2020-07-31 11:48:32

by Arnaud POULIQUEN

[permalink] [raw]
Subject: [PATCH 8/9] rpmsg: virtio: use rpmsg_ns driver to manage ns announcement

Use the new rpmsg_ns API to send the name service announcements if
the VIRTIO_RPMSG_F_NS is set, else just not implement the ops.

Signed-off-by: Arnaud Pouliquen <[email protected]>
---
drivers/rpmsg/virtio_rpmsg_bus.c | 94 +++++---------------------------
1 file changed, 13 insertions(+), 81 deletions(-)

diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
index f771fdae150e..3c771a6392be 100644
--- a/drivers/rpmsg/virtio_rpmsg_bus.c
+++ b/drivers/rpmsg/virtio_rpmsg_bus.c
@@ -91,35 +91,6 @@ struct rpmsg_hdr {
u8 data[];
} __packed;

-/**
- * struct rpmsg_ns_msg - dynamic name service announcement message
- * @name: name of remote service that is published
- * @addr: address of remote service that is published
- * @flags: indicates whether service is created or destroyed
- *
- * This message is sent across to publish a new service, or announce
- * about its removal. When we receive these messages, an appropriate
- * rpmsg channel (i.e device) is created/destroyed. In turn, the ->probe()
- * or ->remove() handler of the appropriate rpmsg driver will be invoked
- * (if/as-soon-as one is registered).
- */
-struct rpmsg_ns_msg {
- char name[RPMSG_NAME_SIZE];
- __virtio32 addr;
- __virtio32 flags;
-} __packed;
-
-/**
- * enum rpmsg_ns_flags - dynamic name service announcement flags
- *
- * @RPMSG_NS_CREATE: a new remote service was just created
- * @RPMSG_NS_DESTROY: a known remote service was just destroyed
- */
-enum rpmsg_ns_flags {
- RPMSG_NS_CREATE = 0,
- RPMSG_NS_DESTROY = 1,
-};
-
/**
* @vrp: the remote processor this channel belongs to
*/
@@ -324,60 +295,18 @@ static void virtio_rpmsg_destroy_ept(struct rpmsg_endpoint *ept)
__rpmsg_destroy_ept(vch->vrp, ept);
}

-static int virtio_rpmsg_announce_create(struct rpmsg_device *rpdev)
-{
- struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
- struct virtproc_info *vrp = vch->vrp;
- struct device *dev = &rpdev->dev;
- int err = 0;
-
- /* need to tell remote processor's name service about this channel ? */
- if (rpdev->announce && rpdev->ept &&
- virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
- struct rpmsg_ns_msg nsm;
-
- strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
- nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
- nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_CREATE);
-
- err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
- if (err)
- dev_err(dev, "failed to announce service %d\n", err);
- }
-
- return err;
-}
-
-static int virtio_rpmsg_announce_destroy(struct rpmsg_device *rpdev)
-{
- struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
- struct virtproc_info *vrp = vch->vrp;
- struct device *dev = &rpdev->dev;
- int err = 0;
-
- /* tell remote processor's name service we're removing this channel */
- if (rpdev->announce && rpdev->ept &&
- virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
- struct rpmsg_ns_msg nsm;
-
- strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
- nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
- nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_DESTROY);
-
- err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
- if (err)
- dev_err(dev, "failed to announce service %d\n", err);
- }
-
- return err;
-}
-
static const struct rpmsg_device_ops virtio_rpmsg_ops = {
.create_channel = virtio_rpmsg_create_channel,
.release_channel = virtio_rpmsg_release_channel,
.create_ept = virtio_rpmsg_create_ept,
- .announce_create = virtio_rpmsg_announce_create,
- .announce_destroy = virtio_rpmsg_announce_destroy,
+};
+
+static const struct rpmsg_device_ops virtio_rpmsg_w_nsa_ops = {
+ .create_channel = virtio_rpmsg_create_channel,
+ .release_channel = virtio_rpmsg_release_channel,
+ .create_ept = virtio_rpmsg_create_ept,
+ .announce_create = rpmsg_ns_announce_create,
+ .announce_destroy = rpmsg_ns_announce_destroy,
};

