Merge remote branch 'drm/drm-next'

40 files changed, 3996 insertions, 1072 deletions

diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index bede10a03407..c6b2e27a446a 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -336,7 +336,7 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 			      struct drm_framebuffer *old_fb)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_display_mode *adjusted_mode, saved_mode;
+	struct drm_display_mode *adjusted_mode, saved_mode, saved_hwmode;
 	struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private;
 	struct drm_encoder_helper_funcs *encoder_funcs;
 	int saved_x, saved_y;
@@ -350,6 +350,7 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 	if (!crtc->enabled)
 		return true;
 
+	saved_hwmode = crtc->hwmode;
 	saved_mode = crtc->mode;
 	saved_x = crtc->x;
 	saved_y = crtc->y;
@@ -427,11 +428,21 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 
 	}
 
+	/* Store real post-adjustment hardware mode. */
+	crtc->hwmode = *adjusted_mode;
+
+	/* Calculate and store various constants which
+	 * are later needed by vblank and swap-completion
+	 * timestamping. They are derived from true hwmode.
+	 */
+	drm_calc_timestamping_constants(crtc);
+
 	/* XXX free adjustedmode */
 	drm_mode_destroy(dev, adjusted_mode);
 	/* FIXME: add subpixel order */
 done:
 	if (!ret) {
+		crtc->hwmode = saved_hwmode;
 		crtc->mode = saved_mode;
 		crtc->x = saved_x;
 		crtc->y = saved_y;
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 722700d5d73e..77c875db5aae 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -40,6 +40,22 @@
 #include <linux/slab.h>
 
 #include <linux/vgaarb.h>
+
+/* Access macro for slots in vblank timestamp ringbuffer. */
+#define vblanktimestamp(dev, crtc, count) ( \
+	(dev)->_vblank_time[(crtc) * DRM_VBLANKTIME_RBSIZE + \
+	((count) % DRM_VBLANKTIME_RBSIZE)])
+
+/* Retry timestamp calculation up to 3 times to satisfy
+ * drm_timestamp_precision before giving up.
+ */
+#define DRM_TIMESTAMP_MAXRETRIES 3
+
+/* Threshold in nanoseconds for detection of redundant
+ * vblank irq in drm_handle_vblank(). 1 msec should be ok.
+ */
+#define DRM_REDUNDANT_VBLIRQ_THRESH_NS 1000000
+
 /**
  * Get interrupt from bus id.
  *
@@ -77,6 +93,87 @@ int drm_irq_by_busid(struct drm_device *dev, void *data,
 	return 0;
 }
 
+/*
+ * Clear vblank timestamp buffer for a crtc.
+ */
+static void clear_vblank_timestamps(struct drm_device *dev, int crtc)
+{
+	memset(&dev->_vblank_time[crtc * DRM_VBLANKTIME_RBSIZE], 0,
+		DRM_VBLANKTIME_RBSIZE * sizeof(struct timeval));
+}
+
+/*
+ * Disable vblank irq's on crtc, make sure that last vblank count
+ * of hardware and corresponding consistent software vblank counter
+ * are preserved, even if there are any spurious vblank irq's after
+ * disable.
+ */
+static void vblank_disable_and_save(struct drm_device *dev, int crtc)
+{
+	unsigned long irqflags;
+	u32 vblcount;
+	s64 diff_ns;
+	int vblrc;
+	struct timeval tvblank;
+
+	/* Prevent vblank irq processing while disabling vblank irqs,
+	 * so no updates of timestamps or count can happen after we've
+	 * disabled. Needed to prevent races in case of delayed irq's.
+	 * Disable preemption, so vblank_time_lock is held as short as
+	 * possible, even under a kernel with PREEMPT_RT patches.
+	 */
+	preempt_disable();
+	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
+
+	dev->driver->disable_vblank(dev, crtc);
+	dev->vblank_enabled[crtc] = 0;
+
+	/* No further vblank irq's will be processed after
+	 * this point. Get current hardware vblank count and
+	 * vblank timestamp, repeat until they are consistent.
+	 *
+	 * FIXME: There is still a race condition here and in
+	 * drm_update_vblank_count() which can cause off-by-one
+	 * reinitialization of software vblank counter. If gpu
+	 * vblank counter doesn't increment exactly at the leading
+	 * edge of a vblank interval, then we can lose 1 count if
+	 * we happen to execute between start of vblank and the
+	 * delayed gpu counter increment.
+	 */
+	do {
+		dev->last_vblank[crtc] = dev->driver->get_vblank_counter(dev, crtc);
+		vblrc = drm_get_last_vbltimestamp(dev, crtc, &tvblank, 0);
+	} while (dev->last_vblank[crtc] != dev->driver->get_vblank_counter(dev, crtc));
+
+	/* Compute time difference to stored timestamp of last vblank
+	 * as updated by last invocation of drm_handle_vblank() in vblank irq.
+	 */
+	vblcount = atomic_read(&dev->_vblank_count[crtc]);
+	diff_ns = timeval_to_ns(&tvblank) -
+		  timeval_to_ns(&vblanktimestamp(dev, crtc, vblcount));
+
+	/* If there is at least 1 msec difference between the last stored
+	 * timestamp and tvblank, then we are currently executing our
+	 * disable inside a new vblank interval, the tvblank timestamp
+	 * corresponds to this new vblank interval and the irq handler
+	 * for this vblank didn't run yet and won't run due to our disable.
+	 * Therefore we need to do the job of drm_handle_vblank() and
+	 * increment the vblank counter by one to account for this vblank.
+	 *
+	 * Skip this step if there isn't any high precision timestamp
+	 * available. In that case we can't account for this and just
+	 * hope for the best.
+	 */
+	if ((vblrc > 0) && (abs(diff_ns) > 1000000))
+		atomic_inc(&dev->_vblank_count[crtc]);
+
+	/* Invalidate all timestamps while vblank irq's are off. */
+	clear_vblank_timestamps(dev, crtc);
+
+	spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
+	preempt_enable();
+}
+
 static void vblank_disable_fn(unsigned long arg)
 {
 	struct drm_device *dev = (struct drm_device *)arg;
@@ -91,10 +188,7 @@ static void vblank_disable_fn(unsigned long arg)
 		if (atomic_read(&dev->vblank_refcount[i]) == 0 &&
 		    dev->vblank_enabled[i]) {
 			DRM_DEBUG("disabling vblank on crtc %d\n", i);
-			dev->last_vblank[i] =
-				dev->driver->get_vblank_counter(dev, i);
-			dev->driver->disable_vblank(dev, i);
-			dev->vblank_enabled[i] = 0;
+			vblank_disable_and_save(dev, i);
 		}
 		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 	}
@@ -117,6 +211,7 @@ void drm_vblank_cleanup(struct drm_device *dev)
 	kfree(dev->last_vblank);
 	kfree(dev->last_vblank_wait);
 	kfree(dev->vblank_inmodeset);
+	kfree(dev->_vblank_time);
 
 	dev->num_crtcs = 0;
 }
@@ -129,6 +224,8 @@ int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 	setup_timer(&dev->vblank_disable_timer, vblank_disable_fn,
 		    (unsigned long)dev);
 	spin_lock_init(&dev->vbl_lock);
+	spin_lock_init(&dev->vblank_time_lock);
+
 	dev->num_crtcs = num_crtcs;
 
 	dev->vbl_queue = kmalloc(sizeof(wait_queue_head_t) * num_crtcs,
@@ -161,6 +258,19 @@ int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 	if (!dev->vblank_inmodeset)
 		goto err;
 
+	dev->_vblank_time = kcalloc(num_crtcs * DRM_VBLANKTIME_RBSIZE,
+				    sizeof(struct timeval), GFP_KERNEL);
+	if (!dev->_vblank_time)
+		goto err;
+
+	DRM_INFO("Supports vblank timestamp caching Rev 1 (10.10.2010).\n");
+
+	/* Driver specific high-precision vblank timestamping supported? */
+	if (dev->driver->get_vblank_timestamp)
+		DRM_INFO("Driver supports precise vblank timestamp query.\n");
+	else
+		DRM_INFO("No driver support for vblank timestamp query.\n");
+
 	/* Zero per-crtc vblank stuff */
 	for (i = 0; i < num_crtcs; i++) {
 		init_waitqueue_head(&dev->vbl_queue[i]);
@@ -279,7 +389,7 @@ EXPORT_SYMBOL(drm_irq_install);
  *
  * Calls the driver's \c drm_driver_irq_uninstall() function, and stops the irq.
  */
-int drm_irq_uninstall(struct drm_device * dev)
+int drm_irq_uninstall(struct drm_device *dev)
 {
 	unsigned long irqflags;
 	int irq_enabled, i;
@@ -335,7 +445,9 @@ int drm_control(struct drm_device *dev, void *data,
 {
 	struct drm_control *ctl = data;
 
-	/* if we haven't irq we fallback for compatibility reasons - this used to be a separate function in drm_dma.h */
+	/* if we haven't irq we fallback for compatibility reasons -
+	 * this used to be a separate function in drm_dma.h
+	 */
 
 
 	switch (ctl->func) {
@@ -360,6 +472,287 @@ int drm_control(struct drm_device *dev, void *data,
 }
 
 /**
+ * drm_calc_timestamping_constants - Calculate and
+ * store various constants which are later needed by
+ * vblank and swap-completion timestamping, e.g, by
+ * drm_calc_vbltimestamp_from_scanoutpos().
+ * They are derived from crtc's true scanout timing,
+ * so they take things like panel scaling or other
+ * adjustments into account.
+ *
+ * @crtc drm_crtc whose timestamp constants should be updated.
+ *
+ */
+void drm_calc_timestamping_constants(struct drm_crtc *crtc)
+{
+	s64 linedur_ns = 0, pixeldur_ns = 0, framedur_ns = 0;
+	u64 dotclock;
+
+	/* Dot clock in Hz: */
+	dotclock = (u64) crtc->hwmode.clock * 1000;
+
+	/* Valid dotclock? */
+	if (dotclock > 0) {
+		/* Convert scanline length in pixels and video dot clock to
+		 * line duration, frame duration and pixel duration in
+		 * nanoseconds:
+		 */
+		pixeldur_ns = (s64) div64_u64(1000000000, dotclock);
+		linedur_ns  = (s64) div64_u64(((u64) crtc->hwmode.crtc_htotal *
+					      1000000000), dotclock);
+		framedur_ns = (s64) crtc->hwmode.crtc_vtotal * linedur_ns;
+	} else
+		DRM_ERROR("crtc %d: Can't calculate constants, dotclock = 0!\n",
+			  crtc->base.id);
+
+	crtc->pixeldur_ns = pixeldur_ns;
+	crtc->linedur_ns  = linedur_ns;
+	crtc->framedur_ns = framedur_ns;
+
+	DRM_DEBUG("crtc %d: hwmode: htotal %d, vtotal %d, vdisplay %d\n",
+		  crtc->base.id, crtc->hwmode.crtc_htotal,
+		  crtc->hwmode.crtc_vtotal, crtc->hwmode.crtc_vdisplay);
+	DRM_DEBUG("crtc %d: clock %d kHz framedur %d linedur %d, pixeldur %d\n",
+		  crtc->base.id, (int) dotclock/1000, (int) framedur_ns,
+		  (int) linedur_ns, (int) pixeldur_ns);
+}
+EXPORT_SYMBOL(drm_calc_timestamping_constants);
+
+/**
+ * drm_calc_vbltimestamp_from_scanoutpos - helper routine for kms
+ * drivers. Implements calculation of exact vblank timestamps from
+ * given drm_display_mode timings and current video scanout position
+ * of a crtc. This can be called from within get_vblank_timestamp()
+ * implementation of a kms driver to implement the actual timestamping.
+ *
+ * Should return timestamps conforming to the OML_sync_control OpenML
+ * extension specification. The timestamp corresponds to the end of
+ * the vblank interval, aka start of scanout of topmost-leftmost display
+ * pixel in the following video frame.
+ *
+ * Requires support for optional dev->driver->get_scanout_position()
+ * in kms driver, plus a bit of setup code to provide a drm_display_mode
+ * that corresponds to the true scanout timing.
+ *
+ * The current implementation only handles standard video modes. It
+ * returns as no operation if a doublescan or interlaced video mode is
+ * active. Higher level code is expected to handle this.
+ *
+ * @dev: DRM device.
+ * @crtc: Which crtc's vblank timestamp to retrieve.
+ * @max_error: Desired maximum allowable error in timestamps (nanosecs).
+ *             On return contains true maximum error of timestamp.
+ * @vblank_time: Pointer to struct timeval which should receive the timestamp.
+ * @flags: Flags to pass to driver:
+ *         0 = Default.
+ *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl irq handler.
+ * @refcrtc: drm_crtc* of crtc which defines scanout timing.
+ *
+ * Returns negative value on error, failure or if not supported in current
+ * video mode:
+ *
+ * -EINVAL   - Invalid crtc.
+ * -EAGAIN   - Temporary unavailable, e.g., called before initial modeset.
+ * -ENOTSUPP - Function not supported in current display mode.
+ * -EIO      - Failed, e.g., due to failed scanout position query.
+ *
+ * Returns or'ed positive status flags on success:
+ *
+ * DRM_VBLANKTIME_SCANOUTPOS_METHOD - Signal this method used for timestamping.
+ * DRM_VBLANKTIME_INVBL - Timestamp taken while scanout was in vblank interval.
+ *
+ */
+int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev, int crtc,
+					  int *max_error,
+					  struct timeval *vblank_time,
+					  unsigned flags,
+					  struct drm_crtc *refcrtc)
+{
+	struct timeval stime, raw_time;
+	struct drm_display_mode *mode;
+	int vbl_status, vtotal, vdisplay;
+	int vpos, hpos, i;
+	s64 framedur_ns, linedur_ns, pixeldur_ns, delta_ns, duration_ns;
+	bool invbl;
+
+	if (crtc < 0 || crtc >= dev->num_crtcs) {
+		DRM_ERROR("Invalid crtc %d\n", crtc);
+		return -EINVAL;
+	}
+
+	/* Scanout position query not supported? Should not happen. */
+	if (!dev->driver->get_scanout_position) {
+		DRM_ERROR("Called from driver w/o get_scanout_position()!?\n");
+		return -EIO;
+	}
+
+	mode = &refcrtc->hwmode;
+	vtotal = mode->crtc_vtotal;
+	vdisplay = mode->crtc_vdisplay;
+
+	/* Durations of frames, lines, pixels in nanoseconds. */
+	framedur_ns = refcrtc->framedur_ns;
+	linedur_ns  = refcrtc->linedur_ns;
+	pixeldur_ns = refcrtc->pixeldur_ns;
+
+	/* If mode timing undefined, just return as no-op:
+	 * Happens during initial modesetting of a crtc.
+	 */
+	if (vtotal <= 0 || vdisplay <= 0 || framedur_ns == 0) {
+		DRM_DEBUG("crtc %d: Noop due to uninitialized mode.\n", crtc);
+		return -EAGAIN;
+	}
+
+	/* Don't know yet how to handle interlaced or
+	 * double scan modes. Just no-op for now.
+	 */
+	if (mode->flags & (DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLSCAN)) {
+		DRM_DEBUG("crtc %d: Noop due to unsupported mode.\n", crtc);
+		return -ENOTSUPP;
+	}
+
+	/* Get current scanout position with system timestamp.
+	 * Repeat query up to DRM_TIMESTAMP_MAXRETRIES times
+	 * if single query takes longer than max_error nanoseconds.
+	 *
+	 * This guarantees a tight bound on maximum error if
+	 * code gets preempted or delayed for some reason.
+	 */
+	for (i = 0; i < DRM_TIMESTAMP_MAXRETRIES; i++) {
+		/* Disable preemption to make it very likely to
+		 * succeed in the first iteration even on PREEMPT_RT kernel.
+		 */
+		preempt_disable();
+
+		/* Get system timestamp before query. */
+		do_gettimeofday(&stime);
+
+		/* Get vertical and horizontal scanout pos. vpos, hpos. */
+		vbl_status = dev->driver->get_scanout_position(dev, crtc, &vpos, &hpos);
+
+		/* Get system timestamp after query. */
+		do_gettimeofday(&raw_time);
+
+		preempt_enable();
+
+		/* Return as no-op if scanout query unsupported or failed. */
+		if (!(vbl_status & DRM_SCANOUTPOS_VALID)) {
+			DRM_DEBUG("crtc %d : scanoutpos query failed [%d].\n",
+				  crtc, vbl_status);
+			return -EIO;
+		}
+
+		duration_ns = timeval_to_ns(&raw_time) - timeval_to_ns(&stime);
+
+		/* Accept result with <  max_error nsecs timing uncertainty. */
+		if (duration_ns <= (s64) *max_error)
+			break;
+	}
+
+	/* Noisy system timing? */
+	if (i == DRM_TIMESTAMP_MAXRETRIES) {
+		DRM_DEBUG("crtc %d: Noisy timestamp %d us > %d us [%d reps].\n",
+			  crtc, (int) duration_ns/1000, *max_error/1000, i);
+	}
+
+	/* Return upper bound of timestamp precision error. */
+	*max_error = (int) duration_ns;
+
+	/* Check if in vblank area:
+	 * vpos is >=0 in video scanout area, but negative
+	 * within vblank area, counting down the number of lines until
+	 * start of scanout.
+	 */
+	invbl = vbl_status & DRM_SCANOUTPOS_INVBL;
+
+	/* Convert scanout position into elapsed time at raw_time query
+	 * since start of scanout at first display scanline. delta_ns
+	 * can be negative if start of scanout hasn't happened yet.
+	 */
+	delta_ns = (s64) vpos * linedur_ns + (s64) hpos * pixeldur_ns;
+
+	/* Is vpos outside nominal vblank area, but less than
+	 * 1/100 of a frame height away from start of vblank?
+	 * If so, assume this isn't a massively delayed vblank
+	 * interrupt, but a vblank interrupt that fired a few
+	 * microseconds before true start of vblank. Compensate
+	 * by adding a full frame duration to the final timestamp.
+	 * Happens, e.g., on ATI R500, R600.
+	 *
+	 * We only do this if DRM_CALLED_FROM_VBLIRQ.
+	 */
+	if ((flags & DRM_CALLED_FROM_VBLIRQ) && !invbl &&
+	    ((vdisplay - vpos) < vtotal / 100)) {
+		delta_ns = delta_ns - framedur_ns;
+
+		/* Signal this correction as "applied". */
+		vbl_status |= 0x8;
+	}
+
+	/* Subtract time delta from raw timestamp to get final
+	 * vblank_time timestamp for end of vblank.
+	 */
+	*vblank_time = ns_to_timeval(timeval_to_ns(&raw_time) - delta_ns);
+
+	DRM_DEBUG("crtc %d : v %d p(%d,%d)@ %d.%d -> %d.%d [e %d us, %d rep]\n",
+		  crtc, (int) vbl_status, hpos, vpos, raw_time.tv_sec,
+		  raw_time.tv_usec, vblank_time->tv_sec, vblank_time->tv_usec,
+		  (int) duration_ns/1000, i);
+
+	vbl_status = DRM_VBLANKTIME_SCANOUTPOS_METHOD;
+	if (invbl)
+		vbl_status |= DRM_VBLANKTIME_INVBL;
+
+	return vbl_status;
+}
+EXPORT_SYMBOL(drm_calc_vbltimestamp_from_scanoutpos);
+
+/**
+ * drm_get_last_vbltimestamp - retrieve raw timestamp for the most recent
+ * vblank interval.
+ *
+ * @dev: DRM device
+ * @crtc: which crtc's vblank timestamp to retrieve
+ * @tvblank: Pointer to target struct timeval which should receive the timestamp
+ * @flags: Flags to pass to driver:
+ *         0 = Default.
+ *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl irq handler.
+ *
+ * Fetches the system timestamp corresponding to the time of the most recent
+ * vblank interval on specified crtc. May call into kms-driver to
+ * compute the timestamp with a high-precision GPU specific method.
+ *
+ * Returns zero if timestamp originates from uncorrected do_gettimeofday()
+ * call, i.e., it isn't very precisely locked to the true vblank.
+ *
+ * Returns non-zero if timestamp is considered to be very precise.
+ */
+u32 drm_get_last_vbltimestamp(struct drm_device *dev, int crtc,
+			      struct timeval *tvblank, unsigned flags)
+{
+	int ret = 0;
+
+	/* Define requested maximum error on timestamps (nanoseconds). */
+	int max_error = (int) drm_timestamp_precision * 1000;
+
+	/* Query driver if possible and precision timestamping enabled. */
+	if (dev->driver->get_vblank_timestamp && (max_error > 0)) {
+		ret = dev->driver->get_vblank_timestamp(dev, crtc, &max_error,
+							tvblank, flags);
+		if (ret > 0)
+			return (u32) ret;
+	}
+
+	/* GPU high precision timestamp query unsupported or failed.
+	 * Return gettimeofday timestamp as best estimate.
+	 */
+	do_gettimeofday(tvblank);
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_get_last_vbltimestamp);
+
+/**
  * drm_vblank_count - retrieve "cooked" vblank counter value
  * @dev: DRM device
  * @crtc: which counter to retrieve
@@ -375,6 +768,40 @@ u32 drm_vblank_count(struct drm_device *dev, int crtc)
 EXPORT_SYMBOL(drm_vblank_count);
 
 /**
+ * drm_vblank_count_and_time - retrieve "cooked" vblank counter value
+ * and the system timestamp corresponding to that vblank counter value.
+ *
+ * @dev: DRM device
+ * @crtc: which counter to retrieve
+ * @vblanktime: Pointer to struct timeval to receive the vblank timestamp.
+ *
+ * Fetches the "cooked" vblank count value that represents the number of
+ * vblank events since the system was booted, including lost events due to
+ * modesetting activity. Returns corresponding system timestamp of the time
+ * of the vblank interval that corresponds to the current value vblank counter
+ * value.
+ */
+u32 drm_vblank_count_and_time(struct drm_device *dev, int crtc,
+			      struct timeval *vblanktime)
+{
+	u32 cur_vblank;
+
+	/* Read timestamp from slot of _vblank_time ringbuffer
+	 * that corresponds to current vblank count. Retry if
+	 * count has incremented during readout. This works like
+	 * a seqlock.
+	 */
+	do {
+		cur_vblank = atomic_read(&dev->_vblank_count[crtc]);
+		*vblanktime = vblanktimestamp(dev, crtc, cur_vblank);
+		smp_rmb();
+	} while (cur_vblank != atomic_read(&dev->_vblank_count[crtc]));
+
+	return cur_vblank;
+}
+EXPORT_SYMBOL(drm_vblank_count_and_time);
+
+/**
  * drm_update_vblank_count - update the master vblank counter
  * @dev: DRM device
  * @crtc: counter to update
@@ -392,7 +819,8 @@ EXPORT_SYMBOL(drm_vblank_count);
  */
 static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 {
-	u32 cur_vblank, diff;
+	u32 cur_vblank, diff, tslot, rc;
+	struct timeval t_vblank;
 
 	/*
 	 * Interrupts were disabled prior to this call, so deal with counter
@@ -400,8 +828,18 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 	 * NOTE!  It's possible we lost a full dev->max_vblank_count events
 	 * here if the register is small or we had vblank interrupts off for
 	 * a long time.
+	 *
+	 * We repeat the hardware vblank counter & timestamp query until
+	 * we get consistent results. This to prevent races between gpu
+	 * updating its hardware counter while we are retrieving the
+	 * corresponding vblank timestamp.
 	 */
-	cur_vblank = dev->driver->get_vblank_counter(dev, crtc);
+	do {
+		cur_vblank = dev->driver->get_vblank_counter(dev, crtc);
+		rc = drm_get_last_vbltimestamp(dev, crtc, &t_vblank, 0);
+	} while (cur_vblank != dev->driver->get_vblank_counter(dev, crtc));
+
+	/* Deal with counter wrap */
 	diff = cur_vblank - dev->last_vblank[crtc];
 	if (cur_vblank < dev->last_vblank[crtc]) {
 		diff += dev->max_vblank_count;
@@ -413,6 +851,16 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 	DRM_DEBUG("enabling vblank interrupts on crtc %d, missed %d\n",
 		  crtc, diff);
 
+	/* Reinitialize corresponding vblank timestamp if high-precision query
+	 * available. Skip this step if query unsupported or failed. Will
+	 * reinitialize delayed at next vblank interrupt in that case.
+	 */
+	if (rc) {
+		tslot = atomic_read(&dev->_vblank_count[crtc]) + diff;
+		vblanktimestamp(dev, crtc, tslot) = t_vblank;
+		smp_wmb();
+	}
+
 	atomic_add(diff, &dev->_vblank_count[crtc]);
 }
 
@@ -429,15 +877,27 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
  */
 int drm_vblank_get(struct drm_device *dev, int crtc)
 {
-	unsigned long irqflags;
+	unsigned long irqflags, irqflags2;
 	int ret = 0;
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/* Going from 0->1 means we have to enable interrupts again */
 	if (atomic_add_return(1, &dev->vblank_refcount[crtc]) == 1) {
+		/* Disable preemption while holding vblank_time_lock. Do
+		 * it explicitely to guard against PREEMPT_RT kernel.
+		 */
+		preempt_disable();
+		spin_lock_irqsave(&dev->vblank_time_lock, irqflags2);
 		if (!dev->vblank_enabled[crtc]) {
+			/* Enable vblank irqs under vblank_time_lock protection.
+			 * All vblank count & timestamp updates are held off
+			 * until we are done reinitializing master counter and
+			 * timestamps. Filtercode in drm_handle_vblank() will
+			 * prevent double-accounting of same vblank interval.
+			 */
 			ret = dev->driver->enable_vblank(dev, crtc);
-			DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n", crtc, ret);
+			DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n",
+				  crtc, ret);
 			if (ret)
 				atomic_dec(&dev->vblank_refcount[crtc]);
 			else {
@@ -445,6 +905,8 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 				drm_update_vblank_count(dev, crtc);
 			}
 		}
+		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2);
+		preempt_enable();
 	} else {
 		if (!dev->vblank_enabled[crtc]) {
 			atomic_dec(&dev->vblank_refcount[crtc]);
@@ -463,15 +925,17 @@ EXPORT_SYMBOL(drm_vblank_get);
  * @crtc: which counter to give up
  *
  * Release ownership of a given vblank counter, turning off interrupts
- * if possible.
+ * if possible. Disable interrupts after drm_vblank_offdelay milliseconds.
  */
 void drm_vblank_put(struct drm_device *dev, int crtc)
 {
-	BUG_ON (atomic_read (&dev->vblank_refcount[crtc]) == 0);
+	BUG_ON(atomic_read(&dev->vblank_refcount[crtc]) == 0);
 
 	/* Last user schedules interrupt disable */
-	if (atomic_dec_and_test(&dev->vblank_refcount[crtc]))
-		mod_timer(&dev->vblank_disable_timer, jiffies + 5*DRM_HZ);
+	if (atomic_dec_and_test(&dev->vblank_refcount[crtc]) &&
+	    (drm_vblank_offdelay > 0))
+		mod_timer(&dev->vblank_disable_timer,
+			  jiffies + ((drm_vblank_offdelay * DRM_HZ)/1000));
 }
 EXPORT_SYMBOL(drm_vblank_put);
 
@@ -480,10 +944,8 @@ void drm_vblank_off(struct drm_device *dev, int crtc)
 	unsigned long irqflags;
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	dev->driver->disable_vblank(dev, crtc);
+	vblank_disable_and_save(dev, crtc);
 	DRM_WAKEUP(&dev->vbl_queue[crtc]);
-	dev->vblank_enabled[crtc] = 0;
-	dev->last_vblank[crtc] = dev->driver->get_vblank_counter(dev, crtc);
 	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 }
 EXPORT_SYMBOL(drm_vblank_off);
@@ -602,7 +1064,6 @@ static int drm_queue_vblank_event(struct drm_device *dev, int pipe,
 	e->base.file_priv = file_priv;
 	e->base.destroy = (void (*) (struct drm_pending_event *)) kfree;
 
-	do_gettimeofday(&now);
 	spin_lock_irqsave(&dev->event_lock, flags);
 
 	if (file_priv->event_space < sizeof e->event) {
@@ -611,7 +1072,8 @@ static int drm_queue_vblank_event(struct drm_device *dev, int pipe,
 	}
 
 	file_priv->event_space -= sizeof e->event;
-	seq = drm_vblank_count(dev, pipe);
+	seq = drm_vblank_count_and_time(dev, pipe, &now);
+
 	if ((vblwait->request.type & _DRM_VBLANK_NEXTONMISS) &&
 	    (seq - vblwait->request.sequence) <= (1 << 23)) {
 		vblwait->request.sequence = seq + 1;
@@ -727,11 +1189,10 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 	if (ret != -EINTR) {
 		struct timeval now;
 
-		do_gettimeofday(&now);
-
+		vblwait->reply.sequence = drm_vblank_count_and_time(dev, crtc, &now);
 		vblwait->reply.tval_sec = now.tv_sec;
 		vblwait->reply.tval_usec = now.tv_usec;
-		vblwait->reply.sequence = drm_vblank_count(dev, crtc);
+
 		DRM_DEBUG("returning %d to client\n",
 			  vblwait->reply.sequence);
 	} else {
@@ -750,8 +1211,7 @@ void drm_handle_vblank_events(struct drm_device *dev, int crtc)
 	unsigned long flags;
 	unsigned int seq;
 
-	do_gettimeofday(&now);
-	seq = drm_vblank_count(dev, crtc);
+	seq = drm_vblank_count_and_time(dev, crtc, &now);
 
 	spin_lock_irqsave(&dev->event_lock, flags);
 
@@ -789,11 +1249,64 @@ void drm_handle_vblank_events(struct drm_device *dev, int crtc)
  */
 void drm_handle_vblank(struct drm_device *dev, int crtc)
 {
+	u32 vblcount;
+	s64 diff_ns;
+	struct timeval tvblank;
+	unsigned long irqflags;
+
 	if (!dev->num_crtcs)
 		return;
 
-	atomic_inc(&dev->_vblank_count[crtc]);
+	/* Need timestamp lock to prevent concurrent execution with
+	 * vblank enable/disable, as this would cause inconsistent
+	 * or corrupted timestamps and vblank counts.
+	 */
+	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
+
+	/* Vblank irq handling disabled. Nothing to do. */
+	if (!dev->vblank_enabled[crtc]) {
+		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
+		return;
+	}
+
+	/* Fetch corresponding timestamp for this vblank interval from
+	 * driver and store it in proper slot of timestamp ringbuffer.
+	 */
+
+	/* Get current timestamp and count. */
+	vblcount = atomic_read(&dev->_vblank_count[crtc]);
+	drm_get_last_vbltimestamp(dev, crtc, &tvblank, DRM_CALLED_FROM_VBLIRQ);
+
+	/* Compute time difference to timestamp of last vblank */
+	diff_ns = timeval_to_ns(&tvblank) -
+		  timeval_to_ns(&vblanktimestamp(dev, crtc, vblcount));
+
+	/* Update vblank timestamp and count if at least
+	 * DRM_REDUNDANT_VBLIRQ_THRESH_NS nanoseconds
+	 * difference between last stored timestamp and current
+	 * timestamp. A smaller difference means basically
+	 * identical timestamps. Happens if this vblank has
+	 * been already processed and this is a redundant call,
+	 * e.g., due to spurious vblank interrupts. We need to
+	 * ignore those for accounting.
+	 */
+	if (abs(diff_ns) > DRM_REDUNDANT_VBLIRQ_THRESH_NS) {
+		/* Store new timestamp in ringbuffer. */
+		vblanktimestamp(dev, crtc, vblcount + 1) = tvblank;
+		smp_wmb();
+
+		/* Increment cooked vblank count. This also atomically commits
+		 * the timestamp computed above.
+		 */
+		atomic_inc(&dev->_vblank_count[crtc]);
+	} else {
+		DRM_DEBUG("crtc %d: Redundant vblirq ignored. diff_ns = %d\n",
+			  crtc, (int) diff_ns);
+	}
+
 	DRM_WAKEUP(&dev->vbl_queue[crtc]);
 	drm_handle_vblank_events(dev, crtc);
+
+	spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
 }
 EXPORT_SYMBOL(drm_handle_vblank);
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index cdc89ee042cc..d59edc18301f 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -40,12 +40,22 @@
 unsigned int drm_debug = 0;	/* 1 to enable debug output */
 EXPORT_SYMBOL(drm_debug);
 
+unsigned int drm_vblank_offdelay = 5000;    /* Default to 5000 msecs. */
+EXPORT_SYMBOL(drm_vblank_offdelay);
+
+unsigned int drm_timestamp_precision = 20;  /* Default to 20 usecs. */
+EXPORT_SYMBOL(drm_timestamp_precision);
+
 MODULE_AUTHOR(CORE_AUTHOR);
 MODULE_DESCRIPTION(CORE_DESC);
 MODULE_LICENSE("GPL and additional rights");
 MODULE_PARM_DESC(debug, "Enable debug output");
+MODULE_PARM_DESC(vblankoffdelay, "Delay until vblank irq auto-disable [msecs]");
+MODULE_PARM_DESC(timestamp_precision_usec, "Max. error on timestamps [usecs]");
 
 module_param_named(debug, drm_debug, int, 0600);
+module_param_named(vblankoffdelay, drm_vblank_offdelay, int, 0600);
+module_param_named(timestamp_precision_usec, drm_timestamp_precision, int, 0600);
 
 struct idr drm_minors_idr;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 9a1fdcf400c2..1f2301d26c0a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -234,10 +234,10 @@ validate_fini_list(struct list_head *list, struct nouveau_fence *fence)
 		if (likely(fence)) {
 			struct nouveau_fence *prev_fence;
 
-			spin_lock(&nvbo->bo.lock);
+			spin_lock(&nvbo->bo.bdev->fence_lock);
 			prev_fence = nvbo->bo.sync_obj;
 			nvbo->bo.sync_obj = nouveau_fence_ref(fence);
-			spin_unlock(&nvbo->bo.lock);
+			spin_unlock(&nvbo->bo.bdev->fence_lock);
 			nouveau_fence_unref((void *)&prev_fence);
 		}
 
@@ -557,9 +557,9 @@ nouveau_gem_pushbuf_reloc_apply(struct drm_device *dev,
 				data |= r->vor;
 		}
 
-		spin_lock(&nvbo->bo.lock);
+		spin_lock(&nvbo->bo.bdev->fence_lock);
 		ret = ttm_bo_wait(&nvbo->bo, false, false, false);
-		spin_unlock(&nvbo->bo.lock);
+		spin_unlock(&nvbo->bo.bdev->fence_lock);
 		if (ret) {
 			NV_ERROR(dev, "reloc wait_idle failed: %d\n", ret);
 			break;
@@ -791,9 +791,9 @@ nouveau_gem_ioctl_cpu_prep(struct drm_device *dev, void *data,
 	}
 
 	if (req->flags & NOUVEAU_GEM_CPU_PREP_NOBLOCK) {
-		spin_lock(&nvbo->bo.lock);
+		spin_lock(&nvbo->bo.bdev->fence_lock);
 		ret = ttm_bo_wait(&nvbo->bo, false, false, no_wait);
-		spin_unlock(&nvbo->bo.lock);
+		spin_unlock(&nvbo->bo.bdev->fence_lock);
 	} else {
 		ret = ttm_bo_synccpu_write_grab(&nvbo->bo, no_wait);
 		if (ret == 0)
diff --git a/drivers/gpu/drm/radeon/ObjectID.h b/drivers/gpu/drm/radeon/ObjectID.h
index c714179d1bfa..c61c3fe9fb98 100644
--- a/drivers/gpu/drm/radeon/ObjectID.h
+++ b/drivers/gpu/drm/radeon/ObjectID.h
@@ -37,6 +37,8 @@
 #define GRAPH_OBJECT_TYPE_CONNECTOR               0x3
 #define GRAPH_OBJECT_TYPE_ROUTER                  0x4
 /* deleted */
+#define GRAPH_OBJECT_TYPE_DISPLAY_PATH            0x6  
+#define GRAPH_OBJECT_TYPE_GENERIC                 0x7
 
 /****************************************************/
 /* Encoder Object ID Definition                     */
@@ -64,6 +66,9 @@
 #define ENCODER_OBJECT_ID_VT1623                  0x10
 #define ENCODER_OBJECT_ID_HDMI_SI1930             0x11
 #define ENCODER_OBJECT_ID_HDMI_INTERNAL           0x12
+#define ENCODER_OBJECT_ID_ALMOND                  0x22
+#define ENCODER_OBJECT_ID_TRAVIS                  0x23
+#define ENCODER_OBJECT_ID_NUTMEG                  0x22
 /* Kaleidoscope (KLDSCP) Class Display Hardware (internal) */
 #define ENCODER_OBJECT_ID_INTERNAL_KLDSCP_TMDS1   0x13
 #define ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1    0x14
@@ -108,6 +113,7 @@
 #define CONNECTOR_OBJECT_ID_DISPLAYPORT           0x13
 #define CONNECTOR_OBJECT_ID_eDP                   0x14
 #define CONNECTOR_OBJECT_ID_MXM                   0x15
+#define CONNECTOR_OBJECT_ID_LVDS_eDP              0x16
 
 /* deleted */
 
@@ -124,6 +130,7 @@
 #define GENERIC_OBJECT_ID_GLSYNC                  0x01
 #define GENERIC_OBJECT_ID_PX2_NON_DRIVABLE        0x02
 #define GENERIC_OBJECT_ID_MXM_OPM                 0x03
+#define GENERIC_OBJECT_ID_STEREO_PIN              0x04        //This object could show up from Misc Object table, it follows ATOM_OBJECT format, and contains one ATOM_OBJECT_GPIO_CNTL_RECORD for the stereo pin
 
 /****************************************************/
 /* Graphics Object ENUM ID Definition               */
@@ -360,6 +367,26 @@
                                                   GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
                                                   ENCODER_OBJECT_ID_GENERAL_EXTERNAL_DVO << OBJECT_ID_SHIFT)
 
+#define ENCODER_ALMOND_ENUM_ID1                  ( GRAPH_OBJECT_TYPE_ENCODER << OBJECT_TYPE_SHIFT |\
+                                                  GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
+                                                  ENCODER_OBJECT_ID_ALMOND << OBJECT_ID_SHIFT)
+
+#define ENCODER_ALMOND_ENUM_ID2                  ( GRAPH_OBJECT_TYPE_ENCODER << OBJECT_TYPE_SHIFT |\
+                                                  GRAPH_OBJECT_ENUM_ID2 << ENUM_ID_SHIFT |\
+                                                  ENCODER_OBJECT_ID_ALMOND << OBJECT_ID_SHIFT)
+
+#define ENCODER_TRAVIS_ENUM_ID1                  ( GRAPH_OBJECT_TYPE_ENCODER << OBJECT_TYPE_SHIFT |\
+                                                  GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
+                                                  ENCODER_OBJECT_ID_TRAVIS << OBJECT_ID_SHIFT)
+
+#define ENCODER_TRAVIS_ENUM_ID2                  ( GRAPH_OBJECT_TYPE_ENCODER << OBJECT_TYPE_SHIFT |\
+                                                  GRAPH_OBJECT_ENUM_ID2 << ENUM_ID_SHIFT |\
+                                                  ENCODER_OBJECT_ID_TRAVIS << OBJECT_ID_SHIFT)
+
+#define ENCODER_NUTMEG_ENUM_ID1                  ( GRAPH_OBJECT_TYPE_ENCODER << OBJECT_TYPE_SHIFT |\
+                                                  GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
+                                                  ENCODER_OBJECT_ID_NUTMEG << OBJECT_ID_SHIFT)
+
 /****************************************************/
 /* Connector Object ID definition - Shared with BIOS */
 /****************************************************/
@@ -421,6 +448,14 @@
                                                  GRAPH_OBJECT_ENUM_ID2 << ENUM_ID_SHIFT |\
                                                  CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_D << OBJECT_ID_SHIFT)
 
+#define CONNECTOR_SINGLE_LINK_DVI_D_ENUM_ID3   ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
+                                                 GRAPH_OBJECT_ENUM_ID3 << ENUM_ID_SHIFT |\
+                                                 CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_D << OBJECT_ID_SHIFT)
+
+#define CONNECTOR_SINGLE_LINK_DVI_D_ENUM_ID4   ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
+                                                 GRAPH_OBJECT_ENUM_ID4 << ENUM_ID_SHIFT |\
+                                                 CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_D << OBJECT_ID_SHIFT)
+
 #define CONNECTOR_DUAL_LINK_DVI_D_ENUM_ID1     ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
                                                  GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
                                                  CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D << OBJECT_ID_SHIFT)
@@ -512,6 +547,7 @@
 #define CONNECTOR_7PIN_DIN_ENUM_ID1            ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
                                                  GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
                                                  CONNECTOR_OBJECT_ID_7PIN_DIN << OBJECT_ID_SHIFT)
+
 #define CONNECTOR_7PIN_DIN_ENUM_ID2            ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
                                                  GRAPH_OBJECT_ENUM_ID2 << ENUM_ID_SHIFT |\
                                                  CONNECTOR_OBJECT_ID_7PIN_DIN << OBJECT_ID_SHIFT)
@@ -593,6 +629,14 @@
                                                  GRAPH_OBJECT_ENUM_ID7 << ENUM_ID_SHIFT |\
                                                  CONNECTOR_OBJECT_ID_MXM << OBJECT_ID_SHIFT)          //Mapping to MXM_DAC
 
+#define CONNECTOR_LVDS_eDP_ENUM_ID1            ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
+                                                 GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
+                                                 CONNECTOR_OBJECT_ID_LVDS_eDP << OBJECT_ID_SHIFT)
+
+#define CONNECTOR_LVDS_eDP_ENUM_ID2            ( GRAPH_OBJECT_TYPE_CONNECTOR << OBJECT_TYPE_SHIFT |\
+                                                 GRAPH_OBJECT_ENUM_ID2 << ENUM_ID_SHIFT |\
+                                                 CONNECTOR_OBJECT_ID_LVDS_eDP << OBJECT_ID_SHIFT)
+
 /****************************************************/
 /* Router Object ID definition - Shared with BIOS   */
 /****************************************************/
@@ -621,6 +665,10 @@
                                                  GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
                                                  GENERIC_OBJECT_ID_MXM_OPM << OBJECT_ID_SHIFT)
 
+#define GENERICOBJECT_STEREO_PIN_ENUM_ID1        (GRAPH_OBJECT_TYPE_GENERIC << OBJECT_TYPE_SHIFT |\
+                                                 GRAPH_OBJECT_ENUM_ID1 << ENUM_ID_SHIFT |\
+                                                 GENERIC_OBJECT_ID_STEREO_PIN << OBJECT_ID_SHIFT)
+
 /****************************************************/
 /* Object Cap definition - Shared with BIOS         */
 /****************************************************/
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index fe359a239df3..58a0cd02c0a2 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -73,8 +73,18 @@
 #define ATOM_PPLL1            0
 #define ATOM_PPLL2            1
 #define ATOM_DCPLL            2
+#define ATOM_PPLL0            2
+#define ATOM_EXT_PLL1         8
+#define ATOM_EXT_PLL2         9
+#define ATOM_EXT_CLOCK        10
 #define ATOM_PPLL_INVALID     0xFF
 
+#define ENCODER_REFCLK_SRC_P1PLL       0       
+#define ENCODER_REFCLK_SRC_P2PLL       1
+#define ENCODER_REFCLK_SRC_DCPLL       2
+#define ENCODER_REFCLK_SRC_EXTCLK      3
+#define ENCODER_REFCLK_SRC_INVALID     0xFF
+
 #define ATOM_SCALER1          0
 #define ATOM_SCALER2          1
 
@@ -192,6 +202,9 @@ typedef struct _ATOM_COMMON_TABLE_HEADER
                                   /*Image can't be updated, while Driver needs to carry the new table! */
 }ATOM_COMMON_TABLE_HEADER;
 
