virtio-balloon: free page hinting fixes
PATCH 1/3 virtio-balloon: fix free page hinting without an iothread
In reply to David HildenbrandIn case we don't have an iothread, we mark the feature as abscent but still add the queue. 'free_page_bh' remains set to NULL.
qemu-system-i386 \
-M microvm \
-nographic \
-device virtio-balloon-device,free-page-hint=true \
-nographic \
-display none \
-monitor none \
-serial none \
-qtest stdioDoing a "write 0xc0000e30 0x24 0x030000000300000003000000030000000300000003000000030000000300000003000000"
We will trigger a SEGFAULT. Let's move the check and bail out.
While at it, move the static initializations to instance_initialize().
--- hw/virtio/virtio-balloon.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-)
hw/virtio/virtio-balloon.c +18 -17
@@ -789,6 +789,13 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)return ; } +if (virtio_has_feature (s->host_features, +VIRTIO_BALLOON_F_FREE_PAGE_HINT ) && !s->iothread) { +error_setg (errp,"'free-page-hint' requires 'iothread' to be set" ); +virtio_cleanup (vdev); +return ; + } + s->ivq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->dvq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->svq =virtio_add_queue (vdev,128 , virtio_balloon_receive_stats);
@@ -797,24 +804,11 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp) VIRTIO_BALLOON_F_FREE_PAGE_HINT)) { s->free_page_vq =virtio_add_queue (vdev,VIRTQUEUE_MAX_SIZE , virtio_balloon_handle_free_page_vq); - s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; - s->free_page_report_cmd_id = -VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; - s->free_page_report_notify.notify = - virtio_balloon_free_page_report_notify;precopy_add_notifier (&s->free_page_report_notify); -if (s->iothread) { -object_ref (OBJECT (s->iothread)); - s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), - virtio_ballloon_get_free_page_hints, s); -qemu_mutex_init (&s->free_page_lock); -qemu_cond_init (&s->free_page_cond); - s->block_iothread = false; - }else { -/* Simply disable this feature if the iothread wasn't created. */ - s->host_features &= ~(1 <<VIRTIO_BALLOON_F_FREE_PAGE_HINT ); -virtio_error (vdev,"iothread is missing" ); - } + +object_ref (OBJECT (s->iothread)); + s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), + virtio_ballloon_get_free_page_hints, s); }reset_stats (s); }
@@ -892,6 +886,13 @@ static void virtio_balloon_instance_init(Object *obj) {VirtIOBalloon *s =VIRTIO_BALLOON (obj); +qemu_mutex_init (&s->free_page_lock); +qemu_cond_init (&s->free_page_cond); + s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; + s->free_page_report_cmd_id =VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; + s->free_page_report_notify.notify = virtio_balloon_free_page_report_notify; + s->block_iothread = false; +object_property_add (obj,"guest-stats" ,"guest statistics" , balloon_stats_get_all,NULL ,NULL , s);
Signature · 1 line
2.25.4
cover virtio-balloon: free page hinting fixes
Some fixes for VIRTIO_BALLOON_F_FREE_PAGE_HINT. First issue was reported by Alexander Bulekov [1], the other ones were discovered by me when digging into the details.
[1] https://lkml.kernel.org/r/[email protected]
David Hildenbrand (3):
virtio-balloon: fix free page hinting without an iothread
virtio-balloon: fix free page hinting check on unrealize
virtio-balloon: unref the iothread when unrealizing
hw/virtio/virtio-balloon.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-)
Signature · 1 line
2.25.4
PATCH 3/3 virtio-balloon: unref the iothread when unrealizing
We took a reference when realizing, so let's drop that reference when unrealizing.
--- hw/virtio/virtio-balloon.c | 1 + 1 file changed, 1 insertion(+)
hw/virtio/virtio-balloon.c +1 -0
@@ -820,6 +820,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh); +object_unref (OBJECT (s->iothread));virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify); }
Signature · 1 line
2.25.4
PATCH 2/3 virtio-balloon: fix free page hinting check on unrealize
In reply to David HildenbrandChecking against guest features is wrong. We allocated data structures based on host features. We can rely on "free_page_bh" as an indicator whether to un-do stuff instead.
