igt-dev.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
To: Ville Syrjala <ville.syrjala@linux.intel.com>,
	igt-dev@lists.freedesktop.org
Subject: Re: [igt-dev] [PATCH i-g-t 13/25] lib/igt_fb: Extract calc_plane_stride()
Date: Fri, 21 Sep 2018 15:33:52 -0700	[thread overview]
Message-ID: <1537569232.3945.41.camel@intel.com> (raw)
In-Reply-To: <20180719150415.935-13-ville.syrjala@linux.intel.com>

Em Qui, 2018-07-19 às 18:04 +0300, Ville Syrjala escreveu:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> 
> Extract the stride calculation from calc_fb_size_packed() to its own
> thing so that we can use it to calculate just the stride.

I think it makes more sense to wait for the rebased versions before
giving r-b tags to patches 13 and 14. Patch 13 looks good, but 14 is
starting to get hard to review without applying it locally.


> 
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> ---
>  lib/igt_fb.c | 53 ++++++++++++++++++++++++++++++++++++++++--------
> -----
>  1 file changed, 40 insertions(+), 13 deletions(-)
> 
> diff --git a/lib/igt_fb.c b/lib/igt_fb.c
> index 0319afa9749f..dd6cd1f6409c 100644
> --- a/lib/igt_fb.c
> +++ b/lib/igt_fb.c
> @@ -226,6 +226,38 @@ static int fb_num_planes(struct
> format_desc_struct *format)
>  	return format->num_planes;
>  }
>  
> +static unsigned calc_plane_stride(int fd,
> +				  struct format_desc_struct *format,
> +				  int width, uint64_t tiling, int
> plane)
> +{
> +	uint32_t min_stride = fb_plane_min_stride(format, width,
> plane);
> +
> +	if (tiling != LOCAL_DRM_FORMAT_MOD_NONE &&
> +	    intel_gen(intel_get_drm_devid(fd)) <= 3) {
> +		uint32_t stride;
> +
> +		/* Round the tiling up to the next power-of-two and
> the region
> +		 * up to the next pot fence size so that this works
> on all
> +		 * generations.
> +		 *
> +		 * This can still fail if the framebuffer is too
> large to be
> +		 * tiled. But then that failure is expected.
> +		 */
> +		for (stride = 512; stride < min_stride; stride *= 2)
> +			;
> +
> +		return stride;
> +	} else {
> +		unsigned int tile_width, tile_height;
> +
> +		igt_get_fb_tile_size(fd, tiling,
> +				     fb_plane_bpp(format, plane),
> +				     &tile_width, &tile_height);
> +
> +		return ALIGN(min_stride, tile_width);
> +	}
> +}
> +
>  static void calc_fb_size_planar(int fd, int width, int height,
>  				struct format_desc_struct *format,
>  				uint64_t tiling, unsigned stride,
> @@ -272,11 +304,10 @@ static void calc_fb_size_packed(int fd, int
> width, int height,
>  				struct format_desc_struct *format,
> uint64_t tiling,
>  				unsigned stride, unsigned *size_ret,
> unsigned *stride_ret)
>  {
> -	unsigned int tile_width, tile_height, size;
> -	int byte_width = fb_plane_min_stride(format, width, 0);
> +	unsigned int size;
>  
> -	igt_get_fb_tile_size(fd, tiling, fb_plane_bpp(format, 0),
> -			     &tile_width, &tile_height);
> +	if (!stride)
> +		stride = calc_plane_stride(fd, format, width,
> tiling, 0);
>  
>  	if (tiling != LOCAL_DRM_FORMAT_MOD_NONE &&
>  	    intel_gen(intel_get_drm_devid(fd)) <= 3) {
> @@ -289,19 +320,15 @@ static void calc_fb_size_packed(int fd, int
> width, int height,
>  		 * This can still fail if the framebuffer is too
> large to be
>  		 * tiled. But then that failure is expected.
>  		 */
> -
> -		if (!stride) {
> -			v = byte_width;
> -			for (stride = 512; stride < v; stride *= 2)
> -				;
> -		}
> -
>  		v = stride * height;
>  		for (size = 1024*1024; size < v; size *= 2)
>  			;
>  	} else {
> -		if (!stride)
> -			stride = ALIGN(byte_width, tile_width);
> +		unsigned int tile_width, tile_height;
> +
> +		igt_get_fb_tile_size(fd, tiling,
> fb_plane_bpp(format, 0),
> +				     &tile_width, &tile_height);
> +
>  		size = stride * ALIGN(height, tile_height);
>  	}
>  
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