+/****************************************************************************/	
+// Structure stores the ROM header.
+/****************************************************************************/	
 typedef struct _ATOM_ROM_HEADER
 {
   ATOM_COMMON_TABLE_HEADER		sHeader;
@@ -221,6 +234,9 @@ typedef struct _ATOM_ROM_HEADER
 	#define	USHORT	void*
 #endif
 
+/****************************************************************************/	
+// Structures used in Command.mtb 
+/****************************************************************************/	
 typedef struct _ATOM_MASTER_LIST_OF_COMMAND_TABLES{
   USHORT ASIC_Init;                              //Function Table, used by various SW components,latest version 1.1
   USHORT GetDisplaySurfaceSize;                  //Atomic Table,  Used by Bios when enabling HW ICON
@@ -312,6 +328,7 @@ typedef struct _ATOM_MASTER_LIST_OF_COMMAND_TABLES{
 #define SetUniphyInstance                        ASIC_StaticPwrMgtStatusChange
 #define HPDInterruptService                      ReadHWAssistedI2CStatus
 #define EnableVGA_Access                         GetSCLKOverMCLKRatio
+#define GetDispObjectInfo                        EnableYUV 
 
 typedef struct _ATOM_MASTER_COMMAND_TABLE
 {
@@ -357,6 +374,24 @@ typedef struct _ATOM_COMMON_ROM_COMMAND_TABLE_HEADER
 /****************************************************************************/	
 #define COMPUTE_MEMORY_PLL_PARAM        1
 #define COMPUTE_ENGINE_PLL_PARAM        2
+#define ADJUST_MC_SETTING_PARAM         3
+
+/****************************************************************************/	
+// Structures used by AdjustMemoryControllerTable
+/****************************************************************************/	
+typedef struct _ATOM_ADJUST_MEMORY_CLOCK_FREQ
+{
+#if ATOM_BIG_ENDIAN
+  ULONG ulPointerReturnFlag:1;      // BYTE_3[7]=1 - Return the pointer to the right Data Block; BYTE_3[7]=0 - Program the right Data Block 
+  ULONG ulMemoryModuleNumber:7;     // BYTE_3[6:0]
+  ULONG ulClockFreq:24;
+#else
+  ULONG ulClockFreq:24;
+  ULONG ulMemoryModuleNumber:7;     // BYTE_3[6:0]
+  ULONG ulPointerReturnFlag:1;      // BYTE_3[7]=1 - Return the pointer to the right Data Block; BYTE_3[7]=0 - Program the right Data Block 
+#endif
+}ATOM_ADJUST_MEMORY_CLOCK_FREQ;
+#define POINTER_RETURN_FLAG             0x80
 
 typedef struct _COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS
 {
@@ -440,6 +475,26 @@ typedef struct _COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS_V4
 #endif
 }COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS_V4;
 
+typedef struct _COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS_V5
+{
+  union
+  {
+    ATOM_COMPUTE_CLOCK_FREQ  ulClock;         //Input Parameter
+    ATOM_S_MPLL_FB_DIVIDER   ulFbDiv;         //Output Parameter
+  };
+  UCHAR   ucRefDiv;                           //Output Parameter      
+  UCHAR   ucPostDiv;                          //Output Parameter      
+  union
+  {
+    UCHAR   ucCntlFlag;                       //Output Flags
+    UCHAR   ucInputFlag;                      //Input Flags. ucInputFlag[0] - Strobe(1)/Performance(0) mode
+  };
+  UCHAR   ucReserved;                       
+}COMPUTE_MEMORY_ENGINE_PLL_PARAMETERS_V5;
+
+// ucInputFlag
+#define ATOM_PLL_INPUT_FLAG_PLL_STROBE_MODE_EN  1   // 1-StrobeMode, 0-PerformanceMode
+
 typedef struct _DYNAMICE_MEMORY_SETTINGS_PARAMETER
 {
   ATOM_COMPUTE_CLOCK_FREQ ulClock;
@@ -583,6 +638,7 @@ typedef struct _DIG_ENCODER_CONTROL_PARAMETERS
 #define ATOM_ENCODER_CONFIG_DPLINKRATE_MASK				0x01
 #define ATOM_ENCODER_CONFIG_DPLINKRATE_1_62GHZ		0x00
 #define ATOM_ENCODER_CONFIG_DPLINKRATE_2_70GHZ		0x01
+#define ATOM_ENCODER_CONFIG_DPLINKRATE_5_40GHZ		0x02
 #define ATOM_ENCODER_CONFIG_LINK_SEL_MASK				  0x04
 #define ATOM_ENCODER_CONFIG_LINKA								  0x00
 #define ATOM_ENCODER_CONFIG_LINKB								  0x04
@@ -608,6 +664,9 @@ typedef struct _DIG_ENCODER_CONTROL_PARAMETERS
 #define ATOM_ENCODER_MODE_TV											13
 #define ATOM_ENCODER_MODE_CV											14
 #define ATOM_ENCODER_MODE_CRT											15
+#define ATOM_ENCODER_MODE_DVO											16
+#define ATOM_ENCODER_MODE_DP_SST                  ATOM_ENCODER_MODE_DP    // For DP1.2
+#define ATOM_ENCODER_MODE_DP_MST                  5                       // For DP1.2
 
 typedef struct _ATOM_DIG_ENCODER_CONFIG_V2
 {
@@ -661,6 +720,7 @@ typedef struct _DIG_ENCODER_CONTROL_PARAMETERS_V2
 #define ATOM_ENCODER_CMD_DP_LINK_TRAINING_START       0x08
 #define ATOM_ENCODER_CMD_DP_LINK_TRAINING_PATTERN1    0x09
 #define ATOM_ENCODER_CMD_DP_LINK_TRAINING_PATTERN2    0x0a
+#define ATOM_ENCODER_CMD_DP_LINK_TRAINING_PATTERN3    0x13
 #define ATOM_ENCODER_CMD_DP_LINK_TRAINING_COMPLETE    0x0b
 #define ATOM_ENCODER_CMD_DP_VIDEO_OFF                 0x0c
 #define ATOM_ENCODER_CMD_DP_VIDEO_ON                  0x0d
@@ -671,24 +731,34 @@ typedef struct _DIG_ENCODER_CONTROL_PARAMETERS_V2
 #define ATOM_ENCODER_STATUS_LINK_TRAINING_COMPLETE    0x10
 #define ATOM_ENCODER_STATUS_LINK_TRAINING_INCOMPLETE  0x00
 
+//ucTableFormatRevision=1
+//ucTableContentRevision=3
 // Following function ENABLE sub-function will be used by driver when TMDS/HDMI/LVDS is used, disable function will be used by driver
 typedef struct _ATOM_DIG_ENCODER_CONFIG_V3
 {
 #if ATOM_BIG_ENDIAN
     UCHAR ucReserved1:1;
-    UCHAR ucDigSel:3;             // =0: DIGA/B/C/D/E/F
+    UCHAR ucDigSel:3;             // =0/1/2/3/4/5: DIG0/1/2/3/4/5 (In register spec also refered as DIGA/B/C/D/E/F)
     UCHAR ucReserved:3;
     UCHAR ucDPLinkRate:1;         // =0: 1.62Ghz, =1: 2.7Ghz
 #else
     UCHAR ucDPLinkRate:1;         // =0: 1.62Ghz, =1: 2.7Ghz
     UCHAR ucReserved:3;
-    UCHAR ucDigSel:3;             // =0: DIGA/B/C/D/E/F
+    UCHAR ucDigSel:3;             // =0/1/2/3/4/5: DIG0/1/2/3/4/5 (In register spec also refered as DIGA/B/C/D/E/F)
     UCHAR ucReserved1:1;
 #endif
 }ATOM_DIG_ENCODER_CONFIG_V3;
 
+#define ATOM_ENCODER_CONFIG_V3_DPLINKRATE_MASK				0x03
+#define ATOM_ENCODER_CONFIG_V3_DPLINKRATE_1_62GHZ		  0x00
+#define ATOM_ENCODER_CONFIG_V3_DPLINKRATE_2_70GHZ		  0x01
 #define ATOM_ENCODER_CONFIG_V3_ENCODER_SEL					  0x70
-
+#define ATOM_ENCODER_CONFIG_V3_DIG0_ENCODER					  0x00
+#define ATOM_ENCODER_CONFIG_V3_DIG1_ENCODER					  0x10
+#define ATOM_ENCODER_CONFIG_V3_DIG2_ENCODER					  0x20
+#define ATOM_ENCODER_CONFIG_V3_DIG3_ENCODER					  0x30
+#define ATOM_ENCODER_CONFIG_V3_DIG4_ENCODER					  0x40
+#define ATOM_ENCODER_CONFIG_V3_DIG5_ENCODER					  0x50
 
 typedef struct _DIG_ENCODER_CONTROL_PARAMETERS_V3
 {
@@ -707,6 +777,56 @@ typedef struct _DIG_ENCODER_CONTROL_PARAMETERS_V3
   UCHAR ucReserved;
 }DIG_ENCODER_CONTROL_PARAMETERS_V3;
 
+//ucTableFormatRevision=1
+//ucTableContentRevision=4
+// start from NI           
+// Following function ENABLE sub-function will be used by driver when TMDS/HDMI/LVDS is used, disable function will be used by driver
+typedef struct _ATOM_DIG_ENCODER_CONFIG_V4
+{
+#if ATOM_BIG_ENDIAN
+    UCHAR ucReserved1:1;
+    UCHAR ucDigSel:3;             // =0/1/2/3/4/5: DIG0/1/2/3/4/5 (In register spec also refered as DIGA/B/C/D/E/F)
+    UCHAR ucReserved:2;
+    UCHAR ucDPLinkRate:2;         // =0: 1.62Ghz, =1: 2.7Ghz, 2=5.4Ghz    <= Changed comparing to previous version
+#else
+    UCHAR ucDPLinkRate:2;         // =0: 1.62Ghz, =1: 2.7Ghz, 2=5.4Ghz    <= Changed comparing to previous version
+    UCHAR ucReserved:2;
+    UCHAR ucDigSel:3;             // =0/1/2/3/4/5: DIG0/1/2/3/4/5 (In register spec also refered as DIGA/B/C/D/E/F)
+    UCHAR ucReserved1:1;
+#endif
+}ATOM_DIG_ENCODER_CONFIG_V4;
+
+#define ATOM_ENCODER_CONFIG_V4_DPLINKRATE_MASK				0x03
+#define ATOM_ENCODER_CONFIG_V4_DPLINKRATE_1_62GHZ		  0x00
+#define ATOM_ENCODER_CONFIG_V4_DPLINKRATE_2_70GHZ		  0x01
+#define ATOM_ENCODER_CONFIG_V4_DPLINKRATE_5_40GHZ		  0x02
+#define ATOM_ENCODER_CONFIG_V4_ENCODER_SEL					  0x70
+#define ATOM_ENCODER_CONFIG_V4_DIG0_ENCODER					  0x00
+#define ATOM_ENCODER_CONFIG_V4_DIG1_ENCODER					  0x10
+#define ATOM_ENCODER_CONFIG_V4_DIG2_ENCODER					  0x20
+#define ATOM_ENCODER_CONFIG_V4_DIG3_ENCODER					  0x30
+#define ATOM_ENCODER_CONFIG_V4_DIG4_ENCODER					  0x40
+#define ATOM_ENCODER_CONFIG_V4_DIG5_ENCODER					  0x50
+
+typedef struct _DIG_ENCODER_CONTROL_PARAMETERS_V4
+{
+  USHORT usPixelClock;      // in 10KHz; for bios convenient
+  union{
+  ATOM_DIG_ENCODER_CONFIG_V4 acConfig;
+  UCHAR ucConfig;
+  };
+  UCHAR ucAction;                              
+  UCHAR ucEncoderMode;
+                            // =0: DP   encoder      
+                            // =1: LVDS encoder          
+                            // =2: DVI  encoder  
+                            // =3: HDMI encoder
+                            // =4: SDVO encoder
+                            // =5: DP audio
+  UCHAR ucLaneNum;          // how many lanes to enable
+  UCHAR ucBitPerColor;      // only valid for DP mode when ucAction = ATOM_ENCODER_CMD_SETUP
+  UCHAR ucHPD_ID;           // HPD ID (1-6). =0 means to skip HDP programming. New comparing to previous version
+}DIG_ENCODER_CONTROL_PARAMETERS_V4;
 
 // define ucBitPerColor: 
 #define PANEL_BPC_UNDEFINE                               0x00
@@ -893,6 +1013,7 @@ typedef struct _ATOM_DIG_TRANSMITTER_CONFIG_V3
 #endif
 }ATOM_DIG_TRANSMITTER_CONFIG_V3;
 
+
 typedef struct _DIG_TRANSMITTER_CONTROL_PARAMETERS_V3
 {
 	union
@@ -936,6 +1057,149 @@ typedef struct _DIG_TRANSMITTER_CONTROL_PARAMETERS_V3
 #define ATOM_TRANSMITTER_CONFIG_V3_TRANSMITTER2           	0x40	//CD
 #define ATOM_TRANSMITTER_CONFIG_V3_TRANSMITTER3           	0x80	//EF
 
+
+/****************************************************************************/	
+// Structures used by UNIPHYTransmitterControlTable V1.4
+// ASIC Families: NI
+// ucTableFormatRevision=1
+// ucTableContentRevision=4
+/****************************************************************************/	
+typedef struct _ATOM_DP_VS_MODE_V4
+{
+  UCHAR ucLaneSel;
+ 	union
+ 	{  
+ 	  UCHAR ucLaneSet;
+ 	  struct {
+#if ATOM_BIG_ENDIAN
+ 		  UCHAR ucPOST_CURSOR2:2;         //Bit[7:6] Post Cursor2 Level      <= New in V4
+ 		  UCHAR ucPRE_EMPHASIS:3;         //Bit[5:3] Pre-emphasis Level
+ 		  UCHAR ucVOLTAGE_SWING:3;        //Bit[2:0] Voltage Swing Level
+#else
+ 		  UCHAR ucVOLTAGE_SWING:3;        //Bit[2:0] Voltage Swing Level
+ 		  UCHAR ucPRE_EMPHASIS:3;         //Bit[5:3] Pre-emphasis Level
+ 		  UCHAR ucPOST_CURSOR2:2;         //Bit[7:6] Post Cursor2 Level      <= New in V4
+#endif
+ 		};
+ 	}; 
+}ATOM_DP_VS_MODE_V4;
+ 
+typedef struct _ATOM_DIG_TRANSMITTER_CONFIG_V4
+{
+#if ATOM_BIG_ENDIAN
+  UCHAR ucTransmitterSel:2;         //bit7:6: =0 Dig Transmitter 1 ( Uniphy AB )
+                                    //        =1 Dig Transmitter 2 ( Uniphy CD )
+                                    //        =2 Dig Transmitter 3 ( Uniphy EF )
+  UCHAR ucRefClkSource:2;           //bit5:4: PPLL1 =0, PPLL2=1, DCPLL=2, EXT_CLK=3   <= New
+  UCHAR ucEncoderSel:1;             //bit3=0: Data/Clk path source from DIGA/C/E. =1: Data/clk path source from DIGB/D/F
+  UCHAR ucLinkSel:1;                //bit2=0: Uniphy LINKA or C or E when fDualLinkConnector=0. when fDualLinkConnector=1, it means master link of dual link is A or C or E
+                                    //    =1: Uniphy LINKB or D or F when fDualLinkConnector=0. when fDualLinkConnector=1, it means master link of dual link is B or D or F
+  UCHAR fCoherentMode:1;            //bit1=1: Coherent Mode ( for DVI/HDMI mode )
+  UCHAR fDualLinkConnector:1;       //bit0=1: Dual Link DVI connector
+#else
+  UCHAR fDualLinkConnector:1;       //bit0=1: Dual Link DVI connector
+  UCHAR fCoherentMode:1;            //bit1=1: Coherent Mode ( for DVI/HDMI mode )
+  UCHAR ucLinkSel:1;                //bit2=0: Uniphy LINKA or C or E when fDualLinkConnector=0. when fDualLinkConnector=1, it means master link of dual link is A or C or E
+                                    //    =1: Uniphy LINKB or D or F when fDualLinkConnector=0. when fDualLinkConnector=1, it means master link of dual link is B or D or F
+  UCHAR ucEncoderSel:1;             //bit3=0: Data/Clk path source from DIGA/C/E. =1: Data/clk path source from DIGB/D/F
+  UCHAR ucRefClkSource:2;           //bit5:4: PPLL1 =0, PPLL2=1, DCPLL=2, EXT_CLK=3   <= New
+  UCHAR ucTransmitterSel:2;         //bit7:6: =0 Dig Transmitter 1 ( Uniphy AB )
+                                    //        =1 Dig Transmitter 2 ( Uniphy CD )
+                                    //        =2 Dig Transmitter 3 ( Uniphy EF )
+#endif
+}ATOM_DIG_TRANSMITTER_CONFIG_V4;
+
+typedef struct _DIG_TRANSMITTER_CONTROL_PARAMETERS_V4
+{
+  union
+  {
+    USHORT usPixelClock;		// in 10KHz; for bios convenient
+    USHORT usInitInfo;			// when init uniphy,lower 8bit is used for connector type defined in objectid.h
+    ATOM_DP_VS_MODE_V4 asMode; // DP Voltage swing mode     Redefined comparing to previous version
+  };
+  union
+  {
+  ATOM_DIG_TRANSMITTER_CONFIG_V4 acConfig;
+  UCHAR ucConfig;
+  };
+  UCHAR ucAction;				    // define as ATOM_TRANSMITER_ACTION_XXX	                        
+  UCHAR ucLaneNum;
+  UCHAR ucReserved[3];
+}DIG_TRANSMITTER_CONTROL_PARAMETERS_V4;
+
+//ucConfig 
+//Bit0
+#define ATOM_TRANSMITTER_CONFIG_V4_DUAL_LINK_CONNECTOR			0x01
+//Bit1
+#define ATOM_TRANSMITTER_CONFIG_V4_COHERENT				          0x02
+//Bit2
+#define ATOM_TRANSMITTER_CONFIG_V4_LINK_SEL_MASK		        0x04
+#define ATOM_TRANSMITTER_CONFIG_V4_LINKA  			            0x00			
+#define ATOM_TRANSMITTER_CONFIG_V4_LINKB				            0x04
+// Bit3
+#define ATOM_TRANSMITTER_CONFIG_V4_ENCODER_SEL_MASK	        0x08
+#define ATOM_TRANSMITTER_CONFIG_V4_DIG1_ENCODER		          0x00				 
+#define ATOM_TRANSMITTER_CONFIG_V4_DIG2_ENCODER		          0x08				
+// Bit5:4
+#define ATOM_TRANSMITTER_CONFIG_V4_REFCLK_SEL_MASK 	        0x30
+#define ATOM_TRANSMITTER_CONFIG_V4_P1PLL         		        0x00
+#define ATOM_TRANSMITTER_CONFIG_V4_P2PLL		                0x10
+#define ATOM_TRANSMITTER_CONFIG_V4_DCPLL		                0x20   // New in _V4
+#define ATOM_TRANSMITTER_CONFIG_V4_REFCLK_SRC_EXT           0x30   // Changed comparing to V3
+// Bit7:6
+#define ATOM_TRANSMITTER_CONFIG_V4_TRANSMITTER_SEL_MASK     0xC0
+#define ATOM_TRANSMITTER_CONFIG_V4_TRANSMITTER1           	0x00	//AB
+#define ATOM_TRANSMITTER_CONFIG_V4_TRANSMITTER2           	0x40	//CD
+#define ATOM_TRANSMITTER_CONFIG_V4_TRANSMITTER3           	0x80	//EF
+
+
+/****************************************************************************/	
+// Structures used by ExternalEncoderControlTable V1.3
+// ASIC Families: Evergreen, Llano, NI
+// ucTableFormatRevision=1
+// ucTableContentRevision=3
+/****************************************************************************/	
+
+typedef struct _EXTERNAL_ENCODER_CONTROL_PARAMETERS_V3
+{
+  union{
+  USHORT usPixelClock;      // pixel clock in 10Khz, valid when ucAction=SETUP/ENABLE_OUTPUT 
+  USHORT usConnectorId;     // connector id, valid when ucAction = INIT
+  };
+  UCHAR  ucConfig;          // indicate which encoder, and DP link rate when ucAction = SETUP/ENABLE_OUTPUT  
+  UCHAR  ucAction;          // 
+  UCHAR  ucEncoderMode;     // encoder mode, only used when ucAction = SETUP/ENABLE_OUTPUT
+  UCHAR  ucLaneNum;         // lane number, only used when ucAction = SETUP/ENABLE_OUTPUT  
+  UCHAR  ucBitPerColor;     // output bit per color, only valid when ucAction = SETUP/ENABLE_OUTPUT and ucEncodeMode= DP
+  UCHAR  ucReserved;        
+}EXTERNAL_ENCODER_CONTROL_PARAMETERS_V3;
+
+// ucAction
+#define EXTERNAL_ENCODER_ACTION_V3_DISABLE_OUTPUT         0x00
+#define EXTERNAL_ENCODER_ACTION_V3_ENABLE_OUTPUT          0x01
+#define EXTERNAL_ENCODER_ACTION_V3_ENCODER_INIT           0x07
+#define EXTERNAL_ENCODER_ACTION_V3_ENCODER_SETUP          0x0f
+#define EXTERNAL_ENCODER_ACTION_V3_ENCODER_BLANKING_OFF   0x10
+#define EXTERNAL_ENCODER_ACTION_V3_ENCODER_BLANKING       0x11
+#define EXTERNAL_ENCODER_ACTION_V3_DACLOAD_DETECTION      0x12
+
+// ucConfig
+#define EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_MASK				0x03
+#define EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_1_62GHZ		  0x00
+#define EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_2_70GHZ		  0x01
+#define EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_5_40GHZ		  0x02
+#define EXTERNAL_ENCODER_CONFIG_V3_ENCODER_SEL_MASK		    0x70
+#define EXTERNAL_ENCODER_CONFIG_V3_ENCODER1		            0x00
+#define EXTERNAL_ENCODER_CONFIG_V3_ENCODER2		            0x10
+#define EXTERNAL_ENCODER_CONFIG_V3_ENCODER3		            0x20
+
+typedef struct _EXTERNAL_ENCODER_CONTROL_PS_ALLOCATION_V3
+{
+  EXTERNAL_ENCODER_CONTROL_PARAMETERS_V3 sExtEncoder;
+  ULONG ulReserved[2];
+}EXTERNAL_ENCODER_CONTROL_PS_ALLOCATION_V3;
+
+
 /****************************************************************************/	
 // Structures used by DAC1OuputControlTable
 //                    DAC2OuputControlTable
@@ -1142,6 +1406,7 @@ typedef struct _PIXEL_CLOCK_PARAMETERS_V2
 #define PIXEL_CLOCK_V4_MISC_SS_ENABLE               0x10
 #define PIXEL_CLOCK_V4_MISC_COHERENT_MODE           0x20
 
+
 typedef struct _PIXEL_CLOCK_PARAMETERS_V3
 {
   USHORT usPixelClock;                // in 10kHz unit; for bios convenient = (RefClk*FB_Div)/(Ref_Div*Post_Div)
@@ -1202,6 +1467,55 @@ typedef struct _PIXEL_CLOCK_PARAMETERS_V5
 #define PIXEL_CLOCK_V5_MISC_HDMI_32BPP              0x08
 #define PIXEL_CLOCK_V5_MISC_REF_DIV_SRC             0x10
 
+typedef struct _CRTC_PIXEL_CLOCK_FREQ
+{
+#if ATOM_BIG_ENDIAN
+  ULONG  ucCRTC:8;            // ATOM_CRTC1~6, indicate the CRTC controller to 
+                              // drive the pixel clock. not used for DCPLL case.
+  ULONG  ulPixelClock:24;     // target the pixel clock to drive the CRTC timing. 
+                              // 0 means disable PPLL/DCPLL. Expanded to 24 bits comparing to previous version.
+#else
+  ULONG  ulPixelClock:24;     // target the pixel clock to drive the CRTC timing. 
+                              // 0 means disable PPLL/DCPLL. Expanded to 24 bits comparing to previous version.
+  ULONG  ucCRTC:8;            // ATOM_CRTC1~6, indicate the CRTC controller to 
+                              // drive the pixel clock. not used for DCPLL case.
+#endif
+}CRTC_PIXEL_CLOCK_FREQ;
+
+typedef struct _PIXEL_CLOCK_PARAMETERS_V6
+{
+  union{
+    CRTC_PIXEL_CLOCK_FREQ ulCrtcPclkFreq;    // pixel clock and CRTC id frequency 
+    ULONG ulDispEngClkFreq;                  // dispclk frequency
+  };
+  USHORT usFbDiv;            // feedback divider integer part. 
+  UCHAR  ucPostDiv;          // post divider. 
+  UCHAR  ucRefDiv;           // Reference divider
+  UCHAR  ucPpll;             // ATOM_PPLL1/ATOM_PPLL2/ATOM_DCPLL
+  UCHAR  ucTransmitterID;    // ASIC encoder id defined in objectId.h, 
+                             // indicate which graphic encoder will be used. 
+  UCHAR  ucEncoderMode;      // Encoder mode: 
+  UCHAR  ucMiscInfo;         // bit[0]= Force program PPLL 
+                             // bit[1]= when VGA timing is used. 
+                             // bit[3:2]= HDMI panel bit depth: =0: 24bpp =1:30bpp, =2:32bpp
+                             // bit[4]= RefClock source for PPLL. 
+                             // =0: XTLAIN( default mode )
+	                           // =1: other external clock source, which is pre-defined                                            
+                             //     by VBIOS depend on the feature required.
+                             // bit[7:5]: reserved.
+  ULONG  ulFbDivDecFrac;     // 20 bit feedback divider decimal fraction part, range from 1~999999 ( 0.000001 to 0.999999 )
+
+}PIXEL_CLOCK_PARAMETERS_V6;
+
+#define PIXEL_CLOCK_V6_MISC_FORCE_PROG_PPLL					0x01
+#define PIXEL_CLOCK_V6_MISC_VGA_MODE								0x02
+#define PIXEL_CLOCK_V6_MISC_HDMI_BPP_MASK           0x0c
+#define PIXEL_CLOCK_V6_MISC_HDMI_24BPP              0x00
+#define PIXEL_CLOCK_V6_MISC_HDMI_36BPP              0x04
+#define PIXEL_CLOCK_V6_MISC_HDMI_30BPP              0x08
+#define PIXEL_CLOCK_V6_MISC_HDMI_48BPP              0x0c
+#define PIXEL_CLOCK_V6_MISC_REF_DIV_SRC             0x10
+
 typedef struct _GET_DISP_PLL_STATUS_INPUT_PARAMETERS_V2
 {
   PIXEL_CLOCK_PARAMETERS_V3 sDispClkInput;
@@ -1241,10 +1555,11 @@ typedef struct _ADJUST_DISPLAY_PLL_PARAMETERS
 typedef struct _ADJUST_DISPLAY_PLL_INPUT_PARAMETERS_V3
 {
 	USHORT usPixelClock;                    // target pixel clock
-	UCHAR ucTransmitterID;                  // transmitter id defined in objectid.h
+	UCHAR ucTransmitterID;                  // GPU transmitter id defined in objectid.h
 	UCHAR ucEncodeMode;                     // encoder mode: CRT, LVDS, DP, TMDS or HDMI
   UCHAR ucDispPllConfig;                 // display pll configure parameter defined as following DISPPLL_CONFIG_XXXX
-	UCHAR ucReserved[3];
+  UCHAR ucExtTransmitterID;               // external encoder id.
+	UCHAR ucReserved[2];
 }ADJUST_DISPLAY_PLL_INPUT_PARAMETERS_V3;
 
 // usDispPllConfig v1.2 for RoadRunner
@@ -1358,6 +1673,7 @@ typedef struct _SET_UP_HW_I2C_DATA_PARAMETERS
 /**************************************************************************/
 #define SPEED_FAN_CONTROL_PS_ALLOCATION   WRITE_ONE_BYTE_HW_I2C_DATA_PARAMETERS
 
+
 /****************************************************************************/	
 // Structures used by PowerConnectorDetectionTable
 /****************************************************************************/	
@@ -1438,6 +1754,31 @@ typedef struct _ENABLE_SPREAD_SPECTRUM_ON_PPLL_V2
 #define ATOM_PPLL_SS_AMOUNT_V2_NFRAC_MASK     0x0F00
 #define ATOM_PPLL_SS_AMOUNT_V2_NFRAC_SHIFT    8
 
+// Used by DCE5.0
+ typedef struct _ENABLE_SPREAD_SPECTRUM_ON_PPLL_V3
+{
+  USHORT  usSpreadSpectrumAmountFrac;   // SS_AMOUNT_DSFRAC New in DCE5.0
+  UCHAR   ucSpreadSpectrumType;	        // Bit[0]: 0-Down Spread,1-Center Spread. 
+                                        // Bit[1]: 1-Ext. 0-Int. 
+                                        // Bit[3:2]: =0 P1PLL =1 P2PLL =2 DCPLL
+                                        // Bits[7:4] reserved
+  UCHAR   ucEnable;	                    // ATOM_ENABLE or ATOM_DISABLE
+  USHORT  usSpreadSpectrumAmount;      	// Includes SS_AMOUNT_FBDIV[7:0] and SS_AMOUNT_NFRAC_SLIP[11:8]    
+  USHORT  usSpreadSpectrumStep;	        // SS_STEP_SIZE_DSFRAC
+}ENABLE_SPREAD_SPECTRUM_ON_PPLL_V3;
+    
+#define ATOM_PPLL_SS_TYPE_V3_DOWN_SPREAD      0x00
+#define ATOM_PPLL_SS_TYPE_V3_CENTRE_SPREAD    0x01
+#define ATOM_PPLL_SS_TYPE_V3_EXT_SPREAD       0x02
+#define ATOM_PPLL_SS_TYPE_V3_PPLL_SEL_MASK    0x0c
+#define ATOM_PPLL_SS_TYPE_V3_P1PLL            0x00
+#define ATOM_PPLL_SS_TYPE_V3_P2PLL            0x04
+#define ATOM_PPLL_SS_TYPE_V3_DCPLL            0x08
+#define ATOM_PPLL_SS_AMOUNT_V3_FBDIV_MASK     0x00FF
+#define ATOM_PPLL_SS_AMOUNT_V3_FBDIV_SHIFT    0
+#define ATOM_PPLL_SS_AMOUNT_V3_NFRAC_MASK     0x0F00
+#define ATOM_PPLL_SS_AMOUNT_V3_NFRAC_SHIFT    8
+
 #define ENABLE_SPREAD_SPECTRUM_ON_PPLL_PS_ALLOCATION  ENABLE_SPREAD_SPECTRUM_ON_PPLL
 
 /**************************************************************************/
@@ -1706,7 +2047,7 @@ typedef struct _ATOM_MASTER_LIST_OF_DATA_TABLES
   USHORT        StandardVESA_Timing;      // Only used by Bios
   USHORT        FirmwareInfo;             // Shared by various SW components,latest version 1.4
   USHORT        DAC_Info;                 // Will be obsolete from R600
-  USHORT        LVDS_Info;                // Shared by various SW components,latest version 1.1 
+  USHORT        LCD_Info;                 // Shared by various SW components,latest version 1.3, was called LVDS_Info 
   USHORT        TMDS_Info;                // Will be obsolete from R600
   USHORT        AnalogTV_Info;            // Shared by various SW components,latest version 1.1 
   USHORT        SupportedDevicesInfo;     // Will be obsolete from R600
@@ -1736,12 +2077,16 @@ typedef struct _ATOM_MASTER_LIST_OF_DATA_TABLES
 	USHORT				PowerSourceInfo;					// Shared by various SW components, latest versoin 1.1
 }ATOM_MASTER_LIST_OF_DATA_TABLES;
 
+// For backward compatible 
+#define LVDS_Info                LCD_Info
+
 typedef struct _ATOM_MASTER_DATA_TABLE
 { 
   ATOM_COMMON_TABLE_HEADER sHeader;  
   ATOM_MASTER_LIST_OF_DATA_TABLES   ListOfDataTables;
 }ATOM_MASTER_DATA_TABLE;
 
+
 /****************************************************************************/	
 // Structure used in MultimediaCapabilityInfoTable
 /****************************************************************************/	
@@ -1776,6 +2121,7 @@ typedef struct _ATOM_MULTIMEDIA_CONFIG_INFO
   UCHAR                    ucVideoInput4Info;// Video Input 4 Type (1:0) F/B setting (2) physical connector ID (5:3) reserved (7:6)
 }ATOM_MULTIMEDIA_CONFIG_INFO;
 
+
 /****************************************************************************/	
 // Structures used in FirmwareInfoTable
 /****************************************************************************/	
@@ -2031,8 +2377,47 @@ typedef struct _ATOM_FIRMWARE_INFO_V2_1
   UCHAR                           ucReserved4[3];
 }ATOM_FIRMWARE_INFO_V2_1;
 
+//the structure below to be used from NI
+//ucTableFormatRevision=2
+//ucTableContentRevision=2
+typedef struct _ATOM_FIRMWARE_INFO_V2_2
+{
+  ATOM_COMMON_TABLE_HEADER        sHeader; 
+  ULONG                           ulFirmwareRevision;
+  ULONG                           ulDefaultEngineClock;       //In 10Khz unit
+  ULONG                           ulDefaultMemoryClock;       //In 10Khz unit
+  ULONG                           ulReserved[2];
+  ULONG                           ulReserved1;                //Was ulMaxEngineClockPLL_Output; //In 10Khz unit*
+  ULONG                           ulReserved2;                //Was ulMaxMemoryClockPLL_Output; //In 10Khz unit*
+  ULONG                           ulMaxPixelClockPLL_Output;  //In 10Khz unit
+  ULONG                           ulBinaryAlteredInfo;        //Was ulASICMaxEngineClock  ?
+  ULONG                           ulDefaultDispEngineClkFreq; //In 10Khz unit. This is the frequency before DCDTO, corresponding to usBootUpVDDCVoltage.          
+  UCHAR                           ucReserved3;                //Was ucASICMaxTemperature;
+  UCHAR                           ucMinAllowedBL_Level;
+  USHORT                          usBootUpVDDCVoltage;        //In MV unit
+  USHORT                          usLcdMinPixelClockPLL_Output; // In MHz unit
+  USHORT                          usLcdMaxPixelClockPLL_Output; // In MHz unit
+  ULONG                           ulReserved4;                //Was ulAsicMaximumVoltage
+  ULONG                           ulMinPixelClockPLL_Output;  //In 10Khz unit
+  ULONG                           ulReserved5;                //Was usMinEngineClockPLL_Input and usMaxEngineClockPLL_Input
+  ULONG                           ulReserved6;                //Was usMinEngineClockPLL_Output and usMinMemoryClockPLL_Input
+  ULONG                           ulReserved7;                //Was usMaxMemoryClockPLL_Input and usMinMemoryClockPLL_Output
+  USHORT                          usReserved11;               //Was usMaxPixelClock;  //In 10Khz unit, Max.  Pclk used only for DAC
+  USHORT                          usMinPixelClockPLL_Input;   //In 10Khz unit
+  USHORT                          usMaxPixelClockPLL_Input;   //In 10Khz unit
+  USHORT                          usBootUpVDDCIVoltage;       //In unit of mv; Was usMinPixelClockPLL_Output;
+  ATOM_FIRMWARE_CAPABILITY_ACCESS usFirmwareCapability;
+  USHORT                          usCoreReferenceClock;       //In 10Khz unit	
+  USHORT                          usMemoryReferenceClock;     //In 10Khz unit	
+  USHORT                          usUniphyDPModeExtClkFreq;   //In 10Khz unit, if it is 0, In DP Mode Uniphy Input clock from internal PPLL, otherwise Input clock from external Spread clock
+  UCHAR                           ucMemoryModule_ID;          //Indicate what is the board design
+  UCHAR                           ucReserved9[3];
+  USHORT                          usBootUpMVDDCVoltage;       //In unit of mv; Was usMinPixelClockPLL_Output;
+  USHORT                          usReserved12;
+  ULONG                           ulReserved10[3];            // New added comparing to previous version
+}ATOM_FIRMWARE_INFO_V2_2;
 
-#define ATOM_FIRMWARE_INFO_LAST  ATOM_FIRMWARE_INFO_V2_1
+#define ATOM_FIRMWARE_INFO_LAST  ATOM_FIRMWARE_INFO_V2_2
 
 /****************************************************************************/	
 // Structures used in IntegratedSystemInfoTable
@@ -2212,7 +2597,7 @@ ulDockingPinCFGInfo: [15:0]-Bus/Device/Function # to CFG to read this Docking Pi
 ucDockingPinBit:     which bit in this register to read the pin status;
 ucDockingPinPolarity:Polarity of the pin when docked;
 
-ulCPUCapInfo:        [7:0]=1:Griffin;[7:0]=2:Greyhound;[7:0]=3:K8, other bits reserved for now and must be 0x0
+ulCPUCapInfo:        [7:0]=1:Griffin;[7:0]=2:Greyhound;[7:0]=3:K8, [7:0]=4:Pharaoh, other bits reserved for now and must be 0x0
 
 usNumberOfCyclesInPeriod:Indicate how many cycles when PWM duty is 100%.
 