--- hw/virtio/virtio-balloon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
hw/virtio/virtio-balloon.c +1 -1
@@ -818,7 +818,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)VirtIODevice *vdev =VIRTIO_DEVICE (dev);VirtIOBalloon *s =VIRTIO_BALLOON (dev); -if (virtio_balloon_free_page_support (s)) { +if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh);virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify);
Signature · 1 line
2.25.4
Re: [PATCH v1 1/3] virtio-balloon: fix free page hinting without an iothread
In reply to David HildenbrandQuoted 36 lines · David Hildenbrand
In case we don't have an iothread, we mark the feature as abscent but still add the queue. '
free_page_bh' remains set to NULL.qemu-system-i386 \ -M microvm \ -nographic \ -device virtio-balloon-device,free-page-hint=true \ -nographic \ -display none \ -monitor none \ -serial none \ -qtest stdioDoing a "write
0xc0000e300x240x030000000300000003000000030000000300000003000000030000000300000003000000"We will trigger a SEGFAULT. Let's move the check and bail out.
While at it, move the static initializations to
instance_initialize().Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Reported-by Alexander Bulekov <[email protected]>Cc Wei Wang <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Cc Alexander Duyck <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-)hw/virtio/virtio-balloon.c +0
-0
hw/virtio/virtio-balloon.c static void virtio_balloon_device_realize(DeviceState *dev, Error **errp) · line 796
Quoted 8 lines
hw/virtio/virtio-balloon.c +5
-0@@ -789,6 +789,13 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)return ; } +if (virtio_has_feature (s->host_features, +VIRTIO_BALLOON_F_FREE_PAGE_HINT ) && !s->iothread) { +error_setg (errp,"'free-page-hint' requires 'iothread' to be set" ); +virtio_cleanup (vdev); +return ;
This is exactly what I had in mind!
Quoted 35 lines
hw/virtio/virtio-balloon.c +6
-17+ } + s->ivq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->dvq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->svq =virtio_add_queue (vdev,128 , virtio_balloon_receive_stats);@@ -797,24 +804,11 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp) VIRTIO_BALLOON_F_FREE_PAGE_HINT)) { s->free_page_vq =virtio_add_queue (vdev,VIRTQUEUE_MAX_SIZE , virtio_balloon_handle_free_page_vq); - s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; - s->free_page_report_cmd_id = -VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; - s->free_page_report_notify.notify = - virtio_balloon_free_page_report_notify;precopy_add_notifier (&s->free_page_report_notify); -if (s->iothread) { -object_ref (OBJECT (s->iothread)); - s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), - virtio_ballloon_get_free_page_hints, s); -qemu_mutex_init (&s->free_page_lock); -qemu_cond_init (&s->free_page_cond); - s->block_iothread = false; - }else { -/* Simply disable this feature if the iothread wasn't created. */ - s->host_features &= ~(1 <<VIRTIO_BALLOON_F_FREE_PAGE_HINT ); -virtio_error (vdev,"iothread is missing" ); - } + +object_ref (OBJECT (s->iothread)); + s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), + virtio_ballloon_get_free_page_hints, s); }reset_stats (s); }
hw/virtio/virtio-balloon.c static void virtio_balloon_instance_init(Object *obj) · line 894
Quoted 9 lines
hw/virtio/virtio-balloon.c +6
-0@@ -892,6 +886,13 @@ static void virtio_balloon_instance_init(Object *obj) {VirtIOBalloon *s =VIRTIO_BALLOON (obj); +qemu_mutex_init (&s->free_page_lock); +qemu_cond_init (&s->free_page_cond); + s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; + s->free_page_report_cmd_id =VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; + s->free_page_report_notify.notify = virtio_balloon_free_page_report_notify; + s->block_iothread = false;
This part is even cleaner.
Quoted 5 lines
hw/virtio/virtio-balloon.c +1
-0+object_property_add (obj,"guest-stats" ,"guest statistics" , balloon_stats_get_all,NULL ,NULL , s);
Re: [PATCH v1 3/3] virtio-balloon: unref the iothread when unrealizing
In reply to David HildenbrandQuoted 19 lines · David Hildenbrand
We took a reference when realizing, so let's drop that reference when unrealizing.
Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Alexander Duyck <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 1 + 1 file changed, 1 insertion(+)hw/virtio/virtio-balloon.c +0
-0
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 826
Quoted 7 lines
hw/virtio/virtio-balloon.c +1
-0@@ -820,6 +820,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh); +object_unref (OBJECT (s->iothread));virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify); }
Quoted 6 lines · David Hildenbrand
Some fixes for
VIRTIO_BALLOON_F_FREE_PAGE_HINT. First issue was reported by Alexander Bulekov [1], the other ones were discovered by me when digging into the details.
Quoted 8 lines
David Hildenbrand (3):
virtio-balloon: fix free page hinting without an iothread
virtio-balloon: fix free page hinting check on unrealize
virtio-balloon: unref the iothread when unrealizinghw/virtio/virtio-balloon.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-)
All this series deserves a 'Cc: [email protected]' tag, right?
Re: [PATCH v1 2/3] virtio-balloon: fix free page hinting check on unrealize
In reply to David HildenbrandQuoted 20 lines · David Hildenbrand
Checking against guest features is wrong. We allocated data structures based on host features. We can rely on "
free_page_bh" as an indicator whether to un-do stuff instead.Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Cc Alexander Duyck <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)hw/virtio/virtio-balloon.c +0
-0
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 824
Quoted 8 lines
hw/virtio/virtio-balloon.c +1
-1@@ -818,7 +818,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)VirtIODevice *vdev =VIRTIO_DEVICE (dev);VirtIOBalloon *s =VIRTIO_BALLOON (dev); -if (virtio_balloon_free_page_support (s)) { +if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh);virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify);
Quoted 12 lines · Philippe Mathieu-Daudé
Quoted 11 lines · David Hildenbrand
Some fixes for
VIRTIO_BALLOON_F_FREE_PAGE_HINT. First issue was reported by Alexander Bulekov [1], the other ones were discovered by me when digging into the details.[1] https://lkml.kernel.org/r/[email protected]
David Hildenbrand (3):
virtio-balloon: fix free page hinting without an iothread
virtio-balloon: fix free page hinting check on unrealize
virtio-balloon: unref the iothread when unrealizing
Quoted 5 lines
Quoted 3 lines
hw/virtio/virtio-balloon.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-)All this series deserves a 'Cc: [email protected]' tag, right?
I was asking myself the same question. People usually don't unplug the balloon device, but yes, #2 and #3 can be triggered by the user. Most probably yes, cc stable for all.
Signature · 3 lines
Thanks, David / dhildenb
Re: [PATCH v1 1/3] virtio-balloon: fix free page hinting without an iothread
In reply to David HildenbrandQuoted 83 lines · David Hildenbrand
In case we don't have an iothread, we mark the feature as abscent but still add the queue. '
free_page_bh' remains set to NULL.qemu-system-i386 \ -M microvm \ -nographic \ -device virtio-balloon-device,free-page-hint=true \ -nographic \ -display none \ -monitor none \ -serial none \ -qtest stdioDoing a "write
0xc0000e300x240x030000000300000003000000030000000300000003000000030000000300000003000000"We will trigger a SEGFAULT. Let's move the check and bail out.
While at it, move the static initializations to
instance_initialize().Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Reported-by Alexander Bulekov <[email protected]>Cc Wei Wang <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Cc Alexander Duyck <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-)hw/virtio/virtio-balloon.c +11
-17@@ -789,6 +789,13 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)return ; } +if (virtio_has_feature (s->host_features, +VIRTIO_BALLOON_F_FREE_PAGE_HINT ) && !s->iothread) { +error_setg (errp,"'free-page-hint' requires 'iothread' to be set" ); +virtio_cleanup (vdev); +return ; + } + s->ivq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->dvq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->svq =virtio_add_queue (vdev,128 , virtio_balloon_receive_stats);@@ -797,24 +804,11 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp) VIRTIO_BALLOON_F_FREE_PAGE_HINT)) { s->free_page_vq =virtio_add_queue (vdev,VIRTQUEUE_MAX_SIZE , virtio_balloon_handle_free_page_vq); - s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; - s->free_page_report_cmd_id = -VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; - s->free_page_report_notify.notify = - virtio_balloon_free_page_report_notify;precopy_add_notifier (&s->free_page_report_notify); -if (s->iothread) { -object_ref (OBJECT (s->iothread)); - s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), - virtio_ballloon_get_free_page_hints, s); -qemu_mutex_init (&s->free_page_lock); -qemu_cond_init (&s->free_page_cond); - s->block_iothread = false; - }else { -/* Simply disable this feature if the iothread wasn't created. */ - s->host_features &= ~(1 <<VIRTIO_BALLOON_F_FREE_PAGE_HINT ); -virtio_error (vdev,"iothread is missing" ); - } + +object_ref (OBJECT (s->iothread)); + s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), + virtio_ballloon_get_free_page_hints, s); }reset_stats (s); }@@ -892,6 +886,13 @@ static void virtio_balloon_instance_init(Object *obj) {VirtIOBalloon *s =VIRTIO_BALLOON (obj);
hw/virtio/virtio-balloon.c static void virtio_balloon_instance_init(Object *obj) · line 898
Quoted 10 lines
hw/virtio/virtio-balloon.c +7
-0+qemu_mutex_init (&s->free_page_lock); +qemu_cond_init (&s->free_page_cond); + s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; + s->free_page_report_cmd_id =VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; + s->free_page_report_notify.notify = virtio_balloon_free_page_report_notify; + s->block_iothread = false; +object_property_add (obj,"guest-stats" ,"guest statistics" , balloon_stats_get_all,NULL ,NULL , s);
So the one nit I have is that I am not sure you need to bother with initializing block_iothread since it should already be initialized to false/0 shouldn't it? Otherwise this all looks good to me.
Re: [PATCH v1 2/3] virtio-balloon: fix free page hinting check on unrealize
In reply to David HildenbrandQuoted 20 lines · David Hildenbrand
Checking against guest features is wrong. We allocated data structures based on host features. We can rely on "
free_page_bh" as an indicator whether to un-do stuff instead.Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Cc Alexander Duyck <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)hw/virtio/virtio-balloon.c +0
-0
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 824
Quoted 8 lines
hw/virtio/virtio-balloon.c +1
-1@@ -818,7 +818,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)VirtIODevice *vdev =VIRTIO_DEVICE (dev);VirtIOBalloon *s =VIRTIO_BALLOON (dev); -if (virtio_balloon_free_page_support (s)) { +if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh);virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify);
Would it make sense to apply the same change to virtio_balloon_device_reset and virtio_balloon_set_status? At least in the case of virtio_balloon_set_status it seems like you could possibly exploit it somehow as clearing the feature in the guest will prevent the toggling of the block_iothread value.
Re: [PATCH v1 1/3] virtio-balloon: fix free page hinting without an iothread
In reply to Alexander DuyckQuoted 84 lines · Alexander Duyck
Quoted 83 lines · David Hildenbrand
In case we don't have an iothread, we mark the feature as abscent but still add the queue. '
free_page_bh' remains set to NULL.qemu-system-i386 \ -M microvm \ -nographic \ -device virtio-balloon-device,free-page-hint=true \ -nographic \ -display none \ -monitor none \ -serial none \ -qtest stdioDoing a "write
0xc0000e300x240x030000000300000003000000030000000300000003000000030000000300000003000000"We will trigger a SEGFAULT. Let's move the check and bail out.
While at it, move the static initializations to
instance_initialize().Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Reported-by Alexander Bulekov <[email protected]>Cc Wei Wang <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Cc Alexander Duyck <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-)hw/virtio/virtio-balloon.c +11
