From: Jani Nikula <jani.nikula@linux.intel.com>
To: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org, intel-xe@lists.freedesktop.org
Subject: Re: [PATCH 2/9] drm/i915: Clean up skl+ plane stride limits
Date: Tue, 07 May 2024 12:02:49 +0300 [thread overview]
Message-ID: <87jzk6kn9i.fsf@intel.com> (raw)
In-Reply-To: <ZjkHaZOdijUrHrci@intel.com>
On Mon, 06 May 2024, Ville Syrjälä <ville.syrjala@linux.intel.com> wrote:
> On Mon, May 06, 2024 at 05:03:59PM +0300, Jani Nikula wrote:
>> On Mon, 06 May 2024, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
>> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>> >
>> > skl_plane_max_stride() is pretty messy. Streamline it and
>> > split it into clear skl+ vs. adl+ variants.
>> >
>> > TODO: Deal with icl and tgl strude limits properly
>> >
>> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
>> > ---
>> > .../drm/i915/display/skl_universal_plane.c | 65 +++++++++++--------
>> > 1 file changed, 37 insertions(+), 28 deletions(-)
>> >
>> > diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
>> > index 0a8e781a3648..b8103d6ebc1f 100644
>> > --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
>> > +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
>> > @@ -461,41 +461,46 @@ static int icl_plane_max_height(const struct drm_framebuffer *fb,
>> > }
>> >
>> > static unsigned int
>> > -skl_plane_max_stride(struct intel_plane *plane,
>> > - u32 pixel_format, u64 modifier,
>> > - unsigned int rotation)
>> > +plane_max_stride(struct intel_plane *plane,
>> > + u32 pixel_format, u64 modifier,
>> > + unsigned int rotation,
>> > + unsigned int max_pixels,
>> > + unsigned int max_bytes)
>> > {
>> > - struct drm_i915_private *i915 = to_i915(plane->base.dev);
>> > const struct drm_format_info *info = drm_format_info(pixel_format);
>> > int cpp = info->cpp[0];
>> > - int max_horizontal_pixels = 8192;
>> > - int max_stride_bytes;
>> > -
>> > - if (DISPLAY_VER(i915) >= 13) {
>> > - /*
>> > - * The stride in bytes must not exceed of the size
>> > - * of 128K bytes. For pixel formats of 64bpp will allow
>> > - * for a 16K pixel surface.
>> > - */
>> > - max_stride_bytes = 131072;
>> > - if (cpp == 8)
>> > - max_horizontal_pixels = 16384;
>>
>> The commit message doesn't mention anything about this being dropped.
>
> 16k pixels * 8 cpp == 128k bytes, so it's completely redundant.
Right. The old one was just so hard to read. :/
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
>
>>
>> BR,
>> Jani.
>>
>> > - else
>> > - max_horizontal_pixels = 65536;
>> > - } else {
>> > - /*
>> > - * "The stride in bytes must not exceed the
>> > - * of the size of 8K pixels and 32K bytes."
>> > - */
>> > - max_stride_bytes = 32768;
>> > - }
>> >
>> > if (drm_rotation_90_or_270(rotation))
>> > - return min(max_horizontal_pixels, max_stride_bytes / cpp);
>> > + return min(max_pixels, max_bytes / cpp);
>> > else
>> > - return min(max_horizontal_pixels * cpp, max_stride_bytes);
>> > + return min(max_pixels * cpp, max_bytes);
>> > }
>> >
>> > +static unsigned int
>> > +adl_plane_max_stride(struct intel_plane *plane,
>> > + u32 pixel_format, u64 modifier,
>> > + unsigned int rotation)
>> > +{
>> > + unsigned int max_pixels = 65536; /* PLANE_OFFSET limit */
>> > + unsigned int max_bytes = 128 * 1024;
>> > +
>> > + return plane_max_stride(plane, pixel_format,
>> > + modifier, rotation,
>> > + max_pixels, max_bytes);
>> > +}
>> > +
>> > +static unsigned int
>> > +skl_plane_max_stride(struct intel_plane *plane,
>> > + u32 pixel_format, u64 modifier,
>> > + unsigned int rotation)
>> > +{
>> > + unsigned int max_pixels = 8192; /* PLANE_OFFSET limit */
>> > + unsigned int max_bytes = 32 * 1024;
>> > +
>> > + return plane_max_stride(plane, pixel_format,
>> > + modifier, rotation,
>> > + max_pixels, max_bytes);
>> > +}
>> >
>> > /* Preoffset values for YUV to RGB Conversion */
>> > #define PREOFF_YUV_TO_RGB_HI 0x1800
>> > @@ -2357,7 +2362,11 @@ skl_universal_plane_create(struct drm_i915_private *dev_priv,
>> > plane->min_cdclk = skl_plane_min_cdclk;
>> > }
>> >
>> > - plane->max_stride = skl_plane_max_stride;
>> > + if (DISPLAY_VER(dev_priv) >= 13)
>> > + plane->max_stride = adl_plane_max_stride;
>> > + else
>> > + plane->max_stride = skl_plane_max_stride;
>> > +
>> > if (DISPLAY_VER(dev_priv) >= 11) {
>> > plane->update_noarm = icl_plane_update_noarm;
>> > plane->update_arm = icl_plane_update_arm;
>>
>> --
>> Jani Nikula, Intel
--
Jani Nikula, Intel
next prev parent reply other threads:[~2024-05-07 9:03 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-06 12:57 [PATCH 0/9] drm/i915: Plane fb refactoring Ville Syrjala
2024-05-06 12:57 ` [PATCH 1/9] drm/i915: Split gen2 vs. gen3 .max_stride() Ville Syrjala
2024-05-06 13:57 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 2/9] drm/i915: Clean up skl+ plane stride limits Ville Syrjala
2024-05-06 14:03 ` Jani Nikula
2024-05-06 16:38 ` Ville Syrjälä
2024-05-07 9:02 ` Jani Nikula [this message]
2024-05-06 12:57 ` [PATCH 3/9] drm/i915: Drop 'uses_fence' parameter from intel_pin_fb_obj_dpt() Ville Syrjala
2024-05-06 14:04 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 4/9] drm/i915: Extract intel_plane_needs_physical() Ville Syrjala
2024-05-06 14:05 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 5/9] drm/i915: Polish types in fb calculations Ville Syrjala
2024-05-06 14:07 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 6/9] drm/i915: Constify 'fb' in during pinning Ville Syrjala
2024-05-06 14:11 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 7/9] drm/i915: Change intel_fbdev_fb_alloc() reuturn type Ville Syrjala
2024-05-06 14:16 ` Jani Nikula
2024-05-06 16:51 ` Ville Syrjälä
2024-05-06 18:19 ` Ville Syrjälä
2024-05-10 10:22 ` [PATCH v2 7/9] drm/i915: Change intel_fbdev_fb_alloc() return type Ville Syrjala
2024-05-10 11:30 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 8/9] drm/i915: Cleanup fbdev fb setup Ville Syrjala
2024-05-10 10:22 ` [PATCH v2 " Ville Syrjala
2024-05-10 11:32 ` Jani Nikula
2024-05-06 12:57 ` [PATCH 9/9] drm/i915: Rename the fb pinning functions to indicate the address space Ville Syrjala
2024-05-10 11:35 ` Jani Nikula
2024-05-06 13:02 ` ✓ CI.Patch_applied: success for drm/i915: Plane fb refactoring Patchwork
2024-05-06 13:03 ` ✓ CI.checkpatch: " Patchwork
2024-05-06 13:03 ` ✓ CI.KUnit: " Patchwork
2024-05-06 13:15 ` ✓ CI.Build: " Patchwork
2024-05-06 13:18 ` ✓ CI.Hooks: " Patchwork
2024-05-06 13:19 ` ✗ CI.checksparse: warning " Patchwork
2024-05-06 13:34 ` ✗ Fi.CI.SPARSE: " Patchwork
2024-05-06 13:42 ` ✓ Fi.CI.BAT: success " Patchwork
2024-05-06 13:53 ` ✓ CI.BAT: " Patchwork
2024-05-06 15:07 ` ✓ CI.FULL: " Patchwork
2024-05-06 18:13 ` ✗ Fi.CI.IGT: failure " Patchwork
2024-05-10 10:52 ` ✓ CI.Patch_applied: success for drm/i915: Plane fb refactoring (rev3) Patchwork
2024-05-10 10:53 ` ✓ CI.checkpatch: " Patchwork
2024-05-10 10:55 ` ✓ CI.KUnit: " Patchwork
2024-05-10 11:07 ` ✓ CI.Build: " Patchwork
2024-05-10 11:10 ` ✓ CI.Hooks: " Patchwork
2024-05-10 11:11 ` ✗ CI.checksparse: warning " Patchwork
2024-05-10 11:43 ` ✗ CI.BAT: failure " Patchwork
2024-05-10 12:37 ` ✓ Fi.CI.BAT: success " Patchwork
2024-05-10 13:54 ` ✓ CI.FULL: " Patchwork
2024-05-10 16:55 ` [PATCH 0/9] drm/i915: Plane fb refactoring Ville Syrjälä
2024-05-11 19:00 ` Lucas De Marchi
2024-05-11 4:12 ` ✗ Fi.CI.IGT: failure for drm/i915: Plane fb refactoring (rev3) 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=87jzk6kn9i.fsf@intel.com \
--to=jani.nikula@linux.intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.