* [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable
@ 2017-02-04 12:21 Ashijeet Acharya
2017-02-04 15:44 ` Paolo Bonzini
2017-02-10 15:04 ` Dr. David Alan Gilbert
0 siblings, 2 replies; 5+ messages in thread
From: Ashijeet Acharya @ 2017-02-04 12:21 UTC (permalink / raw)
To: dgilbert; +Cc: quintela, peter.maydell, pbonzini, qemu-devel, Ashijeet Acharya
Commit a3a3d8c7 introduced a segfault bug while checking for
'dc->vmsd->unmigratable' which caused QEMU to crash when trying to add
devices which do no set their 'dc->vmsd' yet while initialization.
Place a 'dc->vmsd' check prior to it so that we do not segfault for
such devices.
NOTE: This doesn't compromise the functioning of --only-migratable
option as all the unmigratable devices do set their 'dc->vmsd'.
Also, move the only_migratable check inside device_set_realized() to avoid
code duplication and fix the bug to display only_migratable error message
only if the device constructuon is right.
Signed-off-by: Ashijeet Acharya <ashijeetacharya@gmail.com>
---
Changes is v3:
- move only_migratable check inside device_set_realized() to avoid code
- duplication
- I have dropped Juan's R-b tag for this one
Changes in v2:
- place dc->vmsd check in hw/usb/bus.c as well
---
hw/core/qdev.c | 10 ++++++++++
hw/usb/bus.c | 19 -------------------
qdev-monitor.c | 9 ---------
3 files changed, 10 insertions(+), 28 deletions(-)
diff --git a/hw/core/qdev.c b/hw/core/qdev.c
index 5783442..4438cbf 100644
--- a/hw/core/qdev.c
+++ b/hw/core/qdev.c
@@ -37,6 +37,7 @@
#include "hw/boards.h"
#include "hw/sysbus.h"
#include "qapi-event.h"
+#include "migration/migration.h"
int qdev_hotplug = 0;
static bool qdev_hot_added = false;
@@ -896,6 +897,15 @@ static void device_set_realized(Object *obj, bool value, Error **errp)
}
if (value && !dev->realized) {
+ if (only_migratable && dc->vmsd) {
+ if (dc->vmsd->unmigratable) {
+ error_setg(&local_err, "Device %s is not migratable, but "
+ "--only-migratable was specified",
+ object_get_typename(obj));
+ goto fail;
+ }
+ }
+
if (!obj->parent) {
gchar *name = g_strdup_printf("device[%d]", unattached_count++);
diff --git a/hw/usb/bus.c b/hw/usb/bus.c
index 1dcc35c..25913ad 100644
--- a/hw/usb/bus.c
+++ b/hw/usb/bus.c
@@ -8,7 +8,6 @@
#include "monitor/monitor.h"
#include "trace.h"
#include "qemu/cutils.h"
-#include "migration/migration.h"
static void usb_bus_dev_print(Monitor *mon, DeviceState *qdev, int indent);
@@ -687,8 +686,6 @@ USBDevice *usbdevice_create(const char *cmdline)
const char *params;
int len;
USBDevice *dev;
- ObjectClass *klass;
- DeviceClass *dc;
params = strchr(cmdline,':');
if (params) {
@@ -723,22 +720,6 @@ USBDevice *usbdevice_create(const char *cmdline)
return NULL;
}
- klass = object_class_by_name(f->name);
- if (klass == NULL) {
- error_report("Device '%s' not found", f->name);
- return NULL;
- }
-
- dc = DEVICE_CLASS(klass);
-
- if (only_migratable) {
- if (dc->vmsd->unmigratable) {
- error_report("Device %s is not migratable, but --only-migratable "
- "was specified", f->name);
- return NULL;
