| Message ID | 1417067290-20715-1-git-send-email-david@gibson.dropbear.id.au |
|---|---|
| State | New |
| Headers | show |
On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote: > VirtIO devices now remember which endianness they're operating in in order > to support targets which may have guests of either endianness, such as > powerpc. This endianness state is transferred in a subsection of the > virtio device's information. > > With virtio-rng this can lead to an abort after a loadvm hitting the > assert() in virtio_is_big_endian(). This can be reproduced by doing a > migrate and load from file on a bi-endian target with a virtio-rng device. > The actual guest state isn't particularly important to triggering this. > > The cause is that virtio_rng_load_device() calls virtio_rng_process() which > accesses the ring and thus needs the endianness. However, > virtio_rng_process() is called via virtio_load() before it loads the > subsections. Essentially the ->load callback in VirtioDeviceClass should > only be used for actually reading the device state from the stream, not for > post-load re-initialization. Agreed. > This patch fixes the bug by moving the virtio_rng_process() after the call > to virtio_load(). Better yet would be to convert virtio to use vmsd and > have the virtio_rng_process() as a post_load callback, but that's a bigger > project for another day. > > This is bugfix, and should be considered for the 2.2 branch. This is undoing most of 3902d49e13c2428bd6381cfdf183103ca4477c1f , added Greg to CC list. Did you try this on x86 guests, or with multiple rng devices? (keeping context for Greg) > Signed-off-by: David Gibson <david@gibson.dropbear.id.au> > --- > hw/virtio/virtio-rng.c | 15 ++++++++------- > 1 file changed, 8 insertions(+), 7 deletions(-) > > diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c > index e85a979..473c044 100644 > --- a/hw/virtio/virtio-rng.c > +++ b/hw/virtio/virtio-rng.c > @@ -113,20 +113,22 @@ static void virtio_rng_save(QEMUFile *f, void *opaque) > > static int virtio_rng_load(QEMUFile *f, void *opaque, int version_id) > { > + VirtIORNG *vrng = opaque; > + int ret; > + > if (version_id != 1) { > return -EINVAL; > } > - return virtio_load(VIRTIO_DEVICE(opaque), f, version_id); > -} > + ret = virtio_load(VIRTIO_DEVICE(vrng), f, version_id); > + if (ret != 0) { > + return ret; > + } > > -static int virtio_rng_load_device(VirtIODevice *vdev, QEMUFile *f, > - int version_id) > -{ > /* We may have an element ready but couldn't process it due to a quota > * limit. Make sure to try again after live migration when the quota may > * have been reset. > */ > - virtio_rng_process(VIRTIO_RNG(vdev)); > + virtio_rng_process(vrng); > > return 0; > } > @@ -231,7 +233,6 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data) > vdc->realize = virtio_rng_device_realize; > vdc->unrealize = virtio_rng_device_unrealize; > vdc->get_features = get_features; > - vdc->load = virtio_rng_load_device; > } > > static void virtio_rng_initfn(Object *obj) Thanks, Amit
David Gibson <david@gibson.dropbear.id.au> writes: > VirtIO devices now remember which endianness they're operating in in order > to support targets which may have guests of either endianness, such as > powerpc. This endianness state is transferred in a subsection of the > virtio device's information. > > With virtio-rng this can lead to an abort after a loadvm hitting the > assert() in virtio_is_big_endian(). This can be reproduced by doing a > migrate and load from file on a bi-endian target with a virtio-rng device. > The actual guest state isn't particularly important to triggering this. > > The cause is that virtio_rng_load_device() calls virtio_rng_process() which > accesses the ring and thus needs the endianness. However, > virtio_rng_process() is called via virtio_load() before it loads the > subsections. Essentially the ->load callback in VirtioDeviceClass should > only be used for actually reading the device state from the stream, not for > post-load re-initialization. > > This patch fixes the bug by moving the virtio_rng_process() after the call > to virtio_load(). Better yet would be to convert virtio to use vmsd and > have the virtio_rng_process() as a post_load callback, but that's a bigger > project for another day. > > This is bugfix, and should be considered for the 2.2 branch. "[PATCH for-2.2]" would have been a good idea then. Next time :)
