Merge tag 'block-6.0-2022-09-29' of git://git.kernel.dk/linux

Pull block fixes from Jens Axboe: "A single NVMe pull request via Christoph with a few fixes that should go into the 6.0 release: - Fix IOC_PR_CLEAR and IOC_PR_RELEASE ioctls for nvme devices (Michael Kelley) - Disable Write Zeroes on Phison E3C/E4C (Tina Hsu)" * tag 'block-6.0-2022-09-29' of git://git.kernel.dk/linux: nvme-pci: disable Write Zeroes on Phison E3C/E4C nvme: Fix IOC_PR_CLEAR and IOC_PR_RELEASE ioctls for nvme devices
author: Linus Torvalds <torvalds@linux-foundation.org> 2022-09-30 09:33:33 -0700
committer: Linus Torvalds <torvalds@linux-foundation.org> 2022-09-30 09:33:33 -0700
commit: 7bc6e90d7aa4170039abe80b9f4e8c8e4eb35091 (patch)
tree: 14bc5b5b614437063a144a439ab529e1887b003b /arch/arc/plat-sim/git:/git.zx2c4.com/ssh:/git@git.zx2c4.com
parent: Merge tag 'io_uring-6.0-2022-09-29' of git://git.kernel.dk/linux (diff)
parent: Merge tag 'nvme-6.0-2022-09-29' of git://git.infradead.org/nvme into block-6.0 (diff)
0 files changed, 0 insertions, 0 deletions
diff --git a/drivers/acpi/apei/hest.c b/drivers/acpi/apei/hest.c
index abda3786a5d7..181bc2f7bb74 100644
--- a/drivers/acpi/apei/hest.c
+++ b/drivers/acpi/apei/hest.c
@@ -139,13 +139,23 @@ static int __init hest_parse_ghes(struct acpi_hest_header *hest_hdr, void *data)
 {
 	struct platform_device *ghes_dev;
 	struct ghes_arr *ghes_arr = data;
-	int rc;
+	int rc, i;
 
 	if (hest_hdr->type != ACPI_HEST_TYPE_GENERIC_ERROR)
 		return 0;
 
 	if (!((struct acpi_hest_generic *)hest_hdr)->enabled)
 		return 0;
+	for (i = 0; i < ghes_arr->count; i++) {
+		struct acpi_hest_header *hdr;
+		ghes_dev = ghes_arr->ghes_devs[i];
+		hdr = *(struct acpi_hest_header **)ghes_dev->dev.platform_data;
+		if (hdr->source_id == hest_hdr->source_id) {
+			pr_warning(FW_WARN HEST_PFX "Duplicated hardware error source ID: %d.\n",
+				   hdr->source_id);
+			return -EIO;
+		}
+	}
 	ghes_dev = platform_device_alloc("GHES", hest_hdr->source_id);
 	if (!ghes_dev)
 		return -ENOMEM;
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index 52ca9649d769..372f9b70f7f4 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -1333,23 +1333,6 @@ int acpi_resources_are_enforced(void)
 EXPORT_SYMBOL(acpi_resources_are_enforced);
 
 /*
- * Create and initialize a spinlock.
- */
-acpi_status
-acpi_os_create_lock(acpi_spinlock *out_handle)
-{
-	spinlock_t *lock;
-
-	lock = ACPI_ALLOCATE(sizeof(spinlock_t));
-	if (!lock)
-		return AE_NO_MEMORY;
-	spin_lock_init(lock);
-	*out_handle = lock;
-
-	return AE_OK;
-}
-
-/*
  * Deallocate the memory for a spinlock.
  */
 void acpi_os_delete_lock(acpi_spinlock handle)
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index d38c40fe4ddb..41223c7f0206 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -452,7 +452,7 @@ void ahci_save_initial_config(struct device *dev,
 	}
 
 	if (mask_port_map) {
-		dev_printk(KERN_ERR, dev, "masking port_map 0x%x -> 0x%x\n",
+		dev_printk(KERN_WARNING, dev, "masking port_map 0x%x -> 0x%x\n",
 			   port_map,
 			   port_map & mask_port_map);
 		port_map &= mask_port_map;
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 736bee5dafeb..000d03ae6653 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4143,9 +4143,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 	 * Devices which choke on SETXFER.  Applies only if both the
 	 * device and controller are SATA.
 	 */
-	{ "PIONEER DVD-RW  DVRTD08",	"1.00",	ATA_HORKAGE_NOSETXFER },
-	{ "PIONEER DVD-RW  DVR-212D",	"1.28", ATA_HORKAGE_NOSETXFER },
-	{ "PIONEER DVD-RW  DVR-216D",	"1.08", ATA_HORKAGE_NOSETXFER },
+	{ "PIONEER DVD-RW  DVRTD08",	NULL,	ATA_HORKAGE_NOSETXFER },
+	{ "PIONEER DVD-RW  DVR-212D",	NULL,	ATA_HORKAGE_NOSETXFER },
+	{ "PIONEER DVD-RW  DVR-216D",	NULL,	ATA_HORKAGE_NOSETXFER },
 
 	/* End Marker */
 	{ }
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index dfb6e9d3d759..7f099d6e4e0b 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2802,10 +2802,11 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	}
 
 	/*
-	 * Some controllers can't be frozen very well and may set
-	 * spuruious error conditions during reset.  Clear accumulated
-	 * error information.  As reset is the final recovery action,
-	 * nothing is lost by doing this.
+	 * Some controllers can't be frozen very well and may set spurious
+	 * error conditions during reset.  Clear accumulated error
+	 * information and re-thaw the port if frozen.  As reset is the
+	 * final recovery action and we cross check link onlineness against
+	 * device classification later, no hotplug event is lost by this.
 	 */
 	spin_lock_irqsave(link->ap->lock, flags);
 	memset(&link->eh_info, 0, sizeof(link->eh_info));
@@ -2814,6 +2815,9 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	ap->pflags &= ~ATA_PFLAG_EH_PENDING;
 	spin_unlock_irqrestore(link->ap->lock, flags);
 
+	if (ap->pflags & ATA_PFLAG_FROZEN)
+		ata_eh_thaw_port(ap);
+
 	/*
 	 * Make sure onlineness and classification result correspond.
 	 * Hotplug could have happened during reset and some
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index d51f9795c064..927f968e99d9 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -3797,6 +3797,12 @@ EXPORT_SYMBOL_GPL(ata_sas_port_alloc);
  */
 int ata_sas_port_start(struct ata_port *ap)
 {
+	/*
+	 * the port is marked as frozen at allocation time, but if we don't
+	 * have new eh, we won't thaw it
+	 */
+	if (!ap->ops->error_handler)
+		ap->pflags &= ~ATA_PFLAG_FROZEN;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(ata_sas_port_start);
diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c
index 75a6a0c0094f..5d7f58a7e34d 100644
--- a/drivers/ata/pata_marvell.c
+++ b/drivers/ata/pata_marvell.c
@@ -161,6 +161,9 @@ static const struct pci_device_id marvell_pci_tbl[] = {
 	{ PCI_DEVICE(0x11AB, 0x6121), },
 	{ PCI_DEVICE(0x11AB, 0x6123), },
 	{ PCI_DEVICE(0x11AB, 0x6145), },
+	{ PCI_DEVICE(0x1B4B, 0x91A0), },
+	{ PCI_DEVICE(0x1B4B, 0x91A4), },
+
 	{ }	/* terminate list */
 };
 
diff --git a/drivers/ata/sata_dwc_460ex.c b/drivers/ata/sata_dwc_460ex.c
index 1c4b3aa4c7c4..dc88a39e7db8 100644
--- a/drivers/ata/sata_dwc_460ex.c
+++ b/drivers/ata/sata_dwc_460ex.c
@@ -389,7 +389,7 @@ static void sata_dwc_tf_dump(struct ata_taskfile *tf)
 /*
  * Function: get_burst_length_encode
  * arguments: datalength: length in bytes of data
- * returns value to be programmed in register corrresponding to data length
+ * returns value to be programmed in register corresponding to data length
  * This value is effectively the log(base 2) of the length
  */
 static  int get_burst_length_encode(int datalength)
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 9f9b2359f718..45d7c8fc73bd 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -30,7 +30,6 @@
 static DEFINE_MUTEX(mem_sysfs_mutex);
 
 #define MEMORY_CLASS_NAME	"memory"
-#define MIN_MEMORY_BLOCK_SIZE	(1 << SECTION_SIZE_BITS)
 
 static int sections_per_block;
 
diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index 1c291af637b3..6040717b62bb 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -367,7 +367,7 @@ EXPORT_SYMBOL_GPL(platform_device_unregister);
  *
  * Returns &struct platform_device pointer on success, or ERR_PTR() on error.
  */
-struct platform_device *__init_or_module platform_device_register_resndata(
+struct platform_device *platform_device_register_resndata(
 		struct device *parent,
 		const char *name, int id,
 		const struct resource *res, unsigned int num,
diff --git a/drivers/base/power/clock_ops.c b/drivers/base/power/clock_ops.c
index c0dd09df7be8..ad367c4139b1 100644
--- a/drivers/base/power/clock_ops.c
+++ b/drivers/base/power/clock_ops.c
@@ -291,7 +291,7 @@ static int pm_runtime_clk_notify(struct notifier_block *nb,
 {
 	struct pm_clk_notifier_block *clknb;
 	struct device *dev = data;
-	char *con_id;
+	char **con_id;
 	int error;
 
 	dev_dbg(dev, "%s() %ld\n", __func__, action);
@@ -309,8 +309,8 @@ static int pm_runtime_clk_notify(struct notifier_block *nb,
 
 		dev->pwr_domain = clknb->pwr_domain;
 		if (clknb->con_ids[0]) {
-			for (con_id = clknb->con_ids[0]; *con_id; con_id++)
-				pm_runtime_clk_add(dev, con_id);
+			for (con_id = clknb->con_ids; *con_id; con_id++)
+				pm_runtime_clk_add(dev, *con_id);
 		} else {
 			pm_runtime_clk_add(dev, NULL);
 		}
@@ -380,25 +380,25 @@ static int pm_runtime_clk_notify(struct notifier_block *nb,
 {
 	struct pm_clk_notifier_block *clknb;
 	struct device *dev = data;
-	char *con_id;
+	char **con_id;
 
 	dev_dbg(dev, "%s() %ld\n", __func__, action);
 
 	clknb = container_of(nb, struct pm_clk_notifier_block, nb);
 
 	switch (action) {
-	case BUS_NOTIFY_ADD_DEVICE:
+	case BUS_NOTIFY_BIND_DRIVER:
 		if (clknb->con_ids[0]) {
-			for (con_id = clknb->con_ids[0]; *con_id; con_id++)
-				enable_clock(dev, con_id);
+			for (con_id = clknb->con_ids; *con_id; con_id++)
+				enable_clock(dev, *con_id);
 		} else {
 			enable_clock(dev, NULL);
 		}
 		break;
-	case BUS_NOTIFY_DEL_DEVICE:
+	case BUS_NOTIFY_UNBOUND_DRIVER:
 		if (clknb->con_ids[0]) {
-			for (con_id = clknb->con_ids[0]; *con_id; con_id++)
-				disable_clock(dev, con_id);
+			for (con_id = clknb->con_ids; *con_id; con_id++)
+				disable_clock(dev, *con_id);
 		} else {
 			disable_clock(dev, NULL);
 		}
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index aa6320207745..06f09bf89cb2 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -57,7 +57,8 @@ static int async_error;
  */
 void device_pm_init(struct device *dev)
 {
-	dev->power.in_suspend = false;
+	dev->power.is_prepared = false;
+	dev->power.is_suspended = false;
 	init_completion(&dev->power.completion);
 	complete_all(&dev->power.completion);
 	dev->power.wakeup = NULL;
@@ -91,7 +92,7 @@ void device_pm_add(struct device *dev)
 	pr_debug("PM: Adding info for %s:%s\n",
 		 dev->bus ? dev->bus->name : "No Bus", dev_name(dev));
 	mutex_lock(&dpm_list_mtx);
-	if (dev->parent && dev->parent->power.in_suspend)
+	if (dev->parent && dev->parent->power.is_prepared)
 		dev_warn(dev, "parent %s should not be sleeping\n",
 			dev_name(dev->parent));
 	list_add_tail(&dev->power.entry, &dpm_list);
@@ -511,7 +512,14 @@ static int device_resume(struct device *dev, pm_message_t state, bool async)
 	dpm_wait(dev->parent, async);
 	device_lock(dev);
 
-	dev->power.in_suspend = false;
+	/*
+	 * This is a fib.  But we'll allow new children to be added below
+	 * a resumed device, even if the device hasn't been completed yet.
+	 */
+	dev->power.is_prepared = false;
+
+	if (!dev->power.is_suspended)
+		goto Unlock;
 
 	if (dev->pwr_domain) {
 		pm_dev_dbg(dev, state, "power domain ");
@@ -548,6 +556,9 @@ static int device_resume(struct device *dev, pm_message_t state, bool async)
 	}
 
  End:
+	dev->power.is_suspended = false;
+
+ Unlock:
 	device_unlock(dev);
 	complete_all(&dev->power.completion);
 
@@ -670,7 +681,7 @@ void dpm_complete(pm_message_t state)
 		struct device *dev = to_device(dpm_prepared_list.prev);
 
 		get_device(dev);
-		dev->power.in_suspend = false;
+		dev->power.is_prepared = false;
 		list_move(&dev->power.entry, &list);
 		mutex_unlock(&dpm_list_mtx);
 
@@ -835,11 +846,11 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async)
 	device_lock(dev);
 
 	if (async_error)
-		goto End;
+		goto Unlock;
 
 	if (pm_wakeup_pending()) {
 		async_error = -EBUSY;
-		goto End;
+		goto Unlock;
 	}
 
 	if (dev->pwr_domain) {
@@ -877,6 +888,9 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async)
 	}
 
  End:
+	dev->power.is_suspended = !error;
+
+ Unlock:
 	device_unlock(dev);
 	complete_all(&dev->power.completion);
 
@@ -1042,7 +1056,7 @@ int dpm_prepare(pm_message_t state)
 			put_device(dev);
 			break;
 		}
-		dev->power.in_suspend = true;
+		dev->power.is_prepared = true;
 		if (!list_empty(&dev->power.entry))
 			list_move_tail(&dev->power.entry, &dpm_prepared_list);
 		put_device(dev);
diff --git a/drivers/base/syscore.c b/drivers/base/syscore.c
index c126db3cb7d1..e8d11b6630ee 100644
--- a/drivers/base/syscore.c
+++ b/drivers/base/syscore.c
@@ -9,6 +9,7 @@
 #include <linux/syscore_ops.h>
 #include <linux/mutex.h>
 #include <linux/module.h>
+#include <linux/interrupt.h>
 
 static LIST_HEAD(syscore_ops_list);
 static DEFINE_MUTEX(syscore_ops_lock);
@@ -48,6 +49,13 @@ int syscore_suspend(void)
 	struct syscore_ops *ops;
 	int ret = 0;
 
+	pr_debug("Checking wakeup interrupts\n");
+
+	/* Return error code if there are any wakeup interrupts pending. */
+	ret = check_wakeup_irqs();
+	if (ret)
+		return ret;
+
 	WARN_ONCE(!irqs_disabled(),
 		"Interrupts enabled before system core suspend.\n");
 
diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c
index 09ef9a878ef0..cf0e63dd97da 100644
--- a/drivers/block/drbd/drbd_actlog.c
+++ b/drivers/block/drbd/drbd_actlog.c
@@ -79,7 +79,7 @@ static int _drbd_md_sync_page_io(struct drbd_conf *mdev,
 	md_io.error = 0;
 
 	if ((rw & WRITE) && !test_bit(MD_NO_FUA, &mdev->flags))
-		rw |= REQ_FUA;
+		rw |= REQ_FUA | REQ_FLUSH;
 	rw |= REQ_SYNC;
 
 	bio = bio_alloc(GFP_NOIO, 1);
diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c
index f440a02dfdb1..7b976296b564 100644
--- a/drivers/block/drbd/drbd_bitmap.c
+++ b/drivers/block/drbd/drbd_bitmap.c
@@ -112,9 +112,6 @@ struct drbd_bitmap {
 	struct task_struct *bm_task;
 };
 
-static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
-			       unsigned long e, int val, const enum km_type km);
-
 #define bm_print_lock_info(m) __bm_print_lock_info(m, __func__)
 static void __bm_print_lock_info(struct drbd_conf *mdev, const char *func)
 {
@@ -994,6 +991,9 @@ static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must
 		bio_endio(bio, -EIO);
 	} else {
 		submit_bio(rw, bio);
+		/* this should not count as user activity and cause the
+		 * resync to throttle -- see drbd_rs_should_slow_down(). */
+		atomic_add(len >> 9, &mdev->rs_sect_ev);
 	}
 }
 
@@ -1256,7 +1256,7 @@ unsigned long _drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_f
  * expected to be called for only a few bits (e - s about BITS_PER_LONG).
  * Must hold bitmap lock already. */
 static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
-	unsigned long e, int val, const enum km_type km)
+	unsigned long e, int val)
 {
 	struct drbd_bitmap *b = mdev->bitmap;
 	unsigned long *p_addr = NULL;
@@ -1274,14 +1274,14 @@ static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
 		unsigned int page_nr = bm_bit_to_page_idx(b, bitnr);
 		if (page_nr != last_page_nr) {
 			if (p_addr)
-				__bm_unmap(p_addr, km);
+				__bm_unmap(p_addr, KM_IRQ1);
 			if (c < 0)
 				bm_set_page_lazy_writeout(b->bm_pages[last_page_nr]);
 			else if (c > 0)
 				bm_set_page_need_writeout(b->bm_pages[last_page_nr]);
 			changed_total += c;
 			c = 0;
-			p_addr = __bm_map_pidx(b, page_nr, km);
+			p_addr = __bm_map_pidx(b, page_nr, KM_IRQ1);
 			last_page_nr = page_nr;
 		}
 		if (val)
@@ -1290,7 +1290,7 @@ static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
 			c -= (0 != __test_and_clear_bit_le(bitnr & BITS_PER_PAGE_MASK, p_addr));
 	}
 	if (p_addr)
-		__bm_unmap(p_addr, km);
+		__bm_unmap(p_addr, KM_IRQ1);
 	if (c < 0)
 		bm_set_page_lazy_writeout(b->bm_pages[last_page_nr]);
 	else if (c > 0)
@@ -1318,7 +1318,7 @@ static int bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
 	if ((val ? BM_DONT_SET : BM_DONT_CLEAR) & b->bm_flags)
 		bm_print_lock_info(mdev);
 
-	c = __bm_change_bits_to(mdev, s, e, val, KM_IRQ1);
+	c = __bm_change_bits_to(mdev, s, e, val);
 
 	spin_unlock_irqrestore(&b->bm_lock, flags);
 	return c;
@@ -1343,16 +1343,17 @@ static inline void bm_set_full_words_within_one_page(struct drbd_bitmap *b,
 {
 	int i;
 	int bits;
-	unsigned long *paddr = kmap_atomic(b->bm_pages[page_nr], KM_USER0);
+	unsigned long *paddr = kmap_atomic(b->bm_pages[page_nr], KM_IRQ1);
 	for (i = first_word; i < last_word; i++) {
 		bits = hweight_long(paddr[i]);
 		paddr[i] = ~0UL;
 		b->bm_set += BITS_PER_LONG - bits;
 	}
-	kunmap_atomic(paddr, KM_USER0);
+	kunmap_atomic(paddr, KM_IRQ1);
 }
 
-/* Same thing as drbd_bm_set_bits, but without taking the spin_lock_irqsave.
+/* Same thing as drbd_bm_set_bits,
+ * but more efficient for a large bit range.
  * You must first drbd_bm_lock().
  * Can be called to set the whole bitmap in one go.
  * Sets bits from s to e _inclusive_. */
@@ -1366,6 +1367,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
 	 * Do not use memset, because we must account for changes,
 	 * so we need to loop over the words with hweight() anyways.
 	 */
+	struct drbd_bitmap *b = mdev->bitmap;
 	unsigned long sl = ALIGN(s,BITS_PER_LONG);
 	unsigned long el = (e+1) & ~((unsigned long)BITS_PER_LONG-1);
 	int first_page;
@@ -1376,15 +1378,19 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
 
 	if (e - s <= 3*BITS_PER_LONG) {
 		/* don't bother; el and sl may even be wrong. */
-		__bm_change_bits_to(mdev, s, e, 1, KM_USER0);
+		spin_lock_irq(&b->bm_lock);
+		__bm_change_bits_to(mdev, s, e, 1);
+		spin_unlock_irq(&b->bm_lock);
 		return;
 	}
 
 	/* difference is large enough that we can trust sl and el */
 
+	spin_lock_irq(&b->bm_lock);
+
 	/* bits filling the current long */
 	if (sl)
-		__bm_change_bits_to(mdev, s, sl-1, 1, KM_USER0);
+		__bm_change_bits_to(mdev, s, sl-1, 1);
 
 	first_page = sl >> (3 + PAGE_SHIFT);
 	last_page = el >> (3 + PAGE_SHIFT);
@@ -1397,8 +1403,10 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
 	/* first and full pages, unless first page == last page */
 	for (page_nr = first_page; page_nr < last_page; page_nr++) {
 		bm_set_full_words_within_one_page(mdev->bitmap, page_nr, first_word, last_word);
+		spin_unlock_irq(&b->bm_lock);
 		cond_resched();
 		first_word = 0;
+		spin_lock_irq(&b->bm_lock);
 	}
 
 	/* last page (respectively only page, for first page == last page) */
@@ -1411,7 +1419,8 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
 	 * it would trigger an assert in __bm_change_bits_to()
 	 */
 	if (el <= e)
-		__bm_change_bits_to(mdev, el, e, 1, KM_USER0);
+		__bm_change_bits_to(mdev, el, e, 1);
+	spin_unlock_irq(&b->bm_lock);
 }
 
 /* returns bit state
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 25d32c5aa50a..43beaca53179 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -4602,6 +4602,11 @@ int drbd_asender(struct drbd_thread *thi)
 			dev_err(DEV, "meta connection shut down by peer.\n");
 			goto reconnect;
 		} else if (rv == -EAGAIN) {
+			/* If the data socket received something meanwhile,
+			 * that is good enough: peer is still alive. */
+			if (time_after(mdev->last_received,
+				jiffies - mdev->meta.socket->sk->sk_rcvtimeo))
+				continue;
 			if (ping_timeout_active) {
 				dev_err(DEV, "PingAck did not arrive in time.\n");
 				goto reconnect;
@@ -4637,6 +4642,7 @@ int drbd_asender(struct drbd_thread *thi)
 				goto reconnect;
 		}
 		if (received == expect) {
+			mdev->last_received = jiffies;
 			D_ASSERT(cmd != NULL);
 			if (!cmd->process(mdev, h))
 				goto reconnect;
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 4d76b06b6b20..4d3e6f6213ba 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -536,12 +536,7 @@ static int w_make_resync_request(struct drbd_conf *mdev,
 		return 1;
 	}
 
-	/* starting with drbd 8.3.8, we can handle multi-bio EEs,
-	 * if it should be necessary */
-	max_bio_size =
-		mdev->agreed_pro_version < 94 ? queue_max_hw_sectors(mdev->rq_queue) << 9 :
-		mdev->agreed_pro_version < 95 ?	DRBD_MAX_SIZE_H80_PACKET : DRBD_MAX_BIO_SIZE;
-
+	max_bio_size = queue_max_hw_sectors(mdev->rq_queue) << 9;
 	number = drbd_rs_number_requests(mdev);
 	if (number == 0)
 		goto requeue;
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index e6fc716aca45..f533f3375e24 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -192,7 +192,8 @@ static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
 			if (lo->xmit_timeout)
 				del_timer_sync(&ti);
 		} else
-			result = kernel_recvmsg(sock, &msg, &iov, 1, size, 0);
+			result = kernel_recvmsg(sock, &msg, &iov, 1, size,
+						msg.msg_flags);
 
 		if (signal_pending(current)) {
 			siginfo_t info;
@@ -753,9 +754,26 @@ static int __init nbd_init(void)
 		return -ENOMEM;
 
 	part_shift = 0;
-	if (max_part > 0)
+	if (max_part > 0) {
 		part_shift = fls(max_part);
 
+		/*
+		 * Adjust max_part according to part_shift as it is exported
+		 * to user space so that user can know the max number of
+		 * partition kernel should be able to manage.
+		 *
+		 * Note that -1 is required because partition 0 is reserved
+		 * for the whole disk.
+		 */
+		max_part = (1UL << part_shift) - 1;
+	}
+
+	if ((1UL << part_shift) > DISK_MAX_PARTS)
+		return -EINVAL;
+
+	if (nbds_max > 1UL << (MINORBITS - part_shift))
+		return -EINVAL;
+
 	for (i = 0; i < nbds_max; i++) {
 		struct gendisk *disk = alloc_disk(1 << part_shift);
 		if (!disk)
diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index c73910cc28c9..5cf2993a8338 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -809,11 +809,13 @@ static int __init xen_blkif_init(void)
  failed_init:
 	kfree(blkbk->pending_reqs);
 	kfree(blkbk->pending_grant_handles);
-	for (i = 0; i < mmap_pages; i++) {
-		if (blkbk->pending_pages[i])
-			__free_page(blkbk->pending_pages[i]);
+	if (blkbk->pending_pages) {
+		for (i = 0; i < mmap_pages; i++) {
+			if (blkbk->pending_pages[i])
+				__free_page(blkbk->pending_pages[i]);
+		}
+		kfree(blkbk->pending_pages);
 	}
-	kfree(blkbk->pending_pages);
 	kfree(blkbk);
 	blkbk = NULL;
 	return rc;
diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c
index 34570823355b..6cc0db1bf522 100644
--- a/drivers/block/xen-blkback/xenbus.c
+++ b/drivers/block/xen-blkback/xenbus.c
@@ -357,14 +357,13 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 	}
 
 	vbd->bdev = bdev;
-	vbd->size = vbd_sz(vbd);
-
 	if (vbd->bdev->bd_disk == NULL) {
 		DPRINTK("xen_vbd_create: device %08x doesn't exist.\n",
 			vbd->pdevice);
 		xen_vbd_free(vbd);
 		return -ENOENT;
 	}
+	vbd->size = vbd_sz(vbd);
 
 	if (vbd->bdev->bd_disk->flags & GENHD_FL_CD || cdrom)
 		vbd->type |= VDISK_CDROM;
diff --git a/drivers/bluetooth/btmrvl_debugfs.c b/drivers/bluetooth/btmrvl_debugfs.c
index fd6305bf953e..8ecf4c6c2874 100644
--- a/drivers/bluetooth/btmrvl_debugfs.c
+++ b/drivers/bluetooth/btmrvl_debugfs.c
@@ -64,6 +64,8 @@ static ssize_t btmrvl_hscfgcmd_write(struct file *file,
 		return -EFAULT;
 
 	ret = strict_strtol(buf, 10, &result);
+	if (ret)
+		return ret;
 
 	priv->btmrvl_dev.hscfgcmd = result;
 
@@ -108,6 +110,8 @@ static ssize_t btmrvl_psmode_write(struct file *file, const char __user *ubuf,
 		return -EFAULT;
 
 	ret = strict_strtol(buf, 10, &result);
+	if (ret)
+		return ret;
 
 	priv->btmrvl_dev.psmode = result;
 
@@ -147,6 +151,8 @@ static ssize_t btmrvl_pscmd_write(struct file *file, const char __user *ubuf,
 		return -EFAULT;
 
 	ret = strict_strtol(buf, 10, &result);
+	if (ret)
+		return ret;
 
 	priv->btmrvl_dev.pscmd = result;
 
@@ -191,6 +197,8 @@ static ssize_t btmrvl_gpiogap_write(struct file *file, const char __user *ubuf,
 		return -EFAULT;
 
 	ret = strict_strtol(buf, 16, &result);
+	if (ret)
+		return ret;
 
 	priv->btmrvl_dev.gpio_gap = result;
 