static void virtio_rpmsg_release_device(struct device *dev)
@@ -423,7 +352,10 @@ __rpmsg_create_channel(struct virtproc_info *vrp,
rpdev = &vch->rpdev;
rpdev->src = chinfo->src;
rpdev->dst = chinfo->dst;
- rpdev->ops = &virtio_rpmsg_ops;
+ if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS))
+ rpdev->ops = &virtio_rpmsg_w_nsa_ops;
+ else
+ rpdev->ops = &virtio_rpmsg_ops;

/*
* rpmsg server channels has predefined local address (for now),
@@ -933,7 +865,7 @@ static int rpmsg_probe(struct virtio_device *vdev)

/* Assign public information to the rpmsg_device */
rpdev_ns = &vch->rpdev;
- rpdev_ns->ops = &virtio_rpmsg_ops;
+ rpdev_ns->ops = &virtio_rpmsg_w_nsa_ops;

rpdev_ns->dev.parent = &vrp->vdev->dev;
rpdev_ns->dev.release = virtio_rpmsg_release_device;
--
2.17.1


2020-08-25 16:57:25

by Mathieu Poirier

[permalink] [raw]
Subject: Re: [PATCH 8/9] rpmsg: virtio: use rpmsg_ns driver to manage ns announcement

Hi Arnaud,

On Fri, Jul 31, 2020 at 01:47:31PM +0200, Arnaud Pouliquen wrote:
> Use the new rpmsg_ns API to send the name service announcements if
> the VIRTIO_RPMSG_F_NS is set, else just not implement the ops.
>
> Signed-off-by: Arnaud Pouliquen <[email protected]>
> ---
> drivers/rpmsg/virtio_rpmsg_bus.c | 94 +++++---------------------------
> 1 file changed, 13 insertions(+), 81 deletions(-)
>
> diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
> index f771fdae150e..3c771a6392be 100644
> --- a/drivers/rpmsg/virtio_rpmsg_bus.c
> +++ b/drivers/rpmsg/virtio_rpmsg_bus.c
> @@ -91,35 +91,6 @@ struct rpmsg_hdr {
> u8 data[];
> } __packed;
>
> -/**
> - * struct rpmsg_ns_msg - dynamic name service announcement message
> - * @name: name of remote service that is published
> - * @addr: address of remote service that is published
> - * @flags: indicates whether service is created or destroyed
> - *
> - * This message is sent across to publish a new service, or announce
> - * about its removal. When we receive these messages, an appropriate
> - * rpmsg channel (i.e device) is created/destroyed. In turn, the ->probe()
> - * or ->remove() handler of the appropriate rpmsg driver will be invoked
> - * (if/as-soon-as one is registered).
> - */
> -struct rpmsg_ns_msg {
> - char name[RPMSG_NAME_SIZE];
> - __virtio32 addr;
> - __virtio32 flags;
> -} __packed;
> -
> -/**
> - * enum rpmsg_ns_flags - dynamic name service announcement flags
> - *
> - * @RPMSG_NS_CREATE: a new remote service was just created
> - * @RPMSG_NS_DESTROY: a known remote service was just destroyed
> - */
> -enum rpmsg_ns_flags {
> - RPMSG_NS_CREATE = 0,
> - RPMSG_NS_DESTROY = 1,
> -};
> -
> /**
> * @vrp: the remote processor this channel belongs to
> */
> @@ -324,60 +295,18 @@ static void virtio_rpmsg_destroy_ept(struct rpmsg_endpoint *ept)
> __rpmsg_destroy_ept(vch->vrp, ept);
> }
>
> -static int virtio_rpmsg_announce_create(struct rpmsg_device *rpdev)
> -{
> - struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
> - struct virtproc_info *vrp = vch->vrp;
> - struct device *dev = &rpdev->dev;
> - int err = 0;
> -
> - /* need to tell remote processor's name service about this channel ? */
> - if (rpdev->announce && rpdev->ept &&
> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> - struct rpmsg_ns_msg nsm;
> -
> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
> - nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
> - nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_CREATE);
> -
> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
> - if (err)
> - dev_err(dev, "failed to announce service %d\n", err);
> - }
> -
> - return err;
> -}
> -
> -static int virtio_rpmsg_announce_destroy(struct rpmsg_device *rpdev)
> -{
> - struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
> - struct virtproc_info *vrp = vch->vrp;
> - struct device *dev = &rpdev->dev;
> - int err = 0;
> -
> - /* tell remote processor's name service we're removing this channel */
> - if (rpdev->announce && rpdev->ept &&
> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> - struct rpmsg_ns_msg nsm;
> -
> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
> - nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
> - nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_DESTROY);
> -
> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
> - if (err)
> - dev_err(dev, "failed to announce service %d\n", err);
> - }
> -
> - return err;
> -}
> -
> static const struct rpmsg_device_ops virtio_rpmsg_ops = {
> .create_channel = virtio_rpmsg_create_channel,
> .release_channel = virtio_rpmsg_release_channel,
> .create_ept = virtio_rpmsg_create_ept,
> - .announce_create = virtio_rpmsg_announce_create,
> - .announce_destroy = virtio_rpmsg_announce_destroy,
> +};
> +
> +static const struct rpmsg_device_ops virtio_rpmsg_w_nsa_ops = {
> + .create_channel = virtio_rpmsg_create_channel,
> + .release_channel = virtio_rpmsg_release_channel,
> + .create_ept = virtio_rpmsg_create_ept,
> + .announce_create = rpmsg_ns_announce_create,
> + .announce_destroy = rpmsg_ns_announce_destroy,
> };
>
> static void virtio_rpmsg_release_device(struct device *dev)
> @@ -423,7 +352,10 @@ __rpmsg_create_channel(struct virtproc_info *vrp,
> rpdev = &vch->rpdev;
> rpdev->src = chinfo->src;
> rpdev->dst = chinfo->dst;
> - rpdev->ops = &virtio_rpmsg_ops;
> + if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS))
> + rpdev->ops = &virtio_rpmsg_w_nsa_ops;
> + else
> + rpdev->ops = &virtio_rpmsg_ops;