(actually adding Greg) On (Thu) 27 Nov 2014 [14:38:42], Amit Shah wrote: > On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote: > > VirtIO devices now remember which endianness they're operating in in order > > to support targets which may have guests of either endianness, such as > > powerpc. This endianness state is transferred in a subsection of the > > virtio device's information. > > > > With virtio-rng this can lead to an abort after a loadvm hitting the > > assert() in virtio_is_big_endian(). This can be reproduced by doing a > > migrate and load from file on a bi-endian target with a virtio-rng device. > > The actual guest state isn't particularly important to triggering this. > > > > The cause is that virtio_rng_load_device() calls virtio_rng_process() which > > accesses the ring and thus needs the endianness. However, > > virtio_rng_process() is called via virtio_load() before it loads the > > subsections. Essentially the ->load callback in VirtioDeviceClass should > > only be used for actually reading the device state from the stream, not for > > post-load re-initialization. > > Agreed. > > > This patch fixes the bug by moving the virtio_rng_process() after the call > > to virtio_load(). Better yet would be to convert virtio to use vmsd and > > have the virtio_rng_process() as a post_load callback, but that's a bigger > > project for another day. > > > > This is bugfix, and should be considered for the 2.2 branch. > > This is undoing most of 3902d49e13c2428bd6381cfdf183103ca4477c1f , > added Greg to CC list. > > Did you try this on x86 guests, or with multiple rng devices? > > (keeping context for Greg) > > > Signed-off-by: David Gibson <david@gibson.dropbear.id.au> > > --- > > hw/virtio/virtio-rng.c | 15 ++++++++------- > > 1 file changed, 8 insertions(+), 7 deletions(-) > > > > diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c > > index e85a979..473c044 100644 > > --- a/hw/virtio/virtio-rng.c > > +++ b/hw/virtio/virtio-rng.c > > @@ -113,20 +113,22 @@ static void virtio_rng_save(QEMUFile *f, void *opaque) > > > > static int virtio_rng_load(QEMUFile *f, void *opaque, int version_id) > > { > > + VirtIORNG *vrng = opaque; > > + int ret; > > + > > if (version_id != 1) { > > return -EINVAL; > > } > > - return virtio_load(VIRTIO_DEVICE(opaque), f, version_id); > > -} > > + ret = virtio_load(VIRTIO_DEVICE(vrng), f, version_id); > > + if (ret != 0) { > > + return ret; > > + } > > > > -static int virtio_rng_load_device(VirtIODevice *vdev, QEMUFile *f, > > - int version_id) > > -{ > > /* We may have an element ready but couldn't process it due to a quota > > * limit. Make sure to try again after live migration when the quota may > > * have been reset. > > */ > > - virtio_rng_process(VIRTIO_RNG(vdev)); > > + virtio_rng_process(vrng); > > > > return 0; > > } > > @@ -231,7 +233,6 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data) > > vdc->realize = virtio_rng_device_realize; > > vdc->unrealize = virtio_rng_device_unrealize; > > vdc->get_features = get_features; > > - vdc->load = virtio_rng_load_device; > > } > > > > static void virtio_rng_initfn(Object *obj) > > > Thanks, > > Amit Amit
On Thu, 27 Nov 2014 14:38:42 +0530 Amit Shah <amit.shah@redhat.com> wrote: > On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote: > > VirtIO devices now remember which endianness they're operating in in order > > to support targets which may have guests of either endianness, such as > > powerpc. This endianness state is transferred in a subsection of the > > virtio device's information. > > > > With virtio-rng this can lead to an abort after a loadvm hitting the > > assert() in virtio_is_big_endian(). This can be reproduced by doing a > > migrate and load from file on a bi-endian target with a virtio-rng device. > > The actual guest state isn't particularly important to triggering this. > > > > The cause is that virtio_rng_load_device() calls virtio_rng_process() which > > accesses the ring and thus needs the endianness. However, > > virtio_rng_process() is called via