  reply	other threads:[~2018-09-21 22:35 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-19 15:03 [igt-dev] [PATCH i-g-t 01/25] lib/igt_kms: Fill the plane format/mod information for pre-blobifier drivers Ville Syrjala
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 02/25] tests/kms_ccs: Use igt_plane_has_format_mod() Ville Syrjala
2018-09-20 20:56   ` Paulo Zanoni
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 03/25] tests/kms_plane: Add validate-in-formats subtest Ville Syrjala
2018-09-20 21:10   ` Paulo Zanoni
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 04/25] tests/kms_addfb_basic: Check that addfb2 accepts/rejects the expected formats Ville Syrjala
2018-09-20 23:36   ` Paulo Zanoni
2018-09-21 13:37     ` Ville Syrjälä
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 05/25] tests/gem_render_copy: Fix clipped height Ville Syrjala
2018-08-24  3:17   ` Dhinakaran Pandiyan
2018-08-28 14:21     ` Ville Syrjälä
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 06/25] lib/igt_fb: Respect the users choice of stride Ville Syrjala
2018-09-21  0:04   ` Paulo Zanoni
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 07/25] lib: Add DIV_ROUND_UP() Ville Syrjala
2018-09-18 21:17   ` Paulo Zanoni
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 08/25] lib/igt_fb: Use fb_blit_upload as the base class for fb_convert_blit_upload Ville Syrjala
2018-09-21  0:15   ` Paulo Zanoni
2018-07-19 15:03 ` [igt-dev] [PATCH i-g-t 09/25] lib/igt_fb: Pass fb_blit_upload to free_linear_mapping() Ville Syrjala
2018-09-21  0:20   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 10/25] lib/igt_fb: s/planar_foo/fb_plane_foo/ Ville Syrjala
2018-09-21 21:58   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 11/25] lib/igt_fb: Add fb_plane_bpp() Ville Syrjala
2018-09-21 22:02   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 12/25] lib/igt_fb: Add fb_num_planes() Ville Syrjala
2018-09-21 22:05   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 13/25] lib/igt_fb: Extract calc_plane_stride() Ville Syrjala
2018-09-21 22:33   ` Paulo Zanoni [this message]
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 14/25] lib/igt_fb: Consolidate fb size calculation to one function Ville Syrjala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 15/25] lib/kms: Pass strides[] to __kms_addfb Ville Syrjala
2018-09-21 22:45   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 16/25] lib/kms: Pass the number of planes explicitly to __kms_addfb() Ville Syrjala
2018-09-21 22:54   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 17/25] lib/igt_fb: Remove the hand rolled addfb2 Ville Syrjala
2018-09-21 23:15   ` Paulo Zanoni
2018-10-03 10:56     ` Petri Latvala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 18/25] lib/igt_fb: Constify format_desc_struct Ville Syrjala
2018-09-21 23:20   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 19/25] lib/igt_fb: Pass around igt_fb internally Ville Syrjala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 20/25] lib/igt_fb: Refactor blitter usage Ville Syrjala
2018-09-21 23:48   ` Paulo Zanoni
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 21/25] lib/igt_fb: Don't use blitter for large buffers Ville Syrjala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 22/25] lib/rendercopy: Add support for Yf/Ys tiling to gen9 rendercopy Ville Syrjala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 23/25] tests/gem_render_copy: Test Yf tiling Ville Syrjala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 24/25] lib/igt_fb: Use rendercopy for rendering into compressed buffers Ville Syrjala
2018-07-19 15:04 ` [igt-dev] [PATCH i-g-t 25/25] tests/kms_plane: Test all modifiers as well Ville Syrjala
2018-07-19 15:26 ` [igt-dev] ✓ Fi.CI.BAT: success for series starting with [i-g-t,01/25] lib/igt_kms: Fill the plane format/mod information for pre-blobifier drivers Patchwork
2018-07-19 17:40 ` [igt-dev] ✓ Fi.CI.IGT: " Patchwork
2018-09-20 20:26 ` [igt-dev] [PATCH i-g-t 01/25] " Paulo Zanoni
2018-09-21 13:26   ` Ville Syrjälä
2018-09-21 21:39     ` Paulo Zanoni
2018-11-27 20:30       ` Kazlauskas, Nicholas
2018-11-27 21:20         ` Ville Syrjälä

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=1537569232.3945.41.camel@intel.com \
    --to=paulo.r.zanoni@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=ville.syrjala@linux.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;
as well as URLs for NNTP newsgroup(s).