@@ -230,6 +238,8 @@ static ssize_t btmrvl_hscmd_write(struct file *file, const char __user *ubuf,
 		return -EFAULT;
 
 	ret = strict_strtol(buf, 10, &result);
+	if (ret)
+		return ret;
 
 	priv->btmrvl_dev.hscmd = result;
 	if (priv->btmrvl_dev.hscmd) {
@@ -272,6 +282,8 @@ static ssize_t btmrvl_hsmode_write(struct file *file, const char __user *ubuf,
 		return -EFAULT;
 
 	ret = strict_strtol(buf, 10, &result);
+	if (ret)
+		return ret;
 
 	priv->btmrvl_dev.hsmode = result;
 
diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c
index b3f01996318f..48ad2a7ab080 100644
--- a/drivers/bluetooth/hci_ldisc.c
+++ b/drivers/bluetooth/hci_ldisc.c
@@ -355,29 +355,24 @@ static void hci_uart_tty_wakeup(struct tty_struct *tty)
  *             flags        pointer to flags for data
  *             count        count of received data in bytes
  *     
- * Return Value:    Number of bytes received
+ * Return Value:    None
  */
-static unsigned int hci_uart_tty_receive(struct tty_struct *tty,
-		const u8 *data, char *flags, int count)
+static void hci_uart_tty_receive(struct tty_struct *tty, const u8 *data, char *flags, int count)
 {
 	struct hci_uart *hu = (void *)tty->disc_data;
-	int received;
 
 	if (!hu || tty != hu->tty)
-		return -ENODEV;
+		return;
 
 	if (!test_bit(HCI_UART_PROTO_SET, &hu->flags))
-		return -EINVAL;
+		return;
 
 	spin_lock(&hu->rx_lock);
-	received = hu->proto->recv(hu, (void *) data, count);
-	if (received > 0)
-		hu->hdev->stat.byte_rx += received;
+	hu->proto->recv(hu, (void *) data, count);
+	hu->hdev->stat.byte_rx += count;
 	spin_unlock(&hu->rx_lock);
 
 	tty_unthrottle(tty);
-
-	return received;
 }
 
 static int hci_uart_register_dev(struct hci_uart *hu)
diff --git a/drivers/char/agp/intel-agp.h b/drivers/char/agp/intel-agp.h
index 999803ce10dc..5da67f165afa 100644
--- a/drivers/char/agp/intel-agp.h
+++ b/drivers/char/agp/intel-agp.h
@@ -90,9 +90,10 @@
 #define G4x_GMCH_SIZE_MASK	(0xf << 8)
 #define G4x_GMCH_SIZE_1M	(0x1 << 8)
 #define G4x_GMCH_SIZE_2M	(0x3 << 8)
-#define G4x_GMCH_SIZE_VT_1M	(0x9 << 8)
-#define G4x_GMCH_SIZE_VT_1_5M	(0xa << 8)
-#define G4x_GMCH_SIZE_VT_2M	(0xc << 8)
+#define G4x_GMCH_SIZE_VT_EN	(0x8 << 8)
+#define G4x_GMCH_SIZE_VT_1M	(G4x_GMCH_SIZE_1M | G4x_GMCH_SIZE_VT_EN)
+#define G4x_GMCH_SIZE_VT_1_5M	((0x2 << 8) | G4x_GMCH_SIZE_VT_EN)
+#define G4x_GMCH_SIZE_VT_2M	(G4x_GMCH_SIZE_2M | G4x_GMCH_SIZE_VT_EN)
 
 #define GFX_FLSH_CNTL		0x2170 /* 915+ */
 
diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 051474c65b78..34d6a1cab8de 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -163,11 +163,32 @@ static irqreturn_t hpet_interrupt(int irq, void *data)
 	 * This has the effect of treating non-periodic like periodic.
 	 */
 	if ((devp->hd_flags & (HPET_IE | HPET_PERIODIC)) == HPET_IE) {
-		unsigned long m, t;
+		unsigned long m, t, mc, base, k;
+		struct hpet __iomem *hpet = devp->hd_hpet;
+		struct hpets *hpetp = devp->hd_hpets;
 
 		t = devp->hd_ireqfreq;
 		m = read_counter(&devp->hd_timer->hpet_compare);
-		write_counter(t + m, &devp->hd_timer->hpet_compare);
+		mc = read_counter(&hpet->hpet_mc);
+		/* The time for the next interrupt would logically be t + m,
+		 * however, if we are very unlucky and the interrupt is delayed
+		 * for longer than t then we will completely miss the next
+		 * interrupt if we set t + m and an application will hang.
+		 * Therefore we need to make a more complex computation assuming
+		 * that there exists a k for which the following is true:
+		 * k * t + base < mc + delta
+		 * (k + 1) * t + base > mc + delta
+		 * where t is the interval in hpet ticks for the given freq,
+		 * base is the theoretical start value 0 < base < t,
+		 * mc is the main counter value at the time of the interrupt,
+		 * delta is the time it takes to write the a value to the
+		 * comparator.
+		 * k may then be computed as (mc - base + delta) / t .
+		 */
+		base = mc % t;
+		k = (mc - base + hpetp->hp_delta) / t;
+		write_counter(t * (k + 1) + base,
+			      &devp->hd_timer->hpet_compare);
 	}
 
 	if (devp->hd_flags & HPET_SHARED_IRQ)
diff --git a/drivers/clocksource/sh_cmt.c b/drivers/clocksource/sh_cmt.c
index 036e5865eb40..dc7c033ef587 100644
--- a/drivers/clocksource/sh_cmt.c
+++ b/drivers/clocksource/sh_cmt.c
@@ -24,7 +24,6 @@
 #include <linux/ioport.h>
 #include <linux/io.h>
 #include <linux/clk.h>
-#include <linux/pm_runtime.h>
 #include <linux/irq.h>
 #include <linux/err.h>
 #include <linux/clocksource.h>
@@ -153,12 +152,10 @@ static int sh_cmt_enable(struct sh_cmt_priv *p, unsigned long *rate)
 {
 	int ret;
 
-	/* wake up device and enable clock */
-	pm_runtime_get_sync(&p->pdev->dev);
+	/* enable clock */
 	ret = clk_enable(p->clk);
 	if (ret) {
 		dev_err(&p->pdev->dev, "cannot enable clock\n");
-		pm_runtime_put_sync(&p->pdev->dev);
 		return ret;
 	}
 
@@ -190,9 +187,8 @@ static void sh_cmt_disable(struct sh_cmt_priv *p)
 	/* disable interrupts in CMT block */
 	sh_cmt_write(p, CMCSR, 0);
 
-	/* stop clock and mark device as idle */
+	/* stop clock */
 	clk_disable(p->clk);
-	pm_runtime_put_sync(&p->pdev->dev);
 }
 
 /* private flags */
@@ -664,7 +660,6 @@ static int __devinit sh_cmt_probe(struct platform_device *pdev)
 
 	if (p) {
 		dev_info(&pdev->dev, "kept as earlytimer\n");
-		pm_runtime_enable(&pdev->dev);
 		return 0;
 	}
 
@@ -679,9 +674,6 @@ static int __devinit sh_cmt_probe(struct platform_device *pdev)
 		kfree(p);
 		platform_set_drvdata(pdev, NULL);
 	}
-
-	if (!is_early_platform_device(pdev))
-		pm_runtime_enable(&pdev->dev);
 	return ret;
 }
 
diff --git a/drivers/clocksource/sh_tmu.c b/drivers/clocksource/sh_tmu.c
index 17296288a205..808135768617 100644
--- a/drivers/clocksource/sh_tmu.c
+++ b/drivers/clocksource/sh_tmu.c
@@ -25,7 +25,6 @@
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/clk.h>
-#include <linux/pm_runtime.h>
 #include <linux/irq.h>
 #include <linux/err.h>
 #include <linux/clocksource.h>
@@ -110,12 +109,10 @@ static int sh_tmu_enable(struct sh_tmu_priv *p)
 {
 	int ret;
 
-	/* wake up device and enable clock */
-	pm_runtime_get_sync(&p->pdev->dev);
+	/* enable clock */
 	ret = clk_enable(p->clk);
 	if (ret) {
 		dev_err(&p->pdev->dev, "cannot enable clock\n");
-		pm_runtime_put_sync(&p->pdev->dev);
 		return ret;
 	}
 
@@ -144,9 +141,8 @@ static void sh_tmu_disable(struct sh_tmu_priv *p)
 	/* disable interrupts in TMU block */
 	sh_tmu_write(p, TCR, 0x0000);
 
-	/* stop clock and mark device as idle */
+	/* stop clock */
 	clk_disable(p->clk);
-	pm_runtime_put_sync(&p->pdev->dev);
 }
 
 static void sh_tmu_set_next(struct sh_tmu_priv *p, unsigned long delta,
@@ -415,7 +411,6 @@ static int __devinit sh_tmu_probe(struct platform_device *pdev)
 
 	if (p) {
 		dev_info(&pdev->dev, "kept as earlytimer\n");
-		pm_runtime_enable(&pdev->dev);
 		return 0;
 	}
 
@@ -430,9 +425,6 @@ static int __devinit sh_tmu_probe(struct platform_device *pdev)
 		kfree(p);
 		platform_set_drvdata(pdev, NULL);
 	}
-
-	if (!is_early_platform_device(pdev))
-		pm_runtime_enable(&pdev->dev);
 	return ret;
 }
 
diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c
index 219d88a0eeae..dde6a0fad408 100644
--- a/drivers/connector/connector.c
+++ b/drivers/connector/connector.c
@@ -139,6 +139,7 @@ static int cn_call_callback(struct sk_buff *skb)
 	spin_unlock_bh(&dev->cbdev->queue_lock);
 
 	if (cbq != NULL) {
+		err = 0;
 		cbq->callback(msg, nsp);
 		kfree_skb(skb);
 		cn_queue_release_callback(cbq);
diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
index 4e04e1274388..596d5dd32f41 100644
--- a/drivers/cpufreq/acpi-cpufreq.c
+++ b/drivers/cpufreq/acpi-cpufreq.c
@@ -759,7 +759,7 @@ static void __exit acpi_cpufreq_exit(void)
 
 	cpufreq_unregister_driver(&acpi_cpufreq_driver);
 
-	free_percpu(acpi_perf_data);
+	free_acpi_perf_data();
 }
 
 module_param(acpi_pstate_strict, uint, 0644);
diff --git a/drivers/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c
index b60a4c263686..faf7c5217848 100644
--- a/drivers/cpufreq/cpufreq_stats.c
+++ b/drivers/cpufreq/cpufreq_stats.c
@@ -298,11 +298,13 @@ static int cpufreq_stat_notifier_trans(struct notifier_block *nb,
 	old_index = stat->last_index;
 	new_index = freq_table_get_index(stat, freq->new);
 
-	cpufreq_stats_update(freq->cpu);
-	if (old_index == new_index)
+	/* We can't do stat->time_in_state[-1]= .. */
+	if (old_index == -1 || new_index == -1)
 		return 0;
 
-	if (old_index == -1 || new_index == -1)
+	cpufreq_stats_update(freq->cpu);
+
+	if (old_index == new_index)
 		return 0;
 
 	spin_lock(&cpufreq_stats_lock);
@@ -387,6 +389,7 @@ static void __exit cpufreq_stats_exit(void)
 	unregister_hotcpu_notifier(&cpufreq_stat_cpu_notifier);
 	for_each_online_cpu(cpu) {
 		cpufreq_stats_free_table(cpu);
+		cpufreq_stats_free_sysfs(cpu);
 	}
 }
 
diff --git a/drivers/cpufreq/powernow-k8.c b/drivers/cpufreq/powernow-k8.c
index 83479b6fb9a1..bce576d7478e 100644
--- a/drivers/cpufreq/powernow-k8.c
+++ b/drivers/cpufreq/powernow-k8.c
@@ -1079,6 +1079,9 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data,
 	}
 
 	res = transition_fid_vid(data, fid, vid);
+	if (res)
+		return res;
+
 	freqs.new = find_khz_freq_from_fid(data->currfid);
 
 	for_each_cpu(i, data->available_cores) {
@@ -1101,7 +1104,8 @@ static int transition_frequency_pstate(struct powernow_k8_data *data,
 	/* get MSR index for hardware pstate transition */
 	pstate = index & HW_PSTATE_MASK;
 	if (pstate > data->max_hw_pstate)
-		return 0;
+		return -EINVAL;
+
 	freqs.old = find_khz_freq_from_pstate(data->powernow_table,
 			data->currpstate);
 	freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate);
diff --git a/drivers/crypto/caam/caamalg.c b/drivers/crypto/caam/caamalg.c
index d0e65d6ddc77..676d957c22b0 100644
--- a/drivers/crypto/caam/caamalg.c
+++ b/drivers/crypto/caam/caamalg.c
@@ -238,9 +238,9 @@ static int build_sh_desc_ipsec(struct caam_ctx *ctx)
 
 	/* build shared descriptor for this session */
 	sh_desc = kmalloc(CAAM_CMD_SZ * DESC_AEAD_SHARED_TEXT_LEN +
-			  keys_fit_inline ?
-			  ctx->split_key_pad_len + ctx->enckeylen :
-			  CAAM_PTR_SZ * 2, GFP_DMA | GFP_KERNEL);
+			  (keys_fit_inline ?
+			   ctx->split_key_pad_len + ctx->enckeylen :
+			   CAAM_PTR_SZ * 2), GFP_DMA | GFP_KERNEL);
 	if (!sh_desc) {
 		dev_err(jrdev, "could not allocate shared descriptor\n");
 		return -ENOMEM;
diff --git a/drivers/dma/shdma.c b/drivers/dma/shdma.c
index 636e40925b16..028330044201 100644
--- a/drivers/dma/shdma.c
+++ b/drivers/dma/shdma.c
@@ -343,7 +343,7 @@ static int sh_dmae_alloc_chan_resources(struct dma_chan *chan)
 
 		dmae_set_dmars(sh_chan, cfg->mid_rid);
 		dmae_set_chcr(sh_chan, cfg->chcr);
-	} else if ((sh_dmae_readl(sh_chan, CHCR) & 0xf00) != 0x400) {
+	} else {
 		dmae_init(sh_chan);
 	}
 
@@ -1144,6 +1144,8 @@ static int __init sh_dmae_probe(struct platform_device *pdev)
 	/* platform data */
 	shdev->pdata = pdata;
 
+	platform_set_drvdata(pdev, shdev);
+
 	pm_runtime_enable(&pdev->dev);
 	pm_runtime_get_sync(&pdev->dev);
 
@@ -1219,6 +1221,11 @@ static int __init sh_dmae_probe(struct platform_device *pdev)
 	} else {
 		do {
 			for (i = chanirq_res->start; i <= chanirq_res->end; i++) {
+				if (irq_cnt >= SH_DMAC_MAX_CHANNELS) {
+					irq_cap = 1;
+					break;
+				}
+
 				if ((errirq_res->flags & IORESOURCE_BITS) ==
 				    IORESOURCE_IRQ_SHAREABLE)
 					chan_flag[irq_cnt] = IRQF_SHARED;
@@ -1228,15 +1235,11 @@ static int __init sh_dmae_probe(struct platform_device *pdev)
 					"Found IRQ %d for channel %d\n",
 					i, irq_cnt);
 				chan_irq[irq_cnt++] = i;
-
-				if (irq_cnt >= SH_DMAC_MAX_CHANNELS)
-					break;
 			}
 
-			if (irq_cnt >= SH_DMAC_MAX_CHANNELS) {
-				irq_cap = 1;
+			if (irq_cnt >= SH_DMAC_MAX_CHANNELS)
 				break;
-			}
+
 			chanirq_res = platform_get_resource(pdev,
 						IORESOURCE_IRQ, ++irqres);
 		} while (irq_cnt < pdata->channel_num && chanirq_res);
@@ -1256,7 +1259,6 @@ static int __init sh_dmae_probe(struct platform_device *pdev)
 
 	pm_runtime_put(&pdev->dev);
 
-	platform_set_drvdata(pdev, shdev);
 	dma_async_device_register(&shdev->common);
 
 	return err;
@@ -1278,6 +1280,8 @@ rst_err:
 
 	if (dmars)
 		iounmap(shdev->dmars);
+
+	platform_set_drvdata(pdev, NULL);
 emapdmars:
 	iounmap(shdev->chan_reg);
 	synchronize_rcu();
@@ -1316,6 +1320,8 @@ static int __exit sh_dmae_remove(struct platform_device *pdev)
 		iounmap(shdev->dmars);
 	iounmap(shdev->chan_reg);
 
+	platform_set_drvdata(pdev, NULL);
+
 	synchronize_rcu();
 	kfree(shdev);
 
diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c
index 438e6c831170..ebb897329c1e 100644
--- a/drivers/firewire/ohci.c
+++ b/drivers/firewire/ohci.c
@@ -264,6 +264,7 @@ static char ohci_driver_name[] = KBUILD_MODNAME;
 #define PCI_DEVICE_ID_AGERE_FW643	0x5901
 #define PCI_DEVICE_ID_JMICRON_JMB38X_FW	0x2380
 #define PCI_DEVICE_ID_TI_TSB12LV22	0x8009
+#define PCI_VENDOR_ID_PINNACLE_SYSTEMS	0x11bd
 
 #define QUIRK_CYCLE_TIMER		1
 #define QUIRK_RESET_PACKET		2
@@ -3190,6 +3191,11 @@ static int __devinit pci_probe(struct pci_dev *dev,
 	int i, err;
 	size_t size;
 
+	if (dev->vendor == PCI_VENDOR_ID_PINNACLE_SYSTEMS) {
+		dev_err(&dev->dev, "Pinnacle MovieBoard is not yet supported\n");
+		return -ENOSYS;
+	}
+
 	ohci = kzalloc(sizeof(*ohci), GFP_KERNEL);
 	if (ohci == NULL) {
 		err = -ENOMEM;
diff --git a/drivers/firmware/google/Kconfig b/drivers/firmware/google/Kconfig
index 87096b6ca5c9..2f21b0bfe653 100644
--- a/drivers/firmware/google/Kconfig
+++ b/drivers/firmware/google/Kconfig
@@ -13,6 +13,7 @@ menu "Google Firmware Drivers"
 config GOOGLE_SMI
 	tristate "SMI interface for Google platforms"
 	depends on ACPI && DMI
+	select EFI
 	select EFI_VARS
 	help
 	  Say Y here if you want to enable SMI callbacks for Google
diff --git a/drivers/firmware/iscsi_ibft_find.c b/drivers/firmware/iscsi_ibft_find.c
index f032e446fc11..bfe723266fd8 100644
--- a/drivers/firmware/iscsi_ibft_find.c
+++ b/drivers/firmware/iscsi_ibft_find.c
@@ -108,7 +108,9 @@ done:
  */
 unsigned long __init find_ibft_region(unsigned long *sizep)
 {
+#ifdef CONFIG_ACPI
 	int i;
+#endif
 	ibft_addr = NULL;
 
 #ifdef CONFIG_ACPI
diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 4a7f63143455..2967002a9f82 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -87,32 +87,20 @@ config GPIO_IT8761E
 	  Say yes here to support GPIO functionality of IT8761E super I/O chip.
 
 config GPIO_EXYNOS4
-	bool "Samsung Exynos4 GPIO library support"
-	default y if CPU_EXYNOS4210
-	depends on ARM
-	help
-	  Say yes here to support Samsung Exynos4 series SoCs GPIO library
+	def_bool y
+	depends on CPU_EXYNOS4210
 
 config GPIO_PLAT_SAMSUNG
-	bool "Samsung SoCs GPIO library support"
-	default y if SAMSUNG_GPIOLIB_4BIT
-	depends on ARM
-	help
-	  Say yes here to support Samsung SoCs GPIO library
+	def_bool y
+	depends on SAMSUNG_GPIOLIB_4BIT
 
 config GPIO_S5PC100
-	bool "Samsung S5PC100 GPIO library support"
-	default y if CPU_S5PC100
-	depends on ARM
-	help
-	  Say yes here to support Samsung S5PC100 SoCs GPIO library
+	def_bool y
+	depends on CPU_S5PC100
 
 config GPIO_S5PV210
-	bool "Samsung S5PV210/S5PC110 GPIO library support"
-	default y if CPU_S5PV210
-	depends on ARM
-	help
-	  Say yes here to support Samsung S5PV210/S5PC110 SoCs GPIO library
+	def_bool y
+	depends on CPU_S5PV210
 
 config GPIO_PL061
 	bool "PrimeCell PL061 GPIO support"
diff --git a/drivers/gpio/gpio-exynos4.c b/drivers/gpio/gpio-exynos4.c
index d54ca6adb660..9029835112e7 100644
--- a/drivers/gpio/gpio-exynos4.c
+++ b/drivers/gpio/gpio-exynos4.c
@@ -21,16 +21,37 @@
 #include <plat/gpio-cfg.h>
 #include <plat/gpio-cfg-helpers.h>
 
+int s3c_gpio_setpull_exynos4(struct s3c_gpio_chip *chip,
+				unsigned int off, s3c_gpio_pull_t pull)
+{
+	if (pull == S3C_GPIO_PULL_UP)
+		pull = 3;
+
+	return s3c_gpio_setpull_updown(chip, off, pull);
+}
+
+s3c_gpio_pull_t s3c_gpio_getpull_exynos4(struct s3c_gpio_chip *chip,
+						unsigned int off)
+{
+	s3c_gpio_pull_t pull;
+
+	pull = s3c_gpio_getpull_updown(chip, off);
+	if (pull == 3)
+		pull = S3C_GPIO_PULL_UP;
+
+	return pull;
+}
+
 static struct s3c_gpio_cfg gpio_cfg = {
 	.set_config	= s3c_gpio_setcfg_s3c64xx_4bit,
-	.set_pull	= s3c_gpio_setpull_updown,
-	.get_pull	= s3c_gpio_getpull_updown,
+	.set_pull	= s3c_gpio_setpull_exynos4,
+	.get_pull	= s3c_gpio_getpull_exynos4,
 };
 
 static struct s3c_gpio_cfg gpio_cfg_noint = {
 	.set_config	= s3c_gpio_setcfg_s3c64xx_4bit,
-	.set_pull	= s3c_gpio_setpull_updown,
-	.get_pull	= s3c_gpio_getpull_updown,
+	.set_pull	= s3c_gpio_setpull_exynos4,
+	.get_pull	= s3c_gpio_getpull_exynos4,
 };
 
 /*
diff --git a/drivers/gpio/gpio-nomadik.c b/drivers/gpio/gpio-nomadik.c
index 4961ef9bc153..2c212c732d76 100644
--- a/drivers/gpio/gpio-nomadik.c
+++ b/drivers/gpio/gpio-nomadik.c
@@ -4,6 +4,7 @@
  * Copyright (C) 2008,2009 STMicroelectronics
  * Copyright (C) 2009 Alessandro Rubini <rubini@unipv.it>
  *   Rewritten based on work by Prafulla WADASKAR <prafulla.wadaskar@st.com>
+ * Copyright (C) 2011 Linus Walleij <linus.walleij@linaro.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -49,6 +50,7 @@ struct nmk_gpio_chip {
 	u32 (*get_secondary_status)(unsigned int bank);
 	void (*set_ioforce)(bool enable);
 	spinlock_t lock;
+	bool sleepmode;
 	/* Keep track of configured edges */
 	u32 edge_rising;
 	u32 edge_falling;
@@ -393,14 +395,25 @@ EXPORT_SYMBOL(nmk_config_pins_sleep);
  * @gpio: pin number
  * @mode: NMK_GPIO_SLPM_INPUT or NMK_GPIO_SLPM_NOCHANGE,
  *
- * Sets the sleep mode of a pin.  If @mode is NMK_GPIO_SLPM_INPUT, the pin is
- * changed to an input (with pullup/down enabled) in sleep and deep sleep.  If
- * @mode is NMK_GPIO_SLPM_NOCHANGE, the pin remains in the state it was
- * configured even when in sleep and deep sleep.
+ * This register is actually in the pinmux layer, not the GPIO block itself.
+ * The GPIO1B_SLPM register defines the GPIO mode when SLEEP/DEEP-SLEEP
+ * mode is entered (i.e. when signal IOFORCE is HIGH by the platform code).
+ * Each GPIO can be configured to be forced into GPIO mode when IOFORCE is
+ * HIGH, overriding the normal setting defined by GPIO_AFSELx registers.
+ * When IOFORCE returns LOW (by software, after SLEEP/DEEP-SLEEP exit),
+ * the GPIOs return to the normal setting defined by GPIO_AFSELx registers.
  *
- * On DB8500v2 onwards, this setting loses the previous meaning and instead
- * indicates if wakeup detection is enabled on the pin.  Note that
- * enable_irq_wake() will automatically enable wakeup detection.
+ * If @mode is NMK_GPIO_SLPM_INPUT, the corresponding GPIO is switched to GPIO
+ * mode when signal IOFORCE is HIGH (i.e. when SLEEP/DEEP-SLEEP mode is
+ * entered) regardless of the altfunction selected. Also wake-up detection is
+ * ENABLED.
+ *
+ * If @mode is NMK_GPIO_SLPM_NOCHANGE, the corresponding GPIO remains
+ * controlled by NMK_GPIO_DATC, NMK_GPIO_DATS, NMK_GPIO_DIR, NMK_GPIO_PDIS
+ * (for altfunction GPIO) or respective on-chip peripherals (for other
+ * altfuncs) when IOFORCE is HIGH. Also wake-up detection DISABLED.
+ *
+ * Note that enable_irq_wake() will automatically enable wakeup detection.
  */
 int nmk_gpio_set_slpm(int gpio, enum nmk_gpio_slpm mode)
 {
@@ -551,6 +564,12 @@ static void __nmk_gpio_irq_modify(struct nmk_gpio_chip *nmk_chip,
 static void __nmk_gpio_set_wake(struct nmk_gpio_chip *nmk_chip,
 				int gpio, bool on)
 {
+	if (nmk_chip->sleepmode) {
+		__nmk_gpio_set_slpm(nmk_chip, gpio - nmk_chip->chip.base,
+				    on ? NMK_GPIO_SLPM_WAKEUP_ENABLE
+				    : NMK_GPIO_SLPM_WAKEUP_DISABLE);
+	}
+
 	__nmk_gpio_irq_modify(nmk_chip, gpio, WAKE, on);
 }
 
@@ -901,7 +920,7 @@ void nmk_gpio_wakeups_suspend(void)
 		writel(chip->fwimsc & chip->real_wake,
 		       chip->addr + NMK_GPIO_FWIMSC);
 
-		if (cpu_is_u8500v2()) {
+		if (chip->sleepmode) {
 			chip->slpm = readl(chip->addr + NMK_GPIO_SLPC);
 
 			/* 0 -> wakeup enable */
@@ -923,7 +942,7 @@ void nmk_gpio_wakeups_resume(void)
 		writel(chip->rwimsc, chip->addr + NMK_GPIO_RWIMSC);
 		writel(chip->fwimsc, chip->addr + NMK_GPIO_FWIMSC);
 
-		if (cpu_is_u8500v2())
+		if (chip->sleepmode)
 			writel(chip->slpm, chip->addr + NMK_GPIO_SLPC);
 	}
 }
@@ -1010,6 +1029,7 @@ static int __devinit nmk_gpio_probe(struct platform_device *dev)
 	nmk_chip->secondary_parent_irq = secondary_irq;
 	nmk_chip->get_secondary_status = pdata->get_secondary_status;
 	nmk_chip->set_ioforce = pdata->set_ioforce;
+	nmk_chip->sleepmode = pdata->supports_sleepmode;
 	spin_lock_init(&nmk_chip->lock);
 
 	chip = &nmk_chip->chip;
@@ -1065,5 +1085,3 @@ core_initcall(nmk_gpio_init);
 MODULE_AUTHOR("Prafulla WADASKAR and Alessandro Rubini");
 MODULE_DESCRIPTION("Nomadik GPIO Driver");
 MODULE_LICENSE("GPL");
-
-
diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 6c51191da567..35bebde23e83 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -432,7 +432,6 @@ static inline void set_24xx_gpio_triggering(struct gpio_bank *bank, int gpio,
 {
 	void __iomem *base = bank->base;
 	u32 gpio_bit = 1 << gpio;
-	u32 val;
 
 	if (cpu_is_omap44xx()) {
 		MOD_REG_BIT(OMAP4_GPIO_LEVELDETECT0, gpio_bit,
@@ -455,15 +454,8 @@ static inline void set_24xx_gpio_triggering(struct gpio_bank *bank, int gpio,
 	}
 	if (likely(!(bank->non_wakeup_gpios & gpio_bit))) {
 		if (cpu_is_omap44xx()) {
-			if (trigger != 0)
-				__raw_writel(1 << gpio, bank->base+
-						OMAP4_GPIO_IRQWAKEN0);
-			else {
-				val = __raw_readl(bank->base +
-							OMAP4_GPIO_IRQWAKEN0);
-				__raw_writel(val & (~(1 << gpio)), bank->base +
-							 OMAP4_GPIO_IRQWAKEN0);
-			}
+			MOD_REG_BIT(OMAP4_GPIO_IRQWAKEN0, gpio_bit,
+				trigger != 0);
 		} else {
 			/*
 			 * GPIO wakeup request can only be generated on edge
@@ -477,8 +469,9 @@ static inline void set_24xx_gpio_triggering(struct gpio_bank *bank, int gpio,
 					+ OMAP24XX_GPIO_CLEARWKUENA);
 		}
 	}
-	/* This part needs to be executed always for OMAP34xx */
-	if (cpu_is_omap34xx() || (bank->non_wakeup_gpios & gpio_bit)) {
+	/* This part needs to be executed always for OMAP{34xx, 44xx} */
+	if (cpu_is_omap34xx() || cpu_is_omap44xx() ||
+			(bank->non_wakeup_gpios & gpio_bit)) {
 		/*
 		 * Log the edge gpio and manually trigger the IRQ
 		 * after resume if the input level changes
@@ -1134,8 +1127,11 @@ static void gpio_irq_shutdown(struct irq_data *d)
 {
 	unsigned int gpio = d->irq - IH_GPIO_BASE;
 	struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
+	unsigned long flags;
 
+	spin_lock_irqsave(&bank->lock, flags);
 	_reset_gpio(bank, gpio);
+	spin_unlock_irqrestore(&bank->lock, flags);
 }
 
 static void gpio_ack_irq(struct irq_data *d)
@@ -1150,9 +1146,12 @@ static void gpio_mask_irq(struct irq_data *d)
 {
 	unsigned int gpio = d->irq - IH_GPIO_BASE;
 	struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
+	unsigned long flags;
 
+	spin_lock_irqsave(&bank->lock, flags);
 	_set_gpio_irqenable(bank, gpio, 0);
 	_set_gpio_triggering(bank, get_gpio_index(gpio), IRQ_TYPE_NONE);
+	spin_unlock_irqrestore(&bank->lock, flags);
 }
 
 static void gpio_unmask_irq(struct irq_data *d)
@@ -1161,7 +1160,9 @@ static void gpio_unmask_irq(struct irq_data *d)
 	struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
 	unsigned int irq_mask = 1 << get_gpio_index(gpio);
 	u32 trigger = irqd_get_trigger_type(d);
+	unsigned long flags;
 
+	spin_lock_irqsave(&bank->lock, flags);
 	if (trigger)
 		_set_gpio_triggering(bank, get_gpio_index(gpio), trigger);
 
@@ -1173,6 +1174,7 @@ static void gpio_unmask_irq(struct irq_data *d)
 	}
 
 	_set_gpio_irqenable(bank, gpio, 1);
+	spin_unlock_irqrestore(&bank->lock, flags);
 }
 
 static struct irq_chip gpio_irq_chip = {
@@ -1524,7 +1526,7 @@ static void omap_gpio_mod_init(struct gpio_bank *bank, int id)
 	}
 }
 
-static void __init omap_gpio_chip_init(struct gpio_bank *bank)
+static void __devinit omap_gpio_chip_init(struct gpio_bank *bank)
 {
 	int j;
 	static int gpio;
diff --git a/drivers/gpio/langwell_gpio.c b/drivers/gpio/langwell_gpio.c
index bd6571e0097a..644ba1255d3c 100644
--- a/drivers/gpio/langwell_gpio.c
+++ b/drivers/gpio/langwell_gpio.c
@@ -223,7 +223,7 @@ static void lnw_irq_handler(unsigned irq, struct irq_desc *desc)
 		gedr = gpio_reg(&lnw->chip, base, GEDR);
 		pending = readl(gedr);
 		while (pending) {
-			gpio = __ffs(pending) - 1;
+			gpio = __ffs(pending);
 			mask = BIT(gpio);
 			pending &= ~mask;
 			/* Clear before handling so we can't lose an edge */
diff --git a/drivers/gpio/tps65910-gpio.c b/drivers/gpio/tps65910-gpio.c
index 8d1ddfdd63eb..15097ca616d6 100644
--- a/drivers/gpio/tps65910-gpio.c
+++ b/drivers/gpio/tps65910-gpio.c
@@ -81,8 +81,10 @@ void tps65910_gpio_init(struct tps65910 *tps65910, int gpio_base)
 	switch(tps65910_chip_id(tps65910)) {
 	case TPS65910:
 		tps65910->gpio.ngpio	= 6;
+		break;
 	case TPS65911:
 		tps65910->gpio.ngpio	= 9;
+		break;
 	default:
 		return;
 	}
diff --git a/drivers/gpio/wm831x-gpio.c b/drivers/gpio/wm831x-gpio.c
index 309644cf4d9b..2bcfb0be09ff 100644
--- a/drivers/gpio/wm831x-gpio.c
+++ b/drivers/gpio/wm831x-gpio.c
@@ -180,6 +180,7 @@ static void wm831x_gpio_dbg_show(struct seq_file *s, struct gpio_chip *chip)
 			break;
 		case WM831X_GPIO_PULL_UP:
 			pull = "pullup";
+			break;
 		default:
 			pull = "INVALID PULL";
 			break;
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index 3e257a50bf56..61e1ef90d4e5 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -46,10 +46,11 @@ static struct drm_map_list *drm_find_matching_map(struct drm_device *dev,
 	list_for_each_entry(entry, &dev->maplist, head) {
 		/*
 		 * Because the kernel-userspace ABI is fixed at a 32-bit offset
-		 * while PCI resources may live above that, we ignore the map
-		 * offset for maps of type _DRM_FRAMEBUFFER or _DRM_REGISTERS.
-		 * It is assumed that each driver will have only one resource of
-		 * each type.
+		 * while PCI resources may live above that, we only compare the
+		 * lower 32 bits of the map offset for maps of type
+		 * _DRM_FRAMEBUFFER or _DRM_REGISTERS.
+		 * It is assumed that if a driver have more than one resource
+		 * of each type, the lower 32 bits are different.
 		 */
 		if (!entry->map ||
 		    map->type != entry->map->type ||
@@ -59,9 +60,12 @@ static struct drm_map_list *drm_find_matching_map(struct drm_device *dev,
 		case _DRM_SHM:
 			if (map->flags != _DRM_CONTAINS_LOCK)
 				break;
+			return entry;
 		case _DRM_REGISTERS:
 		case _DRM_FRAME_BUFFER:
-			return entry;
+			if ((entry->map->offset & 0xffffffff) ==
+			    (map->offset & 0xffffffff))
+				return entry;
 		default: /* Make gcc happy */
 			;
 		}
@@ -183,9 +187,6 @@ static int drm_addmap_core(struct drm_device * dev, resource_size_t offset,
 			return -EINVAL;
 		}
 #endif
-#ifdef __alpha__
-		map->offset += dev->hose->mem_space->start;
-#endif
 		/* Some drivers preinitialize some maps, without the X Server
 		 * needing to be aware of it.  Therefore, we just return success
 		 * when the server tries to create a duplicate map.
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 872747c5a544..82db18506662 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -886,9 +886,6 @@ int drm_mode_group_init(struct drm_device *dev, struct drm_mode_group *group)
 	total_objects += dev->mode_config.num_connector;
 	total_objects += dev->mode_config.num_encoder;
 
-	if (total_objects == 0)
-		return -EINVAL;
-
 	group->id_list = kzalloc(total_objects * sizeof(uint32_t), GFP_KERNEL);
 	if (!group->id_list)
 		return -ENOMEM;
@@ -1113,7 +1110,7 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 	if (card_res->count_fbs >= fb_count) {
 		copied = 0;
 		fb_id = (uint32_t __user *)(unsigned long)card_res->fb_id_ptr;
-		list_for_each_entry(fb, &file_priv->fbs, head) {
+		list_for_each_entry(fb, &file_priv->fbs, filp_head) {
 			if (put_user(fb->base.id, fb_id + copied)) {
 				ret = -EFAULT;
 				goto out;
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 0a9357c66ff8..09292193dafe 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -184,9 +184,9 @@ drm_edid_block_valid(u8 *raw_edid)
 
 bad:
 	if (raw_edid) {
-		DRM_ERROR("Raw EDID:\n");
+		printk(KERN_ERR "Raw EDID:\n");
 		print_hex_dump_bytes(KERN_ERR, DUMP_PREFIX_NONE, raw_edid, EDID_LENGTH);
-		printk("\n");
+		printk(KERN_ERR "\n");
 	}
 	return 0;
 }
@@ -258,6 +258,17 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
 	return ret == 2 ? 0 : -1;
 }
 
+static bool drm_edid_is_zero(u8 *in_edid, int length)
+{
+	int i;
+	u32 *raw_edid = (u32 *)in_edid;
+
+	for (i = 0; i < length / 4; i++)
+		if (*(raw_edid + i) != 0)
+			return false;
+	return true;
+}
+
 static u8 *
 drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 {
@@ -273,6 +284,10 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 			goto out;
 		if (drm_edid_block_valid(block))
 			break;
+		if (i == 0 && drm_edid_is_zero(block, EDID_LENGTH)) {
+			connector->null_edid_counter++;
+			goto carp;
+		}
 	}
 	if (i == 4)
 		goto carp;
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 74e4ff578017..4012fe423460 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -34,6 +34,7 @@
 #include <linux/module.h>
 #include <linux/mman.h>
 #include <linux/pagemap.h>
+#include <linux/shmem_fs.h>
 #include "drmP.h"
 
 /** @file drm_gem.c
diff --git a/drivers/gpu/drm/drm_ioc32.c b/drivers/gpu/drm/drm_ioc32.c
index d61d185cf040..4a058c7af6c0 100644
--- a/drivers/gpu/drm/drm_ioc32.c
+++ b/drivers/gpu/drm/drm_ioc32.c
@@ -28,6 +28,7 @@
  * IN THE SOFTWARE.
  */
 #include <linux/compat.h>
+#include <linux/ratelimit.h>
 
 #include "drmP.h"
 #include "drm_core.h"
@@ -253,10 +254,10 @@ static int compat_drm_addmap(struct file *file, unsigned int cmd,
 		return -EFAULT;
 
 	m32.handle = (unsigned long)handle;
-	if (m32.handle != (unsigned long)handle && printk_ratelimit())
-		printk(KERN_ERR "compat_drm_addmap truncated handle"
-		       " %p for type %d offset %x\n",
-		       handle, m32.type, m32.offset);
+	if (m32.handle != (unsigned long)handle)
+		printk_ratelimited(KERN_ERR "compat_drm_addmap truncated handle"
+				   " %p for type %d offset %x\n",
+				   handle, m32.type, m32.offset);
 
 	if (copy_to_user(argp, &m32, sizeof(m32)))
 		return -EFAULT;
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index e1aee4f6a7c6..b6a19cb07caf 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -251,7 +251,7 @@ err:
 }
 
 
-int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
+static int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
 {
 	if ((p->busnum >> 8) != drm_get_pci_domain(dev) ||
 	    (p->busnum & 0xff) != dev->pdev->bus->number ||
@@ -292,6 +292,7 @@ static struct drm_bus drm_pci_bus = {
 	.get_name = drm_pci_get_name,
 	.set_busid = drm_pci_set_busid,
 	.set_unique = drm_pci_set_unique,
+	.irq_by_busid = drm_pci_irq_by_busid,
 	.agp_init = drm_pci_agp_init,
 };
 
diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c
index 2c3fcbdfd8ff..5db96d45fc71 100644
--- a/drivers/gpu/drm/drm_vm.c
+++ b/drivers/gpu/drm/drm_vm.c
@@ -526,7 +526,7 @@ static int drm_mmap_dma(struct file *filp, struct vm_area_struct *vma)
 static resource_size_t drm_core_get_reg_ofs(struct drm_device *dev)
 {
 #ifdef __alpha__
-	return dev->hose->dense_mem_base - dev->hose->mem_space->start;
+	return dev->hose->dense_mem_base;
 #else
 	return 0;
 #endif
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 51c2257b11e6..0a893f7400fa 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -776,7 +776,7 @@ static int i915_error_state(struct seq_file *m, void *unused)
 	seq_printf(m, "  INSTPM: 0x%08x\n", error->instpm);
 	seq_printf(m, "  seqno: 0x%08x\n", error->seqno);
 
-	for (i = 0; i < 16; i++)
+	for (i = 0; i < dev_priv->num_fence_regs; i++)
 		seq_printf(m, "  fence[%d] = %08llx\n", i, error->fence[i]);
 
 	if (error->active_bo)
@@ -1207,13 +1207,17 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	if (ret)
 		return ret;
 
-	seq_printf(m, "power context ");
-	describe_obj(m, dev_priv->pwrctx);
-	seq_printf(m, "\n");
+	if (dev_priv->pwrctx) {
+		seq_printf(m, "power context ");
+		describe_obj(m, dev_priv->pwrctx);
+		seq_printf(m, "\n");
+	}
 
-	seq_printf(m, "render context ");
-	describe_obj(m, dev_priv->renderctx);
-	seq_printf(m, "\n");
+	if (dev_priv->renderctx) {
+		seq_printf(m, "render context ");
+		describe_obj(m, dev_priv->renderctx);
+		seq_printf(m, "\n");
+	}
 
 	mutex_unlock(&dev->mode_config.mutex);
 
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 0239e9974bf2..296fbd66f0e1 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1266,30 +1266,6 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	intel_modeset_gem_init(dev);
 
-	if (IS_IVYBRIDGE(dev)) {
-		/* Share pre & uninstall handlers with ILK/SNB */
-		dev->driver->irq_handler = ivybridge_irq_handler;
-		dev->driver->irq_preinstall = ironlake_irq_preinstall;
-		dev->driver->irq_postinstall = ivybridge_irq_postinstall;
-		dev->driver->irq_uninstall = ironlake_irq_uninstall;
-		dev->driver->enable_vblank = ivybridge_enable_vblank;
-		dev->driver->disable_vblank = ivybridge_disable_vblank;
-	} else if (HAS_PCH_SPLIT(dev)) {
-		dev->driver->irq_handler = ironlake_irq_handler;
-		dev->driver->irq_preinstall = ironlake_irq_preinstall;
-		dev->driver->irq_postinstall = ironlake_irq_postinstall;
-		dev->driver->irq_uninstall = ironlake_irq_uninstall;
-		dev->driver->enable_vblank = ironlake_enable_vblank;
-		dev->driver->disable_vblank = ironlake_disable_vblank;
-	} else {
-		dev->driver->irq_preinstall = i915_driver_irq_preinstall;
-		dev->driver->irq_postinstall = i915_driver_irq_postinstall;
-		dev->driver->irq_uninstall = i915_driver_irq_uninstall;
-		dev->driver->irq_handler = i915_driver_irq_handler;
-		dev->driver->enable_vblank = i915_enable_vblank;
-		dev->driver->disable_vblank = i915_disable_vblank;
-	}
-
 	ret = drm_irq_install(dev);
 	if (ret)
 		goto cleanup_gem;
@@ -1967,7 +1943,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (!dev_priv->mm.gtt) {
 		DRM_ERROR("Failed to initialize GTT\n");
 		ret = -ENODEV;
-		goto out_iomapfree;
+		goto out_rmmap;
 	}
 
 	agp_size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
@@ -2011,18 +1987,13 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (dev_priv->wq == NULL) {
 		DRM_ERROR("Failed to create our workqueue.\n");
 		ret = -ENOMEM;
-		goto out_iomapfree;
+		goto out_mtrrfree;
 	}
 
 	/* enable GEM by default */
 	dev_priv->has_gem = 1;
 
-	dev->driver->get_vblank_counter = i915_get_vblank_counter;
-	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
-	if (IS_G4X(dev) || IS_GEN5(dev) || IS_GEN6(dev) || IS_IVYBRIDGE(dev)) {
-		dev->max_vblank_count = 0xffffffff; /* full 32 bit counter */
-		dev->driver->get_vblank_counter = gm45_get_vblank_counter;
-	}
+	intel_irq_init(dev);
 
 	/* Try to make sure MCHBAR is enabled before poking at it */
 	intel_setup_mchbar(dev);
@@ -2103,13 +2074,21 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	return 0;
 
 out_gem_unload:
+	if (dev_priv->mm.inactive_shrinker.shrink)
+		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
+
 	if (dev->pdev->msi_enabled)
 		pci_disable_msi(dev->pdev);
 
 	intel_teardown_gmbus(dev);
 	intel_teardown_mchbar(dev);
 	destroy_workqueue(dev_priv->wq);
-out_iomapfree:
+out_mtrrfree:
+	if (dev_priv->mm.gtt_mtrr >= 0) {
+		mtrr_del(dev_priv->mm.gtt_mtrr, dev->agp->base,
+			 dev->agp->agp_info.aper_size * 1024 * 1024);
+		dev_priv->mm.gtt_mtrr = -1;
+	}
 	io_mapping_free(dev_priv->mm.gtt_mapping);
 out_rmmap:
 	pci_iounmap(dev->pdev, dev_priv->regs);
@@ -2182,9 +2161,8 @@ int i915_driver_unload(struct drm_device *dev)
 		/* Flush any outstanding unpin_work. */
 		flush_workqueue(dev_priv->wq);
 
-		i915_gem_free_all_phys_object(dev);
-
 		mutex_lock(&dev->struct_mutex);
+		i915_gem_free_all_phys_object(dev);
 		i915_gem_cleanup_ringbuffer(dev);
 		mutex_unlock(&dev->struct_mutex);
 		if (I915_HAS_FBC(dev) && i915_powersave)
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 0defd4270594..eb91e2dd7914 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -52,7 +52,7 @@ module_param_named(powersave, i915_powersave, int, 0600);
 unsigned int i915_semaphores = 0;
 module_param_named(semaphores, i915_semaphores, int, 0600);
 
-unsigned int i915_enable_rc6 = 1;
+unsigned int i915_enable_rc6 = 0;
 module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600);
 
 unsigned int i915_enable_fbc = 0;
@@ -577,8 +577,12 @@ int i915_reset(struct drm_device *dev, u8 flags)
 	if (get_seconds() - dev_priv->last_gpu_reset < 5) {
 		DRM_ERROR("GPU hanging too fast, declaring wedged!\n");
 	} else switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		ret = gen6_do_reset(dev, flags);
+		/* If reset with a user forcewake, try to restore */
+		if (atomic_read(&dev_priv->forcewake_count))
+			__gen6_gt_force_wake_get(dev_priv);
 		break;
 	case 5:
 		ret = ironlake_do_reset(dev, flags);
@@ -762,14 +766,6 @@ static struct drm_driver driver = {
 	.resume = i915_resume,
 
 	.device_is_agp = i915_driver_device_is_agp,
-	.enable_vblank = i915_enable_vblank,
-	.disable_vblank = i915_disable_vblank,
-	.get_vblank_timestamp = i915_get_vblank_timestamp,
-	.get_scanout_position = i915_get_crtc_scanoutpos,
-	.irq_preinstall = i915_driver_irq_preinstall,
-	.irq_postinstall = i915_driver_irq_postinstall,
-	.irq_uninstall = i915_driver_irq_uninstall,
-	.irq_handler = i915_driver_irq_handler,
 	.reclaim_buffers = drm_core_reclaim_buffers,
 	.master_create = i915_master_create,
 	.master_destroy = i915_master_destroy,
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ee660355ae68..f245c588ae95 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -211,6 +211,9 @@ struct drm_i915_display_funcs {
 	void (*fdi_link_train)(struct drm_crtc *crtc);
 	void (*init_clock_gating)(struct drm_device *dev);
 	void (*init_pch_clock_gating)(struct drm_device *dev);
+	int (*queue_flip)(struct drm_device *dev, struct drm_crtc *crtc,
+			  struct drm_framebuffer *fb,
+			  struct drm_i915_gem_object *obj);
 	/* clock updates for mode set */
 	/* cursor updates */
 	/* render clock increase/decrease */
@@ -716,6 +719,7 @@ typedef struct drm_i915_private {
 	struct intel_fbdev *fbdev;
 
 	struct drm_property *broadcast_rgb_property;
+	struct drm_property *force_audio_property;
 
 	atomic_t forcewake_count;
 } drm_i915_private_t;
@@ -909,13 +913,6 @@ struct drm_i915_file_private {
 	} mm;
 };
 
-enum intel_chip_family {
-	CHIP_I8XX = 0x01,
-	CHIP_I9XX = 0x02,
-	CHIP_I915 = 0x04,
-	CHIP_I965 = 0x08,
-};
-
 #define INTEL_INFO(dev)	(((struct drm_i915_private *) (dev)->dev_private)->info)
 
 #define IS_I830(dev)		((dev)->pci_device == 0x3577)
@@ -1000,8 +997,6 @@ extern unsigned int i915_enable_fbc;
 
 extern int i915_suspend(struct drm_device *dev, pm_message_t state);
 extern int i915_resume(struct drm_device *dev);
-extern void i915_save_display(struct drm_device *dev);
-extern void i915_restore_display(struct drm_device *dev);
 extern int i915_master_create(struct drm_device *dev, struct drm_master *master);
 extern void i915_master_destroy(struct drm_device *dev, struct drm_master *master);
 
@@ -1036,33 +1031,12 @@ extern int i915_irq_emit(struct drm_device *dev, void *data,
 extern int i915_irq_wait(struct drm_device *dev, void *data,
 			 struct drm_file *file_priv);
 
-extern irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS);
-extern void i915_driver_irq_preinstall(struct drm_device * dev);
-extern int i915_driver_irq_postinstall(struct drm_device *dev);
-extern void i915_driver_irq_uninstall(struct drm_device * dev);
-
-extern irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS);
-extern void ironlake_irq_preinstall(struct drm_device *dev);
-extern int ironlake_irq_postinstall(struct drm_device *dev);
-extern void ironlake_irq_uninstall(struct drm_device *dev);
-
-extern irqreturn_t ivybridge_irq_handler(DRM_IRQ_ARGS);
-extern void ivybridge_irq_preinstall(struct drm_device *dev);
-extern int ivybridge_irq_postinstall(struct drm_device *dev);
-extern void ivybridge_irq_uninstall(struct drm_device *dev);
+extern void intel_irq_init(struct drm_device *dev);
 
 extern int i915_vblank_pipe_set(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 extern int i915_vblank_pipe_get(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
-extern int i915_enable_vblank(struct drm_device *dev, int crtc);
-extern void i915_disable_vblank(struct drm_device *dev, int crtc);
-extern int ironlake_enable_vblank(struct drm_device *dev, int crtc);
-extern void ironlake_disable_vblank(struct drm_device *dev, int crtc);
-extern int ivybridge_enable_vblank(struct drm_device *dev, int crtc);
-extern void ivybridge_disable_vblank(struct drm_device *dev, int crtc);
-extern u32 i915_get_vblank_counter(struct drm_device *dev, int crtc);
-extern u32 gm45_get_vblank_counter(struct drm_device *dev, int crtc);
 extern int i915_vblank_swap(struct drm_device *dev, void *data,
 			    struct drm_file *file_priv);
 
@@ -1073,13 +1047,6 @@ void
 i915_disable_pipestat(drm_i915_private_t *dev_priv, int pipe, u32 mask);
 
 void intel_enable_asle (struct drm_device *dev);
-int i915_get_vblank_timestamp(struct drm_device *dev, int crtc,
-			      int *max_error,
-			      struct timeval *vblank_time,
-			      unsigned flags);
-
-int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
-			     int *vpos, int *hpos);
 
 #ifdef CONFIG_DEBUG_FS
 extern void i915_destroy_error_state(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 0b2e167d2bce..5c0d1247f453 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -31,6 +31,7 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include <linux/shmem_fs.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/pci.h>
@@ -354,13 +355,12 @@ i915_gem_shmem_pread_fast(struct drm_device *dev,
 		 * page_offset = offset within page
 		 * page_length = bytes to copy for this page
 		 */
-		page_offset = offset & (PAGE_SIZE-1);
+		page_offset = offset_in_page(offset);
 		page_length = remain;
 		if ((page_offset + remain) > PAGE_SIZE)
 			page_length = PAGE_SIZE - page_offset;
 
-		page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT,
-					   GFP_HIGHUSER | __GFP_RECLAIMABLE);
+		page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT);
 		if (IS_ERR(page))
 			return PTR_ERR(page);
 
@@ -453,9 +453,9 @@ i915_gem_shmem_pread_slow(struct drm_device *dev,
 		 * data_page_offset = offset with data_page_index page.
 		 * page_length = bytes to copy for this page
 		 */
-		shmem_page_offset = offset & ~PAGE_MASK;
+		shmem_page_offset = offset_in_page(offset);
 		data_page_index = data_ptr / PAGE_SIZE - first_data_page;
-		data_page_offset = data_ptr & ~PAGE_MASK;
+		data_page_offset = offset_in_page(data_ptr);
 
 		page_length = remain;
 		if ((shmem_page_offset + page_length) > PAGE_SIZE)
@@ -463,10 +463,11 @@ i915_gem_shmem_pread_slow(struct drm_device *dev,
 		if ((data_page_offset + page_length) > PAGE_SIZE)
 			page_length = PAGE_SIZE - data_page_offset;
 
-		page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT,
-					   GFP_HIGHUSER | __GFP_RECLAIMABLE);
-		if (IS_ERR(page))
-			return PTR_ERR(page);
+		page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT);
+		if (IS_ERR(page)) {
+			ret = PTR_ERR(page);
+			goto out;
+		}
 
 		if (do_bit17_swizzling) {
 			slow_shmem_bit17_copy(page,
@@ -638,8 +639,8 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev,
 		 * page_offset = offset within page
 		 * page_length = bytes to copy for this page
 		 */
-		page_base = (offset & ~(PAGE_SIZE-1));
-		page_offset = offset & (PAGE_SIZE-1);
+		page_base = offset & PAGE_MASK;
+		page_offset = offset_in_page(offset);
 		page_length = remain;
 		if ((page_offset + remain) > PAGE_SIZE)
 			page_length = PAGE_SIZE - page_offset;
@@ -650,7 +651,6 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev,
 		 */
 		if (fast_user_write(dev_priv->mm.gtt_mapping, page_base,
 				    page_offset, user_data, page_length))
-
 			return -EFAULT;
 
 		remain -= page_length;
@@ -730,9 +730,9 @@ i915_gem_gtt_pwrite_slow(struct drm_device *dev,
 		 * page_length = bytes to copy for this page
 		 */
 		gtt_page_base = offset & PAGE_MASK;
-		gtt_page_offset = offset & ~PAGE_MASK;
+		gtt_page_offset = offset_in_page(offset);
 		data_page_index = data_ptr / PAGE_SIZE - first_data_page;
-		data_page_offset = data_ptr & ~PAGE_MASK;
+		data_page_offset = offset_in_page(data_ptr);
 
 		page_length = remain;
 		if ((gtt_page_offset + page_length) > PAGE_SIZE)
@@ -791,13 +791,12 @@ i915_gem_shmem_pwrite_fast(struct drm_device *dev,
 		 * page_offset = offset within page
 		 * page_length = bytes to copy for this page
 		 */
-		page_offset = offset & (PAGE_SIZE-1);
+		page_offset = offset_in_page(offset);
 		page_length = remain;
 		if ((page_offset + remain) > PAGE_SIZE)
 			page_length = PAGE_SIZE - page_offset;
 
