05.03.2020, 19:17, "Mathieu Poirier" <mathieu.poirier@xxxxxxxxxx>:
> On Wed, 4 Mar 2020 at 07:25, Nikita Shubin <NShubin@xxxxxxxxxx> wrote:
>> add kick method that does nothing, to avoid errors in rproc_virtio_notify.
>>
>> Signed-off-by: Nikita Shubin <NShubin@xxxxxxxxxx>
>> ---
>> drivers/remoteproc/imx_rproc.c | 6 ++++++
>> 1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
>> index 3e72b6f38d4b..796b6b86550a 100644
>> --- a/drivers/remoteproc/imx_rproc.c
>> +++ b/drivers/remoteproc/imx_rproc.c
>> @@ -240,9 +240,15 @@ static void *imx_rproc_da_to_va(struct rproc *rproc, u64 da, int len)
>> return va;
>> }
>>
>> +static void imx_rproc_kick(struct rproc *rproc, int vqid)
>> +{
>> +
>> +}
>> +
>
> If rproc::kick() is empty, how does the MCU know there is packets to
> fetch in the virtio queues?
Well, of course it doesn't i understand this perfectly - just following documentation citing:
| Every remoteproc implementation should at least provide the ->start and ->stop
| handlers. If rpmsg/virtio functionality is also desired, then the ->kick handler
| should be provided as well.
But i as i mentioned in "remoteproc: Fix NULL pointer dereference in rproc_virtio_notify" kick method will be called if
"resource_table exists in firmware and has "Virtio device entry" defined" anyway, the imx_rproc is not in control of what
exactly it is booting, so such situation can occur.
>
>> static const struct rproc_ops imx_rproc_ops = {
>> .start = imx_rproc_start,
>> .stop = imx_rproc_stop,
>> + .kick = imx_rproc_kick,
>> .da_to_va = imx_rproc_da_to_va,
>> };
>>
>> --
>> 2.24.1