- }
- }
-
if (f->usbdevice_init) {
dev = f->usbdevice_init(bus, params);
} else {
diff --git a/qdev-monitor.c b/qdev-monitor.c
index 549f45f..5f2fcdf 100644
--- a/qdev-monitor.c
+++ b/qdev-monitor.c
@@ -29,7 +29,6 @@
#include "qemu/error-report.h"
#include "qemu/help_option.h"
#include "sysemu/block-backend.h"
-#include "migration/migration.h"
/*
* Aliases were a bad idea from the start. Let's keep them
@@ -579,14 +578,6 @@ DeviceState *qdev_device_add(QemuOpts *opts, Error **errp)
return NULL;
}
- if (only_migratable) {
- if (dc->vmsd->unmigratable) {
- error_setg(errp, "Device %s is not migratable, but "
- "--only-migratable was specified", driver);
- return NULL;
- }
- }
-
/* find bus */
path = qemu_opt_get(opts, "bus");
if (path != NULL) {
--
2.6.2
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable
2017-02-04 12:21 [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable Ashijeet Acharya
@ 2017-02-04 15:44 ` Paolo Bonzini
2017-02-10 15:04 ` Dr. David Alan Gilbert
1 sibling, 0 replies; 5+ messages in thread
From: Paolo Bonzini @ 2017-02-04 15:44 UTC (permalink / raw)
To: Ashijeet Acharya, dgilbert; +Cc: peter.maydell, qemu-devel, quintela
On 04/02/2017 04:21, Ashijeet Acharya wrote:
> Commit a3a3d8c7 introduced a segfault bug while checking for
> 'dc->vmsd->unmigratable' which caused QEMU to crash when trying to add
> devices which do no set their 'dc->vmsd' yet while initialization.
> Place a 'dc->vmsd' check prior to it so that we do not segfault for
> such devices.
>
> NOTE: This doesn't compromise the functioning of --only-migratable
> option as all the unmigratable devices do set their 'dc->vmsd'.
>
> Also, move the only_migratable check inside device_set_realized() to avoid
> code duplication and fix the bug to display only_migratable error message
> only if the device constructuon is right.
>
> Signed-off-by: Ashijeet Acharya <ashijeetacharya@gmail.com>
> ---
> Changes is v3:
> - move only_migratable check inside device_set_realized() to avoid code
> - duplication
> - I have dropped Juan's R-b tag for this one
> Changes in v2:
> - place dc->vmsd check in hw/usb/bus.c as well
> ---
> hw/core/qdev.c | 10 ++++++++++
> hw/usb/bus.c | 19 -------------------
> qdev-monitor.c | 9 ---------
> 3 files changed, 10 insertions(+), 28 deletions(-)
>
> diff --git a/hw/core/qdev.c b/hw/core/qdev.c
> index 5783442..4438cbf 100644
> --- a/hw/core/qdev.c
> +++ b/hw/core/qdev.c
> @@ -37,6 +37,7 @@
> #include "hw/boards.h"
> #include "hw/sysbus.h"
> #include "qapi-event.h"
> +#include "migration/migration.h"
>
> int qdev_hotplug = 0;
> static bool qdev_hot_added = false;
> @@ -896,6 +897,15 @@ static void device_set_realized(Object *obj, bool value, Error **errp)
> }
>
> if (value && !dev->realized) {
> + if (only_migratable && dc->vmsd) {
> + if (dc->vmsd->unmigratable) {
> + error_setg(&local_err, "Device %s is not migratable, but "
> + "--only-migratable was specified",
> + object_get_typename(obj));
> + goto fail;
> + }
> + }
> +
> if (!obj->parent) {
> gchar *name = g_strdup_printf("device[%d]", unattached_count++);