-		page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT,
-					   GFP_HIGHUSER | __GFP_RECLAIMABLE);
+		page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT);
 		if (IS_ERR(page))
 			return PTR_ERR(page);
 
@@ -896,9 +895,9 @@ i915_gem_shmem_pwrite_slow(struct drm_device *dev,
 		 * data_page_offset = offset with data_page_index page.
 		 * page_length = bytes to copy for this page
 		 */
-		shmem_page_offset = offset & ~PAGE_MASK;
+		shmem_page_offset = offset_in_page(offset);
 		data_page_index = data_ptr / PAGE_SIZE - first_data_page;
-		data_page_offset = data_ptr & ~PAGE_MASK;
+		data_page_offset = offset_in_page(data_ptr);
 
 		page_length = remain;
 		if ((shmem_page_offset + page_length) > PAGE_SIZE)
@@ -906,8 +905,7 @@ i915_gem_shmem_pwrite_slow(struct drm_device *dev,
 		if ((data_page_offset + page_length) > PAGE_SIZE)
 			page_length = PAGE_SIZE - data_page_offset;
 
-		page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT,
-					   GFP_HIGHUSER | __GFP_RECLAIMABLE);
+		page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT);
 		if (IS_ERR(page)) {
 			ret = PTR_ERR(page);
 			goto out;
@@ -1218,11 +1216,11 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 		ret = i915_gem_object_bind_to_gtt(obj, 0, true);
 		if (ret)
 			goto unlock;
-	}
 
-	ret = i915_gem_object_set_to_gtt_domain(obj, write);
-	if (ret)
-		goto unlock;
+		ret = i915_gem_object_set_to_gtt_domain(obj, write);
+		if (ret)
+			goto unlock;
+	}
 
 	if (obj->tiling_mode == I915_TILING_NONE)
 		ret = i915_gem_object_put_fence(obj);
@@ -1450,8 +1448,9 @@ i915_gem_get_unfenced_gtt_alignment(struct drm_i915_gem_object *obj)
 	 * edge of an even tile row (where tile rows are counted as if the bo is
 	 * placed in a fenced gtt region).
 	 */
-	if (IS_GEN2(dev) ||
-	    (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)))
+	if (IS_GEN2(dev))
+		tile_height = 16;
+	else if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev))
 		tile_height = 32;
 	else
 		tile_height = 8;
@@ -1556,12 +1555,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj,
 
 	inode = obj->base.filp->f_path.dentry->d_inode;
 	mapping = inode->i_mapping;
+	gfpmask |= mapping_gfp_mask(mapping);
+
 	for (i = 0; i < page_count; i++) {
-		page = read_cache_page_gfp(mapping, i,
-					   GFP_HIGHUSER |
-					   __GFP_COLD |
-					   __GFP_RECLAIMABLE |
-					   gfpmask);
+		page = shmem_read_mapping_page_gfp(mapping, i, gfpmask);
 		if (IS_ERR(page))
 			goto err_pages;
 
@@ -1699,13 +1696,10 @@ i915_gem_object_truncate(struct drm_i915_gem_object *obj)
 	/* Our goal here is to return as much of the memory as
 	 * is possible back to the system as we are called from OOM.
 	 * To do this we must instruct the shmfs to drop all of its
-	 * backing pages, *now*. Here we mirror the actions taken
-	 * when by shmem_delete_inode() to release the backing store.
+	 * backing pages, *now*.
 	 */
 	inode = obj->base.filp->f_path.dentry->d_inode;
-	truncate_inode_pages(inode->i_mapping, 0);
-	if (inode->i_op->truncate_range)
-		inode->i_op->truncate_range(inode, 0, (loff_t)-1);
+	shmem_truncate_range(inode, 0, (loff_t)-1);
 
 	obj->madv = __I915_MADV_PURGED;
 }
@@ -2078,8 +2072,8 @@ i915_wait_request(struct intel_ring_buffer *ring,
 		if (!ier) {
 			DRM_ERROR("something (likely vbetool) disabled "
 				  "interrupts, re-enabling\n");
-			i915_driver_irq_preinstall(ring->dev);
-			i915_driver_irq_postinstall(ring->dev);
+			ring->dev->driver->irq_preinstall(ring->dev);
+			ring->dev->driver->irq_postinstall(ring->dev);
 		}
 
 		trace_i915_gem_request_wait_begin(ring, seqno);
@@ -2924,8 +2918,6 @@ i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj)
 	 */
 	wmb();
 
-	i915_gem_release_mmap(obj);
-
 	old_write_domain = obj->base.write_domain;
 	obj->base.write_domain = 0;
 
@@ -3565,6 +3557,7 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
+	struct address_space *mapping;
 
 	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
 	if (obj == NULL)
@@ -3575,6 +3568,9 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 		return NULL;
 	}
 
+	mapping = obj->base.filp->f_path.dentry->d_inode->i_mapping;
+	mapping_set_gfp_mask(mapping, GFP_HIGHUSER | __GFP_RECLAIMABLE);
+
 	i915_gem_info_add_obj(dev_priv, size);
 
 	obj->base.write_domain = I915_GEM_DOMAIN_CPU;
@@ -3950,8 +3946,7 @@ void i915_gem_detach_phys_object(struct drm_device *dev,
 
 	page_count = obj->base.size / PAGE_SIZE;
 	for (i = 0; i < page_count; i++) {
-		struct page *page = read_cache_page_gfp(mapping, i,
-							GFP_HIGHUSER | __GFP_RECLAIMABLE);
+		struct page *page = shmem_read_mapping_page(mapping, i);
 		if (!IS_ERR(page)) {
 			char *dst = kmap_atomic(page);
 			memcpy(dst, vaddr + i*PAGE_SIZE, PAGE_SIZE);
@@ -4012,8 +4007,7 @@ i915_gem_attach_phys_object(struct drm_device *dev,
 		struct page *page;
 		char *dst, *src;
 
-		page = read_cache_page_gfp(mapping, i,
-					   GFP_HIGHUSER | __GFP_RECLAIMABLE);
+		page = shmem_read_mapping_page(mapping, i);
 		if (IS_ERR(page))
 			return PTR_ERR(page);
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 20a4cc5b818f..4934cf84c320 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -187,10 +187,6 @@ i915_gem_object_set_to_gpu_domain(struct drm_i915_gem_object *obj,
 	if ((flush_domains | invalidate_domains) & I915_GEM_DOMAIN_CPU)
 		i915_gem_clflush_object(obj);
 
-	/* blow away mappings if mapped through GTT */
-	if ((flush_domains | invalidate_domains) & I915_GEM_DOMAIN_GTT)
-		i915_gem_release_mmap(obj);
-
 	if (obj->base.pending_write_domain)
 		cd->flips |= atomic_read(&obj->pending_flip);
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index b79619a7b788..3b03f85ea627 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -152,7 +152,7 @@ i915_pipe_enabled(struct drm_device *dev, int pipe)
 /* Called from drm generic code, passed a 'crtc', which
  * we use as a pipe index
  */
-u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
+static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long high_frame;
@@ -184,7 +184,7 @@ u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	return (high1 << 8) | low;
 }
 
-u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
+static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	int reg = PIPE_FRMCOUNT_GM45(pipe);
@@ -198,7 +198,7 @@ u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
 	return I915_READ(reg);
 }
 
-int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
+static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 			     int *vpos, int *hpos)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
@@ -264,7 +264,7 @@ int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	return ret;
 }
 
-int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
+static int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
 			      int *max_error,
 			      struct timeval *vblank_time,
 			      unsigned flags)
@@ -462,7 +462,7 @@ static void pch_irq_handler(struct drm_device *dev)
 		DRM_DEBUG_DRIVER("PCH transcoder A underrun interrupt\n");
 }
 
-irqreturn_t ivybridge_irq_handler(DRM_IRQ_ARGS)
+static irqreturn_t ivybridge_irq_handler(DRM_IRQ_ARGS)
 {
 	struct drm_device *dev = (struct drm_device *) arg;
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
@@ -517,7 +517,7 @@ irqreturn_t ivybridge_irq_handler(DRM_IRQ_ARGS)
 	if (de_iir & DE_PIPEA_VBLANK_IVB)
 		drm_handle_vblank(dev, 0);
 
-	if (de_iir & DE_PIPEB_VBLANK_IVB);
+	if (de_iir & DE_PIPEB_VBLANK_IVB)
 		drm_handle_vblank(dev, 1);
 
 	/* check event from PCH */
@@ -550,7 +550,7 @@ done:
 	return ret;
 }
 
-irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS)
+static irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS)
 {
 	struct drm_device *dev = (struct drm_device *) arg;
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
@@ -1209,7 +1209,7 @@ static void i915_pageflip_stall_check(struct drm_device *dev, int pipe)
 	}
 }
 
-irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS)
+static irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS)
 {
 	struct drm_device *dev = (struct drm_device *) arg;
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
@@ -1454,7 +1454,7 @@ int i915_irq_wait(struct drm_device *dev, void *data,
 /* Called from drm generic code, passed 'crtc' which
  * we use as a pipe index
  */
-int i915_enable_vblank(struct drm_device *dev, int pipe)
+static int i915_enable_vblank(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
@@ -1478,7 +1478,7 @@ int i915_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-int ironlake_enable_vblank(struct drm_device *dev, int pipe)
+static int ironlake_enable_vblank(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
@@ -1494,7 +1494,7 @@ int ironlake_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-int ivybridge_enable_vblank(struct drm_device *dev, int pipe)
+static int ivybridge_enable_vblank(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
@@ -1513,7 +1513,7 @@ int ivybridge_enable_vblank(struct drm_device *dev, int pipe)
 /* Called from drm generic code, passed 'crtc' which
  * we use as a pipe index
  */
-void i915_disable_vblank(struct drm_device *dev, int pipe)
+static void i915_disable_vblank(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
@@ -1529,7 +1529,7 @@ void i915_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-void ironlake_disable_vblank(struct drm_device *dev, int pipe)
+static void ironlake_disable_vblank(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
@@ -1540,7 +1540,7 @@ void ironlake_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-void ivybridge_disable_vblank(struct drm_device *dev, int pipe)
+static void ivybridge_disable_vblank(struct drm_device *dev, int pipe)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	unsigned long irqflags;
@@ -1728,7 +1728,7 @@ repeat:
 
 /* drm_dma.h hooks
 */
-void ironlake_irq_preinstall(struct drm_device *dev)
+static void ironlake_irq_preinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 
@@ -1740,6 +1740,17 @@ void ironlake_irq_preinstall(struct drm_device *dev)
 		INIT_WORK(&dev_priv->rps_work, gen6_pm_rps_work);
 
 	I915_WRITE(HWSTAM, 0xeffe);
+	if (IS_GEN6(dev) || IS_GEN7(dev)) {
+		/* Workaround stalls observed on Sandy Bridge GPUs by
+		 * making the blitter command streamer generate a
+		 * write to the Hardware Status Page for
+		 * MI_USER_INTERRUPT.  This appears to serialize the
+		 * previous seqno write out before the interrupt
+		 * happens.
+		 */
+		I915_WRITE(GEN6_BLITTER_HWSTAM, ~GEN6_BLITTER_USER_INTERRUPT);
+		I915_WRITE(GEN6_BSD_HWSTAM, ~GEN6_BSD_USER_INTERRUPT);
+	}
 
 	/* XXX hotplug from PCH */
 
@@ -1758,7 +1769,7 @@ void ironlake_irq_preinstall(struct drm_device *dev)
 	POSTING_READ(SDEIER);
 }
 
-int ironlake_irq_postinstall(struct drm_device *dev)
+static int ironlake_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	/* enable kind of interrupts always enabled */
@@ -1830,7 +1841,7 @@ int ironlake_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
-int ivybridge_irq_postinstall(struct drm_device *dev)
+static int ivybridge_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	/* enable kind of interrupts always enabled */
@@ -1880,7 +1891,7 @@ int ivybridge_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
-void i915_driver_irq_preinstall(struct drm_device * dev)
+static void i915_driver_irq_preinstall(struct drm_device * dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	int pipe;
@@ -1907,7 +1918,7 @@ void i915_driver_irq_preinstall(struct drm_device * dev)
  * Must be called after intel_modeset_init or hotplug interrupts won't be
  * enabled correctly.
  */
-int i915_driver_irq_postinstall(struct drm_device *dev)
+static int i915_driver_irq_postinstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	u32 enable_mask = I915_INTERRUPT_ENABLE_FIX | I915_INTERRUPT_ENABLE_VAR;
@@ -1983,7 +1994,7 @@ int i915_driver_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
-void ironlake_irq_uninstall(struct drm_device *dev)
+static void ironlake_irq_uninstall(struct drm_device *dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 
@@ -2003,7 +2014,7 @@ void ironlake_irq_uninstall(struct drm_device *dev)
 	I915_WRITE(GTIIR, I915_READ(GTIIR));
 }
 
-void i915_driver_irq_uninstall(struct drm_device * dev)
+static void i915_driver_irq_uninstall(struct drm_device * dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
 	int pipe;
@@ -2029,3 +2040,41 @@ void i915_driver_irq_uninstall(struct drm_device * dev)
 			   I915_READ(PIPESTAT(pipe)) & 0x8000ffff);
 	I915_WRITE(IIR, I915_READ(IIR));
 }
+
+void intel_irq_init(struct drm_device *dev)
+{
+	dev->driver->get_vblank_counter = i915_get_vblank_counter;
+	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
+	if (IS_G4X(dev) || IS_GEN5(dev) || IS_GEN6(dev) || IS_IVYBRIDGE(dev)) {
+		dev->max_vblank_count = 0xffffffff; /* full 32 bit counter */
+		dev->driver->get_vblank_counter = gm45_get_vblank_counter;
+	}
+
+
+	dev->driver->get_vblank_timestamp = i915_get_vblank_timestamp;
+	dev->driver->get_scanout_position = i915_get_crtc_scanoutpos;
+
+	if (IS_IVYBRIDGE(dev)) {
+		/* Share pre & uninstall handlers with ILK/SNB */
+		dev->driver->irq_handler = ivybridge_irq_handler;
+		dev->driver->irq_preinstall = ironlake_irq_preinstall;
+		dev->driver->irq_postinstall = ivybridge_irq_postinstall;
+		dev->driver->irq_uninstall = ironlake_irq_uninstall;
+		dev->driver->enable_vblank = ivybridge_enable_vblank;
+		dev->driver->disable_vblank = ivybridge_disable_vblank;
+	} else if (HAS_PCH_SPLIT(dev)) {
+		dev->driver->irq_handler = ironlake_irq_handler;
+		dev->driver->irq_preinstall = ironlake_irq_preinstall;
+		dev->driver->irq_postinstall = ironlake_irq_postinstall;
+		dev->driver->irq_uninstall = ironlake_irq_uninstall;
+		dev->driver->enable_vblank = ironlake_enable_vblank;
+		dev->driver->disable_vblank = ironlake_disable_vblank;
+	} else {
+		dev->driver->irq_preinstall = i915_driver_irq_preinstall;
+		dev->driver->irq_postinstall = i915_driver_irq_postinstall;
+		dev->driver->irq_uninstall = i915_driver_irq_uninstall;
+		dev->driver->irq_handler = i915_driver_irq_handler;
+		dev->driver->enable_vblank = i915_enable_vblank;
+		dev->driver->disable_vblank = i915_disable_vblank;
+	}
+}
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 2f967af8e62e..5d5def756c9e 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -531,6 +531,7 @@
 #define   GEN6_BSD_SLEEP_PSMI_CONTROL_RC_ILDL_MESSAGE_ENABLE		0
 #define   GEN6_BSD_SLEEP_PSMI_CONTROL_IDLE_INDICATOR			(1 << 3)
 
+#define GEN6_BSD_HWSTAM			0x12098
 #define GEN6_BSD_IMR			0x120a8
 #define   GEN6_BSD_USER_INTERRUPT	(1 << 12)
 
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 60a94d2b5264..5257cfc34c35 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -597,7 +597,7 @@ static void i915_restore_modeset_reg(struct drm_device *dev)
 	return;
 }
 
-void i915_save_display(struct drm_device *dev)
+static void i915_save_display(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -689,7 +689,7 @@ void i915_save_display(struct drm_device *dev)
 	i915_save_vga(dev);
 }
 
-void i915_restore_display(struct drm_device *dev)
+static void i915_restore_display(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -780,6 +780,7 @@ void i915_restore_display(struct drm_device *dev)
 		I915_WRITE(CPU_VGACNTRL, dev_priv->saveVGACNTRL);
 	else
 		I915_WRITE(VGACNTRL, dev_priv->saveVGACNTRL);
+
 	I915_WRITE(VGA0, dev_priv->saveVGA0);
 	I915_WRITE(VGA1, dev_priv->saveVGA1);
 	I915_WRITE(VGA_PD, dev_priv->saveVGA_PD);
@@ -796,6 +797,8 @@ int i915_save_state(struct drm_device *dev)
 
 	pci_read_config_byte(dev->pdev, LBB, &dev_priv->saveLBB);
 
+	mutex_lock(&dev->struct_mutex);
+
 	/* Hardware status page */
 	dev_priv->saveHWS = I915_READ(HWS_PGA);
 
@@ -835,6 +838,8 @@ int i915_save_state(struct drm_device *dev)
 	for (i = 0; i < 3; i++)
 		dev_priv->saveSWF2[i] = I915_READ(SWF30 + (i << 2));
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -845,6 +850,8 @@ int i915_restore_state(struct drm_device *dev)
 
 	pci_write_config_byte(dev->pdev, LBB, dev_priv->saveLBB);
 
+	mutex_lock(&dev->struct_mutex);
+
 	/* Hardware status page */
 	I915_WRITE(HWS_PGA, dev_priv->saveHWS);
 
@@ -862,6 +869,7 @@ int i915_restore_state(struct drm_device *dev)
 		I915_WRITE(IER, dev_priv->saveIER);
 		I915_WRITE(IMR, dev_priv->saveIMR);
 	}
+	mutex_unlock(&dev->struct_mutex);
 
 	intel_init_clock_gating(dev);
 
@@ -873,6 +881,8 @@ int i915_restore_state(struct drm_device *dev)
 	if (IS_GEN6(dev))
 		gen6_enable_rps(dev_priv);
 
+	mutex_lock(&dev->struct_mutex);
+
 	/* Cache mode state */
 	I915_WRITE (CACHE_MODE_0, dev_priv->saveCACHE_MODE_0 | 0xffff0000);
 
@@ -886,6 +896,8 @@ int i915_restore_state(struct drm_device *dev)
 	for (i = 0; i < 3; i++)
 		I915_WRITE(SWF30 + (i << 2), dev_priv->saveSWF2[i]);
 
+	mutex_unlock(&dev->struct_mutex);
+
 	intel_i2c_reset(dev);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index e93f93cc7e78..0979d8877880 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -288,6 +288,8 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector)
 		 * This may be a DVI-I connector with a shared DDC
 		 * link between analog and digital outputs, so we
 		 * have to check the EDID input spec of the attached device.
+		 *
+		 * On the other hand, what should we do if it is a broken EDID?
 		 */
 		if (edid != NULL) {
 			is_digital = edid->input & DRM_EDID_INPUT_DIGITAL;
@@ -298,6 +300,8 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector)
 		if (!is_digital) {
 			DRM_DEBUG_KMS("CRT detected via DDC:0x50 [EDID]\n");
 			return true;
+		} else {
+			DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n");
 		}
 	}
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index f553ddfdc168..21b6f93fe919 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3983,54 +3983,6 @@ static void i830_update_wm(struct drm_device *dev)
 #define ILK_LP0_PLANE_LATENCY		700
 #define ILK_LP0_CURSOR_LATENCY		1300
 
-static bool ironlake_compute_wm0(struct drm_device *dev,
-				 int pipe,
-				 const struct intel_watermark_params *display,
-				 int display_latency_ns,
-				 const struct intel_watermark_params *cursor,
-				 int cursor_latency_ns,
-				 int *plane_wm,
-				 int *cursor_wm)
-{
-	struct drm_crtc *crtc;
-	int htotal, hdisplay, clock, pixel_size;
-	int line_time_us, line_count;
-	int entries, tlb_miss;
-
-	crtc = intel_get_crtc_for_pipe(dev, pipe);
-	if (crtc->fb == NULL || !crtc->enabled)
-		return false;
-
-	htotal = crtc->mode.htotal;
-	hdisplay = crtc->mode.hdisplay;
-	clock = crtc->mode.clock;
-	pixel_size = crtc->fb->bits_per_pixel / 8;
-
-	/* Use the small buffer method to calculate plane watermark */
-	entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000;
-	tlb_miss = display->fifo_size*display->cacheline_size - hdisplay * 8;
-	if (tlb_miss > 0)
-		entries += tlb_miss;
-	entries = DIV_ROUND_UP(entries, display->cacheline_size);
-	*plane_wm = entries + display->guard_size;
-	if (*plane_wm > (int)display->max_wm)
-		*plane_wm = display->max_wm;
-
-	/* Use the large buffer method to calculate cursor watermark */
-	line_time_us = ((htotal * 1000) / clock);
-	line_count = (cursor_latency_ns / line_time_us + 1000) / 1000;
-	entries = line_count * 64 * pixel_size;
-	tlb_miss = cursor->fifo_size*cursor->cacheline_size - hdisplay * 8;
-	if (tlb_miss > 0)
-		entries += tlb_miss;
-	entries = DIV_ROUND_UP(entries, cursor->cacheline_size);
-	*cursor_wm = entries + cursor->guard_size;
-	if (*cursor_wm > (int)cursor->max_wm)
-		*cursor_wm = (int)cursor->max_wm;
-
-	return true;
-}
-
 /*
  * Check the wm result.
  *
@@ -4139,12 +4091,12 @@ static void ironlake_update_wm(struct drm_device *dev)
 	unsigned int enabled;
 
 	enabled = 0;
-	if (ironlake_compute_wm0(dev, 0,
-				 &ironlake_display_wm_info,
-				 ILK_LP0_PLANE_LATENCY,
-				 &ironlake_cursor_wm_info,
-				 ILK_LP0_CURSOR_LATENCY,
-				 &plane_wm, &cursor_wm)) {
+	if (g4x_compute_wm0(dev, 0,
+			    &ironlake_display_wm_info,
+			    ILK_LP0_PLANE_LATENCY,
+			    &ironlake_cursor_wm_info,
+			    ILK_LP0_CURSOR_LATENCY,
+			    &plane_wm, &cursor_wm)) {
 		I915_WRITE(WM0_PIPEA_ILK,
 			   (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm);
 		DRM_DEBUG_KMS("FIFO watermarks For pipe A -"
@@ -4153,12 +4105,12 @@ static void ironlake_update_wm(struct drm_device *dev)
 		enabled |= 1;
 	}
 
-	if (ironlake_compute_wm0(dev, 1,
-				 &ironlake_display_wm_info,
-				 ILK_LP0_PLANE_LATENCY,
-				 &ironlake_cursor_wm_info,
-				 ILK_LP0_CURSOR_LATENCY,
-				 &plane_wm, &cursor_wm)) {
+	if (g4x_compute_wm0(dev, 1,
+			    &ironlake_display_wm_info,
+			    ILK_LP0_PLANE_LATENCY,
+			    &ironlake_cursor_wm_info,
+			    ILK_LP0_CURSOR_LATENCY,
+			    &plane_wm, &cursor_wm)) {
 		I915_WRITE(WM0_PIPEB_ILK,
 			   (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm);
 		DRM_DEBUG_KMS("FIFO watermarks For pipe B -"
@@ -4223,10 +4175,10 @@ static void sandybridge_update_wm(struct drm_device *dev)
 	unsigned int enabled;
 
 	enabled = 0;
-	if (ironlake_compute_wm0(dev, 0,
-				 &sandybridge_display_wm_info, latency,
-				 &sandybridge_cursor_wm_info, latency,
-				 &plane_wm, &cursor_wm)) {
+	if (g4x_compute_wm0(dev, 0,
+			    &sandybridge_display_wm_info, latency,
+			    &sandybridge_cursor_wm_info, latency,
+			    &plane_wm, &cursor_wm)) {
 		I915_WRITE(WM0_PIPEA_ILK,
 			   (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm);
 		DRM_DEBUG_KMS("FIFO watermarks For pipe A -"
@@ -4235,10 +4187,10 @@ static void sandybridge_update_wm(struct drm_device *dev)
 		enabled |= 1;
 	}
 
-	if (ironlake_compute_wm0(dev, 1,
-				 &sandybridge_display_wm_info, latency,
-				 &sandybridge_cursor_wm_info, latency,
-				 &plane_wm, &cursor_wm)) {
+	if (g4x_compute_wm0(dev, 1,
+			    &sandybridge_display_wm_info, latency,
+			    &sandybridge_cursor_wm_info, latency,
+			    &plane_wm, &cursor_wm)) {
 		I915_WRITE(WM0_PIPEB_ILK,
 			   (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm);
 		DRM_DEBUG_KMS("FIFO watermarks For pipe B -"
@@ -4735,6 +4687,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 
 	I915_WRITE(DSPCNTR(plane), dspcntr);
 	POSTING_READ(DSPCNTR(plane));
+	intel_enable_plane(dev_priv, plane, pipe);
 
 	ret = intel_pipe_set_base(crtc, x, y, old_fb);
 
@@ -5265,8 +5218,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 
 	I915_WRITE(DSPCNTR(plane), dspcntr);
 	POSTING_READ(DSPCNTR(plane));
-	if (!HAS_PCH_SPLIT(dev))
-		intel_enable_plane(dev_priv, plane, pipe);
 
 	ret = intel_pipe_set_base(crtc, x, y, old_fb);
 
@@ -6310,6 +6261,197 @@ void intel_prepare_page_flip(struct drm_device *dev, int plane)
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
 
