diff mbox series

virtio_pmem: Add freeze/restore callbacks

Message ID 20240815004617.2325269-1-philipchen@chromium.org (mailing list archive)
State New
Headers show
Series virtio_pmem: Add freeze/restore callbacks | expand

Commit Message

Philip Chen Aug. 15, 2024, 12:46 a.m. UTC
Add basic freeze/restore PM callbacks to support hibernation (S4):
- On freeze, delete vq and quiesce the device to prepare for
  snapshotting.
- On restore, re-init vq and mark DRIVER_OK.

Signed-off-by: Philip Chen <philipchen@chromium.org>
---
 drivers/nvdimm/virtio_pmem.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

Comments

Philip Chen Aug. 24, 2024, 2:06 a.m. UTC | #1
Hi maintainers,

Can anyone let me know if this patch makes sense?
Any comment/feedback is appreciated.
Thanks in advance!

On Wed, Aug 14, 2024 at 5:46 PM Philip Chen <philipchen@chromium.org> wrote:
>
> Add basic freeze/restore PM callbacks to support hibernation (S4):
> - On freeze, delete vq and quiesce the device to prepare for
>   snapshotting.
> - On restore, re-init vq and mark DRIVER_OK.
>
> Signed-off-by: Philip Chen <philipchen@chromium.org>
> ---
>  drivers/nvdimm/virtio_pmem.c | 24 ++++++++++++++++++++++++
>  1 file changed, 24 insertions(+)
>
> diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> index c9b97aeabf85..2396d19ce549 100644
> --- a/drivers/nvdimm/virtio_pmem.c
> +++ b/drivers/nvdimm/virtio_pmem.c
> @@ -143,6 +143,28 @@ static void virtio_pmem_remove(struct virtio_device *vdev)
>         virtio_reset_device(vdev);
>  }
>
> +static int virtio_pmem_freeze(struct virtio_device *vdev)
> +{
> +       vdev->config->del_vqs(vdev);
> +       virtio_reset_device(vdev);
> +
> +       return 0;
> +}
> +
> +static int virtio_pmem_restore(struct virtio_device *vdev)
> +{
> +       int ret;
> +
> +       ret = init_vq(vdev->priv);
> +       if (ret) {
> +               dev_err(&vdev->dev, "failed to initialize virtio pmem's vq\n");
> +               return ret;
> +       }
> +       virtio_device_ready(vdev);
> +
> +       return 0;
> +}
> +
>  static unsigned int features[] = {
>         VIRTIO_PMEM_F_SHMEM_REGION,
>  };
> @@ -155,6 +177,8 @@ static struct virtio_driver virtio_pmem_driver = {
>         .validate               = virtio_pmem_validate,
>         .probe                  = virtio_pmem_probe,
>         .remove                 = virtio_pmem_remove,
> +       .freeze                 = virtio_pmem_freeze,
> +       .restore                = virtio_pmem_restore,
>  };
>
>  module_virtio_driver(virtio_pmem_driver);
> --
> 2.46.0.76.ge559c4bf1a-goog
>
Ira Weiny Sept. 4, 2024, 5:53 p.m. UTC | #2
Philip Chen wrote:
> Hi maintainers,
> 
> Can anyone let me know if this patch makes sense?
> Any comment/feedback is appreciated.
> Thanks in advance!

I'm not an expert on virtio but the code looks ok on the surface.  I've
discussed this with Dan a bit and virtio-pmem is not heavily tested.

Based on our discussion [1] I wonder if there is a way we can recreate this
with QEMU to incorporate into our testing?

Ira

[1] https://lore.kernel.org/lkml/CA+cxXhnb2i5O7_BiOfKLth5Zwp5T62d6F6c39vnuT53cUkU_uw@mail.gmail.com/