Yesterday I struggled with this part and I still do this morning. Function
__rpmsg_create_channel() can only be called if VIRTIO_RPMSG_F_NS is set so there
is no need to check it again. I would also have expected this patch to be a
simple replace of the .announce_create/destroy functions. Adding an ops that
doesn't have the .announce_create/destroy functions looks like a feature to me,
and one that I don't quite get.

Do you think you could expand on the motivation behind this patch?

Thanks,
Mathieu

>
> /*
> * rpmsg server channels has predefined local address (for now),
> @@ -933,7 +865,7 @@ static int rpmsg_probe(struct virtio_device *vdev)
>
> /* Assign public information to the rpmsg_device */
> rpdev_ns = &vch->rpdev;
> - rpdev_ns->ops = &virtio_rpmsg_ops;
> + rpdev_ns->ops = &virtio_rpmsg_w_nsa_ops;
>
> rpdev_ns->dev.parent = &vrp->vdev->dev;
> rpdev_ns->dev.release = virtio_rpmsg_release_device;
> --
> 2.17.1
>

2020-08-26 07:43:54

by Arnaud POULIQUEN

[permalink] [raw]
Subject: Re: [PATCH 8/9] rpmsg: virtio: use rpmsg_ns driver to manage ns announcement

Hi mathieu,

I Sent my V2 few seconds before receiving your comment :)
Please find my answer below