+static int intel_gen2_queue_flip(struct drm_device *dev,
+				 struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	unsigned long offset;
+	u32 flip_mask;
+	int ret;
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv));
+	if (ret)
+		goto out;
+
+	/* Offset into the new buffer for cases of shared fbs between CRTCs */
+	offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
+
+	ret = BEGIN_LP_RING(6);
+	if (ret)
+		goto out;
+
+	/* Can't queue multiple flips, so wait for the previous
+	 * one to finish before executing the next.
+	 */
+	if (intel_crtc->plane)
+		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
+	else
+		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
+	OUT_RING(MI_WAIT_FOR_EVENT | flip_mask);
+	OUT_RING(MI_NOOP);
+	OUT_RING(MI_DISPLAY_FLIP |
+		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+	OUT_RING(fb->pitch);
+	OUT_RING(obj->gtt_offset + offset);
+	OUT_RING(MI_NOOP);
+	ADVANCE_LP_RING();
+out:
+	return ret;
+}
+
+static int intel_gen3_queue_flip(struct drm_device *dev,
+				 struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	unsigned long offset;
+	u32 flip_mask;
+	int ret;
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv));
+	if (ret)
+		goto out;
+
+	/* Offset into the new buffer for cases of shared fbs between CRTCs */
+	offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
+
+	ret = BEGIN_LP_RING(6);
+	if (ret)
+		goto out;
+
+	if (intel_crtc->plane)
+		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
+	else
+		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
+	OUT_RING(MI_WAIT_FOR_EVENT | flip_mask);
+	OUT_RING(MI_NOOP);
+	OUT_RING(MI_DISPLAY_FLIP_I915 |
+		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+	OUT_RING(fb->pitch);
+	OUT_RING(obj->gtt_offset + offset);
+	OUT_RING(MI_NOOP);
+
+	ADVANCE_LP_RING();
+out:
+	return ret;
+}
+
+static int intel_gen4_queue_flip(struct drm_device *dev,
+				 struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	uint32_t pf, pipesrc;
+	int ret;
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv));
+	if (ret)
+		goto out;
+
+	ret = BEGIN_LP_RING(4);
+	if (ret)
+		goto out;
+
+	/* i965+ uses the linear or tiled offsets from the
+	 * Display Registers (which do not change across a page-flip)
+	 * so we need only reprogram the base address.
+	 */
+	OUT_RING(MI_DISPLAY_FLIP |
+		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+	OUT_RING(fb->pitch);
+	OUT_RING(obj->gtt_offset | obj->tiling_mode);
+
+	/* XXX Enabling the panel-fitter across page-flip is so far
+	 * untested on non-native modes, so ignore it for now.
+	 * pf = I915_READ(pipe == 0 ? PFA_CTL_1 : PFB_CTL_1) & PF_ENABLE;
+	 */
+	pf = 0;
+	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
+	OUT_RING(pf | pipesrc);
+	ADVANCE_LP_RING();
+out:
+	return ret;
+}
+
+static int intel_gen6_queue_flip(struct drm_device *dev,
+				 struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	uint32_t pf, pipesrc;
+	int ret;
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv));
+	if (ret)
+		goto out;
+
+	ret = BEGIN_LP_RING(4);
+	if (ret)
+		goto out;
+
+	OUT_RING(MI_DISPLAY_FLIP |
+		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+	OUT_RING(fb->pitch | obj->tiling_mode);
+	OUT_RING(obj->gtt_offset);
+
+	pf = I915_READ(PF_CTL(intel_crtc->pipe)) & PF_ENABLE;
+	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
+	OUT_RING(pf | pipesrc);
+	ADVANCE_LP_RING();
+out:
+	return ret;
+}
+
+/*
+ * On gen7 we currently use the blit ring because (in early silicon at least)
+ * the render ring doesn't give us interrpts for page flip completion, which
+ * means clients will hang after the first flip is queued.  Fortunately the
+ * blit ring generates interrupts properly, so use it instead.
+ */
+static int intel_gen7_queue_flip(struct drm_device *dev,
+				 struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_ring_buffer *ring = &dev_priv->ring[BCS];
+	int ret;
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
+	if (ret)
+		goto out;
+
+	ret = intel_ring_begin(ring, 4);
+	if (ret)
+		goto out;
+
+	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | (intel_crtc->plane << 19));
+	intel_ring_emit(ring, (fb->pitch | obj->tiling_mode));
+	intel_ring_emit(ring, (obj->gtt_offset));
+	intel_ring_emit(ring, (MI_NOOP));
+	intel_ring_advance(ring);
+out:
+	return ret;
+}
+
+static int intel_default_queue_flip(struct drm_device *dev,
+				    struct drm_crtc *crtc,
+				    struct drm_framebuffer *fb,
+				    struct drm_i915_gem_object *obj)
+{
+	return -ENODEV;
+}
+
 static int intel_crtc_page_flip(struct drm_crtc *crtc,
 				struct drm_framebuffer *fb,
 				struct drm_pending_vblank_event *event)
@@ -6320,9 +6462,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct drm_i915_gem_object *obj;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_unpin_work *work;
-	unsigned long flags, offset;
-	int pipe = intel_crtc->pipe;
-	u32 pf, pipesrc;
+	unsigned long flags;
 	int ret;
 
 	work = kzalloc(sizeof *work, GFP_KERNEL);
@@ -6351,9 +6491,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	obj = intel_fb->obj;
 
 	mutex_lock(&dev->struct_mutex);
-	ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv));
-	if (ret)
-		goto cleanup_work;
 
 	/* Reference the objects for the scheduled work. */
 	drm_gem_object_reference(&work->old_fb_obj->base);
@@ -6365,91 +6502,18 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (ret)
 		goto cleanup_objs;
 
-	if (IS_GEN3(dev) || IS_GEN2(dev)) {
-		u32 flip_mask;
-
-		/* Can't queue multiple flips, so wait for the previous
-		 * one to finish before executing the next.
-		 */
-		ret = BEGIN_LP_RING(2);
-		if (ret)
-			goto cleanup_objs;
-
-		if (intel_crtc->plane)
-			flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
-		else
-			flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-		OUT_RING(MI_WAIT_FOR_EVENT | flip_mask);
-		OUT_RING(MI_NOOP);
-		ADVANCE_LP_RING();
-	}
-
 	work->pending_flip_obj = obj;
 
 	work->enable_stall_check = true;
 
-	/* Offset into the new buffer for cases of shared fbs between CRTCs */
-	offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
-
-	ret = BEGIN_LP_RING(4);
-	if (ret)
-		goto cleanup_objs;
-
 	/* Block clients from rendering to the new back buffer until
 	 * the flip occurs and the object is no longer visible.
 	 */
 	atomic_add(1 << intel_crtc->plane, &work->old_fb_obj->pending_flip);
 
-	switch (INTEL_INFO(dev)->gen) {
-	case 2:
-		OUT_RING(MI_DISPLAY_FLIP |
-			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-		OUT_RING(fb->pitch);
-		OUT_RING(obj->gtt_offset + offset);
-		OUT_RING(MI_NOOP);
-		break;
-
-	case 3:
-		OUT_RING(MI_DISPLAY_FLIP_I915 |
-			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-		OUT_RING(fb->pitch);
-		OUT_RING(obj->gtt_offset + offset);
-		OUT_RING(MI_NOOP);
-		break;
-
-	case 4:
-	case 5:
-		/* i965+ uses the linear or tiled offsets from the
-		 * Display Registers (which do not change across a page-flip)
-		 * so we need only reprogram the base address.
-		 */
-		OUT_RING(MI_DISPLAY_FLIP |
-			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-		OUT_RING(fb->pitch);
-		OUT_RING(obj->gtt_offset | obj->tiling_mode);
-
-		/* XXX Enabling the panel-fitter across page-flip is so far
-		 * untested on non-native modes, so ignore it for now.
-		 * pf = I915_READ(pipe == 0 ? PFA_CTL_1 : PFB_CTL_1) & PF_ENABLE;
-		 */
-		pf = 0;
-		pipesrc = I915_READ(PIPESRC(pipe)) & 0x0fff0fff;
-		OUT_RING(pf | pipesrc);
-		break;
-
-	case 6:
-	case 7:
-		OUT_RING(MI_DISPLAY_FLIP |
-			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-		OUT_RING(fb->pitch | obj->tiling_mode);
-		OUT_RING(obj->gtt_offset);
-
-		pf = I915_READ(PF_CTL(pipe)) & PF_ENABLE;
-		pipesrc = I915_READ(PIPESRC(pipe)) & 0x0fff0fff;
-		OUT_RING(pf | pipesrc);
-		break;
-	}
-	ADVANCE_LP_RING();
+	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj);
+	if (ret)
+		goto cleanup_pending;
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -6457,10 +6521,11 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	return 0;
 
+cleanup_pending:
+	atomic_sub(1 << intel_crtc->plane, &work->old_fb_obj->pending_flip);
 cleanup_objs:
 	drm_gem_object_unreference(&work->old_fb_obj->base);
 	drm_gem_object_unreference(&obj->base);
-cleanup_work:
 	mutex_unlock(&dev->struct_mutex);
 
 	spin_lock_irqsave(&dev->event_lock, flags);
@@ -7675,6 +7740,7 @@ static void intel_init_display(struct drm_device *dev)
 			dev_priv->display.update_wm = NULL;
 		} else
 			dev_priv->display.update_wm = pineview_update_wm;
+		dev_priv->display.init_clock_gating = gen3_init_clock_gating;
 	} else if (IS_G4X(dev)) {
 		dev_priv->display.update_wm = g4x_update_wm;
 		dev_priv->display.init_clock_gating = g4x_init_clock_gating;
@@ -7704,6 +7770,31 @@ static void intel_init_display(struct drm_device *dev)
 		else
 			dev_priv->display.get_fifo_size = i830_get_fifo_size;
 	}
+
+	/* Default just returns -ENODEV to indicate unsupported */
+	dev_priv->display.queue_flip = intel_default_queue_flip;
+
+	switch (INTEL_INFO(dev)->gen) {
+	case 2:
+		dev_priv->display.queue_flip = intel_gen2_queue_flip;
+		break;
+
+	case 3:
+		dev_priv->display.queue_flip = intel_gen3_queue_flip;
+		break;
+
+	case 4:
+	case 5:
+		dev_priv->display.queue_flip = intel_gen4_queue_flip;
+		break;
+
+	case 6:
+		dev_priv->display.queue_flip = intel_gen6_queue_flip;
+		break;
+	case 7:
+		dev_priv->display.queue_flip = intel_gen7_queue_flip;
+		break;
+	}
 }
 
 /*
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index a4d80314e7f8..e2aced6eec4c 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -50,7 +50,6 @@ struct intel_dp {
 	bool has_audio;
 	int force_audio;
 	uint32_t color_range;
-	int dpms_mode;
 	uint8_t link_bw;
 	uint8_t lane_count;
 	uint8_t dpcd[4];
@@ -59,8 +58,6 @@ struct intel_dp {
 	bool is_pch_edp;
 	uint8_t	train_set[4];
 	uint8_t link_status[DP_LINK_STATUS_SIZE];
-
-	struct drm_property *force_audio_property;
 };
 
 /**
@@ -140,8 +137,8 @@ intel_dp_max_lane_count(struct intel_dp *intel_dp)
 {
 	int max_lane_count = 4;
 
-	if (intel_dp->dpcd[0] >= 0x11) {
-		max_lane_count = intel_dp->dpcd[2] & 0x1f;
+	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11) {
+		max_lane_count = intel_dp->dpcd[DP_MAX_LANE_COUNT] & 0x1f;
 		switch (max_lane_count) {
 		case 1: case 2: case 4:
 			break;
@@ -155,7 +152,7 @@ intel_dp_max_lane_count(struct intel_dp *intel_dp)
 static int
 intel_dp_max_link_bw(struct intel_dp *intel_dp)
 {
-	int max_link_bw = intel_dp->dpcd[1];
+	int max_link_bw = intel_dp->dpcd[DP_MAX_LINK_RATE];
 
 	switch (max_link_bw) {
 	case DP_LINK_BW_1_62:
@@ -776,7 +773,8 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 	/*
 	 * Check for DPCD version > 1.1 and enhanced framing support
 	 */
-	if (intel_dp->dpcd[0] >= 0x11 && (intel_dp->dpcd[2] & DP_ENHANCED_FRAME_CAP)) {
+	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
+	    (intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP)) {
 		intel_dp->link_configuration[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
 		intel_dp->DP |= DP_ENHANCED_FRAMING;
 	}
@@ -944,11 +942,44 @@ static void ironlake_edp_pll_off(struct drm_encoder *encoder)
 	udelay(200);
 }
 
+/* If the sink supports it, try to set the power state appropriately */
+static void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode)
+{
+	int ret, i;
+
+	/* Should have a valid DPCD by this point */
+	if (intel_dp->dpcd[DP_DPCD_REV] < 0x11)
+		return;
+
+	if (mode != DRM_MODE_DPMS_ON) {
+		ret = intel_dp_aux_native_write_1(intel_dp, DP_SET_POWER,
+						  DP_SET_POWER_D3);
+		if (ret != 1)
+			DRM_DEBUG_DRIVER("failed to write sink power state\n");
+	} else {
+		/*
+		 * When turning on, we need to retry for 1ms to give the sink
+		 * time to wake up.
+		 */
+		for (i = 0; i < 3; i++) {
+			ret = intel_dp_aux_native_write_1(intel_dp,
+							  DP_SET_POWER,
+							  DP_SET_POWER_D0);
+			if (ret == 1)
+				break;
+			msleep(1);
+		}
+	}
+}
+
 static void intel_dp_prepare(struct drm_encoder *encoder)
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct drm_device *dev = encoder->dev;
 
+	/* Wake up the sink first */
+	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+
 	if (is_edp(intel_dp)) {
 		ironlake_edp_backlight_off(dev);
 		ironlake_edp_panel_off(dev);
@@ -992,6 +1023,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 	if (mode != DRM_MODE_DPMS_ON) {
 		if (is_edp(intel_dp))
 			ironlake_edp_backlight_off(dev);
+		intel_dp_sink_dpms(intel_dp, mode);
 		intel_dp_link_down(intel_dp);
 		if (is_edp(intel_dp))
 			ironlake_edp_panel_off(dev);
@@ -1000,6 +1032,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 	} else {
 		if (is_edp(intel_dp))
 			ironlake_edp_panel_vdd_on(intel_dp);
+		intel_dp_sink_dpms(intel_dp, mode);
 		if (!(dp_reg & DP_PORT_EN)) {
 			intel_dp_start_link_train(intel_dp);
 			if (is_edp(intel_dp)) {
@@ -1011,7 +1044,31 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 		if (is_edp(intel_dp))
 			ironlake_edp_backlight_on(dev);
 	}
-	intel_dp->dpms_mode = mode;
+}
+
+/*
+ * Native read with retry for link status and receiver capability reads for
+ * cases where the sink may still be asleep.
+ */
+static bool
+intel_dp_aux_native_read_retry(struct intel_dp *intel_dp, uint16_t address,
+			       uint8_t *recv, int recv_bytes)
+{
+	int ret, i;
+
+	/*
+	 * Sinks are *supposed* to come up within 1ms from an off state,
+	 * but we're also supposed to retry 3 times per the spec.
+	 */
+	for (i = 0; i < 3; i++) {
+		ret = intel_dp_aux_native_read(intel_dp, address, recv,
+					       recv_bytes);
+		if (ret == recv_bytes)
+			return true;
+		msleep(1);
+	}
+
+	return false;
 }
 
 /*
@@ -1021,14 +1078,10 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 static bool
 intel_dp_get_link_status(struct intel_dp *intel_dp)
 {
-	int ret;
-
-	ret = intel_dp_aux_native_read(intel_dp,
-				       DP_LANE0_1_STATUS,
-				       intel_dp->link_status, DP_LINK_STATUS_SIZE);
-	if (ret != DP_LINK_STATUS_SIZE)
-		return false;
-	return true;
+	return intel_dp_aux_native_read_retry(intel_dp,
+					      DP_LANE0_1_STATUS,
+					      intel_dp->link_status,
+					      DP_LINK_STATUS_SIZE);
 }
 
 static uint8_t
@@ -1517,6 +1570,8 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 static void
 intel_dp_check_link_status(struct intel_dp *intel_dp)
 {
+	int ret;
+
 	if (!intel_dp->base.base.crtc)
 		return;
 
@@ -1525,6 +1580,15 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 		return;
 	}
 
+	/* Try to read receiver status if the link appears to be up */
+	ret = intel_dp_aux_native_read(intel_dp,
+				       0x000, intel_dp->dpcd,
+				       sizeof (intel_dp->dpcd));
+	if (ret != sizeof(intel_dp->dpcd)) {
+		intel_dp_link_down(intel_dp);
+		return;
+	}
+
 	if (!intel_channel_eq_ok(intel_dp)) {
 		intel_dp_start_link_train(intel_dp);
 		intel_dp_complete_link_train(intel_dp);
@@ -1535,6 +1599,7 @@ static enum drm_connector_status
 ironlake_dp_detect(struct intel_dp *intel_dp)
 {
 	enum drm_connector_status status;
+	bool ret;
 
 	/* Can't disconnect eDP, but you can close the lid... */
 	if (is_edp(intel_dp)) {
@@ -1545,13 +1610,11 @@ ironlake_dp_detect(struct intel_dp *intel_dp)
 	}
 
 	status = connector_status_disconnected;
-	if (intel_dp_aux_native_read(intel_dp,
-				     0x000, intel_dp->dpcd,
-				     sizeof (intel_dp->dpcd))
-	    == sizeof(intel_dp->dpcd)) {
-		if (intel_dp->dpcd[0] != 0)
-			status = connector_status_connected;
-	}
+	ret = intel_dp_aux_native_read_retry(intel_dp,
+					     0x000, intel_dp->dpcd,
+					     sizeof (intel_dp->dpcd));
+	if (ret && intel_dp->dpcd[DP_DPCD_REV] != 0)
+		status = connector_status_connected;
 	DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx\n", intel_dp->dpcd[0],
 		      intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3]);
 	return status;
@@ -1588,7 +1651,7 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 	if (intel_dp_aux_native_read(intel_dp, 0x000, intel_dp->dpcd,
 				     sizeof (intel_dp->dpcd)) == sizeof (intel_dp->dpcd))
 	{
-		if (intel_dp->dpcd[0] != 0)
+		if (intel_dp->dpcd[DP_DPCD_REV] != 0)
 			status = connector_status_connected;
 	}
 
@@ -1702,7 +1765,7 @@ intel_dp_set_property(struct drm_connector *connector,
 	if (ret)
 		return ret;
 
-	if (property == intel_dp->force_audio_property) {
+	if (property == dev_priv->force_audio_property) {
 		int i = val;
 		bool has_audio;
 
@@ -1792,8 +1855,7 @@ intel_dp_hot_plug(struct intel_encoder *intel_encoder)
 {
 	struct intel_dp *intel_dp = container_of(intel_encoder, struct intel_dp, base);
 
-	if (intel_dp->dpms_mode == DRM_MODE_DPMS_ON)
-		intel_dp_check_link_status(intel_dp);
+	intel_dp_check_link_status(intel_dp);
 }
 
 /* Return which DP Port should be selected for Transcoder DP control */
@@ -1841,16 +1903,7 @@ bool intel_dpd_is_edp(struct drm_device *dev)
 static void
 intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connector)
 {
-	struct drm_device *dev = connector->dev;
-
-	intel_dp->force_audio_property =
-		drm_property_create(dev, DRM_MODE_PROP_RANGE, "force_audio", 2);
-	if (intel_dp->force_audio_property) {
-		intel_dp->force_audio_property->values[0] = -1;
-		intel_dp->force_audio_property->values[1] = 1;
-		drm_connector_attach_property(connector, intel_dp->force_audio_property, 0);
-	}
-
+	intel_attach_force_audio_property(connector);
 	intel_attach_broadcast_rgb_property(connector);
 }
 
@@ -1870,7 +1923,6 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 		return;
 
 	intel_dp->output_reg = output_reg;
-	intel_dp->dpms_mode = -1;
 
 	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
@@ -1965,8 +2017,9 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 					       sizeof(intel_dp->dpcd));
 		ironlake_edp_panel_vdd_off(intel_dp);
 		if (ret == sizeof(intel_dp->dpcd)) {
-			if (intel_dp->dpcd[0] >= 0x11)
-				dev_priv->no_aux_handshake = intel_dp->dpcd[3] &
+			if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
+				dev_priv->no_aux_handshake =
+					intel_dp->dpcd[DP_MAX_DOWNSPREAD] &
 					DP_NO_AUX_HANDSHAKE_LINK_TRAINING;
 		} else {
 			/* if this fails, presume the device is a ghost */
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 831d7a4a0d18..9ffa61eb4d7e 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -236,6 +236,7 @@ struct intel_unpin_work {
 int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
 extern bool intel_ddc_probe(struct intel_encoder *intel_encoder, int ddc_bus);
 
+extern void intel_attach_force_audio_property(struct drm_connector *connector);
 extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
 
 extern void intel_crt_init(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index f289b8642976..aa0a8e83142e 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -45,7 +45,6 @@ struct intel_hdmi {
 	bool has_hdmi_sink;
 	bool has_audio;
 	int force_audio;
-	struct drm_property *force_audio_property;
 };
 
 static struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder)
@@ -194,7 +193,7 @@ static int intel_hdmi_mode_valid(struct drm_connector *connector,
 	if (mode->clock > 165000)
 		return MODE_CLOCK_HIGH;
 	if (mode->clock < 20000)
-		return MODE_CLOCK_HIGH;
+		return MODE_CLOCK_LOW;
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
@@ -287,7 +286,7 @@ intel_hdmi_set_property(struct drm_connector *connector,
 	if (ret)
 		return ret;
 
-	if (property == intel_hdmi->force_audio_property) {
+	if (property == dev_priv->force_audio_property) {
 		int i = val;
 		bool has_audio;
 
@@ -365,16 +364,7 @@ static const struct drm_encoder_funcs intel_hdmi_enc_funcs = {
 static void
 intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *connector)
 {
-	struct drm_device *dev = connector->dev;
-
-	intel_hdmi->force_audio_property =
-		drm_property_create(dev, DRM_MODE_PROP_RANGE, "force_audio", 2);
-	if (intel_hdmi->force_audio_property) {
-		intel_hdmi->force_audio_property->values[0] = -1;
-		intel_hdmi->force_audio_property->values[1] = 1;
-		drm_connector_attach_property(connector, intel_hdmi->force_audio_property, 0);
-	}
-
+	intel_attach_force_audio_property(connector);
 	intel_attach_broadcast_rgb_property(connector);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index d3b903bce7c5..d98cee60b602 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -401,8 +401,7 @@ int intel_setup_gmbus(struct drm_device *dev)
 		bus->reg0 = i | GMBUS_RATE_100KHZ;
 
 		/* XXX force bit banging until GMBUS is fully debugged */
-		if (IS_GEN2(dev))
-			bus->force_bit = intel_gpio_create(dev_priv, i);
+		bus->force_bit = intel_gpio_create(dev_priv, i);
 	}
 
 	intel_i2c_reset(dev_priv->dev);
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 67cb076d271b..b28f7bd9f88a 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -727,6 +727,14 @@ static const struct dmi_system_id intel_no_lvds[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "U800"),
 		},
 	},
+	{
+		.callback = intel_no_lvds_dmi_callback,
+		.ident = "Asus EeeBox PC EB1007",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "EB1007"),
+		},
+	},
 
 	{ }	/* terminating entry */
 };
diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c
index 9034dd8f33c7..3b26a3ba02dd 100644
--- a/drivers/gpu/drm/i915/intel_modes.c
+++ b/drivers/gpu/drm/i915/intel_modes.c
@@ -81,6 +81,36 @@ int intel_ddc_get_modes(struct drm_connector *connector,
 	return ret;
 }
 
+static const char *force_audio_names[] = {
+	"off",
+	"auto",
+	"on",
+};
+
+void
+intel_attach_force_audio_property(struct drm_connector *connector)
+{
+	struct drm_device *dev = connector->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_property *prop;
+	int i;
+
+	prop = dev_priv->force_audio_property;
+	if (prop == NULL) {
+		prop = drm_property_create(dev, DRM_MODE_PROP_ENUM,
+					   "audio",
+					   ARRAY_SIZE(force_audio_names));
+		if (prop == NULL)
+			return;
+
+		for (i = 0; i < ARRAY_SIZE(force_audio_names); i++)
+			drm_property_add_enum(prop, i, i-1, force_audio_names[i]);
+
+		dev_priv->force_audio_property = prop;
+	}
+	drm_connector_attach_property(connector, prop, 0);
+}
+
 static const char *broadcast_rgb_names[] = {
 	"Full",
 	"Limited 16:235",
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index a670c006982e..9e2959bc91cd 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -1409,6 +1409,11 @@ void intel_setup_overlay(struct drm_device *dev)
 	overlay = kzalloc(sizeof(struct intel_overlay), GFP_KERNEL);
 	if (!overlay)
 		return;
+
+	mutex_lock(&dev->struct_mutex);
+	if (WARN_ON(dev_priv->overlay))
+		goto out_free;
+
 	overlay->dev = dev;
 
 	reg_bo = i915_gem_alloc_object(dev, PAGE_SIZE);
@@ -1448,7 +1453,7 @@ void intel_setup_overlay(struct drm_device *dev)
 
 	regs = intel_overlay_map_regs(overlay);
 	if (!regs)
-		goto out_free_bo;
+		goto out_unpin_bo;
 
 	memset(regs, 0, sizeof(struct overlay_registers));
 	update_polyphase_filter(regs);
@@ -1457,14 +1462,17 @@ void intel_setup_overlay(struct drm_device *dev)
 	intel_overlay_unmap_regs(overlay, regs);
 
 	dev_priv->overlay = overlay;
+	mutex_unlock(&dev->struct_mutex);
 	DRM_INFO("initialized overlay support\n");
 	return;
 
 out_unpin_bo:
-	i915_gem_object_unpin(reg_bo);
+	if (!OVERLAY_NEEDS_PHYSICAL(dev))
+		i915_gem_object_unpin(reg_bo);
 out_free_bo:
 	drm_gem_object_unreference(&reg_bo->base);
 out_free:
+	mutex_unlock(&dev->struct_mutex);
 	kfree(overlay);
 	return;
 }
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index c0e0ee63fbf4..39ac2b634ae5 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -165,7 +165,7 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring);
 int __must_check intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n);
 static inline int intel_wait_ring_idle(struct intel_ring_buffer *ring)
 {
-	return intel_wait_ring_buffer(ring, ring->space - 8);
+	return intel_wait_ring_buffer(ring, ring->size - 8);
 }
 
 int __must_check intel_ring_begin(struct intel_ring_buffer *ring, int n);
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 754086f83941..30fe554d8936 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -148,8 +148,6 @@ struct intel_sdvo_connector {
 	int   format_supported_num;
 	struct drm_property *tv_format;
 
-	struct drm_property *force_audio_property;
-
 	/* add the property for the SDVO-TV */
 	struct drm_property *left;
 	struct drm_property *right;
@@ -1712,7 +1710,7 @@ intel_sdvo_set_property(struct drm_connector *connector,
 	if (ret)
 		return ret;
 
-	if (property == intel_sdvo_connector->force_audio_property) {
+	if (property == dev_priv->force_audio_property) {
 		int i = val;
 		bool has_audio;
 
@@ -2037,15 +2035,7 @@ intel_sdvo_add_hdmi_properties(struct intel_sdvo_connector *connector)
 {
 	struct drm_device *dev = connector->base.base.dev;
 
-	connector->force_audio_property =
-		drm_property_create(dev, DRM_MODE_PROP_RANGE, "force_audio", 2);
-	if (connector->force_audio_property) {
-		connector->force_audio_property->values[0] = -1;
-		connector->force_audio_property->values[1] = 1;
-		drm_connector_attach_property(&connector->base.base,
-					      connector->force_audio_property, 0);
-	}
-
+	intel_attach_force_audio_property(&connector->base.base);
 	if (INTEL_INFO(dev)->gen >= 4 && IS_MOBILE(dev))
 		intel_attach_broadcast_rgb_property(&connector->base.base);
 }
diff --git a/drivers/gpu/drm/mga/mga_drv.h b/drivers/gpu/drm/mga/mga_drv.h
index 1084fa4d261b..54558a01969a 100644
--- a/drivers/gpu/drm/mga/mga_drv.h
+++ b/drivers/gpu/drm/mga/mga_drv.h
@@ -195,29 +195,10 @@ extern long mga_compat_ioctl(struct file *filp, unsigned int cmd,
 
 #define mga_flush_write_combine()	DRM_WRITEMEMORYBARRIER()
 
-#if defined(__linux__) && defined(__alpha__)
-#define MGA_BASE(reg)		((unsigned long)(dev_priv->mmio->handle))
-#define MGA_ADDR(reg)		(MGA_BASE(reg) + reg)
-
-#define MGA_DEREF(reg)		(*(volatile u32 *)MGA_ADDR(reg))
-#define MGA_DEREF8(reg)		(*(volatile u8 *)MGA_ADDR(reg))
-
-#define MGA_READ(reg)		(_MGA_READ((u32 *)MGA_ADDR(reg)))
-#define MGA_READ8(reg)		(_MGA_READ((u8 *)MGA_ADDR(reg)))
-#define MGA_WRITE(reg, val)	do { DRM_WRITEMEMORYBARRIER(); MGA_DEREF(reg) = val; } while (0)
-#define MGA_WRITE8(reg, val)	do { DRM_WRITEMEMORYBARRIER(); MGA_DEREF8(reg) = val; } while (0)
-
-static inline u32 _MGA_READ(u32 *addr)
-{
-	DRM_MEMORYBARRIER();
-	return *(volatile u32 *)addr;
-}
-#else
 #define MGA_READ8(reg)		DRM_READ8(dev_priv->mmio, (reg))
 #define MGA_READ(reg)		DRM_READ32(dev_priv->mmio, (reg))
 #define MGA_WRITE8(reg, val)	DRM_WRITE8(dev_priv->mmio, (reg), (val))
 #define MGA_WRITE(reg, val)	DRM_WRITE32(dev_priv->mmio, (reg), (val))
-#endif
 