> 
> On Wed, Aug 14, 2024 at 5:46 PM Philip Chen <philipchen@chromium.org> wrote:
> >
> > Add basic freeze/restore PM callbacks to support hibernation (S4):
> > - On freeze, delete vq and quiesce the device to prepare for
> >   snapshotting.
> > - On restore, re-init vq and mark DRIVER_OK.
> >
> > Signed-off-by: Philip Chen <philipchen@chromium.org>
> > ---
> >  drivers/nvdimm/virtio_pmem.c | 24 ++++++++++++++++++++++++
> >  1 file changed, 24 insertions(+)
> >
> > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > index c9b97aeabf85..2396d19ce549 100644
> > --- a/drivers/nvdimm/virtio_pmem.c
> > +++ b/drivers/nvdimm/virtio_pmem.c
> > @@ -143,6 +143,28 @@ static void virtio_pmem_remove(struct virtio_device *vdev)
> >         virtio_reset_device(vdev);
> >  }
> >
> > +static int virtio_pmem_freeze(struct virtio_device *vdev)
> > +{
> > +       vdev->config->del_vqs(vdev);
> > +       virtio_reset_device(vdev);
> > +
> > +       return 0;
> > +}
> > +
> > +static int virtio_pmem_restore(struct virtio_device *vdev)
> > +{
> > +       int ret;
> > +
> > +       ret = init_vq(vdev->priv);
> > +       if (ret) {
> > +               dev_err(&vdev->dev, "failed to initialize virtio pmem's vq\n");
> > +               return ret;
> > +       }
> > +       virtio_device_ready(vdev);
> > +
> > +       return 0;
> > +}
> > +
> >  static unsigned int features[] = {
> >         VIRTIO_PMEM_F_SHMEM_REGION,
> >  };
> > @@ -155,6 +177,8 @@ static struct virtio_driver virtio_pmem_driver = {
> >         .validate               = virtio_pmem_validate,
> >         .probe                  = virtio_pmem_probe,
> >         .remove                 = virtio_pmem_remove,
> > +       .freeze                 = virtio_pmem_freeze,
> > +       .restore                = virtio_pmem_restore,
> >  };
> >
> >  module_virtio_driver(virtio_pmem_driver);
> > --
> > 2.46.0.76.ge559c4bf1a-goog
> >
Philip Chen Sept. 9, 2024, 11:52 p.m. UTC | #3
Hi

On Wed, Sep 4, 2024 at 10:54 AM Ira Weiny <ira.weiny@intel.com> wrote:
>
> Philip Chen wrote:
> > Hi maintainers,
> >
> > Can anyone let me know if this patch makes sense?
> > Any comment/feedback is appreciated.
> > Thanks in advance!
>
> I'm not an expert on virtio but the code looks ok on the surface.  I've
> discussed this with Dan a bit and virtio-pmem is not heavily tested.

Thanks for your comments.
I think this specific patch is not heavily involved with virtio spec details.
This patch simply provides the basic freeze/restore PM callbacks for
virtio_pmem, like people already did for the other virtio drivers.

>
> Based on our discussion [1] I wonder if there is a way we can recreate this
> with QEMU to incorporate into our testing?

Yes, these are how I test on crosvm, but I believe the same steps can
be applied to QEMU:
(1) Set pm_test_level to TEST_PLATFORM (build time change)
(2) Write something to pmem
(3) Make the device go through a freeze/restore cycle by writing
`disk` to `/sys/power/state`
(4) Validate the data written to pmem in (2) is still preserved

Note:
(a) The freeze/restore PM routines are sometimes called as the backup
for suspend/resume PM routines in a suspend/resume cycle.
In this case, we can also test freeze/restore PM routines with
suspend/resume: i.e. skip (1) and write `mem` to `sys/power/state` in
(3).
(b) I also tried to set up QEMU for testing. But QEMU crashes when I
try to freeze the device even without applying this patch.
Since the issue seems to be irrelevant to pmem and most likely a QEMU
setup problem on my end, I didn't spend more time enabling QEMU.