On 8/25/20 6:54 PM, Mathieu Poirier wrote:
> Hi Arnaud,
>
> On Fri, Jul 31, 2020 at 01:47:31PM +0200, Arnaud Pouliquen wrote:
>> Use the new rpmsg_ns API to send the name service announcements if
>> the VIRTIO_RPMSG_F_NS is set, else just not implement the ops.
>>
>> Signed-off-by: Arnaud Pouliquen <[email protected]>
>> ---
>> drivers/rpmsg/virtio_rpmsg_bus.c | 94 +++++---------------------------
>> 1 file changed, 13 insertions(+), 81 deletions(-)
>>
>> diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
>> index f771fdae150e..3c771a6392be 100644
>> --- a/drivers/rpmsg/virtio_rpmsg_bus.c
>> +++ b/drivers/rpmsg/virtio_rpmsg_bus.c
>> @@ -91,35 +91,6 @@ struct rpmsg_hdr {
>> u8 data[];
>> } __packed;
>>
>> -/**
>> - * struct rpmsg_ns_msg - dynamic name service announcement message
>> - * @name: name of remote service that is published
>> - * @addr: address of remote service that is published
>> - * @flags: indicates whether service is created or destroyed
>> - *
>> - * This message is sent across to publish a new service, or announce
>> - * about its removal. When we receive these messages, an appropriate
>> - * rpmsg channel (i.e device) is created/destroyed. In turn, the ->probe()
>> - * or ->remove() handler of the appropriate rpmsg driver will be invoked
>> - * (if/as-soon-as one is registered).
>> - */
>> -struct rpmsg_ns_msg {
>> - char name[RPMSG_NAME_SIZE];
>> - __virtio32 addr;
>> - __virtio32 flags;
>> -} __packed;
>> -
>> -/**
>> - * enum rpmsg_ns_flags - dynamic name service announcement flags
>> - *
>> - * @RPMSG_NS_CREATE: a new remote service was just created
>> - * @RPMSG_NS_DESTROY: a known remote service was just destroyed
>> - */
>> -enum rpmsg_ns_flags {
>> - RPMSG_NS_CREATE = 0,
>> - RPMSG_NS_DESTROY = 1,
>> -};
>> -
>> /**
>> * @vrp: the remote processor this channel belongs to
>> */
>> @@ -324,60 +295,18 @@ static void virtio_rpmsg_destroy_ept(struct rpmsg_endpoint *ept)
>> __rpmsg_destroy_ept(vch->vrp, ept);
>> }
>>
>> -static int virtio_rpmsg_announce_create(struct rpmsg_device *rpdev)
>> -{
>> - struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
>> - struct virtproc_info *vrp = vch->vrp;
>> - struct device *dev = &rpdev->dev;
>> - int err = 0;
>> -
>> - /* need to tell remote processor's name service about this channel ? */
>> - if (rpdev->announce && rpdev->ept &&
>> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
>> - struct rpmsg_ns_msg nsm;
>> -
>> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
>> - nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
>> - nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_CREATE);
>> -
>> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
>> - if (err)
>> - dev_err(dev, "failed to announce service %d\n", err);
>> - }
>> -
>> - return err;
>> -}
>> -
>> -static int virtio_rpmsg_announce_destroy(struct rpmsg_device *rpdev)
>> -{
>> - struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
>> - struct virtproc_info *vrp = vch->vrp;
>> - struct device *dev = &rpdev->dev;
>> - int err = 0;
>> -
>> - /* tell remote processor's name service we're removing this channel */
>> - if (rpdev->announce && rpdev->ept &&
>> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
>> - struct rpmsg_ns_msg nsm;
>> -
>> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
>> - nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
>> - nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_DESTROY);
>> -
>> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
>> - if (err)
>> - dev_err(dev, "failed to announce service %d\n", err);
>> - }
>> -
>> - return err;
>> -}
>> -
>> static const struct rpmsg_device_ops virtio_rpmsg_ops = {
>> .create_channel = virtio_rpmsg_create_channel,
>> .release_channel = virtio_rpmsg_release_channel,
>> .create_ept = virtio_rpmsg_create_ept,
>> - .announce_create = virtio_rpmsg_announce_create,
>> - .announce_destroy = virtio_rpmsg_announce_destroy,
>> +};
>> +
>> +static const struct rpmsg_device_ops virtio_rpmsg_w_nsa_ops = {
>> + .create_channel = virtio_rpmsg_create_channel,
>> + .release_channel = virtio_rpmsg_release_channel,
>> + .create_ept = virtio_rpmsg_create_ept,
>> + .announce_create = rpmsg_ns_announce_create,
>> + .announce_destroy = rpmsg_ns_announce_destroy,
>> };
>>
>> static void virtio_rpmsg_release_device(struct device *dev)
>> @@ -423,7 +352,10 @@ __rpmsg_create_channel(struct virtproc_info *vrp,
>> rpdev = &vch->rpdev;
>> rpdev->src = chinfo->src;
>> rpdev->dst = chinfo->dst;
>> - rpdev->ops = &virtio_rpmsg_ops;
>> + if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS))
>> + rpdev->ops = &virtio_rpmsg_w_nsa_ops;
>> + else
>> + rpdev->ops = &virtio_rpmsg_ops;
>
> Yesterday I struggled with this part and I still do this morning. Function
> __rpmsg_create_channel() can only be called if VIRTIO_RPMSG_F_NS is set so there
> is no need to check it again.