@@ -2250,6 +2635,14 @@ usMinUpStreamHTLinkWidth:    Asymmetric link width support in the future, to rep
 usMinDownStreamHTLinkWidth:  same as above.
 */
 
+// ATOM_INTEGRATED_SYSTEM_INFO::ulCPUCapInfo  - CPU type definition 
+#define    INTEGRATED_SYSTEM_INFO__UNKNOWN_CPU             0
+#define    INTEGRATED_SYSTEM_INFO__AMD_CPU__GRIFFIN        1
+#define    INTEGRATED_SYSTEM_INFO__AMD_CPU__GREYHOUND      2
+#define    INTEGRATED_SYSTEM_INFO__AMD_CPU__K8             3
+#define    INTEGRATED_SYSTEM_INFO__AMD_CPU__PHARAOH        4
+
+#define    INTEGRATED_SYSTEM_INFO__AMD_CPU__MAX_CODE       INTEGRATED_SYSTEM_INFO__AMD_CPU__PHARAOH    // this deff reflects max defined CPU code
 
 #define SYSTEM_CONFIG_POWEREXPRESS_ENABLE                 0x00000001
 #define SYSTEM_CONFIG_RUN_AT_OVERDRIVE_ENGINE             0x00000002
@@ -2778,8 +3171,88 @@ typedef struct _ATOM_LVDS_INFO_V12
 #define PANEL_RANDOM_DITHER   0x80
 #define PANEL_RANDOM_DITHER_MASK   0x80
 
+#define ATOM_LVDS_INFO_LAST  ATOM_LVDS_INFO_V12   // no need to change this 
+
+/****************************************************************************/	
+// Structures used by LCD_InfoTable V1.3    Note: previous version was called ATOM_LVDS_INFO_V12
+// ASIC Families:  NI
+// ucTableFormatRevision=1
+// ucTableContentRevision=3
+/****************************************************************************/	
+typedef struct _ATOM_LCD_INFO_V13
+{
+  ATOM_COMMON_TABLE_HEADER sHeader;  
+  ATOM_DTD_FORMAT     sLCDTiming;
+  USHORT              usExtInfoTableOffset;
+  USHORT              usSupportedRefreshRate;     //Refer to panel info table in ATOMBIOS extension Spec.
+  ULONG               ulReserved0;
+  UCHAR               ucLCD_Misc;                // Reorganized in V13
+                                                 // Bit0: {=0:single, =1:dual},
+                                                 // Bit1: {=0:LDI format for RGB888, =1 FPDI format for RGB888}  // was {=0:666RGB, =1:888RGB},
+                                                 // Bit3:2: {Grey level}
+                                                 // Bit6:4 Color Bit Depth definition (see below definition in EDID V1.4 @BYTE 14h) 
+                                                 // Bit7   Reserved.  was for ATOM_PANEL_MISC_API_ENABLED, still need it?  
+  UCHAR               ucPanelDefaultRefreshRate;
+  UCHAR               ucPanelIdentification;
+  UCHAR               ucSS_Id;
+  USHORT              usLCDVenderID;
+  USHORT              usLCDProductID;
+  UCHAR               ucLCDPanel_SpecialHandlingCap;  // Reorganized in V13 
+                                                 // Bit0: Once DAL sees this CAP is set, it will read EDID from LCD on its own
+                                                 // Bit1: See LCDPANEL_CAP_DRR_SUPPORTED
+                                                 // Bit2: a quick reference whether an embadded panel (LCD1 ) is LVDS (0) or eDP (1)
+                                                 // Bit7-3: Reserved 
+  UCHAR               ucPanelInfoSize;					 //  start from ATOM_DTD_FORMAT to end of panel info, include ExtInfoTable
+  USHORT              usBacklightPWM;            //  Backlight PWM in Hz. New in _V13
+
+  UCHAR               ucPowerSequenceDIGONtoDE_in4Ms;
+  UCHAR               ucPowerSequenceDEtoVARY_BL_in4Ms;
+  UCHAR               ucPowerSequenceDEtoDIGON_in4Ms;
+  UCHAR               ucPowerSequenceVARY_BLtoDE_in4Ms;
+
+  UCHAR               ucOffDelay_in4Ms;
+  UCHAR               ucPowerSequenceVARY_BLtoBLON_in4Ms;
+  UCHAR               ucPowerSequenceBLONtoVARY_BL_in4Ms;
+  UCHAR               ucReserved1;
+
+  ULONG               ulReserved[4];
+}ATOM_LCD_INFO_V13;  
+
+#define ATOM_LCD_INFO_LAST  ATOM_LCD_INFO_V13    
+
+//Definitions for ucLCD_Misc
+#define ATOM_PANEL_MISC_V13_DUAL                   0x00000001
+#define ATOM_PANEL_MISC_V13_FPDI                   0x00000002
+#define ATOM_PANEL_MISC_V13_GREY_LEVEL             0x0000000C
+#define ATOM_PANEL_MISC_V13_GREY_LEVEL_SHIFT       2
+#define ATOM_PANEL_MISC_V13_COLOR_BIT_DEPTH_MASK   0x70
+#define ATOM_PANEL_MISC_V13_6BIT_PER_COLOR         0x10
+#define ATOM_PANEL_MISC_V13_8BIT_PER_COLOR         0x20
+
+//Color Bit Depth definition in EDID V1.4 @BYTE 14h
+//Bit 6  5  4
+                              //      0  0  0  -  Color bit depth is undefined
+                              //      0  0  1  -  6 Bits per Primary Color
+                              //      0  1  0  -  8 Bits per Primary Color
+                              //      0  1  1  - 10 Bits per Primary Color
+                              //      1  0  0  - 12 Bits per Primary Color
+                              //      1  0  1  - 14 Bits per Primary Color
+                              //      1  1  0  - 16 Bits per Primary Color
+                              //      1  1  1  - Reserved
+ 
+//Definitions for ucLCDPanel_SpecialHandlingCap:
+
+//Once DAL sees this CAP is set, it will read EDID from LCD on its own instead of using sLCDTiming in ATOM_LVDS_INFO_V12. 
+//Other entries in ATOM_LVDS_INFO_V12 are still valid/useful to DAL 
+#define	LCDPANEL_CAP_V13_READ_EDID              0x1        // = LCDPANEL_CAP_READ_EDID no change comparing to previous version
+
+//If a design supports DRR (dynamic refresh rate) on internal panels (LVDS or EDP), this cap is set in ucLCDPanel_SpecialHandlingCap together
+//with multiple supported refresh rates@usSupportedRefreshRate. This cap should not be set when only slow refresh rate is supported (static
+//refresh rate switch by SW. This is only valid from ATOM_LVDS_INFO_V12
+#define	LCDPANEL_CAP_V13_DRR_SUPPORTED          0x2        // = LCDPANEL_CAP_DRR_SUPPORTED no change comparing to previous version
 
-#define ATOM_LVDS_INFO_LAST  ATOM_LVDS_INFO_V12
+//Use this cap bit for a quick reference whether an embadded panel (LCD1 ) is LVDS or eDP.
+#define	LCDPANEL_CAP_V13_eDP                    0x4        // = LCDPANEL_CAP_eDP no change comparing to previous version
 
 typedef struct  _ATOM_PATCH_RECORD_MODE
 {
@@ -2944,9 +3417,9 @@ typedef struct _ATOM_DPCD_INFO
 #define MAX_DTD_MODE_IN_VRAM            6
 #define ATOM_DTD_MODE_SUPPORT_TBL_SIZE  (MAX_DTD_MODE_IN_VRAM*28)    //28= (SIZEOF ATOM_DTD_FORMAT) 
 #define ATOM_STD_MODE_SUPPORT_TBL_SIZE  32*8                         //32 is a predefined number,8= (SIZEOF ATOM_STD_FORMAT)
-#define DFP_ENCODER_TYPE_OFFSET					0x80
-#define DP_ENCODER_LANE_NUM_OFFSET			0x84
-#define DP_ENCODER_LINK_RATE_OFFSET			0x88
+//20 bytes for Encoder Type and DPCD in STD EDID area
+#define DFP_ENCODER_TYPE_OFFSET         (ATOM_EDID_RAW_DATASIZE + ATOM_DTD_MODE_SUPPORT_TBL_SIZE + ATOM_STD_MODE_SUPPORT_TBL_SIZE - 20)    
+#define ATOM_DP_DPCD_OFFSET             (DFP_ENCODER_TYPE_OFFSET + 4 )        
 
 #define ATOM_HWICON1_SURFACE_ADDR       0
 #define ATOM_HWICON2_SURFACE_ADDR       (ATOM_HWICON1_SURFACE_ADDR + ATOM_HWICON_SURFACE_SIZE)
@@ -2997,14 +3470,16 @@ typedef struct _ATOM_DPCD_INFO
 #define ATOM_DFP5_DTD_MODE_TBL_ADDR     (ATOM_DFP5_EDID_ADDR + ATOM_EDID_RAW_DATASIZE)
 #define ATOM_DFP5_STD_MODE_TBL_ADDR     (ATOM_DFP5_DTD_MODE_TBL_ADDR + ATOM_DTD_MODE_SUPPORT_TBL_SIZE)
 
-#define ATOM_DP_TRAINING_TBL_ADDR				(ATOM_DFP5_STD_MODE_TBL_ADDR+ATOM_STD_MODE_SUPPORT_TBL_SIZE)       
+#define ATOM_DP_TRAINING_TBL_ADDR       (ATOM_DFP5_STD_MODE_TBL_ADDR + ATOM_STD_MODE_SUPPORT_TBL_SIZE)
 
-#define ATOM_STACK_STORAGE_START        (ATOM_DP_TRAINING_TBL_ADDR+256)       
-#define ATOM_STACK_STORAGE_END          ATOM_STACK_STORAGE_START+512        
+#define ATOM_STACK_STORAGE_START        (ATOM_DP_TRAINING_TBL_ADDR + 1024)       
+#define ATOM_STACK_STORAGE_END          ATOM_STACK_STORAGE_START + 512        
 
 //The size below is in Kb!
 #define ATOM_VRAM_RESERVE_SIZE         ((((ATOM_STACK_STORAGE_END - ATOM_HWICON1_SURFACE_ADDR)>>10)+4)&0xFFFC)
    
+#define ATOM_VRAM_RESERVE_V2_SIZE      32
+
 #define	ATOM_VRAM_OPERATION_FLAGS_MASK         0xC0000000L
 #define ATOM_VRAM_OPERATION_FLAGS_SHIFT        30
 #define	ATOM_VRAM_BLOCK_NEEDS_NO_RESERVATION   0x1
@@ -3206,6 +3681,15 @@ typedef struct  _ATOM_DISPLAY_OBJECT_PATH
   USHORT    usGraphicObjIds[1];                             //1st Encoder Obj source from GPU to last Graphic Obj destinate to connector.
 }ATOM_DISPLAY_OBJECT_PATH;
 
+typedef struct  _ATOM_DISPLAY_EXTERNAL_OBJECT_PATH
+{
+  USHORT    usDeviceTag;                                   //supported device 
+  USHORT    usSize;                                        //the size of ATOM_DISPLAY_OBJECT_PATH
+  USHORT    usConnObjectId;                                //Connector Object ID 
+  USHORT    usGPUObjectId;                                 //GPU ID 
+  USHORT    usGraphicObjIds[2];                            //usGraphicObjIds[0]= GPU internal encoder, usGraphicObjIds[1]= external encoder 
+}ATOM_DISPLAY_EXTERNAL_OBJECT_PATH;
+
 typedef struct _ATOM_DISPLAY_OBJECT_PATH_TABLE
 {
   UCHAR                           ucNumOfDispPath;
@@ -3261,6 +3745,47 @@ typedef struct _ATOM_SRC_DST_TABLE_FOR_ONE_OBJECT         //usSrcDstTableOffset
 #define EXT_AUXDDC_LUTINDEX_7                   7
 #define MAX_NUMBER_OF_EXT_AUXDDC_LUT_ENTRIES   (EXT_AUXDDC_LUTINDEX_7+1)
 
+//ucChannelMapping are defined as following
+//for DP connector, eDP, DP to VGA/LVDS 
+//Bit[1:0]: Define which pin connect to DP connector DP_Lane0, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+//Bit[3:2]: Define which pin connect to DP connector DP_Lane1, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+//Bit[5:4]: Define which pin connect to DP connector DP_Lane2, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+//Bit[7:6]: Define which pin connect to DP connector DP_Lane3, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+typedef struct _ATOM_DP_CONN_CHANNEL_MAPPING
+{
+#if ATOM_BIG_ENDIAN
+  UCHAR ucDP_Lane3_Source:2;
+  UCHAR ucDP_Lane2_Source:2;
+  UCHAR ucDP_Lane1_Source:2;
+  UCHAR ucDP_Lane0_Source:2;
+#else
+  UCHAR ucDP_Lane0_Source:2;
+  UCHAR ucDP_Lane1_Source:2;
+  UCHAR ucDP_Lane2_Source:2;
+  UCHAR ucDP_Lane3_Source:2;
+#endif
+}ATOM_DP_CONN_CHANNEL_MAPPING;
+
+//for DVI/HDMI, in dual link case, both links have to have same mapping. 
+//Bit[1:0]: Define which pin connect to DVI connector data Lane2, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+//Bit[3:2]: Define which pin connect to DVI connector data Lane1, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+//Bit[5:4]: Define which pin connect to DVI connector data Lane0, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+//Bit[7:6]: Define which pin connect to DVI connector clock lane, =0: source from GPU pin TX0, =1: from GPU pin TX1, =2: from GPU pin TX2, =3 from GPU pin TX3
+typedef struct _ATOM_DVI_CONN_CHANNEL_MAPPING
+{
+#if ATOM_BIG_ENDIAN
+  UCHAR ucDVI_CLK_Source:2;
+  UCHAR ucDVI_DATA0_Source:2;
+  UCHAR ucDVI_DATA1_Source:2;
+  UCHAR ucDVI_DATA2_Source:2;
+#else
+  UCHAR ucDVI_DATA2_Source:2;
+  UCHAR ucDVI_DATA1_Source:2;
+  UCHAR ucDVI_DATA0_Source:2;
+  UCHAR ucDVI_CLK_Source:2;
+#endif
+}ATOM_DVI_CONN_CHANNEL_MAPPING;
+
 typedef struct _EXT_DISPLAY_PATH
 {
   USHORT  usDeviceTag;                    //A bit vector to show what devices are supported 
@@ -3269,7 +3794,13 @@ typedef struct _EXT_DISPLAY_PATH
   UCHAR   ucExtAUXDDCLutIndex;            //An index into external AUX/DDC channel LUT
   UCHAR   ucExtHPDPINLutIndex;            //An index into external HPD pin LUT
   USHORT  usExtEncoderObjId;              //external encoder object id
-  USHORT  usReserved[3]; 
+  union{
+    UCHAR   ucChannelMapping;                  // if ucChannelMapping=0, using default one to one mapping
+    ATOM_DP_CONN_CHANNEL_MAPPING asDPMapping;
+    ATOM_DVI_CONN_CHANNEL_MAPPING asDVIMapping;
+  };
+  UCHAR   ucReserved;
+  USHORT  usReserved[2]; 
 }EXT_DISPLAY_PATH;
    
 #define NUMBER_OF_UCHAR_FOR_GUID          16
@@ -3281,7 +3812,8 @@ typedef  struct _ATOM_EXTERNAL_DISPLAY_CONNECTION_INFO
   UCHAR                    ucGuid [NUMBER_OF_UCHAR_FOR_GUID];     // a GUID is a 16 byte long string
   EXT_DISPLAY_PATH         sPath[MAX_NUMBER_OF_EXT_DISPLAY_PATH]; // total of fixed 7 entries.
   UCHAR                    ucChecksum;                            // a  simple Checksum of the sum of whole structure equal to 0x0. 
-  UCHAR                    Reserved [7];                          // for potential expansion
+  UCHAR                    uc3DStereoPinId;                       // use for eDP panel
+  UCHAR                    Reserved [6];                          // for potential expansion
 }ATOM_EXTERNAL_DISPLAY_CONNECTION_INFO;
 
 //Related definitions, all records are differnt but they have a commond header
@@ -3311,10 +3843,11 @@ typedef struct _ATOM_COMMON_RECORD_HEADER
 #define ATOM_CONNECTOR_AUXDDC_LUT_RECORD_TYPE          17 //This is for the case when connectors are not known to object table
 #define ATOM_OBJECT_LINK_RECORD_TYPE                   18 //Once this record is present under one object, it indicats the oobject is linked to another obj described by the record
 #define ATOM_CONNECTOR_REMOTE_CAP_RECORD_TYPE          19
+#define ATOM_ENCODER_CAP_RECORD_TYPE                   20
 
 
 //Must be updated when new record type is added,equal to that record definition!
-#define ATOM_MAX_OBJECT_RECORD_NUMBER             ATOM_CONNECTOR_REMOTE_CAP_RECORD_TYPE
+#define ATOM_MAX_OBJECT_RECORD_NUMBER             ATOM_ENCODER_CAP_RECORD_TYPE
 
 typedef struct  _ATOM_I2C_RECORD
 {
@@ -3441,6 +3974,26 @@ typedef struct  _ATOM_ENCODER_DVO_CF_RECORD
   UCHAR                       ucPadding[2];
 }ATOM_ENCODER_DVO_CF_RECORD;
 
+// Bit maps for ATOM_ENCODER_CAP_RECORD.ucEncoderCap
+#define ATOM_ENCODER_CAP_RECORD_HBR2     0x01         // DP1.2 HBR2 is supported by this path
+
+typedef struct  _ATOM_ENCODER_CAP_RECORD
+{
+  ATOM_COMMON_RECORD_HEADER   sheader;
+  union {
+    USHORT                    usEncoderCap;         
+    struct {
+#if ATOM_BIG_ENDIAN
+      USHORT                  usReserved:15;        // Bit1-15 may be defined for other capability in future
+      USHORT                  usHBR2Cap:1;          // Bit0 is for DP1.2 HBR2 capability. 
+#else
+      USHORT                  usHBR2Cap:1;          // Bit0 is for DP1.2 HBR2 capability. 
+      USHORT                  usReserved:15;        // Bit1-15 may be defined for other capability in future
+#endif
+    };
+  }; 
+}ATOM_ENCODER_CAP_RECORD;                             
+
 // value for ATOM_CONNECTOR_CF_RECORD.ucConnectedDvoBundle
 #define ATOM_CONNECTOR_CF_RECORD_CONNECTED_UPPER12BITBUNDLEA   1
 #define ATOM_CONNECTOR_CF_RECORD_CONNECTED_LOWER12BITBUNDLEB   2
@@ -3580,6 +4133,11 @@ typedef struct _ATOM_VOLTAGE_CONTROL
 #define	VOLTAGE_CONTROL_ID_DAC								0x02									//I2C control, used for R5xx/R6xx MVDDC,MVDDQ or VDDCI
 #define	VOLTAGE_CONTROL_ID_VT116xM						0x03									//I2C control, used for R6xx Core Voltage
 #define VOLTAGE_CONTROL_ID_DS4402							0x04									
+#define VOLTAGE_CONTROL_ID_UP6266 						0x05									
+#define VOLTAGE_CONTROL_ID_SCORPIO						0x06
+#define	VOLTAGE_CONTROL_ID_VT1556M						0x07									
+#define	VOLTAGE_CONTROL_ID_CHL822x						0x08									
+#define	VOLTAGE_CONTROL_ID_VT1586M						0x09
 
 typedef struct  _ATOM_VOLTAGE_OBJECT
 {
@@ -3670,66 +4228,157 @@ typedef struct _ATOM_POWER_SOURCE_INFO
 #define POWER_SENSOR_GPIO								0x01
 #define POWER_SENSOR_I2C								0x02
 
+typedef struct _ATOM_CLK_VOLT_CAPABILITY
+{
+  ULONG      ulVoltageIndex;                      // The Voltage Index indicated by FUSE, same voltage index shared with SCLK DPM fuse table        
+  ULONG      ulMaximumSupportedCLK;               // Maximum clock supported with specified voltage index, unit in 10kHz
+}ATOM_CLK_VOLT_CAPABILITY;
+
+typedef struct _ATOM_AVAILABLE_SCLK_LIST
+{
+  ULONG      ulSupportedSCLK;               // Maximum clock supported with specified voltage index,  unit in 10kHz
+  USHORT     usVoltageIndex;                // The Voltage Index indicated by FUSE for specified SCLK  
+  USHORT     usVoltageID;                   // The Voltage ID indicated by FUSE for specified SCLK 
+}ATOM_AVAILABLE_SCLK_LIST;
+
+// ATOM_INTEGRATED_SYSTEM_INFO_V6 ulSystemConfig cap definition
+#define ATOM_IGP_INFO_V6_SYSTEM_CONFIG__PCIE_POWER_GATING_ENABLE             1       // refer to ulSystemConfig bit[0]
+
+// this IntegrateSystemInfoTable is used for Liano/Ontario APU
 typedef struct _ATOM_INTEGRATED_SYSTEM_INFO_V6
 {
   ATOM_COMMON_TABLE_HEADER   sHeader;
   ULONG  ulBootUpEngineClock;
   ULONG  ulDentistVCOFreq;          
   ULONG  ulBootUpUMAClock;          
-  ULONG  ulReserved1[8];            
+  ATOM_CLK_VOLT_CAPABILITY   sDISPCLK_Voltage[4];            
   ULONG  ulBootUpReqDisplayVector;
   ULONG  ulOtherDisplayMisc;
   ULONG  ulGPUCapInfo;
-  ULONG  ulReserved2[3];            
+  ULONG  ulSB_MMIO_Base_Addr;
+  USHORT usRequestedPWMFreqInHz;
+  UCHAR  ucHtcTmpLmt;   
+  UCHAR  ucHtcHystLmt;
+  ULONG  ulMinEngineClock;           
   ULONG  ulSystemConfig;            
   ULONG  ulCPUCapInfo;              
-  USHORT usMaxNBVoltage;  
-  USHORT usMinNBVoltage;  
-  USHORT usBootUpNBVoltage;         
-  USHORT usExtDispConnInfoOffset;  
-  UCHAR  ucHtcTmpLmt;   
-  UCHAR  ucTjOffset;    
+  USHORT usNBP0Voltage;               
+  USHORT usNBP1Voltage;
+  USHORT usBootUpNBVoltage;                       
+  USHORT usExtDispConnInfoOffset;
+  USHORT usPanelRefreshRateRange;     
   UCHAR  ucMemoryType;  
   UCHAR  ucUMAChannelNumber;
   ULONG  ulCSR_M3_ARB_CNTL_DEFAULT[10];  
   ULONG  ulCSR_M3_ARB_CNTL_UVD[10]; 
   ULONG  ulCSR_M3_ARB_CNTL_FS3D[10];
-  ULONG  ulReserved3[42]; 
+  ATOM_AVAILABLE_SCLK_LIST   sAvail_SCLK[5];
+  ULONG  ulGMCRestoreResetTime;
+  ULONG  ulMinimumNClk;
+  ULONG  ulIdleNClk;
+  ULONG  ulDDR_DLL_PowerUpTime;
+  ULONG  ulDDR_PLL_PowerUpTime;
+  USHORT usPCIEClkSSPercentage;
+  USHORT usPCIEClkSSType;
+  USHORT usLvdsSSPercentage;
+  USHORT usLvdsSSpreadRateIn10Hz;
+  USHORT usHDMISSPercentage;
+  USHORT usHDMISSpreadRateIn10Hz;
+  USHORT usDVISSPercentage;
+  USHORT usDVISSpreadRateIn10Hz;
+  ULONG  ulReserved3[21]; 
   ATOM_EXTERNAL_DISPLAY_CONNECTION_INFO sExtDispConnInfo;   
 }ATOM_INTEGRATED_SYSTEM_INFO_V6;   
 
+// ulGPUCapInfo
+#define INTEGRATED_SYSTEM_INFO_V6_GPUCAPINFO__TMDSHDMI_COHERENT_SINGLEPLL_MODE       0x01
+#define INTEGRATED_SYSTEM_INFO_V6_GPUCAPINFO__DISABLE_AUX_HW_MODE_DETECTION          0x08
+
+// ulOtherDisplayMisc
+#define INTEGRATED_SYSTEM_INFO__GET_EDID_CALLBACK_FUNC_SUPPORT                       0x01
+
+
 /**********************************************************************************************************************
-// ATOM_INTEGRATED_SYSTEM_INFO_V6 Description
-//ulBootUpEngineClock:              VBIOS bootup Engine clock frequency, in 10kHz unit. 
-//ulDentistVCOFreq:                 Dentist VCO clock in 10kHz unit. 
-//ulBootUpUMAClock:                 System memory boot up clock frequency in 10Khz unit. 
-//ulReserved1[8]                    Reserved by now, must be 0x0. 
-//ulBootUpReqDisplayVector	        VBIOS boot up display IDs
-//                                  ATOM_DEVICE_CRT1_SUPPORT                  0x0001
-//                                  ATOM_DEVICE_CRT2_SUPPORT                  0x0010
-//                                  ATOM_DEVICE_DFP1_SUPPORT                  0x0008 
-//                                  ATOM_DEVICE_DFP6_SUPPORT                  0x0040 
-//                                  ATOM_DEVICE_DFP2_SUPPORT                  0x0080       
-//                                  ATOM_DEVICE_DFP3_SUPPORT                  0x0200       
-//                                  ATOM_DEVICE_DFP4_SUPPORT                  0x0400        
-//                                  ATOM_DEVICE_DFP5_SUPPORT                  0x0800
-//                                  ATOM_DEVICE_LCD1_SUPPORT                  0x0002
-//ulOtherDisplayMisc      	        Other display related flags, not defined yet. 
-//ulGPUCapInfo                      TBD
-//ulReserved2[3]                    must be 0x0 for the reserved.
-//ulSystemConfig                    TBD
-//ulCPUCapInfo                      TBD
-//usMaxNBVoltage                    High NB voltage in unit of mv, calculated using current VDDNB (D24F2xDC) and VDDNB offset fuse. 
-//usMinNBVoltage                    Low NB voltage in unit of mv, calculated using current VDDNB (D24F2xDC) and VDDNB offset fuse.
-//usBootUpNBVoltage                 Boot up NB voltage in unit of mv.
-//ucHtcTmpLmt                       Bit [22:16] of D24F3x64 Thermal Control (HTC) Register.
-//ucTjOffset                        Bit [28:22] of D24F3xE4 Thermtrip Status Register,may not be needed.
-//ucMemoryType                      [3:0]=1:DDR1;=2:DDR2;=3:DDR3.[7:4] is reserved.
-//ucUMAChannelNumber      	        System memory channel numbers. 
-//usExtDispConnectionInfoOffset     ATOM_EXTERNAL_DISPLAY_CONNECTION_INFO offset relative to beginning of this table. 
-//ulCSR_M3_ARB_CNTL_DEFAULT[10]     Arrays with values for CSR M3 arbiter for default
-//ulCSR_M3_ARB_CNTL_UVD[10]         Arrays with values for CSR M3 arbiter for UVD playback.
-//ulCSR_M3_ARB_CNTL_FS3D[10]        Arrays with values for CSR M3 arbiter for Full Screen 3D applications.
+  ATOM_INTEGRATED_SYSTEM_INFO_V6 Description
+ulBootUpEngineClock:              VBIOS bootup Engine clock frequency, in 10kHz unit. if it is equal 0, then VBIOS use pre-defined bootup engine clock
+ulDentistVCOFreq:                 Dentist VCO clock in 10kHz unit. 
+ulBootUpUMAClock:                 System memory boot up clock frequency in 10Khz unit. 
+sDISPCLK_Voltage:                 Report Display clock voltage requirement.
+ 
+ulBootUpReqDisplayVector:         VBIOS boot up display IDs, following are supported devices in Liano/Ontaio projects:
+                                  ATOM_DEVICE_CRT1_SUPPORT                  0x0001
+                                  ATOM_DEVICE_CRT2_SUPPORT                  0x0010
+                                  ATOM_DEVICE_DFP1_SUPPORT                  0x0008 
+                                  ATOM_DEVICE_DFP6_SUPPORT                  0x0040 
+                                  ATOM_DEVICE_DFP2_SUPPORT                  0x0080       
+                                  ATOM_DEVICE_DFP3_SUPPORT                  0x0200       
+                                  ATOM_DEVICE_DFP4_SUPPORT                  0x0400        
+                                  ATOM_DEVICE_DFP5_SUPPORT                  0x0800
+                                  ATOM_DEVICE_LCD1_SUPPORT                  0x0002
+ulOtherDisplayMisc:      	        Other display related flags, not defined yet. 
+ulGPUCapInfo:                     bit[0]=0: TMDS/HDMI Coherent Mode use cascade PLL mode.
+                                        =1: TMDS/HDMI Coherent Mode use signel PLL mode.
+                                  bit[3]=0: Enable HW AUX mode detection logic
+                                        =1: Disable HW AUX mode dettion logic
+ulSB_MMIO_Base_Addr:              Physical Base address to SB MMIO space. Driver needs to initialize it for SMU usage.
+
+usRequestedPWMFreqInHz:           When it's set to 0x0 by SBIOS: the LCD BackLight is not controlled by GPU(SW). 
+                                  Any attempt to change BL using VBIOS function or enable VariBri from PP table is not effective since ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU==0;
+                                  
+                                  When it's set to a non-zero frequency, the BackLight is controlled by GPU (SW) in one of two ways below:
+                                  1. SW uses the GPU BL PWM output to control the BL, in chis case, this non-zero frequency determines what freq GPU should use;
+                                  VBIOS will set up proper PWM frequency and ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU==1,as the result,
+                                  Changing BL using VBIOS function is functional in both driver and non-driver present environment; 
+                                  and enabling VariBri under the driver environment from PP table is optional.
+
+                                  2. SW uses other means to control BL (like DPCD),this non-zero frequency serves as a flag only indicating
+                                  that BL control from GPU is expected.
+                                  VBIOS will NOT set up PWM frequency but make ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU==1
+                                  Changing BL using VBIOS function could be functional in both driver and non-driver present environment,but
+                                  it's per platform 
+                                  and enabling VariBri under the driver environment from PP table is optional.
+
+ucHtcTmpLmt:                      Refer to D18F3x64 bit[22:16], HtcTmpLmt. 
+                                  Threshold on value to enter HTC_active state.
+ucHtcHystLmt:                     Refer to D18F3x64 bit[27:24], HtcHystLmt. 
+                                  To calculate threshold off value to exit HTC_active state, which is Threshold on vlaue minus ucHtcHystLmt.
+ulMinEngineClock:                 Minimum SCLK allowed in 10kHz unit. This is calculated based on WRCK Fuse settings.
+ulSystemConfig:                   Bit[0]=0: PCIE Power Gating Disabled 
+                                        =1: PCIE Power Gating Enabled
+                                  Bit[1]=0: DDR-DLL shut-down feature disabled.
+                                         1: DDR-DLL shut-down feature enabled.
+                                  Bit[2]=0: DDR-PLL Power down feature disabled.
+                                         1: DDR-PLL Power down feature enabled.                                 
+ulCPUCapInfo:                     TBD
+usNBP0Voltage:                    VID for voltage on NB P0 State
+usNBP1Voltage:                    VID for voltage on NB P1 State  
+usBootUpNBVoltage:                Voltage Index of GNB voltage configured by SBIOS, which is suffcient to support VBIOS DISPCLK requirement.
+usExtDispConnInfoOffset:          Offset to sExtDispConnInfo inside the structure
+usPanelRefreshRateRange:          Bit vector for LCD supported refresh rate range. If DRR is requestd by the platform, at least two bits need to be set
+                                  to indicate a range.
+                                  SUPPORTED_LCD_REFRESHRATE_30Hz          0x0004
+                                  SUPPORTED_LCD_REFRESHRATE_40Hz          0x0008
+                                  SUPPORTED_LCD_REFRESHRATE_50Hz          0x0010
+                                  SUPPORTED_LCD_REFRESHRATE_60Hz          0x0020
+ucMemoryType:                     [3:0]=1:DDR1;=2:DDR2;=3:DDR3.[7:4] is reserved.
+ucUMAChannelNumber:      	        System memory channel numbers. 
+ulCSR_M3_ARB_CNTL_DEFAULT[10]:    Arrays with values for CSR M3 arbiter for default
+ulCSR_M3_ARB_CNTL_UVD[10]:        Arrays with values for CSR M3 arbiter for UVD playback.
+ulCSR_M3_ARB_CNTL_FS3D[10]:       Arrays with values for CSR M3 arbiter for Full Screen 3D applications.
+sAvail_SCLK[5]:                   Arrays to provide availabe list of SLCK and corresponding voltage, order from low to high  
+ulGMCRestoreResetTime:            GMC power restore and GMC reset time to calculate data reconnection latency. Unit in ns. 
+ulMinimumNClk:                    Minimum NCLK speed among all NB-Pstates to calcualte data reconnection latency. Unit in 10kHz. 
+ulIdleNClk:                       NCLK speed while memory runs in self-refresh state. Unit in 10kHz.
+ulDDR_DLL_PowerUpTime:            DDR PHY DLL power up time. Unit in ns.
+ulDDR_PLL_PowerUpTime:            DDR PHY PLL power up time. Unit in ns.
+usPCIEClkSSPercentage:            PCIE Clock Spred Spectrum Percentage in unit 0.01%; 100 mean 1%.
+usPCIEClkSSType:                  PCIE Clock Spred Spectrum Type. 0 for Down spread(default); 1 for Center spread.
+usLvdsSSPercentage:               LVDS panel ( not include eDP ) Spread Spectrum Percentage in unit of 0.01%, =0, use VBIOS default setting. 
+usLvdsSSpreadRateIn10Hz:          LVDS panel ( not include eDP ) Spread Spectrum frequency in unit of 10Hz, =0, use VBIOS default setting. 
+usHDMISSPercentage:               HDMI Spread Spectrum Percentage in unit 0.01%; 100 mean 1%,  =0, use VBIOS default setting. 
+usHDMISSpreadRateIn10Hz:          HDMI Spread Spectrum frequency in unit of 10Hz,  =0, use VBIOS default setting. 
+usDVISSPercentage:                DVI Spread Spectrum Percentage in unit 0.01%; 100 mean 1%,  =0, use VBIOS default setting. 
+usDVISSpreadRateIn10Hz:           DVI Spread Spectrum frequency in unit of 10Hz,  =0, use VBIOS default setting. 
 **********************************************************************************************************************/
 
 /**************************************************************************/
@@ -3790,6 +4439,7 @@ typedef struct _ATOM_ASIC_SS_ASSIGNMENT
 #define ASIC_INTERNAL_SS_ON_LVDS    6
 #define ASIC_INTERNAL_SS_ON_DP      7
 #define ASIC_INTERNAL_SS_ON_DCPLL   8
+#define ASIC_EXTERNAL_SS_ON_DP_CLOCK 9
 
 typedef struct _ATOM_ASIC_SS_ASSIGNMENT_V2
 {
@@ -3903,6 +4553,7 @@ typedef struct _ATOM_ASIC_INTERNAL_SS_INFO_V3
 #define ATOM_S0_SYSTEM_POWER_STATE_VALUE_AC     1
 #define ATOM_S0_SYSTEM_POWER_STATE_VALUE_DC     2
 #define ATOM_S0_SYSTEM_POWER_STATE_VALUE_LITEAC 3
+#define ATOM_S0_SYSTEM_POWER_STATE_VALUE_LIT2AC 4
 
 //Byte aligned defintion for BIOS usage
 #define ATOM_S0_CRT1_MONOb0             0x01
@@ -4529,7 +5180,8 @@ typedef struct _ATOM_INIT_REG_BLOCK{
 #define INDEX_ACCESS_RANGE_BEGIN	    (VALUE_DWORD + 1)
 #define INDEX_ACCESS_RANGE_END		    (INDEX_ACCESS_RANGE_BEGIN + 1)
 #define VALUE_INDEX_ACCESS_SINGLE	    (INDEX_ACCESS_RANGE_END + 1)
-
+//#define ACCESS_MCIODEBUGIND            0x40       //defined in BIOS code
+#define ACCESS_PLACEHOLDER             0x80
 
 typedef struct _ATOM_MC_INIT_PARAM_TABLE
 { 
@@ -4554,6 +5206,10 @@ typedef struct _ATOM_MC_INIT_PARAM_TABLE
 #define _32Mx32             0x33
 #define _64Mx8              0x41
 #define _64Mx16             0x42
+#define _64Mx32             0x43
+#define _128Mx8             0x51
+#define _128Mx16            0x52
+#define _256Mx8             0x61
 
 #define SAMSUNG             0x1
 #define INFINEON            0x2
@@ -4569,10 +5225,11 @@ typedef struct _ATOM_MC_INIT_PARAM_TABLE
 #define QIMONDA             INFINEON
 #define PROMOS              MOSEL
 #define KRETON              INFINEON
+#define ELIXIR              NANYA
 
 /////////////Support for GDDR5 MC uCode to reside in upper 64K of ROM/////////////
 
-#define UCODE_ROM_START_ADDRESS		0x1c000
+#define UCODE_ROM_START_ADDRESS		0x1b800
 #define	UCODE_SIGNATURE			0x4375434d // 'MCuC' - MC uCode
 
 //uCode block header for reference
@@ -4903,7 +5560,34 @@ typedef struct _ATOM_VRAM_MODULE_V6
   ATOM_MEMORY_TIMING_FORMAT_V2  asMemTiming[5];//Memory Timing block sort from lower clock to higher clock
 }ATOM_VRAM_MODULE_V6;
 
-
+typedef struct _ATOM_VRAM_MODULE_V7
+{
+// Design Specific Values
+  ULONG	  ulChannelMapCfg;	                // mmMC_SHARED_CHREMAP
+  USHORT  usModuleSize;                     // Size of ATOM_VRAM_MODULE_V7
+  USHORT  usPrivateReserved;                // MC_ARB_RAMCFG (includes NOOFBANK,NOOFRANKS,NOOFROWS,NOOFCOLS)
+  USHORT  usReserved;
+  UCHAR   ucExtMemoryID;                    // Current memory module ID
+  UCHAR   ucMemoryType;                     // MEM_TYPE_DDR2/DDR3/GDDR3/GDDR5
+  UCHAR   ucChannelNum;                     // Number of mem. channels supported in this module
+  UCHAR   ucChannelWidth;                   // CHANNEL_16BIT/CHANNEL_32BIT/CHANNEL_64BIT
+  UCHAR   ucDensity;                        // _8Mx32, _16Mx32, _16Mx16, _32Mx16
+  UCHAR	  ucReserve;                        // Former container for Mx_FLAGS like DBI_AC_MODE_ENABLE_ASIC for GDDR4. Not used now.
+  UCHAR	  ucMisc;                           // RANK_OF_THISMEMORY etc.
+  UCHAR	  ucVREFI;                          // Not used.
+  UCHAR   ucNPL_RT;                         // Round trip delay (MC_SEQ_CAS_TIMING [28:24]:TCL=CL+NPL_RT-2). Always 2.
+  UCHAR	  ucPreamble;                       // [7:4] Write Preamble, [3:0] Read Preamble
+  UCHAR   ucMemorySize;                     // Total memory size in unit of 16MB for CONFIG_MEMSIZE - bit[23:0] zeros
+  UCHAR   ucReserved[3];
+// Memory Module specific values
+  USHORT  usEMRS2Value;                     // EMRS2/MR2 Value. 
+  USHORT  usEMRS3Value;                     // EMRS3/MR3 Value.
+  UCHAR   ucMemoryVenderID;                 // [7:4] Revision, [3:0] Vendor code
+  UCHAR	  ucRefreshRateFactor;              // [1:0]=RefreshFactor (00=8ms, 01=16ms, 10=32ms,11=64ms)
+  UCHAR	  ucFIFODepth;                      // FIFO depth can be detected during vendor detection, here is hardcoded per memory
+  UCHAR   ucCDR_Bandwidth;                  // [0:3]=Read CDR bandwidth, [4:7] - Write CDR Bandwidth
+  char    strMemPNString[20];               // part number end with '0'. 
+}ATOM_VRAM_MODULE_V7;
 
 typedef struct _ATOM_VRAM_INFO_V2
 {
@@ -4942,6 +5626,20 @@ typedef struct _ATOM_VRAM_INFO_V4
 																																						 //	ATOM_INIT_REG_BLOCK				 aMemAdjust;
 }ATOM_VRAM_INFO_V4;
 
+typedef struct _ATOM_VRAM_INFO_HEADER_V2_1
+{
+  ATOM_COMMON_TABLE_HEADER   sHeader;
+	USHORT										 usMemAdjustTblOffset;													 // offset of ATOM_INIT_REG_BLOCK structure for memory vendor specific MC adjust setting
+	USHORT										 usMemClkPatchTblOffset;												 //	offset of ATOM_INIT_REG_BLOCK structure for memory clock specific MC setting
+	USHORT										 usReserved[4];
+  UCHAR                      ucNumOfVRAMModule;                              // indicate number of VRAM module
+  UCHAR                      ucMemoryClkPatchTblVer;                         // version of memory AC timing register list
+  UCHAR                      ucVramModuleVer;                                // indicate ATOM_VRAM_MODUE version
+  UCHAR                      ucReserved; 
+  ATOM_VRAM_MODULE_V7		     aVramInfo[ATOM_MAX_NUMBER_OF_VRAM_MODULE];      // just for allocation, real number of blocks is in ucNumOfVRAMModule;
+}ATOM_VRAM_INFO_HEADER_V2_1;
+
+
 typedef struct _ATOM_VRAM_GPIO_DETECTION_INFO
 {
   ATOM_COMMON_TABLE_HEADER   sHeader;
@@ -5182,6 +5880,16 @@ typedef struct _ASIC_TRANSMITTER_INFO
 	UCHAR  ucReserved;
 }ASIC_TRANSMITTER_INFO;
 
+#define ASIC_TRANSMITTER_INFO_CONFIG__DVO_SDR_MODE          0x01
+#define ASIC_TRANSMITTER_INFO_CONFIG__COHERENT_MODE         0x02
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODEROBJ_ID_MASK    0xc4
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODER_A             0x00
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODER_B             0x04
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODER_C             0x40
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODER_D             0x44
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODER_E             0x80
+#define ASIC_TRANSMITTER_INFO_CONFIG__ENCODER_F             0x84
+
 typedef struct _ASIC_ENCODER_INFO
 {
 	UCHAR ucEncoderID;
@@ -5284,6 +5992,28 @@ typedef struct _DP_ENCODER_SERVICE_PARAMETERS
 /* /obselete */
 #define DP_ENCODER_SERVICE_PS_ALLOCATION				WRITE_ONE_BYTE_HW_I2C_DATA_PARAMETERS
 
+
+typedef struct _DP_ENCODER_SERVICE_PARAMETERS_V2
+{
+	USHORT usExtEncoderObjId;   // External Encoder Object Id, output parameter only, use when ucAction = DP_SERVICE_V2_ACTION_DET_EXT_CONNECTION
+  UCHAR  ucAuxId;
+  UCHAR  ucAction;
+  UCHAR  ucSinkType;          // Iput and Output parameters. 
+  UCHAR  ucHPDId;             // Input parameter, used when ucAction = DP_SERVICE_V2_ACTION_DET_EXT_CONNECTION
+	UCHAR  ucReserved[2];
+}DP_ENCODER_SERVICE_PARAMETERS_V2;
+
+typedef struct _DP_ENCODER_SERVICE_PS_ALLOCATION_V2
+{
+  DP_ENCODER_SERVICE_PARAMETERS_V2 asDPServiceParam;
+  PROCESS_AUX_CHANNEL_TRANSACTION_PARAMETERS_V2 asAuxParam;
+}DP_ENCODER_SERVICE_PS_ALLOCATION_V2;
+
+// ucAction
+#define DP_SERVICE_V2_ACTION_GET_SINK_TYPE							0x01
+#define DP_SERVICE_V2_ACTION_DET_LCD_CONNECTION			    0x02
+
+
 // DP_TRAINING_TABLE
 #define DPCD_SET_LINKRATE_LANENUM_PATTERN1_TBL_ADDR				ATOM_DP_TRAINING_TBL_ADDR		
 #define DPCD_SET_SS_CNTL_TBL_ADDR													(ATOM_DP_TRAINING_TBL_ADDR + 8 )
@@ -5339,6 +6069,7 @@ typedef struct _SET_HWBLOCK_INSTANCE_PARAMETER_V2
 #define SELECT_DCIO_IMPCAL            4
 #define SELECT_DCIO_DIG               6
 #define SELECT_CRTC_PIXEL_RATE        7
+#define SELECT_VGA_BLK                8
 
 /****************************************************************************/	
 //Portion VI: Definitinos for vbios MC scratch registers that driver used
@@ -5744,7 +6475,17 @@ typedef struct _ATOM_PPLIB_THERMALCONTROLLER
 #define ATOM_PP_THERMALCONTROLLER_ADT7473   9
 #define ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO     11
 #define ATOM_PP_THERMALCONTROLLER_EVERGREEN 12
+#define ATOM_PP_THERMALCONTROLLER_EMC2103   13  /* 0x0D */ // Only fan control will be implemented, do NOT show this in PPGen.
+#define ATOM_PP_THERMALCONTROLLER_SUMO      14  /* 0x0E */ // Sumo type, used internally
+#define ATOM_PP_THERMALCONTROLLER_NISLANDS  15
+
+// Thermal controller 'combo type' to use an external controller for Fan control and an internal controller for thermal.
+// We probably should reserve the bit 0x80 for this use.
+// To keep the number of these types low we should also use the same code for all ASICs (i.e. do not distinguish RV6xx and RV7xx Internal here).
+// The driver can pick the correct internal controller based on the ASIC.
+
 #define ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL   0x89    // ADT7473 Fan Control + Internal Thermal Controller
+#define ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL   0x8D    // EMC2103 Fan Control + Internal Thermal Controller
 
 typedef struct _ATOM_PPLIB_STATE
 {
@@ -5841,6 +6582,29 @@ typedef struct _ATOM_PPLIB_POWERPLAYTABLE3
     USHORT                     usExtendendedHeaderOffset;
 } ATOM_PPLIB_POWERPLAYTABLE3, *LPATOM_PPLIB_POWERPLAYTABLE3;
 
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE4
+{
+    ATOM_PPLIB_POWERPLAYTABLE3 basicTable3;
+    ULONG                      ulGoldenPPID;                    // PPGen use only     
+    ULONG                      ulGoldenRevision;                // PPGen use only
+    USHORT                     usVddcDependencyOnSCLKOffset;
+    USHORT                     usVddciDependencyOnMCLKOffset;
+    USHORT                     usVddcDependencyOnMCLKOffset;
+    USHORT                     usMaxClockVoltageOnDCOffset;
+    USHORT                     usReserved[2];  
+} ATOM_PPLIB_POWERPLAYTABLE4, *LPATOM_PPLIB_POWERPLAYTABLE4;
+
+typedef struct _ATOM_PPLIB_POWERPLAYTABLE5
+{
+    ATOM_PPLIB_POWERPLAYTABLE4 basicTable4;
+    ULONG                      ulTDPLimit;
+    ULONG                      ulNearTDPLimit;
+    ULONG                      ulSQRampingThreshold;
+    USHORT                     usCACLeakageTableOffset;         // Points to ATOM_PPLIB_CAC_Leakage_Table
+    ULONG                      ulCACLeakage;                    // TBD, this parameter is still under discussion.  Change to ulReserved if not needed.
+    ULONG                      ulReserved;
+} ATOM_PPLIB_POWERPLAYTABLE5, *LPATOM_PPLIB_POWERPLAYTABLE5;
+
 //// ATOM_PPLIB_NONCLOCK_INFO::usClassification
 #define ATOM_PPLIB_CLASSIFICATION_UI_MASK          0x0007
 #define ATOM_PPLIB_CLASSIFICATION_UI_SHIFT         0
@@ -5864,6 +6628,10 @@ typedef struct _ATOM_PPLIB_POWERPLAYTABLE3
 #define ATOM_PPLIB_CLASSIFICATION_HDSTATE                0x4000
 #define ATOM_PPLIB_CLASSIFICATION_SDSTATE                0x8000
 
+//// ATOM_PPLIB_NONCLOCK_INFO::usClassification2
+#define ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2     0x0001
+#define ATOM_PPLIB_CLASSIFICATION2_ULV                      0x0002
+
 //// ATOM_PPLIB_NONCLOCK_INFO::ulCapsAndSettings
 #define ATOM_PPLIB_SINGLE_DISPLAY_ONLY           0x00000001
 #define ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK         0x00000002
@@ -5896,9 +6664,21 @@ typedef struct _ATOM_PPLIB_POWERPLAYTABLE3
 #define ATOM_PPLIB_M3ARB_MASK                       0x00060000
 #define ATOM_PPLIB_M3ARB_SHIFT                      17
 
+#define ATOM_PPLIB_ENABLE_DRR                       0x00080000
+
+// remaining 16 bits are reserved
+typedef struct _ATOM_PPLIB_THERMAL_STATE
+{
+    UCHAR   ucMinTemperature;
+    UCHAR   ucMaxTemperature;
+    UCHAR   ucThermalAction;
+}ATOM_PPLIB_THERMAL_STATE, *LPATOM_PPLIB_THERMAL_STATE;
+
 // Contained in an array starting at the offset
 // in ATOM_PPLIB_POWERPLAYTABLE::usNonClockInfoArrayOffset.
 // referenced from ATOM_PPLIB_STATE_INFO::ucNonClockStateIndex
+#define ATOM_PPLIB_NONCLOCKINFO_VER1      12
+#define ATOM_PPLIB_NONCLOCKINFO_VER2      24
 typedef struct _ATOM_PPLIB_NONCLOCK_INFO
 {
       USHORT usClassification;
@@ -5906,15 +6686,15 @@ typedef struct _ATOM_PPLIB_NONCLOCK_INFO
       UCHAR  ucMaxTemperature;
       ULONG  ulCapsAndSettings;
       UCHAR  ucRequiredPower;
-      UCHAR  ucUnused1[3];
+      USHORT usClassification2;
+      ULONG  ulVCLK;
+      ULONG  ulDCLK;
+      UCHAR  ucUnused[5];
 } ATOM_PPLIB_NONCLOCK_INFO;
 
