drm/i915: implement drmmode overlay support v4

This implements intel overlay support for kms via a device-specific
ioctl. Thomas Hellstrom brought up the idea of a general ioctl (on
dri-devel). We've reached the conclusion that such an infrastructure
only makes sense when multiple kms overlay implementations exists,
which atm don't (and it doesn't look like this is gonna change).

Open issues:
- Runs in sync with the gpu, i.e. unnecessary waiting. I've decided
  to wait on this because the hw tends to hang when changing something
  in this area. I left some dummy functions as infrastructure.
- polyphase filtering uses a static table.
- uses uninterruptible sleeps. Unfortunately the alternatives may
  unnecessarily wedged the hw if/when we timeout too early (and
  userspace only overloaded the batch buffers with stuff worth a few
  secs of gpu time).

Changes since v1:
- fix off-by-one misconception on my side. This fixes fullscreen
  playback.
Changes since v2:
- add underrun detection as spec'ed for i965.
- flush caches properly, fixing visual corruptions.
Changes since v4:
- fix up cache flushing of overlay memory regs.
- killed require_pipe_a logic - it hangs the chip.

Tested-By: diego.abelenda@gmail.com (on a 865G)
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
[anholt: Resolved against the MADVISE ioctl going in before this one]
Signed-off-by: NEric Anholt <eric@anholt.net>

drivers/gpu/drm/i915/Makefile

@@ -23,6 +23,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o i915_mem.o \
 	  intel_fb.o \
 	  intel_tv.o \
 	  intel_dvo.o \
+	  intel_overlay.o \
 	  dvo_ch7xxx.o \
 	  dvo_ch7017.o \
 	  dvo_ivch.o \

drivers/gpu/drm/i915/i915_dma.c

@@ -807,6 +807,9 @@ static int i915_getparam(struct drm_device *dev, void *data,
 	case I915_PARAM_NUM_FENCES_AVAIL:
 		value = dev_priv->num_fence_regs - dev_priv->fence_reg_start;
 		break;
+	case I915_PARAM_HAS_OVERLAY:
+		value = dev_priv->overlay ? 1 : 0;
+		break;
 	default:
 		DRM_DEBUG_DRIVER("Unknown parameter %d\n",
 					param->param);
@@ -1548,6 +1551,8 @@ int i915_driver_unload(struct drm_device *dev)
 		mutex_unlock(&dev->struct_mutex);
 		drm_mm_takedown(&dev_priv->vram);
 		i915_gem_lastclose(dev);
+
+		intel_cleanup_overlay(dev);
 	}
 
 	pci_dev_put(dev_priv->bridge_dev);
@@ -1656,6 +1661,8 @@ struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_I915_GEM_GET_APERTURE, i915_gem_get_aperture_ioctl, 0),
 	DRM_IOCTL_DEF(DRM_I915_GET_PIPE_FROM_CRTC_ID, intel_get_pipe_from_crtc_id, 0),
 	DRM_IOCTL_DEF(DRM_I915_GEM_MADVISE, i915_gem_madvise_ioctl, 0),
+	DRM_IOCTL_DEF(DRM_I915_OVERLAY_PUT_IMAGE, intel_overlay_put_image, DRM_MASTER|DRM_CONTROL_ALLOW),
+	DRM_IOCTL_DEF(DRM_I915_OVERLAY_ATTRS, intel_overlay_attrs, DRM_MASTER|DRM_CONTROL_ALLOW),
 };
 
 int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls);

drivers/gpu/drm/i915/i915_drv.h

@@ -170,6 +170,8 @@ struct drm_i915_display_funcs {
 	/* clock gating init */
 };
 
+struct intel_overlay;
+
 typedef struct drm_i915_private {
 	struct drm_device *dev;
 
@@ -241,6 +243,9 @@ typedef struct drm_i915_private {
 
 	struct intel_opregion opregion;
 
+	/* overlay */
+	struct intel_overlay *overlay;
+
 	/* LVDS info */
 	int backlight_duty_cycle;  /* restore backlight to this value */
 	bool panel_wants_dither;

drivers/gpu/drm/i915/i915_reg.h

@@ -140,6 +140,7 @@
 #define MI_NOOP			MI_INSTR(0, 0)
 #define MI_USER_INTERRUPT	MI_INSTR(0x02, 0)
 #define MI_WAIT_FOR_EVENT       MI_INSTR(0x03, 0)
+#define   MI_WAIT_FOR_OVERLAY_FLIP	(1<<16)
 #define   MI_WAIT_FOR_PLANE_B_FLIP      (1<<6)
 #define   MI_WAIT_FOR_PLANE_A_FLIP      (1<<2)
 #define   MI_WAIT_FOR_PLANE_A_SCANLINES (1<<1)
@@ -151,6 +152,10 @@
 #define   MI_END_SCENE		(1 << 4) /* flush binner and incr scene count */
 #define MI_BATCH_BUFFER_END	MI_INSTR(0x0a, 0)
 #define MI_REPORT_HEAD		MI_INSTR(0x07, 0)
+#define MI_OVERLAY_FLIP		MI_INSTR(0x11,0)
+#define   MI_OVERLAY_CONTINUE	(0x0<<21)
+#define   MI_OVERLAY_ON		(0x1<<21)
+#define   MI_OVERLAY_OFF	(0x2<<21)
 #define MI_LOAD_SCAN_LINES_INCL MI_INSTR(0x12, 0)
 #define MI_STORE_DWORD_IMM	MI_INSTR(0x20, 1)
 #define   MI_MEM_VIRTUAL	(1 << 22) /* 965+ only */

drivers/gpu/drm/i915/intel_display.c

@@ -1781,6 +1781,22 @@ static void igdng_crtc_dpms(struct drm_crtc *crtc, int mode)
 	}
 }
 