That's right if rpmsg_create_channel is called by the rpmsg_ns only. But it
could not be the case in future, for instance with the rpmsg_ctrl series [1]
As the channel creation is decorrelate from the ns annoucement we need to check it.

[1]https://patchwork.kernel.org/patch/11694787/

I would also have expected this patch to be a
> simple replace of the .announce_create/destroy functions. Adding an ops that
> doesn't have the .announce_create/destroy functions looks like a feature to me,
> and one that I don't quite get.
>
> Do you think you could expand on the motivation behind this patch?

It was my first implementation. It is more of a phylosophical point than anything else.
With this path i tried to implement the following rule:
if VIRTIO_RPMSG_F_NS is not set
no ns announcement support
else
delegate to the ns announcement RPMsg service

Rather, legacy implementation is about to implement the announce ops even if not supported.

If this implementation is confusing i can go back to my first implementation which was
only an update the virtio_rpmsg_announce_xx functions:

@@ -322,15 +304,8 @@ static int virtio_rpmsg_announce_create(struct rpmsg_device *rpdev)
int err = 0;

/* need to tell remote processor's name service about this channel ? */
- if (rpdev->announce && rpdev->ept &&
- virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
- struct rpmsg_ns_msg nsm;
-
- strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
- nsm.addr = rpdev->ept->addr;
- nsm.flags = RPMSG_NS_CREATE;
-
- err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
+ if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
+ err = rpmsg_ctrl_channel_announce(rpdev, RPMSG_NS_CREATE);
if (err)
dev_err(dev, "failed to announce service %d\n", err);
}
@@ -346,15 +321,8 @@ static int virtio_rpmsg_announce_destroy(struct rpmsg_device *rpdev)
int err = 0;

/* tell remote processor's name service we're removing this channel */
- if (rpdev->announce && rpdev->ept &&
- virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
- struct rpmsg_ns_msg nsm;
-
- strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
- nsm.addr = rpdev->ept->addr;
- nsm.flags = RPMSG_NS_DESTROY;
-
- err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
+ if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
+ err = rpmsg_ctrl_channel_announce(rpdev, RPMSG_NS_DESTROY);
if (err)
dev_err(dev, "failed to announce service %d\n", err);
}

Regards,
Arnaud

>
> Thanks,
> Mathieu
>
>>
>> /*
>> * rpmsg server channels has predefined local address (for now),
>> @@ -933,7 +865,7 @@ static int rpmsg_probe(struct virtio_device *vdev)
>>
>> /* Assign public information to the rpmsg_device */
>> rpdev_ns = &vch->rpdev;
>> - rpdev_ns->ops = &virtio_rpmsg_ops;
>> + rpdev_ns->ops = &virtio_rpmsg_w_nsa_ops;
>>
>> rpdev_ns->dev.parent = &vrp->vdev->dev;
>> rpdev_ns->dev.release = virtio_rpmsg_release_device;
>> --
>> 2.17.1
>>

2020-08-26 22:11:37

by Mathieu Poirier

[permalink] [raw]
Subject: Re: [PATCH 8/9] rpmsg: virtio: use rpmsg_ns driver to manage ns announcement

On Wed, Aug 26, 2020 at 09:42:46AM +0200, Arnaud POULIQUEN wrote:
> Hi mathieu,
>
> I Sent my V2 few seconds before receiving your comment :)
> Please find my answer below

Seconds!