virtio_load() before it loads the > > subsections. Essentially the ->load callback in VirtioDeviceClass should > > only be used for actually reading the device state from the stream, not for > > post-load re-initialization. > > Agreed. > > > This patch fixes the bug by moving the virtio_rng_process() after the call > > to virtio_load(). Better yet would be to convert virtio to use vmsd and > > have the virtio_rng_process() as a post_load callback, but that's a bigger > > project for another day. > > I remember discussions on IRC last spring where I agreed I would work on it. :) > > This is bugfix, and should be considered for the 2.2 branch. > > This is undoing most of 3902d49e13c2428bd6381cfdf183103ca4477c1f , > added Greg to CC list. > This commit is indeed completely wrong: the load callback only makes sense when there's something to read which is obviously not the case here... This is definitely post load stuff :-\ Thanks ! :) Reviewed-by: Greg Kurz <gkurz@linux.vnet.ibm.com> -- Greg > Did you try this on x86 guests, or with multiple rng devices? > > (keeping context for Greg) > > > Signed-off-by: David Gibson <david@gibson.dropbear.id.au> > > --- > > hw/virtio/virtio-rng.c | 15 ++++++++------- > > 1 file changed, 8 insertions(+), 7 deletions(-) > > > > diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c > > index e85a979..473c044 100644 > > --- a/hw/virtio/virtio-rng.c > > +++ b/hw/virtio/virtio-rng.c > > @@ -113,20 +113,22 @@ static void virtio_rng_save(QEMUFile *f, void *opaque) > > > > static int virtio_rng_load(QEMUFile *f, void *opaque, int version_id) > > { > > + VirtIORNG *vrng = opaque; > > + int ret; > > + > > if (version_id != 1) { > > return -EINVAL; > > } > > - return virtio_load(VIRTIO_DEVICE(opaque), f, version_id); > > -} > > + ret = virtio_load(VIRTIO_DEVICE(vrng), f, version_id); > > + if (ret != 0) { > > + return ret; > > + } > > > > -static int virtio_rng_load_device(VirtIODevice *vdev, QEMUFile *f, > > - int version_id) > > -{ > > /* We may have an element ready but couldn't process it due to a quota > > * limit. Make sure to try again after live migration when the quota may > > * have been reset. > > */ > > - virtio_rng_process(VIRTIO_RNG(vdev)); > > + virtio_rng_process(vrng); > > > > return 0; > > } > > @@ -231,7 +233,6 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data) > > vdc->realize = virtio_rng_device_realize; > > vdc->unrealize = virtio_rng_device_unrealize; > > vdc->get_features = get_features; > > - vdc->load = virtio_rng_load_device; > > } > > > > static void virtio_rng_initfn(Object *obj) > > > Thanks, > > Amit >
On Thu, Nov 27, 2014 at 02:38:42PM +0530, Amit Shah wrote: > On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote: > > VirtIO devices now remember which endianness they're operating in in order > > to support targets which may have guests of either endianness, such as > > powerpc. This endianness state is transferred in a subsection of the > > virtio device's information. > > > > With virtio-rng this can lead to an abort after a loadvm hitting the > > assert() in virtio_is_big_endian(). This can be reproduced by doing a > > migrate and load from file on a bi-endian target with a virtio-rng device. > > The actual guest state isn't particularly important to triggering this. > > > > The cause is that virtio_rng_load_device() calls virtio_rng_process() which > > accesses the ring and thus needs the endianness. However, > > virtio_rng_process() is called via virtio_load() before it loads the > > subsections. Essentially the ->load callback in VirtioDeviceClass should > > only be used for actually reading the device state from the stream, not for > > post-load re-initialization. > > Agreed. > > > This patch fixes the bug by moving the virtio_rng_process() after the call > > to virtio_load(). Better yet would be to convert virtio to use vmsd and > > have the virtio_rng_process() as a post_load callback, but that's a bigger > > project for another day. > > > > This is bugfix, and should be considered for the 2.2 branch. > > This is undoing most of 3902d49e13c2428bd6381cfdf183103ca4477c1f , > added Greg to CC list. > > Did you try this on x86 guests, or with multiple rng devices? Not so far, I'll see what I can do.