 #define DWGREG0		0x1c00
 #define DWGREG0_END	0x1dff
diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c
index f0d459bb46e4..525744d593c1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_acpi.c
+++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c
@@ -262,7 +262,6 @@ static bool nouveau_dsm_detect(void)
 		vga_count++;
 
 		retval = nouveau_dsm_pci_probe(pdev);
-		printk("ret val is %d\n", retval);
 		if (retval & NOUVEAU_DSM_HAS_MUX)
 			has_dsm |= 1;
 		if (retval & NOUVEAU_DSM_HAS_OPT)
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 4b9f4493c9f9..7347075ca5b8 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -339,11 +339,12 @@ semaphore_acquire(struct nouveau_channel *chan, struct nouveau_semaphore *sema)
 	int ret;
 
 	if (dev_priv->chipset < 0x84) {
-		ret = RING_SPACE(chan, 3);
+		ret = RING_SPACE(chan, 4);
 		if (ret)
 			return ret;
 
-		BEGIN_RING(chan, NvSubSw, NV_SW_SEMAPHORE_OFFSET, 2);
+		BEGIN_RING(chan, NvSubSw, NV_SW_DMA_SEMAPHORE, 3);
+		OUT_RING  (chan, NvSema);
 		OUT_RING  (chan, sema->mem->start);
 		OUT_RING  (chan, 1);
 	} else
@@ -351,10 +352,12 @@ semaphore_acquire(struct nouveau_channel *chan, struct nouveau_semaphore *sema)
 		struct nouveau_vma *vma = &dev_priv->fence.bo->vma;
 		u64 offset = vma->offset + sema->mem->start;
 
-		ret = RING_SPACE(chan, 5);
+		ret = RING_SPACE(chan, 7);
 		if (ret)
 			return ret;
 
+		BEGIN_RING(chan, NvSubSw, NV_SW_DMA_SEMAPHORE, 1);
+		OUT_RING  (chan, chan->vram_handle);
 		BEGIN_RING(chan, NvSubSw, 0x0010, 4);
 		OUT_RING  (chan, upper_32_bits(offset));
 		OUT_RING  (chan, lower_32_bits(offset));
@@ -394,11 +397,12 @@ semaphore_release(struct nouveau_channel *chan, struct nouveau_semaphore *sema)
 	int ret;
 
 	if (dev_priv->chipset < 0x84) {
-		ret = RING_SPACE(chan, 4);
+		ret = RING_SPACE(chan, 5);
 		if (ret)
 			return ret;
 
-		BEGIN_RING(chan, NvSubSw, NV_SW_SEMAPHORE_OFFSET, 1);
+		BEGIN_RING(chan, NvSubSw, NV_SW_DMA_SEMAPHORE, 2);
+		OUT_RING  (chan, NvSema);
 		OUT_RING  (chan, sema->mem->start);
 		BEGIN_RING(chan, NvSubSw, NV_SW_SEMAPHORE_RELEASE, 1);
 		OUT_RING  (chan, 1);
@@ -407,10 +411,12 @@ semaphore_release(struct nouveau_channel *chan, struct nouveau_semaphore *sema)
 		struct nouveau_vma *vma = &dev_priv->fence.bo->vma;
 		u64 offset = vma->offset + sema->mem->start;
 
-		ret = RING_SPACE(chan, 5);
+		ret = RING_SPACE(chan, 7);
 		if (ret)
 			return ret;
 
+		BEGIN_RING(chan, NvSubSw, NV_SW_DMA_SEMAPHORE, 1);
+		OUT_RING  (chan, chan->vram_handle);
 		BEGIN_RING(chan, NvSubSw, 0x0010, 4);
 		OUT_RING  (chan, upper_32_bits(offset));
 		OUT_RING  (chan, lower_32_bits(offset));
@@ -504,22 +510,22 @@ nouveau_fence_channel_init(struct nouveau_channel *chan)
 	struct nouveau_gpuobj *obj = NULL;
 	int ret;
 
-	if (dev_priv->card_type >= NV_C0)
-		goto out_initialised;
+	if (dev_priv->card_type < NV_C0) {
+		/* Create an NV_SW object for various sync purposes */
+		ret = nouveau_gpuobj_gr_new(chan, NvSw, NV_SW);
+		if (ret)
+			return ret;
 
-	/* Create an NV_SW object for various sync purposes */
-	ret = nouveau_gpuobj_gr_new(chan, NvSw, NV_SW);
-	if (ret)
-		return ret;
+		ret = RING_SPACE(chan, 2);
+		if (ret)
+			return ret;
 
-	/* we leave subchannel empty for nvc0 */
-	ret = RING_SPACE(chan, 2);
-	if (ret)
-		return ret;
-	BEGIN_RING(chan, NvSubSw, 0, 1);
-	OUT_RING(chan, NvSw);
+		BEGIN_RING(chan, NvSubSw, 0, 1);
+		OUT_RING  (chan, NvSw);
+		FIRE_RING (chan);
+	}
 
-	/* Create a DMA object for the shared cross-channel sync area. */
+	/* Setup area of memory shared between all channels for x-chan sync */
 	if (USE_SEMA(dev) && dev_priv->chipset < 0x84) {
 		struct ttm_mem_reg *mem = &dev_priv->fence.bo->bo.mem;
 
@@ -534,23 +540,8 @@ nouveau_fence_channel_init(struct nouveau_channel *chan)
 		nouveau_gpuobj_ref(NULL, &obj);
 		if (ret)
 			return ret;
-
-		ret = RING_SPACE(chan, 2);
-		if (ret)
-			return ret;
-		BEGIN_RING(chan, NvSubSw, NV_SW_DMA_SEMAPHORE, 1);
-		OUT_RING(chan, NvSema);
-	} else {
-		ret = RING_SPACE(chan, 2);
-		if (ret)
-			return ret;
-		BEGIN_RING(chan, NvSubSw, NV_SW_DMA_SEMAPHORE, 1);
-		OUT_RING  (chan, chan->vram_handle); /* whole VM */
 	}
 
-	FIRE_RING(chan);
-
-out_initialised:
 	INIT_LIST_HEAD(&chan->fence.pending);
 	spin_lock_init(&chan->fence.lock);
 	atomic_set(&chan->fence.last_sequence_irq, 0);
diff --git a/drivers/gpu/drm/nouveau/nouveau_hw.c b/drivers/gpu/drm/nouveau/nouveau_hw.c
index 053edf9d2f67..ba896e54b799 100644
--- a/drivers/gpu/drm/nouveau/nouveau_hw.c
+++ b/drivers/gpu/drm/nouveau/nouveau_hw.c
@@ -900,6 +900,7 @@ nv_save_state_ext(struct drm_device *dev, int head,
 	}
 	/* NV11 and NV20 don't have this, they stop at 0x52. */
 	if (nv_gf4_disp_arch(dev)) {
+		rd_cio_state(dev, head, regp, NV_CIO_CRE_42);
 		rd_cio_state(dev, head, regp, NV_CIO_CRE_53);
 		rd_cio_state(dev, head, regp, NV_CIO_CRE_54);
 
@@ -1003,6 +1004,7 @@ nv_load_state_ext(struct drm_device *dev, int head,
 			nouveau_wait_eq(dev, 650000000, NV_PRMCIO_INP0__COLOR, 0x8, 0x0);
 		}
 
+		wr_cio_state(dev, head, regp, NV_CIO_CRE_42);
 		wr_cio_state(dev, head, regp, NV_CIO_CRE_53);
 		wr_cio_state(dev, head, regp, NV_CIO_CRE_54);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c
index 2960f583dc38..5ee14d216ce8 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_mem.c
@@ -397,7 +397,7 @@ nouveau_mem_vram_init(struct drm_device *dev)
 		if (pci_dma_supported(dev->pdev, DMA_BIT_MASK(40)))
 			dma_bits = 40;
 	} else
-	if (drm_pci_device_is_pcie(dev) &&
+	if (0 && drm_pci_device_is_pcie(dev) &&
 	    dev_priv->chipset  > 0x40 &&
 	    dev_priv->chipset != 0x45) {
 		if (pci_dma_supported(dev->pdev, DMA_BIT_MASK(39)))
@@ -868,7 +868,9 @@ nouveau_gart_manager_del(struct ttm_mem_type_manager *man,
 		nouveau_vm_unmap(&node->tmp_vma);
 		nouveau_vm_put(&node->tmp_vma);
 	}
+
 	mem->mm_node = NULL;
+	kfree(node);
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/nouveau_perf.c b/drivers/gpu/drm/nouveau/nouveau_perf.c
index 922fb6b664ed..ef9dec0e6f8b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_perf.c
+++ b/drivers/gpu/drm/nouveau/nouveau_perf.c
@@ -182,6 +182,11 @@ nouveau_perf_init(struct drm_device *dev)
 		entries   = perf[2];
 	}
 
+	if (entries > NOUVEAU_PM_MAX_LEVEL) {
+		NV_DEBUG(dev, "perf table has too many entries - buggy vbios?\n");
+		entries = NOUVEAU_PM_MAX_LEVEL;
+	}
+
 	entry = perf + headerlen;
 	for (i = 0; i < entries; i++) {
 		struct nouveau_pm_level *perflvl = &pm->perflvl[pm->nr_perflvl];
diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c
index c77111eca6ac..82fad914e648 100644
--- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c
+++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c
@@ -458,7 +458,7 @@ nouveau_sgdma_init(struct drm_device *dev)
 		dev_priv->gart_info.type = NOUVEAU_GART_HW;
 		dev_priv->gart_info.func = &nv50_sgdma_backend;
 	} else
-	if (drm_pci_device_is_pcie(dev) &&
+	if (0 && drm_pci_device_is_pcie(dev) &&
 	    dev_priv->chipset > 0x40 && dev_priv->chipset != 0x45) {
 		if (nv44_graph_class(dev)) {
 			dev_priv->gart_info.func = &nv44_sgdma_backend;
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index 38ea662568c1..731acea865b5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -563,68 +563,68 @@ nouveau_card_init(struct drm_device *dev)
 	if (ret)
 		goto out_timer;
 
-	switch (dev_priv->card_type) {
-	case NV_04:
-		nv04_graph_create(dev);
-		break;
-	case NV_10:
-		nv10_graph_create(dev);
-		break;
-	case NV_20:
-	case NV_30:
-		nv20_graph_create(dev);
-		break;
-	case NV_40:
-		nv40_graph_create(dev);
-		break;
-	case NV_50:
-		nv50_graph_create(dev);
-		break;
-	case NV_C0:
-		nvc0_graph_create(dev);
-		break;
-	default:
-		break;
-	}
-
-	switch (dev_priv->chipset) {
-	case 0x84:
-	case 0x86:
-	case 0x92:
-	case 0x94:
-	case 0x96:
-	case 0xa0:
-		nv84_crypt_create(dev);
-		break;
-	}
+	if (!nouveau_noaccel) {
+		switch (dev_priv->card_type) {
+		case NV_04:
+			nv04_graph_create(dev);
+			break;
+		case NV_10:
+			nv10_graph_create(dev);
+			break;
+		case NV_20:
+		case NV_30:
+			nv20_graph_create(dev);
+			break;
+		case NV_40:
+			nv40_graph_create(dev);
+			break;
+		case NV_50:
+			nv50_graph_create(dev);
+			break;
+		case NV_C0:
+			nvc0_graph_create(dev);
+			break;
+		default:
+			break;
+		}
 
-	switch (dev_priv->card_type) {
-	case NV_50:
 		switch (dev_priv->chipset) {
-		case 0xa3:
-		case 0xa5:
-		case 0xa8:
-		case 0xaf:
-			nva3_copy_create(dev);
+		case 0x84:
+		case 0x86:
+		case 0x92:
+		case 0x94:
+		case 0x96:
+		case 0xa0:
+			nv84_crypt_create(dev);
 			break;
 		}
-		break;
-	case NV_C0:
-		nvc0_copy_create(dev, 0);
-		nvc0_copy_create(dev, 1);
-		break;
-	default:
-		break;
-	}
 
-	if (dev_priv->card_type == NV_40)
-		nv40_mpeg_create(dev);
-	else
-	if (dev_priv->card_type == NV_50 &&
-	    (dev_priv->chipset < 0x98 || dev_priv->chipset == 0xa0))
-		nv50_mpeg_create(dev);
+		switch (dev_priv->card_type) {
+		case NV_50:
+			switch (dev_priv->chipset) {
+			case 0xa3:
+			case 0xa5:
+			case 0xa8:
+			case 0xaf:
+				nva3_copy_create(dev);
+				break;
+			}
+			break;
+		case NV_C0:
+			nvc0_copy_create(dev, 0);
+			nvc0_copy_create(dev, 1);
+			break;
+		default:
+			break;
+		}
+
+		if (dev_priv->card_type == NV_40)
+			nv40_mpeg_create(dev);
+		else
+		if (dev_priv->card_type == NV_50 &&
+		    (dev_priv->chipset < 0x98 || dev_priv->chipset == 0xa0))
+			nv50_mpeg_create(dev);
 
-	if (!nouveau_noaccel) {
 		for (e = 0; e < NVOBJ_ENGINE_NR; e++) {
 			if (dev_priv->eng[e]) {
 				ret = dev_priv->eng[e]->init(dev, e);
@@ -880,8 +880,8 @@ int nouveau_load(struct drm_device *dev, unsigned long flags)
 
 #ifdef __BIG_ENDIAN
 	/* Put the card in BE mode if it's not */
-	if (nv_rd32(dev, NV03_PMC_BOOT_1))
-		nv_wr32(dev, NV03_PMC_BOOT_1, 0x00000001);
+	if (nv_rd32(dev, NV03_PMC_BOOT_1) != 0x01000001)
+		nv_wr32(dev, NV03_PMC_BOOT_1, 0x01000001);
 
 	DRM_MEMORYBARRIER();
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_vm.c b/drivers/gpu/drm/nouveau/nouveau_vm.c
index 0059e6f58a8b..519a6b4bba46 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_vm.c
@@ -58,6 +58,7 @@ nouveau_vm_map_at(struct nouveau_vma *vma, u64 delta, struct nouveau_mem *node)
 			num -= len;
 			pte += len;
 			if (unlikely(end >= max)) {
+				phys += len << (bits + 12);
 				pde++;
 				pte = 0;
 			}
diff --git a/drivers/gpu/drm/nouveau/nv04_crtc.c b/drivers/gpu/drm/nouveau/nv04_crtc.c
index 3c78bc81357e..f1a3ae491995 100644
--- a/drivers/gpu/drm/nouveau/nv04_crtc.c
+++ b/drivers/gpu/drm/nouveau/nv04_crtc.c
@@ -376,7 +376,10 @@ nv_crtc_mode_set_vga(struct drm_crtc *crtc, struct drm_display_mode *mode)
 	 */
 
 	/* framebuffer can be larger than crtc scanout area. */
-	regp->CRTC[NV_CIO_CRE_RPC0_INDEX] = XLATE(fb->pitch / 8, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
+	regp->CRTC[NV_CIO_CRE_RPC0_INDEX] =
+		XLATE(fb->pitch / 8, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
+	regp->CRTC[NV_CIO_CRE_42] =
+		XLATE(fb->pitch / 8, 11, NV_CIO_CRE_42_OFFSET_11);
 	regp->CRTC[NV_CIO_CRE_RPC1_INDEX] = mode->crtc_hdisplay < 1280 ?
 					    MASK(NV_CIO_CRE_RPC1_LARGE) : 0x00;
 	regp->CRTC[NV_CIO_CRE_LSR_INDEX] = XLATE(horizBlankEnd, 6, NV_CIO_CRE_LSR_HBE_6) |
@@ -824,8 +827,11 @@ nv04_crtc_do_mode_set_base(struct drm_crtc *crtc,
 	regp->CRTC[NV_CIO_CR_OFFSET_INDEX] = drm_fb->pitch >> 3;
 	regp->CRTC[NV_CIO_CRE_RPC0_INDEX] =
 		XLATE(drm_fb->pitch >> 3, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
+	regp->CRTC[NV_CIO_CRE_42] =
+		XLATE(drm_fb->pitch / 8, 11, NV_CIO_CRE_42_OFFSET_11);
 	crtc_wr_cio_state(crtc, regp, NV_CIO_CRE_RPC0_INDEX);
 	crtc_wr_cio_state(crtc, regp, NV_CIO_CR_OFFSET_INDEX);
+	crtc_wr_cio_state(crtc, regp, NV_CIO_CRE_42);
 
 	/* Update the framebuffer location. */
 	regp->fb_start = nv_crtc->fb.offset & ~3;
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 74a3f6872701..08da478ba544 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -409,7 +409,7 @@ nv50_display_flip_next(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	struct nouveau_channel *evo = dispc->sync;
 	int ret;
 
-	ret = RING_SPACE(evo, 24);
+	ret = RING_SPACE(evo, chan ? 25 : 27);
 	if (unlikely(ret))
 		return ret;
 
@@ -458,8 +458,19 @@ nv50_display_flip_next(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	/* queue the flip on the crtc's "display sync" channel */
 	BEGIN_RING(evo, 0, 0x0100, 1);
 	OUT_RING  (evo, 0xfffe0000);
-	BEGIN_RING(evo, 0, 0x0084, 5);
-	OUT_RING  (evo, chan ? 0x00000100 : 0x00000010);
+	if (chan) {
+		BEGIN_RING(evo, 0, 0x0084, 1);
+		OUT_RING  (evo, 0x00000100);
+	} else {
+		BEGIN_RING(evo, 0, 0x0084, 1);
+		OUT_RING  (evo, 0x00000010);
+		/* allows gamma somehow, PDISP will bitch at you if
+		 * you don't wait for vblank before changing this..
+		 */
+		BEGIN_RING(evo, 0, 0x00e0, 1);
+		OUT_RING  (evo, 0x40000000);
+	}
+	BEGIN_RING(evo, 0, 0x0088, 4);
 	OUT_RING  (evo, dispc->sem.offset);
 	OUT_RING  (evo, 0xf00d0000 | dispc->sem.value);
 	OUT_RING  (evo, 0x74b1e000);
diff --git a/drivers/gpu/drm/nouveau/nvreg.h b/drivers/gpu/drm/nouveau/nvreg.h
index fe0f253089ac..bbfb1a68fb11 100644
--- a/drivers/gpu/drm/nouveau/nvreg.h
+++ b/drivers/gpu/drm/nouveau/nvreg.h
@@ -277,6 +277,8 @@
 #		define NV_CIO_CRE_EBR_VDE_11		2:2
 #		define NV_CIO_CRE_EBR_VRS_11		4:4
 #		define NV_CIO_CRE_EBR_VBS_11		6:6
+#	define NV_CIO_CRE_42			0x42
+#		define NV_CIO_CRE_42_OFFSET_11		6:6
 #	define NV_CIO_CRE_43			0x43
 #	define NV_CIO_CRE_44			0x44	/* head control */
 #	define NV_CIO_CRE_CSB			0x45	/* colour saturation boost */
diff --git a/drivers/gpu/drm/radeon/Kconfig b/drivers/gpu/drm/radeon/Kconfig
index 9746fee59f56..ea92bbe3ed37 100644
--- a/drivers/gpu/drm/radeon/Kconfig
+++ b/drivers/gpu/drm/radeon/Kconfig
@@ -28,11 +28,4 @@ config DRM_RADEON_KMS
 	  The kernel will also perform security check on command stream
 	  provided by the user, we want to catch and forbid any illegal use
 	  of the GPU such as DMA into random system memory or into memory
-	  not owned by the process supplying the command stream. This part
-	  of the code is still incomplete and this why we propose that patch
-	  as a staging driver addition, future security might forbid current
-	  experimental userspace to run.
-
-	  This code support the following hardware : R1XX,R2XX,R3XX,R4XX,R5XX
-	  (radeon up to X1950). Works is underway to provide support for R6XX,
-	  R7XX and newer hardware (radeon from HD2XXX to HD4XXX).
+	  not owned by the process supplying the command stream.
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index 49611e2365d9..1b50ad8919d5 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -1200,6 +1200,7 @@ typedef struct _EXTERNAL_ENCODER_CONTROL_PARAMETERS_V3
 #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
+#define EXTERNAL_ENCODER_ACTION_V3_DDC_SETUP              0x14
 
 // ucConfig
 #define EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_MASK				0x03
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index ec848787d7d9..9541995e4b21 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -671,6 +671,13 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 								DISPPLL_CONFIG_DUAL_LINK;
 					}
 				}
+				if (radeon_encoder_is_dp_bridge(encoder)) {
+					struct drm_encoder *ext_encoder = radeon_atom_get_external_encoder(encoder);
+					struct radeon_encoder *ext_radeon_encoder = to_radeon_encoder(ext_encoder);
+					args.v3.sInput.ucExtTransmitterID = ext_radeon_encoder->encoder_id;
+				} else
+					args.v3.sInput.ucExtTransmitterID = 0;
+
 				atom_execute_table(rdev->mode_info.atom_context,
 						   index, (uint32_t *)&args);
 				adjusted_clock = le32_to_cpu(args.v3.sOutput.ulDispPllFreq) * 10;
@@ -1045,7 +1052,7 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 	uint64_t fb_location;
 	uint32_t fb_format, fb_pitch_pixels, tiling_flags;
 	u32 fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_NONE);
-	u32 tmp;
+	u32 tmp, viewport_w, viewport_h;
 	int r;
 
 	/* no fb bound */
@@ -1171,8 +1178,10 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 	y &= ~1;
 	WREG32(EVERGREEN_VIEWPORT_START + radeon_crtc->crtc_offset,
 	       (x << 16) | y);
+	viewport_w = crtc->mode.hdisplay;
+	viewport_h = (crtc->mode.vdisplay + 1) & ~1;
 	WREG32(EVERGREEN_VIEWPORT_SIZE + radeon_crtc->crtc_offset,
-	       (crtc->mode.hdisplay << 16) | crtc->mode.vdisplay);
+	       (viewport_w << 16) | viewport_h);
 
 	/* pageflip setup */
 	/* make sure flip is at vb rather than hb */
@@ -1213,7 +1222,7 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 	uint64_t fb_location;
 	uint32_t fb_format, fb_pitch_pixels, tiling_flags;
 	u32 fb_swap = R600_D1GRPH_SWAP_ENDIAN_NONE;
-	u32 tmp;
+	u32 tmp, viewport_w, viewport_h;
 	int r;
 
 	/* no fb bound */
@@ -1338,8 +1347,10 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc,
 	y &= ~1;
 	WREG32(AVIVO_D1MODE_VIEWPORT_START + radeon_crtc->crtc_offset,
 	       (x << 16) | y);
+	viewport_w = crtc->mode.hdisplay;
+	viewport_h = (crtc->mode.vdisplay + 1) & ~1;
 	WREG32(AVIVO_D1MODE_VIEWPORT_SIZE + radeon_crtc->crtc_offset,
-	       (crtc->mode.hdisplay << 16) | crtc->mode.vdisplay);
+	       (viewport_w << 16) | viewport_h);
 
 	/* pageflip setup */
 	/* make sure flip is at vb rather than hb */
diff --git a/drivers/gpu/drm/radeon/cayman_blit_shaders.c b/drivers/gpu/drm/radeon/cayman_blit_shaders.c
index e148ab04b80b..7b4eeb7b4a8c 100644
--- a/drivers/gpu/drm/radeon/cayman_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/cayman_blit_shaders.c
@@ -39,17 +39,335 @@
 
 const u32 cayman_default_state[] =
 {
-	/* XXX fill in additional blit state */
+	0xc0066900,
+	0x00000000,
+	0x00000060, /* DB_RENDER_CONTROL */
+	0x00000000, /* DB_COUNT_CONTROL */
+	0x00000000, /* DB_DEPTH_VIEW */
+	0x0000002a, /* DB_RENDER_OVERRIDE */
+	0x00000000, /* DB_RENDER_OVERRIDE2 */
+	0x00000000, /* DB_HTILE_DATA_BASE */
 
 	0xc0026900,
-	0x00000316,
-	0x0000000e, /* VGT_VERTEX_REUSE_BLOCK_CNTL */
-	0x00000010, /*  */
+	0x0000000a,
+	0x00000000, /* DB_STENCIL_CLEAR */
+	0x00000000, /* DB_DEPTH_CLEAR */
+
+	0xc0036900,
+	0x0000000f,
+	0x00000000, /* DB_DEPTH_INFO */
+	0x00000000, /* DB_Z_INFO */
+	0x00000000, /* DB_STENCIL_INFO */
+
+	0xc0016900,
+	0x00000080,
+	0x00000000, /* PA_SC_WINDOW_OFFSET */
+
+	0xc00d6900,
+	0x00000083,
+	0x0000ffff, /* PA_SC_CLIPRECT_RULE */
+	0x00000000, /* PA_SC_CLIPRECT_0_TL */
+	0x20002000, /* PA_SC_CLIPRECT_0_BR */
+	0x00000000,
+	0x20002000,
+	0x00000000,
+	0x20002000,
+	0x00000000,
+	0x20002000,
+	0xaaaaaaaa, /* PA_SC_EDGERULE */
+	0x00000000, /* PA_SU_HARDWARE_SCREEN_OFFSET */
+	0x0000000f, /* CB_TARGET_MASK */
+	0x0000000f, /* CB_SHADER_MASK */
+
+	0xc0226900,
+	0x00000094,
+	0x80000000, /* PA_SC_VPORT_SCISSOR_0_TL */
+	0x20002000, /* PA_SC_VPORT_SCISSOR_0_BR */
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x80000000,
+	0x20002000,
+	0x00000000, /* PA_SC_VPORT_ZMIN_0 */
+	0x3f800000, /* PA_SC_VPORT_ZMAX_0 */
+
+	0xc0016900,
+	0x000000d4,
+	0x00000000, /* SX_MISC */
 