>
> diff --git a/hw/usb/bus.c b/hw/usb/bus.c
> index 1dcc35c..25913ad 100644
> --- a/hw/usb/bus.c
> +++ b/hw/usb/bus.c
> @@ -8,7 +8,6 @@
> #include "monitor/monitor.h"
> #include "trace.h"
> #include "qemu/cutils.h"
> -#include "migration/migration.h"
>
> static void usb_bus_dev_print(Monitor *mon, DeviceState *qdev, int indent);
>
> @@ -687,8 +686,6 @@ USBDevice *usbdevice_create(const char *cmdline)
> const char *params;
> int len;
> USBDevice *dev;
> - ObjectClass *klass;
> - DeviceClass *dc;
>
> params = strchr(cmdline,':');
> if (params) {
> @@ -723,22 +720,6 @@ USBDevice *usbdevice_create(const char *cmdline)
> return NULL;
> }
>
> - klass = object_class_by_name(f->name);
> - if (klass == NULL) {
> - error_report("Device '%s' not found", f->name);
> - return NULL;
> - }
> -
> - dc = DEVICE_CLASS(klass);
> -
> - if (only_migratable) {
> - if (dc->vmsd->unmigratable) {
> - error_report("Device %s is not migratable, but --only-migratable "
> - "was specified", f->name);
> - return NULL;
> - }
> - }
> -
> if (f->usbdevice_init) {
> dev = f->usbdevice_init(bus, params);
> } else {
> diff --git a/qdev-monitor.c b/qdev-monitor.c
> index 549f45f..5f2fcdf 100644
> --- a/qdev-monitor.c
> +++ b/qdev-monitor.c
> @@ -29,7 +29,6 @@
> #include "qemu/error-report.h"
> #include "qemu/help_option.h"
> #include "sysemu/block-backend.h"
> -#include "migration/migration.h"
>
> /*
> * Aliases were a bad idea from the start. Let's keep them
> @@ -579,14 +578,6 @@ DeviceState *qdev_device_add(QemuOpts *opts, Error **errp)
> return NULL;
> }
>
> - if (only_migratable) {
> - if (dc->vmsd->unmigratable) {
> - error_setg(errp, "Device %s is not migratable, but "
> - "--only-migratable was specified", driver);
> - return NULL;
> - }
> - }
> -
> /* find bus */
> path = qemu_opt_get(opts, "bus");
> if (path != NULL) {
>
Reviewed-bv: Paolo Bonzini <pbonzinI@redhat.com>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable
2017-02-04 12:21 [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable Ashijeet Acharya
2017-02-04 15:44 ` Paolo Bonzini
@ 2017-02-10 15:04 ` Dr. David Alan Gilbert
2017-02-13 15:14 ` Dr. David Alan Gilbert
1 sibling, 1 reply; 5+ messages in thread
From: Dr. David Alan Gilbert @ 2017-02-10 15:04 UTC (permalink / raw)
To: Ashijeet Acharya; +Cc: quintela, peter.maydell, pbonzini, qemu-devel
* Ashijeet Acharya (ashijeetacharya@gmail.com) wrote:
> Commit a3a3d8c7 introduced a segfault bug while checking for
> 'dc->vmsd->unmigratable' which caused QEMU to crash when trying to add
> devices which do no set their 'dc->vmsd' yet while initialization.
> Place a 'dc->vmsd' check prior to it so that we do not segfault for
> such devices.
>
> NOTE: This doesn't compromise the functioning of --only-migratable
> option as all the unmigratable devices do set their 'dc->vmsd'.
>
> Also, move the only_migratable check inside device_set_realized() to avoid
> code duplication and fix the bug to display only_migratable error message
> only if the device constructuon is right.