 // Contained in an array starting at the offset
 // in ATOM_PPLIB_POWERPLAYTABLE::usClockInfoArrayOffset.
 // referenced from ATOM_PPLIB_STATE::ucClockStateIndices
-#define ATOM_PPLIB_NONCLOCKINFO_VER1      12
-#define ATOM_PPLIB_NONCLOCKINFO_VER2      24
-
 typedef struct _ATOM_PPLIB_R600_CLOCK_INFO
 {
       USHORT usEngineClockLow;
@@ -5985,6 +6765,93 @@ typedef struct _ATOM_PPLIB_RS780_CLOCK_INFO
 #define ATOM_PPLIB_RS780_HTLINKFREQ_LOW        1 
 #define ATOM_PPLIB_RS780_HTLINKFREQ_HIGH       2 
 
+typedef struct _ATOM_PPLIB_SUMO_CLOCK_INFO{
+      USHORT usEngineClockLow;  //clockfrequency & 0xFFFF. The unit is in 10khz
+      UCHAR  ucEngineClockHigh; //clockfrequency >> 16. 
+      UCHAR  vddcIndex;         //2-bit vddc index;
+      UCHAR  leakage;          //please use 8-bit absolute value, not the 6-bit % value 
+      //please initalize to 0
+      UCHAR  rsv;
+      //please initalize to 0
+      USHORT rsv1;
+      //please initialize to 0s
+      ULONG rsv2[2];
+}ATOM_PPLIB_SUMO_CLOCK_INFO;
+
+
+
+typedef struct _ATOM_PPLIB_STATE_V2
+{
+      //number of valid dpm levels in this state; Driver uses it to calculate the whole 
+      //size of the state: sizeof(ATOM_PPLIB_STATE_V2) + (ucNumDPMLevels - 1) * sizeof(UCHAR)
+      UCHAR ucNumDPMLevels;
+      
+      //a index to the array of nonClockInfos
+      UCHAR nonClockInfoIndex;
+      /**
+      * Driver will read the first ucNumDPMLevels in this array
+      */
+      UCHAR clockInfoIndex[1];
+} ATOM_PPLIB_STATE_V2;
+
+typedef struct StateArray{
+    //how many states we have 
+    UCHAR ucNumEntries;
+    
+    ATOM_PPLIB_STATE_V2 states[1];
+}StateArray;
+
+
+typedef struct ClockInfoArray{
+    //how many clock levels we have
+    UCHAR ucNumEntries;
+    
+    //sizeof(ATOM_PPLIB_SUMO_CLOCK_INFO)
+    UCHAR ucEntrySize;
+    
+    //this is for Sumo
+    ATOM_PPLIB_SUMO_CLOCK_INFO clockInfo[1];
+}ClockInfoArray;
+
+typedef struct NonClockInfoArray{
+
+    //how many non-clock levels we have. normally should be same as number of states
+    UCHAR ucNumEntries;
+    //sizeof(ATOM_PPLIB_NONCLOCK_INFO)
+    UCHAR ucEntrySize;
+    
+    ATOM_PPLIB_NONCLOCK_INFO nonClockInfo[1];
+}NonClockInfoArray;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Record
+{
+    USHORT usClockLow;
+    UCHAR  ucClockHigh;
+    USHORT usVoltage;
+}ATOM_PPLIB_Clock_Voltage_Dependency_Record;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Table
+{
+    UCHAR ucNumEntries;                                                // Number of entries.
+    ATOM_PPLIB_Clock_Voltage_Dependency_Record entries[1];             // Dynamically allocate entries.
+}ATOM_PPLIB_Clock_Voltage_Dependency_Table;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Limit_Record
+{
+    USHORT usSclkLow;
+    UCHAR  ucSclkHigh;
+    USHORT usMclkLow;
+    UCHAR  ucMclkHigh;
+    USHORT usVddc;
+    USHORT usVddci;
+}ATOM_PPLIB_Clock_Voltage_Limit_Record;
+
+typedef struct _ATOM_PPLIB_Clock_Voltage_Limit_Table
+{
+    UCHAR ucNumEntries;                                                // Number of entries.
+    ATOM_PPLIB_Clock_Voltage_Limit_Record entries[1];                  // Dynamically allocate entries.
+}ATOM_PPLIB_Clock_Voltage_Limit_Table;
+
 /**************************************************************************/
 
 
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 4dc5b4714c5a..522d29b37007 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -40,6 +40,61 @@
 static void evergreen_gpu_init(struct radeon_device *rdev);
 void evergreen_fini(struct radeon_device *rdev);
 
+void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc];
+	u32 tmp;
+
+	/* make sure flip is at vb rather than hb */
+	tmp = RREG32(EVERGREEN_GRPH_FLIP_CONTROL + radeon_crtc->crtc_offset);
+	tmp &= ~EVERGREEN_GRPH_SURFACE_UPDATE_H_RETRACE_EN;
+	WREG32(EVERGREEN_GRPH_FLIP_CONTROL + radeon_crtc->crtc_offset, tmp);
+
+	/* set pageflip to happen anywhere in vblank interval */
+	WREG32(EVERGREEN_MASTER_UPDATE_MODE + radeon_crtc->crtc_offset, 0);
+
+	/* enable the pflip int */
+	radeon_irq_kms_pflip_irq_get(rdev, crtc);
+}
+
+void evergreen_post_page_flip(struct radeon_device *rdev, int crtc)
+{
+	/* disable the pflip int */
+	radeon_irq_kms_pflip_irq_put(rdev, crtc);
+}
+
+u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
+
+	/* Lock the graphics update lock */
+	tmp |= EVERGREEN_GRPH_UPDATE_LOCK;
+	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+
+	/* update the scanout addresses */
+	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
+	       upper_32_bits(crtc_base));
+	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+	       (u32)crtc_base);
+
+	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
+	       upper_32_bits(crtc_base));
+	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+	       (u32)crtc_base);
+
+	/* Wait for update_pending to go high. */
+	while (!(RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING));
+	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
+
+	/* Unlock the lock, so double-buffering can take place inside vblank */
+	tmp &= ~EVERGREEN_GRPH_UPDATE_LOCK;
+	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+
+	/* Return current update_pending status: */
+	return RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING;
+}
+
 /* get temperature in millidegrees */
 u32 evergreen_get_temp(struct radeon_device *rdev)
 {
@@ -57,6 +112,14 @@ u32 evergreen_get_temp(struct radeon_device *rdev)
 	return actual_temp * 1000;
 }
 
+u32 sumo_get_temp(struct radeon_device *rdev)
+{
+	u32 temp = RREG32(CG_THERMAL_STATUS) & 0xff;
+	u32 actual_temp = (temp >> 1) & 0xff;
+
+	return actual_temp * 1000;
+}
+
 void evergreen_pm_misc(struct radeon_device *rdev)
 {
 	int req_ps_idx = rdev->pm.requested_power_state_index;
@@ -888,31 +951,39 @@ static void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_sa
 	save->vga_hdp_control = RREG32(VGA_HDP_CONTROL);
 	save->crtc_control[0] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET);
 	save->crtc_control[1] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET);
-	save->crtc_control[2] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET);
-	save->crtc_control[3] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET);
-	save->crtc_control[4] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET);
-	save->crtc_control[5] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		save->crtc_control[2] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET);
+		save->crtc_control[3] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET);
+		save->crtc_control[4] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET);
+		save->crtc_control[5] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET);
+	}
 
 	/* Stop all video */
 	WREG32(VGA_RENDER_CONTROL, 0);
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 1);
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 1);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 1);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 1);
+	}
 	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	}
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	}
 
 	WREG32(D1VGA_CONTROL, 0);
 	WREG32(D2VGA_CONTROL, 0);
@@ -942,41 +1013,43 @@ static void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_
 	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC1_REGISTER_OFFSET,
 	       (u32)rdev->mc.vram_start);
 
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC2_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC2_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC2_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC2_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC3_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC3_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC4_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC4_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC4_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC4_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC5_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC5_REGISTER_OFFSET,
-	       upper_32_bits(rdev->mc.vram_start));
-	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC5_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC5_REGISTER_OFFSET,
-	       (u32)rdev->mc.vram_start);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC5_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC5_REGISTER_OFFSET,
+		       upper_32_bits(rdev->mc.vram_start));
+		WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC5_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC5_REGISTER_OFFSET,
+		       (u32)rdev->mc.vram_start);
+	}
 
 	WREG32(EVERGREEN_VGA_MEMORY_BASE_ADDRESS_HIGH, upper_32_bits(rdev->mc.vram_start));
 	WREG32(EVERGREEN_VGA_MEMORY_BASE_ADDRESS, (u32)rdev->mc.vram_start);
@@ -992,22 +1065,28 @@ static void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_
 	WREG32(EVERGREEN_D6VGA_CONTROL, save->vga_control[5]);
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 1);
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 1);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 1);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 1);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 1);
+	}
 	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, save->crtc_control[0]);
 	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, save->crtc_control[1]);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, save->crtc_control[2]);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, save->crtc_control[3]);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, save->crtc_control[4]);
-	WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, save->crtc_control[5]);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, save->crtc_control[2]);
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, save->crtc_control[3]);
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, save->crtc_control[4]);
+		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, save->crtc_control[5]);
+	}
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
-	WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
+		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	}
 	WREG32(VGA_RENDER_CONTROL, save->vga_render_control);
 }
 
@@ -1283,6 +1362,7 @@ static u32 evergreen_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	switch (rdev->family) {
 	case CHIP_CEDAR:
 	case CHIP_REDWOOD:
+	case CHIP_PALM:
 		force_no_swizzle = false;
 		break;
 	case CHIP_CYPRESS:
@@ -1382,6 +1462,43 @@ static u32 evergreen_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	return backend_map;
 }
 
+static void evergreen_program_channel_remap(struct radeon_device *rdev)
+{
+	u32 tcp_chan_steer_lo, tcp_chan_steer_hi, mc_shared_chremap, tmp;
+
+	tmp = RREG32(MC_SHARED_CHMAP);
+	switch ((tmp & NOOFCHAN_MASK) >> NOOFCHAN_SHIFT) {
+	case 0:
+	case 1:
+	case 2:
+	case 3:
+	default:
+		/* default mapping */
+		mc_shared_chremap = 0x00fac688;
+		break;
+	}
+
+	switch (rdev->family) {
+	case CHIP_HEMLOCK:
+	case CHIP_CYPRESS:
+		tcp_chan_steer_lo = 0x54763210;
+		tcp_chan_steer_hi = 0x0000ba98;
+		break;
+	case CHIP_JUNIPER:
+	case CHIP_REDWOOD:
+	case CHIP_CEDAR:
+	case CHIP_PALM:
+	default:
+		tcp_chan_steer_lo = 0x76543210;
+		tcp_chan_steer_hi = 0x0000ba98;
+		break;
+	}
+
+	WREG32(TCP_CHAN_STEER_LO, tcp_chan_steer_lo);
+	WREG32(TCP_CHAN_STEER_HI, tcp_chan_steer_hi);
+	WREG32(MC_SHARED_CHREMAP, mc_shared_chremap);
+}
+
 static void evergreen_gpu_init(struct radeon_device *rdev)
 {
 	u32 cc_rb_backend_disable = 0;
@@ -1493,6 +1610,27 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 		rdev->config.evergreen.sc_hiz_tile_fifo_size = 0x30;
 		rdev->config.evergreen.sc_earlyz_tile_fifo_size = 0x130;
 		break;
+	case CHIP_PALM:
+		rdev->config.evergreen.num_ses = 1;
+		rdev->config.evergreen.max_pipes = 2;
+		rdev->config.evergreen.max_tile_pipes = 2;
+		rdev->config.evergreen.max_simds = 2;
+		rdev->config.evergreen.max_backends = 1 * rdev->config.evergreen.num_ses;
+		rdev->config.evergreen.max_gprs = 256;
+		rdev->config.evergreen.max_threads = 192;
+		rdev->config.evergreen.max_gs_threads = 16;
+		rdev->config.evergreen.max_stack_entries = 256;
+		rdev->config.evergreen.sx_num_of_sets = 4;
+		rdev->config.evergreen.sx_max_export_size = 128;
+		rdev->config.evergreen.sx_max_export_pos_size = 32;
+		rdev->config.evergreen.sx_max_export_smx_size = 96;
+		rdev->config.evergreen.max_hw_contexts = 4;
+		rdev->config.evergreen.sq_num_cf_insts = 1;
+
+		rdev->config.evergreen.sc_prim_fifo_size = 0x40;
+		rdev->config.evergreen.sc_hiz_tile_fifo_size = 0x30;
+		rdev->config.evergreen.sc_earlyz_tile_fifo_size = 0x130;
+		break;
 	}
 
 	/* Initialize HDP */
@@ -1685,6 +1823,8 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	WREG32(DMIF_ADDR_CONFIG, gb_addr_config);
 	WREG32(HDP_ADDR_CONFIG, gb_addr_config);
 
+	evergreen_program_channel_remap(rdev);
+
 	num_shader_engines = ((RREG32(GB_ADDR_CONFIG) & NUM_SHADER_ENGINES(3)) >> 12) + 1;
 	grbm_gfx_index = INSTANCE_BROADCAST_WRITES;
 
@@ -1767,9 +1907,15 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 		      GS_PRIO(2) |
 		      ES_PRIO(3));
 
-	if (rdev->family == CHIP_CEDAR)
+	switch (rdev->family) {
+	case CHIP_CEDAR:
+	case CHIP_PALM:
 		/* no vertex cache */
 		sq_config &= ~VC_ENABLE;
+		break;
+	default:
+		break;
+	}
 
 	sq_lds_resource_mgmt = RREG32(SQ_LDS_RESOURCE_MGMT);
 
@@ -1781,10 +1927,15 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	sq_gpr_resource_mgmt_3 = NUM_HS_GPRS((rdev->config.evergreen.max_gprs - (4 * 2)) * 3 / 32);
 	sq_gpr_resource_mgmt_3 |= NUM_LS_GPRS((rdev->config.evergreen.max_gprs - (4 * 2)) * 3 / 32);
 
-	if (rdev->family == CHIP_CEDAR)
+	switch (rdev->family) {
+	case CHIP_CEDAR:
+	case CHIP_PALM:
 		ps_thread_count = 96;
-	else
+		break;
+	default:
 		ps_thread_count = 128;
+		break;
+	}
 
 	sq_thread_resource_mgmt = NUM_PS_THREADS(ps_thread_count);
 	sq_thread_resource_mgmt |= NUM_VS_THREADS((((rdev->config.evergreen.max_threads - ps_thread_count) / 6) / 8) * 8);
@@ -1815,10 +1966,15 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	WREG32(PA_SC_FORCE_EOV_MAX_CNTS, (FORCE_EOV_MAX_CLK_CNT(4095) |
 					  FORCE_EOV_MAX_REZ_CNT(255)));
 
-	if (rdev->family == CHIP_CEDAR)
+	switch (rdev->family) {
+	case CHIP_CEDAR:
+	case CHIP_PALM:
 		vgt_cache_invalidation = CACHE_INVALIDATION(TC_ONLY);
-	else
+		break;
+	default:
 		vgt_cache_invalidation = CACHE_INVALIDATION(VC_AND_TC);
+		break;
+	}
 	vgt_cache_invalidation |= AUTO_INVLD_EN(ES_AND_GS_AUTO);
 	WREG32(VGT_CACHE_INVALIDATION, vgt_cache_invalidation);
 
@@ -1902,12 +2058,18 @@ int evergreen_mc_init(struct radeon_device *rdev)
 	rdev->mc.aper_base = pci_resource_start(rdev->pdev, 0);
 	rdev->mc.aper_size = pci_resource_len(rdev->pdev, 0);
 	/* Setup GPU memory space */
-	/* size in MB on evergreen */
-	rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
-	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
+	if (rdev->flags & RADEON_IS_IGP) {
+		/* size in bytes on fusion */
+		rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE);
+		rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE);
+	} else {
+		/* size in MB on evergreen */
+		rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
+		rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
+	}
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
 	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
-	r600_vram_gtt_location(rdev, &rdev->mc);
+	r700_vram_gtt_location(rdev, &rdev->mc);
 	radeon_update_bandwidth_info(rdev);
 
 	return 0;
@@ -2024,17 +2186,21 @@ void evergreen_disable_interrupt_state(struct radeon_device *rdev)
 	WREG32(GRBM_INT_CNTL, 0);
 	WREG32(INT_MASK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(INT_MASK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
-	WREG32(INT_MASK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
-	WREG32(INT_MASK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
-	WREG32(INT_MASK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
-	WREG32(INT_MASK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(INT_MASK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
+		WREG32(INT_MASK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+		WREG32(INT_MASK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
+		WREG32(INT_MASK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	}
 
 	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
-	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
-	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
-	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
-	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
+	}
 
 	WREG32(DACA_AUTODETECT_INT_CONTROL, 0);
 	WREG32(DACB_AUTODETECT_INT_CONTROL, 0);
@@ -2060,6 +2226,7 @@ int evergreen_irq_set(struct radeon_device *rdev)
 	u32 crtc1 = 0, crtc2 = 0, crtc3 = 0, crtc4 = 0, crtc5 = 0, crtc6 = 0;
 	u32 hpd1, hpd2, hpd3, hpd4, hpd5, hpd6;
 	u32 grbm_int_cntl = 0;
+	u32 grph1 = 0, grph2 = 0, grph3 = 0, grph4 = 0, grph5 = 0, grph6 = 0;
 
 	if (!rdev->irq.installed) {
 		WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
@@ -2085,27 +2252,33 @@ int evergreen_irq_set(struct radeon_device *rdev)
 		cp_int_cntl |= RB_INT_ENABLE;
 		cp_int_cntl |= TIME_STAMP_INT_ENABLE;
 	}
-	if (rdev->irq.crtc_vblank_int[0]) {
+	if (rdev->irq.crtc_vblank_int[0] ||
+	    rdev->irq.pflip[0]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 0\n");
 		crtc1 |= VBLANK_INT_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[1]) {
+	if (rdev->irq.crtc_vblank_int[1] ||
+	    rdev->irq.pflip[1]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 1\n");
 		crtc2 |= VBLANK_INT_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[2]) {
+	if (rdev->irq.crtc_vblank_int[2] ||
+	    rdev->irq.pflip[2]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 2\n");
 		crtc3 |= VBLANK_INT_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[3]) {
+	if (rdev->irq.crtc_vblank_int[3] ||
+	    rdev->irq.pflip[3]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 3\n");
 		crtc4 |= VBLANK_INT_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[4]) {
+	if (rdev->irq.crtc_vblank_int[4] ||
+	    rdev->irq.pflip[4]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 4\n");
 		crtc5 |= VBLANK_INT_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[5]) {
+	if (rdev->irq.crtc_vblank_int[5] ||
+	    rdev->irq.pflip[5]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 5\n");
 		crtc6 |= VBLANK_INT_MASK;
 	}
@@ -2143,10 +2316,19 @@ int evergreen_irq_set(struct radeon_device *rdev)
 
 	WREG32(INT_MASK + EVERGREEN_CRTC0_REGISTER_OFFSET, crtc1);
 	WREG32(INT_MASK + EVERGREEN_CRTC1_REGISTER_OFFSET, crtc2);
-	WREG32(INT_MASK + EVERGREEN_CRTC2_REGISTER_OFFSET, crtc3);
-	WREG32(INT_MASK + EVERGREEN_CRTC3_REGISTER_OFFSET, crtc4);
-	WREG32(INT_MASK + EVERGREEN_CRTC4_REGISTER_OFFSET, crtc5);
-	WREG32(INT_MASK + EVERGREEN_CRTC5_REGISTER_OFFSET, crtc6);
+	if (!(rdev->flags & RADEON_IS_IGP)) {
+		WREG32(INT_MASK + EVERGREEN_CRTC2_REGISTER_OFFSET, crtc3);
+		WREG32(INT_MASK + EVERGREEN_CRTC3_REGISTER_OFFSET, crtc4);
+		WREG32(INT_MASK + EVERGREEN_CRTC4_REGISTER_OFFSET, crtc5);
+		WREG32(INT_MASK + EVERGREEN_CRTC5_REGISTER_OFFSET, crtc6);
+	}
+
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, grph1);
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, grph2);
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, grph3);
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, grph4);
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, grph5);
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, grph6);
 
 	WREG32(DC_HPD1_INT_CONTROL, hpd1);
 	WREG32(DC_HPD2_INT_CONTROL, hpd2);
@@ -2158,79 +2340,92 @@ int evergreen_irq_set(struct radeon_device *rdev)
 	return 0;
 }
 
-static inline void evergreen_irq_ack(struct radeon_device *rdev,
-				     u32 *disp_int,
-				     u32 *disp_int_cont,
-				     u32 *disp_int_cont2,
-				     u32 *disp_int_cont3,
-				     u32 *disp_int_cont4,
-				     u32 *disp_int_cont5)
+static inline void evergreen_irq_ack(struct radeon_device *rdev)
 {
 	u32 tmp;
 
-	*disp_int = RREG32(DISP_INTERRUPT_STATUS);
-	*disp_int_cont = RREG32(DISP_INTERRUPT_STATUS_CONTINUE);
-	*disp_int_cont2 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE2);
-	*disp_int_cont3 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE3);
-	*disp_int_cont4 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE4);
-	*disp_int_cont5 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE5);
-
-	if (*disp_int & LB_D1_VBLANK_INTERRUPT)
+	rdev->irq.stat_regs.evergreen.disp_int = RREG32(DISP_INTERRUPT_STATUS);
+	rdev->irq.stat_regs.evergreen.disp_int_cont = RREG32(DISP_INTERRUPT_STATUS_CONTINUE);
+	rdev->irq.stat_regs.evergreen.disp_int_cont2 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE2);
+	rdev->irq.stat_regs.evergreen.disp_int_cont3 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE3);
+	rdev->irq.stat_regs.evergreen.disp_int_cont4 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE4);
+	rdev->irq.stat_regs.evergreen.disp_int_cont5 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE5);
+	rdev->irq.stat_regs.evergreen.d1grph_int = RREG32(GRPH_INT_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET);
+	rdev->irq.stat_regs.evergreen.d2grph_int = RREG32(GRPH_INT_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET);
+	rdev->irq.stat_regs.evergreen.d3grph_int = RREG32(GRPH_INT_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET);
+	rdev->irq.stat_regs.evergreen.d4grph_int = RREG32(GRPH_INT_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET);
+	rdev->irq.stat_regs.evergreen.d5grph_int = RREG32(GRPH_INT_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET);
+	rdev->irq.stat_regs.evergreen.d6grph_int = RREG32(GRPH_INT_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET);
+
+	if (rdev->irq.stat_regs.evergreen.d1grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET, GRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.evergreen.d2grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET, GRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.evergreen.d3grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, GRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.evergreen.d4grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, GRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.evergreen.d5grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, GRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.evergreen.d6grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, GRPH_PFLIP_INT_CLEAR);
+
+	if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET, VBLANK_ACK);
-	if (*disp_int & LB_D1_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET, VLINE_ACK);
 
-	if (*disp_int_cont & LB_D2_VBLANK_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET, VBLANK_ACK);
-	if (*disp_int_cont & LB_D2_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET, VLINE_ACK);
 
-	if (*disp_int_cont2 & LB_D3_VBLANK_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VBLANK_ACK);
-	if (*disp_int_cont2 & LB_D3_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VLINE_ACK);
 
-	if (*disp_int_cont3 & LB_D4_VBLANK_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, VBLANK_ACK);
-	if (*disp_int_cont3 & LB_D4_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, VLINE_ACK);
 
-	if (*disp_int_cont4 & LB_D5_VBLANK_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VBLANK_ACK);
-	if (*disp_int_cont4 & LB_D5_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VLINE_ACK);
 
-	if (*disp_int_cont5 & LB_D6_VBLANK_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, VBLANK_ACK);
-	if (*disp_int_cont5 & LB_D6_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, VLINE_ACK);
 
-	if (*disp_int & DC_HPD1_INTERRUPT) {
+	if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT) {
 		tmp = RREG32(DC_HPD1_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD1_INT_CONTROL, tmp);
 	}
-	if (*disp_int_cont & DC_HPD2_INTERRUPT) {
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_INTERRUPT) {
 		tmp = RREG32(DC_HPD2_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD2_INT_CONTROL, tmp);
 	}
-	if (*disp_int_cont2 & DC_HPD3_INTERRUPT) {
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_INTERRUPT) {
 		tmp = RREG32(DC_HPD3_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD3_INT_CONTROL, tmp);
 	}
-	if (*disp_int_cont3 & DC_HPD4_INTERRUPT) {
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_INTERRUPT) {
 		tmp = RREG32(DC_HPD4_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD4_INT_CONTROL, tmp);
 	}
-	if (*disp_int_cont4 & DC_HPD5_INTERRUPT) {
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_INTERRUPT) {
 		tmp = RREG32(DC_HPD5_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD5_INT_CONTROL, tmp);
 	}
-	if (*disp_int_cont5 & DC_HPD6_INTERRUPT) {
+	if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_INTERRUPT) {
 		tmp = RREG32(DC_HPD5_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD6_INT_CONTROL, tmp);
@@ -2239,14 +2434,10 @@ static inline void evergreen_irq_ack(struct radeon_device *rdev,
 
 void evergreen_irq_disable(struct radeon_device *rdev)
 {
-	u32 disp_int, disp_int_cont, disp_int_cont2;
-	u32 disp_int_cont3, disp_int_cont4, disp_int_cont5;
-
 	r600_disable_interrupts(rdev);
 	/* Wait and acknowledge irq */
 	mdelay(1);
-	evergreen_irq_ack(rdev, &disp_int, &disp_int_cont, &disp_int_cont2,
-			  &disp_int_cont3, &disp_int_cont4, &disp_int_cont5);
+	evergreen_irq_ack(rdev);
 	evergreen_disable_interrupt_state(rdev);
 }
 
@@ -2286,8 +2477,6 @@ int evergreen_irq_process(struct radeon_device *rdev)
 	u32 rptr = rdev->ih.rptr;
 	u32 src_id, src_data;
 	u32 ring_index;
-	u32 disp_int, disp_int_cont, disp_int_cont2;
-	u32 disp_int_cont3, disp_int_cont4, disp_int_cont5;
 	unsigned long flags;
 	bool queue_hotplug = false;
 
@@ -2308,8 +2497,7 @@ int evergreen_irq_process(struct radeon_device *rdev)
 
 restart_ih:
 	/* display interrupts */
-	evergreen_irq_ack(rdev, &disp_int, &disp_int_cont, &disp_int_cont2,
-			  &disp_int_cont3, &disp_int_cont4, &disp_int_cont5);
+	evergreen_irq_ack(rdev);
 
 	rdev->ih.wptr = wptr;
 	while (rptr != wptr) {
@@ -2322,17 +2510,21 @@ restart_ih:
 		case 1: /* D1 vblank/vline */
 			switch (src_data) {
 			case 0: /* D1 vblank */
-				if (disp_int & LB_D1_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 0);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int &= ~LB_D1_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[0]) {
+						drm_handle_vblank(rdev->ddev, 0);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[0])
+						radeon_crtc_handle_flip(rdev, 0);
+					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
 				break;
 			case 1: /* D1 vline */
-				if (disp_int & LB_D1_VLINE_INTERRUPT) {
-					disp_int &= ~LB_D1_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D1 vline\n");
 				}
 				break;
@@ -2344,17 +2536,21 @@ restart_ih:
 		case 2: /* D2 vblank/vline */
 			switch (src_data) {
 			case 0: /* D2 vblank */
-				if (disp_int_cont & LB_D2_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 1);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[1]) {
+						drm_handle_vblank(rdev->ddev, 1);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[1])
+						radeon_crtc_handle_flip(rdev, 1);
+					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
 				break;
 			case 1: /* D2 vline */
-				if (disp_int_cont & LB_D2_VLINE_INTERRUPT) {
-					disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D2 vline\n");
 				}
 				break;
@@ -2366,17 +2562,21 @@ restart_ih:
 		case 3: /* D3 vblank/vline */
 			switch (src_data) {
 			case 0: /* D3 vblank */
-				if (disp_int_cont2 & LB_D3_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 2);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[2]) {
+						drm_handle_vblank(rdev->ddev, 2);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[2])
+						radeon_crtc_handle_flip(rdev, 2);
+					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
 				break;
 			case 1: /* D3 vline */
-				if (disp_int_cont2 & LB_D3_VLINE_INTERRUPT) {
-					disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D3 vline\n");
 				}
 				break;
@@ -2388,17 +2588,21 @@ restart_ih:
 		case 4: /* D4 vblank/vline */
 			switch (src_data) {
 			case 0: /* D4 vblank */
-				if (disp_int_cont3 & LB_D4_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 3);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[3]) {
+						drm_handle_vblank(rdev->ddev, 3);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[3])
+						radeon_crtc_handle_flip(rdev, 3);
+					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
 				break;
 			case 1: /* D4 vline */
-				if (disp_int_cont3 & LB_D4_VLINE_INTERRUPT) {
-					disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D4 vline\n");
 				}
 				break;
@@ -2410,17 +2614,21 @@ restart_ih:
 		case 5: /* D5 vblank/vline */
 			switch (src_data) {
 			case 0: /* D5 vblank */
-				if (disp_int_cont4 & LB_D5_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 4);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[4]) {
+						drm_handle_vblank(rdev->ddev, 4);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[4])
+						radeon_crtc_handle_flip(rdev, 4);
+					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
 				break;
 			case 1: /* D5 vline */
-				if (disp_int_cont4 & LB_D5_VLINE_INTERRUPT) {
-					disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D5 vline\n");
 				}
 				break;
@@ -2432,17 +2640,21 @@ restart_ih:
 		case 6: /* D6 vblank/vline */
 			switch (src_data) {
 			case 0: /* D6 vblank */
-				if (disp_int_cont5 & LB_D6_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 5);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[5]) {
+						drm_handle_vblank(rdev->ddev, 5);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[5])
+						radeon_crtc_handle_flip(rdev, 5);
+					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
 				break;
 			case 1: /* D6 vline */
-				if (disp_int_cont5 & LB_D6_VLINE_INTERRUPT) {
-					disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D6 vline\n");
 				}
 				break;
@@ -2454,43 +2666,43 @@ restart_ih:
 		case 42: /* HPD hotplug */
 			switch (src_data) {
 			case 0:
-				if (disp_int & DC_HPD1_INTERRUPT) {
-					disp_int &= ~DC_HPD1_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD1\n");
 				}
 				break;
 			case 1:
-				if (disp_int_cont & DC_HPD2_INTERRUPT) {
-					disp_int_cont &= ~DC_HPD2_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD2\n");
 				}
 				break;
 			case 2:
-				if (disp_int_cont2 & DC_HPD3_INTERRUPT) {
-					disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD3\n");
 				}
 				break;
 			case 3:
-				if (disp_int_cont3 & DC_HPD4_INTERRUPT) {
-					disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD4\n");
 				}
 				break;
 			case 4:
-				if (disp_int_cont4 & DC_HPD5_INTERRUPT) {
-					disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD5\n");
 				}
 				break;
 			case 5:
-				if (disp_int_cont5 & DC_HPD6_INTERRUPT) {
-					disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
+				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_INTERRUPT) {
+					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD6\n");
 				}
@@ -2666,12 +2878,16 @@ static bool evergreen_card_posted(struct radeon_device *rdev)
 	u32 reg;
 
 	/* first check CRTCs */
-	reg = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET) |
-		RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET) |
-		RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET) |
-		RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET) |
-		RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET) |
-		RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET);
+	if (rdev->flags & RADEON_IS_IGP)
+		reg = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET);
+	else
+		reg = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET);
 	if (reg & EVERGREEN_CRTC_MASTER_EN)
 		return true;
 
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
index e0e590110dd4..2ccd1f0545fe 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
@@ -147,7 +147,8 @@ set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
 	radeon_ring_write(rdev, 0);
 	radeon_ring_write(rdev, SQ_TEX_VTX_VALID_BUFFER << 30);
 
-	if (rdev->family == CHIP_CEDAR)
+	if ((rdev->family == CHIP_CEDAR) ||
+	    (rdev->family == CHIP_PALM))
 		cp_set_surface_sync(rdev,
 				    PACKET3_TC_ACTION_ENA, 48, gpu_addr);
 	else
@@ -331,9 +332,31 @@ set_default_state(struct radeon_device *rdev)
 		num_hs_stack_entries = 85;
 		num_ls_stack_entries = 85;
 		break;
+	case CHIP_PALM:
+		num_ps_gprs = 93;
+		num_vs_gprs = 46;
+		num_temp_gprs = 4;
+		num_gs_gprs = 31;
+		num_es_gprs = 31;
+		num_hs_gprs = 23;
+		num_ls_gprs = 23;
+		num_ps_threads = 96;
+		num_vs_threads = 16;
+		num_gs_threads = 16;
+		num_es_threads = 16;
+		num_hs_threads = 16;
+		num_ls_threads = 16;
+		num_ps_stack_entries = 42;
+		num_vs_stack_entries = 42;
+		num_gs_stack_entries = 42;
+		num_es_stack_entries = 42;
+		num_hs_stack_entries = 42;
+		num_ls_stack_entries = 42;
+		break;
 	}
 
-	if (rdev->family == CHIP_CEDAR)
+	if ((rdev->family == CHIP_CEDAR) ||
+	    (rdev->family == CHIP_PALM))
 		sq_config = 0;
 	else
 		sq_config = VC_ENABLE;
diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h
index 2330f3a36fd5..c781c92c3451 100644
--- a/drivers/gpu/drm/radeon/evergreen_reg.h
+++ b/drivers/gpu/drm/radeon/evergreen_reg.h
@@ -105,6 +105,11 @@
 #define EVERGREEN_GRPH_Y_START                          0x6830
 #define EVERGREEN_GRPH_X_END                            0x6834
 #define EVERGREEN_GRPH_Y_END                            0x6838
+#define EVERGREEN_GRPH_UPDATE                           0x6844
+#       define EVERGREEN_GRPH_SURFACE_UPDATE_PENDING    (1 << 2)
+#       define EVERGREEN_GRPH_UPDATE_LOCK               (1 << 16)
+#define EVERGREEN_GRPH_FLIP_CONTROL                     0x6848
+#       define EVERGREEN_GRPH_SURFACE_UPDATE_H_RETRACE_EN (1 << 0)
 
 /* CUR blocks at 0x6998, 0x7598, 0x10198, 0x10d98, 0x11998, 0x12598 */
 #define EVERGREEN_CUR_CONTROL                           0x6998
@@ -178,6 +183,7 @@
 #       define EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE (1 << 24)
 #define EVERGREEN_CRTC_STATUS                           0x6e8c
 #define EVERGREEN_CRTC_STATUS_POSITION                  0x6e90
+#define EVERGREEN_MASTER_UPDATE_MODE                    0x6ef8
 #define EVERGREEN_CRTC_UPDATE_LOCK                      0x6ed4
 
 #define EVERGREEN_DC_GPIO_HPD_MASK                      0x64b0
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index 113c70cc8b39..87fcaba76695 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -164,11 +164,13 @@
 #define		SE_SC_BUSY					(1 << 29)
 #define		SE_DB_BUSY					(1 << 30)
 #define		SE_CB_BUSY					(1 << 31)
-
+/* evergreen */
 #define	CG_MULT_THERMAL_STATUS				0x740
 #define		ASIC_T(x)			        ((x) << 16)
 #define		ASIC_T_MASK			        0x7FF0000
 #define		ASIC_T_SHIFT			        16
+/* APU */
+#define	CG_THERMAL_STATUS			        0x678
 
 #define	HDP_HOST_PATH_CNTL				0x2C00
 #define	HDP_NONSURFACE_BASE				0x2C04
@@ -180,6 +182,7 @@
 #define MC_SHARED_CHMAP						0x2004
 #define		NOOFCHAN_SHIFT					12
 #define		NOOFCHAN_MASK					0x00003000
+#define MC_SHARED_CHREMAP					0x2008
 
 #define	MC_ARB_RAMCFG					0x2760
 #define		NOOFBANK_SHIFT					0
@@ -348,6 +351,9 @@
 #define		SYNC_WALKER					(1 << 25)
 #define		SYNC_ALIGNER					(1 << 26)
 
+#define	TCP_CHAN_STEER_LO				0x960c
+#define	TCP_CHAN_STEER_HI				0x9610
+
 #define	VGT_CACHE_INVALIDATION				0x88C4
 #define		CACHE_INVALIDATION(x)				((x) << 0)
 #define			VC_ONLY						0
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 8e10aa9f74b0..2316f73db6c0 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -68,6 +68,54 @@ MODULE_FIRMWARE(FIRMWARE_R520);
  * r100,rv100,rs100,rv200,rs200,r200,rv250,rs300,rv280
  */
 
+void r100_pre_page_flip(struct radeon_device *rdev, int crtc)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc];
+	u32 tmp;
+
+	/* make sure flip is at vb rather than hb */
+	tmp = RREG32(RADEON_CRTC_OFFSET_CNTL + radeon_crtc->crtc_offset);
+	tmp &= ~RADEON_CRTC_OFFSET_FLIP_CNTL;
+	WREG32(RADEON_CRTC_OFFSET_CNTL + radeon_crtc->crtc_offset, tmp);
+
+	/* set pageflip to happen as late as possible in the vblank interval.
+	 * same field for crtc1/2
+	 */
+	tmp = RREG32(RADEON_CRTC_GEN_CNTL);
+	tmp &= ~RADEON_CRTC_VSTAT_MODE_MASK;
+	WREG32(RADEON_CRTC_GEN_CNTL, tmp);
+
+	/* enable the pflip int */
+	radeon_irq_kms_pflip_irq_get(rdev, crtc);
+}
+
+void r100_post_page_flip(struct radeon_device *rdev, int crtc)
+{
+	/* disable the pflip int */
+	radeon_irq_kms_pflip_irq_put(rdev, crtc);
+}
+
+u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	u32 tmp = ((u32)crtc_base) | RADEON_CRTC_OFFSET__OFFSET_LOCK;
+
+	/* Lock the graphics update lock */
+	/* update the scanout addresses */
+	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
+
+	/* Note: We don't wait for update_pending to assert, as this never
+	 * happens for some reason on R1xx - R4xx. Adds a bit of imprecision.
+	 */
+
+	/* Unlock the lock, so double-buffering can take place inside vblank */
+	tmp &= ~RADEON_CRTC_OFFSET__OFFSET_LOCK;
+	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
+
+	/* Return current update_pending status: */
+	return RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET;
+}
+
 void r100_pm_get_dynpm_state(struct radeon_device *rdev)
 {
 	int i;
@@ -526,10 +574,12 @@ int r100_irq_set(struct radeon_device *rdev)
 	if (rdev->irq.gui_idle) {
 		tmp |= RADEON_GUI_IDLE_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[0]) {
+	if (rdev->irq.crtc_vblank_int[0] ||
+	    rdev->irq.pflip[0]) {
 		tmp |= RADEON_CRTC_VBLANK_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[1]) {
+	if (rdev->irq.crtc_vblank_int[1] ||
+	    rdev->irq.pflip[1]) {
 		tmp |= RADEON_CRTC2_VBLANK_MASK;
 	}
 	if (rdev->irq.hpd[0]) {
@@ -600,14 +650,22 @@ int r100_irq_process(struct radeon_device *rdev)
 		}
 		/* Vertical blank interrupts */
 		if (status & RADEON_CRTC_VBLANK_STAT) {
-			drm_handle_vblank(rdev->ddev, 0);
-			rdev->pm.vblank_sync = true;
-			wake_up(&rdev->irq.vblank_queue);
+			if (rdev->irq.crtc_vblank_int[0]) {
+				drm_handle_vblank(rdev->ddev, 0);
+				rdev->pm.vblank_sync = true;
+				wake_up(&rdev->irq.vblank_queue);
+			}
+			if (rdev->irq.pflip[0])
+				radeon_crtc_handle_flip(rdev, 0);
 		}
 		if (status & RADEON_CRTC2_VBLANK_STAT) {
-			drm_handle_vblank(rdev->ddev, 1);
-			rdev->pm.vblank_sync = true;
-			wake_up(&rdev->irq.vblank_queue);
+			if (rdev->irq.crtc_vblank_int[1]) {
+				drm_handle_vblank(rdev->ddev, 1);
+				rdev->pm.vblank_sync = true;
+				wake_up(&rdev->irq.vblank_queue);
+			}
+			if (rdev->irq.pflip[1])
+				radeon_crtc_handle_flip(rdev, 1);
 		}
 		if (status & RADEON_FP_DETECT_STAT) {
 			queue_hotplug = true;
diff --git a/drivers/gpu/drm/radeon/r500_reg.h b/drivers/gpu/drm/radeon/r500_reg.h
index 6ac1f604e29b..fc437059918f 100644
--- a/drivers/gpu/drm/radeon/r500_reg.h
+++ b/drivers/gpu/drm/radeon/r500_reg.h
@@ -355,6 +355,8 @@
 #define AVIVO_D1CRTC_FRAME_COUNT                                0x60a4
 #define AVIVO_D1CRTC_STEREO_CONTROL                             0x60c4
 
+#define AVIVO_D1MODE_MASTER_UPDATE_MODE                         0x60e4
+
 /* master controls */
 #define AVIVO_DC_CRTC_MASTER_EN                                 0x60f8
 #define AVIVO_DC_CRTC_TV_CONTROL                                0x60fc
@@ -409,8 +411,10 @@
 #define AVIVO_D1GRPH_X_END                                      0x6134
 #define AVIVO_D1GRPH_Y_END                                      0x6138
 #define AVIVO_D1GRPH_UPDATE                                     0x6144
+#       define AVIVO_D1GRPH_SURFACE_UPDATE_PENDING              (1 << 2)
 #       define AVIVO_D1GRPH_UPDATE_LOCK                         (1 << 16)
 #define AVIVO_D1GRPH_FLIP_CONTROL                               0x6148
+#       define AVIVO_D1GRPH_SURFACE_UPDATE_H_RETRACE_EN         (1 << 0)
 
 #define AVIVO_D1CUR_CONTROL                     0x6400
 #       define AVIVO_D1CURSOR_EN                (1 << 0)
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index a322d4f647bd..c6a37e036f11 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -83,6 +83,9 @@ MODULE_FIRMWARE("radeon/JUNIPER_rlc.bin");
 MODULE_FIRMWARE("radeon/CYPRESS_pfp.bin");
 MODULE_FIRMWARE("radeon/CYPRESS_me.bin");
 MODULE_FIRMWARE("radeon/CYPRESS_rlc.bin");
+MODULE_FIRMWARE("radeon/PALM_pfp.bin");
+MODULE_FIRMWARE("radeon/PALM_me.bin");
+MODULE_FIRMWARE("radeon/SUMO_rlc.bin");
 
 int r600_debugfs_mc_info_init(struct radeon_device *rdev);
 
@@ -1161,7 +1164,7 @@ static void r600_mc_program(struct radeon_device *rdev)
  * Note: GTT start, end, size should be initialized before calling this
  * function on AGP platform.
  */
-void r600_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc)
+static void r600_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc)
 {
 	u64 size_bf, size_af;
 
@@ -2000,6 +2003,10 @@ int r600_init_microcode(struct radeon_device *rdev)
 		chip_name = "CYPRESS";
 		rlc_chip_name = "CYPRESS";
 		break;
+	case CHIP_PALM:
+		chip_name = "PALM";
+		rlc_chip_name = "SUMO";
+		break;
 	default: BUG();
 	}
 