>
> Ira
>
> [1] https://lore.kernel.org/lkml/CA+cxXhnb2i5O7_BiOfKLth5Zwp5T62d6F6c39vnuT53cUkU_uw@mail.gmail.com/
>
> >
> > On Wed, Aug 14, 2024 at 5:46 PM Philip Chen <philipchen@chromium.org> wrote:
> > >
> > > Add basic freeze/restore PM callbacks to support hibernation (S4):
> > > - On freeze, delete vq and quiesce the device to prepare for
> > >   snapshotting.
> > > - On restore, re-init vq and mark DRIVER_OK.
> > >
> > > Signed-off-by: Philip Chen <philipchen@chromium.org>
> > > ---
> > >  drivers/nvdimm/virtio_pmem.c | 24 ++++++++++++++++++++++++
> > >  1 file changed, 24 insertions(+)
> > >
> > > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > > index c9b97aeabf85..2396d19ce549 100644
> > > --- a/drivers/nvdimm/virtio_pmem.c
> > > +++ b/drivers/nvdimm/virtio_pmem.c
> > > @@ -143,6 +143,28 @@ static void virtio_pmem_remove(struct virtio_device *vdev)
> > >         virtio_reset_device(vdev);
> > >  }
> > >
> > > +static int virtio_pmem_freeze(struct virtio_device *vdev)
> > > +{
> > > +       vdev->config->del_vqs(vdev);
> > > +       virtio_reset_device(vdev);
> > > +
> > > +       return 0;
> > > +}
> > > +
> > > +static int virtio_pmem_restore(struct virtio_device *vdev)
> > > +{
> > > +       int ret;
> > > +
> > > +       ret = init_vq(vdev->priv);
> > > +       if (ret) {
> > > +               dev_err(&vdev->dev, "failed to initialize virtio pmem's vq\n");
> > > +               return ret;
> > > +       }
> > > +       virtio_device_ready(vdev);
> > > +
> > > +       return 0;
> > > +}
> > > +
> > >  static unsigned int features[] = {
> > >         VIRTIO_PMEM_F_SHMEM_REGION,
> > >  };
> > > @@ -155,6 +177,8 @@ static struct virtio_driver virtio_pmem_driver = {
> > >         .validate               = virtio_pmem_validate,
> > >         .probe                  = virtio_pmem_probe,
> > >         .remove                 = virtio_pmem_remove,
> > > +       .freeze                 = virtio_pmem_freeze,
> > > +       .restore                = virtio_pmem_restore,
> > >  };
> > >
> > >  module_virtio_driver(virtio_pmem_driver);
> > > --
> > > 2.46.0.76.ge559c4bf1a-goog
> > >
>
>
Philip Chen Sept. 23, 2024, 4:47 p.m. UTC | #4
Hi maintainers,

Are there any other concerns I should address for this patch?

