Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Nautiyal, Ankit K" <ankit.k.nautiyal@intel.com>
To: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>,
	<intel-gfx@lists.freedesktop.org>,
	<intel-xe@lists.freedesktop.org>
Cc: <jani.nikula@intel.com>, <ville.syrjala@intel.com>
Subject: Re: [PATCH v4 7/7] drm/i915/dsc: Check if vblank is sufficient for dsc prefill
Date: Mon, 13 Jan 2025 17:07:08 +0530	[thread overview]
Message-ID: <b86ae2ce-8edb-46ac-91bb-4395ffb2b265@intel.com> (raw)
In-Reply-To: <10a4a32d-1a5e-483d-9f73-3887fb5a1864@intel.com>


On 1/13/2025 4:43 PM, Nautiyal, Ankit K wrote:
>
> On 1/8/2025 6:05 PM, Mitul Golani wrote:
>> Check if vblank is sufficient for dsc prefill latency.
>>
>> --v2:
>> - Consider chroma downscaling factor in latency calculation. [Ankit]
>> - Replace with appropriate function name.
>>
>> --v3:
>> - Remove FIXME tag.[Ankit]
>> - Replace Ycbcr444 to Ycbcr420.[Anit]
>> - Correct precision. [Ankit]
>> - Use some local valiables like linetime_factor and latency to
>> adjust precision.
>> - Declare latency to 0 initially to avoid returning any garbage values.
>> - Account for second scaler downscaling factor as well. [Ankit]
>>
>> Signed-off-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
>> ---
>>   drivers/gpu/drm/i915/display/skl_watermark.c | 37 ++++++++++++++++++++
>>   1 file changed, 37 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c 
>> b/drivers/gpu/drm/i915/display/skl_watermark.c
>> index fe91854e456c..d275ec687748 100644
>> --- a/drivers/gpu/drm/i915/display/skl_watermark.c
>> +++ b/drivers/gpu/drm/i915/display/skl_watermark.c
>> @@ -2292,6 +2292,42 @@ static int icl_build_plane_wm(struct 
>> intel_crtc_state *crtc_state,
>>       return 0;
>>   }
>>   +static int
>> +dsc_prefill_latency(const struct intel_crtc_state *crtc_state)
>> +{
>> +    const struct intel_crtc_scaler_state *scaler_state =
>> +                        &crtc_state->scaler_state;
>> +    int latency = 0;
>> +    int count = hweight32(scaler_state->scaler_users);
>> +    long long hscale_k[2] = {1, 1};
>> +    long long vscale_k[2] = {1, 1};
>> +
>> +    if (!crtc_state->dsc.compression_enable)
>> +        return latency;
>> +
>> +    for (int i = 0; i < count; i++) {
>> +        hscale_k[i] = mul_u32_u32(scaler_state->scalers[i].hscale, 
>> 1000) >> 16;
>> +        vscale_k[i] = mul_u32_u32(scaler_state->scalers[i].vscale, 
>> 1000) >> 16;
>> +    }
>> +
>> +    if (count) {
>> +        int chroma_downscaling_factor =
>> +            crtc_state->output_format == 
>> INTEL_OUTPUT_FORMAT_YCBCR420 ? 2 : 1;
>> +        long long total_scaling_factor;
>> +        int linetime_factor = DIV_ROUND_UP(15 * 
>> crtc_state->linetime, 10);
>> +
>> +        total_scaling_factor  = DIV_ROUND_UP_ULL(hscale_k[0] * 
>> vscale_k[0], 1000000);
>
> I think this might end up being 0.
>
> Consider hscale and vscale to be 0.5 each. hscale_k and vscale_k will 
> be 500 each.
>
> total_scaling_factor will become 500*500/ 10,00,000 or 2,50,000 
> /10,00,000 = 0.
>
> I think you should compute latency for scaler 0 and then if there is 
> another scaler, multiply it with the total scaling factor for scaler 1.