On (Fri) 28 Nov 2014 [11:50:51], David Gibson wrote: > On Thu, Nov 27, 2014 at 02:38:42PM +0530, Amit Shah wrote: > > On (Thu) 27 Nov 2014 [16:48:10], David Gibson wrote: > > > VirtIO devices now remember which endianness they're operating in in order > > > to support targets which may have guests of either endianness, such as > > > powerpc. This endianness state is transferred in a subsection of the > > > virtio device's information. > > > > > > With virtio-rng this can lead to an abort after a loadvm hitting the > > > assert() in virtio_is_big_endian(). This can be reproduced by doing a > > > migrate and load from file on a bi-endian target with a virtio-rng device. > > > The actual guest state isn't particularly important to triggering this. > > > > > > The cause is that virtio_rng_load_device() calls virtio_rng_process() which > > > accesses the ring and thus needs the endianness. However, > > > virtio_rng_process() is called via virtio_load() before it loads the > > > subsections. Essentially the ->load callback in VirtioDeviceClass should > > > only be used for actually reading the device state from the stream, not for > > > post-load re-initialization. > > > > Agreed. > > > > > This patch fixes the bug by moving the virtio_rng_process() after the call > > > to virtio_load(). Better yet would be to convert virtio to use vmsd and > > > have the virtio_rng_process() as a post_load callback, but that's a bigger > > > project for another day. > > > > > > This is bugfix, and should be considered for the 2.2 branch. > > > > This is undoing most of 3902d49e13c2428bd6381cfdf183103ca4477c1f , > > added Greg to CC list. > > > > Did you try this on x86 guests, or with multiple rng devices? > > Not so far, I'll see what I can do. Thanks. Amit
On 27 November 2014 at 09:26, Markus Armbruster <armbru@redhat.com> wrote: > David Gibson <david@gibson.dropbear.id.au> writes: > >> VirtIO devices now remember which endianness they're operating in in order >> to support targets which may have guests of either endianness, such as >> powerpc. This endianness state is transferred in a subsection of the >> virtio device's information. >> >> With virtio-rng this can lead to an abort after a loadvm hitting the >> assert() in virtio_is_big_endian(). This can be reproduced by doing a >> migrate and load from file on a bi-endian target with a virtio-rng device. >> The actual guest state isn't particularly important to triggering this. >> >> The cause is that virtio_rng_load_device() calls virtio_rng_process() which >> accesses the ring and thus needs the endianness. However, >> virtio_rng_process() is called via virtio_load() before it loads the >> subsections. Essentially the ->load callback in VirtioDeviceClass should >> only be used for actually reading the device state from the stream, not for >> post-load re-initialization. >> >> This patch fixes the bug by moving the virtio_rng_process() after the call >> to virtio_load(). Better yet would be to convert virtio to use vmsd and >> have the virtio_rng_process() as a post_load callback, but that's a bigger >> project for another day. >> >> This is bugfix, and should be considered for the 2.2 branch. > > "[PATCH for-2.2]" would have been a good idea then. Next time :) So do you want this patch in 2.2? I was planning to put in the virtio-vs-xen fixes today and tag rc4, so it's not too late if you're confident this patch is good. Let me know if you think it should go in, and I can apply it to master directly. -- PMM
On Fri, Nov 28, 2014 at 09:14:46AM +0000, Peter Maydell wrote: > On 27 November 2014 at 09:26, Markus Armbruster <armbru@redhat.com> wrote: > > David Gibson <david@gibson.dropbear.id.au> writes: > > > >> VirtIO devices now remember which endianness they're operating in in order > >> to support targets which may have guests of either endianness, such as > >> powerpc. This endianness state is transferred in a subsection of the > >> virtio device's information. > >> > >> With virtio-rng this can lead to an abort after a loadvm hitting the > >> assert() in virtio_is_big_endian(). This can be reproduced by doing a > >> migrate and load from file on a bi-endian target with a virtio-rng device. > >> The actual guest state isn't particularly important to triggering this. > >> > >> The cause is that virtio_rng_load_device() calls virtio_rng_process() which > >> accesses the ring and thus needs the endianness. However, > >> virtio_rng_process() is called via virtio_load() before it loads the > >> subsections. Essentially the ->load callback in VirtioDeviceClass should > >> only be used for actually reading the device state from the stream, not for > >> post-load re-initialization. > >> > >> This patch fixes the bug by moving the virtio_rng_process() after the call > >> to virtio_load(). Better yet would be to convert virtio to use vmsd and > >> have the virtio_rng_process() as a post_load callback, but that's a bigger > >> project for another day. > >> > >> This is bugfix, and should be considered for the 2.2 branch. > > > > "[PATCH for-2.2]" would have been a good idea then. Next time :) > > So do you want this patch in 2.2? I was planning to put in the > virtio-vs-xen fixes today and tag rc4, so it's not too late if you're > confident this patch is good. Let me know if you think it should go in, > and I can apply it to master directly. Yes, I think it should be applied.