-17@@ -789,6 +789,13 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)return ; } +if (virtio_has_feature (s->host_features, +VIRTIO_BALLOON_F_FREE_PAGE_HINT ) && !s->iothread) { +error_setg (errp,"'free-page-hint' requires 'iothread' to be set" ); +virtio_cleanup (vdev); +return ; + } + s->ivq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->dvq =virtio_add_queue (vdev,128 , virtio_balloon_handle_output); s->svq =virtio_add_queue (vdev,128 , virtio_balloon_receive_stats);@@ -797,24 +804,11 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp) VIRTIO_BALLOON_F_FREE_PAGE_HINT)) { s->free_page_vq =virtio_add_queue (vdev,VIRTQUEUE_MAX_SIZE , virtio_balloon_handle_free_page_vq); - s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; - s->free_page_report_cmd_id = -VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; - s->free_page_report_notify.notify = - virtio_balloon_free_page_report_notify;precopy_add_notifier (&s->free_page_report_notify); -if (s->iothread) { -object_ref (OBJECT (s->iothread)); - s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), - virtio_ballloon_get_free_page_hints, s); -qemu_mutex_init (&s->free_page_lock); -qemu_cond_init (&s->free_page_cond); - s->block_iothread = false; - }else { -/* Simply disable this feature if the iothread wasn't created. */ - s->host_features &= ~(1 <<VIRTIO_BALLOON_F_FREE_PAGE_HINT ); -virtio_error (vdev,"iothread is missing" ); - } + +object_ref (OBJECT (s->iothread)); + s->free_page_bh =aio_bh_new (iothread_get_aio_context (s->iothread), + virtio_ballloon_get_free_page_hints, s); }reset_stats (s); }@@ -892,6 +886,13 @@ static void virtio_balloon_instance_init(Object *obj) {VirtIOBalloon *s =VIRTIO_BALLOON (obj);
Quoted 14 lines
hw/virtio/virtio-balloon.c static void virtio_balloon_instance_init(Object *obj) · line 898
Quoted 10 lines
hw/virtio/virtio-balloon.c +7
-0+qemu_mutex_init (&s->free_page_lock); +qemu_cond_init (&s->free_page_cond); + s->free_page_report_status =FREE_PAGE_REPORT_S_STOP ; + s->free_page_report_cmd_id =VIRTIO_BALLOON_FREE_PAGE_REPORT_CMD_ID_MIN ; + s->free_page_report_notify.notify = virtio_balloon_free_page_report_notify; + s->block_iothread = false; +object_property_add (obj,"guest-stats" ,"guest statistics" , balloon_stats_get_all,NULL ,NULL , s);So the one nit I have is that I am not sure you need to bother with initializing
block_iothreadsince it should already be initialized to false/0 shouldn't it? Otherwise this all looks good to me.
Yes, and as "FREE_PAGE_REPORT_S_STOP = 0", that's implicitly set as well. Both can go IMHO.
Quoted 3 lines
Reviewed-by Alexander Duyck <[email protected]>
Thanks!
Signature · 3 lines
Thanks, David / dhildenb
Re: [PATCH v1 3/3] virtio-balloon: unref the iothread when unrealizing
In reply to David HildenbrandQuoted 19 lines · David Hildenbrand
We took a reference when realizing, so let's drop that reference when unrealizing.
Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Alexander Duyck <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 1 + 1 file changed, 1 insertion(+)hw/virtio/virtio-balloon.c +0
-0
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 826
Quoted 7 lines
hw/virtio/virtio-balloon.c +1
-0@@ -820,6 +820,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh); +object_unref (OBJECT (s->iothread));virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify); }
I'm not entirely sure about this order of operations. It seems like it would make more sense to remove the notifier, stop the hinting, delete the bh, and then release the IO thread.
Re: [PATCH v1 3/3] virtio-balloon: unref the iothread when unrealizing
Quoted 21 lines · Alexander Duyck
Quoted 20 lines · David Hildenbrand
We took a reference when realizing, so let's drop that reference when unrealizing.
Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Alexander Duyck <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 1 + 1 file changed, 1 insertion(+)hw/virtio/virtio-balloon.c +0
-0@@ -820,6 +820,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)
Quoted 10 lines
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 826
Quoted 6 lines
hw/virtio/virtio-balloon.c +1
-0if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh); +object_unref (OBJECT (s->iothread));virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify); }I'm not entirely sure about this order of operations. It seems like it would make more sense to remove the notifier, stop the hinting, delete the bh, and then release the IO thread.
This is the reverse order of the steps in virtio_balloon_device_realize(). And I guess it should be fine. The notifier cannot really be active/trigger while we are removing devices (cannot happen with concurrent migration). After qemu_bh_delete(), the iothread is effectively unused.
I am unsure about many things regarding free page hinting (e.g., if the virtio_balloon_free_page_stop() is of any use while we are ripping out the device and it will be gone in a second).
