts = true;
break;
--
2.7.4
Michael Roth
2017-08-29 00:13:38 UTC
Permalink
From: Max Reitz <***@redhat.com>

Otherwise the qemu-img process will exit with EXIT_SUCCESS instead of
EXIT_FAILURE.

Cc: qemu-stable <qemu-***@nongnu.org>
Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <***@amsat.org>
* applied directly to stable, upstream code has issue fixed via a
refactoring introduced by 9fd77f9, which isn't targetted for stable
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
qemu-img.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/qemu-img.c b/qemu-img.c
index b220cf7..9aa7823 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -2069,6 +2069,7 @@ static int img_convert(int argc, char **argv)
opts = qemu_opts_parse_noisily(&qemu_object_opts,
optarg, true);
if (!opts) {
+ ret = -1;
goto fail_getopt;
}
break;
@@ -2081,6 +2082,7 @@ static int img_convert(int argc, char **argv)
if (qemu_opts_foreach(&qemu_object_opts,
user_creatable_add_opts_foreach,
NULL, NULL)) {
+ ret = -1;
goto fail_getopt;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:09 UTC
Permalink
From: Fam Zheng <***@redhat.com>

This matches the qbus_set_hotplug_handler in realize, and it releases
the final reference to the embedded VirtIODevice so that it is
properly finalized.

A use-after-free is fixed with this patch, indirectly:
virtio_device_instance_finalize wasn't called at hot-unplug, and the
vdev->listener would be a dangling pointer in the global and the per
address space listener list. See also RHBZ 1449031.

Cc: qemu-***@nongnu.org
Signed-off-by: Fam Zheng <***@redhat.com>
Message-Id: <20170518102808.30046-1-***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
(cherry picked from commit 2cbe2de5454cf9af44b620b2b40d56361a12a45f)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/scsi/virtio-scsi.c | 3 +++
1 file changed, 3 insertions(+)

diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index bd62d08..cb1c123 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -915,6 +915,9 @@ void virtio_scsi_common_unrealize(DeviceState *dev, Error **errp)

static void virtio_scsi_device_unrealize(DeviceState *dev, Error **errp)
{
+ VirtIOSCSI *s = VIRTIO_SCSI(dev);
+
+ qbus_set_hotplug_handler(BUS(&s->bus), NULL, &error_abort);
virtio_scsi_common_unrealize(dev, errp);
}
--
2.7.4
Michael Roth
2017-08-29 00:14:27 UTC
Permalink
From: Greg Kurz <***@kaod.org>

This function has three implementations. Two are stubs that do nothing
and the third one only passes the obj_path argument to:

Object *object_resolve_path(const char *path, bool *ambiguous);

Signed-off-by: Greg Kurz <***@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <***@amsat.org>
Reviewed-by: Thomas Huth <***@redhat.com>
Signed-off-by: David Gibson <***@gibson.dropbear.id.au>
(cherry picked from commit ec69355beffe138c0f97306e65410e5dbc605554)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
target/ppc/kvm.c | 4 ++--
target/ppc/kvm_ppc.h | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c
index 9f1f132..89f3586 100644
--- a/target/ppc/kvm.c
+++ b/target/ppc/kvm.c
@@ -415,7 +415,7 @@ static void kvm_fixup_page_sizes(PowerPCCPU *cpu)
}
}

-bool kvmppc_is_mem_backend_page_size_ok(char *obj_path)
+bool kvmppc_is_mem_backend_page_size_ok(const char *obj_path)
{
Object *mem_obj = object_resolve_path(obj_path, NULL);
char *mempath = object_property_get_str(mem_obj, "mem-path", NULL);
@@ -436,7 +436,7 @@ static inline void kvm_fixup_page_sizes(PowerPCCPU *cpu)
{
}

-bool kvmppc_is_mem_backend_page_size_ok(char *obj_path)
+bool kvmppc_is_mem_backend_page_size_ok(const char *obj_path)
{
return true;
}
diff --git a/target/ppc/kvm_ppc.h b/target/ppc/kvm_ppc.h
index 8e9f42d..eec5cca 100644
--- a/target/ppc/kvm_ppc.h
+++ b/target/ppc/kvm_ppc.h
@@ -57,7 +57,7 @@ int kvmppc_enable_hwrng(void);
int kvmppc_put_books_sregs(PowerPCCPU *cpu);
PowerPCCPUClass *kvm_ppc_get_host_cpu_class(void);

-bool kvmppc_is_mem_backend_page_size_ok(char *obj_path);
+bool kvmppc_is_mem_backend_page_size_ok(const char *obj_path);

#else

@@ -191,7 +191,7 @@ static inline uint64_t kvmppc_rma_size(uint64_t current_size,
return ram_size;
}

-static inline bool kvmppc_is_mem_backend_page_size_ok(char *obj_path)
+static inline bool kvmppc_is_mem_backend_page_size_ok(const char *obj_path)
{
return true;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:28 UTC
Permalink
From: Greg Kurz <***@kaod.org>

The string returned by object_property_get_str() is dynamically allocated.

Signed-off-by: Greg Kurz <***@kaod.org>
Reviewed-by: Thomas Huth <***@redhat.com>
Signed-off-by: David Gibson <***@gibson.dropbear.id.au>
(cherry picked from commit 2d3e302ec2246d703ffa8d8f8769a3fa448d8145)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
target/ppc/kvm.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c
index 89f3586..b1dd41c 100644
--- a/target/ppc/kvm.c
+++ b/target/ppc/kvm.c
@@ -423,6 +423,7 @@ bool kvmppc_is_mem_backend_page_size_ok(const char *obj_path)

if (mempath) {
pagesize = qemu_mempath_getpagesize(mempath);
+ g_free(mempath);
} else {
pagesize = getpagesize();
}
--
2.7.4
Michael Roth
2017-08-29 00:14:18 UTC
Permalink
From: Eric Blake <***@redhat.com>

Use blkdebug's new geometry constraints to emulate setups that
have needed past regression fixes: write zeroes asserting
when running through a loopback block device with max-transfer
smaller than cluster size, and discard rounding away portions
of requests not aligned to preferred boundaries. Also, add
coverage that the block layer is honoring max transfer limits.

For now, a single iotest performs all actions, with the idea
that we can add future blkdebug constraint test cases in the
same file; but it can be split into multiple iotests if we find
reason to run one portion of the test in more setups than what
are possible in the other.

For reference, the final portion of the test (checking whether
discard passes as much as possible to the lowest layers of the
stack) works as follows:

qemu-io: discard 30M at 80000001, passed to blkdebug
blkdebug: discard 511 bytes at 80000001, -ENOTSUP (smaller than
blkdebug's 512 align)
blkdebug: discard 14371328 bytes at 80000512, passed to qcow2
qcow2: discard 739840 bytes at 80000512, -ENOTSUP (smaller than
qcow2's 1M align)
qcow2: discard 13M bytes at 77M, succeeds
blkdebug: discard 15M bytes at 90M, passed to qcow2
qcow2: discard 15M bytes at 90M, succeeds
blkdebug: discard 1356800 bytes at 105M, passed to qcow2
qcow2: discard 1M at 105M, succeeds
qcow2: discard 308224 bytes at 106M, -ENOTSUP (smaller than qcow2's
1M align)
blkdebug: discard 1 byte at 111457280, -ENOTSUP (smaller than
blkdebug's 512 align)

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170429191419.30051-10-***@redhat.com
[mreitz: For cooperation with image locking, add -r to the qemu-io
invocation which verifies the image content]
Signed-off-by: Max Reitz <***@redhat.com>

(cherry picked from commit 40812d937392fddc11f72a668aef251039cc15ce)
Conflicts:
tests/qemu-iotests/group
* dropped context dependency on other test groups
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
tests/qemu-iotests/177 | 114 +++++++++++++++++++++++++++++++++++++++++++++
tests/qemu-iotests/177.out | 49 +++++++++++++++++++
tests/qemu-iotests/group | 1 +
3 files changed, 164 insertions(+)
create mode 100755 tests/qemu-iotests/177
create mode 100644 tests/qemu-iotests/177.out

diff --git a/tests/qemu-iotests/177 b/tests/qemu-iotests/177
new file mode 100755
index 0000000..2005c17
--- /dev/null
+++ b/tests/qemu-iotests/177
@@ -0,0 +1,114 @@
+#!/bin/bash
+#
+# Test corner cases with unusual block geometries
+#
+# Copyright (C) 2016-2017 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program 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 General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+# creator
+owner=***@redhat.com
+
+seq=`basename $0`
+echo "QA output created by $seq"
+
+here=`pwd`
+status=1 # failure is the default!
+
+_cleanup()
+{
+ _cleanup_test_img
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+
+_supported_fmt qcow2
+_supported_proto file
+
+CLUSTER_SIZE=1M
+size=128M
+options=driver=blkdebug,image.driver=qcow2
+
+echo
+echo "== setting up files =="
+
+TEST_IMG="$TEST_IMG.base" _make_test_img $size
+$QEMU_IO -c "write -P 11 0 $size" "$TEST_IMG.base" | _filter_qemu_io
+_make_test_img -b "$TEST_IMG.base"
+$QEMU_IO -c "write -P 22 0 $size" "$TEST_IMG" | _filter_qemu_io
+
+# Limited to 64k max-transfer
+echo
+echo "== constrained alignment and max-transfer =="
+limits=align=4k,max-transfer=64k
+$QEMU_IO -c "open -o $options,$limits blkdebug::$TEST_IMG" \
+ -c "write -P 33 1000 128k" -c "read -P 33 1000 128k" | _filter_qemu_io
+
+echo
+echo "== write zero with constrained max-transfer =="
+limits=align=512,max-transfer=64k,opt-write-zero=$CLUSTER_SIZE
+$QEMU_IO -c "open -o $options,$limits blkdebug::$TEST_IMG" \
+ -c "write -z 8003584 2093056" | _filter_qemu_io
+
+# non-power-of-2 write-zero/discard alignments
+echo
+echo "== non-power-of-2 write zeroes limits =="
+
+limits=align=512,opt-write-zero=15M,max-write-zero=15M,opt-discard=15M,max-discard=15M
+$QEMU_IO -c "open -o $options,$limits blkdebug::$TEST_IMG" \
+ -c "write -z 32M 32M" | _filter_qemu_io
+
+echo
+echo "== non-power-of-2 discard limits =="
+
+limits=align=512,opt-write-zero=15M,max-write-zero=15M,opt-discard=15M,max-discard=15M
+$QEMU_IO -c "open -o $options,$limits blkdebug::$TEST_IMG" \
+ -c "discard 80000001 30M" | _filter_qemu_io
+
+echo
+echo "== verify image content =="
+
+function verify_io()
+{
+ if ($QEMU_IMG info -f "$IMGFMT" "$TEST_IMG" |
+ grep "compat: 0.10" > /dev/null); then
+ # For v2 images, discarded clusters are read from the backing file
+ discarded=11
+ else
+ # Discarded clusters are zeroed for v3 or later
+ discarded=0
+ fi
+
+ echo read -P 22 0 1000
+ echo read -P 33 1000 128k
+ echo read -P 22 132072 7871512
+ echo read -P 0 8003584 2093056
+ echo read -P 22 10096640 23457792
+ echo read -P 0 32M 32M
+ echo read -P 22 64M 13M
+ echo read -P $discarded 77M 29M
+ echo read -P 22 106M 22M
+}
+
+verify_io | $QEMU_IO -r "$TEST_IMG" | _filter_qemu_io
+
+_check_test_img
+
+# success, all done
+echo "*** done"
+status=0
diff --git a/tests/qemu-iotests/177.out b/tests/qemu-iotests/177.out
new file mode 100644
index 0000000..e887542
--- /dev/null
+++ b/tests/qemu-iotests/177.out
@@ -0,0 +1,49 @@
+QA output created by 177
+
+== setting up files ==
+Formatting 'TEST_DIR/t.IMGFMT.base', fmt=IMGFMT size=134217728
+wrote 134217728/134217728 bytes at offset 0
+128 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/t.IMGFMT.base
+wrote 134217728/134217728 bytes at offset 0
+128 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== constrained alignment and max-transfer ==
+wrote 131072/131072 bytes at offset 1000
+128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 131072/131072 bytes at offset 1000
+128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== write zero with constrained max-transfer ==
+wrote 2093056/2093056 bytes at offset 8003584
+1.996 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== non-power-of-2 write zeroes limits ==
+wrote 33554432/33554432 bytes at offset 33554432
+32 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== non-power-of-2 discard limits ==
+discard 31457280/31457280 bytes at offset 80000001
+30 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== verify image content ==
+read 1000/1000 bytes at offset 0
+1000 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 131072/131072 bytes at offset 1000
+128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 7871512/7871512 bytes at offset 132072
+7.507 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 2093056/2093056 bytes at offset 8003584
+1.996 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 23457792/23457792 bytes at offset 10096640
+22.371 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 33554432/33554432 bytes at offset 33554432
+32 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 13631488/13631488 bytes at offset 67108864
+13 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 30408704/30408704 bytes at offset 80740352
+29 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 23068672/23068672 bytes at offset 111149056
+22 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+No errors were found on the image.
+*** done
diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group
index 43142dd..a513ec7 100644
--- a/tests/qemu-iotests/group
+++ b/tests/qemu-iotests/group
@@ -169,3 +169,4 @@
174 auto
175 auto quick
176 rw auto backing
+177 rw auto quick
--
2.7.4
Michael Roth
2017-08-29 00:14:32 UTC
Permalink
From: Kevin Wolf <***@redhat.com>

The final bdrv_set_backing_hd() could be working on already freed nodes
because the commit job drops its references (through BlockBackends) to
both overlay_bs and top already a bit earlier.

One way to trigger the bug is hot unplugging a disk for which
blockdev_mark_auto_del() cancels the block job.

Fix this by taking BDS-level references while we're still using the
nodes.

Cc: qemu-***@nongnu.org
Signed-off-by: Kevin Wolf <***@redhat.com>
Reviewed-by: John Snow <***@redhat.com>
(cherry picked from commit 19ebd13ed45ad5d5f277f5914d55b83f13eb09eb)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/commit.c | 7 +++++++
1 file changed, 7 insertions(+)

diff --git a/block/commit.c b/block/commit.c
index 76a0d98..3bae46e 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -89,6 +89,10 @@ static void commit_complete(BlockJob *job, void *opaque)
int ret = data->ret;
bool remove_commit_top_bs = false;

+ /* Make sure overlay_bs and top stay around until bdrv_set_backing_hd() */
+ bdrv_ref(top);
+ bdrv_ref(overlay_bs);
+
/* Remove base node parent that still uses BLK_PERM_WRITE/RESIZE before
* the normal backing chain can be restored. */
blk_unref(s->base);
@@ -124,6 +128,9 @@ static void commit_complete(BlockJob *job, void *opaque)
if (remove_commit_top_bs) {
bdrv_set_backing_hd(overlay_bs, top, &error_abort);
}
+
+ bdrv_unref(overlay_bs);
+ bdrv_unref(top);
}

static void coroutine_fn commit_run(void *opaque)
--
2.7.4
Michael Roth
2017-08-29 00:14:11 UTC
Permalink
From: Halil Pasic <***@linux.vnet.ibm.com>

Prior to the virtio-ccw-2.7 machine (and commit 2a79eb1a), our virtio
devices residing under the virtual-css bus do not have qdev_path based
migration stream identifiers (because their qdev_path is NULL). The ids
are instead generated when the device is registered as a composition of
the so called idstr, which takes the vmsd name as its value, and an
instance_id, which is which is calculated as a maximal instance_id
registered with the same idstr plus one, or zero (if none was registered
previously).

That means, under certain circumstances, one device might try, and even
succeed, to load the state of a different device. This can lead to
trouble.

Let us fail the migration if the above problem is detected during load.