@@ -2865,6 +2872,8 @@ static void r600_disable_interrupt_state(struct radeon_device *rdev)
 	WREG32(CP_INT_CNTL, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
 	WREG32(GRBM_INT_CNTL, 0);
 	WREG32(DxMODE_INT_MASK, 0);
+	WREG32(D1GRPH_INTERRUPT_CONTROL, 0);
+	WREG32(D2GRPH_INTERRUPT_CONTROL, 0);
 	if (ASIC_IS_DCE3(rdev)) {
 		WREG32(DCE3_DACA_AUTODETECT_INT_CONTROL, 0);
 		WREG32(DCE3_DACB_AUTODETECT_INT_CONTROL, 0);
@@ -2989,6 +2998,7 @@ int r600_irq_set(struct radeon_device *rdev)
 	u32 hpd1, hpd2, hpd3, hpd4 = 0, hpd5 = 0, hpd6 = 0;
 	u32 grbm_int_cntl = 0;
 	u32 hdmi1, hdmi2;
+	u32 d1grph = 0, d2grph = 0;
 
 	if (!rdev->irq.installed) {
 		WARN(1, "Can't enable IRQ/MSI because no handler is installed\n");
@@ -3025,11 +3035,13 @@ int r600_irq_set(struct radeon_device *rdev)
 		cp_int_cntl |= RB_INT_ENABLE;
 		cp_int_cntl |= TIME_STAMP_INT_ENABLE;
 	}
-	if (rdev->irq.crtc_vblank_int[0]) {
+	if (rdev->irq.crtc_vblank_int[0] ||
+	    rdev->irq.pflip[0]) {
 		DRM_DEBUG("r600_irq_set: vblank 0\n");
 		mode_int |= D1MODE_VBLANK_INT_MASK;
 	}
-	if (rdev->irq.crtc_vblank_int[1]) {
+	if (rdev->irq.crtc_vblank_int[1] ||
+	    rdev->irq.pflip[1]) {
 		DRM_DEBUG("r600_irq_set: vblank 1\n");
 		mode_int |= D2MODE_VBLANK_INT_MASK;
 	}
@@ -3072,6 +3084,8 @@ int r600_irq_set(struct radeon_device *rdev)
 
 	WREG32(CP_INT_CNTL, cp_int_cntl);
 	WREG32(DxMODE_INT_MASK, mode_int);
+	WREG32(D1GRPH_INTERRUPT_CONTROL, d1grph);
+	WREG32(D2GRPH_INTERRUPT_CONTROL, d2grph);
 	WREG32(GRBM_INT_CNTL, grbm_int_cntl);
 	WREG32(R600_HDMI_BLOCK1 + R600_HDMI_CNTL, hdmi1);
 	if (ASIC_IS_DCE3(rdev)) {
@@ -3094,32 +3108,35 @@ int r600_irq_set(struct radeon_device *rdev)
 	return 0;
 }
 
-static inline void r600_irq_ack(struct radeon_device *rdev,
-				u32 *disp_int,
-				u32 *disp_int_cont,
-				u32 *disp_int_cont2)
+static inline void r600_irq_ack(struct radeon_device *rdev)
 {
 	u32 tmp;
 
 	if (ASIC_IS_DCE3(rdev)) {
-		*disp_int = RREG32(DCE3_DISP_INTERRUPT_STATUS);
-		*disp_int_cont = RREG32(DCE3_DISP_INTERRUPT_STATUS_CONTINUE);
-		*disp_int_cont2 = RREG32(DCE3_DISP_INTERRUPT_STATUS_CONTINUE2);
+		rdev->irq.stat_regs.r600.disp_int = RREG32(DCE3_DISP_INTERRUPT_STATUS);
+		rdev->irq.stat_regs.r600.disp_int_cont = RREG32(DCE3_DISP_INTERRUPT_STATUS_CONTINUE);
+		rdev->irq.stat_regs.r600.disp_int_cont2 = RREG32(DCE3_DISP_INTERRUPT_STATUS_CONTINUE2);
 	} else {
-		*disp_int = RREG32(DISP_INTERRUPT_STATUS);
-		*disp_int_cont = RREG32(DISP_INTERRUPT_STATUS_CONTINUE);
-		*disp_int_cont2 = 0;
-	}
-
-	if (*disp_int & LB_D1_VBLANK_INTERRUPT)
+		rdev->irq.stat_regs.r600.disp_int = RREG32(DISP_INTERRUPT_STATUS);
+		rdev->irq.stat_regs.r600.disp_int_cont = RREG32(DISP_INTERRUPT_STATUS_CONTINUE);
+		rdev->irq.stat_regs.r600.disp_int_cont2 = 0;
+	}
+	rdev->irq.stat_regs.r600.d1grph_int = RREG32(D1GRPH_INTERRUPT_STATUS);
+	rdev->irq.stat_regs.r600.d2grph_int = RREG32(D2GRPH_INTERRUPT_STATUS);
+
+	if (rdev->irq.stat_regs.r600.d1grph_int & DxGRPH_PFLIP_INT_OCCURRED)
+		WREG32(D1GRPH_INTERRUPT_STATUS, DxGRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.r600.d2grph_int & DxGRPH_PFLIP_INT_OCCURRED)
+		WREG32(D2GRPH_INTERRUPT_STATUS, DxGRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.r600.disp_int & LB_D1_VBLANK_INTERRUPT)
 		WREG32(D1MODE_VBLANK_STATUS, DxMODE_VBLANK_ACK);
-	if (*disp_int & LB_D1_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.r600.disp_int & LB_D1_VLINE_INTERRUPT)
 		WREG32(D1MODE_VLINE_STATUS, DxMODE_VLINE_ACK);
-	if (*disp_int & LB_D2_VBLANK_INTERRUPT)
+	if (rdev->irq.stat_regs.r600.disp_int & LB_D2_VBLANK_INTERRUPT)
 		WREG32(D2MODE_VBLANK_STATUS, DxMODE_VBLANK_ACK);
-	if (*disp_int & LB_D2_VLINE_INTERRUPT)
+	if (rdev->irq.stat_regs.r600.disp_int & LB_D2_VLINE_INTERRUPT)
 		WREG32(D2MODE_VLINE_STATUS, DxMODE_VLINE_ACK);
-	if (*disp_int & DC_HPD1_INTERRUPT) {
+	if (rdev->irq.stat_regs.r600.disp_int & DC_HPD1_INTERRUPT) {
 		if (ASIC_IS_DCE3(rdev)) {
 			tmp = RREG32(DC_HPD1_INT_CONTROL);
 			tmp |= DC_HPDx_INT_ACK;
@@ -3130,7 +3147,7 @@ static inline void r600_irq_ack(struct radeon_device *rdev,
 			WREG32(DC_HOT_PLUG_DETECT1_INT_CONTROL, tmp);
 		}
 	}
-	if (*disp_int & DC_HPD2_INTERRUPT) {
+	if (rdev->irq.stat_regs.r600.disp_int & DC_HPD2_INTERRUPT) {
 		if (ASIC_IS_DCE3(rdev)) {
 			tmp = RREG32(DC_HPD2_INT_CONTROL);
 			tmp |= DC_HPDx_INT_ACK;
@@ -3141,7 +3158,7 @@ static inline void r600_irq_ack(struct radeon_device *rdev,
 			WREG32(DC_HOT_PLUG_DETECT2_INT_CONTROL, tmp);
 		}
 	}
-	if (*disp_int_cont & DC_HPD3_INTERRUPT) {
+	if (rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD3_INTERRUPT) {
 		if (ASIC_IS_DCE3(rdev)) {
 			tmp = RREG32(DC_HPD3_INT_CONTROL);
 			tmp |= DC_HPDx_INT_ACK;
@@ -3152,18 +3169,18 @@ static inline void r600_irq_ack(struct radeon_device *rdev,
 			WREG32(DC_HOT_PLUG_DETECT3_INT_CONTROL, tmp);
 		}
 	}
-	if (*disp_int_cont & DC_HPD4_INTERRUPT) {
+	if (rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD4_INTERRUPT) {
 		tmp = RREG32(DC_HPD4_INT_CONTROL);
 		tmp |= DC_HPDx_INT_ACK;
 		WREG32(DC_HPD4_INT_CONTROL, tmp);
 	}
 	if (ASIC_IS_DCE32(rdev)) {
-		if (*disp_int_cont2 & DC_HPD5_INTERRUPT) {
+		if (rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD5_INTERRUPT) {
 			tmp = RREG32(DC_HPD5_INT_CONTROL);
 			tmp |= DC_HPDx_INT_ACK;
 			WREG32(DC_HPD5_INT_CONTROL, tmp);
 		}
-		if (*disp_int_cont2 & DC_HPD6_INTERRUPT) {
+		if (rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD6_INTERRUPT) {
 			tmp = RREG32(DC_HPD5_INT_CONTROL);
 			tmp |= DC_HPDx_INT_ACK;
 			WREG32(DC_HPD6_INT_CONTROL, tmp);
@@ -3185,12 +3202,10 @@ static inline void r600_irq_ack(struct radeon_device *rdev,
 
 void r600_irq_disable(struct radeon_device *rdev)
 {
-	u32 disp_int, disp_int_cont, disp_int_cont2;
-
 	r600_disable_interrupts(rdev);
 	/* Wait and acknowledge irq */
 	mdelay(1);
-	r600_irq_ack(rdev, &disp_int, &disp_int_cont, &disp_int_cont2);
+	r600_irq_ack(rdev);
 	r600_disable_interrupt_state(rdev);
 }
 
@@ -3253,7 +3268,7 @@ int r600_irq_process(struct radeon_device *rdev)
 	u32 wptr = r600_get_ih_wptr(rdev);
 	u32 rptr = rdev->ih.rptr;
 	u32 src_id, src_data;
-	u32 ring_index, disp_int, disp_int_cont, disp_int_cont2;
+	u32 ring_index;
 	unsigned long flags;
 	bool queue_hotplug = false;
 
@@ -3274,7 +3289,7 @@ int r600_irq_process(struct radeon_device *rdev)
 
 restart_ih:
 	/* display interrupts */
-	r600_irq_ack(rdev, &disp_int, &disp_int_cont, &disp_int_cont2);
+	r600_irq_ack(rdev);
 
 	rdev->ih.wptr = wptr;
 	while (rptr != wptr) {
@@ -3287,17 +3302,21 @@ restart_ih:
 		case 1: /* D1 vblank/vline */
 			switch (src_data) {
 			case 0: /* D1 vblank */
-				if (disp_int & LB_D1_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 0);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int &= ~LB_D1_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int & LB_D1_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[0]) {
+						drm_handle_vblank(rdev->ddev, 0);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[0])
+						radeon_crtc_handle_flip(rdev, 0);
+					rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
 				break;
 			case 1: /* D1 vline */
-				if (disp_int & LB_D1_VLINE_INTERRUPT) {
-					disp_int &= ~LB_D1_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int & LB_D1_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D1 vline\n");
 				}
 				break;
@@ -3309,17 +3328,21 @@ restart_ih:
 		case 5: /* D2 vblank/vline */
 			switch (src_data) {
 			case 0: /* D2 vblank */
-				if (disp_int & LB_D2_VBLANK_INTERRUPT) {
-					drm_handle_vblank(rdev->ddev, 1);
-					rdev->pm.vblank_sync = true;
-					wake_up(&rdev->irq.vblank_queue);
-					disp_int &= ~LB_D2_VBLANK_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int & LB_D2_VBLANK_INTERRUPT) {
+					if (rdev->irq.crtc_vblank_int[1]) {
+						drm_handle_vblank(rdev->ddev, 1);
+						rdev->pm.vblank_sync = true;
+						wake_up(&rdev->irq.vblank_queue);
+					}
+					if (rdev->irq.pflip[1])
+						radeon_crtc_handle_flip(rdev, 1);
+					rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
 				break;
 			case 1: /* D1 vline */
-				if (disp_int & LB_D2_VLINE_INTERRUPT) {
-					disp_int &= ~LB_D2_VLINE_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int & LB_D2_VLINE_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VLINE_INTERRUPT;
 					DRM_DEBUG("IH: D2 vline\n");
 				}
 				break;
@@ -3331,43 +3354,43 @@ restart_ih:
 		case 19: /* HPD/DAC hotplug */
 			switch (src_data) {
 			case 0:
-				if (disp_int & DC_HPD1_INTERRUPT) {
-					disp_int &= ~DC_HPD1_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int & DC_HPD1_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int &= ~DC_HPD1_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD1\n");
 				}
 				break;
 			case 1:
-				if (disp_int & DC_HPD2_INTERRUPT) {
-					disp_int &= ~DC_HPD2_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int & DC_HPD2_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int &= ~DC_HPD2_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD2\n");
 				}
 				break;
 			case 4:
-				if (disp_int_cont & DC_HPD3_INTERRUPT) {
-					disp_int_cont &= ~DC_HPD3_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD3_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int_cont &= ~DC_HPD3_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD3\n");
 				}
 				break;
 			case 5:
-				if (disp_int_cont & DC_HPD4_INTERRUPT) {
-					disp_int_cont &= ~DC_HPD4_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD4_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int_cont &= ~DC_HPD4_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD4\n");
 				}
 				break;
 			case 10:
-				if (disp_int_cont2 & DC_HPD5_INTERRUPT) {
-					disp_int_cont2 &= ~DC_HPD5_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD5_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int_cont2 &= ~DC_HPD5_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD5\n");
 				}
 				break;
 			case 12:
-				if (disp_int_cont2 & DC_HPD6_INTERRUPT) {
-					disp_int_cont2 &= ~DC_HPD6_INTERRUPT;
+				if (rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD6_INTERRUPT) {
+					rdev->irq.stat_regs.r600.disp_int_cont2 &= ~DC_HPD6_INTERRUPT;
 					queue_hotplug = true;
 					DRM_DEBUG("IH: HPD6\n");
 				}
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index bff4dc4f410f..c89cfa8e0c05 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -728,6 +728,15 @@
 /* DCE 3.2 */
 #       define DC_HPDx_EN                                 (1 << 28)
 
+#define D1GRPH_INTERRUPT_STATUS                           0x6158
+#define D2GRPH_INTERRUPT_STATUS                           0x6958
+#       define DxGRPH_PFLIP_INT_OCCURRED                  (1 << 0)
+#       define DxGRPH_PFLIP_INT_CLEAR                     (1 << 8)
+#define D1GRPH_INTERRUPT_CONTROL                          0x615c
+#define D2GRPH_INTERRUPT_CONTROL                          0x695c
+#       define DxGRPH_PFLIP_INT_MASK                      (1 << 0)
+#       define DxGRPH_PFLIP_INT_TYPE                      (1 << 8)
+
 /*
  * PM4
  */
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 3a7095743d44..431d4186ddf0 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -69,6 +69,7 @@
 #include <ttm/ttm_bo_driver.h>
 #include <ttm/ttm_placement.h>
 #include <ttm/ttm_module.h>
+#include <ttm/ttm_execbuf_util.h>
 
 #include "radeon_family.h"
 #include "radeon_mode.h"
@@ -180,6 +181,7 @@ void rs690_pm_info(struct radeon_device *rdev);
 extern u32 rv6xx_get_temp(struct radeon_device *rdev);
 extern u32 rv770_get_temp(struct radeon_device *rdev);
 extern u32 evergreen_get_temp(struct radeon_device *rdev);
+extern u32 sumo_get_temp(struct radeon_device *rdev);
 
 /*
  * Fences.
@@ -259,13 +261,12 @@ struct radeon_bo {
 };
 
 struct radeon_bo_list {
-	struct list_head	list;
+	struct ttm_validate_buffer tv;
 	struct radeon_bo	*bo;
 	uint64_t		gpu_offset;
 	unsigned		rdomain;
 	unsigned		wdomain;
 	u32			tiling_flags;
-	bool			reserved;
 };
 
 /*
@@ -377,11 +378,56 @@ void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg);
 /*
  * IRQS.
  */
+
+struct radeon_unpin_work {
+	struct work_struct work;
+	struct radeon_device *rdev;
+	int crtc_id;
+	struct radeon_fence *fence;
+	struct drm_pending_vblank_event *event;
+	struct radeon_bo *old_rbo;
+	u64 new_crtc_base;
+};
+
+struct r500_irq_stat_regs {
+	u32 disp_int;
+};
+
+struct r600_irq_stat_regs {
+	u32 disp_int;
+	u32 disp_int_cont;
+	u32 disp_int_cont2;
+	u32 d1grph_int;
+	u32 d2grph_int;
+};
+
+struct evergreen_irq_stat_regs {
+	u32 disp_int;
+	u32 disp_int_cont;
+	u32 disp_int_cont2;
+	u32 disp_int_cont3;
+	u32 disp_int_cont4;
+	u32 disp_int_cont5;
+	u32 d1grph_int;
+	u32 d2grph_int;
+	u32 d3grph_int;
+	u32 d4grph_int;
+	u32 d5grph_int;
+	u32 d6grph_int;
+};
+
+union radeon_irq_stat_regs {
+	struct r500_irq_stat_regs r500;
+	struct r600_irq_stat_regs r600;
+	struct evergreen_irq_stat_regs evergreen;
+};
+
 struct radeon_irq {
 	bool		installed;
 	bool		sw_int;
 	/* FIXME: use a define max crtc rather than hardcode it */
 	bool		crtc_vblank_int[6];
+	bool		pflip[6];
 	wait_queue_head_t	vblank_queue;
 	/* FIXME: use defines for max hpd/dacs */
 	bool            hpd[6];
@@ -392,12 +438,17 @@ struct radeon_irq {
 	bool		hdmi[2];
 	spinlock_t sw_lock;
 	int sw_refcount;
+	union radeon_irq_stat_regs stat_regs;
+	spinlock_t pflip_lock[6];
+	int pflip_refcount[6];
 };
 
 int radeon_irq_kms_init(struct radeon_device *rdev);
 void radeon_irq_kms_fini(struct radeon_device *rdev);
 void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev);
 void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev);
+void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc);
+void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc);
 
 /*
  * CP & ring.
@@ -687,6 +738,7 @@ enum radeon_int_thermal_type {
 	THERMAL_TYPE_RV6XX,
 	THERMAL_TYPE_RV770,
 	THERMAL_TYPE_EVERGREEN,
+	THERMAL_TYPE_SUMO,
 };
 
 struct radeon_voltage {
@@ -881,6 +933,10 @@ struct radeon_asic {
 	void (*pm_finish)(struct radeon_device *rdev);
 	void (*pm_init_profile)(struct radeon_device *rdev);
 	void (*pm_get_dynpm_state)(struct radeon_device *rdev);
+	/* pageflipping */
+	void (*pre_page_flip)(struct radeon_device *rdev, int crtc);
+	u32 (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base);
+	void (*post_page_flip)(struct radeon_device *rdev, int crtc);
 };
 
 /*
@@ -1269,6 +1325,7 @@ void r100_pll_errata_after_index(struct radeon_device *rdev);
 #define ASIC_IS_DCE3(rdev) ((rdev->family >= CHIP_RV620))
 #define ASIC_IS_DCE32(rdev) ((rdev->family >= CHIP_RV730))
 #define ASIC_IS_DCE4(rdev) ((rdev->family >= CHIP_CEDAR))
+#define ASIC_IS_DCE41(rdev) ((rdev->family >= CHIP_PALM))
 
 /*
  * BIOS helpers.
@@ -1344,6 +1401,9 @@ static inline void radeon_ring_write(struct radeon_device *rdev, uint32_t v)
 #define radeon_pm_finish(rdev) (rdev)->asic->pm_finish((rdev))
 #define radeon_pm_init_profile(rdev) (rdev)->asic->pm_init_profile((rdev))
 #define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm_get_dynpm_state((rdev))
+#define radeon_pre_page_flip(rdev, crtc) rdev->asic->pre_page_flip((rdev), (crtc))
+#define radeon_page_flip(rdev, crtc, base) rdev->asic->page_flip((rdev), (crtc), (base))
+#define radeon_post_page_flip(rdev, crtc) rdev->asic->post_page_flip((rdev), (crtc))
 
 /* Common functions */
 /* AGP */
@@ -1432,7 +1492,6 @@ extern void rs690_line_buffer_adjust(struct radeon_device *rdev,
 					struct drm_display_mode *mode2);
 
 /* r600, rv610, rv630, rv620, rv635, rv670, rs780, rs880 */
-extern void r600_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 extern bool r600_card_posted(struct radeon_device *rdev);
 extern void r600_cp_stop(struct radeon_device *rdev);
 extern int r600_cp_start(struct radeon_device *rdev);
@@ -1478,6 +1537,7 @@ extern void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mo
 extern int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder);
 extern void r600_hdmi_update_audio_settings(struct drm_encoder *encoder);
 
+extern void r700_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 extern void r700_cp_stop(struct radeon_device *rdev);
 extern void r700_cp_fini(struct radeon_device *rdev);
 extern void evergreen_disable_interrupt_state(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 64fb89ecbf74..3d73fe484f42 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -171,6 +171,9 @@ static struct radeon_asic r100_asic = {
 	.pm_finish = &r100_pm_finish,
 	.pm_init_profile = &r100_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &r100_pre_page_flip,
+	.page_flip = &r100_page_flip,
+	.post_page_flip = &r100_post_page_flip,
 };
 
 static struct radeon_asic r200_asic = {
@@ -215,6 +218,9 @@ static struct radeon_asic r200_asic = {
 	.pm_finish = &r100_pm_finish,
 	.pm_init_profile = &r100_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &r100_pre_page_flip,
+	.page_flip = &r100_page_flip,
+	.post_page_flip = &r100_post_page_flip,
 };
 
 static struct radeon_asic r300_asic = {
@@ -260,6 +266,9 @@ static struct radeon_asic r300_asic = {
 	.pm_finish = &r100_pm_finish,
 	.pm_init_profile = &r100_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &r100_pre_page_flip,
+	.page_flip = &r100_page_flip,
+	.post_page_flip = &r100_post_page_flip,
 };
 
 static struct radeon_asic r300_asic_pcie = {
@@ -304,6 +313,9 @@ static struct radeon_asic r300_asic_pcie = {
 	.pm_finish = &r100_pm_finish,
 	.pm_init_profile = &r100_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &r100_pre_page_flip,
+	.page_flip = &r100_page_flip,
+	.post_page_flip = &r100_post_page_flip,
 };
 
 static struct radeon_asic r420_asic = {
@@ -349,6 +361,9 @@ static struct radeon_asic r420_asic = {
 	.pm_finish = &r100_pm_finish,
 	.pm_init_profile = &r420_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &r100_pre_page_flip,
+	.page_flip = &r100_page_flip,
+	.post_page_flip = &r100_post_page_flip,
 };
 
 static struct radeon_asic rs400_asic = {
@@ -394,6 +409,9 @@ static struct radeon_asic rs400_asic = {
 	.pm_finish = &r100_pm_finish,
 	.pm_init_profile = &r100_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &r100_pre_page_flip,
+	.page_flip = &r100_page_flip,
+	.post_page_flip = &r100_post_page_flip,
 };
 
 static struct radeon_asic rs600_asic = {
@@ -439,6 +457,9 @@ static struct radeon_asic rs600_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &r420_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rs600_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic rs690_asic = {
@@ -484,6 +505,9 @@ static struct radeon_asic rs690_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &r420_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rs600_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic rv515_asic = {
@@ -529,6 +553,9 @@ static struct radeon_asic rv515_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &r420_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rs600_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic r520_asic = {
@@ -574,6 +601,9 @@ static struct radeon_asic r520_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &r420_pm_init_profile,
 	.pm_get_dynpm_state = &r100_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rs600_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic r600_asic = {
@@ -618,6 +648,9 @@ static struct radeon_asic r600_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &r600_pm_init_profile,
 	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rs600_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic rs780_asic = {
@@ -662,6 +695,9 @@ static struct radeon_asic rs780_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &rs780_pm_init_profile,
 	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rs600_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic rv770_asic = {
@@ -706,6 +742,9 @@ static struct radeon_asic rv770_asic = {
 	.pm_finish = &rs600_pm_finish,
 	.pm_init_profile = &r600_pm_init_profile,
 	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
+	.pre_page_flip = &rs600_pre_page_flip,
+	.page_flip = &rv770_page_flip,
+	.post_page_flip = &rs600_post_page_flip,
 };
 
 static struct radeon_asic evergreen_asic = {
@@ -749,6 +788,52 @@ static struct radeon_asic evergreen_asic = {
 	.pm_finish = &evergreen_pm_finish,
 	.pm_init_profile = &r600_pm_init_profile,
 	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
+	.pre_page_flip = &evergreen_pre_page_flip,
+	.page_flip = &evergreen_page_flip,
+	.post_page_flip = &evergreen_post_page_flip,
+};
+
+static struct radeon_asic sumo_asic = {
+	.init = &evergreen_init,
+	.fini = &evergreen_fini,
+	.suspend = &evergreen_suspend,
+	.resume = &evergreen_resume,
+	.cp_commit = &r600_cp_commit,
+	.gpu_is_lockup = &evergreen_gpu_is_lockup,
+	.asic_reset = &evergreen_asic_reset,
+	.vga_set_state = &r600_vga_set_state,
+	.gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
+	.gart_set_page = &rs600_gart_set_page,
+	.ring_test = &r600_ring_test,
+	.ring_ib_execute = &r600_ring_ib_execute,
+	.irq_set = &evergreen_irq_set,
+	.irq_process = &evergreen_irq_process,
+	.get_vblank_counter = &evergreen_get_vblank_counter,
+	.fence_ring_emit = &r600_fence_ring_emit,
+	.cs_parse = &evergreen_cs_parse,
+	.copy_blit = &evergreen_copy_blit,
+	.copy_dma = &evergreen_copy_blit,
+	.copy = &evergreen_copy_blit,
+	.get_engine_clock = &radeon_atom_get_engine_clock,
+	.set_engine_clock = &radeon_atom_set_engine_clock,
+	.get_memory_clock = NULL,
+	.set_memory_clock = NULL,
+	.get_pcie_lanes = NULL,
+	.set_pcie_lanes = NULL,
+	.set_clock_gating = NULL,
+	.set_surface_reg = r600_set_surface_reg,
+	.clear_surface_reg = r600_clear_surface_reg,
+	.bandwidth_update = &evergreen_bandwidth_update,
+	.hpd_init = &evergreen_hpd_init,
+	.hpd_fini = &evergreen_hpd_fini,
+	.hpd_sense = &evergreen_hpd_sense,
+	.hpd_set_polarity = &evergreen_hpd_set_polarity,
+	.gui_idle = &r600_gui_idle,
+	.pm_misc = &evergreen_pm_misc,
+	.pm_prepare = &evergreen_pm_prepare,
+	.pm_finish = &evergreen_pm_finish,
+	.pm_init_profile = &rs780_pm_init_profile,
+	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
 };
 
 int radeon_asic_init(struct radeon_device *rdev)
@@ -835,6 +920,9 @@ int radeon_asic_init(struct radeon_device *rdev)
 	case CHIP_HEMLOCK:
 		rdev->asic = &evergreen_asic;
 		break;
+	case CHIP_PALM:
+		rdev->asic = &sumo_asic;
+		break;
 	default:
 		/* FIXME: not supported yet */
 		return -EINVAL;
@@ -849,7 +937,9 @@ int radeon_asic_init(struct radeon_device *rdev)
 	if (rdev->flags & RADEON_SINGLE_CRTC)
 		rdev->num_crtc = 1;
 	else {
-		if (ASIC_IS_DCE4(rdev))
+		if (ASIC_IS_DCE41(rdev))
+			rdev->num_crtc = 2;
+		else if (ASIC_IS_DCE4(rdev))
 			rdev->num_crtc = 6;
 		else
 			rdev->num_crtc = 2;
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 740988244143..4970eda1bd41 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -130,6 +130,9 @@ extern void r100_pm_prepare(struct radeon_device *rdev);
 extern void r100_pm_finish(struct radeon_device *rdev);
 extern void r100_pm_init_profile(struct radeon_device *rdev);
 extern void r100_pm_get_dynpm_state(struct radeon_device *rdev);
+extern void r100_pre_page_flip(struct radeon_device *rdev, int crtc);
+extern u32 r100_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+extern void r100_post_page_flip(struct radeon_device *rdev, int crtc);
 
 /*
  * r200,rv250,rs300,rv280
@@ -205,6 +208,9 @@ void rs600_hpd_set_polarity(struct radeon_device *rdev,
 extern void rs600_pm_misc(struct radeon_device *rdev);
 extern void rs600_pm_prepare(struct radeon_device *rdev);
 extern void rs600_pm_finish(struct radeon_device *rdev);
+extern void rs600_pre_page_flip(struct radeon_device *rdev, int crtc);
+extern u32 rs600_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+extern void rs600_post_page_flip(struct radeon_device *rdev, int crtc);
 
 /*
  * rs690,rs740
@@ -287,6 +293,7 @@ void rv770_fini(struct radeon_device *rdev);
 int rv770_suspend(struct radeon_device *rdev);
 int rv770_resume(struct radeon_device *rdev);
 extern void rv770_pm_misc(struct radeon_device *rdev);
+extern u32 rv770_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
 
 /*
  * evergreen
@@ -314,5 +321,8 @@ extern int evergreen_cs_parse(struct radeon_cs_parser *p);
 extern void evergreen_pm_misc(struct radeon_device *rdev);
 extern void evergreen_pm_prepare(struct radeon_device *rdev);
 extern void evergreen_pm_finish(struct radeon_device *rdev);
+extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc);
+extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc);
 
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index bc5a2c3382d9..8e82f672263f 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -1337,6 +1337,43 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev,
 	return false;
 }
 
+static void radeon_atombios_get_igp_ss_overrides(struct radeon_device *rdev,
+						 struct radeon_atom_ss *ss,
+						 int id)
+{
+	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	int index = GetIndexIntoMasterTable(DATA, IntegratedSystemInfo);
+	u16 data_offset, size;
+	struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 *igp_info;
+	u8 frev, crev;
+	u16 percentage = 0, rate = 0;
+
+	/* get any igp specific overrides */
+	if (atom_parse_data_header(mode_info->atom_context, index, &size,
+				   &frev, &crev, &data_offset)) {
+		igp_info = (struct _ATOM_INTEGRATED_SYSTEM_INFO_V6 *)
+			(mode_info->atom_context->bios + data_offset);
+		switch (id) {
+		case ASIC_INTERNAL_SS_ON_TMDS:
+			percentage = le16_to_cpu(igp_info->usDVISSPercentage);
+			rate = le16_to_cpu(igp_info->usDVISSpreadRateIn10Hz);
+			break;
+		case ASIC_INTERNAL_SS_ON_HDMI:
+			percentage = le16_to_cpu(igp_info->usHDMISSPercentage);
+			rate = le16_to_cpu(igp_info->usHDMISSpreadRateIn10Hz);
+			break;
+		case ASIC_INTERNAL_SS_ON_LVDS:
+			percentage = le16_to_cpu(igp_info->usLvdsSSPercentage);
+			rate = le16_to_cpu(igp_info->usLvdsSSpreadRateIn10Hz);
+			break;
+		}
+		if (percentage)
+			ss->percentage = percentage;
+		if (rate)
+			ss->rate = rate;
+	}
+}
+
 union asic_ss_info {
 	struct _ATOM_ASIC_INTERNAL_SS_INFO info;
 	struct _ATOM_ASIC_INTERNAL_SS_INFO_V2 info_2;
@@ -1401,6 +1438,8 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
 						le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
 					ss->type = ss_info->info_3.asSpreadSpectrum[i].ucSpreadSpectrumMode;
 					ss->rate = le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadRateIn10Hz);
+					if (rdev->flags & RADEON_IS_IGP)
+						radeon_atombios_get_igp_ss_overrides(rdev, ss, id);
 					return true;
 				}
 			}
@@ -1740,495 +1779,600 @@ static const char *pp_lib_thermal_controller_names[] = {
 	"RV6xx",
 	"RV770",
 	"adt7473",
+	"NONE",
 	"External GPIO",
 	"Evergreen",
-	"adt7473 with internal",
-
+	"emc2103",
+	"Sumo",
 };
 
 union power_info {
 	struct _ATOM_POWERPLAY_INFO info;
 	struct _ATOM_POWERPLAY_INFO_V2 info_2;
 	struct _ATOM_POWERPLAY_INFO_V3 info_3;
-	struct _ATOM_PPLIB_POWERPLAYTABLE info_4;
+	struct _ATOM_PPLIB_POWERPLAYTABLE pplib;
+	struct _ATOM_PPLIB_POWERPLAYTABLE2 pplib2;
+	struct _ATOM_PPLIB_POWERPLAYTABLE3 pplib3;
 };
 
-void radeon_atombios_get_power_modes(struct radeon_device *rdev)
+union pplib_clock_info {
+	struct _ATOM_PPLIB_R600_CLOCK_INFO r600;
+	struct _ATOM_PPLIB_RS780_CLOCK_INFO rs780;
+	struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO evergreen;
+	struct _ATOM_PPLIB_SUMO_CLOCK_INFO sumo;
+};
+
+union pplib_power_state {
+	struct _ATOM_PPLIB_STATE v1;
+	struct _ATOM_PPLIB_STATE_V2 v2;
+};
+
+static void radeon_atombios_parse_misc_flags_1_3(struct radeon_device *rdev,
+						 int state_index,
+						 u32 misc, u32 misc2)
+{
+	rdev->pm.power_state[state_index].misc = misc;
+	rdev->pm.power_state[state_index].misc2 = misc2;
+	/* order matters! */
+	if (misc & ATOM_PM_MISCINFO_POWER_SAVING_MODE)
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_POWERSAVE;
+	if (misc & ATOM_PM_MISCINFO_DEFAULT_DC_STATE_ENTRY_TRUE)
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_BATTERY;
+	if (misc & ATOM_PM_MISCINFO_DEFAULT_LOW_DC_STATE_ENTRY_TRUE)
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_BATTERY;
+	if (misc & ATOM_PM_MISCINFO_LOAD_BALANCE_EN)
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_BALANCED;
+	if (misc & ATOM_PM_MISCINFO_3D_ACCELERATION_EN) {
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_PERFORMANCE;
+		rdev->pm.power_state[state_index].flags &=
+			~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
+	}
+	if (misc2 & ATOM_PM_MISCINFO2_SYSTEM_AC_LITE_MODE)
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_BALANCED;
+	if (misc & ATOM_PM_MISCINFO_DRIVER_DEFAULT_MODE) {
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_DEFAULT;
+		rdev->pm.default_power_state_index = state_index;
+		rdev->pm.power_state[state_index].default_clock_mode =
+			&rdev->pm.power_state[state_index].clock_info[0];
+	} else if (state_index == 0) {
+		rdev->pm.power_state[state_index].clock_info[0].flags |=
+			RADEON_PM_MODE_NO_DISPLAY;
+	}
+}
+
+static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev)
 {
 	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	u32 misc, misc2 = 0;
+	int num_modes = 0, i;
+	int state_index = 0;
+	struct radeon_i2c_bus_rec i2c_bus;
+	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-	u16 data_offset;
+        u16 data_offset;
 	u8 frev, crev;
-	u32 misc, misc2 = 0, sclk, mclk;
-	union power_info *power_info;
-	struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info;
-	struct _ATOM_PPLIB_STATE *power_state;
-	int num_modes = 0, i, j;
-	int state_index = 0, mode_index = 0;
-	struct radeon_i2c_bus_rec i2c_bus;
-
-	rdev->pm.default_power_state_index = -1;
 