>
> Signed-off-by: Ashijeet Acharya <ashijeetacharya@gmail.com>
Queued
> ---
> Changes is v3:
> - move only_migratable check inside device_set_realized() to avoid code
> - duplication
> - I have dropped Juan's R-b tag for this one
> Changes in v2:
> - place dc->vmsd check in hw/usb/bus.c as well
> ---
> hw/core/qdev.c | 10 ++++++++++
> hw/usb/bus.c | 19 -------------------
> qdev-monitor.c | 9 ---------
> 3 files changed, 10 insertions(+), 28 deletions(-)
>
> diff --git a/hw/core/qdev.c b/hw/core/qdev.c
> index 5783442..4438cbf 100644
> --- a/hw/core/qdev.c
> +++ b/hw/core/qdev.c
> @@ -37,6 +37,7 @@
> #include "hw/boards.h"
> #include "hw/sysbus.h"
> #include "qapi-event.h"
> +#include "migration/migration.h"
>
> int qdev_hotplug = 0;
> static bool qdev_hot_added = false;
> @@ -896,6 +897,15 @@ static void device_set_realized(Object *obj, bool value, Error **errp)
> }
>
> if (value && !dev->realized) {
> + if (only_migratable && dc->vmsd) {
> + if (dc->vmsd->unmigratable) {
> + error_setg(&local_err, "Device %s is not migratable, but "
> + "--only-migratable was specified",
> + object_get_typename(obj));
> + goto fail;
> + }
> + }
> +
> if (!obj->parent) {
> gchar *name = g_strdup_printf("device[%d]", unattached_count++);
>
> diff --git a/hw/usb/bus.c b/hw/usb/bus.c
> index 1dcc35c..25913ad 100644
> --- a/hw/usb/bus.c
> +++ b/hw/usb/bus.c
> @@ -8,7 +8,6 @@
> #include "monitor/monitor.h"
> #include "trace.h"
> #include "qemu/cutils.h"
> -#include "migration/migration.h"
>
> static void usb_bus_dev_print(Monitor *mon, DeviceState *qdev, int indent);
>
> @@ -687,8 +686,6 @@ USBDevice *usbdevice_create(const char *cmdline)
> const char *params;
> int len;
> USBDevice *dev;
> - ObjectClass *klass;
> - DeviceClass *dc;
>
> params = strchr(cmdline,':');
> if (params) {
> @@ -723,22 +720,6 @@ USBDevice *usbdevice_create(const char *cmdline)
> return NULL;
> }
>
> - klass = object_class_by_name(f->name);
> - if (klass == NULL) {
> - error_report("Device '%s' not found", f->name);
> - return NULL;
> - }
> -
> - dc = DEVICE_CLASS(klass);
> -
> - if (only_migratable) {
> - if (dc->vmsd->unmigratable) {
> - error_report("Device %s is not migratable, but --only-migratable "
> - "was specified", f->name);
> - return NULL;
> - }
> - }
> -
> if (f->usbdevice_init) {
> dev = f->usbdevice_init(bus, params);
> } else {
> diff --git a/qdev-monitor.c b/qdev-monitor.c
> index 549f45f..5f2fcdf 100644
> --- a/qdev-monitor.c
> +++ b/qdev-monitor.c
> @@ -29,7 +29,6 @@
> #include "qemu/error-report.h"
> #include "qemu/help_option.h"
> #include "sysemu/block-backend.h"
> -#include "migration/migration.h"
>
> /*
> * Aliases were a bad idea from the start. Let's keep them
> @@ -579,14 +578,6 @@ DeviceState *qdev_device_add(QemuOpts *opts, Error **errp)
> return NULL;
> }
>
> - if (only_migratable) {
> - if (dc->vmsd->unmigratable) {
> - error_setg(errp, "Device %s is not migratable, but "
> - "--only-migratable was specified", driver);
> - return NULL;
> - }
> - }
> -
> /* find bus */
> path = qemu_opt_get(opts, "bus");
> if (path != NULL) {
> --
> 2.6.2
>
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable
2017-02-10 15:04 ` Dr. David Alan Gilbert
@ 2017-02-13 15:14 ` Dr. David Alan Gilbert
2017-02-13 15:42 ` Paolo Bonzini
0 siblings, 1 reply; 5+ messages in thread
From: Dr. David Alan Gilbert @ 2017-02-13 15:14 UTC (permalink / raw)
To: Ashijeet Acharya; +Cc: quintela, peter.maydell, pbonzini, qemu-devel
* Dr. David Alan Gilbert (dgilbert@redhat.com) wrote:
> * Ashijeet Acharya (ashijeetacharya@gmail.com) wrote:
> > Commit a3a3d8c7 introduced a segfault bug while checking for
> > 'dc->vmsd->unmigratable' which caused QEMU to crash when trying to add
> > devices which do no set their 'dc->vmsd' yet while initialization.
> > Place a 'dc->vmsd' check prior to it so that we do not segfault for
> > such devices.
> >
> > NOTE: This doesn't compromise the functioning of --only-migratable
> > option as all the unmigratable devices do set their 'dc->vmsd'.
> >
> > Also, move the only_migratable check inside device_set_realized() to avoid
> > code duplication and fix the bug to display only_migratable error message
> > only if the device constructuon is right.
> >
> > Signed-off-by: Ashijeet Acharya <ashijeetacharya@gmail.com>
>
> Queued
I've had to remove this; it fails to build user-mode qemu's.
I think the problem is only_migratable is in vl.c and doesn't
get linked in user mode qemu; I tried adding a #ifndef CONFIG_USER_ONLY
around the block and it didn't help, so back to you.
Dave
> > ---
> > Changes is v3:
> > - move only_migratable check inside device_set_realized() to avoid code
> > - duplication
> > - I have dropped Juan's R-b tag for this one
> > Changes in v2:
> > - place dc->vmsd check in hw/usb/bus.c as well
> > ---
> > hw/core/qdev.c | 10 ++++++++++
> > hw/usb/bus.c | 19 -------------------
> > qdev-monitor.c | 9 ---------
> > 3 files changed, 10 insertions(+), 28 deletions(-)
> >
> > diff --git a/hw/core/qdev.c b/hw/core/qdev.c
> > index 5783442..4438cbf 100644
> > --- a/hw/core/qdev.c
> > +++ b/hw/core/qdev.c
> > @@ -37,6 +37,7 @@
> > #include "hw/boards.h"
> > #include "hw/sysbus.h"
> > #include "qapi-event.h"
> > +#include "migration/migration.h"
> >
> > int qdev_hotplug = 0;
> > static bool qdev_hot_added = false;
> > @@ -896,6 +897,15 @@ static void device_set_realized(Object *obj, bool value, Error **errp)
> > }
> >
> > if (value && !dev->realized) {
> > + if (only_migratable && dc->vmsd) {
> > + if (dc->vmsd->unmigratable) {
> > + error_setg(&local_err, "Device %s is not migratable, but "
> > + "--only-migratable was specified",
> > + object_get_typename(obj));
> > + goto fail;
> > + }
> > + }
> > +
> > if (!obj->parent) {
> > gchar *name = g_strdup_printf("device[%d]", unattached_count++);
> >
> > diff --git a/hw/usb/bus.c b/hw/usb/bus.c
> > index 1dcc35c..25913ad 100644
> > --- a/hw/usb/bus.c
> > +++ b/hw/usb/bus.c
> > @@ -8,7 +8,6 @@
> > #include "monitor/monitor.h"
> > #include "trace.h"
> > #include "qemu/cutils.h"
> > -#include "migration/migration.h"
> >
> > static void usb_bus_dev_print(Monitor *mon, DeviceState *qdev, int indent);
> >
> > @@ -687,8 +686,6 @@ USBDevice *usbdevice_create(const char *cmdline)
> > const char *params;
> > int len;
> > USBDevice *dev;
> > - ObjectClass *klass;
> > - DeviceClass *dc;
> >
> > params = strchr(cmdline,':');
> > if (params) {
> > @@ -723,22 +720,6 @@ USBDevice *usbdevice_create(const char *cmdline)
> > return NULL;
> > }
> >
> > - klass = object_class_by_name(f->name);
> > - if (klass == NULL) {
> > - error_report("Device '%s' not found", f->name);
> > - return NULL;
> > - }
> > -
> > - dc = DEVICE_CLASS(klass);
> > -
> > - if (only_migratable) {
> > - if (dc->vmsd->unmigratable) {
> > - error_report("Device %s is not migratable, but --only-migratable "