>
> On 8/25/20 6:54 PM, Mathieu Poirier wrote:
> > Hi Arnaud,
> >
> > On Fri, Jul 31, 2020 at 01:47:31PM +0200, Arnaud Pouliquen wrote:
> >> Use the new rpmsg_ns API to send the name service announcements if
> >> the VIRTIO_RPMSG_F_NS is set, else just not implement the ops.
> >>
> >> Signed-off-by: Arnaud Pouliquen <[email protected]>
> >> ---
> >> drivers/rpmsg/virtio_rpmsg_bus.c | 94 +++++---------------------------
> >> 1 file changed, 13 insertions(+), 81 deletions(-)
> >>
> >> diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
> >> index f771fdae150e..3c771a6392be 100644
> >> --- a/drivers/rpmsg/virtio_rpmsg_bus.c
> >> +++ b/drivers/rpmsg/virtio_rpmsg_bus.c
> >> @@ -91,35 +91,6 @@ struct rpmsg_hdr {
> >> u8 data[];
> >> } __packed;
> >>
> >> -/**
> >> - * struct rpmsg_ns_msg - dynamic name service announcement message
> >> - * @name: name of remote service that is published
> >> - * @addr: address of remote service that is published
> >> - * @flags: indicates whether service is created or destroyed
> >> - *
> >> - * This message is sent across to publish a new service, or announce
> >> - * about its removal. When we receive these messages, an appropriate
> >> - * rpmsg channel (i.e device) is created/destroyed. In turn, the ->probe()
> >> - * or ->remove() handler of the appropriate rpmsg driver will be invoked
> >> - * (if/as-soon-as one is registered).
> >> - */
> >> -struct rpmsg_ns_msg {
> >> - char name[RPMSG_NAME_SIZE];
> >> - __virtio32 addr;
> >> - __virtio32 flags;
> >> -} __packed;
> >> -
> >> -/**
> >> - * enum rpmsg_ns_flags - dynamic name service announcement flags
> >> - *
> >> - * @RPMSG_NS_CREATE: a new remote service was just created
> >> - * @RPMSG_NS_DESTROY: a known remote service was just destroyed
> >> - */
> >> -enum rpmsg_ns_flags {
> >> - RPMSG_NS_CREATE = 0,
> >> - RPMSG_NS_DESTROY = 1,
> >> -};
> >> -
> >> /**
> >> * @vrp: the remote processor this channel belongs to
> >> */
> >> @@ -324,60 +295,18 @@ static void virtio_rpmsg_destroy_ept(struct rpmsg_endpoint *ept)
> >> __rpmsg_destroy_ept(vch->vrp, ept);
> >> }
> >>
> >> -static int virtio_rpmsg_announce_create(struct rpmsg_device *rpdev)
> >> -{
> >> - struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
> >> - struct virtproc_info *vrp = vch->vrp;
> >> - struct device *dev = &rpdev->dev;
> >> - int err = 0;
> >> -
> >> - /* need to tell remote processor's name service about this channel ? */
> >> - if (rpdev->announce && rpdev->ept &&
> >> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> >> - struct rpmsg_ns_msg nsm;
> >> -
> >> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
> >> - nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
> >> - nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_CREATE);
> >> -
> >> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
> >> - if (err)
> >> - dev_err(dev, "failed to announce service %d\n", err);
> >> - }
> >> -
> >> - return err;
> >> -}
> >> -
> >> -static int virtio_rpmsg_announce_destroy(struct rpmsg_device *rpdev)
