Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <e1294f23-5eda-6aa1-0363-a4d503d2fa59@quicinc.com>
Date:   Fri, 21 Apr 2023 14:30:55 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.9.0
Subject: Re: [PATCH v1 4/5] drm/msm/dpu: calculate DSC encoder parameters
 dynamically
Content-Language: en-US
To:     Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
        <dri-devel@lists.freedesktop.org>, <robdclark@gmail.com>,
        <sean@poorly.run>, <swboyd@chromium.org>, <dianders@chromium.org>,
        <vkoul@kernel.org>, <daniel@ffwll.ch>, <airlied@gmail.com>,
        <agross@kernel.org>, <andersson@kernel.org>
CC:     <quic_abhinavk@quicinc.com>, <quic_sbillaka@quicinc.com>,
        <marijn.suijten@somainline.org>, <freedreno@lists.freedesktop.org>,
        <linux-arm-msm@vger.kernel.org>, <linux-kernel@vger.kernel.org>
References: <1682033114-28483-1-git-send-email-quic_khsieh@quicinc.com>
 <1682033114-28483-5-git-send-email-quic_khsieh@quicinc.com>
 <c8089d0f-c8cd-6a24-718f-682145d04f02@linaro.org>
 <95a37ba2-29b5-47b9-48df-1bd4e4c409c5@quicinc.com>
 <62df2f48-2553-72f0-43e4-9a5d4027ab81@linaro.org>
From:   Kuogee Hsieh <quic_khsieh@quicinc.com>
In-Reply-To: <62df2f48-2553-72f0-43e4-9a5d4027ab81@linaro.org>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 4/21/2023 2:13 PM, Dmitry Baryshkov wrote:
> On 22/04/2023 00:07, Kuogee Hsieh wrote:
>>
>> On 4/20/2023 5:27 PM, Dmitry Baryshkov wrote:
>>> On 21/04/2023 02:25, Kuogee Hsieh wrote:
>>>> During DSC preparation, add run time calculation to figure out what
>>>> usage modes, split mode and merge mode, is going to be setup.
>>>>
>>>> Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
>>>> ---
>>>>   drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 56 
>>>> ++++++++++++++++-------------
>>>>   1 file changed, 32 insertions(+), 24 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c 
>>>> b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
>>>> index 2fdacf1..5677728 100644
>>>> --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
>>>> +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
>>>> @@ -529,17 +529,9 @@ void dpu_encoder_helper_split_config(
>>>>   bool dpu_encoder_use_dsc_merge(struct drm_encoder *drm_enc)
>>>>   {
>>>>       struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc);
>>>> -    int i, intf_count = 0, num_dsc = 0;
>>>> +    struct msm_display_topology *topology = &dpu_enc->topology;
>>>>   -    for (i = 0; i < MAX_PHYS_ENCODERS_PER_VIRTUAL; i++)
>>>> -        if (dpu_enc->phys_encs[i])
>>>> -            intf_count++;
>>>> -
>>>> -    /* See dpu_encoder_get_topology, we only support 2:2:1 
>>>> topology */
>>>> -    if (dpu_enc->dsc)
>>>> -        num_dsc = 2;
>>>> -
>>>> -    return (num_dsc > 0) && (num_dsc > intf_count);
>>>> +    return (topology->num_dsc > topology->num_intf);
>>>>   }
>>>>     static void dpu_encoder_get_topology(
>>>> @@ -1861,41 +1853,57 @@ static void dpu_encoder_prep_dsc(struct 
>>>> dpu_encoder_virt *dpu_enc,
>>>>       struct dpu_encoder_phys *enc_master = dpu_enc->cur_master;
>>>>       struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC];
>>>>       struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC];
>>>> +    struct msm_display_topology *topology = &dpu_enc->topology;
>>>>       int this_frame_slices;
>>>>       int intf_ip_w, enc_ip_w;
>>>> -    int dsc_common_mode;
>>>> +    int dsc_common_mode = 0;
>>>>       int pic_width;
>>>>       u32 initial_lines;
>>>> +    int num_dsc, num_intf;
>>>>       int i;
>>>>         for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) {
>>>>           hw_pp[i] = dpu_enc->hw_pp[i];
>>>>           hw_dsc[i] = dpu_enc->hw_dsc[i];
>>>> -
>>>> -        if (!hw_pp[i] || !hw_dsc[i]) {
>>>> -            DPU_ERROR_ENC(dpu_enc, "invalid params for DSC\n");
>>>> -            return;
>>>> -        }
>>>
>> Why?
