Re: [PATCH v8 16/28] media: iris: implement vb2 streaming ops

[Stefan Schmidt]

Hello Dikshita,

On Wed, 11 Dec 2024 at 10:34, Dikshita Agarwal
<quic_dikshita@xxxxxxxxxxx> wrote:
>
>
>
> On 12/10/2024 6:20 PM, Stefan Schmidt wrote:
> > Hello Dikshita,
> >
> > On Tue, 10 Dec 2024 at 12:07, Dikshita Agarwal
> > <quic_dikshita@xxxxxxxxxxx> wrote:
> >>
> >> +static int iris_hfi_gen1_session_stop(struct iris_inst *inst, u32 plane)
> >> +{
> >> +       struct hfi_session_flush_pkt flush_pkt;
> >> +       struct iris_core *core = inst->core;
> >> +       struct hfi_session_pkt pkt;
> >> +       u32 flush_type = 0;
> >> +       int ret = 0;
> >> +
> >> +       if ((V4L2_TYPE_IS_OUTPUT(plane) &&
> >> +            inst->state == IRIS_INST_INPUT_STREAMING) ||
> >> +           (V4L2_TYPE_IS_CAPTURE(plane) &&
> >> +            inst->state == IRIS_INST_OUTPUT_STREAMING) ||
> >> +           inst->state == IRIS_INST_ERROR) {
> >> +               reinit_completion(&inst->completion);
> >> +               iris_hfi_gen1_packet_session_cmd(inst, &pkt, HFI_CMD_SESSION_STOP);
> >> +               ret = iris_hfi_queue_cmd_write(core, &pkt, pkt.shdr.hdr.size);
> >> +               if (!ret)
> >> +                       ret = iris_wait_for_session_response(inst, false);
> >> +
> >> +               reinit_completion(&inst->completion);
> >> +               iris_hfi_gen1_packet_session_cmd(inst, &pkt, HFI_CMD_SESSION_RELEASE_RESOURCES);
> >> +               ret = iris_hfi_queue_cmd_write(core, &pkt, pkt.shdr.hdr.size);
> >> +               if (!ret)
> >> +                       ret = iris_wait_for_session_response(inst, false);
> >> +       } else if (inst->state == IRIS_INST_STREAMING) {
> >> +               if (V4L2_TYPE_IS_OUTPUT(plane))
> >> +                       flush_type = HFI_FLUSH_ALL;
> >> +               else if (V4L2_TYPE_IS_CAPTURE(plane))
> >> +                       flush_type = HFI_FLUSH_OUTPUT;
> >
> > Below there is also HFI_FLUSH_OUTPUT2 defined. Do we need to handle
> > this flush type here as well?
> The behavior for HFI_FLUSH_OUTPUT2 is same as HFI_FLUSH_OUTPUT so there is
> no need to add specific handling for HFI_FLUSH_OUTPUT2.

Thanks, in that case we are good.

regards
Stefan Schmidt