> >> -{
> >> - struct virtio_rpmsg_channel *vch = to_virtio_rpmsg_channel(rpdev);
> >> - struct virtproc_info *vrp = vch->vrp;
> >> - struct device *dev = &rpdev->dev;
> >> - int err = 0;
> >> -
> >> - /* tell remote processor's name service we're removing this channel */
> >> - if (rpdev->announce && rpdev->ept &&
> >> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> >> - struct rpmsg_ns_msg nsm;
> >> -
> >> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
> >> - nsm.addr = cpu_to_virtio32(vrp->vdev, rpdev->ept->addr);
> >> - nsm.flags = cpu_to_virtio32(vrp->vdev, RPMSG_NS_DESTROY);
> >> -
> >> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
> >> - if (err)
> >> - dev_err(dev, "failed to announce service %d\n", err);
> >> - }
> >> -
> >> - return err;
> >> -}
> >> -
> >> static const struct rpmsg_device_ops virtio_rpmsg_ops = {
> >> .create_channel = virtio_rpmsg_create_channel,
> >> .release_channel = virtio_rpmsg_release_channel,
> >> .create_ept = virtio_rpmsg_create_ept,
> >> - .announce_create = virtio_rpmsg_announce_create,
> >> - .announce_destroy = virtio_rpmsg_announce_destroy,
> >> +};
> >> +
> >> +static const struct rpmsg_device_ops virtio_rpmsg_w_nsa_ops = {
> >> + .create_channel = virtio_rpmsg_create_channel,
> >> + .release_channel = virtio_rpmsg_release_channel,
> >> + .create_ept = virtio_rpmsg_create_ept,
> >> + .announce_create = rpmsg_ns_announce_create,
> >> + .announce_destroy = rpmsg_ns_announce_destroy,
> >> };
> >>
> >> static void virtio_rpmsg_release_device(struct device *dev)
> >> @@ -423,7 +352,10 @@ __rpmsg_create_channel(struct virtproc_info *vrp,
> >> rpdev = &vch->rpdev;
> >> rpdev->src = chinfo->src;
> >> rpdev->dst = chinfo->dst;
> >> - rpdev->ops = &virtio_rpmsg_ops;
> >> + if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS))
> >> + rpdev->ops = &virtio_rpmsg_w_nsa_ops;
> >> + else
> >> + rpdev->ops = &virtio_rpmsg_ops;
> >
> > Yesterday I struggled with this part and I still do this morning. Function
> > __rpmsg_create_channel() can only be called if VIRTIO_RPMSG_F_NS is set so there
> > is no need to check it again.
>
> That's right if rpmsg_create_channel is called by the rpmsg_ns only. But it
> could not be the case in future, for instance with the rpmsg_ctrl series [1]
> As the channel creation is decorrelate from the ns annoucement we need to check it.
>
> [1]https://patchwork.kernel.org/patch/11694787/
>
> I would also have expected this patch to be a
> > simple replace of the .announce_create/destroy functions. Adding an ops that
> > doesn't have the .announce_create/destroy functions looks like a feature to me,
> > and one that I don't quite get.
> >
> > Do you think you could expand on the motivation behind this patch?
>
> It was my first implementation. It is more of a phylosophical point than anything else.
> With this path i tried to implement the following rule:
> if VIRTIO_RPMSG_F_NS is not set
> no ns announcement support
> else
> delegate to the ns announcement RPMsg service