-	if (atom_parse_data_header(mode_info->atom_context, index, NULL,
-				   &frev, &crev, &data_offset)) {
-		power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
-		if (frev < 4) {
-			/* add the i2c bus for thermal/fan chip */
-			if (power_info->info.ucOverdriveThermalController > 0) {
-				DRM_INFO("Possible %s thermal controller at 0x%02x\n",
-					 thermal_controller_names[power_info->info.ucOverdriveThermalController],
-					 power_info->info.ucOverdriveControllerAddress >> 1);
-				i2c_bus = radeon_lookup_i2c_gpio(rdev, power_info->info.ucOverdriveI2cLine);
-				rdev->pm.i2c_bus = radeon_i2c_lookup(rdev, &i2c_bus);
-				if (rdev->pm.i2c_bus) {
-					struct i2c_board_info info = { };
-					const char *name = thermal_controller_names[power_info->info.
-										    ucOverdriveThermalController];
-					info.addr = power_info->info.ucOverdriveControllerAddress >> 1;
-					strlcpy(info.type, name, sizeof(info.type));
-					i2c_new_device(&rdev->pm.i2c_bus->adapter, &info);
-				}
+	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
+				   &frev, &crev, &data_offset))
+		return state_index;
+	power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
+
+	/* add the i2c bus for thermal/fan chip */
+	if (power_info->info.ucOverdriveThermalController > 0) {
+		DRM_INFO("Possible %s thermal controller at 0x%02x\n",
+			 thermal_controller_names[power_info->info.ucOverdriveThermalController],
+			 power_info->info.ucOverdriveControllerAddress >> 1);
+		i2c_bus = radeon_lookup_i2c_gpio(rdev, power_info->info.ucOverdriveI2cLine);
+		rdev->pm.i2c_bus = radeon_i2c_lookup(rdev, &i2c_bus);
+		if (rdev->pm.i2c_bus) {
+			struct i2c_board_info info = { };
+			const char *name = thermal_controller_names[power_info->info.
+								    ucOverdriveThermalController];
+			info.addr = power_info->info.ucOverdriveControllerAddress >> 1;
+			strlcpy(info.type, name, sizeof(info.type));
+			i2c_new_device(&rdev->pm.i2c_bus->adapter, &info);
+		}
+	}
+	num_modes = power_info->info.ucNumOfPowerModeEntries;
+	if (num_modes > ATOM_MAX_NUMBEROF_POWER_BLOCK)
+		num_modes = ATOM_MAX_NUMBEROF_POWER_BLOCK;
+	/* last mode is usually default, array is low to high */
+	for (i = 0; i < num_modes; i++) {
+		rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
+		switch (frev) {
+		case 1:
+			rdev->pm.power_state[state_index].num_clock_modes = 1;
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				le16_to_cpu(power_info->info.asPowerPlayInfo[i].usMemoryClock);
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				le16_to_cpu(power_info->info.asPowerPlayInfo[i].usEngineClock);
+			/* skip invalid modes */
+			if ((rdev->pm.power_state[state_index].clock_info[0].mclk == 0) ||
+			    (rdev->pm.power_state[state_index].clock_info[0].sclk == 0))
+				continue;
+			rdev->pm.power_state[state_index].pcie_lanes =
+				power_info->info.asPowerPlayInfo[i].ucNumPciELanes;
+			misc = le32_to_cpu(power_info->info.asPowerPlayInfo[i].ulMiscInfo);
+			if ((misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_SUPPORT) ||
+			    (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)) {
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type =
+					VOLTAGE_GPIO;
+				rdev->pm.power_state[state_index].clock_info[0].voltage.gpio =
+					radeon_lookup_gpio(rdev,
+							   power_info->info.asPowerPlayInfo[i].ucVoltageDropIndex);
+				if (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)
+					rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
+						true;
+				else
+					rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
+						false;
+			} else if (misc & ATOM_PM_MISCINFO_PROGRAM_VOLTAGE) {
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type =
+					VOLTAGE_VDDC;
+				rdev->pm.power_state[state_index].clock_info[0].voltage.vddc_id =
+					power_info->info.asPowerPlayInfo[i].ucVoltageDropIndex;
 			}
-			num_modes = power_info->info.ucNumOfPowerModeEntries;
-			if (num_modes > ATOM_MAX_NUMBEROF_POWER_BLOCK)
-				num_modes = ATOM_MAX_NUMBEROF_POWER_BLOCK;
-			/* last mode is usually default, array is low to high */
-			for (i = 0; i < num_modes; i++) {
-				rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
-				switch (frev) {
-				case 1:
-					rdev->pm.power_state[state_index].num_clock_modes = 1;
-					rdev->pm.power_state[state_index].clock_info[0].mclk =
-						le16_to_cpu(power_info->info.asPowerPlayInfo[i].usMemoryClock);
-					rdev->pm.power_state[state_index].clock_info[0].sclk =
-						le16_to_cpu(power_info->info.asPowerPlayInfo[i].usEngineClock);
-					/* skip invalid modes */
-					if ((rdev->pm.power_state[state_index].clock_info[0].mclk == 0) ||
-					    (rdev->pm.power_state[state_index].clock_info[0].sclk == 0))
-						continue;
-					rdev->pm.power_state[state_index].pcie_lanes =
-						power_info->info.asPowerPlayInfo[i].ucNumPciELanes;
-					misc = le32_to_cpu(power_info->info.asPowerPlayInfo[i].ulMiscInfo);
-					if ((misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_SUPPORT) ||
-					    (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)) {
-						rdev->pm.power_state[state_index].clock_info[0].voltage.type =
-							VOLTAGE_GPIO;
-						rdev->pm.power_state[state_index].clock_info[0].voltage.gpio =
-							radeon_lookup_gpio(rdev,
-							power_info->info.asPowerPlayInfo[i].ucVoltageDropIndex);
-						if (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)
-							rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
-								true;
-						else
-							rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
-								false;
-					} else if (misc & ATOM_PM_MISCINFO_PROGRAM_VOLTAGE) {
-						rdev->pm.power_state[state_index].clock_info[0].voltage.type =
-							VOLTAGE_VDDC;
-						rdev->pm.power_state[state_index].clock_info[0].voltage.vddc_id =
-							power_info->info.asPowerPlayInfo[i].ucVoltageDropIndex;
-					}
-					rdev->pm.power_state[state_index].flags = RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					rdev->pm.power_state[state_index].misc = misc;
-					/* order matters! */
-					if (misc & ATOM_PM_MISCINFO_POWER_SAVING_MODE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_POWERSAVE;
-					if (misc & ATOM_PM_MISCINFO_DEFAULT_DC_STATE_ENTRY_TRUE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-					if (misc & ATOM_PM_MISCINFO_DEFAULT_LOW_DC_STATE_ENTRY_TRUE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-					if (misc & ATOM_PM_MISCINFO_LOAD_BALANCE_EN)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BALANCED;
-					if (misc & ATOM_PM_MISCINFO_3D_ACCELERATION_EN) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_PERFORMANCE;
-						rdev->pm.power_state[state_index].flags &=
-							~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					}
-					if (misc & ATOM_PM_MISCINFO_DRIVER_DEFAULT_MODE) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_DEFAULT;
-						rdev->pm.default_power_state_index = state_index;
-						rdev->pm.power_state[state_index].default_clock_mode =
-							&rdev->pm.power_state[state_index].clock_info[0];
-						rdev->pm.power_state[state_index].flags &=
-							~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					} else if (state_index == 0) {
-						rdev->pm.power_state[state_index].clock_info[0].flags |=
-							RADEON_PM_MODE_NO_DISPLAY;
-					}
-					state_index++;
-					break;
-				case 2:
-					rdev->pm.power_state[state_index].num_clock_modes = 1;
-					rdev->pm.power_state[state_index].clock_info[0].mclk =
-						le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMemoryClock);
-					rdev->pm.power_state[state_index].clock_info[0].sclk =
-						le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulEngineClock);
-					/* skip invalid modes */
-					if ((rdev->pm.power_state[state_index].clock_info[0].mclk == 0) ||
-					    (rdev->pm.power_state[state_index].clock_info[0].sclk == 0))
-						continue;
-					rdev->pm.power_state[state_index].pcie_lanes =
-						power_info->info_2.asPowerPlayInfo[i].ucNumPciELanes;
-					misc = le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMiscInfo);
-					misc2 = le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMiscInfo2);
-					if ((misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_SUPPORT) ||
-					    (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)) {
-						rdev->pm.power_state[state_index].clock_info[0].voltage.type =
-							VOLTAGE_GPIO;
-						rdev->pm.power_state[state_index].clock_info[0].voltage.gpio =
-							radeon_lookup_gpio(rdev,
-							power_info->info_2.asPowerPlayInfo[i].ucVoltageDropIndex);
-						if (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)
-							rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
-								true;
-						else
-							rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
-								false;
-					} else if (misc & ATOM_PM_MISCINFO_PROGRAM_VOLTAGE) {
-						rdev->pm.power_state[state_index].clock_info[0].voltage.type =
-							VOLTAGE_VDDC;
-						rdev->pm.power_state[state_index].clock_info[0].voltage.vddc_id =
-							power_info->info_2.asPowerPlayInfo[i].ucVoltageDropIndex;
-					}
-					rdev->pm.power_state[state_index].flags = RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					rdev->pm.power_state[state_index].misc = misc;
-					rdev->pm.power_state[state_index].misc2 = misc2;
-					/* order matters! */
-					if (misc & ATOM_PM_MISCINFO_POWER_SAVING_MODE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_POWERSAVE;
-					if (misc & ATOM_PM_MISCINFO_DEFAULT_DC_STATE_ENTRY_TRUE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-					if (misc & ATOM_PM_MISCINFO_DEFAULT_LOW_DC_STATE_ENTRY_TRUE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-					if (misc & ATOM_PM_MISCINFO_LOAD_BALANCE_EN)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BALANCED;
-					if (misc & ATOM_PM_MISCINFO_3D_ACCELERATION_EN) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_PERFORMANCE;
-						rdev->pm.power_state[state_index].flags &=
-							~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					}
-					if (misc2 & ATOM_PM_MISCINFO2_SYSTEM_AC_LITE_MODE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BALANCED;
-					if (misc2 & ATOM_PM_MISCINFO2_MULTI_DISPLAY_SUPPORT)
-						rdev->pm.power_state[state_index].flags &=
-							~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					if (misc & ATOM_PM_MISCINFO_DRIVER_DEFAULT_MODE) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_DEFAULT;
-						rdev->pm.default_power_state_index = state_index;
-						rdev->pm.power_state[state_index].default_clock_mode =
-							&rdev->pm.power_state[state_index].clock_info[0];
-						rdev->pm.power_state[state_index].flags &=
-							~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					} else if (state_index == 0) {
-						rdev->pm.power_state[state_index].clock_info[0].flags |=
-							RADEON_PM_MODE_NO_DISPLAY;
-					}
-					state_index++;
-					break;
-				case 3:
-					rdev->pm.power_state[state_index].num_clock_modes = 1;
-					rdev->pm.power_state[state_index].clock_info[0].mclk =
-						le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMemoryClock);
-					rdev->pm.power_state[state_index].clock_info[0].sclk =
-						le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulEngineClock);
-					/* skip invalid modes */
-					if ((rdev->pm.power_state[state_index].clock_info[0].mclk == 0) ||
-					    (rdev->pm.power_state[state_index].clock_info[0].sclk == 0))
-						continue;
-					rdev->pm.power_state[state_index].pcie_lanes =
-						power_info->info_3.asPowerPlayInfo[i].ucNumPciELanes;
-					misc = le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMiscInfo);
-					misc2 = le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMiscInfo2);
-					if ((misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_SUPPORT) ||
-					    (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)) {
-						rdev->pm.power_state[state_index].clock_info[0].voltage.type =
-							VOLTAGE_GPIO;
-						rdev->pm.power_state[state_index].clock_info[0].voltage.gpio =
-							radeon_lookup_gpio(rdev,
-							power_info->info_3.asPowerPlayInfo[i].ucVoltageDropIndex);
-						if (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)
-							rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
-								true;
-						else
-							rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
-								false;
-					} else if (misc & ATOM_PM_MISCINFO_PROGRAM_VOLTAGE) {
-						rdev->pm.power_state[state_index].clock_info[0].voltage.type =
-							VOLTAGE_VDDC;
-						rdev->pm.power_state[state_index].clock_info[0].voltage.vddc_id =
-							power_info->info_3.asPowerPlayInfo[i].ucVoltageDropIndex;
-						if (misc2 & ATOM_PM_MISCINFO2_VDDCI_DYNAMIC_VOLTAGE_EN) {
-							rdev->pm.power_state[state_index].clock_info[0].voltage.vddci_enabled =
-								true;
-							rdev->pm.power_state[state_index].clock_info[0].voltage.vddci_id =
-							power_info->info_3.asPowerPlayInfo[i].ucVDDCI_VoltageDropIndex;
-						}
-					}
-					rdev->pm.power_state[state_index].flags = RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					rdev->pm.power_state[state_index].misc = misc;
-					rdev->pm.power_state[state_index].misc2 = misc2;
-					/* order matters! */
-					if (misc & ATOM_PM_MISCINFO_POWER_SAVING_MODE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_POWERSAVE;
-					if (misc & ATOM_PM_MISCINFO_DEFAULT_DC_STATE_ENTRY_TRUE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-					if (misc & ATOM_PM_MISCINFO_DEFAULT_LOW_DC_STATE_ENTRY_TRUE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-					if (misc & ATOM_PM_MISCINFO_LOAD_BALANCE_EN)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BALANCED;
-					if (misc & ATOM_PM_MISCINFO_3D_ACCELERATION_EN) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_PERFORMANCE;
-						rdev->pm.power_state[state_index].flags &=
-							~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					}
-					if (misc2 & ATOM_PM_MISCINFO2_SYSTEM_AC_LITE_MODE)
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BALANCED;
-					if (misc & ATOM_PM_MISCINFO_DRIVER_DEFAULT_MODE) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_DEFAULT;
-						rdev->pm.default_power_state_index = state_index;
-						rdev->pm.power_state[state_index].default_clock_mode =
-							&rdev->pm.power_state[state_index].clock_info[0];
-					} else if (state_index == 0) {
-						rdev->pm.power_state[state_index].clock_info[0].flags |=
-							RADEON_PM_MODE_NO_DISPLAY;
-					}
-					state_index++;
-					break;
-				}
+			rdev->pm.power_state[state_index].flags = RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
+			radeon_atombios_parse_misc_flags_1_3(rdev, state_index, misc, 0);
+			state_index++;
+			break;
+		case 2:
+			rdev->pm.power_state[state_index].num_clock_modes = 1;
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMemoryClock);
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulEngineClock);
+			/* skip invalid modes */
+			if ((rdev->pm.power_state[state_index].clock_info[0].mclk == 0) ||
+			    (rdev->pm.power_state[state_index].clock_info[0].sclk == 0))
+				continue;
+			rdev->pm.power_state[state_index].pcie_lanes =
+				power_info->info_2.asPowerPlayInfo[i].ucNumPciELanes;
+			misc = le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMiscInfo);
+			misc2 = le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMiscInfo2);
+			if ((misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_SUPPORT) ||
+			    (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)) {
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type =
+					VOLTAGE_GPIO;
+				rdev->pm.power_state[state_index].clock_info[0].voltage.gpio =
+					radeon_lookup_gpio(rdev,
+							   power_info->info_2.asPowerPlayInfo[i].ucVoltageDropIndex);
+				if (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)
+					rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
+						true;
+				else
+					rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
+						false;
+			} else if (misc & ATOM_PM_MISCINFO_PROGRAM_VOLTAGE) {
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type =
+					VOLTAGE_VDDC;
+				rdev->pm.power_state[state_index].clock_info[0].voltage.vddc_id =
+					power_info->info_2.asPowerPlayInfo[i].ucVoltageDropIndex;
 			}
-			/* last mode is usually default */
-			if (rdev->pm.default_power_state_index == -1) {
-				rdev->pm.power_state[state_index - 1].type =
-					POWER_STATE_TYPE_DEFAULT;
-				rdev->pm.default_power_state_index = state_index - 1;
-				rdev->pm.power_state[state_index - 1].default_clock_mode =
-					&rdev->pm.power_state[state_index - 1].clock_info[0];
-				rdev->pm.power_state[state_index].flags &=
-					~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-				rdev->pm.power_state[state_index].misc = 0;
-				rdev->pm.power_state[state_index].misc2 = 0;
+			rdev->pm.power_state[state_index].flags = RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
+			radeon_atombios_parse_misc_flags_1_3(rdev, state_index, misc, misc2);
+			state_index++;
+			break;
+		case 3:
+			rdev->pm.power_state[state_index].num_clock_modes = 1;
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMemoryClock);
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulEngineClock);
+			/* skip invalid modes */
+			if ((rdev->pm.power_state[state_index].clock_info[0].mclk == 0) ||
+			    (rdev->pm.power_state[state_index].clock_info[0].sclk == 0))
+				continue;
+			rdev->pm.power_state[state_index].pcie_lanes =
+				power_info->info_3.asPowerPlayInfo[i].ucNumPciELanes;
+			misc = le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMiscInfo);
+			misc2 = le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMiscInfo2);
+			if ((misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_SUPPORT) ||
+			    (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)) {
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type =
+					VOLTAGE_GPIO;
+				rdev->pm.power_state[state_index].clock_info[0].voltage.gpio =
+					radeon_lookup_gpio(rdev,
+							   power_info->info_3.asPowerPlayInfo[i].ucVoltageDropIndex);
+				if (misc & ATOM_PM_MISCINFO_VOLTAGE_DROP_ACTIVE_HIGH)
+					rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
+						true;
+				else
+					rdev->pm.power_state[state_index].clock_info[0].voltage.active_high =
+						false;
+			} else if (misc & ATOM_PM_MISCINFO_PROGRAM_VOLTAGE) {
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type =
+					VOLTAGE_VDDC;
+				rdev->pm.power_state[state_index].clock_info[0].voltage.vddc_id =
+					power_info->info_3.asPowerPlayInfo[i].ucVoltageDropIndex;
+				if (misc2 & ATOM_PM_MISCINFO2_VDDCI_DYNAMIC_VOLTAGE_EN) {
+					rdev->pm.power_state[state_index].clock_info[0].voltage.vddci_enabled =
+						true;
+					rdev->pm.power_state[state_index].clock_info[0].voltage.vddci_id =
+						power_info->info_3.asPowerPlayInfo[i].ucVDDCI_VoltageDropIndex;
+				}
 			}
+			rdev->pm.power_state[state_index].flags = RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
+			radeon_atombios_parse_misc_flags_1_3(rdev, state_index, misc, misc2);
+			state_index++;
+			break;
+		}
+	}
+	/* last mode is usually default */
+	if (rdev->pm.default_power_state_index == -1) {
+		rdev->pm.power_state[state_index - 1].type =
+			POWER_STATE_TYPE_DEFAULT;
+		rdev->pm.default_power_state_index = state_index - 1;
+		rdev->pm.power_state[state_index - 1].default_clock_mode =
+			&rdev->pm.power_state[state_index - 1].clock_info[0];
+		rdev->pm.power_state[state_index].flags &=
+			~RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
+		rdev->pm.power_state[state_index].misc = 0;
+		rdev->pm.power_state[state_index].misc2 = 0;
+	}
+	return state_index;
+}
+
+static void radeon_atombios_add_pplib_thermal_controller(struct radeon_device *rdev,
+							 ATOM_PPLIB_THERMALCONTROLLER *controller)
+{
+	struct radeon_i2c_bus_rec i2c_bus;
+
+	/* add the i2c bus for thermal/fan chip */
+	if (controller->ucType > 0) {
+		if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) {
+			DRM_INFO("Internal thermal controller %s fan control\n",
+				 (controller->ucFanParameters &
+				  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
+			rdev->pm.int_thermal_type = THERMAL_TYPE_RV6XX;
+		} else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) {
+			DRM_INFO("Internal thermal controller %s fan control\n",
+				 (controller->ucFanParameters &
+				  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
+			rdev->pm.int_thermal_type = THERMAL_TYPE_RV770;
+		} else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EVERGREEN) {
+			DRM_INFO("Internal thermal controller %s fan control\n",
+				 (controller->ucFanParameters &
+				  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
+			rdev->pm.int_thermal_type = THERMAL_TYPE_EVERGREEN;
+		} else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SUMO) {
+			DRM_INFO("Internal thermal controller %s fan control\n",
+				 (controller->ucFanParameters &
+				  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
+			rdev->pm.int_thermal_type = THERMAL_TYPE_SUMO;
+		} else if ((controller->ucType ==
+			    ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) ||
+			   (controller->ucType ==
+			    ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL) ||
+			   (controller->ucType ==
+			    ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL)) {
+			DRM_INFO("Special thermal controller config\n");
 		} else {
-			int fw_index = GetIndexIntoMasterTable(DATA, FirmwareInfo);
-			uint8_t fw_frev, fw_crev;
-			uint16_t fw_data_offset, vddc = 0;
-			union firmware_info *firmware_info;
-			ATOM_PPLIB_THERMALCONTROLLER *controller = &power_info->info_4.sThermalController;
-
-			if (atom_parse_data_header(mode_info->atom_context, fw_index, NULL,
-						   &fw_frev, &fw_crev, &fw_data_offset)) {
-				firmware_info =
-					(union firmware_info *)(mode_info->atom_context->bios +
-								fw_data_offset);
-				vddc = firmware_info->info_14.usBootUpVDDCVoltage;
+			DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n",
+				 pp_lib_thermal_controller_names[controller->ucType],
+				 controller->ucI2cAddress >> 1,
+				 (controller->ucFanParameters &
+				  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
+			i2c_bus = radeon_lookup_i2c_gpio(rdev, controller->ucI2cLine);
+			rdev->pm.i2c_bus = radeon_i2c_lookup(rdev, &i2c_bus);
+			if (rdev->pm.i2c_bus) {
+				struct i2c_board_info info = { };
+				const char *name = pp_lib_thermal_controller_names[controller->ucType];
+				info.addr = controller->ucI2cAddress >> 1;
+				strlcpy(info.type, name, sizeof(info.type));
+				i2c_new_device(&rdev->pm.i2c_bus->adapter, &info);
 			}
+		}
+	}
+}
 
-			/* add the i2c bus for thermal/fan chip */
-			if (controller->ucType > 0) {
-				if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) {
-					DRM_INFO("Internal thermal controller %s fan control\n",
-						 (controller->ucFanParameters &
-						  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
-					rdev->pm.int_thermal_type = THERMAL_TYPE_RV6XX;
-				} else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) {
-					DRM_INFO("Internal thermal controller %s fan control\n",
-						 (controller->ucFanParameters &
-						  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
-					rdev->pm.int_thermal_type = THERMAL_TYPE_RV770;
-				} else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EVERGREEN) {
-					DRM_INFO("Internal thermal controller %s fan control\n",
-						 (controller->ucFanParameters &
-						  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
-					rdev->pm.int_thermal_type = THERMAL_TYPE_EVERGREEN;
-				} else if ((controller->ucType ==
-					    ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) ||
-					   (controller->ucType ==
-					    ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL)) {
-					DRM_INFO("Special thermal controller config\n");
-				} else {
-					DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n",
-						 pp_lib_thermal_controller_names[controller->ucType],
-						 controller->ucI2cAddress >> 1,
-						 (controller->ucFanParameters &
-						  ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with");
-					i2c_bus = radeon_lookup_i2c_gpio(rdev, controller->ucI2cLine);
-					rdev->pm.i2c_bus = radeon_i2c_lookup(rdev, &i2c_bus);
-					if (rdev->pm.i2c_bus) {
-						struct i2c_board_info info = { };
-						const char *name = pp_lib_thermal_controller_names[controller->ucType];
-						info.addr = controller->ucI2cAddress >> 1;
-						strlcpy(info.type, name, sizeof(info.type));
-						i2c_new_device(&rdev->pm.i2c_bus->adapter, &info);
-					}
+static u16 radeon_atombios_get_default_vddc(struct radeon_device *rdev)
+{
+	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	int index = GetIndexIntoMasterTable(DATA, FirmwareInfo);
+	u8 frev, crev;
+	u16 data_offset;
+	union firmware_info *firmware_info;
+	u16 vddc = 0;
 
-				}
-			}
-			/* first mode is usually default, followed by low to high */
-			for (i = 0; i < power_info->info_4.ucNumStates; i++) {
-				mode_index = 0;
-				power_state = (struct _ATOM_PPLIB_STATE *)
-					(mode_info->atom_context->bios +
-					 data_offset +
-					 le16_to_cpu(power_info->info_4.usStateArrayOffset) +
-					 i * power_info->info_4.ucStateEntrySize);
-				non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
-					(mode_info->atom_context->bios +
-					 data_offset +
-					 le16_to_cpu(power_info->info_4.usNonClockInfoArrayOffset) +
-					 (power_state->ucNonClockStateIndex *
-					  power_info->info_4.ucNonClockSize));
-				for (j = 0; j < (power_info->info_4.ucStateEntrySize - 1); j++) {
-					if (rdev->flags & RADEON_IS_IGP) {
-						struct _ATOM_PPLIB_RS780_CLOCK_INFO *clock_info =
-							(struct _ATOM_PPLIB_RS780_CLOCK_INFO *)
-							(mode_info->atom_context->bios +
-							 data_offset +
-							 le16_to_cpu(power_info->info_4.usClockInfoArrayOffset) +
-							 (power_state->ucClockStateIndices[j] *
-							  power_info->info_4.ucClockInfoSize));
-						sclk = le16_to_cpu(clock_info->usLowEngineClockLow);
-						sclk |= clock_info->ucLowEngineClockHigh << 16;
-						rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
-						/* skip invalid modes */
-						if (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0)
-							continue;
-						/* voltage works differently on IGPs */
-						mode_index++;
-					} else if (ASIC_IS_DCE4(rdev)) {
-						struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO *clock_info =
-							(struct _ATOM_PPLIB_EVERGREEN_CLOCK_INFO *)
-							(mode_info->atom_context->bios +
-							 data_offset +
-							 le16_to_cpu(power_info->info_4.usClockInfoArrayOffset) +
-							 (power_state->ucClockStateIndices[j] *
-							  power_info->info_4.ucClockInfoSize));
-						sclk = le16_to_cpu(clock_info->usEngineClockLow);
-						sclk |= clock_info->ucEngineClockHigh << 16;
-						mclk = le16_to_cpu(clock_info->usMemoryClockLow);
-						mclk |= clock_info->ucMemoryClockHigh << 16;
-						rdev->pm.power_state[state_index].clock_info[mode_index].mclk = mclk;
-						rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
-						/* skip invalid modes */
-						if ((rdev->pm.power_state[state_index].clock_info[mode_index].mclk == 0) ||
-						    (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0))
-							continue;
-						rdev->pm.power_state[state_index].clock_info[mode_index].voltage.type =
-							VOLTAGE_SW;
-						rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage =
-							clock_info->usVDDC;
-						/* XXX usVDDCI */
-						mode_index++;
-					} else {
-						struct _ATOM_PPLIB_R600_CLOCK_INFO *clock_info =
-							(struct _ATOM_PPLIB_R600_CLOCK_INFO *)
-							(mode_info->atom_context->bios +
-							 data_offset +
-							 le16_to_cpu(power_info->info_4.usClockInfoArrayOffset) +
-							 (power_state->ucClockStateIndices[j] *
-							  power_info->info_4.ucClockInfoSize));
-						sclk = le16_to_cpu(clock_info->usEngineClockLow);
-						sclk |= clock_info->ucEngineClockHigh << 16;
-						mclk = le16_to_cpu(clock_info->usMemoryClockLow);
-						mclk |= clock_info->ucMemoryClockHigh << 16;
-						rdev->pm.power_state[state_index].clock_info[mode_index].mclk = mclk;
-						rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
-						/* skip invalid modes */
-						if ((rdev->pm.power_state[state_index].clock_info[mode_index].mclk == 0) ||
-						    (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0))
-							continue;
-						rdev->pm.power_state[state_index].clock_info[mode_index].voltage.type =
-							VOLTAGE_SW;
-						rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage =
-							clock_info->usVDDC;
-						mode_index++;
-					}
-				}
-				rdev->pm.power_state[state_index].num_clock_modes = mode_index;
-				if (mode_index) {
-					misc = le32_to_cpu(non_clock_info->ulCapsAndSettings);
-					misc2 = le16_to_cpu(non_clock_info->usClassification);
-					rdev->pm.power_state[state_index].misc = misc;
-					rdev->pm.power_state[state_index].misc2 = misc2;
-					rdev->pm.power_state[state_index].pcie_lanes =
-						((misc & ATOM_PPLIB_PCIE_LINK_WIDTH_MASK) >>
-						ATOM_PPLIB_PCIE_LINK_WIDTH_SHIFT) + 1;
-					switch (misc2 & ATOM_PPLIB_CLASSIFICATION_UI_MASK) {
-					case ATOM_PPLIB_CLASSIFICATION_UI_BATTERY:
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BATTERY;
-						break;
-					case ATOM_PPLIB_CLASSIFICATION_UI_BALANCED:
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_BALANCED;
-						break;
-					case ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE:
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_PERFORMANCE;
-						break;
-					case ATOM_PPLIB_CLASSIFICATION_UI_NONE:
-						if (misc2 & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE)
-							rdev->pm.power_state[state_index].type =
-								POWER_STATE_TYPE_PERFORMANCE;
-						break;
-					}
-					rdev->pm.power_state[state_index].flags = 0;
-					if (misc & ATOM_PPLIB_SINGLE_DISPLAY_ONLY)
-						rdev->pm.power_state[state_index].flags |=
-							RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
-					if (misc2 & ATOM_PPLIB_CLASSIFICATION_BOOT) {
-						rdev->pm.power_state[state_index].type =
-							POWER_STATE_TYPE_DEFAULT;
-						rdev->pm.default_power_state_index = state_index;
-						rdev->pm.power_state[state_index].default_clock_mode =
-							&rdev->pm.power_state[state_index].clock_info[mode_index - 1];
-						/* patch the table values with the default slck/mclk from firmware info */
-						for (j = 0; j < mode_index; j++) {
-							rdev->pm.power_state[state_index].clock_info[j].mclk =
-								rdev->clock.default_mclk;
-							rdev->pm.power_state[state_index].clock_info[j].sclk =
-								rdev->clock.default_sclk;
-							if (vddc)
-								rdev->pm.power_state[state_index].clock_info[j].voltage.voltage =
-									vddc;
-						}
-					}
-					state_index++;
-				}
-			}
-			/* if multiple clock modes, mark the lowest as no display */
-			for (i = 0; i < state_index; i++) {
-				if (rdev->pm.power_state[i].num_clock_modes > 1)
-					rdev->pm.power_state[i].clock_info[0].flags |=
-						RADEON_PM_MODE_NO_DISPLAY;
-			}
-			/* first mode is usually default */
-			if (rdev->pm.default_power_state_index == -1) {
-				rdev->pm.power_state[0].type =
-					POWER_STATE_TYPE_DEFAULT;
-				rdev->pm.default_power_state_index = 0;
-				rdev->pm.power_state[0].default_clock_mode =
-					&rdev->pm.power_state[0].clock_info[0];
-			}
+	if (atom_parse_data_header(mode_info->atom_context, index, NULL,
+				   &frev, &crev, &data_offset)) {
+		firmware_info =
+			(union firmware_info *)(mode_info->atom_context->bios +
+						data_offset);
+		vddc = firmware_info->info_14.usBootUpVDDCVoltage;
+	}
+
+	return vddc;
+}
+
+static void radeon_atombios_parse_pplib_non_clock_info(struct radeon_device *rdev,
+						       int state_index, int mode_index,
+						       struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info)
+{
+	int j;
+	u32 misc = le32_to_cpu(non_clock_info->ulCapsAndSettings);
+	u32 misc2 = le16_to_cpu(non_clock_info->usClassification);
+	u16 vddc = radeon_atombios_get_default_vddc(rdev);
+
+	rdev->pm.power_state[state_index].misc = misc;
+	rdev->pm.power_state[state_index].misc2 = misc2;
+	rdev->pm.power_state[state_index].pcie_lanes =
+		((misc & ATOM_PPLIB_PCIE_LINK_WIDTH_MASK) >>
+		 ATOM_PPLIB_PCIE_LINK_WIDTH_SHIFT) + 1;
+	switch (misc2 & ATOM_PPLIB_CLASSIFICATION_UI_MASK) {
+	case ATOM_PPLIB_CLASSIFICATION_UI_BATTERY:
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_BATTERY;
+		break;
+	case ATOM_PPLIB_CLASSIFICATION_UI_BALANCED:
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_BALANCED;
+		break;
+	case ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE:
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_PERFORMANCE;
+		break;
+	case ATOM_PPLIB_CLASSIFICATION_UI_NONE:
+		if (misc2 & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE)
+			rdev->pm.power_state[state_index].type =
+				POWER_STATE_TYPE_PERFORMANCE;
+		break;
+	}
+	rdev->pm.power_state[state_index].flags = 0;
+	if (misc & ATOM_PPLIB_SINGLE_DISPLAY_ONLY)
+		rdev->pm.power_state[state_index].flags |=
+			RADEON_PM_STATE_SINGLE_DISPLAY_ONLY;
+	if (misc2 & ATOM_PPLIB_CLASSIFICATION_BOOT) {
+		rdev->pm.power_state[state_index].type =
+			POWER_STATE_TYPE_DEFAULT;
+		rdev->pm.default_power_state_index = state_index;
+		rdev->pm.power_state[state_index].default_clock_mode =
+			&rdev->pm.power_state[state_index].clock_info[mode_index - 1];
+		/* patch the table values with the default slck/mclk from firmware info */
+		for (j = 0; j < mode_index; j++) {
+			rdev->pm.power_state[state_index].clock_info[j].mclk =
+				rdev->clock.default_mclk;
+			rdev->pm.power_state[state_index].clock_info[j].sclk =
+				rdev->clock.default_sclk;
+			if (vddc)
+				rdev->pm.power_state[state_index].clock_info[j].voltage.voltage =
+					vddc;
+		}
+	}
+}
+
+static bool radeon_atombios_parse_pplib_clock_info(struct radeon_device *rdev,
+						   int state_index, int mode_index,
+						   union pplib_clock_info *clock_info)
+{
+	u32 sclk, mclk;
+
+	if (rdev->flags & RADEON_IS_IGP) {
+		if (rdev->family >= CHIP_PALM) {
+			sclk = le16_to_cpu(clock_info->sumo.usEngineClockLow);
+			sclk |= clock_info->sumo.ucEngineClockHigh << 16;
+			rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
+		} else {
+			sclk = le16_to_cpu(clock_info->rs780.usLowEngineClockLow);
+			sclk |= clock_info->rs780.ucLowEngineClockHigh << 16;
+			rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
+		}
+	} else if (ASIC_IS_DCE4(rdev)) {
+		sclk = le16_to_cpu(clock_info->evergreen.usEngineClockLow);
+		sclk |= clock_info->evergreen.ucEngineClockHigh << 16;
+		mclk = le16_to_cpu(clock_info->evergreen.usMemoryClockLow);
+		mclk |= clock_info->evergreen.ucMemoryClockHigh << 16;
+		rdev->pm.power_state[state_index].clock_info[mode_index].mclk = mclk;
+		rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
+		rdev->pm.power_state[state_index].clock_info[mode_index].voltage.type =
+			VOLTAGE_SW;
+		rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage =
+			clock_info->evergreen.usVDDC;
+	} else {
+		sclk = le16_to_cpu(clock_info->r600.usEngineClockLow);
+		sclk |= clock_info->r600.ucEngineClockHigh << 16;
+		mclk = le16_to_cpu(clock_info->r600.usMemoryClockLow);
+		mclk |= clock_info->r600.ucMemoryClockHigh << 16;
+		rdev->pm.power_state[state_index].clock_info[mode_index].mclk = mclk;
+		rdev->pm.power_state[state_index].clock_info[mode_index].sclk = sclk;
+		rdev->pm.power_state[state_index].clock_info[mode_index].voltage.type =
+			VOLTAGE_SW;
+		rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage =
+			clock_info->r600.usVDDC;
+	}
+
+	if (rdev->flags & RADEON_IS_IGP) {
+		/* skip invalid modes */
+		if (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0)
+			return false;
+	} else {
+		/* skip invalid modes */
+		if ((rdev->pm.power_state[state_index].clock_info[mode_index].mclk == 0) ||
+		    (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0))
+			return false;
+	}
+	return true;
+}
+
+static int radeon_atombios_parse_power_table_4_5(struct radeon_device *rdev)
+{
+	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info;
+	union pplib_power_state *power_state;
+	int i, j;
+	int state_index = 0, mode_index = 0;
+	union pplib_clock_info *clock_info;
+	bool valid;
+	union power_info *power_info;
+	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
+        u16 data_offset;
+	u8 frev, crev;
+
+	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
+				   &frev, &crev, &data_offset))
+		return state_index;
+	power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
+
+	radeon_atombios_add_pplib_thermal_controller(rdev, &power_info->pplib.sThermalController);
+	/* first mode is usually default, followed by low to high */
+	for (i = 0; i < power_info->pplib.ucNumStates; i++) {
+		mode_index = 0;
+		power_state = (union pplib_power_state *)
+			(mode_info->atom_context->bios + data_offset +
+			 le16_to_cpu(power_info->pplib.usStateArrayOffset) +
+			 i * power_info->pplib.ucStateEntrySize);
+		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
+			(mode_info->atom_context->bios + data_offset +
+			 le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset) +
+			 (power_state->v1.ucNonClockStateIndex *
+			  power_info->pplib.ucNonClockSize));
+		for (j = 0; j < (power_info->pplib.ucStateEntrySize - 1); j++) {
+			clock_info = (union pplib_clock_info *)
+				(mode_info->atom_context->bios + data_offset +
+				 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset) +
+				 (power_state->v1.ucClockStateIndices[j] *
+				  power_info->pplib.ucClockInfoSize));
+			valid = radeon_atombios_parse_pplib_clock_info(rdev,
+								       state_index, mode_index,
+								       clock_info);
+			if (valid)
+				mode_index++;
+		}
+		rdev->pm.power_state[state_index].num_clock_modes = mode_index;
+		if (mode_index) {
+			radeon_atombios_parse_pplib_non_clock_info(rdev, state_index, mode_index,
+								   non_clock_info);
+			state_index++;
+		}
+	}
+	/* if multiple clock modes, mark the lowest as no display */
+	for (i = 0; i < state_index; i++) {
+		if (rdev->pm.power_state[i].num_clock_modes > 1)
+			rdev->pm.power_state[i].clock_info[0].flags |=
+				RADEON_PM_MODE_NO_DISPLAY;
+	}
+	/* first mode is usually default */
+	if (rdev->pm.default_power_state_index == -1) {
+		rdev->pm.power_state[0].type =
+			POWER_STATE_TYPE_DEFAULT;
+		rdev->pm.default_power_state_index = 0;
+		rdev->pm.power_state[0].default_clock_mode =
+			&rdev->pm.power_state[0].clock_info[0];
+	}
+	return state_index;
+}
+
+static int radeon_atombios_parse_power_table_6(struct radeon_device *rdev)
+{
+	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	struct _ATOM_PPLIB_NONCLOCK_INFO *non_clock_info;
+	union pplib_power_state *power_state;
+	int i, j, non_clock_array_index, clock_array_index;
+	int state_index = 0, mode_index = 0;
+	union pplib_clock_info *clock_info;
+	struct StateArray *state_array;
+	struct ClockInfoArray *clock_info_array;
+	struct NonClockInfoArray *non_clock_info_array;
+	bool valid;
+	union power_info *power_info;
+	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
+        u16 data_offset;
+	u8 frev, crev;
+
+	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
+				   &frev, &crev, &data_offset))
+		return state_index;
+	power_info = (union power_info *)(mode_info->atom_context->bios + data_offset);
+
+	radeon_atombios_add_pplib_thermal_controller(rdev, &power_info->pplib.sThermalController);
+	state_array = (struct StateArray *)
+		(mode_info->atom_context->bios + data_offset +
+		 power_info->pplib.usStateArrayOffset);
+	clock_info_array = (struct ClockInfoArray *)
+		(mode_info->atom_context->bios + data_offset +
+		 power_info->pplib.usClockInfoArrayOffset);
+	non_clock_info_array = (struct NonClockInfoArray *)
+		(mode_info->atom_context->bios + data_offset +
+		 power_info->pplib.usNonClockInfoArrayOffset);
+	for (i = 0; i < state_array->ucNumEntries; i++) {
+		mode_index = 0;
+		power_state = (union pplib_power_state *)&state_array->states[i];
+		/* XXX this might be an inagua bug... */
+		non_clock_array_index = i; /* power_state->v2.nonClockInfoIndex */
+		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
+			&non_clock_info_array->nonClockInfo[non_clock_array_index];
+		for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
+			clock_array_index = power_state->v2.clockInfoIndex[j];
+			/* XXX this might be an inagua bug... */
+			if (clock_array_index >= clock_info_array->ucNumEntries)
+				continue;
+			clock_info = (union pplib_clock_info *)
+				&clock_info_array->clockInfo[clock_array_index];
+			valid = radeon_atombios_parse_pplib_clock_info(rdev,
+								       state_index, mode_index,
+								       clock_info);
+			if (valid)
+				mode_index++;
+		}
+		rdev->pm.power_state[state_index].num_clock_modes = mode_index;
+		if (mode_index) {
+			radeon_atombios_parse_pplib_non_clock_info(rdev, state_index, mode_index,
+								   non_clock_info);
+			state_index++;
+		}
+	}
+	/* if multiple clock modes, mark the lowest as no display */
+	for (i = 0; i < state_index; i++) {
+		if (rdev->pm.power_state[i].num_clock_modes > 1)
+			rdev->pm.power_state[i].clock_info[0].flags |=
+				RADEON_PM_MODE_NO_DISPLAY;
+	}
+	/* first mode is usually default */
+	if (rdev->pm.default_power_state_index == -1) {
+		rdev->pm.power_state[0].type =
+			POWER_STATE_TYPE_DEFAULT;
+		rdev->pm.default_power_state_index = 0;
+		rdev->pm.power_state[0].default_clock_mode =
+			&rdev->pm.power_state[0].clock_info[0];
+	}
+	return state_index;
+}
+
+void radeon_atombios_get_power_modes(struct radeon_device *rdev)
+{
+	struct radeon_mode_info *mode_info = &rdev->mode_info;
+	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
+	u16 data_offset;
+	u8 frev, crev;
+	int state_index = 0;
+
+	rdev->pm.default_power_state_index = -1;
+
+	if (atom_parse_data_header(mode_info->atom_context, index, NULL,
+				   &frev, &crev, &data_offset)) {
+		switch (frev) {
+		case 1:
+		case 2:
+		case 3:
+			state_index = radeon_atombios_parse_power_table_1_3(rdev);
+			break;
+		case 4:
+		case 5:
+			state_index = radeon_atombios_parse_power_table_4_5(rdev);
+			break;
+		case 6:
+			state_index = radeon_atombios_parse_power_table_6(rdev);
+			break;
+		default:
+			break;
 		}
 	} else {
 		/* add the default mode */
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 6d64a2705f12..35b5eb8fbe2a 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -77,13 +77,13 @@ int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 			p->relocs_ptr[i] = &p->relocs[i];
 			p->relocs[i].robj = p->relocs[i].gobj->driver_private;
 			p->relocs[i].lobj.bo = p->relocs[i].robj;
-			p->relocs[i].lobj.rdomain = r->read_domains;
 			p->relocs[i].lobj.wdomain = r->write_domain;
+			p->relocs[i].lobj.rdomain = r->read_domains;
+			p->relocs[i].lobj.tv.bo = &p->relocs[i].robj->tbo;
 			p->relocs[i].handle = r->handle;
 			p->relocs[i].flags = r->flags;
-			INIT_LIST_HEAD(&p->relocs[i].lobj.list);
 			radeon_bo_list_add_object(&p->relocs[i].lobj,
-						&p->validated);
+						  &p->validated);
 		}
 	}
 	return radeon_bo_list_validate(&p->validated);
@@ -189,10 +189,13 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error)
 {
 	unsigned i;
 
-	if (!error && parser->ib) {
-		radeon_bo_list_fence(&parser->validated, parser->ib->fence);
-	}
-	radeon_bo_list_unreserve(&parser->validated);
+
+	if (!error && parser->ib)
+		ttm_eu_fence_buffer_objects(&parser->validated,
+					    parser->ib->fence);
+	else
+		ttm_eu_backoff_reservation(&parser->validated);
+
 	if (parser->relocs != NULL) {
 		for (i = 0; i < parser->nrelocs; i++) {
 			if (parser->relocs[i].gobj)
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index e12e79326cb1..be1278721f45 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -335,7 +335,12 @@ bool radeon_card_posted(struct radeon_device *rdev)
 	uint32_t reg;
 
 	/* first check CRTCs */
-	if (ASIC_IS_DCE4(rdev)) {
+	if (ASIC_IS_DCE41(rdev)) {
+		reg = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET) |
+			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET);
+		if (reg & EVERGREEN_CRTC_MASTER_EN)
+			return true;
+	} else if (ASIC_IS_DCE4(rdev)) {
 		reg = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET) |
 			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET) |
 			RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET) |
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 1df4dc6c063c..7b17e639ab32 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -183,12 +183,272 @@ static void radeon_crtc_destroy(struct drm_crtc *crtc)
 	kfree(radeon_crtc);
 }
 
+/*
+ * Handle unpin events outside the interrupt handler proper.
+ */
+static void radeon_unpin_work_func(struct work_struct *__work)
+{
+	struct radeon_unpin_work *work =
+		container_of(__work, struct radeon_unpin_work, work);
+	int r;
+
+	/* unpin of the old buffer */
+	r = radeon_bo_reserve(work->old_rbo, false);
+	if (likely(r == 0)) {
+		r = radeon_bo_unpin(work->old_rbo);
+		if (unlikely(r != 0)) {
+			DRM_ERROR("failed to unpin buffer after flip\n");
+		}
+		radeon_bo_unreserve(work->old_rbo);
+	} else
+		DRM_ERROR("failed to reserve buffer after flip\n");
+	kfree(work);
+}
+
+void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	struct radeon_unpin_work *work;
+	struct drm_pending_vblank_event *e;
+	struct timeval now;
+	unsigned long flags;
+	u32 update_pending;
+	int vpos, hpos;
+
+	spin_lock_irqsave(&rdev->ddev->event_lock, flags);
+	work = radeon_crtc->unpin_work;
+	if (work == NULL ||
+	    !radeon_fence_signaled(work->fence)) {
+		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
+		return;
+	}
+	/* New pageflip, or just completion of a previous one? */
+	if (!radeon_crtc->deferred_flip_completion) {
+		/* do the flip (mmio) */
+		update_pending = radeon_page_flip(rdev, crtc_id, work->new_crtc_base);
+	} else {
+		/* This is just a completion of a flip queued in crtc
+		 * at last invocation. Make sure we go directly to
+		 * completion routine.
+		 */
+		update_pending = 0;
+		radeon_crtc->deferred_flip_completion = 0;
+	}
+
+	/* Has the pageflip already completed in crtc, or is it certain
+	 * to complete in this vblank?
+	 */
+	if (update_pending &&
+	    (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id,
+							       &vpos, &hpos)) &&
+	    (vpos >=0) &&
+	    (vpos < (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100)) {
+		/* crtc didn't flip in this target vblank interval,
+		 * but flip is pending in crtc. It will complete it
+		 * in next vblank interval, so complete the flip at
+		 * next vblank irq.
+		 */
+		radeon_crtc->deferred_flip_completion = 1;
+		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
+		return;
+	}
+
+	/* Pageflip (will be) certainly completed in this vblank. Clean up. */
+	radeon_crtc->unpin_work = NULL;
+
+	/* wakeup userspace */
+	if (work->event) {
+		e = work->event;
+		e->event.sequence = drm_vblank_count_and_time(rdev->ddev, crtc_id, &now);
+		e->event.tv_sec = now.tv_sec;
+		e->event.tv_usec = now.tv_usec;
+		list_add_tail(&e->base.link, &e->base.file_priv->event_list);
+		wake_up_interruptible(&e->base.file_priv->event_wait);
+	}
+	spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
+
+	drm_vblank_put(rdev->ddev, radeon_crtc->crtc_id);
+	radeon_fence_unref(&work->fence);
+	radeon_post_page_flip(work->rdev, work->crtc_id);
+	schedule_work(&work->work);
+}
+
+static int radeon_crtc_page_flip(struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_pending_vblank_event *event)
+{
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+	struct radeon_framebuffer *old_radeon_fb;
+	struct radeon_framebuffer *new_radeon_fb;
+	struct drm_gem_object *obj;
+	struct radeon_bo *rbo;
+	struct radeon_fence *fence;
+	struct radeon_unpin_work *work;
+	unsigned long flags;
+	u32 tiling_flags, pitch_pixels;
+	u64 base;
+	int r;
+
+	work = kzalloc(sizeof *work, GFP_KERNEL);
+	if (work == NULL)
+		return -ENOMEM;
+
+	r = radeon_fence_create(rdev, &fence);
+	if (unlikely(r != 0)) {
+		kfree(work);
+		DRM_ERROR("flip queue: failed to create fence.\n");
+		return -ENOMEM;
+	}
+	work->event = event;
+	work->rdev = rdev;
+	work->crtc_id = radeon_crtc->crtc_id;
+	work->fence = radeon_fence_ref(fence);
+	old_radeon_fb = to_radeon_framebuffer(crtc->fb);
+	new_radeon_fb = to_radeon_framebuffer(fb);
+	/* schedule unpin of the old buffer */
+	obj = old_radeon_fb->obj;
+	rbo = obj->driver_private;
+	work->old_rbo = rbo;
+	INIT_WORK(&work->work, radeon_unpin_work_func);
+
+	/* We borrow the event spin lock for protecting unpin_work */
+	spin_lock_irqsave(&dev->event_lock, flags);
+	if (radeon_crtc->unpin_work) {
+		spin_unlock_irqrestore(&dev->event_lock, flags);
+		kfree(work);
+		radeon_fence_unref(&fence);
+
+		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
+		return -EBUSY;
+	}
+	radeon_crtc->unpin_work = work;
+	radeon_crtc->deferred_flip_completion = 0;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+
+	/* pin the new buffer */
+	obj = new_radeon_fb->obj;
+	rbo = obj->driver_private;
+
+	DRM_DEBUG_DRIVER("flip-ioctl() cur_fbo = %p, cur_bbo = %p\n",
+			 work->old_rbo, rbo);
+
+	r = radeon_bo_reserve(rbo, false);
+	if (unlikely(r != 0)) {
+		DRM_ERROR("failed to reserve new rbo buffer before flip\n");
+		goto pflip_cleanup;
+	}
+	r = radeon_bo_pin(rbo, RADEON_GEM_DOMAIN_VRAM, &base);
+	if (unlikely(r != 0)) {
+		radeon_bo_unreserve(rbo);
+		r = -EINVAL;
+		DRM_ERROR("failed to pin new rbo buffer before flip\n");
+		goto pflip_cleanup;
+	}
+	radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL);
+	radeon_bo_unreserve(rbo);
+
+	if (!ASIC_IS_AVIVO(rdev)) {
+		/* crtc offset is from display base addr not FB location */
+		base -= radeon_crtc->legacy_display_base_addr;
+		pitch_pixels = fb->pitch / (fb->bits_per_pixel / 8);
+
+		if (tiling_flags & RADEON_TILING_MACRO) {
+			if (ASIC_IS_R300(rdev)) {
+				base &= ~0x7ff;
+			} else {
+				int byteshift = fb->bits_per_pixel >> 4;
+				int tile_addr = (((crtc->y >> 3) * pitch_pixels +  crtc->x) >> (8 - byteshift)) << 11;
+				base += tile_addr + ((crtc->x << byteshift) % 256) + ((crtc->y % 8) << 8);
+			}
+		} else {
+			int offset = crtc->y * pitch_pixels + crtc->x;
+			switch (fb->bits_per_pixel) {
+			case 8:
+			default:
+				offset *= 1;
+				break;
+			case 15:
+			case 16:
+				offset *= 2;
+				break;
+			case 24:
+				offset *= 3;
+				break;
+			case 32:
+				offset *= 4;
+				break;
+			}
+			base += offset;
+		}
+		base &= ~7;
+	}
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+	work->new_crtc_base = base;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+
+	/* update crtc fb */
+	crtc->fb = fb;
+
+	r = drm_vblank_get(dev, radeon_crtc->crtc_id);
+	if (r) {
+		DRM_ERROR("failed to get vblank before flip\n");
+		goto pflip_cleanup1;
+	}
+
+	/* 32 ought to cover us */
+	r = radeon_ring_lock(rdev, 32);
+	if (r) {
+		DRM_ERROR("failed to lock the ring before flip\n");
+		goto pflip_cleanup2;
+	}
+
+	/* emit the fence */
+	radeon_fence_emit(rdev, fence);
+	/* set the proper interrupt */
+	radeon_pre_page_flip(rdev, radeon_crtc->crtc_id);
+	/* fire the ring */
+	radeon_ring_unlock_commit(rdev);
+
+	return 0;
+
+pflip_cleanup2:
+	drm_vblank_put(dev, radeon_crtc->crtc_id);
+
+pflip_cleanup1:
+	r = radeon_bo_reserve(rbo, false);
+	if (unlikely(r != 0)) {
+		DRM_ERROR("failed to reserve new rbo in error path\n");
+		goto pflip_cleanup;
+	}
+	r = radeon_bo_unpin(rbo);
+	if (unlikely(r != 0)) {
+		radeon_bo_unreserve(rbo);
+		r = -EINVAL;
+		DRM_ERROR("failed to unpin new rbo in error path\n");
+		goto pflip_cleanup;
+	}
+	radeon_bo_unreserve(rbo);
+
+pflip_cleanup:
+	spin_lock_irqsave(&dev->event_lock, flags);
+	radeon_crtc->unpin_work = NULL;
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+	radeon_fence_unref(&fence);
+	kfree(work);
+
+	return r;
+}
+
 static const struct drm_crtc_funcs radeon_crtc_funcs = {
 	.cursor_set = radeon_crtc_cursor_set,
 	.cursor_move = radeon_crtc_cursor_move,
 	.gamma_set = radeon_crtc_gamma_set,
 	.set_config = drm_crtc_helper_set_config,
 	.destroy = radeon_crtc_destroy,
+	.page_flip = radeon_crtc_page_flip,
 };
 
 static void radeon_crtc_init(struct drm_device *dev, int index)
@@ -225,7 +485,7 @@ static void radeon_crtc_init(struct drm_device *dev, int index)
 		radeon_legacy_init_crtc(dev, radeon_crtc);
 }
 
-static const char *encoder_names[34] = {
+static const char *encoder_names[36] = {
 	"NONE",
 	"INTERNAL_LVDS",
 	"INTERNAL_TMDS1",
@@ -260,6 +520,8 @@ static const char *encoder_names[34] = {
 	"INTERNAL_KLDSCP_LVTMA",
 	"INTERNAL_UNIPHY1",
 	"INTERNAL_UNIPHY2",
+	"NUTMEG",
+	"TRAVIS",
 };
 
 static const char *connector_names[15] = {
@@ -1019,7 +1281,7 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
 /*
  * Retrieve current video scanout position of crtc on a given gpu.
  *
- * \param rdev Device to query.
+ * \param dev Device to query.
  * \param crtc Crtc to query.
  * \param *vpos Location where vertical scanout position should be stored.
  * \param *hpos Location where horizontal scanout position should go.
@@ -1031,72 +1293,74 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  *
  * \return Flags, or'ed together as follows:
  *
- * RADEON_SCANOUTPOS_VALID = Query successfull.
- * RADEON_SCANOUTPOS_INVBL = Inside vblank.
- * RADEON_SCANOUTPOS_ACCURATE = Returned position is accurate. A lack of
+ * DRM_SCANOUTPOS_VALID = Query successfull.
+ * DRM_SCANOUTPOS_INVBL = Inside vblank.
+ * DRM_SCANOUTPOS_ACCURATE = Returned position is accurate. A lack of
  * this flag means that returned position may be offset by a constant but
  * unknown small number of scanlines wrt. real scanout position.
  *
  */
-int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos, int *hpos)
+int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, int *vpos, int *hpos)
 {
 	u32 stat_crtc = 0, vbl = 0, position = 0;
 	int vbl_start, vbl_end, vtotal, ret = 0;
 	bool in_vbl = true;
 
+	struct radeon_device *rdev = dev->dev_private;
+
 	if (ASIC_IS_DCE4(rdev)) {
 		if (crtc == 0) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC0_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC0_REGISTER_OFFSET);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 1) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC1_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC1_REGISTER_OFFSET);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 2) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC2_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC2_REGISTER_OFFSET);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 3) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC3_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC3_REGISTER_OFFSET);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 4) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC4_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC4_REGISTER_OFFSET);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 5) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC5_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC5_REGISTER_OFFSET);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	} else if (ASIC_IS_AVIVO(rdev)) {
 		if (crtc == 0) {
 			vbl = RREG32(AVIVO_D1CRTC_V_BLANK_START_END);
 			position = RREG32(AVIVO_D1CRTC_STATUS_POSITION);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 1) {
 			vbl = RREG32(AVIVO_D2CRTC_V_BLANK_START_END);
 			position = RREG32(AVIVO_D2CRTC_STATUS_POSITION);
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	} else {
 		/* Pre-AVIVO: Different encoding of scanout pos and vblank interval. */
@@ -1112,7 +1376,7 @@ int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos,
 			if (!(stat_crtc & 1))
 				in_vbl = false;
 
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 		if (crtc == 1) {
 			vbl = (RREG32(RADEON_CRTC2_V_TOTAL_DISP) &
@@ -1122,7 +1386,7 @@ int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos,
 			if (!(stat_crtc & 1))
 				in_vbl = false;
 
-			ret |= RADEON_SCANOUTPOS_VALID;
+			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	}
 
@@ -1133,13 +1397,13 @@ int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos,
 	/* Valid vblank area boundaries from gpu retrieved? */
 	if (vbl > 0) {
 		/* Yes: Decode. */
-		ret |= RADEON_SCANOUTPOS_ACCURATE;
+		ret |= DRM_SCANOUTPOS_ACCURATE;
 		vbl_start = vbl & 0x1fff;
 		vbl_end = (vbl >> 16) & 0x1fff;
 	}
 	else {
 		/* No: Fake something reasonable which gives at least ok results. */
-		vbl_start = rdev->mode_info.crtcs[crtc]->base.mode.crtc_vdisplay;
+		vbl_start = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
 		vbl_end = 0;
 	}
 
@@ -1155,7 +1419,7 @@ int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos,
 
 	/* Inside "upper part" of vblank area? Apply corrective offset if so: */
 	if (in_vbl && (*vpos >= vbl_start)) {
-		vtotal = rdev->mode_info.crtcs[crtc]->base.mode.crtc_vtotal;
+		vtotal = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
 		*vpos = *vpos - vtotal;
 	}
 
@@ -1164,7 +1428,7 @@ int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos,
 
 	/* In vblank? */
 	if (in_vbl)
-		ret |= RADEON_SCANOUTPOS_INVBL;
+		ret |= DRM_SCANOUTPOS_INVBL;
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 88e4ea925900..a92d2a5cea90 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -48,9 +48,10 @@
  * - 2.5.0 - add get accel 2 to work around ddx breakage for evergreen
  * - 2.6.0 - add tiling config query (r6xx+), add initial HiZ support (r300->r500)
  *   2.7.0 - fixups for r600 2D tiling support. (no external ABI change), add eg dyn gpr regs
+ *   2.8.0 - pageflip support
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	7
+#define KMS_DRIVER_MINOR	8
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
@@ -66,6 +67,10 @@ int radeon_resume_kms(struct drm_device *dev);
 u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc);
 int radeon_enable_vblank_kms(struct drm_device *dev, int crtc);
 void radeon_disable_vblank_kms(struct drm_device *dev, int crtc);
+int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
+				    int *max_error,
+				    struct timeval *vblank_time,
+				    unsigned flags);
 void radeon_driver_irq_preinstall_kms(struct drm_device *dev);
 int radeon_driver_irq_postinstall_kms(struct drm_device *dev);
 void radeon_driver_irq_uninstall_kms(struct drm_device *dev);
@@ -74,6 +79,8 @@ int radeon_dma_ioctl_kms(struct drm_device *dev, void *data,
 			 struct drm_file *file_priv);
 int radeon_gem_object_init(struct drm_gem_object *obj);
 void radeon_gem_object_free(struct drm_gem_object *obj);
+extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
+				      int *vpos, int *hpos);
 extern struct drm_ioctl_desc radeon_ioctls_kms[];
 extern int radeon_max_kms_ioctl;
 int radeon_mmap(struct file *filp, struct vm_area_struct *vma);
@@ -277,6 +284,8 @@ static struct drm_driver kms_driver = {
 	.get_vblank_counter = radeon_get_vblank_counter_kms,
 	.enable_vblank = radeon_enable_vblank_kms,
 	.disable_vblank = radeon_disable_vblank_kms,
+	.get_vblank_timestamp = radeon_get_vblank_timestamp_kms,
+	.get_scanout_position = radeon_get_crtc_scanoutpos,
 #if defined(CONFIG_DEBUG_FS)
 	.debugfs_init = radeon_debugfs_init,
 	.debugfs_cleanup = radeon_debugfs_cleanup,
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index 041943df966b..e4e64a80b58d 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -713,7 +713,7 @@ atombios_get_encoder_mode(struct drm_encoder *encoder)
  * DIG1/2 can drive UNIPHY0/1/2 link A or link B
  *
  * DCE 4.0
- * - 3 DIG transmitter blocks UNPHY0/1/2 (links A and B).
+ * - 3 DIG transmitter blocks UNIPHY0/1/2 (links A and B).
  * Supports up to 6 digital outputs
  * - 6 DIG encoder blocks.
  * - DIG to PHY mapping is hardcoded
@@ -724,6 +724,12 @@ atombios_get_encoder_mode(struct drm_encoder *encoder)
  * DIG5 drives UNIPHY2 link A, A+B
  * DIG6 drives UNIPHY2 link B
  *
+ * DCE 4.1
+ * - 3 DIG transmitter blocks UNIPHY0/1/2 (links A and B).
+ * Supports up to 6 digital outputs
+ * - 2 DIG encoder blocks.
+ * DIG1/2 can drive UNIPHY0/1/2 link A or link B
+ *
  * Routing
  * crtc -> dig encoder -> UNIPHY/LVTMA (1 or 2 links)
  * Examples:
@@ -904,9 +910,15 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t
 		else
 			args.v3.ucLaneNum = 4;
 
-		if (dig->linkb) {
-			args.v3.acConfig.ucLinkSel = 1;
-			args.v3.acConfig.ucEncoderSel = 1;
+		if (ASIC_IS_DCE41(rdev)) {
+			args.v3.acConfig.ucEncoderSel = dig->dig_encoder;
+			if (dig->linkb)
+				args.v3.acConfig.ucLinkSel = 1;
+		} else {
+			if (dig->linkb) {
+				args.v3.acConfig.ucLinkSel = 1;
+				args.v3.acConfig.ucEncoderSel = 1;
+			}
 		}
 
 		/* Select the PLL for the PHY
@@ -1044,6 +1056,7 @@ atombios_set_edp_panel_power(struct drm_connector *connector, int action)
 
 union external_encoder_control {
 	EXTERNAL_ENCODER_CONTROL_PS_ALLOCATION v1;
+	EXTERNAL_ENCODER_CONTROL_PS_ALLOCATION_V3 v3;
 };
 
 static void
@@ -1054,6 +1067,7 @@ atombios_external_encoder_setup(struct drm_encoder *encoder,
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_encoder *ext_radeon_encoder = to_radeon_encoder(ext_encoder);
 	union external_encoder_control args;
 	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	int index = GetIndexIntoMasterTable(COMMAND, ExternalEncoderControl);
@@ -1061,6 +1075,7 @@ atombios_external_encoder_setup(struct drm_encoder *encoder,
 	int dp_clock = 0;
 	int dp_lane_count = 0;
 	int connector_object_id = 0;
+	u32 ext_enum = (ext_radeon_encoder->encoder_enum & ENUM_ID_MASK) >> ENUM_ID_SHIFT;
 
 	if (connector) {
 		struct radeon_connector *radeon_connector = to_radeon_connector(connector);
@@ -1099,6 +1114,37 @@ atombios_external_encoder_setup(struct drm_encoder *encoder,
 			else
 				args.v1.sDigEncoder.ucLaneNum = 4;
 			break;
+		case 3:
+			args.v3.sExtEncoder.ucAction = action;
+			if (action == EXTERNAL_ENCODER_ACTION_V3_ENCODER_INIT)
+				args.v3.sExtEncoder.usConnectorId = connector_object_id;
+			else
+				args.v3.sExtEncoder.usPixelClock = cpu_to_le16(radeon_encoder->pixel_clock / 10);
+			args.v3.sExtEncoder.ucEncoderMode = atombios_get_encoder_mode(encoder);
+
+			if (args.v3.sExtEncoder.ucEncoderMode == ATOM_ENCODER_MODE_DP) {
+				if (dp_clock == 270000)
+					args.v3.sExtEncoder.ucConfig |= EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_2_70GHZ;
+				else if (dp_clock == 540000)
+					args.v3.sExtEncoder.ucConfig |= EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_5_40GHZ;
+				args.v3.sExtEncoder.ucLaneNum = dp_lane_count;
+			} else if (radeon_encoder->pixel_clock > 165000)
+				args.v3.sExtEncoder.ucLaneNum = 8;
+			else
+				args.v3.sExtEncoder.ucLaneNum = 4;
+			switch (ext_enum) {
+			case GRAPH_OBJECT_ENUM_ID1:
+				args.v3.sExtEncoder.ucConfig |= EXTERNAL_ENCODER_CONFIG_V3_ENCODER1;
+				break;
+			case GRAPH_OBJECT_ENUM_ID2:
+				args.v3.sExtEncoder.ucConfig |= EXTERNAL_ENCODER_CONFIG_V3_ENCODER2;
+				break;
+			case GRAPH_OBJECT_ENUM_ID3:
+				args.v3.sExtEncoder.ucConfig |= EXTERNAL_ENCODER_CONFIG_V3_ENCODER3;
+				break;
+			}
+			args.v3.sExtEncoder.ucBitPerColor = PANEL_8BIT_PER_COLOR;
+			break;
 		default:
 			DRM_ERROR("Unknown table version: %d, %d\n", frev, crev);
 			return;
@@ -1289,12 +1335,18 @@ radeon_atom_encoder_dpms(struct drm_encoder *encoder, int mode)
 		switch (mode) {
 		case DRM_MODE_DPMS_ON:
 		default:
-			action = ATOM_ENABLE;
+			if (ASIC_IS_DCE41(rdev) && (rdev->flags & RADEON_IS_IGP))
+				action = EXTERNAL_ENCODER_ACTION_V3_ENABLE_OUTPUT;
+			else
+				action = ATOM_ENABLE;
 			break;
 		case DRM_MODE_DPMS_STANDBY:
 		case DRM_MODE_DPMS_SUSPEND:
 		case DRM_MODE_DPMS_OFF:
-			action = ATOM_DISABLE;
+			if (ASIC_IS_DCE41(rdev) && (rdev->flags & RADEON_IS_IGP))
+				action = EXTERNAL_ENCODER_ACTION_V3_DISABLE_OUTPUT;
+			else
+				action = ATOM_DISABLE;
 			break;
 		}
 		atombios_external_encoder_setup(encoder, ext_encoder, action);
@@ -1483,6 +1535,11 @@ static int radeon_atom_pick_dig_encoder(struct drm_encoder *encoder)
 	struct radeon_encoder_atom_dig *dig;
 	uint32_t dig_enc_in_use = 0;
 
+	/* on DCE41 and encoder can driver any phy so just crtc id */
+	if (ASIC_IS_DCE41(rdev)) {
+		return radeon_crtc->crtc_id;
+	}
+
 	if (ASIC_IS_DCE4(rdev)) {
 		dig = radeon_encoder->enc_priv;
 		switch (radeon_encoder->encoder_id) {
@@ -1610,7 +1667,13 @@ radeon_atom_encoder_mode_set(struct drm_encoder *encoder,
 	}
 
 	if (ext_encoder) {
-		atombios_external_encoder_setup(encoder, ext_encoder, ATOM_ENABLE);
+		if (ASIC_IS_DCE41(rdev) && (rdev->flags & RADEON_IS_IGP)) {
+			atombios_external_encoder_setup(encoder, ext_encoder,
+							EXTERNAL_ENCODER_ACTION_V3_ENCODER_INIT);
+			atombios_external_encoder_setup(encoder, ext_encoder,
+							EXTERNAL_ENCODER_ACTION_V3_ENCODER_SETUP);
+		} else
+			atombios_external_encoder_setup(encoder, ext_encoder, ATOM_ENABLE);
 	}
 
 	atombios_apply_encoder_quirks(encoder, adjusted_mode);
@@ -2029,6 +2092,8 @@ radeon_add_atom_encoder(struct drm_device *dev, uint32_t encoder_enum, uint32_t
 	case ENCODER_OBJECT_ID_TITFP513:
 	case ENCODER_OBJECT_ID_VT1623:
 	case ENCODER_OBJECT_ID_HDMI_SI1930:
+	case ENCODER_OBJECT_ID_TRAVIS:
+	case ENCODER_OBJECT_ID_NUTMEG:
 		/* these are handled by the primary encoders */
 		radeon_encoder->is_ext_encoder = true;
 		if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT))
diff --git a/drivers/gpu/drm/radeon/radeon_family.h b/drivers/gpu/drm/radeon/radeon_family.h
index e329066dcabd..4c222d5437d1 100644
--- a/drivers/gpu/drm/radeon/radeon_family.h
+++ b/drivers/gpu/drm/radeon/radeon_family.h
@@ -80,6 +80,7 @@ enum radeon_family {
 	CHIP_JUNIPER,
 	CHIP_CYPRESS,
 	CHIP_HEMLOCK,
+	CHIP_PALM,
 	CHIP_LAST,
 };
 
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index a108c7ed14f5..c6861bb751ad 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -71,8 +71,10 @@ void radeon_driver_irq_preinstall_kms(struct drm_device *dev)
 	rdev->irq.gui_idle = false;
 	for (i = 0; i < rdev->num_crtc; i++)
 		rdev->irq.crtc_vblank_int[i] = false;
-	for (i = 0; i < 6; i++)
+	for (i = 0; i < 6; i++) {
 		rdev->irq.hpd[i] = false;
+		rdev->irq.pflip[i] = false;
+	}
 	radeon_irq_set(rdev);
 	/* Clear bits */
 	radeon_irq_process(rdev);
@@ -101,8 +103,10 @@ void radeon_driver_irq_uninstall_kms(struct drm_device *dev)
 	rdev->irq.gui_idle = false;
 	for (i = 0; i < rdev->num_crtc; i++)
 		rdev->irq.crtc_vblank_int[i] = false;
-	for (i = 0; i < 6; i++)
+	for (i = 0; i < 6; i++) {
 		rdev->irq.hpd[i] = false;
+		rdev->irq.pflip[i] = false;
+	}
 	radeon_irq_set(rdev);
 }
 
@@ -121,7 +125,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
 	 * chips.  Disable MSI on them for now.
 	 */
 	if ((rdev->family >= CHIP_RV380) &&
-	    (!(rdev->flags & RADEON_IS_IGP)) &&
+	    ((!(rdev->flags & RADEON_IS_IGP)) || (rdev->family >= CHIP_PALM)) &&
 	    (!(rdev->flags & RADEON_IS_AGP))) {
 		int ret = pci_enable_msi(rdev->pdev);
 		if (!ret) {
@@ -175,3 +179,34 @@ void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev)
 	spin_unlock_irqrestore(&rdev->irq.sw_lock, irqflags);
 }
 
+void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc)
+{
+	unsigned long irqflags;
+
+	if (crtc < 0 || crtc >= rdev->num_crtc)
+		return;
+
+	spin_lock_irqsave(&rdev->irq.pflip_lock[crtc], irqflags);
+	if (rdev->ddev->irq_enabled && (++rdev->irq.pflip_refcount[crtc] == 1)) {
+		rdev->irq.pflip[crtc] = true;
+		radeon_irq_set(rdev);
+	}
+	spin_unlock_irqrestore(&rdev->irq.pflip_lock[crtc], irqflags);
+}
+
+void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc)
+{
+	unsigned long irqflags;
+
+	if (crtc < 0 || crtc >= rdev->num_crtc)
+		return;
+
+	spin_lock_irqsave(&rdev->irq.pflip_lock[crtc], irqflags);
+	BUG_ON(rdev->ddev->irq_enabled && rdev->irq.pflip_refcount[crtc] <= 0);
+	if (rdev->ddev->irq_enabled && (--rdev->irq.pflip_refcount[crtc] == 0)) {
+		rdev->irq.pflip[crtc] = false;
+		radeon_irq_set(rdev);
+	}
+	spin_unlock_irqrestore(&rdev->irq.pflip_lock[crtc], irqflags);
+}
+
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 8fbbe1c6ebbd..4bf423ca4c12 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -277,6 +277,27 @@ void radeon_disable_vblank_kms(struct drm_device *dev, int crtc)
 	radeon_irq_set(rdev);
 }
 
+int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
+				    int *max_error,
+				    struct timeval *vblank_time,
+				    unsigned flags)
+{
+	struct drm_crtc *drmcrtc;
+	struct radeon_device *rdev = dev->dev_private;
+
+	if (crtc < 0 || crtc >= dev->num_crtcs) {
+		DRM_ERROR("Invalid crtc %d\n", crtc);
+		return -EINVAL;
+	}
+
+	/* Get associated drm_crtc: */
+	drmcrtc = &rdev->mode_info.crtcs[crtc]->base;
+
+	/* Helper routine in DRM core does all the work: */
+	return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error,
+						     vblank_time, flags,
+						     drmcrtc);
+}
 
 /*
  * IOCTL.
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index e301c6f9e059..f406f02bf14e 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -277,6 +277,9 @@ struct radeon_crtc {
 	fixed20_12 hsc;
 	struct drm_display_mode native_mode;
 	int pll_id;
+	/* page flipping */
+	struct radeon_unpin_work *unpin_work;
+	int deferred_flip_completion;
 };
 
 struct radeon_encoder_primary_dac {
@@ -442,10 +445,6 @@ struct radeon_framebuffer {
 	struct drm_gem_object *obj;
 };
 
-/* radeon_get_crtc_scanoutpos() return flags */
-#define RADEON_SCANOUTPOS_VALID        (1 << 0)
-#define RADEON_SCANOUTPOS_INVBL        (1 << 1)
-#define RADEON_SCANOUTPOS_ACCURATE     (1 << 2)
 
 extern enum radeon_tv_std
 radeon_combios_get_tv_info(struct radeon_device *rdev);
@@ -562,7 +561,8 @@ extern int radeon_crtc_cursor_set(struct drm_crtc *crtc,
 extern int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 				   int x, int y);
 
-extern int radeon_get_crtc_scanoutpos(struct radeon_device *rdev, int crtc, int *vpos, int *hpos);
+extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
+				      int *vpos, int *hpos);
 
 extern bool radeon_combios_check_hardcoded_edid(struct radeon_device *rdev);
 extern struct edid *
@@ -662,4 +662,7 @@ int radeon_fbdev_total_size(struct radeon_device *rdev);
 bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj);
 
 void radeon_fb_output_poll_changed(struct radeon_device *rdev);
+
+void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id);
+
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index a598d0049aa5..82b032670782 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -302,34 +302,9 @@ void radeon_bo_list_add_object(struct radeon_bo_list *lobj,
 				struct list_head *head)
 {
 	if (lobj->wdomain) {
-		list_add(&lobj->list, head);
+		list_add(&lobj->tv.head, head);
 	} else {
-		list_add_tail(&lobj->list, head);
-	}
-}
-
-int radeon_bo_list_reserve(struct list_head *head)
-{
-	struct radeon_bo_list *lobj;
-	int r;
-
-	list_for_each_entry(lobj, head, list){
-		r = radeon_bo_reserve(lobj->bo, false);
-		if (unlikely(r != 0))
-			return r;
-		lobj->reserved = true;
-	}
-	return 0;
-}
-
-void radeon_bo_list_unreserve(struct list_head *head)
-{
-	struct radeon_bo_list *lobj;
-
-	list_for_each_entry(lobj, head, list) {
-		/* only unreserve object we successfully reserved */
-		if (lobj->reserved && radeon_bo_is_reserved(lobj->bo))
-			radeon_bo_unreserve(lobj->bo);
+		list_add_tail(&lobj->tv.head, head);
 	}
 }
 
@@ -340,14 +315,11 @@ int radeon_bo_list_validate(struct list_head *head)
 	u32 domain;
 	int r;
 
-	list_for_each_entry(lobj, head, list) {
-		lobj->reserved = false;
-	}
-	r = radeon_bo_list_reserve(head);
+	r = ttm_eu_reserve_buffers(head);
 	if (unlikely(r != 0)) {
 		return r;
 	}
-	list_for_each_entry(lobj, head, list) {
+	list_for_each_entry(lobj, head, tv.head) {
 		bo = lobj->bo;
 		if (!bo->pin_count) {
 			domain = lobj->wdomain ? lobj->wdomain : lobj->rdomain;
@@ -370,25 +342,6 @@ int radeon_bo_list_validate(struct list_head *head)
 	return 0;
 }
 
-void radeon_bo_list_fence(struct list_head *head, void *fence)
-{
-	struct radeon_bo_list *lobj;
-	struct radeon_bo *bo;
-	struct radeon_fence *old_fence = NULL;
-
-	list_for_each_entry(lobj, head, list) {
-		bo = lobj->bo;
-		spin_lock(&bo->tbo.lock);
-		old_fence = (struct radeon_fence *)bo->tbo.sync_obj;
-		bo->tbo.sync_obj = radeon_fence_ref(fence);
-		bo->tbo.sync_obj_arg = NULL;
-		spin_unlock(&bo->tbo.lock);
-		if (old_fence) {
-			radeon_fence_unref(&old_fence);
-		}
-	}
-}
-
 int radeon_bo_fbdev_mmap(struct radeon_bo *bo,
 			     struct vm_area_struct *vma)
 {
diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
index d143702b244a..22d4c237dea5 100644
--- a/drivers/gpu/drm/radeon/radeon_object.h
+++ b/drivers/gpu/drm/radeon/radeon_object.h
@@ -126,12 +126,12 @@ static inline int radeon_bo_wait(struct radeon_bo *bo, u32 *mem_type,
 	r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, 0);
 	if (unlikely(r != 0))
 		return r;
-	spin_lock(&bo->tbo.lock);
+	spin_lock(&bo->tbo.bdev->fence_lock);
 	if (mem_type)
 		*mem_type = bo->tbo.mem.mem_type;
 	if (bo->tbo.sync_obj)
 		r = ttm_bo_wait(&bo->tbo, true, true, no_wait);
-	spin_unlock(&bo->tbo.lock);
+	spin_unlock(&bo->tbo.bdev->fence_lock);
 	ttm_bo_unreserve(&bo->tbo);
 	return r;
 }
@@ -152,10 +152,7 @@ extern int radeon_bo_init(struct radeon_device *rdev);
 extern void radeon_bo_fini(struct radeon_device *rdev);
 extern void radeon_bo_list_add_object(struct radeon_bo_list *lobj,
 				struct list_head *head);
-extern int radeon_bo_list_reserve(struct list_head *head);
-extern void radeon_bo_list_unreserve(struct list_head *head);
 extern int radeon_bo_list_validate(struct list_head *head);
-extern void radeon_bo_list_fence(struct list_head *head, void *fence);
 extern int radeon_bo_fbdev_mmap(struct radeon_bo *bo,
 				struct vm_area_struct *vma);
 extern int radeon_bo_set_tiling_flags(struct radeon_bo *bo,
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 8c9b2ef32c68..4de7776bd1c5 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -449,6 +449,9 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev,
 	case THERMAL_TYPE_EVERGREEN:
 		temp = evergreen_get_temp(rdev);
 		break;
+	case THERMAL_TYPE_SUMO:
+		temp = sumo_get_temp(rdev);
+		break;
 	default:
 		temp = 0;
 		break;
@@ -487,6 +490,7 @@ static int radeon_hwmon_init(struct radeon_device *rdev)
 	case THERMAL_TYPE_RV6XX:
 	case THERMAL_TYPE_RV770:
 	case THERMAL_TYPE_EVERGREEN:
+	case THERMAL_TYPE_SUMO:
 		rdev->pm.int_hwmon_dev = hwmon_device_register(rdev->dev);
 		if (IS_ERR(rdev->pm.int_hwmon_dev)) {
 			err = PTR_ERR(rdev->pm.int_hwmon_dev);
@@ -720,9 +724,9 @@ static bool radeon_pm_in_vbl(struct radeon_device *rdev)
 	 */
 	for (crtc = 0; (crtc < rdev->num_crtc) && in_vbl; crtc++) {
 		if (rdev->pm.active_crtcs & (1 << crtc)) {
-			vbl_status = radeon_get_crtc_scanoutpos(rdev, crtc, &vpos, &hpos);
-			if ((vbl_status & RADEON_SCANOUTPOS_VALID) &&
-			    !(vbl_status & RADEON_SCANOUTPOS_INVBL))
+			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, &vpos, &hpos);
+			if ((vbl_status & DRM_SCANOUTPOS_VALID) &&
+			    !(vbl_status & DRM_SCANOUTPOS_INVBL))
 				in_vbl = false;
 		}
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_reg.h b/drivers/gpu/drm/radeon/radeon_reg.h
index 64928814de53..26c43e234350 100644
--- a/drivers/gpu/drm/radeon/radeon_reg.h
+++ b/drivers/gpu/drm/radeon/radeon_reg.h
@@ -422,6 +422,7 @@
 #       define RADEON_CRTC_CSYNC_EN         (1 <<  4)
 #       define RADEON_CRTC_ICON_EN          (1 << 15)
 #       define RADEON_CRTC_CUR_EN           (1 << 16)
+#       define RADEON_CRTC_VSTAT_MODE_MASK  (3 << 17)
 #       define RADEON_CRTC_CUR_MODE_MASK    (7 << 20)
 #       define RADEON_CRTC_CUR_MODE_SHIFT   20
 #       define RADEON_CRTC_CUR_MODE_MONO    0
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index f1c6e02c2e6b..9a85b1614c86 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -46,6 +46,56 @@
 void rs600_gpu_init(struct radeon_device *rdev);
 int rs600_mc_wait_for_idle(struct radeon_device *rdev);
 
+void rs600_pre_page_flip(struct radeon_device *rdev, int crtc)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc];
+	u32 tmp;
+
+	/* make sure flip is at vb rather than hb */
+	tmp = RREG32(AVIVO_D1GRPH_FLIP_CONTROL + radeon_crtc->crtc_offset);
+	tmp &= ~AVIVO_D1GRPH_SURFACE_UPDATE_H_RETRACE_EN;
+	WREG32(AVIVO_D1GRPH_FLIP_CONTROL + radeon_crtc->crtc_offset, tmp);
+
+	/* set pageflip to happen anywhere in vblank interval */
+	WREG32(AVIVO_D1MODE_MASTER_UPDATE_MODE + radeon_crtc->crtc_offset, 0);
+
+	/* enable the pflip int */
+	radeon_irq_kms_pflip_irq_get(rdev, crtc);
+}
+
+void rs600_post_page_flip(struct radeon_device *rdev, int crtc)
+{
+	/* disable the pflip int */
+	radeon_irq_kms_pflip_irq_put(rdev, crtc);
+}
+
+u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
+
+	/* Lock the graphics update lock */
+	tmp |= AVIVO_D1GRPH_UPDATE_LOCK;
+	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+
+	/* update the scanout addresses */
+	WREG32(AVIVO_D1GRPH_SECONDARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+	       (u32)crtc_base);
+	WREG32(AVIVO_D1GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+	       (u32)crtc_base);
+
+	/* Wait for update_pending to go high. */
+	while (!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING));
+	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
+
+	/* Unlock the lock, so double-buffering can take place inside vblank */
+	tmp &= ~AVIVO_D1GRPH_UPDATE_LOCK;
+	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+
+	/* Return current update_pending status: */
+	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
+}
+
 void rs600_pm_misc(struct radeon_device *rdev)
 {
 	int requested_index = rdev->pm.requested_power_state_index;
@@ -515,10 +565,12 @@ int rs600_irq_set(struct radeon_device *rdev)
 	if (rdev->irq.gui_idle) {
 		tmp |= S_000040_GUI_IDLE(1);
 	}
-	if (rdev->irq.crtc_vblank_int[0]) {
+	if (rdev->irq.crtc_vblank_int[0] ||
+	    rdev->irq.pflip[0]) {
 		mode_int |= S_006540_D1MODE_VBLANK_INT_MASK(1);
 	}
-	if (rdev->irq.crtc_vblank_int[1]) {
+	if (rdev->irq.crtc_vblank_int[1] ||
+	    rdev->irq.pflip[1]) {
 		mode_int |= S_006540_D2MODE_VBLANK_INT_MASK(1);
 	}
 	if (rdev->irq.hpd[0]) {
@@ -534,7 +586,7 @@ int rs600_irq_set(struct radeon_device *rdev)
 	return 0;
 }
 
-static inline uint32_t rs600_irq_ack(struct radeon_device *rdev, u32 *r500_disp_int)
+static inline u32 rs600_irq_ack(struct radeon_device *rdev)
 {
 	uint32_t irqs = RREG32(R_000044_GEN_INT_STATUS);
 	uint32_t irq_mask = S_000044_SW_INT(1);
@@ -547,27 +599,27 @@ static inline uint32_t rs600_irq_ack(struct radeon_device *rdev, u32 *r500_disp_
 	}
 
 	if (G_000044_DISPLAY_INT_STAT(irqs)) {
-		*r500_disp_int = RREG32(R_007EDC_DISP_INTERRUPT_STATUS);
-		if (G_007EDC_LB_D1_VBLANK_INTERRUPT(*r500_disp_int)) {
+		rdev->irq.stat_regs.r500.disp_int = RREG32(R_007EDC_DISP_INTERRUPT_STATUS);
+		if (G_007EDC_LB_D1_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			WREG32(R_006534_D1MODE_VBLANK_STATUS,
 				S_006534_D1MODE_VBLANK_ACK(1));
 		}
-		if (G_007EDC_LB_D2_VBLANK_INTERRUPT(*r500_disp_int)) {
+		if (G_007EDC_LB_D2_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			WREG32(R_006D34_D2MODE_VBLANK_STATUS,
 				S_006D34_D2MODE_VBLANK_ACK(1));
 		}
-		if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(*r500_disp_int)) {
+		if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			tmp = RREG32(R_007D08_DC_HOT_PLUG_DETECT1_INT_CONTROL);
 			tmp |= S_007D08_DC_HOT_PLUG_DETECT1_INT_ACK(1);
 			WREG32(R_007D08_DC_HOT_PLUG_DETECT1_INT_CONTROL, tmp);
 		}
-		if (G_007EDC_DC_HOT_PLUG_DETECT2_INTERRUPT(*r500_disp_int)) {
+		if (G_007EDC_DC_HOT_PLUG_DETECT2_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			tmp = RREG32(R_007D18_DC_HOT_PLUG_DETECT2_INT_CONTROL);
 			tmp |= S_007D18_DC_HOT_PLUG_DETECT2_INT_ACK(1);
 			WREG32(R_007D18_DC_HOT_PLUG_DETECT2_INT_CONTROL, tmp);
 		}
 	} else {
-		*r500_disp_int = 0;
+		rdev->irq.stat_regs.r500.disp_int = 0;
 	}
 
 	if (irqs) {
@@ -578,32 +630,30 @@ static inline uint32_t rs600_irq_ack(struct radeon_device *rdev, u32 *r500_disp_
 
 void rs600_irq_disable(struct radeon_device *rdev)
 {
-	u32 tmp;
-
 	WREG32(R_000040_GEN_INT_CNTL, 0);
 	WREG32(R_006540_DxMODE_INT_MASK, 0);
 	/* Wait and acknowledge irq */
 	mdelay(1);
-	rs600_irq_ack(rdev, &tmp);
+	rs600_irq_ack(rdev);
 }
 
 int rs600_irq_process(struct radeon_device *rdev)
 {
-	uint32_t status, msi_rearm;
-	uint32_t r500_disp_int;
+	u32 status, msi_rearm;
 	bool queue_hotplug = false;
 
 	/* reset gui idle ack.  the status bit is broken */
 	rdev->irq.gui_idle_acked = false;
 
-	status = rs600_irq_ack(rdev, &r500_disp_int);
-	if (!status && !r500_disp_int) {
+	status = rs600_irq_ack(rdev);
+	if (!status && !rdev->irq.stat_regs.r500.disp_int) {
 		return IRQ_NONE;
 	}
-	while (status || r500_disp_int) {
+	while (status || rdev->irq.stat_regs.r500.disp_int) {
 		/* SW interrupt */
-		if (G_000044_SW_INT(status))
+		if (G_000044_SW_INT(status)) {
 			radeon_fence_process(rdev);
+		}
 		/* GUI idle */
 		if (G_000040_GUI_IDLE(status)) {
 			rdev->irq.gui_idle_acked = true;
@@ -611,25 +661,33 @@ int rs600_irq_process(struct radeon_device *rdev)
 			wake_up(&rdev->irq.idle_queue);
 		}
 		/* Vertical blank interrupts */
-		if (G_007EDC_LB_D1_VBLANK_INTERRUPT(r500_disp_int)) {
-			drm_handle_vblank(rdev->ddev, 0);
-			rdev->pm.vblank_sync = true;
-			wake_up(&rdev->irq.vblank_queue);
+		if (G_007EDC_LB_D1_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
+			if (rdev->irq.crtc_vblank_int[0]) {
+				drm_handle_vblank(rdev->ddev, 0);
+				rdev->pm.vblank_sync = true;
+				wake_up(&rdev->irq.vblank_queue);
+			}
+			if (rdev->irq.pflip[0])
+				radeon_crtc_handle_flip(rdev, 0);
 		}
-		if (G_007EDC_LB_D2_VBLANK_INTERRUPT(r500_disp_int)) {
-			drm_handle_vblank(rdev->ddev, 1);
-			rdev->pm.vblank_sync = true;
-			wake_up(&rdev->irq.vblank_queue);
+		if (G_007EDC_LB_D2_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
+			if (rdev->irq.crtc_vblank_int[1]) {
+				drm_handle_vblank(rdev->ddev, 1);
+				rdev->pm.vblank_sync = true;
+				wake_up(&rdev->irq.vblank_queue);
+			}
+			if (rdev->irq.pflip[1])
+				radeon_crtc_handle_flip(rdev, 1);
 		}
-		if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(r500_disp_int)) {
+		if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			queue_hotplug = true;
 			DRM_DEBUG("HPD1\n");
 		}
-		if (G_007EDC_DC_HOT_PLUG_DETECT2_INTERRUPT(r500_disp_int)) {
+		if (G_007EDC_DC_HOT_PLUG_DETECT2_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			queue_hotplug = true;
 			DRM_DEBUG("HPD2\n");
 		}
-		status = rs600_irq_ack(rdev, &r500_disp_int);
+		status = rs600_irq_ack(rdev);
 	}
 	/* reset gui idle ack.  the status bit is broken */
 	rdev->irq.gui_idle_acked = false;
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 4dfead8cee33..7c2e0b19a558 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -42,6 +42,40 @@
 static void rv770_gpu_init(struct radeon_device *rdev);
 void rv770_fini(struct radeon_device *rdev);
 
+u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
+
+	/* Lock the graphics update lock */
+	tmp |= AVIVO_D1GRPH_UPDATE_LOCK;
+	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+
+	/* update the scanout addresses */
+	if (radeon_crtc->crtc_id) {
+		WREG32(D2GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, upper_32_bits(crtc_base));
+		WREG32(D2GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, upper_32_bits(crtc_base));
+	} else {
+		WREG32(D1GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, upper_32_bits(crtc_base));
+		WREG32(D1GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, upper_32_bits(crtc_base));
+	}
+	WREG32(D1GRPH_SECONDARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+	       (u32)crtc_base);
+	WREG32(D1GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+	       (u32)crtc_base);
+
+	/* Wait for update_pending to go high. */
+	while (!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING));
+	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
+
+	/* Unlock the lock, so double-buffering can take place inside vblank */
+	tmp &= ~AVIVO_D1GRPH_UPDATE_LOCK;
+	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+
+	/* Return current update_pending status: */
+	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
+}
+
 /* get temperature in millidegrees */
 u32 rv770_get_temp(struct radeon_device *rdev)
 {
@@ -237,6 +271,12 @@ static void rv770_mc_program(struct radeon_device *rdev)
 			rdev->mc.vram_end >> 12);
 	}
 	WREG32(MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR, 0);
+	if (rdev->flags & RADEON_IS_IGP) {
+		tmp = RREG32(MC_FUS_VM_FB_OFFSET) & 0x000FFFFF;
+		tmp |= ((rdev->mc.vram_end >> 20) & 0xF) << 24;
+		tmp |= ((rdev->mc.vram_start >> 20) & 0xF) << 20;
+		WREG32(MC_FUS_VM_FB_OFFSET, tmp);
+	}
 	tmp = ((rdev->mc.vram_end >> 24) & 0xFFFF) << 16;
 	tmp |= ((rdev->mc.vram_start >> 24) & 0xFFFF);
 	WREG32(MC_VM_FB_LOCATION, tmp);
@@ -489,6 +529,49 @@ static u32 r700_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	return backend_map;
 }
 
+static void rv770_program_channel_remap(struct radeon_device *rdev)
+{
+	u32 tcp_chan_steer, mc_shared_chremap, tmp;
+	bool force_no_swizzle;
+
+	switch (rdev->family) {
+	case CHIP_RV770:
+	case CHIP_RV730:
+		force_no_swizzle = false;
+		break;
+	case CHIP_RV710:
+	case CHIP_RV740:
+	default:
+		force_no_swizzle = true;
+		break;
+	}
+
+	tmp = RREG32(MC_SHARED_CHMAP);
+	switch ((tmp & NOOFCHAN_MASK) >> NOOFCHAN_SHIFT) {
+	case 0:
+	case 1:
+	default:
+		/* default mapping */
+		mc_shared_chremap = 0x00fac688;
+		break;
+	case 2:
+	case 3:
+		if (force_no_swizzle)
+			mc_shared_chremap = 0x00fac688;
+		else
+			mc_shared_chremap = 0x00bbc298;
+		break;
+	}
+
+	if (rdev->family == CHIP_RV740)
+		tcp_chan_steer = 0x00ef2a60;
+	else
+		tcp_chan_steer = 0x00fac688;
+
+	WREG32(TCP_CHAN_STEER, tcp_chan_steer);
+	WREG32(MC_SHARED_CHREMAP, mc_shared_chremap);
+}
+
 static void rv770_gpu_init(struct radeon_device *rdev)
 {
 	int i, j, num_qd_pipes;
@@ -688,6 +771,8 @@ static void rv770_gpu_init(struct radeon_device *rdev)
 	WREG32(DCP_TILING_CONFIG, (gb_tiling_config & 0xffff));
 	WREG32(HDP_TILING_CONFIG, (gb_tiling_config & 0xffff));
 
+	rv770_program_channel_remap(rdev);
+
 	WREG32(CC_RB_BACKEND_DISABLE,      cc_rb_backend_disable);
 	WREG32(CC_GC_SHADER_PIPE_CONFIG,   cc_gc_shader_pipe_config);
 	WREG32(GC_USER_SHADER_PIPE_CONFIG, cc_gc_shader_pipe_config);
@@ -956,6 +1041,50 @@ static void rv770_vram_scratch_fini(struct radeon_device *rdev)
 	radeon_bo_unref(&rdev->vram_scratch.robj);
 }
 