On Mon, Sep 9, 2024 at 4:52 PM Philip Chen <philipchen@chromium.org> wrote:
>
> Hi
>
> On Wed, Sep 4, 2024 at 10:54 AM Ira Weiny <ira.weiny@intel.com> wrote:
> >
> > Philip Chen wrote:
> > > Hi maintainers,
> > >
> > > Can anyone let me know if this patch makes sense?
> > > Any comment/feedback is appreciated.
> > > Thanks in advance!
> >
> > I'm not an expert on virtio but the code looks ok on the surface.  I've
> > discussed this with Dan a bit and virtio-pmem is not heavily tested.
>
> Thanks for your comments.
> I think this specific patch is not heavily involved with virtio spec details.
> This patch simply provides the basic freeze/restore PM callbacks for
> virtio_pmem, like people already did for the other virtio drivers.
>
> >
> > Based on our discussion [1] I wonder if there is a way we can recreate this
> > with QEMU to incorporate into our testing?
>
> Yes, these are how I test on crosvm, but I believe the same steps can
> be applied to QEMU:
> (1) Set pm_test_level to TEST_PLATFORM (build time change)
> (2) Write something to pmem
> (3) Make the device go through a freeze/restore cycle by writing
> `disk` to `/sys/power/state`
> (4) Validate the data written to pmem in (2) is still preserved
>
> Note:
> (a) The freeze/restore PM routines are sometimes called as the backup
> for suspend/resume PM routines in a suspend/resume cycle.
> In this case, we can also test freeze/restore PM routines with
> suspend/resume: i.e. skip (1) and write `mem` to `sys/power/state` in
> (3).
> (b) I also tried to set up QEMU for testing. But QEMU crashes when I
> try to freeze the device even without applying this patch.
> Since the issue seems to be irrelevant to pmem and most likely a QEMU
> setup problem on my end, I didn't spend more time enabling QEMU.
>
>
>
> >
> > Ira
> >
> > [1] https://lore.kernel.org/lkml/CA+cxXhnb2i5O7_BiOfKLth5Zwp5T62d6F6c39vnuT53cUkU_uw@mail.gmail.com/
> >
> > >
> > > On Wed, Aug 14, 2024 at 5:46 PM Philip Chen <philipchen@chromium.org> wrote:
> > > >
> > > > Add basic freeze/restore PM callbacks to support hibernation (S4):
> > > > - On freeze, delete vq and quiesce the device to prepare for
> > > >   snapshotting.
> > > > - On restore, re-init vq and mark DRIVER_OK.
> > > >
> > > > Signed-off-by: Philip Chen <philipchen@chromium.org>
> > > > ---
> > > >  drivers/nvdimm/virtio_pmem.c | 24 ++++++++++++++++++++++++
> > > >  1 file changed, 24 insertions(+)
> > > >
> > > > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > > > index c9b97aeabf85..2396d19ce549 100644
> > > > --- a/drivers/nvdimm/virtio_pmem.c
> > > > +++ b/drivers/nvdimm/virtio_pmem.c
> > > > @@ -143,6 +143,28 @@ static void virtio_pmem_remove(struct virtio_device *vdev)
> > > >         virtio_reset_device(vdev);
> > > >  }
> > > >
> > > > +static int virtio_pmem_freeze(struct virtio_device *vdev)
> > > > +{
> > > > +       vdev->config->del_vqs(vdev);
> > > > +       virtio_reset_device(vdev);
> > > > +
> > > > +       return 0;
> > > > +}
> > > > +
> > > > +static int virtio_pmem_restore(struct virtio_device *vdev)
> > > > +{
> > > > +       int ret;
> > > > +
> > > > +       ret = init_vq(vdev->priv);
> > > > +       if (ret) {
> > > > +               dev_err(&vdev->dev, "failed to initialize virtio pmem's vq\n");
> > > > +               return ret;
> > > > +       }
> > > > +       virtio_device_ready(vdev);
> > > > +
> > > > +       return 0;
> > > > +}
> > > > +
> > > >  static unsigned int features[] = {
> > > >         VIRTIO_PMEM_F_SHMEM_REGION,
> > > >  };
> > > > @@ -155,6 +177,8 @@ static struct virtio_driver virtio_pmem_driver = {
> > > >         .validate               = virtio_pmem_validate,
> > > >         .probe                  = virtio_pmem_probe,
> > > >         .remove                 = virtio_pmem_remove,
> > > > +       .freeze                 = virtio_pmem_freeze,
> > > > +       .restore                = virtio_pmem_restore,
> > > >  };
> > > >
> > > >  module_virtio_driver(virtio_pmem_driver);
> > > > --
> > > > 2.46.0.76.ge559c4bf1a-goog
> > > >
> >
> >
Pankaj Gupta Sept. 25, 2024, 7:30 a.m. UTC | #5
+CC MST

> > Philip Chen wrote:
> > > Hi maintainers,
> > >
> > > Can anyone let me know if this patch makes sense?
> > > Any comment/feedback is appreciated.
> > > Thanks in advance!
> >
> > I'm not an expert on virtio but the code looks ok on the surface.  I've
> > discussed this with Dan a bit and virtio-pmem is not heavily tested.
>
> Thanks for your comments.
> I think this specific patch is not heavily involved with virtio spec details.
> This patch simply provides the basic freeze/restore PM callbacks for
> virtio_pmem, like people already did for the other virtio drivers.
>
> >
> > Based on our discussion [1] I wonder if there is a way we can recreate this
> > with QEMU to incorporate into our testing?
>
> Yes, these are how I test on crosvm, but I believe the same steps can
> be applied to QEMU:
> (1) Set pm_test_level to TEST_PLATFORM (build time change)
> (2) Write something to pmem
> (3) Make the device go through a freeze/restore cycle by writing
> `disk` to `/sys/power/state`
> (4) Validate the data written to pmem in (2) is still preserved
>
> Note:
> (a) The freeze/restore PM routines are sometimes called as the backup
> for suspend/resume PM routines in a suspend/resume cycle.
> In this case, we can also test freeze/restore PM routines with
> suspend/resume: i.e. skip (1) and write `mem` to `sys/power/state` in
> (3).
> (b) I also tried to set up QEMU for testing. But QEMU crashes when I
> try to freeze the device even without applying this patch.
> Since the issue seems to be irrelevant to pmem and most likely a QEMU
> setup problem on my end, I didn't spend more time enabling QEMU.

Thanks for the work!

Did you check why QEMU was crashing, maybe because we did not support
freeze/restore before
so some missing NULL check? Any back trace you got?

If it works in crossVM, it should work in Qemu as well?