 	0xc0026900,
 	0x000000d9,
 	0x00000000, /* CP_RINGID */
 	0x00000000, /* CP_VMID */
+
+	0xc0096900,
+	0x00000100,
+	0x00ffffff, /* VGT_MAX_VTX_INDX */
+	0x00000000, /* VGT_MIN_VTX_INDX */
+	0x00000000, /* VGT_INDX_OFFSET */
+	0x00000000, /* VGT_MULTI_PRIM_IB_RESET_INDX */
+	0x00000000, /* SX_ALPHA_TEST_CONTROL */
+	0x00000000, /* CB_BLEND_RED */
+	0x00000000, /* CB_BLEND_GREEN */
+	0x00000000, /* CB_BLEND_BLUE */
+	0x00000000, /* CB_BLEND_ALPHA */
+
+	0xc0016900,
+	0x00000187,
+	0x00000100, /* SPI_VS_OUT_ID_0 */
+
+	0xc0026900,
+	0x00000191,
+	0x00000100, /* SPI_PS_INPUT_CNTL_0 */
+	0x00000101, /* SPI_PS_INPUT_CNTL_1 */
+
+	0xc0016900,
+	0x000001b1,
+	0x00000000, /* SPI_VS_OUT_CONFIG */
+
+	0xc0106900,
+	0x000001b3,
+	0x20000001, /* SPI_PS_IN_CONTROL_0 */
+	0x00000000, /* SPI_PS_IN_CONTROL_1 */
+	0x00000000, /* SPI_INTERP_CONTROL_0 */
+	0x00000000, /* SPI_INPUT_Z */
+	0x00000000, /* SPI_FOG_CNTL */
+	0x00100000, /* SPI_BARYC_CNTL */
+	0x00000000, /* SPI_PS_IN_CONTROL_2 */
+	0x00000000, /* SPI_COMPUTE_INPUT_CNTL */
+	0x00000000, /* SPI_COMPUTE_NUM_THREAD_X */
+	0x00000000, /* SPI_COMPUTE_NUM_THREAD_Y */
+	0x00000000, /* SPI_COMPUTE_NUM_THREAD_Z */
+	0x00000000, /* SPI_GPR_MGMT */
+	0x00000000, /* SPI_LDS_MGMT */
+	0x00000000, /* SPI_STACK_MGMT */
+	0x00000000, /* SPI_WAVE_MGMT_1 */
+	0x00000000, /* SPI_WAVE_MGMT_2 */
+
+	0xc0016900,
+	0x000001e0,
+	0x00000000, /* CB_BLEND0_CONTROL */
+
+	0xc00e6900,
+	0x00000200,
+	0x00000000, /* DB_DEPTH_CONTROL */
+	0x00000000, /* DB_EQAA */
+	0x00cc0010, /* CB_COLOR_CONTROL */
+	0x00000210, /* DB_SHADER_CONTROL */
+	0x00010000, /* PA_CL_CLIP_CNTL */
+	0x00000004, /* PA_SU_SC_MODE_CNTL */
+	0x00000100, /* PA_CL_VTE_CNTL */
+	0x00000000, /* PA_CL_VS_OUT_CNTL */
+	0x00000000, /* PA_CL_NANINF_CNTL */
+	0x00000000, /* PA_SU_LINE_STIPPLE_CNTL */
+	0x00000000, /* PA_SU_LINE_STIPPLE_SCALE */
+	0x00000000, /* PA_SU_PRIM_FILTER_CNTL */
+	0x00000000, /*  */
+	0x00000000, /*  */
+
+	0xc0026900,
+	0x00000229,
+	0x00000000, /* SQ_PGM_START_FS */
+	0x00000000,
+
+	0xc0016900,
+	0x0000023b,
+	0x00000000, /* SQ_LDS_ALLOC_PS */
+
+	0xc0066900,
+	0x00000240,
+	0x00000000, /* SQ_ESGS_RING_ITEMSIZE */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+
+	0xc0046900,
+	0x00000247,
+	0x00000000, /* SQ_GS_VERT_ITEMSIZE */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+
+	0xc0116900,
+	0x00000280,
+	0x00000000, /* PA_SU_POINT_SIZE */
+	0x00000000, /* PA_SU_POINT_MINMAX */
+	0x00000008, /* PA_SU_LINE_CNTL */
+	0x00000000, /* PA_SC_LINE_STIPPLE */
+	0x00000000, /* VGT_OUTPUT_PATH_CNTL */
+	0x00000000, /* VGT_HOS_CNTL */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000, /* VGT_GS_MODE */
+
+	0xc0026900,
+	0x00000292,
+	0x00000000, /* PA_SC_MODE_CNTL_0 */
+	0x00000000, /* PA_SC_MODE_CNTL_1 */
+
+	0xc0016900,
+	0x000002a1,
+	0x00000000, /* VGT_PRIMITIVEID_EN */
+
+	0xc0016900,
+	0x000002a5,
+	0x00000000, /* VGT_MULTI_PRIM_IB_RESET_EN */
+
+	0xc0026900,
+	0x000002a8,
+	0x00000000, /* VGT_INSTANCE_STEP_RATE_0 */
+	0x00000000,
+
+	0xc0026900,
+	0x000002ad,
+	0x00000000, /* VGT_REUSE_OFF */
+	0x00000000,
+
+	0xc0016900,
+	0x000002d5,
+	0x00000000, /* VGT_SHADER_STAGES_EN */
+
+	0xc0016900,
+	0x000002dc,
+	0x0000aa00, /* DB_ALPHA_TO_MASK */
+
+	0xc0066900,
+	0x000002de,
+	0x00000000, /* PA_SU_POLY_OFFSET_DB_FMT_CNTL */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+
+	0xc0026900,
+	0x000002e5,
+	0x00000000, /* VGT_STRMOUT_CONFIG */
+	0x00000000,
+
+	0xc01b6900,
+	0x000002f5,
+	0x76543210, /* PA_SC_CENTROID_PRIORITY_0 */
+	0xfedcba98, /* PA_SC_CENTROID_PRIORITY_1 */
+	0x00000000, /* PA_SC_LINE_CNTL */
+	0x00000000, /* PA_SC_AA_CONFIG */
+	0x00000005, /* PA_SU_VTX_CNTL */
+	0x3f800000, /* PA_CL_GB_VERT_CLIP_ADJ */
+	0x3f800000, /* PA_CL_GB_VERT_DISC_ADJ */
+	0x3f800000, /* PA_CL_GB_HORZ_CLIP_ADJ */
+	0x3f800000, /* PA_CL_GB_HORZ_DISC_ADJ */
+	0x00000000, /* PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0 */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xffffffff, /* PA_SC_AA_MASK_X0Y0_X1Y0 */
+	0xffffffff,
+
+	0xc0026900,
+	0x00000316,
+	0x0000000e, /* VGT_VERTEX_REUSE_BLOCK_CNTL */
+	0x00000010, /*  */
+};
+
+const u32 cayman_vs[] =
+{
+	0x00000004,
+	0x80400400,
+	0x0000a03c,
+	0x95000688,
+	0x00004000,
+	0x15000688,
+	0x00000000,
+	0x88000000,
+	0x04000000,
+	0x67961001,
+#ifdef __BIG_ENDIAN
+	0x00020000,
+#else
+	0x00000000,
+#endif
+	0x00000000,
+	0x04000000,
+	0x67961000,
+#ifdef __BIG_ENDIAN
+	0x00020008,
+#else
+	0x00000008,
+#endif
+	0x00000000,
+};
+
+const u32 cayman_ps[] =
+{
+	0x00000004,
+	0xa00c0000,
+	0x00000008,
+	0x80400000,
+	0x00000000,
+	0x95000688,
+	0x00000000,
+	0x88000000,
+	0x00380400,
+	0x00146b10,
+	0x00380000,
+	0x20146b10,
+	0x00380400,
+	0x40146b00,
+	0x80380000,
+	0x60146b00,
+	0x00000010,
+	0x000d1000,
+	0xb0800000,
+	0x00000000,
 };
 
+const u32 cayman_ps_size = ARRAY_SIZE(cayman_ps);
+const u32 cayman_vs_size = ARRAY_SIZE(cayman_vs);
 const u32 cayman_default_size = ARRAY_SIZE(cayman_default_state);
diff --git a/drivers/gpu/drm/radeon/cayman_blit_shaders.h b/drivers/gpu/drm/radeon/cayman_blit_shaders.h
index 33b75e5d0fa4..f5d0e9a60267 100644
--- a/drivers/gpu/drm/radeon/cayman_blit_shaders.h
+++ b/drivers/gpu/drm/radeon/cayman_blit_shaders.h
@@ -25,8 +25,11 @@
 #ifndef CAYMAN_BLIT_SHADERS_H
 #define CAYMAN_BLIT_SHADERS_H
 
+extern const u32 cayman_ps[];
+extern const u32 cayman_vs[];
 extern const u32 cayman_default_state[];
 
+extern const u32 cayman_ps_size, cayman_vs_size;
 extern const u32 cayman_default_size;
 
 #endif
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 7c37638095f7..15bd0477a3e8 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -88,21 +88,40 @@ u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 /* get temperature in millidegrees */
 int evergreen_get_temp(struct radeon_device *rdev)
 {
-	u32 temp = (RREG32(CG_MULT_THERMAL_STATUS) & ASIC_T_MASK) >>
-		ASIC_T_SHIFT;
-	u32 actual_temp = 0;
+	u32 temp, toffset;
+	int actual_temp = 0;
 
-	if (temp & 0x400)
-		actual_temp = -256;
-	else if (temp & 0x200)
-		actual_temp = 255;
-	else if (temp & 0x100) {
-		actual_temp = temp & 0x1ff;
-		actual_temp |= ~0x1ff;
-	} else
-		actual_temp = temp & 0xff;
+	if (rdev->family == CHIP_JUNIPER) {
+		toffset = (RREG32(CG_THERMAL_CTRL) & TOFFSET_MASK) >>
+			TOFFSET_SHIFT;
+		temp = (RREG32(CG_TS0_STATUS) & TS0_ADC_DOUT_MASK) >>
+			TS0_ADC_DOUT_SHIFT;
+
+		if (toffset & 0x100)
+			actual_temp = temp / 2 - (0x200 - toffset);
+		else
+			actual_temp = temp / 2 + toffset;
+
+		actual_temp = actual_temp * 1000;
+
+	} else {
+		temp = (RREG32(CG_MULT_THERMAL_STATUS) & ASIC_T_MASK) >>
+			ASIC_T_SHIFT;
 
-	return (actual_temp * 1000) / 2;
+		if (temp & 0x400)
+			actual_temp = -256;
+		else if (temp & 0x200)
+			actual_temp = 255;
+		else if (temp & 0x100) {
+			actual_temp = temp & 0x1ff;
+			actual_temp |= ~0x1ff;
+		} else
+			actual_temp = temp & 0xff;
+
+		actual_temp = (actual_temp * 1000) / 2;
+	}
+
+	return actual_temp;
 }
 
 int sumo_get_temp(struct radeon_device *rdev)
@@ -121,11 +140,17 @@ void evergreen_pm_misc(struct radeon_device *rdev)
 	struct radeon_voltage *voltage = &ps->clock_info[req_cm_idx].voltage;
 
 	if (voltage->type == VOLTAGE_SW) {
+		/* 0xff01 is a flag rather then an actual voltage */
+		if (voltage->voltage == 0xff01)
+			return;
 		if (voltage->voltage && (voltage->voltage != rdev->pm.current_vddc)) {
 			radeon_atom_set_voltage(rdev, voltage->voltage, SET_VOLTAGE_TYPE_ASIC_VDDC);
 			rdev->pm.current_vddc = voltage->voltage;
 			DRM_DEBUG("Setting: vddc: %d\n", voltage->voltage);
 		}
+		/* 0xff01 is a flag rather then an actual voltage */
+		if (voltage->vddci == 0xff01)
+			return;
 		if (voltage->vddci && (voltage->vddci != rdev->pm.current_vddci)) {
 			radeon_atom_set_voltage(rdev, voltage->vddci, SET_VOLTAGE_TYPE_ASIC_VDDCI);
 			rdev->pm.current_vddci = voltage->vddci;
@@ -960,17 +985,19 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav
 {
 	save->vga_control[0] = RREG32(D1VGA_CONTROL);
 	save->vga_control[1] = RREG32(D2VGA_CONTROL);
-	save->vga_control[2] = RREG32(EVERGREEN_D3VGA_CONTROL);
-	save->vga_control[3] = RREG32(EVERGREEN_D4VGA_CONTROL);
-	save->vga_control[4] = RREG32(EVERGREEN_D5VGA_CONTROL);
-	save->vga_control[5] = RREG32(EVERGREEN_D6VGA_CONTROL);
 	save->vga_render_control = RREG32(VGA_RENDER_CONTROL);
 	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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
+		save->vga_control[2] = RREG32(EVERGREEN_D3VGA_CONTROL);
+		save->vga_control[3] = RREG32(EVERGREEN_D4VGA_CONTROL);
 		save->crtc_control[2] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET);
 		save->crtc_control[3] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET);
+	}
+	if (rdev->num_crtc >= 6) {
+		save->vga_control[4] = RREG32(EVERGREEN_D5VGA_CONTROL);
+		save->vga_control[5] = RREG32(EVERGREEN_D6VGA_CONTROL);
 		save->crtc_control[4] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET);
 		save->crtc_control[5] = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET);
 	}
@@ -979,35 +1006,45 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav
 	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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1);
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1);
+	}
+	if (rdev->num_crtc >= 6) {
 		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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
 		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+	}
+	if (rdev->num_crtc >= 6) {
 		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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+	}
+	if (rdev->num_crtc >= 6) {
 		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);
-	WREG32(EVERGREEN_D3VGA_CONTROL, 0);
-	WREG32(EVERGREEN_D4VGA_CONTROL, 0);
-	WREG32(EVERGREEN_D5VGA_CONTROL, 0);
-	WREG32(EVERGREEN_D6VGA_CONTROL, 0);
+	if (rdev->num_crtc >= 4) {
+		WREG32(EVERGREEN_D3VGA_CONTROL, 0);
+		WREG32(EVERGREEN_D4VGA_CONTROL, 0);
+	}
+	if (rdev->num_crtc >= 6) {
+		WREG32(EVERGREEN_D5VGA_CONTROL, 0);
+		WREG32(EVERGREEN_D6VGA_CONTROL, 0);
+	}
 }
 
 void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *save)
@@ -1030,7 +1067,7 @@ void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *s
 	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC1_REGISTER_OFFSET,
 	       (u32)rdev->mc.vram_start);
 
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		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,
@@ -1048,7 +1085,8 @@ void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *s
 		       (u32)rdev->mc.vram_start);
 		WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET,
 		       (u32)rdev->mc.vram_start);
-
+	}
+	if (rdev->num_crtc >= 6) {
 		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,
@@ -1076,31 +1114,41 @@ void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *s
 	/* Restore video state */
 	WREG32(D1VGA_CONTROL, save->vga_control[0]);
 	WREG32(D2VGA_CONTROL, save->vga_control[1]);
-	WREG32(EVERGREEN_D3VGA_CONTROL, save->vga_control[2]);
-	WREG32(EVERGREEN_D4VGA_CONTROL, save->vga_control[3]);
-	WREG32(EVERGREEN_D5VGA_CONTROL, save->vga_control[4]);
-	WREG32(EVERGREEN_D6VGA_CONTROL, save->vga_control[5]);
+	if (rdev->num_crtc >= 4) {
+		WREG32(EVERGREEN_D3VGA_CONTROL, save->vga_control[2]);
+		WREG32(EVERGREEN_D4VGA_CONTROL, save->vga_control[3]);
+	}
+	if (rdev->num_crtc >= 6) {
+		WREG32(EVERGREEN_D5VGA_CONTROL, save->vga_control[4]);
+		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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1);
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1);
+	}
+	if (rdev->num_crtc >= 6) {
 		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]);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, save->crtc_control[2]);
 		WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, save->crtc_control[3]);
+	}
+	if (rdev->num_crtc >= 6) {
 		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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+	}
+	if (rdev->num_crtc >= 6) {
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
 		WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
 	}
@@ -1415,6 +1463,8 @@ static u32 evergreen_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	case CHIP_CEDAR:
 	case CHIP_REDWOOD:
 	case CHIP_PALM:
+	case CHIP_SUMO:
+	case CHIP_SUMO2:
 	case CHIP_TURKS:
 	case CHIP_CAICOS:
 		force_no_swizzle = false;
@@ -1544,6 +1594,8 @@ static void evergreen_program_channel_remap(struct radeon_device *rdev)
 	case CHIP_REDWOOD:
 	case CHIP_CEDAR:
 	case CHIP_PALM:
+	case CHIP_SUMO:
+	case CHIP_SUMO2:
 	case CHIP_TURKS:
 	case CHIP_CAICOS:
 	default:
@@ -1689,6 +1741,54 @@ 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_SUMO:
+		rdev->config.evergreen.num_ses = 1;
+		rdev->config.evergreen.max_pipes = 4;
+		rdev->config.evergreen.max_tile_pipes = 2;
+		if (rdev->pdev->device == 0x9648)
+			rdev->config.evergreen.max_simds = 3;
+		else if ((rdev->pdev->device == 0x9647) ||
+			 (rdev->pdev->device == 0x964a))
+			rdev->config.evergreen.max_simds = 4;
+		else
+			rdev->config.evergreen.max_simds = 5;
+		rdev->config.evergreen.max_backends = 2 * rdev->config.evergreen.num_ses;
+		rdev->config.evergreen.max_gprs = 256;
+		rdev->config.evergreen.max_threads = 248;
+		rdev->config.evergreen.max_gs_threads = 32;
+		rdev->config.evergreen.max_stack_entries = 256;
+		rdev->config.evergreen.sx_num_of_sets = 4;
+		rdev->config.evergreen.sx_max_export_size = 256;
+		rdev->config.evergreen.sx_max_export_pos_size = 64;
+		rdev->config.evergreen.sx_max_export_smx_size = 192;
+		rdev->config.evergreen.max_hw_contexts = 8;
+		rdev->config.evergreen.sq_num_cf_insts = 2;
+
+		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;
+	case CHIP_SUMO2:
+		rdev->config.evergreen.num_ses = 1;
+		rdev->config.evergreen.max_pipes = 4;
+		rdev->config.evergreen.max_tile_pipes = 4;
+		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 = 248;
+		rdev->config.evergreen.max_gs_threads = 32;
+		rdev->config.evergreen.max_stack_entries = 512;
+		rdev->config.evergreen.sx_num_of_sets = 4;
+		rdev->config.evergreen.sx_max_export_size = 256;
+		rdev->config.evergreen.sx_max_export_pos_size = 64;
+		rdev->config.evergreen.sx_max_export_smx_size = 192;
+		rdev->config.evergreen.max_hw_contexts = 8;
+		rdev->config.evergreen.sq_num_cf_insts = 2;
+
+		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;
 	case CHIP_BARTS:
 		rdev->config.evergreen.num_ses = 2;
 		rdev->config.evergreen.max_pipes = 4;
@@ -1900,7 +2000,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 			gb_backend_map = 0x66442200;
 			break;
 		case CHIP_JUNIPER:
-			gb_backend_map = 0x00006420;
+			gb_backend_map = 0x00002200;
 			break;
 		default:
 			gb_backend_map =
@@ -1936,9 +2036,9 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 		rdev->config.evergreen.tile_config |= (3 << 0);
 		break;
 	}
-	/* num banks is 8 on all fusion asics */
+	/* num banks is 8 on all fusion asics. 0 = 4, 1 = 8, 2 = 16 */
 	if (rdev->flags & RADEON_IS_IGP)
-		rdev->config.evergreen.tile_config |= 8 << 4;
+		rdev->config.evergreen.tile_config |= 1 << 4;
 	else
 		rdev->config.evergreen.tile_config |=
 			((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) << 4;
@@ -2039,6 +2139,8 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	switch (rdev->family) {
 	case CHIP_CEDAR:
 	case CHIP_PALM:
+	case CHIP_SUMO:
+	case CHIP_SUMO2:
 	case CHIP_CAICOS:
 		/* no vertex cache */
 		sq_config &= ~VC_ENABLE;
@@ -2060,6 +2162,8 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	switch (rdev->family) {
 	case CHIP_CEDAR:
 	case CHIP_PALM:
+	case CHIP_SUMO:
+	case CHIP_SUMO2:
 		ps_thread_count = 96;
 		break;
 	default:
@@ -2099,6 +2203,8 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	switch (rdev->family) {
 	case CHIP_CEDAR:
 	case CHIP_PALM:
+	case CHIP_SUMO:
+	case CHIP_SUMO2:
 	case CHIP_CAICOS:
 		vgt_cache_invalidation = CACHE_INVALIDATION(TC_ONLY);
 		break;
@@ -2165,7 +2271,10 @@ int evergreen_mc_init(struct radeon_device *rdev)
 
 	/* Get VRAM informations */
 	rdev->mc.vram_is_ddr = true;
-	tmp = RREG32(MC_ARB_RAMCFG);
+	if (rdev->flags & RADEON_IS_IGP)
+		tmp = RREG32(FUS_MC_ARB_RAMCFG);
+	else
+		tmp = RREG32(MC_ARB_RAMCFG);
 	if (tmp & CHANSIZE_OVERRIDE) {
 		chansize = 16;
 	} else if (tmp & CHANSIZE_MASK) {
@@ -2331,18 +2440,22 @@ 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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(INT_MASK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
 		WREG32(INT_MASK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+	}
+	if (rdev->num_crtc >= 6) {
 		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);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0);
 		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0);
+	}
+	if (rdev->num_crtc >= 6) {
 		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
 		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
 	}
@@ -2461,19 +2574,25 @@ int evergreen_irq_set(struct radeon_device *rdev)
 
 	WREG32(INT_MASK + EVERGREEN_CRTC0_REGISTER_OFFSET, crtc1);
 	WREG32(INT_MASK + EVERGREEN_CRTC1_REGISTER_OFFSET, crtc2);
-	if (!(rdev->flags & RADEON_IS_IGP)) {
+	if (rdev->num_crtc >= 4) {
 		WREG32(INT_MASK + EVERGREEN_CRTC2_REGISTER_OFFSET, crtc3);
 		WREG32(INT_MASK + EVERGREEN_CRTC3_REGISTER_OFFSET, crtc4);
+	}
+	if (rdev->num_crtc >= 6) {
 		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);
+	if (rdev->num_crtc >= 4) {
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, grph3);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, grph4);
+	}
+	if (rdev->num_crtc >= 6) {
+		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);
@@ -2497,53 +2616,57 @@ static inline void evergreen_irq_ack(struct radeon_device *rdev)
 	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->num_crtc >= 4) {
+		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);
+	}
+	if (rdev->num_crtc >= 6) {
+		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 (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET, VLINE_ACK);
-
 	if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT)
 		WREG32(VBLANK_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET, VBLANK_ACK);
 	if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT)
 		WREG32(VLINE_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET, VLINE_ACK);
 
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT)
-		WREG32(VBLANK_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VBLANK_ACK);
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT)
-		WREG32(VLINE_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VLINE_ACK);
-
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT)
-		WREG32(VBLANK_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, VBLANK_ACK);
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT)
-		WREG32(VLINE_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, VLINE_ACK);
-
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT)
-		WREG32(VBLANK_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VBLANK_ACK);
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT)
-		WREG32(VLINE_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VLINE_ACK);
+	if (rdev->num_crtc >= 4) {
+		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.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT)
+			WREG32(VBLANK_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VBLANK_ACK);
+		if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT)
+			WREG32(VLINE_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VLINE_ACK);
+		if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT)
+			WREG32(VBLANK_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, VBLANK_ACK);
+		if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT)
+			WREG32(VLINE_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET, VLINE_ACK);
+	}
 
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT)
-		WREG32(VBLANK_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, VBLANK_ACK);
-	if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT)
-		WREG32(VLINE_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, VLINE_ACK);
+	if (rdev->num_crtc >= 6) {
+		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_cont4 & LB_D5_VBLANK_INTERRUPT)
+			WREG32(VBLANK_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VBLANK_ACK);
+		if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT)
+			WREG32(VLINE_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VLINE_ACK);
+		if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT)
+			WREG32(VBLANK_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, VBLANK_ACK);
+		if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT)
+			WREG32(VLINE_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET, VLINE_ACK);
+	}
 
 	if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT) {
 		tmp = RREG32(DC_HPD1_INT_CONTROL);
@@ -2618,28 +2741,25 @@ static inline u32 evergreen_get_ih_wptr(struct radeon_device *rdev)
 
 int evergreen_irq_process(struct radeon_device *rdev)
 {
-	u32 wptr = evergreen_get_ih_wptr(rdev);
-	u32 rptr = rdev->ih.rptr;
+	u32 wptr;
+	u32 rptr;
 	u32 src_id, src_data;
 	u32 ring_index;
 	unsigned long flags;
 	bool queue_hotplug = false;
 
-	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
-	if (!rdev->ih.enabled)
+	if (!rdev->ih.enabled || rdev->shutdown)
 		return IRQ_NONE;
 
-	spin_lock_irqsave(&rdev->ih.lock, flags);
+	wptr = evergreen_get_ih_wptr(rdev);
+	rptr = rdev->ih.rptr;
+	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
 
+	spin_lock_irqsave(&rdev->ih.lock, flags);
 	if (rptr == wptr) {
 		spin_unlock_irqrestore(&rdev->ih.lock, flags);
 		return IRQ_NONE;
 	}
-	if (rdev->shutdown) {
-		spin_unlock_irqrestore(&rdev->ih.lock, flags);
-		return IRQ_NONE;
-	}
-
 restart_ih:
 	/* display interrupts */
 	evergreen_irq_ack(rdev);
@@ -2868,7 +2988,7 @@ restart_ih:
 			radeon_fence_process(rdev);
 			break;
 		case 233: /* GUI IDLE */
-			DRM_DEBUG("IH: CP EOP\n");
+			DRM_DEBUG("IH: GUI idle\n");
 			rdev->pm.gui_idle = true;
 			wake_up(&rdev->irq.idle_queue);
 			break;
@@ -3154,6 +3274,7 @@ void evergreen_fini(struct radeon_device *rdev)
 	r700_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
+	radeon_ib_pool_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	evergreen_pcie_gart_fini(rdev);
 	radeon_gem_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
index ba06a69c6de8..2eb251858e72 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
@@ -31,6 +31,7 @@
 
 #include "evergreend.h"
 #include "evergreen_blit_shaders.h"
+#include "cayman_blit_shaders.h"
 
 #define DI_PT_RECTLIST        0x11
 #define DI_INDEX_SIZE_16_BIT  0x0
@@ -152,6 +153,8 @@ set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
 
 	if ((rdev->family == CHIP_CEDAR) ||
 	    (rdev->family == CHIP_PALM) ||
+	    (rdev->family == CHIP_SUMO) ||
+	    (rdev->family == CHIP_SUMO2) ||
 	    (rdev->family == CHIP_CAICOS))
 		cp_set_surface_sync(rdev,
 				    PACKET3_TC_ACTION_ENA, 48, gpu_addr);
@@ -199,6 +202,16 @@ static void
 set_scissors(struct radeon_device *rdev, int x1, int y1,
 	     int x2, int y2)
 {
+	/* workaround some hw bugs */
+	if (x2 == 0)
+		x1 = 1;
+	if (y2 == 0)
+		y1 = 1;
+	if (rdev->family == CHIP_CAYMAN) {
+		if ((x2 == 1) && (y2 == 1))
+			x2 = 2;
+	}
+
 	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
 	radeon_ring_write(rdev, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
 	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16));
@@ -239,7 +252,7 @@ draw_auto(struct radeon_device *rdev)
 
 }
 
-/* emits 36 */
+/* emits 39 */
 static void
 set_default_state(struct radeon_device *rdev)
 {
@@ -255,238 +268,289 @@ set_default_state(struct radeon_device *rdev)
 	u64 gpu_addr;
 	int dwords;
 
-	switch (rdev->family) {
-	case CHIP_CEDAR:
-	default:
-		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;
-	case CHIP_REDWOOD:
-		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 = 128;
-		num_vs_threads = 20;
-		num_gs_threads = 20;
-		num_es_threads = 20;
-		num_hs_threads = 20;
-		num_ls_threads = 20;
-		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;
-	case CHIP_JUNIPER:
-		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 = 128;
-		num_vs_threads = 20;
-		num_gs_threads = 20;
-		num_es_threads = 20;
-		num_hs_threads = 20;
-		num_ls_threads = 20;
-		num_ps_stack_entries = 85;
-		num_vs_stack_entries = 85;
-		num_gs_stack_entries = 85;
-		num_es_stack_entries = 85;
-		num_hs_stack_entries = 85;
-		num_ls_stack_entries = 85;
-		break;
-	case CHIP_CYPRESS:
-	case CHIP_HEMLOCK:
-		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 = 128;
-		num_vs_threads = 20;
-		num_gs_threads = 20;
-		num_es_threads = 20;
-		num_hs_threads = 20;
-		num_ls_threads = 20;
-		num_ps_stack_entries = 85;
-		num_vs_stack_entries = 85;
-		num_gs_stack_entries = 85;
-		num_es_stack_entries = 85;
-		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;
-	case CHIP_BARTS:
-		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 = 128;
-		num_vs_threads = 20;
-		num_gs_threads = 20;
-		num_es_threads = 20;
-		num_hs_threads = 20;
-		num_ls_threads = 20;
-		num_ps_stack_entries = 85;
-		num_vs_stack_entries = 85;
-		num_gs_stack_entries = 85;
-		num_es_stack_entries = 85;
-		num_hs_stack_entries = 85;
-		num_ls_stack_entries = 85;
-		break;
-	case CHIP_TURKS:
-		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 = 128;
-		num_vs_threads = 20;
-		num_gs_threads = 20;
-		num_es_threads = 20;
-		num_hs_threads = 20;
-		num_ls_threads = 20;
-		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;
-	case CHIP_CAICOS:
-		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 = 128;
-		num_vs_threads = 10;
-		num_gs_threads = 10;
-		num_es_threads = 10;
-		num_hs_threads = 10;
-		num_ls_threads = 10;
-		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;
-	}
+	/* set clear context state */
+	radeon_ring_write(rdev, PACKET3(PACKET3_CLEAR_STATE, 0));
+	radeon_ring_write(rdev, 0);
 