How to reproduce the problem:
1) start qemu-system-s390x making sure you have the following devices
defined on your command line:
-device virtio-rng-ccw,id=rng1,devno=fe.0.0001
-device virtio-rng-ccw,id=rng2,devno=fe.0.0002
2) detach the devices and reattach in reverse order using the monitor:
(qemu) device_del rng1
(qemu) device_del rng2
(qemu) device_add virtio-rng-ccw,id=rng2,devno=fe.0.0002
(qemu) device_add virtio-rng-ccw,id=rng1,devno=fe.0.0001
3) save the state of the vm into a temporary file and quit QEMU:
(qemu) migrate "exec:gzip -c > /tmp/tmp_vmstate.gz"
(qemu) q
4) use your command line from step 1 with
-incoming "exec:gzip -c -d /tmp/tmp_vmstate.gz"
appended to reproduce the problem (while trying to to load the saved vm)

CC: qemu-***@nongnu.org
Signed-off-by: Halil Pasic <***@linux.vnet.ibm.com>
Reviewed-by: Dong Jia Shi <***@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <***@de.ibm.com>
Message-Id: <20170518111405.56947-1-***@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <***@de.ibm.com>
(cherry picked from commit 8ed179c937830143dc0e03daac30a55272ed89e3)
* removed context dep on d8d98db5
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/s390x/css.c | 14 ++++++++++++++
hw/s390x/virtio-ccw.c | 6 +++++-
2 files changed, 19 insertions(+), 1 deletion(-)

diff --git a/hw/s390x/css.c b/hw/s390x/css.c
index 37caa98..b24e8b7 100644
--- a/hw/s390x/css.c
+++ b/hw/s390x/css.c
@@ -14,6 +14,7 @@
#include "qapi/visitor.h"
#include "hw/qdev.h"
#include "qemu/bitops.h"
+#include "qemu/error-report.h"
#include "exec/address-spaces.h"
#include "cpu.h"
#include "hw/s390x/ioinst.h"
@@ -1676,13 +1677,26 @@ void subch_device_save(SubchDev *s, QEMUFile *f)
int subch_device_load(SubchDev *s, QEMUFile *f)
{
SubchDev *old_s;
+ Error *err = NULL;
uint16_t old_schid = s->schid;
+ uint16_t old_devno = s->devno;
int i;

s->cssid = qemu_get_byte(f);
s->ssid = qemu_get_byte(f);
s->schid = qemu_get_be16(f);
s->devno = qemu_get_be16(f);
+ if (s->devno != old_devno) {
+ /* Only possible if machine < 2.7 (no css_dev_path) */
+
+ error_setg(&err, "%x != %x", old_devno, s->devno);
+ error_append_hint(&err, "Devno mismatch, tried to load wrong section!"
+ " Likely reason: some sequences of plug and unplug"
+ " can break migration for machine versions prior to"
+ " 2.7 (known design flaw).\n");
+ error_report_err(err);
+ return -EINVAL;
+ }
/* Re-assign subch. */
if (old_schid != s->schid) {
old_s = channel_subsys.css[s->cssid]->sch_set[s->ssid]->sch[old_schid];
diff --git a/hw/s390x/virtio-ccw.c b/hw/s390x/virtio-ccw.c
index 00b3bde..c0c1db8 100644
--- a/hw/s390x/virtio-ccw.c
+++ b/hw/s390x/virtio-ccw.c
@@ -1264,9 +1264,13 @@ static int virtio_ccw_load_config(DeviceState *d, QEMUFile *f)
SubchDev *s = ccw_dev->sch;
VirtIODevice *vdev = virtio_ccw_get_vdev(s);
int len;
+ int ret;

s->driver_data = dev;
- subch_device_load(s, f);
+ ret = subch_device_load(s, f);
+ if (ret) {
+ return ret;
+ }
len = qemu_get_be32(f);
if (len != 0) {
dev->indicators = get_indicator(qemu_get_be64(f), len);
--
2.7.4
Michael Roth
2017-08-29 00:14:30 UTC
Permalink
From: Greg Kurz <***@kaod.org>

The string returned by object_property_get_str() is dynamically allocated.

(Spotted by Coverity, CID 1375942)

Signed-off-by: Greg Kurz <***@kaod.org>
Signed-off-by: David Gibson <***@gibson.dropbear.id.au>
(cherry picked from commit 8a9e0e7b890b2598da94646bf6a7272f3d3924de)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/ppc/spapr.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index 5564f78..58c15ef 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -2547,8 +2547,11 @@ static void spapr_memory_pre_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
if (mem_dev && !kvmppc_is_mem_backend_page_size_ok(mem_dev)) {
error_setg(errp, "Memory backend has bad page size. "
"Use 'memory-backend-file' with correct mem-path.");
- return;
+ goto out;
}
+
+out:
+ g_free(mem_dev);
}

typedef struct sPAPRDIMMState {
--
2.7.4
Michael Roth
2017-08-29 00:14:34 UTC
Permalink
From: Eric Blake <***@redhat.com>

Back in qemu 2.5, qemu-nbd was immune to port probes (a transient
server would not quit, regardless of how many probe connections
came and went, until a connection actually negotiated). But we
broke that in commit ee7d7aa when removing the return value to
nbd_client_new(), although that patch also introduced a bug causing
an assertion failure on a client that fails negotiation. We then
made it worse during refactoring in commit 1a6245a (a segfault
before we could even assert); the (masked) assertion was cleaned
up in d3780c2 (still in 2.6), and just recently we finally fixed
the segfault ("nbd: Fully intialize client in case of failed
negotiation"). But that still means that ever since we added
TLS support to qemu-nbd, we have been vulnerable to an ill-timed
port-scan being able to cause a denial of service by taking down
qemu-nbd before a real client has a chance to connect.

Since negotiation is now handled asynchronously via coroutines,
we no longer have a synchronous point of return by re-adding a
return value to nbd_client_new(). So this patch instead wires
things up to pass the negotiation status through the close_fn
callback function.

Simple test across two terminals:
$ qemu-nbd -f raw -p 30001 file
$ nmap 127.0.0.1 -p 30001 && \
qemu-io -c 'r 0 512' -f raw nbd://localhost:30001

Note that this patch does not change what constitutes successful
negotiation (thus, a client must enter transmission phase before
that client can be considered as a reason to terminate the server
when the connection ends). Perhaps we may want to tweak things
in a later patch to also treat a client that uses NBD_OPT_ABORT
as being a 'successful' negotiation (the client correctly talked
the NBD protocol, and informed us it was not going to use our
export after all), but that's a discussion for another day.

Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1451614

Signed-off-by: Eric Blake <***@redhat.com>
Message-Id: <20170608222617.20376-1-***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
(cherry picked from commit 0c9390d978cbf61e8f16c9f580fa96b305c43568)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
blockdev-nbd.c | 6 +++++-
include/block/nbd.h | 2 +-
nbd/server.c | 24 +++++++++++++++---------
qemu-nbd.c | 4 ++--
4 files changed, 23 insertions(+), 13 deletions(-)

diff --git a/blockdev-nbd.c b/blockdev-nbd.c
index 8a11807..8d7284a 100644
--- a/blockdev-nbd.c
+++ b/blockdev-nbd.c
@@ -27,6 +27,10 @@ typedef struct NBDServerData {

static NBDServerData *nbd_server;

+static void nbd_blockdev_client_closed(NBDClient *client, bool ignored)
+{
+ nbd_client_put(client);
+}

static gboolean nbd_accept(QIOChannel *ioc, GIOCondition condition,
gpointer opaque)
@@ -46,7 +50,7 @@ static gboolean nbd_accept(QIOChannel *ioc, GIOCondition condition,
qio_channel_set_name(QIO_CHANNEL(cioc), "nbd-server");
nbd_client_new(NULL, cioc,
nbd_server->tlscreds, NULL,
- nbd_client_put);
+ nbd_blockdev_client_closed);
object_unref(OBJECT(cioc));
return TRUE;
}
diff --git a/include/block/nbd.h b/include/block/nbd.h
index 3e373f0..b69c30d 100644
--- a/include/block/nbd.h
+++ b/include/block/nbd.h
@@ -160,7 +160,7 @@ void nbd_client_new(NBDExport *exp,
QIOChannelSocket *sioc,
QCryptoTLSCreds *tlscreds,
const char *tlsaclname,
- void (*close)(NBDClient *));
+ void (*close_fn)(NBDClient *, bool));
void nbd_client_get(NBDClient *client);
void nbd_client_put(NBDClient *client);

diff --git a/nbd/server.c b/nbd/server.c
index edfda84..a98bb21 100644
--- a/nbd/server.c
+++ b/nbd/server.c
@@ -81,7 +81,7 @@ static QTAILQ_HEAD(, NBDExport) exports = QTAILQ_HEAD_INITIALIZER(exports);

struct NBDClient {
int refcount;
- void (*close)(NBDClient *client);
+ void (*close_fn)(NBDClient *client, bool negotiated);

bool no_zeroes;
NBDExport *exp;
@@ -796,7 +796,7 @@ void nbd_client_put(NBDClient *client)
}
}

-static void client_close(NBDClient *client)
+static void client_close(NBDClient *client, bool negotiated)
{
if (client->closing) {
return;
@@ -811,8 +811,8 @@ static void client_close(NBDClient *client)
NULL);

/* Also tell the client, so that they release their reference. */
- if (client->close) {
- client->close(client);
+ if (client->close_fn) {
+ client->close_fn(client, negotiated);
}
}

@@ -993,7 +993,7 @@ void nbd_export_close(NBDExport *exp)

nbd_export_get(exp);
QTAILQ_FOREACH_SAFE(client, &exp->clients, next, next) {
- client_close(client);
+ client_close(client, true);
}
nbd_export_set_name(exp, NULL);
nbd_export_set_description(exp, NULL);
@@ -1355,7 +1355,7 @@ done:

out:
nbd_request_put(req);
- client_close(client);
+ client_close(client, true);
nbd_client_put(client);
}

@@ -1381,7 +1381,7 @@ static coroutine_fn void nbd_co_client_start(void *opaque)
qemu_co_mutex_init(&client->send_lock);

if (nbd_negotiate(data)) {
- client_close(client);
+ client_close(client, false);
goto out;
}

@@ -1391,11 +1391,17 @@ out:
g_free(data);
}

+/*
+ * Create a new client listener on the given export @exp, using the
+ * given channel @sioc. Begin servicing it in a coroutine. When the
+ * connection closes, call @close_fn with an indication of whether the
+ * client completed negotiation.
+ */
void nbd_client_new(NBDExport *exp,
QIOChannelSocket *sioc,
QCryptoTLSCreds *tlscreds,
const char *tlsaclname,
- void (*close_fn)(NBDClient *))
+ void (*close_fn)(NBDClient *, bool))
{
NBDClient *client;
NBDClientNewData *data = g_new(NBDClientNewData, 1);
@@ -1412,7 +1418,7 @@ void nbd_client_new(NBDExport *exp,
object_ref(OBJECT(client->sioc));
client->ioc = QIO_CHANNEL(sioc);
object_ref(OBJECT(client->ioc));
- client->close = close_fn;
+ client->close_fn = close_fn;

data->client = client;
data->co = qemu_coroutine_create(nbd_co_client_start, data);
diff --git a/qemu-nbd.c b/qemu-nbd.c
index 14e7947..3b55ffa 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -335,10 +335,10 @@ static void nbd_export_closed(NBDExport *exp)

static void nbd_update_server_watch(void);

-static void nbd_client_closed(NBDClient *client)
+static void nbd_client_closed(NBDClient *client, bool negotiated)
{
nb_fds--;
- if (nb_fds == 0 && !persistent && state == RUNNING) {
+ if (negotiated && nb_fds == 0 && !persistent && state == RUNNING) {
state = TERMINATE;
}
nbd_update_server_watch();
--
2.7.4
Michael Roth
2017-08-29 00:14:13 UTC
Permalink
From: Eric Blake <***@redhat.com>

Commits 04ed95f4 and 1a62d0ac updated the block layer to auto-fragment
any I/O to fit within device boundaries. Additionally, when using a
minimum alignment of 4k, we want to ensure the block layer does proper
read-modify-write rather than requesting I/O on a slice of a sector.
Let's enforce that the contract is obeyed when using blkdebug. For
now, blkdebug only allows alignment overrides, and just inherits other
limits from whatever device it is wrapping, but a future patch will
further enhance things.

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Kevin Wolf <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170429191419.30051-5-***@redhat.com
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit e0ef439588ce1ede747f82b77d893190c1cc9f4d)
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index 46bd08a..b2cee0c 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -429,6 +429,13 @@ blkdebug_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
BDRVBlkdebugState *s = bs->opaque;
BlkdebugRule *rule = NULL;

+ /* Sanity check block layer guarantees */
+ assert(QEMU_IS_ALIGNED(offset, bs->bl.request_alignment));
+ assert(QEMU_IS_ALIGNED(bytes, bs->bl.request_alignment));
+ if (bs->bl.max_transfer) {
+ assert(bytes <= bs->bl.max_transfer);
+ }
+
QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
uint64_t inject_offset = rule->options.inject.offset;

@@ -453,6 +460,13 @@ blkdebug_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
BDRVBlkdebugState *s = bs->opaque;
BlkdebugRule *rule = NULL;

+ /* Sanity check block layer guarantees */
+ assert(QEMU_IS_ALIGNED(offset, bs->bl.request_alignment));
+ assert(QEMU_IS_ALIGNED(bytes, bs->bl.request_alignment));
+ if (bs->bl.max_transfer) {
+ assert(bytes <= bs->bl.max_transfer);
+ }
+
QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
uint64_t inject_offset = rule->options.inject.offset;
--
2.7.4
Michael Roth
2017-08-29 00:14:26 UTC
Permalink
From: Eduardo Habkost <***@redhat.com>

Since the automatic cpuid-level code was introduced in commit
c39c0edf9bb3b968ba95484465a50c7b19f4aa3a ("target-i386: Automatically
set level/xlevel/xlevel2 when needed"), the CPU model tables just define
the default CPUID level code (set using "min-level"). Setting
"[x]level" forces CPUID level to a specific value and disable the
automatic-level logic.

But the PC compat code was not updated and the existing "[x]level"
compat properties broke compatibility for people using features that
triggered the auto-level code. To keep previous behavior, we should set
"min-[x]level" instead of "[x]level" on compat_props.

This was not a problem for most cases, because old machine-types don't
have full-cpuid-auto-level enabled. The only common use case it broke
was the CPUID[7] auto-level code, that was already enabled since the
first CPUID[7] feature was introduced (in QEMU 1.4.0).

This causes the regression reported at:
https://bugzilla.redhat.com/show_bug.cgi?id=1454641

Change the PC compat code to use "min-[x]level" instead of "[x]level" on
compat_props, and add new test cases to ensure we don't break this
again.

Reported-by: "Guo, Zhiyi" <***@redhat.com>
Fixes: c39c0edf9bb ("target-i386: Automatically set level/xlevel/xlevel2 when needed")
Cc: qemu-***@nongnu.org
Acked-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Eduardo Habkost <***@redhat.com>
(cherry picked from commit 1f43571604da85c62f25f3ba6d275b1b5ea76ca2)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
include/hw/i386/pc.h | 42 +++++++++++++++++++++---------------------
tests/test-x86-cpuid-compat.c | 38 ++++++++++++++++++++++++++++++++++++++
2 files changed, 59 insertions(+), 21 deletions(-)

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index f278b3a..564486f 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -558,75 +558,75 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
.value = "off",\
},{\
.driver = "qemu64" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(4),\
},{\
.driver = "kvm64" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(5),\
},{\
.driver = "pentium3" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(2),\
},{\
.driver = "n270" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(5),\
},{\
.driver = "Conroe" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(4),\
},{\
.driver = "Penryn" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(4),\
},{\
.driver = "Nehalem" "-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(4),\
},{\
.driver = "n270" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Penryn" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Conroe" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Nehalem" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Westmere" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "SandyBridge" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "IvyBridge" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Haswell" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Haswell-noTSX" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Broadwell" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = "Broadwell-noTSX" "-" TYPE_X86_CPU,\
- .property = "xlevel",\
+ .property = "min-xlevel",\
.value = stringify(0x8000000a),\
},{\
.driver = TYPE_X86_CPU,\
@@ -852,7 +852,7 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
.value = stringify(2),\
},{\
.driver = "Conroe-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(2),\
},{\
.driver = "Penryn-" TYPE_X86_CPU,\
@@ -860,7 +860,7 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
.value = stringify(2),\
},{\
.driver = "Penryn-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(2),\
},{\
.driver = "Nehalem-" TYPE_X86_CPU,\
@@ -868,7 +868,7 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
.value = stringify(2),\
},{\
.driver = "Nehalem-" TYPE_X86_CPU,\
- .property = "level",\
+ .property = "min-level",\
.value = stringify(2),\
},{\
.driver = "virtio-net-pci",\
diff --git a/tests/test-x86-cpuid-compat.c b/tests/test-x86-cpuid-compat.c
index 79a2e69..b449f54 100644
--- a/tests/test-x86-cpuid-compat.c
+++ b/tests/test-x86-cpuid-compat.c
@@ -228,6 +228,44 @@ int main(int argc, char **argv)
add_cpuid_test("x86/cpuid/auto-xlevel2/pc-2.7",
"-machine pc-i440fx-2.7 -cpu 486,+xstore",
"xlevel2", 0);
+ /*
+ * QEMU 1.4.0 had auto-level enabled for CPUID[7], already,
+ * and the compat code that sets default level shouldn't
+ * disable the auto-level=7 code:
+ */
+ add_cpuid_test("x86/cpuid/auto-level7/pc-i440fx-1.4/off",
+ "-machine pc-i440fx-1.4 -cpu Nehalem",
+ "level", 2);
+ add_cpuid_test("x86/cpuid/auto-level7/pc-i440fx-1.5/on",
+ "-machine pc-i440fx-1.4 -cpu Nehalem,+smap",
+ "level", 7);
+ add_cpuid_test("x86/cpuid/auto-level7/pc-i440fx-2.3/off",
+ "-machine pc-i440fx-2.3 -cpu Penryn",
+ "level", 4);
+ add_cpuid_test("x86/cpuid/auto-level7/pc-i440fx-2.3/on",
+ "-machine pc-i440fx-2.3 -cpu Penryn,+erms",
+ "level", 7);
+ add_cpuid_test("x86/cpuid/auto-level7/pc-i440fx-2.9/off",
+ "-machine pc-i440fx-2.9 -cpu Conroe",
+ "level", 10);
+ add_cpuid_test("x86/cpuid/auto-level7/pc-i440fx-2.9/on",
+ "-machine pc-i440fx-2.9 -cpu Conroe,+erms",
+ "level", 10);
+
+ /*
+ * xlevel doesn't have any feature that triggers auto-level
+ * code on old machine-types. Just check that the compat code
+ * is working correctly:
+ */
+ add_cpuid_test("x86/cpuid/xlevel-compat/pc-i440fx-2.3",
+ "-machine pc-i440fx-2.3 -cpu SandyBridge",
+ "xlevel", 0x8000000a);
+ add_cpuid_test("x86/cpuid/xlevel-compat/pc-i440fx-2.4/npt-off",
+ "-machine pc-i440fx-2.4 -cpu SandyBridge,",
+ "xlevel", 0x80000008);
+ add_cpuid_test("x86/cpuid/xlevel-compat/pc-i440fx-2.4/npt-on",
+ "-machine pc-i440fx-2.4 -cpu SandyBridge,+npt",
+ "xlevel", 0x80000008);

return g_test_run();
}
--
2.7.4
Michael Roth
2017-08-29 00:14:31 UTC
Permalink
From: Max Filippov <***@gmail.com>

Xtensa cores may have registers of types/sizes not supported by the
gdbstub accessors. Ignore writes to such registers and return zero on
read, but always return correct register size, so that gdb on the other
side is able to access all registers in the packet holding unsupported
registers in the middle. This fixes gdb interaction with cores that have
vector/custom TIE registers.

