linux-kernel - Re: [PATCH] media: venus: add support for key frame

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <f1bb2ead-fe8e-af6a-1b96-9460a7b01f29@linaro.org>
Date:   Fri, 12 Oct 2018 10:37:31 +0300
From:   Stanimir Varbanov <stanimir.varbanov@...aro.org>
To:     Alexandre Courbot <acourbot@...omium.org>, mgottam@...eaurora.org
Cc:     Hans Verkuil <hverkuil@...all.nl>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Linux Media Mailing List <linux-media@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        linux-arm-msm@...r.kernel.org, vgarodia@...eaurora.org
Subject: Re: [PATCH] media: venus: add support for key frame

Hi Alex,

On 10/12/2018 08:26 AM, Alexandre Courbot wrote:
> On Tue, Oct 9, 2018 at 4:54 PM Malathi Gottam <mgottam@...eaurora.org> wrote:
>>
>> When client requests for a keyframe, set the property
>> to hardware to generate the sync frame.
>>
>> Signed-off-by: Malathi Gottam <mgottam@...eaurora.org>
>> ---
>>  drivers/media/platform/qcom/venus/venc_ctrls.c | 13 +++++++++++++
>>  1 file changed, 13 insertions(+)
>>
>> diff --git a/drivers/media/platform/qcom/venus/venc_ctrls.c b/drivers/media/platform/qcom/venus/venc_ctrls.c
>> index 45910172..f332c8e 100644
>> --- a/drivers/media/platform/qcom/venus/venc_ctrls.c
>> +++ b/drivers/media/platform/qcom/venus/venc_ctrls.c
>> @@ -81,6 +81,8 @@ static int venc_op_s_ctrl(struct v4l2_ctrl *ctrl)
>>         struct venc_controls *ctr = &inst->controls.enc;
>>         u32 bframes;
>>         int ret;
>> +       void *ptr;
>> +       u32 ptype;
>>
>>         switch (ctrl->id) {
>>         case V4L2_CID_MPEG_VIDEO_BITRATE_MODE:
>> @@ -173,6 +175,14 @@ static int venc_op_s_ctrl(struct v4l2_ctrl *ctrl)
>>
>>                 ctr->num_b_frames = bframes;
>>                 break;
>> +       case V4L2_CID_MPEG_VIDEO_FORCE_KEY_FRAME:
>> +               ptype = HFI_PROPERTY_CONFIG_VENC_REQUEST_SYNC_FRAME;
>> +               ret = hfi_session_set_property(inst, ptype, ptr);
> 
> The test bot already said it, but ptr is passed to
> hfi_session_set_property() uninitialized. And as can be expected the
> call returns -EINVAL on my board.
> 
> Looking at other uses of HFI_PROPERTY_CONFIG_VENC_REQUEST_SYNC_FRAME I
> see that the packet sent to the firmware does not have room for an
> argument, so I tried to pass NULL but got the same result.

yes, because pdata cannot be NULL. I'd suggest to make a pointer to
struct hfi_enable and pass it to the set_property function.

-- 
regards,
Stan