Scratch that.

Just checked the Bspec. We need to have downscale amount <for each 
direction and scaler> = MAX(1, scaler input / scaler output)

So at min the scaling factor will be one. This check is missing and we 
need to consider this while computing the scaling factor.


Regards,

Ankit


>
> Regards,
>
> Ankit
>
>
>> +
>> +        if (count > 1)
>> +            total_scaling_factor *= DIV_ROUND_UP_ULL(hscale_k[1] * 
>> vscale_k[1],
>> +                                 1000000);
>> +
>> +        latency = total_scaling_factor * linetime_factor * 
>> chroma_downscaling_factor;
>> +    }
>> +
>> +    return latency;
>> +}
>> +
>>   static int
>>   scaler_prefill_latency(const struct intel_crtc_state *crtc_state)
>>   {
>> @@ -2333,6 +2369,7 @@ skl_is_vblank_too_short(const struct 
>> intel_crtc_state *crtc_state,
>>       return crtc_state->framestart_delay +
>>           intel_usecs_to_scanlines(adjusted_mode, latency) +
>>           scaler_prefill_latency(crtc_state) +
>> +        dsc_prefill_latency(crtc_state) +
>>           wm0_lines >
>>           adjusted_mode->crtc_vtotal - adjusted_mode->crtc_vblank_start;
>>   }

  reply	other threads:[~2025-01-13 11:38 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-08 12:35 [PATCH v4 0/7] Check Scaler and DSC Prefill Latency Against Vblank Mitul Golani
2025-01-08 12:35 ` [PATCH v4 1/7] drm/i915/scaler: Add and compute scaling factors Mitul Golani
2025-01-13  6:20   ` Nautiyal, Ankit K
2025-01-08 12:35 ` [PATCH v4 2/7] drm/i915/scaler: Use crtc_state to setup plane or pipe scaler Mitul Golani
2025-01-08 12:35 ` [PATCH v4 3/7] drm/i915/scaler: Refactor max_scale computation Mitul Golani
2025-01-13  8:29   ` Nautiyal, Ankit K
2025-01-08 12:35 ` [PATCH v4 4/7] drm/i915/scaler: Compute scaling factors for pipe scaler Mitul Golani
2025-01-08 12:35 ` [PATCH v4 5/7] drm/i915/scaler: Limit pipe scaler downscaling factors for YUV420 Mitul Golani
2025-01-08 12:35 ` [PATCH v4 6/7] drm/i915/scaler: Check if vblank is sufficient for scaler Mitul Golani
2025-01-13  9:45   ` Nautiyal, Ankit K
2025-01-08 12:35 ` [PATCH v4 7/7] drm/i915/dsc: Check if vblank is sufficient for dsc prefill Mitul Golani
2025-01-13 11:13   ` Nautiyal, Ankit K
2025-01-13 11:37     ` Nautiyal, Ankit K [this message]
2025-01-08 13:24 ` ✓ CI.Patch_applied: success for Check Scaler and DSC Prefill Latency Against Vblank (rev2) Patchwork
2025-01-08 13:24 ` ✓ CI.checkpatch: " Patchwork
2025-01-08 13:26 ` ✓ CI.KUnit: " Patchwork
2025-01-08 13:44 ` ✓ CI.Build: " Patchwork
2025-01-08 13:46 ` ✓ CI.Hooks: " Patchwork
2025-01-08 13:48 ` ✗ CI.checksparse: warning " Patchwork
2025-01-08 14:16 ` ✓ Xe.CI.BAT: success " Patchwork
2025-01-10  9:49 ` ✗ Xe.CI.Full: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b86ae2ce-8edb-46ac-91bb-4395ffb2b265@intel.com \
    --to=ankit.k.nautiyal@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=jani.nikula@intel.com \
    --cc=mitulkumar.ajitkumar.golani@intel.com \
    --cc=ville.syrjala@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox