[Intel-gfx] [PATCH 37/37] drm/i915: Implement .get_format_info() hook for CCS

[Ville Syrjälä]

On Fri, Nov 18, 2016 at 03:31:48PM -0800, Ben Widawsky wrote:
> On 16-11-18 21:53:13, Ville Syrjälä wrote:
> >From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> >
> >By providing our own format information for the CCS formats, we should
> >be able to make framebuffer_check() do the right thing for the CCS
> >surface as well.
> >
> 
> I was hoping to see that patch as well :-). If you're adding the new fb
> modifiers, I think it'd make sense to make it part of this series.
> Alternatively, I can take 36, and 37 and make it part of my series, then
> integrate that last bit. It's up to you.
> 
> >Note that we'll return the same format info for both Y and Yf tiled
> >format as that's what happens with the non-CCS Y vs. Yf as well. If
> >desired, we could potentially return a unique pointer for each
> >pixel_format+tiling+ccs combination, in which case we immediately be
> >able to tell if any of that stuff changed by just comparing the
> >pointers. But that does sound a bit wasteful space wise.
> >
> >Cc: Ben Widawsky <ben at bwidawsk.net>
> >Cc: intel-gfx at lists.freedesktop.org
> >Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 
> I have a comment below however, you can consider it:
> Reviewed-by: Ben Widawsky <ben at bwidawsk.net>
> 
> >---
> > drivers/gpu/drm/i915/intel_display.c | 37 ++++++++++++++++++++++++++++++++++++
> > include/uapi/drm/drm_fourcc.h        |  3 +++
> > 2 files changed, 40 insertions(+)
> >
> >diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> >index 7b7135be3b9e..de6909770c68 100644
> >--- a/drivers/gpu/drm/i915/intel_display.c
> >+++ b/drivers/gpu/drm/i915/intel_display.c
> >@@ -2488,6 +2488,42 @@ static unsigned int intel_fb_modifier_to_tiling(uint64_t fb_modifier)
> > 	}
> > }
> >
> >+static const struct drm_format_info ccs_formats[] = {
> >+	{ .format = DRM_FORMAT_XRGB8888, .depth = 24, .num_planes = 2, .cpp = { 4, 1, }, .hsub = 16, .vsub = 8, },
> >+	{ .format = DRM_FORMAT_XBGR8888, .depth = 24, .num_planes = 2, .cpp = { 4, 1, }, .hsub = 16, .vsub = 8, },
> >+	{ .format = DRM_FORMAT_ARGB8888, .depth = 32, .num_planes = 2, .cpp = { 4, 1, }, .hsub = 16, .vsub = 8, },
> >+	{ .format = DRM_FORMAT_ABGR8888, .depth = 32, .num_planes = 2, .cpp = { 4, 1, }, .hsub = 16, .vsub = 8, },
> >+};
> >+
> >+static const struct drm_format_info *
> >+lookup_format_info(const struct drm_format_info formats[],
> >+		   int num_formats, u32 format)
> >+{
> >+	int i;
> >+
> >+	for (i = 0; i < num_formats; i++) {
> >+		if (formats[i].format == format)
> >+			return &formats[i];
> >+	}
> >+
> >+	return NULL;
> >+}
> >+
> >+static const struct drm_format_info *
> >+intel_get_format_info(struct drm_device *dev,
> >+		      const struct drm_mode_fb_cmd2 *cmd)
> >+{
> >+	switch (cmd->modifier[0]) {
> >+	case I915_FORMAT_MOD_Y_TILED_CCS:
> >+	case I915_FORMAT_MOD_Yf_TILED_CCS:
> >+		return lookup_format_info(ccs_formats,
> >+					  ARRAY_SIZE(ccs_formats),
> >+					  cmd->pixel_format);
> >+	default:
> >+		return NULL;
> >+	}
> >+}
> >+
> 
> It sort of seems like somewhat of a waste to provide this if implementations are
> allowed to return NULL. It's like saying, "DRM core will check stuff for you if
> you provide the info, but you don't have to do it if you don't want to."

The core will check the stuff anyway. The NULL just means "I don't have
any special requirements, so the core format info will be sufficient".

> If
> that's the case you may as well provide a driver hook to just do the check, ie.
> s/mod_funcs->get_format_info/mode_functs->check_format/

Drivers already have to do a bunch of checks in .fb_create(). In
addition the core does some basic sanity checks before the driver
even sees the mode_cmd (except for the new .get_format_info() hook
that is). I don't want every driver to have to duplicate all of these
basic sanity checks.

One alternative to this scheme would be have a helper function that
every driver would call in .fb_create() that would do these basic sanity
checks. That way we wouldn't need the extra hook, with a slight risk
that driver would forget to call the helper.

> 
> > static int
> > intel_fill_fb_info(struct drm_i915_private *dev_priv,
> > 		   struct drm_framebuffer *fb)
> >@@ -15922,6 +15958,7 @@ intel_user_framebuffer_create(struct drm_device *dev,
> >
> > static const struct drm_mode_config_funcs intel_mode_funcs = {
> > 	.fb_create = intel_user_framebuffer_create,
> >+	.get_format_info = intel_get_format_info,
> > 	.output_poll_changed = intel_fbdev_output_poll_changed,
> > 	.atomic_check = intel_atomic_check,
> > 	.atomic_commit = intel_atomic_commit,
> >diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h
> >index a5890bf44c0a..2926d916f199 100644
> >--- a/include/uapi/drm/drm_fourcc.h
> >+++ b/include/uapi/drm/drm_fourcc.h
> >@@ -218,6 +218,9 @@ extern "C" {
> >  */
> > #define I915_FORMAT_MOD_Yf_TILED fourcc_mod_code(INTEL, 3)
> >
> >+#define I915_FORMAT_MOD_Y_TILED_CCS	fourcc_mod_code(INTEL, 4)
> >+#define I915_FORMAT_MOD_Yf_TILED_CCS	fourcc_mod_code(INTEL, 5)
> >+
> > /*
> >  * Tiled, NV12MT, grouped in 64 (pixels) x 32 (lines) -sized macroblocks
> >  *
> 
> -- 
> Ben Widawsky, Intel Open Source Technology Center

-- 
Ville Syrjälä
Intel OTC