Cc: qemu-***@nongnu.org
Signed-off-by: Max Filippov <***@gmail.com>
(cherry picked from commit dd7b952b793e341c905355581a21cdbaa8b13c31)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
target/xtensa/gdbstub.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/target/xtensa/gdbstub.c b/target/xtensa/gdbstub.c
index fa5469a..da131ae 100644
--- a/target/xtensa/gdbstub.c
+++ b/target/xtensa/gdbstub.c
@@ -58,7 +58,10 @@ int xtensa_cpu_gdb_read_register(CPUState *cs, uint8_t *mem_buf, int n)
case 8:
return gdb_get_reg64(mem_buf, float64_val(env->fregs[i].f64));
default:
- return 0;
+ qemu_log_mask(LOG_UNIMP, "%s from reg %d of unsupported size %d\n",
+ __func__, n, reg->size);
+ memset(mem_buf, 0, reg->size);
+ return reg->size;
}

case 8: /*a*/
@@ -67,6 +70,8 @@ int xtensa_cpu_gdb_read_register(CPUState *cs, uint8_t *mem_buf, int n)
default:
qemu_log_mask(LOG_UNIMP, "%s from reg %d of unsupported type %d\n",
__func__, n, reg->type);
+ memset(mem_buf, 0, reg->size);
+ return reg->size;
return 0;
}
}
@@ -111,7 +116,9 @@ int xtensa_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n)
env->fregs[reg->targno & 0x0f].f64 = make_float64(tmp);
return 8;
default:
- return 0;
+ qemu_log_mask(LOG_UNIMP, "%s to reg %d of unsupported size %d\n",
+ __func__, n, reg->size);
+ return reg->size;
}

case 8: /*a*/
@@ -121,7 +128,7 @@ int xtensa_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n)
default:
qemu_log_mask(LOG_UNIMP, "%s to reg %d of unsupported type %d\n",
__func__, n, reg->type);
- return 0;
+ return reg->size;
}

return 4;
--
2.7.4
Michael Roth
2017-08-29 00:14:38 UTC
Permalink
From: Paolo Bonzini <***@redhat.com>

When attaching the NBD QIOChannel to an AioContext, the TLS channel should
be used, not the underlying socket channel. This is because, trivially,
the TLS channel will be the one that we read/write to and thus the one
that will get the qio_channel_yield() call.

Fixes: ff82911cd3f69f028f2537825c9720ff78bc3f19
Cc: qemu-***@nongnu.org
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Reviewed-by: Daniel P. Berrange <***@redhat.com>
Tested-by: Daniel P. Berrange <***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
(cherry picked from commit 96d06835dc40007673cdfab6322e9042c4077113)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/nbd-client.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/block/nbd-client.c b/block/nbd-client.c
index 1e2952f..56eb0e2 100644
--- a/block/nbd-client.c
+++ b/block/nbd-client.c
@@ -352,14 +352,14 @@ int nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int count)
void nbd_client_detach_aio_context(BlockDriverState *bs)
{
NBDClientSession *client = nbd_get_client_session(bs);
- qio_channel_detach_aio_context(QIO_CHANNEL(client->sioc));
+ qio_channel_detach_aio_context(QIO_CHANNEL(client->ioc));
}

void nbd_client_attach_aio_context(BlockDriverState *bs,
AioContext *new_context)
{
NBDClientSession *client = nbd_get_client_session(bs);
- qio_channel_attach_aio_context(QIO_CHANNEL(client->sioc), new_context);
+ qio_channel_attach_aio_context(QIO_CHANNEL(client->ioc), new_context);
aio_co_schedule(new_context, client->read_reply_co);
}
--
2.7.4
Michael Roth
2017-08-29 00:14:29 UTC
Permalink
From: Laurent Vivier <***@redhat.com>

This allows to manage errors before the memory
has started to be hotplugged. We already have
the function for the CPU cores.

Signed-off-by: Laurent Vivier <***@redhat.com>
Reviewed-by: Greg Kurz <***@kaod.org>
[dwg: Fixed a couple of style nits]
Signed-off-by: David Gibson <***@gibson.dropbear.id.au>

(cherry picked from commit c871bc70bb22d1d70451bc813ecb008fe98cc92b)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/ppc/spapr.c | 41 ++++++++++++++++++++++++++---------------
1 file changed, 26 insertions(+), 15 deletions(-)

diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index 35db949..5564f78 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -2508,20 +2508,6 @@ static void spapr_memory_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
uint64_t align = memory_region_get_alignment(mr);
uint64_t size = memory_region_size(mr);
uint64_t addr;
- char *mem_dev;
-
- if (size % SPAPR_MEMORY_BLOCK_SIZE) {
- error_setg(&local_err, "Hotplugged memory size must be a multiple of "
- "%lld MB", SPAPR_MEMORY_BLOCK_SIZE/M_BYTE);
- goto out;
- }
-
- mem_dev = object_property_get_str(OBJECT(dimm), PC_DIMM_MEMDEV_PROP, NULL);
- if (mem_dev && !kvmppc_is_mem_backend_page_size_ok(mem_dev)) {
- error_setg(&local_err, "Memory backend has bad page size. "
- "Use 'memory-backend-file' with correct mem-path.");
- goto out;
- }

pc_dimm_memory_plug(dev, &ms->hotplug_memory, mr, align, &local_err);
if (local_err) {
@@ -2542,6 +2528,29 @@ out:
error_propagate(errp, local_err);
}

+static void spapr_memory_pre_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
+ Error **errp)
+{
+ PCDIMMDevice *dimm = PC_DIMM(dev);
+ PCDIMMDeviceClass *ddc = PC_DIMM_GET_CLASS(dimm);
+ MemoryRegion *mr = ddc->get_memory_region(dimm);
+ uint64_t size = memory_region_size(mr);
+ char *mem_dev;
+
+ if (size % SPAPR_MEMORY_BLOCK_SIZE) {
+ error_setg(errp, "Hotplugged memory size must be a multiple of "
+ "%lld MB", SPAPR_MEMORY_BLOCK_SIZE / M_BYTE);
+ return;
+ }
+
+ mem_dev = object_property_get_str(OBJECT(dimm), PC_DIMM_MEMDEV_PROP, NULL);
+ if (mem_dev && !kvmppc_is_mem_backend_page_size_ok(mem_dev)) {
+ error_setg(errp, "Memory backend has bad page size. "
+ "Use 'memory-backend-file' with correct mem-path.");
+ return;
+ }
+}
+
typedef struct sPAPRDIMMState {
uint32_t nr_lmbs;
} sPAPRDIMMState;
@@ -2912,7 +2921,9 @@ static void spapr_machine_device_unplug_request(HotplugHandler *hotplug_dev,
static void spapr_machine_device_pre_plug(HotplugHandler *hotplug_dev,
DeviceState *dev, Error **errp)
{
- if (object_dynamic_cast(OBJECT(dev), TYPE_SPAPR_CPU_CORE)) {
+ if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
+ spapr_memory_pre_plug(hotplug_dev, dev, errp);
+ } else if (object_dynamic_cast(OBJECT(dev), TYPE_SPAPR_CPU_CORE)) {
spapr_core_pre_plug(hotplug_dev, dev, errp);
}
}
--
2.7.4
Michael Roth
2017-08-29 00:14:15 UTC
Permalink
From: Eric Blake <***@redhat.com>

In order to test the effects of artificial geometry constraints
on operations like write zero or discard, we first need blkdebug
to manage these actions. It also allows us to inject errors on
those operations, just like we can for read/write/flush.

We can also test the contract promised by the block layer; namely,
if a device has specified limits on alignment or maximum size,
then those limits must be obeyed (for now, the blkdebug driver
merely inherits limits from whatever it is wrapping, but the next
patch will further enhance it to allow specific limit overrides).

This patch intentionally refuses to service requests smaller than
the requested alignments; this is because an upcoming patch adds
a qemu-iotest to prove that the block layer is correctly handling
fragmentation, but the test only works if there is a way to tell
the difference at artificial alignment boundaries when blkdebug is
using a larger-than-default alignment. If we let the blkdebug
layer always defer to the underlying layer, which potentially has
a smaller granularity, the iotest will be thwarted.

Tested by setting up an NBD server with export 'foo', then invoking:
$ ./qemu-io
qemu-io> open -o driver=blkdebug blkdebug::nbd://localhost:10809/foo
qemu-io> d 0 15M
qemu-io> w -z 0 15M

Pre-patch, the server never sees the discard (it was silently
eaten by the block layer); post-patch it is passed across the
wire. Likewise, pre-patch the write is always passed with
NBD_WRITE (with 15M of zeroes on the wire), while post-patch
it can utilize NBD_WRITE_ZEROES (for less traffic).

Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Message-id: 20170429191419.30051-7-***@redhat.com
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit 63188c245013dbe383e8b031e665f813e2452ea5)
* prereq for 81c219a
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 74 insertions(+)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index c5d2edb..a3dc5f6 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -1,6 +1,7 @@
/*
* Block protocol for I/O error injection
*
+ * Copyright (C) 2016-2017 Red Hat, Inc.
* Copyright (c) 2010 Kevin Wolf <***@redhat.com>
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
@@ -382,6 +383,11 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
goto out;
}

+ bs->supported_write_flags = BDRV_REQ_FUA &
+ bs->file->bs->supported_write_flags;
+ bs->supported_zero_flags = (BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP) &
+ bs->file->bs->supported_zero_flags;
+
/* Set request alignment */
align = qemu_opt_get_size(opts, "align", 0);
if (align < INT_MAX && is_power_of_2(align)) {
@@ -494,6 +500,72 @@ static int blkdebug_co_flush(BlockDriverState *bs)
return bdrv_co_flush(bs->file->bs);
}

+static int coroutine_fn blkdebug_co_pwrite_zeroes(BlockDriverState *bs,
+ int64_t offset, int count,
+ BdrvRequestFlags flags)
+{
+ uint32_t align = MAX(bs->bl.request_alignment,
+ bs->bl.pwrite_zeroes_alignment);
+ int err;
+
+ /* Only pass through requests that are larger than requested
+ * preferred alignment (so that we test the fallback to writes on
+ * unaligned portions), and check that the block layer never hands
+ * us anything unaligned that crosses an alignment boundary. */
+ if (count < align) {
+ assert(QEMU_IS_ALIGNED(offset, align) ||
+ QEMU_IS_ALIGNED(offset + count, align) ||
+ DIV_ROUND_UP(offset, align) ==
+ DIV_ROUND_UP(offset + count, align));
+ return -ENOTSUP;
+ }
+ assert(QEMU_IS_ALIGNED(offset, align));
+ assert(QEMU_IS_ALIGNED(count, align));
+ if (bs->bl.max_pwrite_zeroes) {
+ assert(count <= bs->bl.max_pwrite_zeroes);
+ }
+
+ err = rule_check(bs, offset, count);
+ if (err) {
+ return err;
+ }
+
+ return bdrv_co_pwrite_zeroes(bs->file, offset, count, flags);
+}
+
+static int coroutine_fn blkdebug_co_pdiscard(BlockDriverState *bs,
+ int64_t offset, int count)
+{
+ uint32_t align = bs->bl.pdiscard_alignment;
+ int err;
+
+ /* Only pass through requests that are larger than requested
+ * minimum alignment, and ensure that unaligned requests do not
+ * cross optimum discard boundaries. */
+ if (count < bs->bl.request_alignment) {
+ assert(QEMU_IS_ALIGNED(offset, align) ||
+ QEMU_IS_ALIGNED(offset + count, align) ||
+ DIV_ROUND_UP(offset, align) ==
+ DIV_ROUND_UP(offset + count, align));
+ return -ENOTSUP;
+ }
+ assert(QEMU_IS_ALIGNED(offset, bs->bl.request_alignment));
+ assert(QEMU_IS_ALIGNED(count, bs->bl.request_alignment));
+ if (align && count >= align) {
+ assert(QEMU_IS_ALIGNED(offset, align));
+ assert(QEMU_IS_ALIGNED(count, align));
+ }
+ if (bs->bl.max_pdiscard) {
+ assert(count <= bs->bl.max_pdiscard);
+ }
+
+ err = rule_check(bs, offset, count);
+ if (err) {
+ return err;
+ }
+
+ return bdrv_co_pdiscard(bs->file->bs, offset, count);
+}

static void blkdebug_close(BlockDriverState *bs)
{
@@ -748,6 +820,8 @@ static BlockDriver bdrv_blkdebug = {
.bdrv_co_preadv = blkdebug_co_preadv,
.bdrv_co_pwritev = blkdebug_co_pwritev,
.bdrv_co_flush_to_disk = blkdebug_co_flush,
+ .bdrv_co_pwrite_zeroes = blkdebug_co_pwrite_zeroes,
+ .bdrv_co_pdiscard = blkdebug_co_pdiscard,

.bdrv_debug_event = blkdebug_debug_event,
.bdrv_debug_breakpoint = blkdebug_debug_breakpoint,
--
2.7.4
Michael Roth
2017-08-29 00:13:40 UTC
Permalink
From: Max Reitz <***@redhat.com>

It does not make much sense to use a backing image for the target when
you concatenate multiple images (because then there is no correspondence
between the source images' backing files and the target's); but it was
still possible to give one by using -o backing_file=X instead of -B X.

Fix this by moving the check.

(Also, change the error message because -B is not the only way to
specify the backing file, evidently.)

Cc: qemu-stable <qemu-***@nongnu.org>
Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
* applied patch from v1 of series as suggested by author
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
qemu-img.c | 14 +++++++-------
tests/qemu-iotests/122.out | 4 ++--
2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/qemu-img.c b/qemu-img.c
index cfc3bc3..4b2d59a 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -2113,13 +2113,6 @@ static int img_convert(int argc, char **argv)
}


- if (bs_n > 1 && out_baseimg) {
- error_report("-B makes no sense when concatenating multiple input "
- "images");
- ret = -1;
- goto out;
- }
-
src_flags = 0;
ret = bdrv_parse_cache_mode(src_cache, &src_flags, &src_writethrough);
if (ret < 0) {
@@ -2229,6 +2222,13 @@ static int img_convert(int argc, char **argv)
out_baseimg = out_baseimg_param;
}

+ if (bs_n > 1 && out_baseimg) {
+ error_report("Having a backing file for the target makes no sense when "
+ "concatenating multiple input images");
+ ret = -1;
+ goto out;
+ }
+
/* Check if compression is supported */
if (compress) {
bool encryption =
diff --git a/tests/qemu-iotests/122.out b/tests/qemu-iotests/122.out
index 98814de..9317d80 100644
--- a/tests/qemu-iotests/122.out
+++ b/tests/qemu-iotests/122.out
@@ -61,8 +61,8 @@ read 65536/65536 bytes at offset 4194304
64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
read 65536/65536 bytes at offset 8388608
64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-qemu-img: -B makes no sense when concatenating multiple input images
-qemu-img: -B makes no sense when concatenating multiple input images
+qemu-img: Having a backing file for the target makes no sense when concatenating multiple input images
+qemu-img: Having a backing file for the target makes no sense when concatenating multiple input images

=== Compression with misaligned allocations and image sizes ===
--
2.7.4
Michael Roth
2017-08-29 00:14:42 UTC
Permalink
From: Laurent Vivier <***@redhat.com>

since commit 5c4537bd ("spapr: Fix 2.7<->2.8 migration of PCI host bridge"),
some migration fields are forged from the new ones in spapr_pci_pre_save().

It works well, except when the number of MSI devices is 0,
because in this case the function exits immediately.

This fix moves the migration code before the exit code.