+static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
+{
+	struct intel_overlay *overlay;
+
+	if (!enable && intel_crtc->overlay) {
+		overlay = intel_crtc->overlay;
+		mutex_lock(&overlay->dev->struct_mutex);
+		intel_overlay_switch_off(overlay);
+		mutex_unlock(&overlay->dev->struct_mutex);
+	}
+	/* Let userspace switch the overlay on again. In most cases userspace
+	 * has to recompute where to put it anyway. */
+
+	return;
+}
+
 static void i9xx_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct drm_device *dev = crtc->dev;
@@ -1839,12 +1855,13 @@ static void i9xx_crtc_dpms(struct drm_crtc *crtc, int mode)
 			intel_update_fbc(crtc, &crtc->mode);
 
 		/* Give the overlay scaler a chance to enable if it's on this pipe */
-		//intel_crtc_dpms_video(crtc, true); TODO
+		intel_crtc_dpms_overlay(intel_crtc, true);
 	break;
 	case DRM_MODE_DPMS_OFF:
 		intel_update_watermarks(dev);
+
 		/* Give the overlay scaler a chance to disable if it's on this pipe */
-		//intel_crtc_dpms_video(crtc, FALSE); TODO
+		intel_crtc_dpms_overlay(intel_crtc, false);
 
 		if (dev_priv->cfb_plane == plane &&
 		    dev_priv->display.disable_fbc)
@@ -2039,7 +2056,7 @@ static int i830_get_display_clock_speed(struct drm_device *dev)
  * Return the pipe currently connected to the panel fitter,
  * or -1 if the panel fitter is not present or not in use
  */
-static int intel_panel_fitter_pipe (struct drm_device *dev)
+int intel_panel_fitter_pipe (struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32  pfit_control;
@@ -4458,6 +4475,8 @@ void intel_modeset_init(struct drm_device *dev)
 	INIT_WORK(&dev_priv->idle_work, intel_idle_update);
 	setup_timer(&dev_priv->idle_timer, intel_gpu_idle_timer,
 		    (unsigned long)dev);
+
+	intel_setup_overlay(dev);
 }
 
 void intel_modeset_cleanup(struct drm_device *dev)

drivers/gpu/drm/i915/intel_drv.h

@@ -110,6 +110,25 @@ struct intel_output {
 	int clone_mask;
 };
 
+struct intel_crtc;
+struct intel_overlay {
+	struct drm_device *dev;
+	struct intel_crtc *crtc;
+	struct drm_i915_gem_object *vid_bo;
+	struct drm_i915_gem_object *old_vid_bo;
+	int active;
+	int pfit_active;
+	u32 pfit_vscale_ratio; /* shifted-point number, (1<<12) == 1.0 */
+	u32 color_key;
+	u32 brightness, contrast, saturation;
+	u32 old_xscale, old_yscale;
+	/* register access */
+	u32 flip_addr;
+	struct drm_i915_gem_object *reg_bo;
+	void *virt_addr;
+	int hw_wedged;
+};
+
 struct intel_crtc {
 	struct drm_crtc base;
 	enum pipe pipe;
@@ -121,6 +140,7 @@ struct intel_crtc {
 	bool busy; /* is scanout buffer being updated frequently? */
 	struct timer_list idle_timer;
 	bool lowfreq_avail;
+	struct intel_overlay *overlay;
 };
 
 #define to_intel_crtc(x) container_of(x, struct intel_crtc, base)
@@ -148,6 +168,7 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 extern void intel_edp_link_config (struct intel_output *, int *, int *);
 
 
+extern int intel_panel_fitter_pipe (struct drm_device *dev);
 extern void intel_crtc_load_lut(struct drm_crtc *crtc);
 extern void intel_encoder_prepare (struct drm_encoder *encoder);
 extern void intel_encoder_commit (struct drm_encoder *encoder);
@@ -183,4 +204,11 @@ extern int intel_framebuffer_create(struct drm_device *dev,
 				    struct drm_framebuffer **fb,
 				    struct drm_gem_object *obj);
 