+void r700_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc)
+{
+	u64 size_bf, size_af;
+
+	if (mc->mc_vram_size > 0xE0000000) {
+		/* leave room for at least 512M GTT */
+		dev_warn(rdev->dev, "limiting VRAM\n");
+		mc->real_vram_size = 0xE0000000;
+		mc->mc_vram_size = 0xE0000000;
+	}
+	if (rdev->flags & RADEON_IS_AGP) {
+		size_bf = mc->gtt_start;
+		size_af = 0xFFFFFFFF - mc->gtt_end + 1;
+		if (size_bf > size_af) {
+			if (mc->mc_vram_size > size_bf) {
+				dev_warn(rdev->dev, "limiting VRAM\n");
+				mc->real_vram_size = size_bf;
+				mc->mc_vram_size = size_bf;
+			}
+			mc->vram_start = mc->gtt_start - mc->mc_vram_size;
+		} else {
+			if (mc->mc_vram_size > size_af) {
+				dev_warn(rdev->dev, "limiting VRAM\n");
+				mc->real_vram_size = size_af;
+				mc->mc_vram_size = size_af;
+			}
+			mc->vram_start = mc->gtt_end;
+		}
+		mc->vram_end = mc->vram_start + mc->mc_vram_size - 1;
+		dev_info(rdev->dev, "VRAM: %lluM 0x%08llX - 0x%08llX (%lluM used)\n",
+				mc->mc_vram_size >> 20, mc->vram_start,
+				mc->vram_end, mc->real_vram_size >> 20);
+	} else {
+		u64 base = 0;
+		if (rdev->flags & RADEON_IS_IGP) {
+			base = (RREG32(MC_VM_FB_LOCATION) & 0xFFFF) << 24;
+			base |= RREG32(MC_FUS_VM_FB_OFFSET) & 0x00F00000;
+		}
+		radeon_vram_location(rdev, &rdev->mc, base);
+		rdev->mc.gtt_base_align = 0;
+		radeon_gtt_location(rdev, mc);
+	}
+}
+
 int rv770_mc_init(struct radeon_device *rdev)
 {
 	u32 tmp;
@@ -996,7 +1125,7 @@ int rv770_mc_init(struct radeon_device *rdev)
 	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE);
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
 	rdev->mc.active_vram_size = rdev->mc.visible_vram_size;
-	r600_vram_gtt_location(rdev, &rdev->mc);
+	r700_vram_gtt_location(rdev, &rdev->mc);
 	radeon_update_bandwidth_info(rdev);
 
 	return 0;
diff --git a/drivers/gpu/drm/radeon/rv770d.h b/drivers/gpu/drm/radeon/rv770d.h
index b7a5a20e81dc..98f9ad256d3d 100644
--- a/drivers/gpu/drm/radeon/rv770d.h
+++ b/drivers/gpu/drm/radeon/rv770d.h
@@ -138,6 +138,7 @@
 #define MC_SHARED_CHMAP						0x2004
 #define		NOOFCHAN_SHIFT					12
 #define		NOOFCHAN_MASK					0x00003000
+#define MC_SHARED_CHREMAP					0x2008
 
 #define	MC_ARB_RAMCFG					0x2760
 #define		NOOFBANK_SHIFT					0
@@ -157,6 +158,7 @@
 #define	MC_VM_AGP_BOT					0x202C
 #define	MC_VM_AGP_BASE					0x2030
 #define	MC_VM_FB_LOCATION				0x2024
+#define	MC_FUS_VM_FB_OFFSET				0x2898
 #define	MC_VM_MB_L1_TLB0_CNTL				0x2234
 #define	MC_VM_MB_L1_TLB1_CNTL				0x2238
 #define	MC_VM_MB_L1_TLB2_CNTL				0x223C
@@ -303,6 +305,7 @@
 #define		BILINEAR_PRECISION_8_BIT			(1 << 31)
 
 #define	TCP_CNTL					0x9610
+#define	TCP_CHAN_STEER					0x9614
 
 #define	VGT_CACHE_INVALIDATION				0x88C4
 #define		CACHE_INVALIDATION(x)				((x)<<0)
@@ -351,4 +354,11 @@
 
 #define	SRBM_STATUS				        0x0E50
 
+#define D1GRPH_PRIMARY_SURFACE_ADDRESS                    0x6110
+#define D1GRPH_PRIMARY_SURFACE_ADDRESS_HIGH               0x6914
+#define D2GRPH_PRIMARY_SURFACE_ADDRESS_HIGH               0x6114
+#define D1GRPH_SECONDARY_SURFACE_ADDRESS                  0x6118
+#define D1GRPH_SECONDARY_SURFACE_ADDRESS_HIGH             0x691c
+#define D2GRPH_SECONDARY_SURFACE_ADDRESS_HIGH             0x611c
+
 #endif
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 148a322d8f5d..cf2ec562550e 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -169,7 +169,7 @@ int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo, bool interruptible)
 }
 EXPORT_SYMBOL(ttm_bo_wait_unreserved);
 
-static void ttm_bo_add_to_lru(struct ttm_buffer_object *bo)
+void ttm_bo_add_to_lru(struct ttm_buffer_object *bo)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
 	struct ttm_mem_type_manager *man;
@@ -191,11 +191,7 @@ static void ttm_bo_add_to_lru(struct ttm_buffer_object *bo)
 	}
 }
 
-/**
- * Call with the lru_lock held.
- */
-
-static int ttm_bo_del_from_lru(struct ttm_buffer_object *bo)
+int ttm_bo_del_from_lru(struct ttm_buffer_object *bo)
 {
 	int put_count = 0;
 
@@ -227,9 +223,18 @@ int ttm_bo_reserve_locked(struct ttm_buffer_object *bo,
 		/**
 		 * Deadlock avoidance for multi-bo reserving.
 		 */
-		if (use_sequence && bo->seq_valid &&
-			(sequence - bo->val_seq < (1 << 31))) {
-			return -EAGAIN;
+		if (use_sequence && bo->seq_valid) {
+			/**
+			 * We've already reserved this one.
+			 */
+			if (unlikely(sequence == bo->val_seq))
+				return -EDEADLK;
+			/**
+			 * Already reserved by a thread that will not back
+			 * off for us. We need to back off.
+			 */
+			if (unlikely(sequence - bo->val_seq < (1 << 31)))
+				return -EAGAIN;
 		}
 
 		if (no_wait)
@@ -267,6 +272,13 @@ static void ttm_bo_ref_bug(struct kref *list_kref)
 	BUG();
 }
 
+void ttm_bo_list_ref_sub(struct ttm_buffer_object *bo, int count,
+			 bool never_free)
+{
+	kref_sub(&bo->list_kref, count,
+		 (never_free) ? ttm_bo_ref_bug : ttm_bo_release_list);
+}
+
 int ttm_bo_reserve(struct ttm_buffer_object *bo,
 		   bool interruptible,
 		   bool no_wait, bool use_sequence, uint32_t sequence)
@@ -282,20 +294,24 @@ int ttm_bo_reserve(struct ttm_buffer_object *bo,
 		put_count = ttm_bo_del_from_lru(bo);
 	spin_unlock(&glob->lru_lock);
 
-	while (put_count--)
-		kref_put(&bo->list_kref, ttm_bo_ref_bug);
+	ttm_bo_list_ref_sub(bo, put_count, true);
 
 	return ret;
 }
 
+void ttm_bo_unreserve_locked(struct ttm_buffer_object *bo)
+{
+	ttm_bo_add_to_lru(bo);
+	atomic_set(&bo->reserved, 0);
+	wake_up_all(&bo->event_queue);
+}
+
 void ttm_bo_unreserve(struct ttm_buffer_object *bo)
 {
 	struct ttm_bo_global *glob = bo->glob;
 
 	spin_lock(&glob->lru_lock);
-	ttm_bo_add_to_lru(bo);
-	atomic_set(&bo->reserved, 0);
-	wake_up_all(&bo->event_queue);
+	ttm_bo_unreserve_locked(bo);
 	spin_unlock(&glob->lru_lock);
 }
 EXPORT_SYMBOL(ttm_bo_unreserve);
@@ -362,8 +378,13 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo,
 	int ret = 0;
 
 	if (old_is_pci || new_is_pci ||
-	    ((mem->placement & bo->mem.placement & TTM_PL_MASK_CACHING) == 0))
-		ttm_bo_unmap_virtual(bo);
+	    ((mem->placement & bo->mem.placement & TTM_PL_MASK_CACHING) == 0)) {
+		ret = ttm_mem_io_lock(old_man, true);
+		if (unlikely(ret != 0))
+			goto out_err;
+		ttm_bo_unmap_virtual_locked(bo);
+		ttm_mem_io_unlock(old_man);
+	}
 
 	/*
 	 * Create and bind a ttm if required.
@@ -416,11 +437,9 @@ moved:
 	}
 
 	if (bo->mem.mm_node) {
-		spin_lock(&bo->lock);
 		bo->offset = (bo->mem.start << PAGE_SHIFT) +
 		    bdev->man[bo->mem.mem_type].gpu_offset;
 		bo->cur_placement = bo->mem.placement;
-		spin_unlock(&bo->lock);
 	} else
 		bo->offset = 0;
 
@@ -452,7 +471,6 @@ static void ttm_bo_cleanup_memtype_use(struct ttm_buffer_object *bo)
 		ttm_tt_destroy(bo->ttm);
 		bo->ttm = NULL;
 	}
-
 	ttm_bo_mem_put(bo, &bo->mem);
 
 	atomic_set(&bo->reserved, 0);
@@ -474,14 +492,14 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 	int put_count;
 	int ret;
 
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	(void) ttm_bo_wait(bo, false, false, true);
 	if (!bo->sync_obj) {
 
 		spin_lock(&glob->lru_lock);
 
 		/**
-		 * Lock inversion between bo::reserve and bo::lock here,
+		 * Lock inversion between bo:reserve and bdev::fence_lock here,
 		 * but that's OK, since we're only trylocking.
 		 */
 
@@ -490,14 +508,13 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 		if (unlikely(ret == -EBUSY))
 			goto queue;
 
-		spin_unlock(&bo->lock);
+		spin_unlock(&bdev->fence_lock);
 		put_count = ttm_bo_del_from_lru(bo);
 
 		spin_unlock(&glob->lru_lock);
 		ttm_bo_cleanup_memtype_use(bo);
 
-		while (put_count--)
-			kref_put(&bo->list_kref, ttm_bo_ref_bug);
+		ttm_bo_list_ref_sub(bo, put_count, true);
 
 		return;
 	} else {
@@ -512,7 +529,7 @@ queue:
 	kref_get(&bo->list_kref);
 	list_add_tail(&bo->ddestroy, &bdev->ddestroy);
 	spin_unlock(&glob->lru_lock);
-	spin_unlock(&bo->lock);
+	spin_unlock(&bdev->fence_lock);
 
 	if (sync_obj) {
 		driver->sync_obj_flush(sync_obj, sync_obj_arg);
@@ -537,14 +554,15 @@ static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo,
 			       bool no_wait_reserve,
 			       bool no_wait_gpu)
 {
+	struct ttm_bo_device *bdev = bo->bdev;
 	struct ttm_bo_global *glob = bo->glob;
 	int put_count;
 	int ret = 0;
 
 retry:
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	ret = ttm_bo_wait(bo, false, interruptible, no_wait_gpu);
-	spin_unlock(&bo->lock);
+	spin_unlock(&bdev->fence_lock);
 
 	if (unlikely(ret != 0))
 		return ret;
@@ -580,8 +598,7 @@ retry:
 	spin_unlock(&glob->lru_lock);
 	ttm_bo_cleanup_memtype_use(bo);
 
-	while (put_count--)
-		kref_put(&bo->list_kref, ttm_bo_ref_bug);
+	ttm_bo_list_ref_sub(bo, put_count, true);
 
 	return 0;
 }
@@ -652,6 +669,7 @@ static void ttm_bo_release(struct kref *kref)
 	struct ttm_buffer_object *bo =
 	    container_of(kref, struct ttm_buffer_object, kref);
 	struct ttm_bo_device *bdev = bo->bdev;
+	struct ttm_mem_type_manager *man = &bdev->man[bo->mem.mem_type];
 
 	if (likely(bo->vm_node != NULL)) {
 		rb_erase(&bo->vm_rb, &bdev->addr_space_rb);
@@ -659,6 +677,9 @@ static void ttm_bo_release(struct kref *kref)
 		bo->vm_node = NULL;
 	}
 	write_unlock(&bdev->vm_lock);
+	ttm_mem_io_lock(man, false);
+	ttm_mem_io_free_vm(bo);
+	ttm_mem_io_unlock(man);
 	ttm_bo_cleanup_refs_or_queue(bo);
 	kref_put(&bo->list_kref, ttm_bo_release_list);
 	write_lock(&bdev->vm_lock);
@@ -698,9 +719,9 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo, bool interruptible,
 	struct ttm_placement placement;
 	int ret = 0;
 
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	ret = ttm_bo_wait(bo, false, interruptible, no_wait_gpu);
-	spin_unlock(&bo->lock);
+	spin_unlock(&bdev->fence_lock);
 
 	if (unlikely(ret != 0)) {
 		if (ret != -ERESTARTSYS) {
@@ -715,7 +736,8 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo, bool interruptible,
 
 	evict_mem = bo->mem;
 	evict_mem.mm_node = NULL;
-	evict_mem.bus.io_reserved = false;
+	evict_mem.bus.io_reserved_vm = false;
+	evict_mem.bus.io_reserved_count = 0;
 
 	placement.fpfn = 0;
 	placement.lpfn = 0;
@@ -802,8 +824,7 @@ retry:
 
 	BUG_ON(ret != 0);
 
-	while (put_count--)
-		kref_put(&bo->list_kref, ttm_bo_ref_bug);
+	ttm_bo_list_ref_sub(bo, put_count, true);
 
 	ret = ttm_bo_evict(bo, interruptible, no_wait_reserve, no_wait_gpu);
 	ttm_bo_unreserve(bo);
@@ -1036,6 +1057,7 @@ int ttm_bo_move_buffer(struct ttm_buffer_object *bo,
 {
 	int ret = 0;
 	struct ttm_mem_reg mem;
+	struct ttm_bo_device *bdev = bo->bdev;
 
 	BUG_ON(!atomic_read(&bo->reserved));
 
@@ -1044,15 +1066,16 @@ int ttm_bo_move_buffer(struct ttm_buffer_object *bo,
 	 * Have the driver move function wait for idle when necessary,
 	 * instead of doing it here.
 	 */
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	ret = ttm_bo_wait(bo, false, interruptible, no_wait_gpu);
-	spin_unlock(&bo->lock);
+	spin_unlock(&bdev->fence_lock);
 	if (ret)
 		return ret;
 	mem.num_pages = bo->num_pages;
 	mem.size = mem.num_pages << PAGE_SHIFT;
 	mem.page_alignment = bo->mem.page_alignment;
-	mem.bus.io_reserved = false;
+	mem.bus.io_reserved_vm = false;
+	mem.bus.io_reserved_count = 0;
 	/*
 	 * Determine where to move the buffer.
 	 */
@@ -1163,7 +1186,6 @@ int ttm_bo_init(struct ttm_bo_device *bdev,
 	}
 	bo->destroy = destroy;
 
-	spin_lock_init(&bo->lock);
 	kref_init(&bo->kref);
 	kref_init(&bo->list_kref);
 	atomic_set(&bo->cpu_writers, 0);
@@ -1172,6 +1194,7 @@ int ttm_bo_init(struct ttm_bo_device *bdev,
 	INIT_LIST_HEAD(&bo->lru);
 	INIT_LIST_HEAD(&bo->ddestroy);
 	INIT_LIST_HEAD(&bo->swap);
+	INIT_LIST_HEAD(&bo->io_reserve_lru);
 	bo->bdev = bdev;
 	bo->glob = bdev->glob;
 	bo->type = type;
@@ -1181,7 +1204,8 @@ int ttm_bo_init(struct ttm_bo_device *bdev,
 	bo->mem.num_pages = bo->num_pages;
 	bo->mem.mm_node = NULL;
 	bo->mem.page_alignment = page_alignment;
-	bo->mem.bus.io_reserved = false;
+	bo->mem.bus.io_reserved_vm = false;
+	bo->mem.bus.io_reserved_count = 0;
 	bo->buffer_start = buffer_start & PAGE_MASK;
 	bo->priv_flags = 0;
 	bo->mem.placement = (TTM_PL_FLAG_SYSTEM | TTM_PL_FLAG_CACHED);
@@ -1355,6 +1379,10 @@ int ttm_bo_init_mm(struct ttm_bo_device *bdev, unsigned type,
 	BUG_ON(type >= TTM_NUM_MEM_TYPES);
 	man = &bdev->man[type];
 	BUG_ON(man->has_type);
+	man->io_reserve_fastpath = true;
+	man->use_io_reserve_lru = false;
+	mutex_init(&man->io_reserve_mutex);
+	INIT_LIST_HEAD(&man->io_reserve_lru);
 
 	ret = bdev->driver->init_mem_type(bdev, type, man);
 	if (ret)
@@ -1527,7 +1555,8 @@ int ttm_bo_device_init(struct ttm_bo_device *bdev,
 	bdev->dev_mapping = NULL;
 	bdev->glob = glob;
 	bdev->need_dma32 = need_dma32;
-
+	bdev->val_seq = 0;
+	spin_lock_init(&bdev->fence_lock);
 	mutex_lock(&glob->device_list_mutex);
 	list_add_tail(&bdev->device_list, &glob->device_list);
 	mutex_unlock(&glob->device_list_mutex);
@@ -1561,7 +1590,7 @@ bool ttm_mem_reg_is_pci(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
 	return true;
 }
 
-void ttm_bo_unmap_virtual(struct ttm_buffer_object *bo)
+void ttm_bo_unmap_virtual_locked(struct ttm_buffer_object *bo)
 {
 	struct ttm_bo_device *bdev = bo->bdev;
 	loff_t offset = (loff_t) bo->addr_space_offset;
@@ -1570,8 +1599,20 @@ void ttm_bo_unmap_virtual(struct ttm_buffer_object *bo)
 	if (!bdev->dev_mapping)
 		return;
 	unmap_mapping_range(bdev->dev_mapping, offset, holelen, 1);
-	ttm_mem_io_free(bdev, &bo->mem);
+	ttm_mem_io_free_vm(bo);
+}
+
+void ttm_bo_unmap_virtual(struct ttm_buffer_object *bo)
+{
+	struct ttm_bo_device *bdev = bo->bdev;
+	struct ttm_mem_type_manager *man = &bdev->man[bo->mem.mem_type];
+
+	ttm_mem_io_lock(man, false);
+	ttm_bo_unmap_virtual_locked(bo);
+	ttm_mem_io_unlock(man);
 }
+
+
 EXPORT_SYMBOL(ttm_bo_unmap_virtual);
 
 static void ttm_bo_vm_insert_rb(struct ttm_buffer_object *bo)
@@ -1651,6 +1692,7 @@ int ttm_bo_wait(struct ttm_buffer_object *bo,
 		bool lazy, bool interruptible, bool no_wait)
 {
 	struct ttm_bo_driver *driver = bo->bdev->driver;
+	struct ttm_bo_device *bdev = bo->bdev;
 	void *sync_obj;
 	void *sync_obj_arg;
 	int ret = 0;
@@ -1664,9 +1706,9 @@ int ttm_bo_wait(struct ttm_buffer_object *bo,
 			void *tmp_obj = bo->sync_obj;
 			bo->sync_obj = NULL;
 			clear_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags);
-			spin_unlock(&bo->lock);
+			spin_unlock(&bdev->fence_lock);
 			driver->sync_obj_unref(&tmp_obj);
-			spin_lock(&bo->lock);
+			spin_lock(&bdev->fence_lock);
 			continue;
 		}
 
@@ -1675,29 +1717,29 @@ int ttm_bo_wait(struct ttm_buffer_object *bo,
 
 		sync_obj = driver->sync_obj_ref(bo->sync_obj);
 		sync_obj_arg = bo->sync_obj_arg;
-		spin_unlock(&bo->lock);
+		spin_unlock(&bdev->fence_lock);
 		ret = driver->sync_obj_wait(sync_obj, sync_obj_arg,
 					    lazy, interruptible);
 		if (unlikely(ret != 0)) {
 			driver->sync_obj_unref(&sync_obj);
-			spin_lock(&bo->lock);
+			spin_lock(&bdev->fence_lock);
 			return ret;
 		}
-		spin_lock(&bo->lock);
+		spin_lock(&bdev->fence_lock);
 		if (likely(bo->sync_obj == sync_obj &&
 			   bo->sync_obj_arg == sync_obj_arg)) {
 			void *tmp_obj = bo->sync_obj;
 			bo->sync_obj = NULL;
 			clear_bit(TTM_BO_PRIV_FLAG_MOVING,
 				  &bo->priv_flags);
-			spin_unlock(&bo->lock);
+			spin_unlock(&bdev->fence_lock);
 			driver->sync_obj_unref(&sync_obj);
 			driver->sync_obj_unref(&tmp_obj);
-			spin_lock(&bo->lock);
+			spin_lock(&bdev->fence_lock);
 		} else {
-			spin_unlock(&bo->lock);
+			spin_unlock(&bdev->fence_lock);
 			driver->sync_obj_unref(&sync_obj);
-			spin_lock(&bo->lock);
+			spin_lock(&bdev->fence_lock);
 		}
 	}
 	return 0;
@@ -1706,6 +1748,7 @@ EXPORT_SYMBOL(ttm_bo_wait);
 
 int ttm_bo_synccpu_write_grab(struct ttm_buffer_object *bo, bool no_wait)
 {
+	struct ttm_bo_device *bdev = bo->bdev;
 	int ret = 0;
 
 	/*
@@ -1715,9 +1758,9 @@ int ttm_bo_synccpu_write_grab(struct ttm_buffer_object *bo, bool no_wait)
 	ret = ttm_bo_reserve(bo, true, no_wait, false, 0);
 	if (unlikely(ret != 0))
 		return ret;
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	ret = ttm_bo_wait(bo, false, true, no_wait);
-	spin_unlock(&bo->lock);
+	spin_unlock(&bdev->fence_lock);
 	if (likely(ret == 0))
 		atomic_inc(&bo->cpu_writers);
 	ttm_bo_unreserve(bo);
@@ -1783,16 +1826,15 @@ static int ttm_bo_swapout(struct ttm_mem_shrink *shrink)
 	put_count = ttm_bo_del_from_lru(bo);
 	spin_unlock(&glob->lru_lock);
 
-	while (put_count--)
-		kref_put(&bo->list_kref, ttm_bo_ref_bug);
+	ttm_bo_list_ref_sub(bo, put_count, true);
 
 	/**
 	 * Wait for GPU, then move to system cached.
 	 */
 
-	spin_lock(&bo->lock);
+	spin_lock(&bo->bdev->fence_lock);
 	ret = ttm_bo_wait(bo, false, false, false);
-	spin_unlock(&bo->lock);
+	spin_unlock(&bo->bdev->fence_lock);
 
 	if (unlikely(ret != 0))
 		goto out;
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 3106d5bcce32..a89839f83f6c 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -75,37 +75,123 @@ int ttm_bo_move_ttm(struct ttm_buffer_object *bo,
 }
 EXPORT_SYMBOL(ttm_bo_move_ttm);
 
-int ttm_mem_io_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
+int ttm_mem_io_lock(struct ttm_mem_type_manager *man, bool interruptible)
 {
-	int ret;
+	if (likely(man->io_reserve_fastpath))
+		return 0;
+
+	if (interruptible)
+		return mutex_lock_interruptible(&man->io_reserve_mutex);
+
+	mutex_lock(&man->io_reserve_mutex);
+	return 0;
+}
 
-	if (!mem->bus.io_reserved) {
-		mem->bus.io_reserved = true;
+void ttm_mem_io_unlock(struct ttm_mem_type_manager *man)
+{
+	if (likely(man->io_reserve_fastpath))
+		return;
+
+	mutex_unlock(&man->io_reserve_mutex);
+}
+
+static int ttm_mem_io_evict(struct ttm_mem_type_manager *man)
+{
+	struct ttm_buffer_object *bo;
+
+	if (!man->use_io_reserve_lru || list_empty(&man->io_reserve_lru))
+		return -EAGAIN;
+
+	bo = list_first_entry(&man->io_reserve_lru,
+			      struct ttm_buffer_object,
+			      io_reserve_lru);
+	list_del_init(&bo->io_reserve_lru);
+	ttm_bo_unmap_virtual_locked(bo);
+
+	return 0;
+}
+
+static int ttm_mem_io_reserve(struct ttm_bo_device *bdev,
+			      struct ttm_mem_reg *mem)
+{
+	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
+	int ret = 0;
+
+	if (!bdev->driver->io_mem_reserve)
+		return 0;
+	if (likely(man->io_reserve_fastpath))
+		return bdev->driver->io_mem_reserve(bdev, mem);
+
+	if (bdev->driver->io_mem_reserve &&
+	    mem->bus.io_reserved_count++ == 0) {
+retry:
 		ret = bdev->driver->io_mem_reserve(bdev, mem);
+		if (ret == -EAGAIN) {
+			ret = ttm_mem_io_evict(man);
+			if (ret == 0)
+				goto retry;
+		}
+	}
+	return ret;
+}
+
+static void ttm_mem_io_free(struct ttm_bo_device *bdev,
+			    struct ttm_mem_reg *mem)
+{
+	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
+
+	if (likely(man->io_reserve_fastpath))
+		return;
+
+	if (bdev->driver->io_mem_reserve &&
+	    --mem->bus.io_reserved_count == 0 &&
+	    bdev->driver->io_mem_free)
+		bdev->driver->io_mem_free(bdev, mem);
+
+}
+
+int ttm_mem_io_reserve_vm(struct ttm_buffer_object *bo)
+{
+	struct ttm_mem_reg *mem = &bo->mem;
+	int ret;
+
+	if (!mem->bus.io_reserved_vm) {
+		struct ttm_mem_type_manager *man =
+			&bo->bdev->man[mem->mem_type];
+
+		ret = ttm_mem_io_reserve(bo->bdev, mem);
 		if (unlikely(ret != 0))
 			return ret;
+		mem->bus.io_reserved_vm = true;
+		if (man->use_io_reserve_lru)
+			list_add_tail(&bo->io_reserve_lru,
+				      &man->io_reserve_lru);
 	}
 	return 0;
 }
 
-void ttm_mem_io_free(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
+void ttm_mem_io_free_vm(struct ttm_buffer_object *bo)
 {
-	if (bdev->driver->io_mem_reserve) {
-		if (mem->bus.io_reserved) {
-			mem->bus.io_reserved = false;
-			bdev->driver->io_mem_free(bdev, mem);
-		}
+	struct ttm_mem_reg *mem = &bo->mem;
+
+	if (mem->bus.io_reserved_vm) {
+		mem->bus.io_reserved_vm = false;
+		list_del_init(&bo->io_reserve_lru);
+		ttm_mem_io_free(bo->bdev, mem);
 	}
 }
 
 int ttm_mem_reg_ioremap(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem,
 			void **virtual)
 {
+	struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
 	int ret;
 	void *addr;
 
 	*virtual = NULL;
+	(void) ttm_mem_io_lock(man, false);
 	ret = ttm_mem_io_reserve(bdev, mem);
+	ttm_mem_io_unlock(man);
 	if (ret || !mem->bus.is_iomem)
 		return ret;
 
@@ -117,7 +203,9 @@ int ttm_mem_reg_ioremap(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem,
 		else
 			addr = ioremap_nocache(mem->bus.base + mem->bus.offset, mem->bus.size);
 		if (!addr) {
+			(void) ttm_mem_io_lock(man, false);
 			ttm_mem_io_free(bdev, mem);
+			ttm_mem_io_unlock(man);
 			return -ENOMEM;
 		}
 	}
@@ -134,7 +222,9 @@ void ttm_mem_reg_iounmap(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem,
 
 	if (virtual && mem->bus.addr == NULL)
 		iounmap(virtual);
+	(void) ttm_mem_io_lock(man, false);
 	ttm_mem_io_free(bdev, mem);
+	ttm_mem_io_unlock(man);
 }
 
 static int ttm_copy_io_page(void *dst, void *src, unsigned long page)
@@ -231,7 +321,7 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	struct ttm_mem_type_manager *man = &bdev->man[new_mem->mem_type];
 	struct ttm_tt *ttm = bo->ttm;
 	struct ttm_mem_reg *old_mem = &bo->mem;
-	struct ttm_mem_reg old_copy = *old_mem;
+	struct ttm_mem_reg old_copy;
 	void *old_iomap;
 	void *new_iomap;
 	int ret;
@@ -281,7 +371,7 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 	mb();
 out2:
 	ttm_bo_free_old_node(bo);
-
+	old_copy = *old_mem;
 	*old_mem = *new_mem;
 	new_mem->mm_node = NULL;
 
@@ -292,7 +382,7 @@ out2:
 	}
 
 out1:
-	ttm_mem_reg_iounmap(bdev, new_mem, new_iomap);
+	ttm_mem_reg_iounmap(bdev, old_mem, new_iomap);
 out:
 	ttm_mem_reg_iounmap(bdev, &old_copy, old_iomap);
 	return ret;
@@ -337,11 +427,11 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
 	 * TODO: Explicit member copy would probably be better here.
 	 */
 
-	spin_lock_init(&fbo->lock);
 	init_waitqueue_head(&fbo->event_queue);
 	INIT_LIST_HEAD(&fbo->ddestroy);
 	INIT_LIST_HEAD(&fbo->lru);
 	INIT_LIST_HEAD(&fbo->swap);
+	INIT_LIST_HEAD(&fbo->io_reserve_lru);
 	fbo->vm_node = NULL;
 	atomic_set(&fbo->cpu_writers, 0);
 
@@ -453,6 +543,8 @@ int ttm_bo_kmap(struct ttm_buffer_object *bo,
 		unsigned long start_page, unsigned long num_pages,
 		struct ttm_bo_kmap_obj *map)
 {
+	struct ttm_mem_type_manager *man =
+		&bo->bdev->man[bo->mem.mem_type];
 	unsigned long offset, size;
 	int ret;
 
@@ -467,7 +559,9 @@ int ttm_bo_kmap(struct ttm_buffer_object *bo,
 	if (num_pages > 1 && !DRM_SUSER(DRM_CURPROC))
 		return -EPERM;
 #endif
+	(void) ttm_mem_io_lock(man, false);
 	ret = ttm_mem_io_reserve(bo->bdev, &bo->mem);
+	ttm_mem_io_unlock(man);
 	if (ret)
 		return ret;
 	if (!bo->mem.bus.is_iomem) {
@@ -482,12 +576,15 @@ EXPORT_SYMBOL(ttm_bo_kmap);
 
 void ttm_bo_kunmap(struct ttm_bo_kmap_obj *map)
 {
+	struct ttm_buffer_object *bo = map->bo;
+	struct ttm_mem_type_manager *man =
+		&bo->bdev->man[bo->mem.mem_type];
+
 	if (!map->virtual)
 		return;
 	switch (map->bo_kmap_type) {
 	case ttm_bo_map_iomap:
 		iounmap(map->virtual);
-		ttm_mem_io_free(map->bo->bdev, &map->bo->mem);
 		break;
 	case ttm_bo_map_vmap:
 		vunmap(map->virtual);
@@ -500,6 +597,9 @@ void ttm_bo_kunmap(struct ttm_bo_kmap_obj *map)
 	default:
 		BUG();
 	}
+	(void) ttm_mem_io_lock(man, false);
+	ttm_mem_io_free(map->bo->bdev, &map->bo->mem);
+	ttm_mem_io_unlock(man);
 	map->virtual = NULL;
 	map->page = NULL;
 }
@@ -520,7 +620,7 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
 	struct ttm_buffer_object *ghost_obj;
 	void *tmp_obj = NULL;
 
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	if (bo->sync_obj) {
 		tmp_obj = bo->sync_obj;
 		bo->sync_obj = NULL;
@@ -529,7 +629,7 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
 	bo->sync_obj_arg = sync_obj_arg;
 	if (evict) {
 		ret = ttm_bo_wait(bo, false, false, false);
-		spin_unlock(&bo->lock);
+		spin_unlock(&bdev->fence_lock);
 		if (tmp_obj)
 			driver->sync_obj_unref(&tmp_obj);
 		if (ret)
@@ -552,7 +652,7 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
 		 */
 
 		set_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags);
-		spin_unlock(&bo->lock);
+		spin_unlock(&bdev->fence_lock);
 		if (tmp_obj)
 			driver->sync_obj_unref(&tmp_obj);
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index fe6cb77899f4..221b924acebe 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -83,6 +83,8 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	int i;
 	unsigned long address = (unsigned long)vmf->virtual_address;
 	int retval = VM_FAULT_NOPAGE;
+	struct ttm_mem_type_manager *man =
+		&bdev->man[bo->mem.mem_type];
 
 	/*
 	 * Work around locking order reversal in fault / nopfn
@@ -118,24 +120,28 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	 * move.
 	 */
 
-	spin_lock(&bo->lock);
+	spin_lock(&bdev->fence_lock);
 	if (test_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags)) {
 		ret = ttm_bo_wait(bo, false, true, false);
-		spin_unlock(&bo->lock);
+		spin_unlock(&bdev->fence_lock);
 		if (unlikely(ret != 0)) {
 			retval = (ret != -ERESTARTSYS) ?
 			    VM_FAULT_SIGBUS : VM_FAULT_NOPAGE;
 			goto out_unlock;
 		}
 	} else
-		spin_unlock(&bo->lock);
+		spin_unlock(&bdev->fence_lock);
 
-
-	ret = ttm_mem_io_reserve(bdev, &bo->mem);
-	if (ret) {
-		retval = VM_FAULT_SIGBUS;
+	ret = ttm_mem_io_lock(man, true);
+	if (unlikely(ret != 0)) {
+		retval = VM_FAULT_NOPAGE;
 		goto out_unlock;
 	}
+	ret = ttm_mem_io_reserve_vm(bo);
+	if (unlikely(ret != 0)) {
+		retval = VM_FAULT_SIGBUS;
+		goto out_io_unlock;
+	}
 
 	page_offset = ((address - vma->vm_start) >> PAGE_SHIFT) +
 	    bo->vm_node->start - vma->vm_pgoff;
@@ -144,7 +150,7 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	if (unlikely(page_offset >= bo->num_pages)) {
 		retval = VM_FAULT_SIGBUS;
-		goto out_unlock;
+		goto out_io_unlock;
 	}
 
 	/*
@@ -182,7 +188,7 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 			page = ttm_tt_get_page(ttm, page_offset);
 			if (unlikely(!page && i == 0)) {
 				retval = VM_FAULT_OOM;
-				goto out_unlock;
+				goto out_io_unlock;
 			} else if (unlikely(!page)) {
 				break;
 			}
@@ -200,14 +206,15 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 		else if (unlikely(ret != 0)) {
 			retval =
 			    (ret == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS;
-			goto out_unlock;
+			goto out_io_unlock;
 		}
 
 		address += PAGE_SIZE;
 		if (unlikely(++page_offset >= page_last))
 			break;
 	}
-
+out_io_unlock:
+	ttm_mem_io_unlock(man);
 out_unlock:
 	ttm_bo_unreserve(bo);
 	return retval;
diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
index c285c2902d15..3832fe10b4df 100644
--- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c
+++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
@@ -32,7 +32,7 @@
 #include <linux/sched.h>
 #include <linux/module.h>
 
-void ttm_eu_backoff_reservation(struct list_head *list)
+static void ttm_eu_backoff_reservation_locked(struct list_head *list)
 {
 	struct ttm_validate_buffer *entry;
 
@@ -41,10 +41,77 @@ void ttm_eu_backoff_reservation(struct list_head *list)
 		if (!entry->reserved)
 			continue;
 
+		if (entry->removed) {
+			ttm_bo_add_to_lru(bo);
+			entry->removed = false;
+
+		}
 		entry->reserved = false;
-		ttm_bo_unreserve(bo);
+		atomic_set(&bo->reserved, 0);
+		wake_up_all(&bo->event_queue);
+	}
+}
+
+static void ttm_eu_del_from_lru_locked(struct list_head *list)
+{
+	struct ttm_validate_buffer *entry;
+
+	list_for_each_entry(entry, list, head) {
+		struct ttm_buffer_object *bo = entry->bo;
+		if (!entry->reserved)
+			continue;
+
+		if (!entry->removed) {
+			entry->put_count = ttm_bo_del_from_lru(bo);
+			entry->removed = true;
+		}
 	}
 }
+
+static void ttm_eu_list_ref_sub(struct list_head *list)
+{
+	struct ttm_validate_buffer *entry;
+
+	list_for_each_entry(entry, list, head) {
+		struct ttm_buffer_object *bo = entry->bo;
+
+		if (entry->put_count) {
+			ttm_bo_list_ref_sub(bo, entry->put_count, true);
+			entry->put_count = 0;
+		}
+	}
+}
+
+static int ttm_eu_wait_unreserved_locked(struct list_head *list,
+					 struct ttm_buffer_object *bo)
+{
+	struct ttm_bo_global *glob = bo->glob;
+	int ret;
+
+	ttm_eu_del_from_lru_locked(list);
+	spin_unlock(&glob->lru_lock);
+	ret = ttm_bo_wait_unreserved(bo, true);
+	spin_lock(&glob->lru_lock);
+	if (unlikely(ret != 0))
+		ttm_eu_backoff_reservation_locked(list);
+	return ret;
+}
+
+
+void ttm_eu_backoff_reservation(struct list_head *list)
+{
+	struct ttm_validate_buffer *entry;
+	struct ttm_bo_global *glob;
+
+	if (list_empty(list))
+		return;
+
+	entry = list_first_entry(list, struct ttm_validate_buffer, head);
+	glob = entry->bo->glob;
+	spin_lock(&glob->lru_lock);
+	ttm_eu_backoff_reservation_locked(list);
+	spin_unlock(&glob->lru_lock);
+}
 EXPORT_SYMBOL(ttm_eu_backoff_reservation);
 
 /*
@@ -59,37 +126,76 @@ EXPORT_SYMBOL(ttm_eu_backoff_reservation);
  * buffers in different orders.
  */
 
-int ttm_eu_reserve_buffers(struct list_head *list, uint32_t val_seq)
+int ttm_eu_reserve_buffers(struct list_head *list)
 {
+	struct ttm_bo_global *glob;
 	struct ttm_validate_buffer *entry;
 	int ret;
+	uint32_t val_seq;
+
+	if (list_empty(list))
+		return 0;
+
+	list_for_each_entry(entry, list, head) {
+		entry->reserved = false;
+		entry->put_count = 0;
+		entry->removed = false;
+	}
+
+	entry = list_first_entry(list, struct ttm_validate_buffer, head);
+	glob = entry->bo->glob;
 
 retry:
+	spin_lock(&glob->lru_lock);
+	val_seq = entry->bo->bdev->val_seq++;
+
 	list_for_each_entry(entry, list, head) {
 		struct ttm_buffer_object *bo = entry->bo;
 
-		entry->reserved = false;
-		ret = ttm_bo_reserve(bo, true, false, true, val_seq);
-		if (ret != 0) {
-			ttm_eu_backoff_reservation(list);
-			if (ret == -EAGAIN) {
-				ret = ttm_bo_wait_unreserved(bo, true);
-				if (unlikely(ret != 0))
-					return ret;
-				goto retry;
-			} else
+retry_this_bo:
+		ret = ttm_bo_reserve_locked(bo, true, true, true, val_seq);
+		switch (ret) {
+		case 0:
+			break;
+		case -EBUSY:
+			ret = ttm_eu_wait_unreserved_locked(list, bo);
+			if (unlikely(ret != 0)) {
+				spin_unlock(&glob->lru_lock);
+				ttm_eu_list_ref_sub(list);
 				return ret;
+			}
+			goto retry_this_bo;
+		case -EAGAIN:
+			ttm_eu_backoff_reservation_locked(list);
+			spin_unlock(&glob->lru_lock);
+			ttm_eu_list_ref_sub(list);
+			ret = ttm_bo_wait_unreserved(bo, true);
+			if (unlikely(ret != 0))
+				return ret;
+			goto retry;
+		default:
+			ttm_eu_backoff_reservation_locked(list);
+			spin_unlock(&glob->lru_lock);
+			ttm_eu_list_ref_sub(list);
+			return ret;
 		}
 
 		entry->reserved = true;
 		if (unlikely(atomic_read(&bo->cpu_writers) > 0)) {
-			ttm_eu_backoff_reservation(list);
+			ttm_eu_backoff_reservation_locked(list);
+			spin_unlock(&glob->lru_lock);
+			ttm_eu_list_ref_sub(list);
 			ret = ttm_bo_wait_cpu(bo, false);
 			if (ret)
 				return ret;
 			goto retry;
 		}
 	}
+
+	ttm_eu_del_from_lru_locked(list);
+	spin_unlock(&glob->lru_lock);
+	ttm_eu_list_ref_sub(list);
+
 	return 0;
 }
 EXPORT_SYMBOL(ttm_eu_reserve_buffers);
@@ -97,21 +203,36 @@ EXPORT_SYMBOL(ttm_eu_reserve_buffers);
 void ttm_eu_fence_buffer_objects(struct list_head *list, void *sync_obj)
 {
 	struct ttm_validate_buffer *entry;
+	struct ttm_buffer_object *bo;
+	struct ttm_bo_global *glob;
+	struct ttm_bo_device *bdev;
+	struct ttm_bo_driver *driver;
 
-	list_for_each_entry(entry, list, head) {
-		struct ttm_buffer_object *bo = entry->bo;
-		struct ttm_bo_driver *driver = bo->bdev->driver;
-		void *old_sync_obj;
+	if (list_empty(list))
+		return;
+
+	bo = list_first_entry(list, struct ttm_validate_buffer, head)->bo;
+	bdev = bo->bdev;
+	driver = bdev->driver;
+	glob = bo->glob;
 
-		spin_lock(&bo->lock);
-		old_sync_obj = bo->sync_obj;
+	spin_lock(&bdev->fence_lock);
+	spin_lock(&glob->lru_lock);
+
+	list_for_each_entry(entry, list, head) {
+		bo = entry->bo;
+		entry->old_sync_obj = bo->sync_obj;
 		bo->sync_obj = driver->sync_obj_ref(sync_obj);
 		bo->sync_obj_arg = entry->new_sync_obj_arg;
-		spin_unlock(&bo->lock);
-		ttm_bo_unreserve(bo);
+		ttm_bo_unreserve_locked(bo);
 		entry->reserved = false;
-		if (old_sync_obj)
-			driver->sync_obj_unref(&old_sync_obj);
+	}
+	spin_unlock(&glob->lru_lock);
+	spin_unlock(&bdev->fence_lock);
+
+	list_for_each_entry(entry, list, head) {
+		if (entry->old_sync_obj)
+			driver->sync_obj_unref(&entry->old_sync_obj);
 	}
 }
 EXPORT_SYMBOL(ttm_eu_fence_buffer_objects);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index e7a58d055041..10fc01f69c40 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -264,7 +264,6 @@ struct vmw_private {
 	 */
 
 	struct vmw_sw_context ctx;
-	uint32_t val_seq;
 	struct mutex cmdbuf_mutex;
 
 	/**
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index 76954e3528c1..41b95ed6dbcd 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -653,8 +653,7 @@ int vmw_execbuf_ioctl(struct drm_device *dev, void *data,
 	ret = vmw_cmd_check_all(dev_priv, sw_context, cmd, arg->command_size);
 	if (unlikely(ret != 0))
 		goto out_err;
-	ret = ttm_eu_reserve_buffers(&sw_context->validate_nodes,
-				     dev_priv->val_seq++);
+	ret = ttm_eu_reserve_buffers(&sw_context->validate_nodes);
 	if (unlikely(ret != 0))
 		goto out_err;