>>
>> MAX_CHANNELS_PER_ENC == 2
>>
>> This works for dsi since it use 2 dsc encoder.
>>
>> Since DP only use one dsc encoder, this will cause it return at loop 
>> 2 without execute dpu_encoder_dsc_pipe_cfg().
>
> Then the loop should go up to num_dsc rather than MAX_CHANNELS_PER_ENC
>
>>
>>>
>>>>       }
>>>>   -    dsc_common_mode = 0;
>>>> +    num_dsc = topology->num_dsc;
>>>> +    num_intf = topology->num_intf;
>>>> +
>>>>       pic_width = dsc->pic_width;
>>>>   -    dsc_common_mode = DSC_MODE_MULTIPLEX | DSC_MODE_SPLIT_PANEL;
>>>>       if (enc_master->intf_mode == INTF_MODE_VIDEO)
>>>>           dsc_common_mode |= DSC_MODE_VIDEO;
>>>>   +    /*
>>>> +     * If this encoder is driving more than one DSC encoder, they
>>>> +     * operate in tandem, same pic dimension needs to be used by
>>>> +     * each of them.(pp-split is assumed to be not supported)
>>>> +     *
>>>> +     */
>>>> +
>>>>       this_frame_slices = pic_width / dsc->slice_width;
>>>>       intf_ip_w = this_frame_slices * dsc->slice_width;
>>>> +    enc_ip_w = intf_ip_w;
>>>> +
>>>> +    intf_ip_w /= num_intf;
>>>> +
>>>> +    if (num_dsc > 1)
>>>> +        dsc_common_mode |= DSC_MODE_SPLIT_PANEL;
>>>> +
>>>> +    if (dpu_encoder_use_dsc_merge(&dpu_enc->base)) {
>>>> +        dsc_common_mode |= DSC_MODE_MULTIPLEX;
>>>> +        /*
>>>> +         * in dsc merge case: when using 2 encoders for the same
>>>> +         * stream, no. of slices need to be same on both the
>>>> +         * encoders.
>>>> +         */
>>>> +        enc_ip_w = intf_ip_w / 2;
>>>
>>> So do you want to get enc_ip_w / 2 or enc_ip_w / num_intf / 2 here?
>> enc_ip_w / num_intf / 2
>
> But previously we had enc_ip_w = intf_ip_w / 2. Was it because of the 
> assumption that num_intf = 1?
i think so since there is no num_intf involve at previous code.
>
>>>
>>>> +    }
>>>>   -    /*
>>>> -     * dsc merge case: when using 2 encoders for the same stream,
>>>> -     * no. of slices need to be same on both the encoders.
>>>> -     */
>>>> -    enc_ip_w = intf_ip_w / 2;
>>>>       initial_lines = dpu_encoder_dsc_initial_line_calc(dsc, 
>>>> enc_ip_w);
>>>>   -    for (i = 0; i < MAX_CHANNELS_PER_ENC; i++)
>>>> +    for (i = 0; i < num_dsc; i++)
>>>>           dpu_encoder_dsc_pipe_cfg(dpu_enc, hw_dsc[i], hw_pp[i], dsc,
>>>>                       dsc_common_mode, initial_lines);
>>>>   }
>>>
>