Thanks,
Pankaj

>
>
>
> >
> > Ira
> >
> > [1] https://lore.kernel.org/lkml/CA+cxXhnb2i5O7_BiOfKLth5Zwp5T62d6F6c39vnuT53cUkU_uw@mail.gmail.com/
> >
> > >
> > > On Wed, Aug 14, 2024 at 5:46 PM Philip Chen <philipchen@chromium.org> wrote:
> > > >
> > > > Add basic freeze/restore PM callbacks to support hibernation (S4):
> > > > - On freeze, delete vq and quiesce the device to prepare for
> > > >   snapshotting.
> > > > - On restore, re-init vq and mark DRIVER_OK.
> > > >
> > > > Signed-off-by: Philip Chen <philipchen@chromium.org>
> > > > ---
> > > >  drivers/nvdimm/virtio_pmem.c | 24 ++++++++++++++++++++++++
> > > >  1 file changed, 24 insertions(+)
> > > >
> > > > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > > > index c9b97aeabf85..2396d19ce549 100644
> > > > --- a/drivers/nvdimm/virtio_pmem.c
> > > > +++ b/drivers/nvdimm/virtio_pmem.c
> > > > @@ -143,6 +143,28 @@ static void virtio_pmem_remove(struct virtio_device *vdev)
> > > >         virtio_reset_device(vdev);
> > > >  }
> > > >
> > > > +static int virtio_pmem_freeze(struct virtio_device *vdev)
> > > > +{
> > > > +       vdev->config->del_vqs(vdev);
> > > > +       virtio_reset_device(vdev);
> > > > +
> > > > +       return 0;
> > > > +}
> > > > +
> > > > +static int virtio_pmem_restore(struct virtio_device *vdev)
> > > > +{
> > > > +       int ret;
> > > > +
> > > > +       ret = init_vq(vdev->priv);
> > > > +       if (ret) {
> > > > +               dev_err(&vdev->dev, "failed to initialize virtio pmem's vq\n");
> > > > +               return ret;
> > > > +       }
> > > > +       virtio_device_ready(vdev);
> > > > +
> > > > +       return 0;
> > > > +}
> > > > +
> > > >  static unsigned int features[] = {
> > > >         VIRTIO_PMEM_F_SHMEM_REGION,
> > > >  };
> > > > @@ -155,6 +177,8 @@ static struct virtio_driver virtio_pmem_driver = {
> > > >         .validate               = virtio_pmem_validate,
> > > >         .probe                  = virtio_pmem_probe,
> > > >         .remove                 = virtio_pmem_remove,
> > > > +       .freeze                 = virtio_pmem_freeze,
> > > > +       .restore                = virtio_pmem_restore,
> > > >  };
> > > >
> > > >  module_virtio_driver(virtio_pmem_driver);
> > > > --
> > > > 2.46.0.76.ge559c4bf1a-goog
> > > >
> >
> >
diff mbox series

Patch

diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
index c9b97aeabf85..2396d19ce549 100644
--- a/drivers/nvdimm/virtio_pmem.c
+++ b/drivers/nvdimm/virtio_pmem.c
@@ -143,6 +143,28 @@  static void virtio_pmem_remove(struct virtio_device *vdev)
 	virtio_reset_device(vdev);
 }
 
+static int virtio_pmem_freeze(struct virtio_device *vdev)
+{
+	vdev->config->del_vqs(vdev);
+	virtio_reset_device(vdev);
+
+	return 0;
+}
+
+static int virtio_pmem_restore(struct virtio_device *vdev)
+{
+	int ret;
+
+	ret = init_vq(vdev->priv);
+	if (ret) {
+		dev_err(&vdev->dev, "failed to initialize virtio pmem's vq\n");
+		return ret;
+	}
+	virtio_device_ready(vdev);
+
+	return 0;
+}
+
 static unsigned int features[] = {
 	VIRTIO_PMEM_F_SHMEM_REGION,
 };
@@ -155,6 +177,8 @@  static struct virtio_driver virtio_pmem_driver = {
 	.validate		= virtio_pmem_validate,
 	.probe			= virtio_pmem_probe,
 	.remove			= virtio_pmem_remove,
+	.freeze			= virtio_pmem_freeze,
+	.restore		= virtio_pmem_restore,
 };
 
 module_virtio_driver(virtio_pmem_driver);