I had another very long look at this... I haven't had the time to look in your
next serie so the end result is not yet clear in my head. But...

In __rpmsg_create_channel() the new code is testing for VIRTIO_RPMSG_F_NS in
order to allocate the ops, which means that the rpdev that is fed to
virtio_rpmsg_announce_create/destroy() are associated with a virtproc_info.
Moreover there is a test in virtio_rpmsg_announce_create/destroy() that checks
for VIRTIO_RPMSG_F_NS already. So if the rpdev was created from another
interface than the name announcement then no message would be sent.

So it seems to me that the last two patches could be dropped and things would
still work properly.

The good news is that I reviewed V2 today and things look good. I will wait to
review your next set before adding RB tags. The down side is that having spent
a fair amount of time on your set, I need to look at other people's work if I
want to be fair to everyone. As such I have to push back the review of your
other set to next week.

Thanks,
Mathieu

>
> Rather, legacy implementation is about to implement the announce ops even if not supported.
>
> If this implementation is confusing i can go back to my first implementation which was
> only an update the virtio_rpmsg_announce_xx functions:
>
> @@ -322,15 +304,8 @@ static int virtio_rpmsg_announce_create(struct rpmsg_device *rpdev)
> int err = 0;
>
> /* need to tell remote processor's name service about this channel ? */
> - if (rpdev->announce && rpdev->ept &&
> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> - struct rpmsg_ns_msg nsm;
> -
> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
> - nsm.addr = rpdev->ept->addr;
> - nsm.flags = RPMSG_NS_CREATE;
> -
> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
> + if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> + err = rpmsg_ctrl_channel_announce(rpdev, RPMSG_NS_CREATE);
> if (err)
> dev_err(dev, "failed to announce service %d\n", err);
> }
> @@ -346,15 +321,8 @@ static int virtio_rpmsg_announce_destroy(struct rpmsg_device *rpdev)
> int err = 0;
>
> /* tell remote processor's name service we're removing this channel */
> - if (rpdev->announce && rpdev->ept &&
> - virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> - struct rpmsg_ns_msg nsm;
> -
> - strncpy(nsm.name, rpdev->id.name, RPMSG_NAME_SIZE);
> - nsm.addr = rpdev->ept->addr;
> - nsm.flags = RPMSG_NS_DESTROY;
> -
> - err = rpmsg_sendto(rpdev->ept, &nsm, sizeof(nsm), RPMSG_NS_ADDR);
> + if (virtio_has_feature(vrp->vdev, VIRTIO_RPMSG_F_NS)) {
> + err = rpmsg_ctrl_channel_announce(rpdev, RPMSG_NS_DESTROY);
> if (err)
> dev_err(dev, "failed to announce service %d\n", err);
> }
>
> Regards,
> Arnaud
>
> >
> > Thanks,
> > Mathieu
> >
> >>
> >> /*
> >> * rpmsg server channels has predefined local address (for now),
> >> @@ -933,7 +865,7 @@ static int rpmsg_probe(struct virtio_device *vdev)
> >>
> >> /* Assign public information to the rpmsg_device */
> >> rpdev_ns = &vch->rpdev;
> >> - rpdev_ns->ops = &virtio_rpmsg_ops;
> >> + rpdev_ns->ops = &virtio_rpmsg_w_nsa_ops;
> >>
> >> rpdev_ns->dev.parent = &vrp->vdev->dev;
> >> rpdev_ns->dev.release = virtio_rpmsg_release_device;
> >> --
> >> 2.17.1
> >>

2020-08-27 07:06:17

by Arnaud POULIQUEN

[permalink] [raw]
Subject: Re: [PATCH 8/9] rpmsg: virtio: use rpmsg_ns driver to manage ns announcement



On 8/27/20 12:10 AM, Mathieu Poirier wrote:
> I had another very long look at this... I haven't had the time to look in your
> next serie so the end result is not yet clear in my head. But...
>
> In __rpmsg_create_channel() the new code is testing for VIRTIO_RPMSG_F_NS in
> order to allocate the ops, which means that the rpdev that is fed to
> virtio_rpmsg_announce_create/destroy() are associated with a virtproc_info.
> Moreover there is a test in virtio_rpmsg_announce_create/destroy() that checks
> for VIRTIO_RPMSG_F_NS already. So if the rpdev was created from another
> interface than the name announcement then no message would be sent.
>
> So it seems to me that the last two patches could be dropped and things would
> still work properly.

Yes the last 2 patches concerns the TX part. Drop them could also be an option.

>
> The good news is that I reviewed V2 today and things look good. I will wait to
> review your next set before adding RB tags. The down side is that having spent
> a fair amount of time on your set, I need to look at other people's work if I
> want to be fair to everyone. As such I have to push back the review of your
> other set to next week.

That's fair! And this will give me time to have a look to the detach series.

Thanks,
Arnaud

>
> Thanks,
> Mathieu
>