> > - "was specified", f->name);
> > - return NULL;
> > - }
> > - }
> > -
> > if (f->usbdevice_init) {
> > dev = f->usbdevice_init(bus, params);
> > } else {
> > diff --git a/qdev-monitor.c b/qdev-monitor.c
> > index 549f45f..5f2fcdf 100644
> > --- a/qdev-monitor.c
> > +++ b/qdev-monitor.c
> > @@ -29,7 +29,6 @@
> > #include "qemu/error-report.h"
> > #include "qemu/help_option.h"
> > #include "sysemu/block-backend.h"
> > -#include "migration/migration.h"
> >
> > /*
> > * Aliases were a bad idea from the start. Let's keep them
> > @@ -579,14 +578,6 @@ DeviceState *qdev_device_add(QemuOpts *opts, Error **errp)
> > return NULL;
> > }
> >
> > - if (only_migratable) {
> > - if (dc->vmsd->unmigratable) {
> > - error_setg(errp, "Device %s is not migratable, but "
> > - "--only-migratable was specified", driver);
> > - return NULL;
> > - }
> > - }
> > -
> > /* find bus */
> > path = qemu_opt_get(opts, "bus");
> > if (path != NULL) {
> > --
> > 2.6.2
> >
> >
> --
> Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable
2017-02-13 15:14 ` Dr. David Alan Gilbert
@ 2017-02-13 15:42 ` Paolo Bonzini
0 siblings, 0 replies; 5+ messages in thread
From: Paolo Bonzini @ 2017-02-13 15:42 UTC (permalink / raw)
To: Dr. David Alan Gilbert, Ashijeet Acharya
Cc: quintela, peter.maydell, qemu-devel
On 13/02/2017 16:14, Dr. David Alan Gilbert wrote:
> * Dr. David Alan Gilbert (dgilbert@redhat.com) wrote:
>> * Ashijeet Acharya (ashijeetacharya@gmail.com) wrote:
>>> Commit a3a3d8c7 introduced a segfault bug while checking for
>>> 'dc->vmsd->unmigratable' which caused QEMU to crash when trying to add
>>> devices which do no set their 'dc->vmsd' yet while initialization.
>>> Place a 'dc->vmsd' check prior to it so that we do not segfault for
>>> such devices.
>>>
>>> NOTE: This doesn't compromise the functioning of --only-migratable
>>> option as all the unmigratable devices do set their 'dc->vmsd'.
>>>
>>> Also, move the only_migratable check inside device_set_realized() to avoid
>>> code duplication and fix the bug to display only_migratable error message
>>> only if the device constructuon is right.
>>>
>>> Signed-off-by: Ashijeet Acharya <ashijeetacharya@gmail.com>
>>
>> Queued
>
> I've had to remove this; it fails to build user-mode qemu's.
> I think the problem is only_migratable is in vl.c and doesn't
> get linked in user mode qemu; I tried adding a #ifndef CONFIG_USER_ONLY
> around the block and it didn't help, so back to you.\
It doesn't help because the file is compiled just once. You can add it
to stubs/ (for example stubs/vmstate.c).
Paolo
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-02-13 15:42 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-02-04 12:21 [Qemu-devel] [PATCH v3] migrate: Introduce a 'dc->vmsd' check to avoid segfault for --only-migratable Ashijeet Acharya
2017-02-04 15:44 ` Paolo Bonzini
2017-02-10 15:04 ` Dr. David Alan Gilbert
2017-02-13 15:14 ` Dr. David Alan Gilbert
2017-02-13 15:42 ` Paolo Bonzini
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).