Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Govindapillai, Vinod" <vinod.govindapillai@intel.com>
To: "Murthy, Arun R" <arun.r.murthy@intel.com>,
	"intel-xe@lists.freedesktop.org" <intel-xe@lists.freedesktop.org>,
	"dri-devel@lists.freedesktop.org"
	<dri-devel@lists.freedesktop.org>,
	"intel-gfx@lists.freedesktop.org"
	<intel-gfx@lists.freedesktop.org>
Cc: "Deak, Imre" <imre.deak@intel.com>
Subject: Re: [PATCH v6 1/2] drm/display/dp: Export fn to calculate link symbol cycles
Date: Thu, 24 Apr 2025 09:40:14 +0000	[thread overview]
Message-ID: <d441c161e65a4e028abce954298ad5a680c1c70e.camel@intel.com> (raw)
In-Reply-To: <20250424-hblank-v6-1-3d10442d9a31@intel.com>

Hi,


On Thu, 2025-04-24 at 13:52 +0530, Arun R Murthy wrote:
> Unify the function to calculate the link symbol cycles for both dsc and
> non-dsc case and export the function so that it can be used in the
> respective platform display drivers for other calculations.
> 
> v2: unify the fn for both dsc and non-dsc case (Imre)
> v3: rename drm_dp_link_symbol_cycles to drm_dp_link_data_symbol_cycles
>     retain slice_eoc_cycles as is (Imre)
> v4: Expose only drm_dp_link_symbol_cycles() (Imre)
> 
> Reviewed-by: Imre Deak <imre.deak@intel.com>
> Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
> ---
>  drivers/gpu/drm/display/drm_dp_helper.c | 53 +++++++++++++++++++++------------
>  include/drm/display/drm_dp_helper.h     |  2 ++
>  2 files changed, 36 insertions(+), 19 deletions(-)
> 
> diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
> index 57828f2b7b5a0582ca4a6f2a9be2d5909fe8ad24..6b451c9053a77e3e3889ae4cef64caaf942247c8 100644
> --- a/drivers/gpu/drm/display/drm_dp_helper.c
> +++ b/drivers/gpu/drm/display/drm_dp_helper.c
> @@ -4393,8 +4393,9 @@ EXPORT_SYMBOL(drm_panel_dp_aux_backlight);
>  #endif
>  
>  /* See DP Standard v2.1 2.6.4.4.1.1, 2.8.4.4, 2.8.7 */
> -static int drm_dp_link_symbol_cycles(int lane_count, int pixels, int bpp_x16,
> -				     int symbol_size, bool is_mst)
> +static int drm_dp_link_data_symbol_cycles(int lane_count, int pixels,
> +					  int bpp_x16, int symbol_size,
> +					  bool is_mst)
>  {
>  	int cycles = DIV_ROUND_UP(pixels * bpp_x16, 16 * symbol_size * lane_count);
>  	int align = is_mst ? 4 / lane_count : 1;
> @@ -4402,22 +4403,41 @@ static int drm_dp_link_symbol_cycles(int lane_count, int pixels, int
> bpp_x16,
>  	return ALIGN(cycles, align);
>  }
>  
> -static int drm_dp_link_dsc_symbol_cycles(int lane_count, int pixels, int slice_count,
> -					 int bpp_x16, int symbol_size, bool is_mst)
> +/**
> + * drm_dp_link_symbol_cycles - calculate the link symbol count with/without dsc
> + * @lane_count: DP link lane count
> + * @pixels: number of pixels in a scanline
> + * @dsc_slice_count: number of slices for DSC or '0' for non-DSC
> + * @bpp_x16: bits per pixel in .4 binary fixed format
> + * @symbol_size: DP symbol size
> + * @is_mst: %true for MST and %false for SST
> + *
> + * Calculate the link symbol cycles for both DSC (@dsc_slice_count !=0) and
> + * non-DSC case (@dsc_slice_count == 0) and return the count.
> + */
> +int drm_dp_link_symbol_cycles(int lane_count, int pixels, int dsc_slice_count,
> +			      int bpp_x16, int symbol_size, bool is_mst)
>  {
> -	int slice_pixels = DIV_ROUND_UP(pixels, slice_count);
> -	int slice_data_cycles = drm_dp_link_symbol_cycles(lane_count, slice_pixels,
> -							  bpp_x16, symbol_size, is_mst);
> -	int slice_eoc_cycles = is_mst ? 4 / lane_count : 1;
> +	int slice_count = dsc_slice_count ? : 1;
> +	int slice_data_cycles = drm_dp_link_data_symbol_cycles(lane_count,
> +							       pixels,
> +							       bpp_x16,
> +							       symbol_size,
> +							       is_mst);

Btw, now we pass "pixels" in drm_dp_link_data_symbol_cycles instead of "pixels / slice_count" in
case of dsc as well!

Now I see the some of the modesets for example 6k @60 are failing!

BR
Vinod

> +	int slice_eoc_cycles = 0;
> +
> +	if (dsc_slice_count)
> +		slice_eoc_cycles = is_mst ? 4 / lane_count : 1;
>  
>  	return slice_count * (slice_data_cycles + slice_eoc_cycles);
>  }
> +EXPORT_SYMBOL(drm_dp_link_symbol_cycles);
>  
>  /**
>   * drm_dp_bw_overhead - Calculate the BW overhead of a DP link stream
>   * @lane_count: DP link lane count
>   * @hactive: pixel count of the active period in one scanline of the stream
> - * @dsc_slice_count: DSC slice count if @flags/DRM_DP_LINK_BW_OVERHEAD_DSC is set
> + * @dsc_slice_count: number of slices for DSC or '0' for non-DSC
>   * @bpp_x16: bits per pixel in .4 binary fixed point
>   * @flags: DRM_DP_OVERHEAD_x flags
>   *
> @@ -4431,7 +4451,7 @@ static int drm_dp_link_dsc_symbol_cycles(int lane_count, int pixels, int
> slice_c
>   * as well as the stream's
>   * - @hactive timing
>   * - @bpp_x16 color depth
> - * - compression mode (@flags / %DRM_DP_OVERHEAD_DSC).
> + * - compression mode (@dsc_slice_count != 0)
>   * Note that this overhead doesn't account for the 8b/10b, 128b/132b
>   * channel coding efficiency, for that see
>   * @drm_dp_link_bw_channel_coding_efficiency().
> @@ -4486,15 +4506,10 @@ int drm_dp_bw_overhead(int lane_count, int hactive,
>  	WARN_ON((flags & DRM_DP_BW_OVERHEAD_UHBR) &&
>  		(flags & DRM_DP_BW_OVERHEAD_FEC));
>  
> -	if (flags & DRM_DP_BW_OVERHEAD_DSC)
> -		symbol_cycles = drm_dp_link_dsc_symbol_cycles(lane_count, hactive,
> -							      dsc_slice_count,
> -							      bpp_x16, symbol_size,
> -							      is_mst);
> -	else
> -		symbol_cycles = drm_dp_link_symbol_cycles(lane_count, hactive,
> -							  bpp_x16, symbol_size,
> -							  is_mst);
> +	symbol_cycles = drm_dp_link_symbol_cycles(lane_count, hactive,
> +						  dsc_slice_count,
> +						  bpp_x16, symbol_size,
> +						  is_mst);
>  
>  	return DIV_ROUND_UP_ULL(mul_u32_u32(symbol_cycles * symbol_size * lane_count,
>  					    overhead * 16),
> diff --git a/include/drm/display/drm_dp_helper.h b/include/drm/display/drm_dp_helper.h
> index d9614e2c89397536f44bb7258e894628ae1dccc9..7b19192c70313d66dce1b7ba40dd59c14f80a182 100644
> --- a/include/drm/display/drm_dp_helper.h
> +++ b/include/drm/display/drm_dp_helper.h
> @@ -971,5 +971,7 @@ int drm_dp_bw_channel_coding_efficiency(bool is_uhbr);
>  int drm_dp_max_dprx_data_rate(int max_link_rate, int max_lanes);
>  
>  ssize_t drm_dp_vsc_sdp_pack(const struct drm_dp_vsc_sdp *vsc, struct dp_sdp *sdp);
> +int drm_dp_link_symbol_cycles(int lane_count, int pixels, int dsc_slice_count,
> +			      int bpp_x16, int symbol_size, bool is_mst);
>  
>  #endif /* _DRM_DP_HELPER_H_ */
> 


  reply	other threads:[~2025-04-24  9:40 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-24  8:22 [PATCH v6 0/2] Rework/Correction on minimum hblank calculation Arun R Murthy
2025-04-24  8:22 ` [PATCH v6 1/2] drm/display/dp: Export fn to calculate link symbol cycles Arun R Murthy
2025-04-24  9:40   ` Govindapillai, Vinod [this message]
2025-04-24 11:15     ` Imre Deak
2025-04-24  8:22 ` [PATCH v6 2/2] drm/i915/display: move min_hblank from dp_mst.c to dp.c Arun R Murthy
2025-04-24 14:59 ` ✓ CI.Patch_applied: success for Rework/Correction on minimum hblank calculation (rev6) Patchwork
2025-04-24 14:59 ` ✓ CI.checkpatch: " Patchwork
2025-04-24 15:01 ` ✓ CI.KUnit: " Patchwork
2025-04-24 15:09 ` ✓ CI.Build: " Patchwork
2025-04-24 15:11 ` ✗ CI.Hooks: failure " Patchwork
2025-04-24 15:13 ` ✗ CI.checksparse: warning " Patchwork
2025-04-24 15:32 ` ✓ Xe.CI.BAT: success " Patchwork
2025-04-25 11:44 ` ✗ 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=d441c161e65a4e028abce954298ad5a680c1c70e.camel@intel.com \
    --to=vinod.govindapillai@intel.com \
    --cc=arun.r.murthy@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=imre.deak@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    /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