The problem can be reproduced with these commands:

source qemu-2.9:

qemu-system-ppc64 -monitor stdio -M pseries-2.6 -nodefaults -S

destination qemu-2.6:

qemu-system-ppc64 -monitor stdio -M pseries-2.6 -nodefaults \
-incoming tcp:0:4444

on the source:

migrate tcp:localhost:4444

Destination fails with the following error:

qemu-system-ppc64: error while loading state for
instance 0x0 of device 'spapr_pci'
qemu-system-ppc64: load of migration failed: Invalid argument

Signed-off-by: Laurent Vivier <***@redhat.com>
Reviewed-by: Greg Kurz <***@kaod.org>
Signed-off-by: David Gibson <***@gibson.dropbear.id.au>
(cherry picked from commit e806b4db1477a1c6bfda7bba28c7f26c47f18e1e)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/ppc/spapr_pci.c | 28 ++++++++++++++--------------
1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 98c52e4..ecfbf01 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -1893,20 +1893,6 @@ static void spapr_pci_pre_save(void *opaque)
gpointer key, value;
int i;

- g_free(sphb->msi_devs);
- sphb->msi_devs = NULL;
- sphb->msi_devs_num = g_hash_table_size(sphb->msi);
- if (!sphb->msi_devs_num) {
- return;
- }
- sphb->msi_devs = g_malloc(sphb->msi_devs_num * sizeof(spapr_pci_msi_mig));
-
- g_hash_table_iter_init(&iter, sphb->msi);
- for (i = 0; g_hash_table_iter_next(&iter, &key, &value); ++i) {
- sphb->msi_devs[i].key = *(uint32_t *) key;
- sphb->msi_devs[i].value = *(spapr_pci_msi *) value;
- }
-
if (sphb->pre_2_8_migration) {
sphb->mig_liobn = sphb->dma_liobn[0];
sphb->mig_mem_win_addr = sphb->mem_win_addr;
@@ -1920,6 +1906,20 @@ static void spapr_pci_pre_save(void *opaque)
sphb->mig_mem_win_size += sphb->mem64_win_size;
}
}
+
+ g_free(sphb->msi_devs);
+ sphb->msi_devs = NULL;
+ sphb->msi_devs_num = g_hash_table_size(sphb->msi);
+ if (!sphb->msi_devs_num) {
+ return;
+ }
+ sphb->msi_devs = g_malloc(sphb->msi_devs_num * sizeof(spapr_pci_msi_mig));
+
+ g_hash_table_iter_init(&iter, sphb->msi);
+ for (i = 0; g_hash_table_iter_next(&iter, &key, &value); ++i) {
+ sphb->msi_devs[i].key = *(uint32_t *) key;
+ sphb->msi_devs[i].value = *(spapr_pci_msi *) value;
+ }
}

static int spapr_pci_post_load(void *opaque, int version_id)
--
2.7.4
Michael Roth
2017-08-29 00:14:33 UTC
Permalink
From: Eric Blake <***@redhat.com>

If a non-NBD client connects to qemu-nbd, we would end up with
a SIGSEGV in nbd_client_put() because we were trying to
unregister the client's association to the export, even though
we skipped inserting the client into that list. Easy trigger
in two terminals:

$ qemu-nbd -p 30001 --format=raw file
$ nmap 127.0.0.1 -p 30001

nmap claims that it thinks it connected to a pago-services1
server (which probably means nmap could be updated to learn the
NBD protocol and give a more accurate diagnosis of the open
port - but that's not our problem), then terminates immediately,
so our call to nbd_negotiate() fails. The fix is to reorder
nbd_co_client_start() to ensure that all initialization occurs
before we ever try talking to a client in nbd_negotiate(), so
that the teardown sequence on negotiation failure doesn't fault
while dereferencing a half-initialized object.

While debugging this, I also noticed that nbd_update_server_watch()
called by nbd_client_closed() was still adding a channel to accept
the next client, even when the state was no longer RUNNING. That
is fixed by making nbd_can_accept() pay attention to the current
state.

Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1451614

Signed-off-by: Eric Blake <***@redhat.com>
Message-Id: <20170527030421.28366-1-***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
(cherry picked from commit df8ad9f128c15aa0a0ebc7b24e9a22c9775b67af)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
nbd/server.c | 8 +++-----
qemu-nbd.c | 2 +-
2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/nbd/server.c b/nbd/server.c
index 924a1fe..edfda84 100644
--- a/nbd/server.c
+++ b/nbd/server.c
@@ -1376,16 +1376,14 @@ static coroutine_fn void nbd_co_client_start(void *opaque)

if (exp) {
nbd_export_get(exp);
+ QTAILQ_INSERT_TAIL(&exp->clients, client, next);
}
+ qemu_co_mutex_init(&client->send_lock);
+
if (nbd_negotiate(data)) {
client_close(client);
goto out;
}
- qemu_co_mutex_init(&client->send_lock);
-
- if (exp) {
- QTAILQ_INSERT_TAIL(&exp->clients, client, next);
- }

nbd_client_receive_next_request(client);

diff --git a/qemu-nbd.c b/qemu-nbd.c
index e4f00e2..14e7947 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -324,7 +324,7 @@ out:

static int nbd_can_accept(void)
{
- return nb_fds < shared;
+ return state == RUNNING && nb_fds < shared;
}

static void nbd_export_closed(NBDExport *exp)
--
2.7.4
Michael Roth
2017-08-29 00:14:41 UTC
Permalink
From: Alexander Graf <***@suse.de>

When resetting the keyboard, we need to reset not just the pending keystrokes,
but also any pending modifiers. Otherwise there's a race when we're getting
reset while running an escape sequence (modifier 0x100).

Cc: qemu-***@nongnu.org
Signed-off-by: Alexander Graf <***@suse.de>
Message-id: 1498117295-162030-1-git-send-email-***@suse.de
Signed-off-by: Gerd Hoffmann <***@redhat.com>
(cherry picked from commit 51dbea77a29ea46173373a6dad4ebd95d4661f42)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/input/hid.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/hw/input/hid.c b/hw/input/hid.c
index fa9cc4c..a0892e8 100644
--- a/hw/input/hid.c
+++ b/hw/input/hid.c
@@ -483,6 +483,7 @@ void hid_reset(HIDState *hs)
memset(hs->kbd.keycodes, 0, sizeof(hs->kbd.keycodes));
memset(hs->kbd.key, 0, sizeof(hs->kbd.key));
hs->kbd.keys = 0;
+ hs->kbd.modifiers = 0;
break;
case HID_MOUSE:
case HID_TABLET:
--
2.7.4
Michael Roth
2017-08-29 00:14:20 UTC
Permalink
From: Eric Blake <***@redhat.com>

We document that *file is valid if the return is not an error and
includes BDRV_BLOCK_OFFSET_VALID, but forgot to obey this contract
when a driver (such as blkdebug) lacks a callback. Messed up in
commit 67a0fd2 (v2.6), when we added the file parameter.

Enhance qemu-iotest 177 to cover this, using a sequence that would
print garbage or even SEGV, because it was dererefencing through
uninitialized memory. [The resulting test output shows that we
have less-than-ideal block status from the blkdebug driver, but
that's a separate fix coming up soon.]

Setting *file on all paths that return BDRV_BLOCK_OFFSET_VALID is
enough to fix the crash, but we can go one step further: always
setting *file, even on error, means that a broken caller that
blindly dereferences file without checking for error is now more
likely to get a reliable SEGV instead of randomly acting on garbage,
making it easier to diagnose such buggy callers. Adding an
assertion that file is set where expected doesn't hurt either.

CC: qemu-***@nongnu.org
Signed-off-by: Eric Blake <***@redhat.com>
Reviewed-by: Fam Zheng <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Reviewed-by: John Snow <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 81c219ac6ce0d6182e35f3976f2caa4cefcaf9f0)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/io.c | 5 +++--
tests/qemu-iotests/177 | 3 +++
tests/qemu-iotests/177.out | 2 ++
3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/block/io.c b/block/io.c
index fe0c867..a8589ee 100644
--- a/block/io.c
+++ b/block/io.c
@@ -1757,6 +1757,7 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
int64_t n;
int64_t ret, ret2;

+ *file = NULL;
total_sectors = bdrv_nb_sectors(bs);
if (total_sectors < 0) {
return total_sectors;
@@ -1777,11 +1778,11 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
ret = BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED;
if (bs->drv->protocol_name) {
ret |= BDRV_BLOCK_OFFSET_VALID | (sector_num * BDRV_SECTOR_SIZE);
+ *file = bs;
}
return ret;
}

- *file = NULL;
bdrv_inc_in_flight(bs);
ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, pnum,
file);
@@ -1791,7 +1792,7 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
}

if (ret & BDRV_BLOCK_RAW) {
- assert(ret & BDRV_BLOCK_OFFSET_VALID);
+ assert(ret & BDRV_BLOCK_OFFSET_VALID && *file);
ret = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS,
*pnum, pnum, file);
goto out;
diff --git a/tests/qemu-iotests/177 b/tests/qemu-iotests/177
index 2005c17..f8ed8fb 100755
--- a/tests/qemu-iotests/177
+++ b/tests/qemu-iotests/177
@@ -43,6 +43,7 @@ _supported_proto file
CLUSTER_SIZE=1M
size=128M
options=driver=blkdebug,image.driver=qcow2
+nested_opts=image.file.driver=file,image.file.filename=$TEST_IMG

echo
echo "== setting up files =="
@@ -106,6 +107,8 @@ function verify_io()
}

verify_io | $QEMU_IO -r "$TEST_IMG" | _filter_qemu_io
+$QEMU_IMG map --image-opts "$options,$nested_opts,align=4k" \
+ | _filter_qemu_img_map

_check_test_img

diff --git a/tests/qemu-iotests/177.out b/tests/qemu-iotests/177.out
index e887542..fcfbfa3 100644
--- a/tests/qemu-iotests/177.out
+++ b/tests/qemu-iotests/177.out
@@ -45,5 +45,7 @@ read 30408704/30408704 bytes at offset 80740352
29 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
read 23068672/23068672 bytes at offset 111149056
22 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+Offset Length File
+0 0x8000000 json:{"image": {"driver": "IMGFMT", "file": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}}, "driver": "blkdebug", "align": "4k"}
No errors were found on the image.
*** done
--
2.7.4
Michael Roth
2017-08-29 00:14:39 UTC
Permalink
From: Max Reitz <***@redhat.com>

uri_parse(...)->scheme may be NULL. In fact, probably every field may be
NULL, and the callers do test this for all of the other fields but not
for scheme (except for block/gluster.c; block/vxhs.c does not access
that field at all).

We can easily fix this by using g_strcmp0() instead of strcmp().

Cc: qemu-***@nongnu.org
Signed-off-by: Max Reitz <***@redhat.com>
Message-id: 20170613205726.13544-1-***@redhat.com
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit f69165a8feca055cf4a37d13ab0fc5beec3cb372)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/nbd.c | 6 +++---
block/nfs.c | 2 +-
block/sheepdog.c | 6 +++---
block/ssh.c | 2 +-
4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/block/nbd.c b/block/nbd.c
index b3545f5..9a54edf 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -65,11 +65,11 @@ static int nbd_parse_uri(const char *filename, QDict *options)
}

/* transport */
- if (!strcmp(uri->scheme, "nbd")) {
+ if (!g_strcmp0(uri->scheme, "nbd")) {
is_unix = false;
- } else if (!strcmp(uri->scheme, "nbd+tcp")) {
+ } else if (!g_strcmp0(uri->scheme, "nbd+tcp")) {
is_unix = false;
- } else if (!strcmp(uri->scheme, "nbd+unix")) {
+ } else if (!g_strcmp0(uri->scheme, "nbd+unix")) {
is_unix = true;
} else {
ret = -EINVAL;
diff --git a/block/nfs.c b/block/nfs.c
index bfeebc1..344186f 100644
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -83,7 +83,7 @@ static int nfs_parse_uri(const char *filename, QDict *options, Error **errp)
error_setg(errp, "Invalid URI specified");
goto out;
}
- if (strcmp(uri->scheme, "nfs") != 0) {
+ if (g_strcmp0(uri->scheme, "nfs") != 0) {
error_setg(errp, "URI scheme must be 'nfs'");
goto out;
}
diff --git a/block/sheepdog.c b/block/sheepdog.c
index fb9203e..2d8d8c8 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -1052,11 +1052,11 @@ static void sd_parse_uri(SheepdogConfig *cfg, const char *filename,
}

/* transport */
- if (!strcmp(uri->scheme, "sheepdog")) {
+ if (!g_strcmp0(uri->scheme, "sheepdog")) {
is_unix = false;
- } else if (!strcmp(uri->scheme, "sheepdog+tcp")) {
+ } else if (!g_strcmp0(uri->scheme, "sheepdog+tcp")) {
is_unix = false;
- } else if (!strcmp(uri->scheme, "sheepdog+unix")) {
+ } else if (!g_strcmp0(uri->scheme, "sheepdog+unix")) {
is_unix = true;
} else {
error_setg(&err, "URI scheme must be 'sheepdog', 'sheepdog+tcp',"
diff --git a/block/ssh.c b/block/ssh.c
index 34a2f79..139b05d 100644
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -205,7 +205,7 @@ static int parse_uri(const char *filename, QDict *options, Error **errp)
return -EINVAL;
}

- if (strcmp(uri->scheme, "ssh") != 0) {
+ if (g_strcmp0(uri->scheme, "ssh") != 0) {
error_setg(errp, "URI scheme must be 'ssh'");
goto err;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:36 UTC
Permalink
From: Max Reitz <***@redhat.com>

The bs->exact_filename field may not be sufficient to store the full
blkdebug node filename. In this case, we should not generate a filename
at all instead of an unusable one.

Cc: qemu-***@nongnu.org
Reported-by: Qu Wenruo <***@cn.fujitsu.com>
Signed-off-by: Max Reitz <***@redhat.com>
Message-id: 20170613172006.19685-2-***@redhat.com
Reviewed-by: Alberto Garcia <***@igalia.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit de81d72d3d13a19edf4d461be3b0f5a877be0234)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkdebug.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index 5ccb9ce..63bc5d4 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -840,9 +840,13 @@ static void blkdebug_refresh_filename(BlockDriverState *bs, QDict *options)
}

if (!force_json && bs->file->bs->exact_filename[0]) {
- snprintf(bs->exact_filename, sizeof(bs->exact_filename),
- "blkdebug:%s:%s", s->config_file ?: "",
- bs->file->bs->exact_filename);
+ int ret = snprintf(bs->exact_filename, sizeof(bs->exact_filename),
+ "blkdebug:%s:%s", s->config_file ?: "",
+ bs->file->bs->exact_filename);
+ if (ret >= sizeof(bs->exact_filename)) {
+ /* An overflow makes the filename unusable, so do not report any */
+ bs->exact_filename[0] = 0;
+ }
}

opts = qdict_new();
--
2.7.4
Michael Roth
2017-08-29 00:14:35 UTC
Permalink
From: Kevin Wolf <***@redhat.com>

commit_complete() can't assume that after its block_job_completed() the
job is actually immediately freed; someone else may still be holding
references. In this case, the op blockers on the intermediate nodes make
the graph reconfiguration in the completion code fail.

Call block_job_remove_all_bdrv() manually so that we know for sure that
any blockers on intermediate nodes are given up.

Cc: qemu-***@nongnu.org
Signed-off-by: Kevin Wolf <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
(cherry picked from commit 4f78a16fee462471416dc49b409d57b2071cf3d9)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/commit.c | 7 +++++++
1 file changed, 7 insertions(+)

diff --git a/block/commit.c b/block/commit.c
index 3bae46e..fba25e2 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -119,6 +119,13 @@ static void commit_complete(BlockJob *job, void *opaque)
}
g_free(s->backing_file_str);
blk_unref(s->top);
+
+ /* If there is more than one reference to the job (e.g. if called from
+ * block_job_finish_sync()), block_job_completed() won't free it and
+ * therefore the blockers on the intermediate nodes remain. This would
+ * cause bdrv_set_backing_hd() to fail. */
+ block_job_remove_all_bdrv(job);
+
block_job_completed(&s->common, ret);
g_free(data);
--
2.7.4
Michael Roth
2017-08-29 00:14:40 UTC
Permalink
From: Bruce Rogers <***@suse.com>

Commit a0e640a8 introduced a path processing error.
Pass fstatat the dirpath based path component instead
of the entire path.

Signed-off-by: Bruce Rogers <***@suse.com>
Signed-off-by: Greg Kurz <***@kaod.org>
(cherry picked from commit 790db7efdbe1536acf1c4f4f95a0316dbda59433)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/9pfs/9p-local.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/9pfs/9p-local.c b/hw/9pfs/9p-local.c
index 7a0c383..b13ff21 100644
--- a/hw/9pfs/9p-local.c
+++ b/hw/9pfs/9p-local.c
@@ -1009,7 +1009,7 @@ static int local_remove(FsContext *ctx, const char *path)
goto out;
}

- if (fstatat(dirfd, path, &stbuf, AT_SYMLINK_NOFOLLOW) < 0) {
+ if (fstatat(dirfd, name, &stbuf, AT_SYMLINK_NOFOLLOW) < 0) {
goto err_out;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:44 UTC
Permalink
From: Kevin Wolf <***@redhat.com>

I can't see how overlay_bs could become NULL with the current code, but
other code in this function already checks it and we can make Coverity
happy with this check, so let's add it.

Cc: qemu-***@nongnu.org
Signed-off-by: Kevin Wolf <***@redhat.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
(cherry picked from commit b1e1fa0c3afc7f671fbc24645bdf67949a5657e5)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/commit.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/block/commit.c b/block/commit.c
index fba25e2..66e3418 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -91,7 +91,9 @@ static void commit_complete(BlockJob *job, void *opaque)

/* Make sure overlay_bs and top stay around until bdrv_set_backing_hd() */
bdrv_ref(top);
- bdrv_ref(overlay_bs);
+ if (overlay_bs) {
+ bdrv_ref(overlay_bs);
+ }

/* Remove base node parent that still uses BLK_PERM_WRITE/RESIZE before
* the normal backing chain can be restored. */
--
2.7.4
Michael Roth
2017-08-29 00:14:43 UTC
Permalink
From: Jason Wang <***@redhat.com>

After converting to use DMA api for virtio devices, we should use
dma_as instead of address_space_memory. Otherwise it won't work if
IOMMU is enabled.

Fixes: commit 8607f5c3072c ("virtio: convert to use DMA api")
Cc: qemu-***@nongnu.org
Signed-off-by: Jason Wang <***@redhat.com>
Message-Id: <1499170866-9068-1-git-send-email-***@redhat.com>
Signed-off-by: Paolo Bonzini <***@redhat.com>
(cherry picked from commit 025bdeab3c163aee9604a60b2332a5fcbcc00f8d)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/scsi/virtio-scsi.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index cb1c123..3915c2e 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -43,12 +43,13 @@ static inline SCSIDevice *virtio_scsi_device_find(VirtIOSCSI *s, uint8_t *lun)

void virtio_scsi_init_req(VirtIOSCSI *s, VirtQueue *vq, VirtIOSCSIReq *req)
{
+ VirtIODevice *vdev = VIRTIO_DEVICE(s);
const size_t zero_skip =
offsetof(VirtIOSCSIReq, resp_iov) + sizeof(req->resp_iov);

req->vq = vq;
req->dev = s;
- qemu_sglist_init(&req->qsgl, DEVICE(s), 8, &address_space_memory);
+ qemu_sglist_init(&req->qsgl, DEVICE(s), 8, vdev->dma_as);
qemu_iovec_init(&req->resp_iov, 1);
memset((uint8_t *)req + zero_skip, 0, sizeof(*req) - zero_skip);
}
--
2.7.4
Michael Roth
2017-08-29 00:14:45 UTC
Permalink
From: Greg Kurz <***@kaod.org>

In case of error, we must ensure the dynamically allocated base_core_type
is freed, like it is done everywhere else in this function.

This is a regression introduced in QEMU 2.9 by commit 8149e2992f78.

Signed-off-by: Greg Kurz <***@kaod.org>
Signed-off-by: David Gibson <***@gibson.dropbear.id.au>
(cherry picked from commit df8658de43db242ea82183d75cc957c2b0fa013a)
Conflicts:
hw/ppc/spapr.c
* fix context dep on 459264ef2
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/ppc/spapr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index 58c15ef..bda3854 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -2805,7 +2805,7 @@ static void spapr_core_pre_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
if (cc->nr_threads != smp_threads) {
error_setg(errp, "invalid nr-threads %d, must be %d",
cc->nr_threads, smp_threads);
- return;
+ goto out;
}

core_slot = spapr_find_cpu_slot(MACHINE(hotplug_dev), cc->core_id, &index);
--
2.7.4
Michael Roth
2017-08-29 00:14:48 UTC
Permalink
From: Alexander Graf <***@suse.de>

Delays in the input layer are special cased input events. Every input
event is accounted for in a global intput queue count. The special cased
delays however did not get removed from the queue, leading to queue overruns
and thus silent key drops after typing quite a few characters.

Signed-off-by: Alexander Graf <***@suse.de>
Message-id: 1498117318-162102-1-git-send-email-***@suse.de
Fixes: be1a7176 ("input: add support for kbd delays")
Cc: qemu-***@nongnu.org
Signed-off-by: Gerd Hoffmann <***@redhat.com>
(cherry picked from commit 77b0359bf414ad666d1714dc9888f1017c08e283)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
ui/input.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/ui/input.c b/ui/input.c
index fb1f404..94ba3d5 100644
--- a/ui/input.c
+++ b/ui/input.c
@@ -252,6 +252,7 @@ static void qemu_input_queue_process(void *opaque)
item = QTAILQ_FIRST(queue);
g_assert(item->type == QEMU_INPUT_QUEUE_DELAY);
QTAILQ_REMOVE(queue, item, node);
+ queue_count--;
g_free(item);

while (!QTAILQ_EMPTY(queue)) {
--
2.7.4
Michael Roth
2017-08-29 00:14:37 UTC
Permalink
From: Max Reitz <***@redhat.com>

The bs->exact_filename field may not be sufficient to store the full
blkverify node filename. In this case, we should not generate a filename
at all instead of an unusable one.

Cc: qemu-***@nongnu.org
Reported-by: Qu Wenruo <***@cn.fujitsu.com>
Signed-off-by: Max Reitz <***@redhat.com>
Message-id: 20170613172006.19685-3-***@redhat.com
Reviewed-by: Alberto Garcia <***@igalia.com>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Signed-off-by: Max Reitz <***@redhat.com>
(cherry picked from commit 05cc758a3dfc79488d0a8eb7f5830a41871e78d0)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/blkverify.c | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/block/blkverify.c b/block/blkverify.c
index 6b0a603..06369f9 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -301,10 +301,14 @@ static void blkverify_refresh_filename(BlockDriverState *bs, QDict *options)
if (bs->file->bs->exact_filename[0]
&& s->test_file->bs->exact_filename[0])
{
- snprintf(bs->exact_filename, sizeof(bs->exact_filename),
- "blkverify:%s:%s",
- bs->file->bs->exact_filename,
- s->test_file->bs->exact_filename);
+ int ret = snprintf(bs->exact_filename, sizeof(bs->exact_filename),
+ "blkverify:%s:%s",
+ bs->file->bs->exact_filename,
+ s->test_file->bs->exact_filename);
+ if (ret >= sizeof(bs->exact_filename)) {
+ /* An overflow makes the filename unusable, so do not report any */
+ bs->exact_filename[0] = 0;
+ }
}
}
--
2.7.4
Michael Roth
2017-08-29 00:14:46 UTC
Permalink
From: Jason Wang <***@redhat.com>

Spec said offloads should be le64, so use virtio_ldq_p() to guarantee
valid endian.

Fixes: 644c98587d4c ("virtio-net: dynamic network offloads configuration")
Cc: qemu-***@nongnu.org
Cc: Dmitry Fleytman <***@redhat.com>
Signed-off-by: Jason Wang <***@redhat.com>
(cherry picked from commit 189ae6bb5ce1f5a322f8691d00fe942ba43dd601)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/net/virtio-net.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 98bd683..7695f75 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -723,6 +723,8 @@ static int virtio_net_handle_offloads(VirtIONet *n, uint8_t cmd,
if (cmd == VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET) {
uint64_t supported_offloads;

+ offloads = virtio_ldq_p(vdev, &offloads);
+
if (!n->has_vnet_hdr) {
return VIRTIO_NET_ERR;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:47 UTC
Permalink
From: Gerd Hoffmann <***@redhat.com>

Apply a limit to the number of items we accept into the keyboard queue.

Impact: Without this limit vnc clients can exhaust host memory by
sending keyboard events faster than qemu feeds them to the guest.

Fixes: CVE-2017-8379
Cc: P J P <***@redhat.com>
Cc: Huawei PSIRT <***@huawei.com>
Reported-by: ***@huawei.com
Signed-off-by: Gerd Hoffmann <***@redhat.com>
Message-id: 20170428084237.23960-1-***@redhat.com
(cherry picked from commit fa18f36a461984eae50ab957e47ec78dae3c14fc)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
ui/input.c | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/ui/input.c b/ui/input.c
index ed88cda..fb1f404 100644
--- a/ui/input.c
+++ b/ui/input.c
@@ -41,6 +41,8 @@ static QTAILQ_HEAD(QemuInputEventQueueHead, QemuInputEventQueue) kbd_queue =
QTAILQ_HEAD_INITIALIZER(kbd_queue);
static QEMUTimer *kbd_timer;
static uint32_t kbd_default_delay_ms = 10;
+static uint32_t queue_count;
+static uint32_t queue_limit = 1024;

QemuInputHandlerState *qemu_input_handler_register(DeviceState *dev,
QemuInputHandler *handler)
@@ -268,6 +270,7 @@ static void qemu_input_queue_process(void *opaque)
break;
}
QTAILQ_REMOVE(queue, item, node);
+ queue_count--;
g_free(item);
}
}
@@ -282,6 +285,7 @@ static void qemu_input_queue_delay(struct QemuInputEventQueueHead *queue,
item->delay_ms = delay_ms;
item->timer = timer;
QTAILQ_INSERT_TAIL(queue, item, node);
+ queue_count++;

if (start_timer) {
timer_mod(item->timer, qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL)
@@ -298,6 +302,7 @@ static void qemu_input_queue_event(struct QemuInputEventQueueHead *queue,
item->src = src;
item->evt = evt;
QTAILQ_INSERT_TAIL(queue, item, node);
+ queue_count++;
}

static void qemu_input_queue_sync(struct QemuInputEventQueueHead *queue)
@@ -306,6 +311,7 @@ static void qemu_input_queue_sync(struct QemuInputEventQueueHead *queue)

item->type = QEMU_INPUT_QUEUE_SYNC;
QTAILQ_INSERT_TAIL(queue, item, node);
+ queue_count++;
}

void qemu_input_event_send_impl(QemuConsole *src, InputEvent *evt)
@@ -381,7 +387,7 @@ void qemu_input_event_send_key(QemuConsole *src, KeyValue *key, bool down)
qemu_input_event_send(src, evt);
qemu_input_event_sync();
qapi_free_InputEvent(evt);
- } else {
+ } else if (queue_count < queue_limit) {
qemu_input_queue_event(&kbd_queue, src, evt);
qemu_input_queue_sync(&kbd_queue);
}
@@ -409,8 +415,10 @@ void qemu_input_event_send_key_delay(uint32_t delay_ms)
kbd_timer = timer_new_ms(QEMU_CLOCK_VIRTUAL, qemu_input_queue_process,
&kbd_queue);
}
- qemu_input_queue_delay(&kbd_queue, kbd_timer,
- delay_ms ? delay_ms : kbd_default_delay_ms);
+ if (queue_count < queue_limit) {
+ qemu_input_queue_delay(&kbd_queue, kbd_timer,
+ delay_ms ? delay_ms : kbd_default_delay_ms);
+ }
}

InputEvent *qemu_input_event_new_btn(InputButton btn, bool down)
--
2.7.4
Michael Roth
2017-08-29 00:14:54 UTC
Permalink
From: Greg Kurz <***@kaod.org>

This function has to ensure it doesn't follow a symlink that could be used
to escape the virtfs directory. This could be easily achieved if fchmodat()
on linux honored the AT_SYMLINK_NOFOLLOW flag as described in POSIX, but
it doesn't. There was a tentative to implement a new fchmodat2() syscall
with the correct semantics:

https://patchwork.kernel.org/patch/9596301/

but it didn't gain much momentum. Also it was suggested to look at an O_PATH
based solution in the first place.

The current implementation covers most use-cases, but it notably fails if:
- the target path has access rights equal to 0000 (openat() returns EPERM),
=> once you've done chmod(0000) on a file, you can never chmod() again
- the target path is UNIX domain socket (openat() returns ENXIO)
=> bind() of UNIX domain sockets fails if the file is on 9pfs

The solution is to use O_PATH: openat() now succeeds in both cases, and we
can ensure the path isn't a symlink with fstat(). The associated entry in
"/proc/self/fd" can hence be safely passed to the regular chmod() syscall.

The previous behavior is kept for older systems that don't have O_PATH.

Signed-off-by: Greg Kurz <***@kaod.org>
Reviewed-by: Eric Blake <***@redhat.com>
Tested-by: Zhi Yong Wu <***@ucloud.cn>
Acked-by: Philippe Mathieu-Daudé <***@amsat.org>
(cherry picked from commit 4751fd5328dfcd4fe2f9055728a72a0e3ae56512)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/9pfs/9p-local.c | 42 +++++++++++++++++++++++++++++++++++-------
hw/9pfs/9p-util.h | 24 +++++++++++++++---------
2 files changed, 50 insertions(+), 16 deletions(-)

diff --git a/hw/9pfs/9p-local.c b/hw/9pfs/9p-local.c
index b13ff21..a83dbfb 100644
--- a/hw/9pfs/9p-local.c
+++ b/hw/9pfs/9p-local.c
@@ -278,17 +278,27 @@ update_map_file:

static int fchmodat_nofollow(int dirfd, const char *name, mode_t mode)
{
+ struct stat stbuf;
int fd, ret;

/* FIXME: this should be handled with fchmodat(AT_SYMLINK_NOFOLLOW).
- * Unfortunately, the linux kernel doesn't implement it yet. As an
- * alternative, let's open the file and use fchmod() instead. This
- * may fail depending on the permissions of the file, but it is the
- * best we can do to avoid TOCTTOU. We first try to open read-only
- * in case name points to a directory. If that fails, we try write-only
- * in case name doesn't point to a directory.
+ * Unfortunately, the linux kernel doesn't implement it yet.
*/
- fd = openat_file(dirfd, name, O_RDONLY, 0);
+
+ /* First, we clear non-racing symlinks out of the way. */
+ if (fstatat(dirfd, name, &stbuf, AT_SYMLINK_NOFOLLOW)) {
+ return -1;
+ }
+ if (S_ISLNK(stbuf.st_mode)) {
+ errno = ELOOP;
+ return -1;
+ }
+
+ /* Access modes are ignored when O_PATH is supported. We try O_RDONLY and
+ * O_WRONLY for old-systems that don't support O_PATH.
+ */
+ fd = openat_file(dirfd, name, O_RDONLY | O_PATH_9P_UTIL, 0);
+#if O_PATH_9P_UTIL == 0
if (fd == -1) {
/* In case the file is writable-only and isn't a directory. */
if (errno == EACCES) {
@@ -302,6 +312,24 @@ static int fchmodat_nofollow(int dirfd, const char *name, mode_t mode)
return -1;
}
ret = fchmod(fd, mode);
+#else
+ if (fd == -1) {
+ return -1;
+ }
+
+ /* Now we handle racing symlinks. */
+ ret = fstat(fd, &stbuf);
+ if (!ret) {
+ if (S_ISLNK(stbuf.st_mode)) {
+ errno = ELOOP;
+ ret = -1;
+ } else {
+ char *proc_path = g_strdup_printf("/proc/self/fd/%d", fd);
+ ret = chmod(proc_path, mode);
+ g_free(proc_path);
+ }
+ }
+#endif
close_preserve_errno(fd);
return ret;
}
diff --git a/hw/9pfs/9p-util.h b/hw/9pfs/9p-util.h
index 517027c..033e7e6 100644
--- a/hw/9pfs/9p-util.h
+++ b/hw/9pfs/9p-util.h
@@ -13,6 +13,12 @@
#ifndef QEMU_9P_UTIL_H
#define QEMU_9P_UTIL_H

+#ifdef O_PATH
+#define O_PATH_9P_UTIL O_PATH
+#else
+#define O_PATH_9P_UTIL 0
+#endif
+
static inline void close_preserve_errno(int fd)
{
int serrno = errno;
@@ -22,13 +28,8 @@ static inline void close_preserve_errno(int fd)

static inline int openat_dir(int dirfd, const char *name)
{
-#ifdef O_PATH
-#define OPENAT_DIR_O_PATH O_PATH
-#else
-#define OPENAT_DIR_O_PATH 0
-#endif
return openat(dirfd, name,
- O_DIRECTORY | O_RDONLY | O_NOFOLLOW | OPENAT_DIR_O_PATH);
+ O_DIRECTORY | O_RDONLY | O_NOFOLLOW | O_PATH_9P_UTIL);
}

static inline int openat_file(int dirfd, const char *name, int flags,
@@ -43,9 +44,14 @@ static inline int openat_file(int dirfd, const char *name, int flags,
}

serrno = errno;
- /* O_NONBLOCK was only needed to open the file. Let's drop it. */
- ret = fcntl(fd, F_SETFL, flags);
- assert(!ret);
+ /* O_NONBLOCK was only needed to open the file. Let's drop it. We don't
+ * do that with O_PATH since fcntl(F_SETFL) isn't supported, and openat()
+ * ignored it anyway.
+ */
+ if (!(flags & O_PATH_9P_UTIL)) {
+ ret = fcntl(fd, F_SETFL, flags);
+ assert(!ret);
+ }
errno = serrno;
return fd;
}
--
2.7.4
Michael Roth
2017-08-29 00:13:41 UTC
Permalink
From: Max Reitz <***@redhat.com>

Reproducer:
$ ./qemu-img info ''
qemu-img: ./block.c:1008: bdrv_open_driver: Assertion
`!drv->bdrv_needs_filename || bs->filename[0]' failed.
[1] 26105 abort (core dumped) ./qemu-img info ''

This patch fixes this to be:
$ ./qemu-img info ''
qemu-img: Could not open '': The 'file' block driver requires a file
name

Cc: qemu-stable <qemu-***@nongnu.org>
Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Fam Zheng <***@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <***@amsat.org>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 4a0082401a770261b85625a41eef4a4e89ad7a74)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/block.c b/block.c
index 1fbbb8d..46da908 100644
--- a/block.c
+++ b/block.c
@@ -1167,7 +1167,7 @@ static int bdrv_open_common(BlockDriverState *bs, BlockBackend *file,
filename = qdict_get_try_str(options, "filename");
}

- if (drv->bdrv_needs_filename && !filename) {
+ if (drv->bdrv_needs_filename && (!filename || !filename[0])) {
error_setg(errp, "The '%s' block driver requires a file name",
drv->format_name);
ret = -EINVAL;
--
2.7.4
Michael Roth
2017-08-29 00:14:49 UTC
Permalink
From: Kevin Wolf <***@redhat.com>

Signed-off-by: Kevin Wolf <***@redhat.com>
Reviewed-by: John Snow <***@redhat.com>
(cherry picked from commit c3971b883a596abc6af45f53d2f43fb2f59ccd3b)
*prereq for d3c8c674
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
tests/qemu-iotests/040 | 35 +++++++++++++++++++++++++++++++++--
tests/qemu-iotests/040.out | 4 ++--
2 files changed, 35 insertions(+), 4 deletions(-)

diff --git a/tests/qemu-iotests/040 b/tests/qemu-iotests/040
index 5bdaf3d..9d381d9 100755
--- a/tests/qemu-iotests/040
+++ b/tests/qemu-iotests/040
@@ -70,7 +70,9 @@ class ImageCommitTestCase(iotests.QMPTestCase):
self.wait_for_complete()

class TestSingleDrive(ImageCommitTestCase):
- image_len = 1 * 1024 * 1024
+ # Need some space after the copied data so that throttling is effective in
+ # tests that use it rather than just completing the job immediately
+ image_len = 2 * 1024 * 1024
test_len = 1 * 1024 * 256

def setUp(self):
@@ -79,7 +81,9 @@ class TestSingleDrive(ImageCommitTestCase):
qemu_img('create', '-f', iotests.imgfmt, '-o', 'backing_file=%s' % mid_img, test_img)
qemu_io('-f', 'raw', '-c', 'write -P 0xab 0 524288', backing_img)
qemu_io('-f', iotests.imgfmt, '-c', 'write -P 0xef 524288 524288', mid_img)
- self.vm = iotests.VM().add_drive(test_img)
+ self.vm = iotests.VM().add_drive(test_img, interface="none")
+ self.vm.add_device("virtio-scsi-pci")
+ self.vm.add_device("scsi-hd,id=scsi0,drive=drive0")
self.vm.launch()

def tearDown(self):
@@ -131,6 +135,33 @@ class TestSingleDrive(ImageCommitTestCase):
self.assert_qmp(result, 'error/class', 'GenericError')
self.assert_qmp(result, 'error/desc', 'Base \'%s\' not found' % mid_img)

+ # When the job is running on a BB that is automatically deleted on hot
+ # unplug, the job is cancelled when the device disappears
+ def test_hot_unplug(self):
+ if self.image_len == 0:
+ return
+
+ self.assert_no_active_block_jobs()
+ result = self.vm.qmp('block-commit', device='drive0', top=mid_img,
+ base=backing_img, speed=(self.image_len / 4))
+ self.assert_qmp(result, 'return', {})
+ result = self.vm.qmp('device_del', id='scsi0')
+ self.assert_qmp(result, 'return', {})
+
+ cancelled = False
+ deleted = False
+ while not cancelled or not deleted:
+ for event in self.vm.get_qmp_events(wait=True):
+ if event['event'] == 'DEVICE_DELETED':
+ self.assert_qmp(event, 'data/device', 'scsi0')
+ deleted = True
+ elif event['event'] == 'BLOCK_JOB_CANCELLED':
+ self.assert_qmp(event, 'data/device', 'drive0')
+ cancelled = True
+ else:
+ self.fail("Unexpected event %s" % (event['event']))
+
+ self.assert_no_active_block_jobs()

class TestRelativePaths(ImageCommitTestCase):
image_len = 1 * 1024 * 1024
diff --git a/tests/qemu-iotests/040.out b/tests/qemu-iotests/040.out
index 4fd1c2d..6d9bee1 100644
--- a/tests/qemu-iotests/040.out
+++ b/tests/qemu-iotests/040.out
@@ -1,5 +1,5 @@
-.........................
+...........................
----------------------------------------------------------------------
-Ran 25 tests
+Ran 27 tests

OK
--
2.7.4
Michael Roth
2017-08-29 00:14:51 UTC
Permalink
From: Laurent Vivier <***@redhat.com>

With pseries machine type a negative core-id is not managed properly:
-1 gives an inaccurate error message ("core -1 already populated"),
-2 crashes QEMU (core dump)

As it seems a negative value is invalid for any architecture,
instead of checking this in spapr_core_pre_plug() I think it's better
to check this in the generic part, core_prop_set_core_id()

Signed-off-by: Laurent Vivier <***@redhat.com>
Message-Id: <20170802103259.25940-1-***@redhat.com>
Reviewed-by: Greg Kurz <***@kaod.org>
Reviewed-by: David Gibson <***@gibson.dropbear.id.au>
Reviewed-by: Eduardo Habkost <***@redhat.com>
Signed-off-by: Eduardo Habkost <***@redhat.com>
(cherry picked from commit be2960baae07e5257cde8c814cbd91647e235147)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/cpu/core.c | 5 +++++
1 file changed, 5 insertions(+)

diff --git a/hw/cpu/core.c b/hw/cpu/core.c
index 2bf960d..bd578ab 100644
--- a/hw/cpu/core.c
+++ b/hw/cpu/core.c
@@ -33,6 +33,11 @@ static void core_prop_set_core_id(Object *obj, Visitor *v, const char *name,
return;
}

+ if (value < 0) {
+ error_setg(errp, "Invalid core id %"PRId64, value);
+ return;
+ }
+
core->core_id = value;
}
--
2.7.4
Michael Roth
2017-08-29 00:14:50 UTC
Permalink
From: Kevin Wolf <***@redhat.com>

Commits 0db832f and 6cdbceb introduced the automatic insertion of filter
nodes above the top layer of mirror and commit block jobs. The
assumption made there was that since libvirt doesn't do node-level
management of the block layer yet, it shouldn't be affected by added
nodes.

This is true as far as commands issued by libvirt are concerned. It only
uses BlockBackend names to address nodes, so any operations it performs
still operate on the root of the tree as intended.

However, the assumption breaks down when you consider query commands,
which return data for the wrong node now. These commands also return
information on some child nodes (bs->file and/or bs->backing), which
libvirt does make use of, and which refer to the wrong nodes, too.

One of the consequences is that oVirt gets wrong information about the
image size and stops the VM in response as long as a mirror or commit
job is running:

https://bugzilla.redhat.com/show_bug.cgi?id=1470634

This patch fixes the problem by hiding the implicit nodes created
automatically by the mirror and commit block jobs in the output of
query-block and BlockBackend-based query-blockstats as long as the user
doesn't indicate that they are aware of those nodes by providing a node
name for them in the QMP command to start the block job.

The node-based commands query-named-block-nodes and query-blockstats
with query-nodes=true still show all nodes, including implicit ones.
This ensures that users that are capable of node-level management can
still access the full information; users that only know BlockBackends
won't use these commands.

Cc: qemu-***@nongnu.org
Signed-off-by: Kevin Wolf <***@redhat.com>
Reviewed-by: Peter Krempa <***@redhat.com>
Reviewed-by: Max Reitz <***@redhat.com>
Tested-by: Eric Blake <***@redhat.com>
(cherry picked from commit d3c8c67469ee70fcae116d5abc277a7ebc8a19fd)
Conflicts:
block/qapi.c
include/block/block_int.h
* fix context deps on 46eade7b and 5a9347c6
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block.c | 13 -------------
block/commit.c | 3 +++
block/mirror.c | 3 +++
block/qapi.c | 34 ++++++++++++++++++++++++++++------
include/block/block.h | 1 -
include/block/block_int.h | 1 +
qapi/block-core.json | 6 ++++--
tests/qemu-iotests/040 | 30 +++++++++++++++++++++++++++++-
tests/qemu-iotests/040.out | 4 ++--
tests/qemu-iotests/041 | 38 +++++++++++++++++++++++++++++++++++++-
tests/qemu-iotests/041.out | 4 ++--
11 files changed, 109 insertions(+), 28 deletions(-)

diff --git a/block.c b/block.c
index 21cb65e..46ea4a3 100644
--- a/block.c
+++ b/block.c
@@ -3863,19 +3863,6 @@ BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
return retval;
}

-int bdrv_get_backing_file_depth(BlockDriverState *bs)
-{
- if (!bs->drv) {
- return 0;
- }
-
- if (!bs->backing) {
- return 0;
- }
-
- return 1 + bdrv_get_backing_file_depth(bs->backing->bs);
-}
-
void bdrv_init(void)
{
module_call_init(MODULE_INIT_BLOCK);
diff --git a/block/commit.c b/block/commit.c
index 66e3418..b4c1be7 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -351,6 +351,9 @@ void commit_start(const char *job_id, BlockDriverState *bs,
if (commit_top_bs == NULL) {
goto fail;
}
+ if (!filter_node_name) {
+ commit_top_bs->implicit = true;
+ }
commit_top_bs->total_sectors = top->total_sectors;
bdrv_set_aio_context(commit_top_bs, bdrv_get_aio_context(top));

diff --git a/block/mirror.c b/block/mirror.c
index 4e8f124..9bf5f89 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -1152,6 +1152,9 @@ static void mirror_start_job(const char *job_id, BlockDriverState *bs,
if (mirror_top_bs == NULL) {
return;
}
+ if (!filter_node_name) {
+ mirror_top_bs->implicit = true;
+ }
mirror_top_bs->total_sectors = bs->total_sectors;
bdrv_set_aio_context(mirror_top_bs, bdrv_get_aio_context(bs));

diff --git a/block/qapi.c b/block/qapi.c
index a40922e..803205b 100644
--- a/block/qapi.c
+++ b/block/qapi.c
@@ -64,7 +64,6 @@ BlockDeviceInfo *bdrv_block_device_info(BlockBackend *blk,
info->backing_file = g_strdup(bs->backing_file);
}

- info->backing_file_depth = bdrv_get_backing_file_depth(bs);
info->detect_zeroes = bs->detect_zeroes;

if (blk && blk_get_public(blk)->throttle_state) {
@@ -125,6 +124,7 @@ BlockDeviceInfo *bdrv_block_device_info(BlockBackend *blk,

bs0 = bs;
p_image_info = &info->image;
+ info->backing_file_depth = 0;
while (1) {
Error *local_err = NULL;
bdrv_query_image_info(bs0, p_image_info, &local_err);
@@ -133,13 +133,21 @@ BlockDeviceInfo *bdrv_block_device_info(BlockBackend *blk,
qapi_free_BlockDeviceInfo(info);
return NULL;
}
+
if (bs0->drv && bs0->backing) {
+ info->backing_file_depth++;
bs0 = bs0->backing->bs;
(*p_image_info)->has_backing_image = true;
p_image_info = &((*p_image_info)->backing_image);
} else {
break;
}
+
+ /* Skip automatically inserted nodes that the user isn't aware of for
+ * query-block (blk != NULL), but not for query-named-block-nodes */
+ while (blk && bs0 && bs0->drv && bs0->implicit) {
+ bs0 = backing_bs(bs0);
+ }
}

return info;
@@ -322,6 +330,12 @@ static void bdrv_query_info(BlockBackend *blk, BlockInfo **p_info,
{
BlockInfo *info = g_malloc0(sizeof(*info));
BlockDriverState *bs = blk_bs(blk);
+
+ /* Skip automatically inserted nodes that the user isn't aware of */
+ while (bs && bs->drv && bs->implicit) {
+ bs = backing_bs(bs);
+ }
+
info->device = g_strdup(blk_name(blk));
info->type = g_strdup("unknown");
info->locked = blk_dev_is_medium_locked(blk);
@@ -424,8 +438,8 @@ static void bdrv_query_blk_stats(BlockDeviceStats *ds, BlockBackend *blk)
}
}

-static BlockStats *bdrv_query_bds_stats(const BlockDriverState *bs,
- bool query_backing)
+static BlockStats *bdrv_query_bds_stats(BlockDriverState *bs,
+ bool blk_level)
{
BlockStats *s = NULL;

@@ -436,6 +450,14 @@ static BlockStats *bdrv_query_bds_stats(const BlockDriverState *bs,
return s;
}

+ /* Skip automatically inserted nodes that the user isn't aware of in
+ * a BlockBackend-level command. Stay at the exact node for a node-level
+ * command. */
+ while (blk_level && bs->drv && bs->implicit) {
+ bs = backing_bs(bs);
+ assert(bs);
+ }
+
if (bdrv_get_node_name(bs)[0]) {
s->has_node_name = true;
s->node_name = g_strdup(bdrv_get_node_name(bs));
@@ -445,12 +467,12 @@ static BlockStats *bdrv_query_bds_stats(const BlockDriverState *bs,

if (bs->file) {
s->has_parent = true;
- s->parent = bdrv_query_bds_stats(bs->file->bs, query_backing);
+ s->parent = bdrv_query_bds_stats(bs->file->bs, blk_level);
}

- if (query_backing && bs->backing) {
+ if (blk_level && bs->backing) {
s->has_backing = true;
- s->backing = bdrv_query_bds_stats(bs->backing->bs, query_backing);
+ s->backing = bdrv_query_bds_stats(bs->backing->bs, blk_level);
}

return s;
diff --git a/include/block/block.h b/include/block/block.h
index 4bf4843..060c08c 100644
--- a/include/block/block.h
+++ b/include/block/block.h
@@ -292,7 +292,6 @@ int coroutine_fn bdrv_co_pwrite_zeroes(BdrvChild *child, int64_t offset,
int count, BdrvRequestFlags flags);
BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
const char *backing_file);
-int bdrv_get_backing_file_depth(BlockDriverState *bs);
void bdrv_refresh_filename(BlockDriverState *bs);
int bdrv_truncate(BdrvChild *child, int64_t offset, Error **errp);
int64_t bdrv_nb_sectors(BlockDriverState *bs);
diff --git a/include/block/block_int.h b/include/block/block_int.h
index 59400bd..89d7b45 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -518,6 +518,7 @@ struct BlockDriverState {
bool valid_key; /* if true, a valid encryption key has been set */
bool sg; /* if true, the device is a /dev/sg* */
bool probed; /* if true, format was probed rather than specified */
+ bool implicit; /* if true, this filter node was automatically inserted */

BlockDriver *drv; /* NULL means no media */
void *opaque;
diff --git a/qapi/block-core.json b/qapi/block-core.json
index 38edada..7ce90ec 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -467,7 +467,8 @@
#
# Get a list of BlockInfo for all virtual block devices.
#
-# Returns: a list of @BlockInfo describing each virtual block device
+# Returns: a list of @BlockInfo describing each virtual block device. Filter
+# nodes that were created implicitly are skipped over.
#
# Since: 0.14.0
#
@@ -723,7 +724,8 @@
# information, but not "backing".
# If false or omitted, the behavior is as before - query all the
# device backends, recursively including their "parent" and
-# "backing". (Since 2.3)
+# "backing". Filter nodes that were created implicitly are
+# skipped over in this mode. (Since 2.3)
#
# Returns: A list of @BlockStats for each virtual block devices.
#
diff --git a/tests/qemu-iotests/040 b/tests/qemu-iotests/040
index 9d381d9..95b7510 100755
--- a/tests/qemu-iotests/040
+++ b/tests/qemu-iotests/040
@@ -81,7 +81,7 @@ class TestSingleDrive(ImageCommitTestCase):
qemu_img('create', '-f', iotests.imgfmt, '-o', 'backing_file=%s' % mid_img, test_img)
qemu_io('-f', 'raw', '-c', 'write -P 0xab 0 524288', backing_img)
qemu_io('-f', iotests.imgfmt, '-c', 'write -P 0xef 524288 524288', mid_img)
- self.vm = iotests.VM().add_drive(test_img, interface="none")
+ self.vm = iotests.VM().add_drive(test_img, "node-name=top,backing.node-name=mid,backing.backing.node-name=base", interface="none")
self.vm.add_device("virtio-scsi-pci")
self.vm.add_device("scsi-hd,id=scsi0,drive=drive0")
self.vm.launch()
@@ -163,6 +163,34 @@ class TestSingleDrive(ImageCommitTestCase):

self.assert_no_active_block_jobs()

+ # Tests that the insertion of the commit_top filter node doesn't make a
+ # difference to query-blockstat
+ def test_implicit_node(self):
+ if self.image_len == 0:
+ return
+
+ self.assert_no_active_block_jobs()
+ result = self.vm.qmp('block-commit', device='drive0', top=mid_img,
+ base=backing_img, speed=(self.image_len / 4))
+ self.assert_qmp(result, 'return', {})
+
+ result = self.vm.qmp('query-block')
+ self.assert_qmp(result, 'return[0]/inserted/file', test_img)
+ self.assert_qmp(result, 'return[0]/inserted/drv', iotests.imgfmt)
+ self.assert_qmp(result, 'return[0]/inserted/backing_file', mid_img)
+ self.assert_qmp(result, 'return[0]/inserted/backing_file_depth', 2)
+ self.assert_qmp(result, 'return[0]/inserted/image/filename', test_img)
+ self.assert_qmp(result, 'return[0]/inserted/image/backing-image/filename', mid_img)
+ self.assert_qmp(result, 'return[0]/inserted/image/backing-image/backing-image/filename', backing_img)
+
+ result = self.vm.qmp('query-blockstats')
+ self.assert_qmp(result, 'return[0]/node-name', 'top')
+ self.assert_qmp(result, 'return[0]/backing/node-name', 'mid')
+ self.assert_qmp(result, 'return[0]/backing/backing/node-name', 'base')
+
+ self.cancel_and_wait()
+ self.assert_no_active_block_jobs()
+
class TestRelativePaths(ImageCommitTestCase):
image_len = 1 * 1024 * 1024
test_len = 1 * 1024 * 256
diff --git a/tests/qemu-iotests/040.out b/tests/qemu-iotests/040.out
index 6d9bee1..e20a75c 100644
--- a/tests/qemu-iotests/040.out
+++ b/tests/qemu-iotests/040.out
@@ -1,5 +1,5 @@
-...........................
+.............................
----------------------------------------------------------------------
-Ran 27 tests
+Ran 29 tests

OK
diff --git a/tests/qemu-iotests/041 b/tests/qemu-iotests/041
index 2f54986..60f09cc 100755
--- a/tests/qemu-iotests/041
+++ b/tests/qemu-iotests/041
@@ -42,7 +42,7 @@ class TestSingleDrive(iotests.QMPTestCase):
def setUp(self):
iotests.create_image(backing_img, self.image_len)
qemu_img('create', '-f', iotests.imgfmt, '-o', 'backing_file=%s' % backing_img, test_img)
- self.vm = iotests.VM().add_drive(test_img)
+ self.vm = iotests.VM().add_drive(test_img, "node-name=top,backing.node-name=base")
if iotests.qemu_default_machine == 'pc':
self.vm.add_drive(None, 'media=cdrom', 'ide')
self.vm.launch()
@@ -169,6 +169,42 @@ class TestSingleDrive(iotests.QMPTestCase):
self.assertTrue(iotests.compare_images(test_img, target_img),
'target image does not match source after mirroring')

+ # Tests that the insertion of the mirror_top filter node doesn't make a
+ # difference to query-block
+ def test_implicit_node(self):
+ self.assert_no_active_block_jobs()
+
+ result = self.vm.qmp(self.qmp_cmd, device='drive0', sync='full',
+ target=self.qmp_target)
+ self.assert_qmp(result, 'return', {})
+
+ result = self.vm.qmp('query-block')
+ self.assert_qmp(result, 'return[0]/inserted/file', test_img)
+ self.assert_qmp(result, 'return[0]/inserted/drv', iotests.imgfmt)
+ self.assert_qmp(result, 'return[0]/inserted/backing_file', backing_img)
+ self.assert_qmp(result, 'return[0]/inserted/backing_file_depth', 1)
+ self.assert_qmp(result, 'return[0]/inserted/image/filename', test_img)
+ self.assert_qmp(result, 'return[0]/inserted/image/backing-image/filename', backing_img)
+
+ result = self.vm.qmp('query-blockstats')
+ self.assert_qmp(result, 'return[0]/node-name', 'top')
+ self.assert_qmp(result, 'return[0]/backing/node-name', 'base')
+
+ self.cancel_and_wait(force=True)
+ result = self.vm.qmp('query-block')
+ self.assert_qmp(result, 'return[0]/inserted/file', test_img)
+ self.assert_qmp(result, 'return[0]/inserted/drv', iotests.imgfmt)
+ self.assert_qmp(result, 'return[0]/inserted/backing_file', backing_img)
+ self.assert_qmp(result, 'return[0]/inserted/backing_file_depth', 1)
+ self.assert_qmp(result, 'return[0]/inserted/image/filename', test_img)
+ self.assert_qmp(result, 'return[0]/inserted/image/backing-image/filename', backing_img)
+
+ result = self.vm.qmp('query-blockstats')
+ self.assert_qmp(result, 'return[0]/node-name', 'top')
+ self.assert_qmp(result, 'return[0]/backing/node-name', 'base')
+
+ self.vm.shutdown()
+
def test_medium_not_found(self):
if iotests.qemu_default_machine != 'pc':
return
diff --git a/tests/qemu-iotests/041.out b/tests/qemu-iotests/041.out
index e30fd3b..c28b392 100644
--- a/tests/qemu-iotests/041.out
+++ b/tests/qemu-iotests/041.out
@@ -1,5 +1,5 @@
-...............................................................................
+.....................................................................................
----------------------------------------------------------------------
-Ran 79 tests
+Ran 85 tests

OK
--
2.7.4
Michael Roth
2017-08-29 00:13:43 UTC
Permalink
From: "Daniel P. Berrange" <***@redhat.com>

Historically the migration data channel has only needed to be
unidirectional. Thus the 'exec:' protocol was requesting an
I/O channel with O_RDONLY on incoming side, and O_WRONLY on
the outgoing side.

This is fine for classic migration, but if you then try to run
TLS over it, this fails because the TLS handshake requires a
bi-directional channel.

Signed-off-by: Daniel P. Berrange <***@redhat.com>
Reviewed-by: Juan Quintela <***@redhat.com>
Signed-off-by: Juan Quintela <***@redhat.com>
(cherry picked from commit 062d81f0e968fe1597474735f3ea038065027372)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
migration/exec.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/migration/exec.c b/migration/exec.c
index 9157721..aba9089 100644
--- a/migration/exec.c
+++ b/migration/exec.c
@@ -32,7 +32,7 @@ void exec_start_outgoing_migration(MigrationState *s, const char *command, Error

trace_migration_exec_outgoing(command);
ioc = QIO_CHANNEL(qio_channel_command_new_spawn(argv,
- O_WRONLY,
+ O_RDWR,
errp));
if (!ioc) {
return;
@@ -59,7 +59,7 @@ void exec_start_incoming_migration(const char *command, Error **errp)

trace_migration_exec_incoming(command);
ioc = QIO_CHANNEL(qio_channel_command_new_spawn(argv,
- O_RDONLY,
+ O_RDWR,
errp));
if (!ioc) {
return;
--
2.7.4
Michael Roth
2017-08-29 00:14:53 UTC
Permalink
From: Jeff Cody <***@redhat.com>

Commit c096358e747e88fc7364e40e3c354ee0bb683960 introduced assertion
checks for when qemu_mutex() functions are called without the
corresponding qemu_mutex_init() having initialized the mutex.

This uncovered a latent bug in qemu's nfs driver - in
nfs_client_close(), the NFSClient structure is overwritten with zeros,
prior to the mutex being destroyed.

Go ahead and destroy the mutex in nfs_client_close(), and change where
we call qemu_mutex_init() so that it is correctly balanced.

There are also a couple of memory leaks obscured by the memset, so this
fixes those as well.

Finally, we should be able to get rid of the memset(), as it isn't
necessary.

Cc: qemu-***@nongnu.org
Signed-off-by: Jeff Cody <***@redhat.com>
Reviewed-by: Peter Lieven <***@kamp.de>
Reviewed-by: Stefan Hajnoczi <***@redhat.com>
Reviewed-by: John Snow <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 113fe792fd4931dd0538f03859278b8719ee4fa2)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
block/nfs.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/block/nfs.c b/block/nfs.c
index 344186f..7e1bea1 100644
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -434,19 +434,23 @@ static void nfs_client_close(NFSClient *client)
if (client->context) {
if (client->fh) {
nfs_close(client->context, client->fh);
+ client->fh = NULL;
}
aio_set_fd_handler(client->aio_context, nfs_get_fd(client->context),
false, NULL, NULL, NULL, NULL);
nfs_destroy_context(client->context);
+ client->context = NULL;
}
- memset(client, 0, sizeof(NFSClient));
+ g_free(client->path);
+ qemu_mutex_destroy(&client->mutex);
+ qapi_free_NFSServer(client->server);
+ client->server = NULL;
}

static void nfs_file_close(BlockDriverState *bs)
{
NFSClient *client = bs->opaque;
nfs_client_close(client);
- qemu_mutex_destroy(&client->mutex);
}

static NFSServer *nfs_config(QDict *options, Error **errp)
@@ -499,6 +503,7 @@ static int64_t nfs_client_open(NFSClient *client, QDict *options,
struct stat st;
char *file = NULL, *strp = NULL;

+ qemu_mutex_init(&client->mutex);
opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
qemu_opts_absorb_qdict(opts, options, &local_err);
if (local_err) {
@@ -661,7 +666,7 @@ static int nfs_file_open(BlockDriverState *bs, QDict *options, int flags,
if (ret < 0) {
return ret;
}
- qemu_mutex_init(&client->mutex);
+
bs->total_sectors = ret;
ret = 0;
return ret;
--
2.7.4
Michael Roth
2017-08-29 00:14:52 UTC
Permalink
From: Aleksandr Bezzubikov <***@gmail.com>

Unmask previously masked SHPC feature in _OSC method.

Signed-off-by: Aleksandr Bezzubikov <***@gmail.com>
Reviewed-by: Marcel Apfelbaum <***@redhat.com>
Reviewed-by: Michael S. Tsirkin <***@redhat.com>
Signed-off-by: Michael S. Tsirkin <***@redhat.com>
(cherry picked from commit a41c78c135eb1850826e96b2154690323ff66719)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
hw/i386/acpi-build.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
index 2073108..76fa455 100644
--- a/hw/i386/acpi-build.c
+++ b/hw/i386/acpi-build.c
@@ -1818,9 +1818,9 @@ static Aml *build_q35_osc_method(void)

/*
* Always allow native PME, AER (no dependencies)
- * Never allow SHPC (no SHPC controller in this system)
+ * Allow SHPC (PCI bridges can have SHPC controller)
*/
- aml_append(if_ctx, aml_and(a_ctrl, aml_int(0x1D), a_ctrl));
+ aml_append(if_ctx, aml_and(a_ctrl, aml_int(0x1F), a_ctrl));

if_ctx2 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(1))));
/* Unknown revision */
--
2.7.4
Michael Roth
2017-08-29 00:13:42 UTC
Permalink
From: Max Reitz <***@redhat.com>

Signed-off-by: Max Reitz <***@redhat.com>
Reviewed-by: Eric Blake <***@redhat.com>
Reviewed-by: Fam Zheng <***@redhat.com>
Signed-off-by: Kevin Wolf <***@redhat.com>
(cherry picked from commit 42dc10f17a7f1754d419e715114c37f5c5fde12f)
Signed-off-by: Michael Roth <***@linux.vnet.ibm.com>
---
tests/qemu-iotests/051 | 1 +
tests/qemu-iotests/051.out | 3 +++
tests/qemu-iotests/051.pc.out | 3 +++
3 files changed, 7 insertions(+)

diff --git a/tests/qemu-iotests/051 b/tests/qemu-iotests/051
index 630cb7a..4fe6760 100755
--- a/tests/qemu-iotests/051
+++ b/tests/qemu-iotests/051
@@ -231,6 +231,7 @@ echo === Leaving out required options ===
echo

run_qemu -drive driver=file
+run_qemu -drive driver=file,filename=
run_qemu -drive driver=nbd
run_qemu -drive driver=raw
run_qemu -drive file.driver=file
diff --git a/tests/qemu-iotests/051.out b/tests/qemu-iotests/051.out
index 7524c62..57bee08 100644
--- a/tests/qemu-iotests/051.out
+++ b/tests/qemu-iotests/051.out
@@ -221,6 +221,9 @@ QEMU X.Y.Z monitor - type 'help' for more information
Testing: -drive driver=file
QEMU_PROG: -drive driver=file: The 'file' block driver requires a file name

+Testing: -drive driver=file,filename=
+QEMU_PROG: -drive driver=file,filename=: The 'file' block driver requires a file name
+
Testing: -drive driver=nbd
QEMU_PROG: -drive driver=nbd: NBD server address missing

diff --git a/tests/qemu-iotests/051.pc.out b/tests/qemu-iotests/051.pc.out
index c6f4eef..19426e5 100644
--- a/tests/qemu-iotests/051.pc.out
+++ b/tests/qemu-iotests/051.pc.out
@@ -319,6 +319,9 @@ QEMU X.Y.Z monitor - type 'help' for more information
Testing: -drive driver=file
QEMU_PROG: -drive driver=file: The 'file' block driver requires a file name

+Testing: -drive driver=file,filename=
+QEMU_PROG: -drive driver=file,filename=: The 'file' block driver requires a file name
+
Testing: -drive driver=nbd
QEMU_PROG: -drive driver=nbd: NBD server address missing
--
2.7.4
Michael Roth
2017-08-29 00:47:58 UTC
Permalink
Quoting Michael Roth (2017-08-28 19:13:35)
Hi everyone,
https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_mdroth_qemu_commits_stable-2D2.9-2Dstaging&d=DwIFaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=Mw1bU8iEiV5THnZe_RluoHefJMDFgKus3DOUY40AbVA&s=wB11-59-V11-yisUGoowZ4UvmSBfZNqhjDdEk9QwqAk&e=
https://urldefense.proofpoint.com/v2/url?u=http-3A__wiki.qemu.org_Planning_2.9&d=DwIFaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=Mw1bU8iEiV5THnZe_RluoHefJMDFgKus3DOUY40AbVA&s=aKWE0XkaM9D2OJvn5Etwst9lR3FUDED9C_m5ue7HB6w&e=
Sorry for this. I've sent some other emails to see if this behavior
continued from my SMTP relay, and it seems to have been some sort of
temporary issue. The original URLs were (assuming I don't get bit by
this again):

https://github.com/mdroth/qemu/commits/stable-2.9-staging

and

http://wiki.qemu.org/Planning/2.9
think should be included in the release.
Testing/feedback is greatly appreciated.
Thanks!
----------------------------------------------------------------
stream: fix crash in stream_start() when block_job_create() fails
hw/i386: allow SHPC for Q35 machine
hid: Reset kbd modifiers on reset
input: Decrement queue count on kbd delay
qemu-img: wait for convert coroutines to complete
ACPI: don't call acpi_pcihp_device_plug_cb on xen
9pfs: local: remove: use correct path component
migration: setup bi-directional I/O channel for exec: protocol
pc: Use "min-[x]level" on compat_props
dirty-bitmap: Report BlockDirtyInfo.count in bytes, as documented
coccinelle: Add script to remove useless QObject casts
qobject: Drop useless QObject casts
qobject: Add helper macros for common scalar insertions
s390x: Drop useless casts
qobject: Use simpler QDict/QList scalar insertion macros
blkdebug: Sanity check block layer guarantees
blkdebug: Refactor error injection
blkdebug: Add pass-through write_zero and discard support
blkdebug: Simplify override logic
blkdebug: Add ability to override unmap geometries
tests: Add coverage for recent block geometry fixes
block: Simplify BDRV_BLOCK_RAW recursion
block: Guarantee that *file is set on bdrv_get_block_status()
nbd: Fully initialize client in case of failed negotiation
nbd: Fix regression on resiliency to port scan
block: Reuse bs as backing hd for drive-backup sync=none
virtio-scsi: Unset hotplug handler when unrealize
input: limit kbd queue depth
9pfs: local: fix unlink of alien files in mapped-file mode
virtio: allow broken device to notify guest
target/ppc: pass const string to kvmppc_is_mem_backend_page_size_ok()
target/ppc: fix memory leak in kvmppc_is_mem_backend_page_size_ok()
spapr: fix memory leak in spapr_memory_pre_plug()
spapr: fix memory leak in spapr_core_pre_plug()
9pfs: local: fix fchmodat_nofollow() limitations
s390x/css: catch section mismatch on load
pci: deassert intx when pci device unrealize
vvfat: fix qemu-img map and qemu-img convert
virtio-scsi: finalize IOMMU support
virtio-net: fix offload ctrl endian
block/nfs: fix mutex assertion in nfs_file_close()
blockdev: use drained_begin/end for qmp_block_resize
mirror: Drop permissions on s->target on completion
commit: Fix use after free in completion
commit: Fix completion with extra reference
commit: Add NULL check for overlay_bs
qemu-iotests: Test automatic commit job cancel on hot unplug
block: Skip implicit nodes in query-block/blockstats
virtio-serial-bus: Unset hotplug handler when unrealize
spapr: add pre_plug function for memory
spapr: fix migration to pseries machine < 2.8
cpu: don't allow negative core id
replication: Make --disable-replication compile again
target/xtensa: fix mapping direction in read/write simcalls
target/xtensa: fix return value of read/write simcalls
target/xtensa: handle unknown registers in gdbstub
qemu-img/convert: Always set ret < 0 on error
qemu-img/convert: Move bs_n > 1 && -B check down
block: An empty filename counts as no filename
iotests/051: Add test for empty filename
block: Do not unref bs->file on error in BD's open
block/vhdx: Make vhdx_create() always set errp
block: Add errp to b{lk,drv}_truncate()
blkdebug: Catch bs->exact_filename overflow
blkverify: Catch bs->exact_filename overflow
block: Do not strcmp() with NULL uri->scheme
tests: check-qom-proplist: add checks for cmdline-created objects
monitor: fix object_del for command-line-created objects
curl: strengthen assertion in curl_clean_state
curl: never invoke callbacks with s->mutex held
curl: avoid recursive locking of BDRVCURLState mutex
linuxboot_dma: compile for i486
nbd: fix NBD over TLS
qga-win: Enable 'can-offline' field in 'guest-get-vcpus' reply
e1000e: Fix ICR "Other" causes clear logic
aio: add missing aio_notify() to aio_enable_external()
virtio-net: fix wild pointer when remove virtio-net queues
hw/virtio: fix vhost user fails to startup when MQ
MAINTAINERS | 1 +
block.c | 82 +++++-----
block/blkdebug.c | 288 ++++++++++++++++++++++++++++--------
block/blkreplay.c | 3 -
block/blkverify.c | 26 ++--
block/block-backend.c | 5 +-
block/commit.c | 24 ++-
block/crypto.c | 2 +-
block/curl.c | 33 ++++-
block/dirty-bitmap.c | 2 +-
block/file-posix.c | 8 +-
block/file-win32.c | 4 +-
block/io.c | 9 +-
block/mirror.c | 12 +-
block/nbd-client.c | 4 +-
block/nbd.c | 47 +++---
block/nfs.c | 56 ++++---
block/null.c | 2 +-
block/parallels.c | 13 +-
block/qapi.c | 34 ++++-
block/qcow.c | 6 +-
block/qcow2-refcount.c | 5 +-
block/qcow2.c | 18 ++-
block/qed.c | 2 +-
block/quorum.c | 16 +-
block/raw-format.c | 2 +-
block/rbd.c | 16 +-
block/sheepdog.c | 6 +-
block/snapshot.c | 2 +-
block/ssh.c | 18 +--
block/stream.c | 2 +-
block/vdi.c | 4 +-
block/vhdx-log.c | 2 +-
block/vhdx.c | 29 +++-
block/vmdk.c | 13 +-
block/vpc.c | 13 +-
block/vvfat.c | 13 +-
blockdev-nbd.c | 6 +-
blockdev.c | 71 ++++-----
hw/9pfs/9p-local.c | 78 ++++++----
hw/9pfs/9p-util.h | 24 +--
hw/acpi/piix4.c | 11 +-
hw/block/xen_disk.c | 2 +-
hw/char/virtio-serial-bus.c | 3 +
hw/cpu/core.c | 5 +
hw/i386/acpi-build.c | 4 +-
hw/input/hid.c | 1 +
hw/net/e1000e_core.c | 10 +-
hw/net/virtio-net.c | 5 +
hw/pci/pci.c | 1 +
hw/ppc/spapr.c | 46 ++++--
hw/ppc/spapr_pci.c | 28 ++--
hw/s390x/css.c | 14 ++
hw/s390x/virtio-ccw.c | 6 +-
hw/scsi/virtio-scsi.c | 6 +-
hw/usb/xen-usb.c | 12 +-
hw/virtio/vhost-user.c | 21 ++-
hw/virtio/virtio.c | 4 +-
include/block/aio.h | 10 +-
include/block/block.h | 3 +-
include/block/block_int.h | 1 +
include/block/nbd.h | 2 +-
include/hw/i386/pc.h | 42 +++---
include/qapi/qmp/qdict.h | 8 +
include/qapi/qmp/qlist.h | 8 +
include/sysemu/block-backend.h | 2 +-
migration/colo.c | 12 ++
migration/exec.c | 4 +-
monitor.c | 28 ++--
nbd/server.c | 32 ++--
pc-bios/linuxboot_dma.bin | Bin 1536 -> 1536 bytes
pc-bios/optionrom/Makefile | 1 +
qapi/block-core.json | 39 ++++-
qapi/qmp-event.c | 2 +-
qemu-img.c | 75 +++++-----
qemu-io-cmds.c | 5 +-
qemu-io.c | 2 +-
qemu-nbd.c | 8 +-
qga/commands-win32.c | 2 +-
qobject/qdict.c | 2 +-
qom/object_interfaces.c | 9 ++
scripts/coccinelle/qobject.cocci | 35 +++++
target/ppc/kvm.c | 5 +-
target/ppc/kvm_ppc.h | 4 +-
target/s390x/cpu_models.c | 4 +-
target/xtensa/gdbstub.c | 13 +-
target/xtensa/xtensa-semi.c | 27 +++-
tests/check-qdict.c | 142 +++++++++---------
tests/check-qlist.c | 4 +-
tests/check-qom-proplist.c | 56 +++++++
tests/device-introspect-test.c | 4 +-
tests/qemu-iotests/040 | 63 +++++++-
tests/qemu-iotests/040.out | 4 +-
tests/qemu-iotests/041 | 38 ++++-
tests/qemu-iotests/041.out | 4 +-
tests/qemu-iotests/051 | 1 +
tests/qemu-iotests/051.out | 3 +
tests/qemu-iotests/051.pc.out | 3 +
tests/qemu-iotests/122.out | 4 +-
tests/qemu-iotests/177 | 117 +++++++++++++++
tests/qemu-iotests/177.out | 51 +++++++
tests/qemu-iotests/group | 1 +
tests/test-qemu-opts.c | 4 +-
tests/test-qmp-commands.c | 30 ++--
tests/test-qmp-event.c | 30 ++--
tests/test-qobject-output-visitor.c | 6 +-
tests/test-x86-cpuid-compat.c | 38 +++++
ui/input.c | 15 +-
util/qemu-option.c | 2 +-
109 files changed, 1507 insertions(+), 678 deletions(-)
create mode 100644 scripts/coccinelle/qobject.cocci
create mode 100755 tests/qemu-iotests/177
create mode 100644 tests/qemu-iotests/177.out
Thomas Huth
2017-08-29 02:18:20 UTC
Permalink
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
I'd like to suggest the following patches:

601b9a9008c5a612d76073bb - target-s390x: Mask the SIGP order_code ...
b7da97eef74bf834be244de0 - monitor: Check whether TCG is enabled ...
17eb587aeb492fe68f8130b0 - slirp: tftp, copy sockaddr_size
99efaa2696caaf6182958e27 - hw/s390x/ipl: Fix crash with ...
36bed541ca886da735bef1e8 - fix qemu-system-unicore32 crashing ...
b190f477e29c7cd03a8fee49 - qemu-system-tricore: segfault when ...
8ccccff9dd7ba24c7a788611 - hw/ppc/spapr_rtc: Mark the RTC device ...
1f98e55385d11da1dc0de644 - hw/ppc/spapr_iommu: Fix crash when ...

Not sure, but maybe the following patch should be included, too, since
there were some bogus files in the old version of the U-Boot sources:

73663d71ef2bab201475d58e - PPC: E500: Update u-boot to v2017.07

Thomas
Cornelia Huck
2017-08-29 07:41:30 UTC
Permalink
On Tue, 29 Aug 2017 04:18:20 +0200
Post by Thomas Huth
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
601b9a9008c5a612d76073bb - target-s390x: Mask the SIGP order_code ...
99efaa2696caaf6182958e27 - hw/s390x/ipl: Fix crash with ...
Agreed on the s390x patches.
Michael Roth
2017-08-31 16:42:45 UTC
Permalink
Quoting Thomas Huth (2017-08-28 21:18:20)
Post by Thomas Huth
Hi everyone,
https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_mdroth_qemu_commits_stable-2D2.9-2Dstaging&d=DwIDaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=Id5ItcTzhCqn35tC8JynLtLuRcfupmsTlJGwTYEDdIg&s=fqHIfooeKQPNEWX7AqyC93OMzAs-U-UwZ6Yu0trfn0Y&e=
https://urldefense.proofpoint.com/v2/url?u=http-3A__wiki.qemu.org_Planning_2.9&d=DwIDaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=Id5ItcTzhCqn35tC8JynLtLuRcfupmsTlJGwTYEDdIg&s=RVxHUyHJDN1hk2AsMfiZmguXpEhz0pFHGijG75NIReY&e=
think should be included in the release.
601b9a9008c5a612d76073bb - target-s390x: Mask the SIGP order_code ...
b7da97eef74bf834be244de0 - monitor: Check whether TCG is enabled ...
17eb587aeb492fe68f8130b0 - slirp: tftp, copy sockaddr_size
99efaa2696caaf6182958e27 - hw/s390x/ipl: Fix crash with ...
36bed541ca886da735bef1e8 - fix qemu-system-unicore32 crashing ...
b190f477e29c7cd03a8fee49 - qemu-system-tricore: segfault when ...
8ccccff9dd7ba24c7a788611 - hw/ppc/spapr_rtc: Mark the RTC device ...
1f98e55385d11da1dc0de644 - hw/ppc/spapr_iommu: Fix crash when ...
Not sure, but maybe the following patch should be included, too, since
73663d71ef2bab201475d58e - PPC: E500: Update u-boot to v2017.07
Do you have more background on any issues caused by these bogus files?
As it stands I think I would opt not to update unless there are specific
user-visible bugs we're trying to address which warrant the risk of any
regressions which might get pulled in in the process.
Post by Thomas Huth
Thomas
Peter Maydell
2017-08-31 17:07:08 UTC
Permalink
Post by Michael Roth
Quoting Thomas Huth (2017-08-28 21:18:20)
Post by Thomas Huth
Not sure, but maybe the following patch should be included, too, since
73663d71ef2bab201475d58e - PPC: E500: Update u-boot to v2017.07
Do you have more background on any issues caused by these bogus files?
As it stands I think I would opt not to update unless there are specific
user-visible bugs we're trying to address which warrant the risk of any
regressions which might get pulled in in the process.
These are the relevant threads:
https://lists.gnu.org/archive/html/qemu-discuss/2017-07/msg00005.html
and
https://lists.gnu.org/archive/html/qemu-devel/2017-07/msg02956.html

The summary is
(1) one of the u-boot source files which is distributed as part
of the QEMU tarball has a comment which makes it a bit unclear
whether it's something that's redistributable (the source file
isn't actually used in the u-boot target we care about)
(2) the u-boot binary blob we were shipping doesn't correspond
to the sources we were shipping

and we fixed those in master by updating the blob and the
submodule to the most recent u-boot.

I guess the low-risk fix for the stable branch would be to
update the u-boot submodule to 79c884d7e4 as suggested in
https://lists.gnu.org/archive/html/qemu-devel/2017-07/msg03174.html
which would bring the distributed sources into line with
the binary blob in stable, so no need to change the
blob we're distributing. I think it makes sense to do that
for stable.

thanks
-- PMM
Michael Roth
2017-08-31 18:09:52 UTC
Permalink
Quoting Peter Maydell (2017-08-31 12:07:08)
Post by Michael Roth
Quoting Thomas Huth (2017-08-28 21:18:20)
Post by Thomas Huth
Not sure, but maybe the following patch should be included, too, since
73663d71ef2bab201475d58e - PPC: E500: Update u-boot to v2017.07
Do you have more background on any issues caused by these bogus files?
As it stands I think I would opt not to update unless there are specific
user-visible bugs we're trying to address which warrant the risk of any
regressions which might get pulled in in the process.
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.gnu.org_archive_html_qemu-2Ddiscuss_2017-2D07_msg00005.html&d=DwIBaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=QzqXTgvEiqKKSlIJgVfNCEiYXPQ5oVFxHFdUcgtf_L8&s=4BVFXlpHawdLmHljZCHMSNEVaj8JzUuNJw6HgKZzvn0&e=
and
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.gnu.org_archive_html_qemu-2Ddevel_2017-2D07_msg02956.html&d=DwIBaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=QzqXTgvEiqKKSlIJgVfNCEiYXPQ5oVFxHFdUcgtf_L8&s=iO72EaulRM4jy_9QBnIcqH5K_hIWtAmBOMqy6QORs2M&e=
The summary is
(1) one of the u-boot source files which is distributed as part
of the QEMU tarball has a comment which makes it a bit unclear
whether it's something that's redistributable (the source file
isn't actually used in the u-boot target we care about)
(2) the u-boot binary blob we were shipping doesn't correspond
to the sources we were shipping
and we fixed those in master by updating the blob and the
submodule to the most recent u-boot.
I guess the low-risk fix for the stable branch would be to
update the u-boot submodule to 79c884d7e4 as suggested in
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.gnu.org_archive_html_qemu-2Ddevel_2017-2D07_msg03174.html&d=DwIBaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=QzqXTgvEiqKKSlIJgVfNCEiYXPQ5oVFxHFdUcgtf_L8&s=WY3VTHQGDP63Rw7hykVtVSbAqb8db-of8rkUG3hrlUg&e=
which would bring the distributed sources into line with
the binary blob in stable, so no need to change the
blob we're distributing. I think it makes sense to do that
for stable.
Thanks for the background/suggestion, I think I'll take this
approach.
thanks
-- PMM
Cole Robinson
2017-08-29 14:04:49 UTC
Permalink
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
Here's extra patches we are carrying for Fedora 26:

Applying: vmw_pvscsi: check message ring page count at initialisation
Applying: audio: release capture buffers
Applying: scsi: avoid an off-by-one error in megasas_mmio_write
Applying: 9pfs: local: forbid client access to metadata (CVE-2017-7493)
Applying: megasas: do not read sense length more than once from frame
Applying: megasas: do not read iovec count more than once from frame
Applying: megasas: do not read DCMD opcode more than once from frame
Applying: megasas: do not read command more than once from frame
Applying: megasas: do not read SCSI req parameters more than once from frame
Applying: megasas: always store SCSIRequest* into MegasasCmd
Applying: usb-redir: fix stack overflow in usbredir_log_data
Applying: qemu-nbd: Ignore SIGPIPE
Applying: vnc: Set default kbd delay to 10ms


The last one was part of agraf's input fixes, not sure if it's strictly
required to fix the root issue but when all 3 patches were applied it fixed
fedora's automated QA system.

The rest of the patches are for CVEs but not sure about the severity

Thanks,
Cole
Peter Maydell
2017-08-31 10:21:28 UTC
Permalink
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
I would suggest also commit 1201d308519f1e915866d7583d5136d03cc1d384
("slirp: fix clearing ifq_so from pending packets") which I've
just applied to master, as it fixes a use-after-free if the
guest sends suitable bogus packets and the VM is using slirp
networking.

thanks
-- PMM
Michael Roth
2017-08-31 18:00:00 UTC
Permalink
Quoting Michael Roth (2017-08-28 19:13:35)
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
Thank you for the recommendations. Branch updated with the following
additional patches:

s390-ccw: Fix alignment for CCW1
vnc: Set default kbd delay to 10ms
qemu-nbd: Ignore SIGPIPE
usb-redir: fix stack overflow in usbredir_log_data
megasas: do not read SCSI req parameters more than once from frame
megasas: do not read command more than once from frame
megasas: do not read DCMD opcode more than once from frame
megasas: do not read iovec count more than once from frame
megasas: do not read sense length more than once from frame
9pfs: local: forbid client access to metadata (CVE-2017-7493)
scsi: avoid an off-by-one error in megasas_mmio_write
audio: release capture buffers
vmw_pvscsi: check message ring page count at initialisation
hw/ppc/spapr_iommu: Fix crash when removing the "spapr-tce-table" device
hw/ppc/spapr_rtc: Mark the RTC device with user_creatable = false
qdev: Replace cannot_instantiate_with_device_add_yet with !user_creatable
fix qemu-system-unicore32 crashing when calling without -kernel
hw/s390x/ipl: Fix crash with virtio-scsi-pci device
slirp: fix clearing ifq_so from pending packets
slirp: tftp, copy sockaddr_size
monitor: Check whether TCG is enabled before running the "info jit" code
target-s390x: Mask the SIGP order_code to 8bit.
Michael Roth
2017-09-01 23:22:51 UTC
Permalink
Quoting Michael Roth (2017-08-31 13:00:00)
Post by Michael Roth
Quoting Michael Roth (2017-08-28 19:13:35)
Hi everyone,
https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_mdroth_qemu_commits_stable-2D2.9-2Dstaging&d=DwIFaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=sThPI1c0u5x-3sg5Nw8wNqjg_5Z5xLzfPGC18E94zn8&m=hoyg_NnavbPJKZEF_zct-EdpBD-Nb3rhYLms_7k54A4&s=dc2xpZH9f2d88cyYQblkq-w_a99BSN8i6ypQkAYf5Pw&e=
Thank you for the recommendations. Branch updated with the following
s390-ccw: Fix alignment for CCW1
vnc: Set default kbd delay to 10ms
qemu-nbd: Ignore SIGPIPE
usb-redir: fix stack overflow in usbredir_log_data
megasas: do not read SCSI req parameters more than once from frame
megasas: do not read command more than once from frame
megasas: do not read DCMD opcode more than once from frame
megasas: do not read iovec count more than once from frame
megasas: do not read sense length more than once from frame
9pfs: local: forbid client access to metadata (CVE-2017-7493)
scsi: avoid an off-by-one error in megasas_mmio_write
audio: release capture buffers
vmw_pvscsi: check message ring page count at initialisation
hw/ppc/spapr_iommu: Fix crash when removing the "spapr-tce-table" device
hw/ppc/spapr_rtc: Mark the RTC device with user_creatable = false
qdev: Replace cannot_instantiate_with_device_add_yet with !user_creatable
fix qemu-system-unicore32 crashing when calling without -kernel
hw/s390x/ipl: Fix crash with virtio-scsi-pci device
slirp: fix clearing ifq_so from pending packets
slirp: tftp, copy sockaddr_size
monitor: Check whether TCG is enabled before running the "info jit" code
target-s390x: Mask the SIGP order_code to 8bit.
Branch updated with the following additional patches:

exec: Add lock parameter to qemu_ram_ptr_length
xen/mapcache: store dma information in revmapcache entries for debugging
exec: use qemu_ram_ptr_length to access guest ram
xhci: only update dequeue ptr on completed transfers
vl.c/exit: pause cpus before closing block devices
PPC: E500: update u-boot to match shipped binary
Peter Lieven
2017-08-31 18:19:08 UTC
Permalink
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
I also have:

452589b vl.c/exit: pause cpus before closing block devices

Peter
Bruce Rogers
2017-09-01 19:03:42 UTC
Permalink
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
In addition to ones already mentioned by others might I also suggest the
following which we are carrying in our v2.9.0 based qemu for SUSE / openSUSE
releases, which seem to be stable material:

d54fddea xhci: only update dequeue ptr on completed transfers
04bf2526 exec: use qemu_ram_ptr_length to access guest ram

This second patch has a follow on due to negatively affecting xen usage:
f5aa69bd exec: Add locking parameter to qemu_ram_ptr_length

Thanks.

Bruce Rogers
Peter Maydell
2017-09-04 17:33:57 UTC
Permalink
Hi everyone,
https://github.com/mdroth/qemu/commits/stable-2.9-staging
http://wiki.qemu.org/Planning/2.9
think should be included in the release.
Testing/feedback is greatly appreciated.
If it's not too late for this stable release, you might consider
3e4d91b94ce400326fae0850578d9e9f30a71adb
(which just hit master).

This is a pretty long-standing bug, so it's not the end of
the world if it doesn't get in, but it does fix a bug which
was causing real-world FreeBSD AArch64 guest binaries to segfault:
https://bugs.launchpad.net/qemu/+bug/1711316
https://bugs.launchpad.net/qemu/+bug/1713066

thanks
-- PMM

Loading...