On Fri, 28 Nov 2014 09:14:46 +0000 Peter Maydell <peter.maydell@linaro.org> wrote: > On 27 November 2014 at 09:26, Markus Armbruster <armbru@redhat.com> wrote: > > David Gibson <david@gibson.dropbear.id.au> writes: > > > >> VirtIO devices now remember which endianness they're operating in in order > >> to support targets which may have guests of either endianness, such as > >> powerpc. This endianness state is transferred in a subsection of the > >> virtio device's information. > >> > >> With virtio-rng this can lead to an abort after a loadvm hitting the > >> assert() in virtio_is_big_endian(). This can be reproduced by doing a > >> migrate and load from file on a bi-endian target with a virtio-rng device. > >> The actual guest state isn't particularly important to triggering this. > >> > >> The cause is that virtio_rng_load_device() calls virtio_rng_process() which > >> accesses the ring and thus needs the endianness. However, > >> virtio_rng_process() is called via virtio_load() before it loads the > >> subsections. Essentially the ->load callback in VirtioDeviceClass should > >> only be used for actually reading the device state from the stream, not for > >> post-load re-initialization. > >> > >> This patch fixes the bug by moving the virtio_rng_process() after the call > >> to virtio_load(). Better yet would be to convert virtio to use vmsd and > >> have the virtio_rng_process() as a post_load callback, but that's a bigger > >> project for another day. > >> > >> This is bugfix, and should be considered for the 2.2 branch. > > > > "[PATCH for-2.2]" would have been a good idea then. Next time :) > > So do you want this patch in 2.2? I was planning to put in the > virtio-vs-xen fixes today and tag rc4, so it's not too late if you're > confident this patch is good. Let me know if you think it should go in, > and I can apply it to master directly. > > -- PMM > Peter, FWIW I think it should. Commit 3902d49e13c2428bd6381cfdf183103ca4477c1f is clearly bad: virtio-rng does not need the .load callback obviously... and the fact it breaks migration makes it even worse... :( Please apply to 2.2. Cheers. -- Greg
On 28 November 2014 at 11:47, Greg Kurz <gkurz@linux.vnet.ibm.com> wrote: > FWIW I think it should. Commit 3902d49e13c2428bd6381cfdf183103ca4477c1f is > clearly bad: virtio-rng does not need the .load callback obviously... and > the fact it breaks migration makes it even worse... :( > > Please apply to 2.2. Applied to master, thanks. -- PMM
diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c index e85a979..473c044 100644 --- a/hw/virtio/virtio-rng.c +++ b/hw/virtio/virtio-rng.c @@ -113,20 +113,22 @@ static void virtio_rng_save(QEMUFile *f, void *opaque) static int virtio_rng_load(QEMUFile *f, void *opaque, int version_id) { + VirtIORNG *vrng = opaque; + int ret; + if (version_id != 1) { return -EINVAL; } - return virtio_load(VIRTIO_DEVICE(opaque), f, version_id); -} + ret = virtio_load(VIRTIO_DEVICE(vrng), f, version_id); + if (ret != 0) { + return ret; + } -static int virtio_rng_load_device(VirtIODevice *vdev, QEMUFile *f, - int version_id) -{ /* We may have an element ready but couldn't process it due to a quota * limit. Make sure to try again after live migration when the quota may * have been reset. */ - virtio_rng_process(VIRTIO_RNG(vdev)); + virtio_rng_process(vrng); return 0; } @@ -231,7 +233,6 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data) vdc->realize = virtio_rng_device_realize; vdc->unrealize = virtio_rng_device_unrealize; vdc->get_features = get_features; - vdc->load = virtio_rng_load_device; } static void virtio_rng_initfn(Object *obj)
VirtIO devices now remember which endianness they're operating in in order to support targets which may have guests of either endianness, such as powerpc. This endianness state is transferred in a subsection of the virtio device's information. With virtio-rng this can lead to an abort after a loadvm hitting the assert() in virtio_is_big_endian(). This can be reproduced by doing a migrate and load from file on a bi-endian target with a virtio-rng device. The actual guest state isn't particularly important to triggering this. The cause is that virtio_rng_load_device() calls virtio_rng_process() which accesses the ring and thus needs the endianness. However, virtio_rng_process() is called via virtio_load() before it loads the subsections. Essentially the ->load callback in VirtioDeviceClass should only be used for actually reading the device state from the stream, not for post-load re-initialization. This patch fixes the bug by moving the virtio_rng_process() after the call to virtio_load(). Better yet would be to convert virtio to use vmsd and have the virtio_rng_process() as a post_load callback, but that's a bigger project for another day. This is bugfix, and should be considered for the 2.2 branch. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> --- hw/virtio/virtio-rng.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-)