Re: [PATCH v5] remoteproc: Fix NULL pointer dereference in rproc_virtio_notify

[Arnaud POULIQUEN]

Hi Bjorn

On 3/11/20 9:01 PM, Bjorn Andersson wrote:
> On Tue 10 Mar 06:19 PDT 2020, Arnaud POULIQUEN wrote:
> 
>>
>>
>>> -----Original Message-----
>>> From: linux-remoteproc-owner@xxxxxxxxxxxxxxx <linux-remoteproc-
>>> owner@xxxxxxxxxxxxxxx> On Behalf Of Nikita Shubin
>>> Sent: mardi 10 mars 2020 13:09
>>> To: Arnaud POULIQUEN <arnaud.pouliquen@xxxxxx>
>>> Cc: stable@xxxxxxxxxxxxxxx; Ohad Ben-Cohen <ohad@xxxxxxxxxx>; Bjorn
>>> Andersson <bjorn.andersson@xxxxxxxxxx>; linux-
>>> remoteproc@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Mathieu Poirier
>>> <mathieu.poirier@xxxxxxxxxx>
>>> Subject: Re: [PATCH v5] remoteproc: Fix NULL pointer dereference in
>>> rproc_virtio_notify
>>>
>>> On Mon, Mar 09, 2020 at 03:22:24PM +0100, Arnaud POULIQUEN wrote:
>>>> Hi,
>>>>
>>>> sorry for the late answer...
>>>>
>>>> On 3/6/20 8:24 AM, Nikita Shubin wrote:
>>>>> Undefined rproc_ops .kick method in remoteproc driver will result in
>>>>> "Unable to handle kernel NULL pointer dereference" in
>>>>> rproc_virtio_notify, after firmware loading if:
>>>>>
>>>>>  1) .kick method wasn't defined in driver
>>>>>  2) resource_table exists in firmware and has "Virtio device entry"
>>>>> defined
>>>>>
>>>>> Let's refuse to register an rproc-induced virtio device if no kick
>>>>> method was defined for rproc.
>>>>>

[...]

>>>>>
>>>>> Signed-off-by: Nikita Shubin <NShubin@xxxxxxxxxx>
>>>>> Fixes: 7a186941626d ("remoteproc: remove the single rpmsg vdev
>>>>> limitation")
>>>>> Cc: stable@xxxxxxxxxxxxxxx
>>>>> ---
>>>>>  drivers/remoteproc/remoteproc_virtio.c | 7 +++++++
>>>>>  1 file changed, 7 insertions(+)
>>>>>
>>>>> diff --git a/drivers/remoteproc/remoteproc_virtio.c
>>>>> b/drivers/remoteproc/remoteproc_virtio.c
>>>>> index 8c07cb2ca8ba..31a62a0b470e 100644
>>>>> --- a/drivers/remoteproc/remoteproc_virtio.c
>>>>> +++ b/drivers/remoteproc/remoteproc_virtio.c
>>>>> @@ -334,6 +334,13 @@ int rproc_add_virtio_dev(struct rproc_vdev
>>> *rvdev, int id)
>>>>>  	struct rproc_mem_entry *mem;
>>>>>  	int ret;
>>>>>
>>>>> +	if (rproc->ops->kick == NULL) {
>>>>> +		ret = -EINVAL;
>>>>> +		dev_err(dev, ".kick method not defined for %s",
>>>>> +				rproc->name);
>>>>> +		goto out;
>>>>> +	}
>>>>> +
>>>> Should the kick ops be mandatory for all the platforms? How about making
>>> it optional instead?
>>>
>>> Hi, Arnaud.
>>>
>>> It is not mandatory, currently it must be provided if specified vdev entry is in
>>> resourse table. Otherwise it looks like there is no point in creating vdev.
>>
>> Yes, my question was about having it optional for vdev also. A platform could implement the vdev
>> without kick mechanism but by polling depending due to hardware capability...
>> This could be an alternative avoiding to implement a dummy function in platform driver.
>>
> 
> Is this a real thing or a theoretical suggestion?
Only a theoretical suggestion, trigged by the IMX platform patchset which implement a "temporary" dummy kick.
and based on OpenAMP lib implementation which does not request a doorbell.
Anyway no issue to keep it mandatory here. 

Regards,
Arnaud

> 
> Regards,
> Bjorn
> 
>> Anyway it just a proposal that makes sense from MPOV. If Bjorn is ok with your patch, nothing blocking on my side.
>>
>> Regards
>> Arnaud
>>
>>>
>>>
>>>>
>>>> Regards,
>>>> Arnaud
>>>>
>>>>>  	/* Try to find dedicated vdev buffer carveout */
>>>>>  	mem = rproc_find_carveout_by_name(rproc, "vdev%dbuffer",
>>> rvdev->index);
>>>>>  	if (mem) {
>>>>>