+extern void intel_setup_overlay(struct drm_device *dev);
+extern void intel_cleanup_overlay(struct drm_device *dev);
+extern int intel_overlay_switch_off(struct intel_overlay *overlay);
+extern int intel_overlay_put_image(struct drm_device *dev, void *data,
+				   struct drm_file *file_priv);
+extern int intel_overlay_attrs(struct drm_device *dev, void *data,
+			       struct drm_file *file_priv);
 #endif /* __INTEL_DRV_H__ */

include/drm/i915_drm.h

@@ -186,6 +186,8 @@ typedef struct _drm_i915_sarea {
 #define DRM_I915_GEM_MMAP_GTT	0x24
 #define DRM_I915_GET_PIPE_FROM_CRTC_ID	0x25
 #define DRM_I915_GEM_MADVISE	0x26
+#define DRM_I915_OVERLAY_PUT_IMAGE	0x27
+#define DRM_I915_OVERLAY_ATTRS	0x28
 
 #define DRM_IOCTL_I915_INIT		DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t)
 #define DRM_IOCTL_I915_FLUSH		DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH)
@@ -223,6 +225,8 @@ typedef struct _drm_i915_sarea {
 #define DRM_IOCTL_I915_GEM_GET_APERTURE	DRM_IOR  (DRM_COMMAND_BASE + DRM_I915_GEM_GET_APERTURE, struct drm_i915_gem_get_aperture)
 #define DRM_IOCTL_I915_GET_PIPE_FROM_CRTC_ID DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GET_PIPE_FROM_CRTC_ID, struct drm_intel_get_pipe_from_crtc_id)
 #define DRM_IOCTL_I915_GEM_MADVISE	DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_MADVISE, struct drm_i915_gem_madvise)
+#define DRM_IOCTL_I915_OVERLAY_PUT_IMAGE	DRM_IOW(DRM_COMMAND_BASE + DRM_IOCTL_I915_OVERLAY_ATTRS, struct drm_intel_overlay_put_image)
+#define DRM_IOCTL_I915_OVERLAY_ATTRS	DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_OVERLAY_ATTRS, struct drm_intel_overlay_attrs)
 
 /* Allow drivers to submit batchbuffers directly to hardware, relying
  * on the security mechanisms provided by hardware.
@@ -266,6 +270,7 @@ typedef struct drm_i915_irq_wait {
 #define I915_PARAM_CHIPSET_ID            4
 #define I915_PARAM_HAS_GEM               5
 #define I915_PARAM_NUM_FENCES_AVAIL      6
+#define I915_PARAM_HAS_OVERLAY           7
 
 typedef struct drm_i915_getparam {
 	int param;
@@ -686,4 +691,70 @@ struct drm_i915_gem_madvise {
 	__u32 retained;
 };
 
+/* flags */
+#define I915_OVERLAY_TYPE_MASK 		0xff
+#define I915_OVERLAY_YUV_PLANAR 	0x01
+#define I915_OVERLAY_YUV_PACKED 	0x02
+#define I915_OVERLAY_RGB		0x03
+
+#define I915_OVERLAY_DEPTH_MASK		0xff00
+#define I915_OVERLAY_RGB24		0x1000
+#define I915_OVERLAY_RGB16		0x2000
+#define I915_OVERLAY_RGB15		0x3000
+#define I915_OVERLAY_YUV422		0x0100
+#define I915_OVERLAY_YUV411		0x0200
+#define I915_OVERLAY_YUV420		0x0300
+#define I915_OVERLAY_YUV410		0x0400
+
+#define I915_OVERLAY_SWAP_MASK		0xff0000
+#define I915_OVERLAY_NO_SWAP		0x000000
+#define I915_OVERLAY_UV_SWAP		0x010000
+#define I915_OVERLAY_Y_SWAP		0x020000
+#define I915_OVERLAY_Y_AND_UV_SWAP	0x030000
+
+#define I915_OVERLAY_FLAGS_MASK		0xff000000
+#define I915_OVERLAY_ENABLE		0x01000000
+
+struct drm_intel_overlay_put_image {
+	/* various flags and src format description */
+	__u32 flags;
+	/* source picture description */
+	__u32 bo_handle;
+	/* stride values and offsets are in bytes, buffer relative */
+	__u16 stride_Y; /* stride for packed formats */
+	__u16 stride_UV;
+	__u32 offset_Y; /* offset for packet formats */
+	__u32 offset_U;
+	__u32 offset_V;
+	/* in pixels */
+	__u16 src_width;
+	__u16 src_height;
+	/* to compensate the scaling factors for partially covered surfaces */
+	__u16 src_scan_width;
+	__u16 src_scan_height;
+	/* output crtc description */
+	__u32 crtc_id;
+	__u16 dst_x;
+	__u16 dst_y;
+	__u16 dst_width;
+	__u16 dst_height;
+};
+
+/* flags */
+#define I915_OVERLAY_UPDATE_ATTRS	(1<<0)
+#define I915_OVERLAY_UPDATE_GAMMA	(1<<1)
+struct drm_intel_overlay_attrs {
+	__u32 flags;
+	__u32 color_key;
+	__s32 brightness;
+	__u32 contrast;
+	__u32 saturation;
+	__u32 gamma0;
+	__u32 gamma1;
+	__u32 gamma2;
+	__u32 gamma3;
+	__u32 gamma4;
+	__u32 gamma5;
+};
+
 #endif				/* _I915_DRM_H_ */