Signature · 3 lines
Thanks, David / dhildenb
Re: [PATCH v1 2/3] virtio-balloon: fix free page hinting check on unrealize
In reply to Alexander DuyckQuoted 24 lines · Alexander Duyck
Quoted 23 lines · David Hildenbrand
Checking against guest features is wrong. We allocated data structures based on host features. We can rely on "
free_page_bh" as an indicator whether to un-do stuff instead.Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Cc Alexander Duyck <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)hw/virtio/virtio-balloon.c +0
-0@@ -818,7 +818,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)VirtIODevice *vdev =VIRTIO_DEVICE (dev);VirtIOBalloon *s =VIRTIO_BALLOON (dev);
Quoted 8 lines
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 824
Quoted 5 lines
hw/virtio/virtio-balloon.c +1
-1-if (virtio_balloon_free_page_support (s)) { +if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh);virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify);Would it make sense to apply the same change to
virtio_balloon_device_resetandvirtio_balloon_set_status? At least in
Good question ...
Quoted 3 lines
the case of
virtio_balloon_set_statusit seems like you could possibly exploit it somehow as clearing the feature in the guest will prevent the toggling of theblock_iothreadvalue.
The guest cannot change features at random points in time. However, what happens if we are stopped and trigger a system_reset? The the vdev features are stale ... I'll play with it, see if this is required, and add a separate patch if necessary!
Quoted 3 lines
Reviewed-by Alexander Duyck <[email protected]>
Thanks!
Signature · 3 lines
Thanks, David / dhildenb
Re: [PATCH v1 3/3] virtio-balloon: unref the iothread when unrealizing
In reply to David Hildenbrand+Stefan/Paolo in case they have a better clue.
Quoted 43 lines · David Hildenbrand
Quoted 21 lines · Alexander Duyck
Quoted 20 lines · David Hildenbrand
We took a reference when realizing, so let's drop that reference when unrealizing.
Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Alexander Duyck <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 1 + 1 file changed, 1 insertion(+)hw/virtio/virtio-balloon.c +0
-0@@ -820,6 +820,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)Quoted 10 lines
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 826
Quoted 6 lines
hw/virtio/virtio-balloon.c +1
-0if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh); +object_unref (OBJECT (s->iothread));virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify); }I'm not entirely sure about this order of operations. It seems like it would make more sense to remove the notifier, stop the hinting, delete the bh, and then release the IO thread.
This is the reverse order of the steps in
virtio_balloon_device_realize(). And I guess it should be fine. The notifier cannot really be active/trigger while we are removing devices (cannot happen with concurrent migration). Afterqemu_bh_delete(), the iothread is effectively unused.I am unsure about many things regarding free page hinting (e.g., if the
virtio_balloon_free_page_stop()is of any use while we are ripping out the device and it will be gone in a second).
Alexander Duyck
Reviewed-byRe: [PATCH v1 3/3] virtio-balloon: unref the iothread when unrealizing
In reply to David HildenbrandQuoted 40 lines · David Hildenbrand
Quoted 21 lines · Alexander Duyck
Quoted 20 lines · David Hildenbrand
We took a reference when realizing, so let's drop that reference when unrealizing.
Fixes c13c4153f76d ("virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_HINT")Cc Wei Wang <[email protected]>Cc Alexander Duyck <[email protected]>Cc Michael S. Tsirkin <[email protected]>Cc Philippe Mathieu-Daudé <[email protected]>Signed-off-by David Hildenbrand <[email protected]>--- hw/virtio/virtio-balloon.c | 1 + 1 file changed, 1 insertion(+)hw/virtio/virtio-balloon.c +0
-0@@ -820,6 +820,7 @@ static void virtio_balloon_device_unrealize(DeviceState *dev)Quoted 10 lines
hw/virtio/virtio-balloon.c static void virtio_balloon_device_unrealize(DeviceState *dev) · line 826
Quoted 6 lines
hw/virtio/virtio-balloon.c +1
-0if (s->free_page_bh) {qemu_bh_delete (s->free_page_bh); +object_unref (OBJECT (s->iothread));virtio_balloon_free_page_stop (s);precopy_remove_notifier (&s->free_page_report_notify); }I'm not entirely sure about this order of operations. It seems like it would make more sense to remove the notifier, stop the hinting, delete the bh, and then release the IO thread.
This is the reverse order of the steps in
virtio_balloon_device_realize(). And I guess it should be fine. The notifier cannot really be active/trigger while we are removing devices (cannot happen with concurrent migration). Afterqemu_bh_delete(), the iothread is effectively unused.
Quoted 3 lines
I am unsure about many things regarding free page hinting (e.g., if the
virtio_balloon_free_page_stop()is of any use while we are ripping out the device and it will be gone in a second).
Agreed. This is probably fine as is.