-	if ((rdev->family == CHIP_CEDAR) ||
-	    (rdev->family == CHIP_PALM) ||
-	    (rdev->family == CHIP_CAICOS))
-		sq_config = 0;
-	else
-		sq_config = VC_ENABLE;
+	if (rdev->family < CHIP_CAYMAN) {
+		switch (rdev->family) {
+		case CHIP_CEDAR:
+		default:
+			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;
+		case CHIP_REDWOOD:
+			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 = 128;
+			num_vs_threads = 20;
+			num_gs_threads = 20;
+			num_es_threads = 20;
+			num_hs_threads = 20;
+			num_ls_threads = 20;
+			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;
+		case CHIP_JUNIPER:
+			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 = 128;
+			num_vs_threads = 20;
+			num_gs_threads = 20;
+			num_es_threads = 20;
+			num_hs_threads = 20;
+			num_ls_threads = 20;
+			num_ps_stack_entries = 85;
+			num_vs_stack_entries = 85;
+			num_gs_stack_entries = 85;
+			num_es_stack_entries = 85;
+			num_hs_stack_entries = 85;
+			num_ls_stack_entries = 85;
+			break;
+		case CHIP_CYPRESS:
+		case CHIP_HEMLOCK:
+			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 = 128;
+			num_vs_threads = 20;
+			num_gs_threads = 20;
+			num_es_threads = 20;
+			num_hs_threads = 20;
+			num_ls_threads = 20;
+			num_ps_stack_entries = 85;
+			num_vs_stack_entries = 85;
+			num_gs_stack_entries = 85;
+			num_es_stack_entries = 85;
+			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;
+		case CHIP_SUMO:
+			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 = 25;
+			num_gs_threads = 25;
+			num_es_threads = 25;
+			num_hs_threads = 25;
+			num_ls_threads = 25;
+			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;
+		case CHIP_SUMO2:
+			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 = 25;
+			num_gs_threads = 25;
+			num_es_threads = 25;
+			num_hs_threads = 25;
+			num_ls_threads = 25;
+			num_ps_stack_entries = 85;
+			num_vs_stack_entries = 85;
+			num_gs_stack_entries = 85;
+			num_es_stack_entries = 85;
+			num_hs_stack_entries = 85;
+			num_ls_stack_entries = 85;
+			break;
+		case CHIP_BARTS:
+			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 = 128;
+			num_vs_threads = 20;
+			num_gs_threads = 20;
+			num_es_threads = 20;
+			num_hs_threads = 20;
+			num_ls_threads = 20;
+			num_ps_stack_entries = 85;
+			num_vs_stack_entries = 85;
+			num_gs_stack_entries = 85;
+			num_es_stack_entries = 85;
+			num_hs_stack_entries = 85;
+			num_ls_stack_entries = 85;
+			break;
+		case CHIP_TURKS:
+			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 = 128;
+			num_vs_threads = 20;
+			num_gs_threads = 20;
+			num_es_threads = 20;
+			num_hs_threads = 20;
+			num_ls_threads = 20;
+			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;
+		case CHIP_CAICOS:
+			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 = 128;
+			num_vs_threads = 10;
+			num_gs_threads = 10;
+			num_es_threads = 10;
+			num_hs_threads = 10;
+			num_ls_threads = 10;
+			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;
+		}
 
-	sq_config |= (EXPORT_SRC_C |
-		      CS_PRIO(0) |
-		      LS_PRIO(0) |
-		      HS_PRIO(0) |
-		      PS_PRIO(0) |
-		      VS_PRIO(1) |
-		      GS_PRIO(2) |
-		      ES_PRIO(3));
+		if ((rdev->family == CHIP_CEDAR) ||
+		    (rdev->family == CHIP_PALM) ||
+		    (rdev->family == CHIP_SUMO) ||
+		    (rdev->family == CHIP_SUMO2) ||
+		    (rdev->family == CHIP_CAICOS))
+			sq_config = 0;
+		else
+			sq_config = VC_ENABLE;
 
-	sq_gpr_resource_mgmt_1 = (NUM_PS_GPRS(num_ps_gprs) |
-				  NUM_VS_GPRS(num_vs_gprs) |
-				  NUM_CLAUSE_TEMP_GPRS(num_temp_gprs));
-	sq_gpr_resource_mgmt_2 = (NUM_GS_GPRS(num_gs_gprs) |
-				  NUM_ES_GPRS(num_es_gprs));
-	sq_gpr_resource_mgmt_3 = (NUM_HS_GPRS(num_hs_gprs) |
-				  NUM_LS_GPRS(num_ls_gprs));
-	sq_thread_resource_mgmt = (NUM_PS_THREADS(num_ps_threads) |
-				   NUM_VS_THREADS(num_vs_threads) |
-				   NUM_GS_THREADS(num_gs_threads) |
-				   NUM_ES_THREADS(num_es_threads));
-	sq_thread_resource_mgmt_2 = (NUM_HS_THREADS(num_hs_threads) |
-				     NUM_LS_THREADS(num_ls_threads));
-	sq_stack_resource_mgmt_1 = (NUM_PS_STACK_ENTRIES(num_ps_stack_entries) |
-				    NUM_VS_STACK_ENTRIES(num_vs_stack_entries));
-	sq_stack_resource_mgmt_2 = (NUM_GS_STACK_ENTRIES(num_gs_stack_entries) |
-				    NUM_ES_STACK_ENTRIES(num_es_stack_entries));
-	sq_stack_resource_mgmt_3 = (NUM_HS_STACK_ENTRIES(num_hs_stack_entries) |
-				    NUM_LS_STACK_ENTRIES(num_ls_stack_entries));
+		sq_config |= (EXPORT_SRC_C |
+			      CS_PRIO(0) |
+			      LS_PRIO(0) |
+			      HS_PRIO(0) |
+			      PS_PRIO(0) |
+			      VS_PRIO(1) |
+			      GS_PRIO(2) |
+			      ES_PRIO(3));
 
-	/* set clear context state */
-	radeon_ring_write(rdev, PACKET3(PACKET3_CLEAR_STATE, 0));
-	radeon_ring_write(rdev, 0);
+		sq_gpr_resource_mgmt_1 = (NUM_PS_GPRS(num_ps_gprs) |
+					  NUM_VS_GPRS(num_vs_gprs) |
+					  NUM_CLAUSE_TEMP_GPRS(num_temp_gprs));
+		sq_gpr_resource_mgmt_2 = (NUM_GS_GPRS(num_gs_gprs) |
+					  NUM_ES_GPRS(num_es_gprs));
+		sq_gpr_resource_mgmt_3 = (NUM_HS_GPRS(num_hs_gprs) |
+					  NUM_LS_GPRS(num_ls_gprs));
+		sq_thread_resource_mgmt = (NUM_PS_THREADS(num_ps_threads) |
+					   NUM_VS_THREADS(num_vs_threads) |
+					   NUM_GS_THREADS(num_gs_threads) |
+					   NUM_ES_THREADS(num_es_threads));
+		sq_thread_resource_mgmt_2 = (NUM_HS_THREADS(num_hs_threads) |
+					     NUM_LS_THREADS(num_ls_threads));
+		sq_stack_resource_mgmt_1 = (NUM_PS_STACK_ENTRIES(num_ps_stack_entries) |
+					    NUM_VS_STACK_ENTRIES(num_vs_stack_entries));
+		sq_stack_resource_mgmt_2 = (NUM_GS_STACK_ENTRIES(num_gs_stack_entries) |
+					    NUM_ES_STACK_ENTRIES(num_es_stack_entries));
+		sq_stack_resource_mgmt_3 = (NUM_HS_STACK_ENTRIES(num_hs_stack_entries) |
+					    NUM_LS_STACK_ENTRIES(num_ls_stack_entries));
 
-	/* disable dyn gprs */
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-	radeon_ring_write(rdev, (SQ_DYN_GPR_CNTL_PS_FLUSH_REQ - PACKET3_SET_CONFIG_REG_START) >> 2);
-	radeon_ring_write(rdev, 0);
+		/* disable dyn gprs */
+		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(rdev, (SQ_DYN_GPR_CNTL_PS_FLUSH_REQ - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(rdev, 0);
 
-	/* SQ config */
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 11));
-	radeon_ring_write(rdev, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_START) >> 2);
-	radeon_ring_write(rdev, sq_config);
-	radeon_ring_write(rdev, sq_gpr_resource_mgmt_1);
-	radeon_ring_write(rdev, sq_gpr_resource_mgmt_2);
-	radeon_ring_write(rdev, sq_gpr_resource_mgmt_3);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, sq_thread_resource_mgmt);
-	radeon_ring_write(rdev, sq_thread_resource_mgmt_2);
-	radeon_ring_write(rdev, sq_stack_resource_mgmt_1);
-	radeon_ring_write(rdev, sq_stack_resource_mgmt_2);
-	radeon_ring_write(rdev, sq_stack_resource_mgmt_3);
+		/* setup LDS */
+		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(rdev, (SQ_LDS_RESOURCE_MGMT - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(rdev, 0x10001000);
+
+		/* SQ config */
+		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 11));
+		radeon_ring_write(rdev, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(rdev, sq_config);
+		radeon_ring_write(rdev, sq_gpr_resource_mgmt_1);
+		radeon_ring_write(rdev, sq_gpr_resource_mgmt_2);
+		radeon_ring_write(rdev, sq_gpr_resource_mgmt_3);
+		radeon_ring_write(rdev, 0);
+		radeon_ring_write(rdev, 0);
+		radeon_ring_write(rdev, sq_thread_resource_mgmt);
+		radeon_ring_write(rdev, sq_thread_resource_mgmt_2);
+		radeon_ring_write(rdev, sq_stack_resource_mgmt_1);
+		radeon_ring_write(rdev, sq_stack_resource_mgmt_2);
+		radeon_ring_write(rdev, sq_stack_resource_mgmt_3);
+	}
 
 	/* CONTEXT_CONTROL */
 	radeon_ring_write(rdev, 0xc0012800);
@@ -560,7 +624,10 @@ int evergreen_blit_init(struct radeon_device *rdev)
 	mutex_init(&rdev->r600_blit.mutex);
 	rdev->r600_blit.state_offset = 0;
 
-	rdev->r600_blit.state_len = evergreen_default_size;
+	if (rdev->family < CHIP_CAYMAN)
+		rdev->r600_blit.state_len = evergreen_default_size;
+	else
+		rdev->r600_blit.state_len = cayman_default_size;
 
 	dwords = rdev->r600_blit.state_len;
 	while (dwords & 0xf) {
@@ -572,11 +639,17 @@ int evergreen_blit_init(struct radeon_device *rdev)
 	obj_size = ALIGN(obj_size, 256);
 
 	rdev->r600_blit.vs_offset = obj_size;
-	obj_size += evergreen_vs_size * 4;
+	if (rdev->family < CHIP_CAYMAN)
+		obj_size += evergreen_vs_size * 4;
+	else
+		obj_size += cayman_vs_size * 4;
 	obj_size = ALIGN(obj_size, 256);
 
 	rdev->r600_blit.ps_offset = obj_size;
-	obj_size += evergreen_ps_size * 4;
+	if (rdev->family < CHIP_CAYMAN)
+		obj_size += evergreen_ps_size * 4;
+	else
+		obj_size += cayman_ps_size * 4;
 	obj_size = ALIGN(obj_size, 256);
 
 	r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true, RADEON_GEM_DOMAIN_VRAM,
@@ -599,16 +672,29 @@ int evergreen_blit_init(struct radeon_device *rdev)
 		return r;
 	}
 
-	memcpy_toio(ptr + rdev->r600_blit.state_offset,
-		    evergreen_default_state, rdev->r600_blit.state_len * 4);
+	if (rdev->family < CHIP_CAYMAN) {
+		memcpy_toio(ptr + rdev->r600_blit.state_offset,
+			    evergreen_default_state, rdev->r600_blit.state_len * 4);
 
-	if (num_packet2s)
-		memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
-			    packet2s, num_packet2s * 4);
-	for (i = 0; i < evergreen_vs_size; i++)
-		*(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(evergreen_vs[i]);
-	for (i = 0; i < evergreen_ps_size; i++)
-		*(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(evergreen_ps[i]);
+		if (num_packet2s)
+			memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
+				    packet2s, num_packet2s * 4);
+		for (i = 0; i < evergreen_vs_size; i++)
+			*(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(evergreen_vs[i]);
+		for (i = 0; i < evergreen_ps_size; i++)
+			*(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(evergreen_ps[i]);
+	} else {
+		memcpy_toio(ptr + rdev->r600_blit.state_offset,
+			    cayman_default_state, rdev->r600_blit.state_len * 4);
+
+		if (num_packet2s)
+			memcpy_toio(ptr + rdev->r600_blit.state_offset + (rdev->r600_blit.state_len * 4),
+				    packet2s, num_packet2s * 4);
+		for (i = 0; i < cayman_vs_size; i++)
+			*(u32 *)((unsigned long)ptr + rdev->r600_blit.vs_offset + i * 4) = cpu_to_le32(cayman_vs[i]);
+		for (i = 0; i < cayman_ps_size; i++)
+			*(u32 *)((unsigned long)ptr + rdev->r600_blit.ps_offset + i * 4) = cpu_to_le32(cayman_ps[i]);
+	}
 	radeon_bo_kunmap(rdev->r600_blit.shader_obj);
 	radeon_bo_unreserve(rdev->r600_blit.shader_obj);
 
@@ -692,7 +778,7 @@ int evergreen_blit_prepare_copy(struct radeon_device *rdev, int size_bytes)
 	/* calculate number of loops correctly */
 	ring_size = num_loops * dwords_per_loop;
 	/* set default  + shaders */
-	ring_size += 52; /* shaders + def state */
+	ring_size += 55; /* shaders + def state */
 	ring_size += 10; /* fence emit for VB IB */
 	ring_size += 5; /* done copy */
 	ring_size += 10; /* fence emit for done copy */
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index f37e91ee8a11..b7b2714f0b32 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -168,10 +168,16 @@
 #define		SE_DB_BUSY					(1 << 30)
 #define		SE_CB_BUSY					(1 << 31)
 /* evergreen */
+#define	CG_THERMAL_CTRL					0x72c
+#define		TOFFSET_MASK			        0x00003FE0
+#define		TOFFSET_SHIFT			        5
 #define	CG_MULT_THERMAL_STATUS				0x740
 #define		ASIC_T(x)			        ((x) << 16)
-#define		ASIC_T_MASK			        0x7FF0000
+#define		ASIC_T_MASK			        0x07FF0000
 #define		ASIC_T_SHIFT			        16
+#define	CG_TS0_STATUS					0x760
+#define		TS0_ADC_DOUT_MASK			0x000003FF
+#define		TS0_ADC_DOUT_SHIFT			0
 /* APU */
 #define	CG_THERMAL_STATUS			        0x678
 
@@ -460,7 +466,7 @@
 #define IH_RB_WPTR_ADDR_LO                                0x3e14
 #define IH_CNTL                                           0x3e18
 #       define ENABLE_INTR                                (1 << 0)
-#       define IH_MC_SWAP(x)                              ((x) << 2)
+#       define IH_MC_SWAP(x)                              ((x) << 1)
 #       define IH_MC_SWAP_NONE                            0
 #       define IH_MC_SWAP_16BIT                           1
 #       define IH_MC_SWAP_32BIT                           2
@@ -541,7 +547,7 @@
 #       define LB_D5_VBLANK_INTERRUPT                   (1 << 3)
 #       define DC_HPD5_INTERRUPT                        (1 << 17)
 #       define DC_HPD5_RX_INTERRUPT                     (1 << 18)
-#define DISP_INTERRUPT_STATUS_CONTINUE5                 0x6050
+#define DISP_INTERRUPT_STATUS_CONTINUE5                 0x6150
 #       define LB_D6_VLINE_INTERRUPT                    (1 << 2)
 #       define LB_D6_VBLANK_INTERRUPT                   (1 << 3)
 #       define DC_HPD6_INTERRUPT                        (1 << 17)
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index b205ba1cdd8f..559dbd412906 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1387,14 +1387,12 @@ static int cayman_startup(struct radeon_device *rdev)
 		return r;
 	cayman_gpu_init(rdev);
 
-#if 0
-	r = cayman_blit_init(rdev);
+	r = evergreen_blit_init(rdev);
 	if (r) {
-		cayman_blit_fini(rdev);
+		evergreen_blit_fini(rdev);
 		rdev->asic->copy = NULL;
 		dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r);
 	}
-#endif
 
 	/* allocate wb buffer */
 	r = radeon_wb_init(rdev);
@@ -1452,7 +1450,7 @@ int cayman_resume(struct radeon_device *rdev)
 
 int cayman_suspend(struct radeon_device *rdev)
 {
-	/* int r; */
+	int r;
 
 	/* FIXME: we should wait for ring to be empty */
 	cayman_cp_enable(rdev, false);
@@ -1461,14 +1459,13 @@ int cayman_suspend(struct radeon_device *rdev)
 	radeon_wb_disable(rdev);
 	cayman_pcie_gart_disable(rdev);
 
-#if 0
 	/* unpin shaders bo */
 	r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
 	if (likely(r == 0)) {
 		radeon_bo_unpin(rdev->r600_blit.shader_obj);
 		radeon_bo_unreserve(rdev->r600_blit.shader_obj);
 	}
-#endif
+
 	return 0;
 }
 
@@ -1580,10 +1577,11 @@ int cayman_init(struct radeon_device *rdev)
 
 void cayman_fini(struct radeon_device *rdev)
 {
-	/* cayman_blit_fini(rdev); */
+	evergreen_blit_fini(rdev);
 	cayman_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
+	radeon_ib_pool_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	cayman_pcie_gart_fini(rdev);
 	radeon_gem_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index 9736746da2d6..4672869cdb26 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -320,7 +320,7 @@
 #define	CGTS_USER_TCC_DISABLE				0x914C
 #define		TCC_DISABLE_MASK				0xFFFF0000
 #define		TCC_DISABLE_SHIFT				16
-#define	CGTS_SM_CTRL_REG				0x915C
+#define	CGTS_SM_CTRL_REG				0x9150
 #define		OVERRIDE				(1 << 21)
 
 #define	TA_CNTL_AUX					0x9508
diff --git a/drivers/gpu/drm/radeon/r100_track.h b/drivers/gpu/drm/radeon/r100_track.h
index 2fef9de7f363..686f9dc5d4bd 100644
--- a/drivers/gpu/drm/radeon/r100_track.h
+++ b/drivers/gpu/drm/radeon/r100_track.h
@@ -63,7 +63,7 @@ struct r100_cs_track {
 	unsigned			num_arrays;
 	unsigned			max_indx;
 	unsigned			color_channel_mask;
-	struct r100_cs_track_array	arrays[11];
+	struct r100_cs_track_array	arrays[16];
 	struct r100_cs_track_cb 	cb[R300_MAX_CB];
 	struct r100_cs_track_cb 	zb;
 	struct r100_cs_track_cb 	aa;
@@ -146,6 +146,12 @@ static inline int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 	ib = p->ib->ptr;
 	track = (struct r100_cs_track *)p->track;
 	c = radeon_get_ib_value(p, idx++) & 0x1F;
+	if (c > 16) {
+	    DRM_ERROR("Only 16 vertex buffers are allowed %d\n",
+		      pkt->opcode);
+	    r100_cs_dump_packet(p, pkt);
+	    return -EINVAL;
+	}
 	track->num_arrays = c;
 	for (i = 0; i < (c - 1); i+=2, idx+=3) {
 		r = r100_cs_packet_next_reloc(p, &reloc);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 6f27593901c7..bc54b26cb32f 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -87,6 +87,10 @@ MODULE_FIRMWARE("radeon/CYPRESS_rlc.bin");
 MODULE_FIRMWARE("radeon/PALM_pfp.bin");
 MODULE_FIRMWARE("radeon/PALM_me.bin");
 MODULE_FIRMWARE("radeon/SUMO_rlc.bin");
+MODULE_FIRMWARE("radeon/SUMO_pfp.bin");
+MODULE_FIRMWARE("radeon/SUMO_me.bin");
+MODULE_FIRMWARE("radeon/SUMO2_pfp.bin");
+MODULE_FIRMWARE("radeon/SUMO2_me.bin");
 
 int r600_debugfs_mc_info_init(struct radeon_device *rdev);
 
@@ -586,6 +590,9 @@ void r600_pm_misc(struct radeon_device *rdev)
 	struct radeon_voltage *voltage = &ps->clock_info[req_cm_idx].voltage;
 
 	if ((voltage->type == VOLTAGE_SW) && voltage->voltage) {
+		/* 0xff01 is a flag rather then an actual voltage */
+		if (voltage->voltage == 0xff01)
+			return;
 		if (voltage->voltage != rdev->pm.current_vddc) {
 			radeon_atom_set_voltage(rdev, voltage->voltage, SET_VOLTAGE_TYPE_ASIC_VDDC);
 			rdev->pm.current_vddc = voltage->voltage;
@@ -2024,6 +2031,14 @@ int r600_init_microcode(struct radeon_device *rdev)
 		chip_name = "PALM";
 		rlc_chip_name = "SUMO";
 		break;
+	case CHIP_SUMO:
+		chip_name = "SUMO";
+		rlc_chip_name = "SUMO";
+		break;
+	case CHIP_SUMO2:
+		chip_name = "SUMO2";
+		rlc_chip_name = "SUMO";
+		break;
 	default: BUG();
 	}
 
@@ -2613,6 +2628,7 @@ void r600_fini(struct radeon_device *rdev)
 	r600_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
+	radeon_ib_pool_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	r600_pcie_gart_fini(rdev);
 	radeon_agp_fini(rdev);
@@ -3282,27 +3298,26 @@ static inline u32 r600_get_ih_wptr(struct radeon_device *rdev)
 
 int r600_irq_process(struct radeon_device *rdev)
 {
-	u32 wptr = r600_get_ih_wptr(rdev);
-	u32 rptr = rdev->ih.rptr;
+	u32 wptr;
+	u32 rptr;
 	u32 src_id, src_data;
 	u32 ring_index;
 	unsigned long flags;
 	bool queue_hotplug = false;
 
-	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
-	if (!rdev->ih.enabled)
+	if (!rdev->ih.enabled || rdev->shutdown)
 		return IRQ_NONE;
 
+	wptr = r600_get_ih_wptr(rdev);
+	rptr = rdev->ih.rptr;
+	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
+
 	spin_lock_irqsave(&rdev->ih.lock, flags);
 
 	if (rptr == wptr) {
 		spin_unlock_irqrestore(&rdev->ih.lock, flags);
 		return IRQ_NONE;
 	}
-	if (rdev->shutdown) {
-		spin_unlock_irqrestore(&rdev->ih.lock, flags);
-		return IRQ_NONE;
-	}
 
 restart_ih:
 	/* display interrupts */
@@ -3432,7 +3447,7 @@ restart_ih:
 			radeon_fence_process(rdev);
 			break;
 		case 233: /* GUI IDLE */
-			DRM_DEBUG("IH: CP EOP\n");
+			DRM_DEBUG("IH: GUI idle\n");
 			rdev->pm.gui_idle = true;
 			wake_up(&rdev->irq.idle_queue);
 			break;
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index fd18be9871ab..909bda8dd550 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -71,20 +71,21 @@ struct r600_cs_track {
 	u64			db_bo_mc;
 };
 
-#define FMT_8_BIT(fmt, vc) [fmt] = { 1, 1, 1, vc }
-#define FMT_16_BIT(fmt, vc) [fmt] = { 1, 1, 2, vc }
-#define FMT_24_BIT(fmt) [fmt] = { 1, 1, 3, 0 }
-#define FMT_32_BIT(fmt, vc) [fmt] = { 1, 1, 4, vc }
-#define FMT_48_BIT(fmt) [fmt] = { 1, 1, 6, 0 }
-#define FMT_64_BIT(fmt, vc) [fmt] = { 1, 1, 8, vc }
-#define FMT_96_BIT(fmt) [fmt] = { 1, 1, 12, 0 }
-#define FMT_128_BIT(fmt, vc) [fmt] = { 1, 1, 16, vc }
+#define FMT_8_BIT(fmt, vc)   [fmt] = { 1, 1, 1, vc, CHIP_R600 }
+#define FMT_16_BIT(fmt, vc)  [fmt] = { 1, 1, 2, vc, CHIP_R600 }
+#define FMT_24_BIT(fmt)      [fmt] = { 1, 1, 3,  0, CHIP_R600 }
+#define FMT_32_BIT(fmt, vc)  [fmt] = { 1, 1, 4, vc, CHIP_R600 }
+#define FMT_48_BIT(fmt)      [fmt] = { 1, 1, 6,  0, CHIP_R600 }
+#define FMT_64_BIT(fmt, vc)  [fmt] = { 1, 1, 8, vc, CHIP_R600 }
+#define FMT_96_BIT(fmt)      [fmt] = { 1, 1, 12, 0, CHIP_R600 }
+#define FMT_128_BIT(fmt, vc) [fmt] = { 1, 1, 16,vc, CHIP_R600 }
 
 struct gpu_formats {
 	unsigned blockwidth;
 	unsigned blockheight;
 	unsigned blocksize;
 	unsigned valid_color;
+	enum radeon_family min_family;
 };
 
 static const struct gpu_formats color_formats_table[] = {
@@ -154,7 +155,11 @@ static const struct gpu_formats color_formats_table[] = {
 	[V_038004_FMT_BC3] = { 4, 4, 16, 0 },
 	[V_038004_FMT_BC4] = { 4, 4, 8, 0 },
 	[V_038004_FMT_BC5] = { 4, 4, 16, 0},
+	[V_038004_FMT_BC6] = { 4, 4, 16, 0, CHIP_CEDAR}, /* Evergreen-only */
+	[V_038004_FMT_BC7] = { 4, 4, 16, 0, CHIP_CEDAR}, /* Evergreen-only */
 
+	/* The other Evergreen formats */
+	[V_038004_FMT_32_AS_32_32_32_32] = { 1, 1, 4, 0, CHIP_CEDAR},
 };
 
 static inline bool fmt_is_valid_color(u32 format)
@@ -168,11 +173,14 @@ static inline bool fmt_is_valid_color(u32 format)
 	return false;
 }
 
-static inline bool fmt_is_valid_texture(u32 format)
+static inline bool fmt_is_valid_texture(u32 format, enum radeon_family family)
 {
 	if (format >= ARRAY_SIZE(color_formats_table))
 		return false;
 	
+	if (family < color_formats_table[format].min_family)
+		return false;
+
 	if (color_formats_table[format].blockwidth > 0)
 		return true;
 
@@ -1325,7 +1333,7 @@ static inline int r600_check_texture_resource(struct radeon_cs_parser *p,  u32 i
 		return -EINVAL;
 	}
 	format = G_038004_DATA_FORMAT(word1);
-	if (!fmt_is_valid_texture(format)) {
+	if (!fmt_is_valid_texture(format, p->family)) {
 		dev_warn(p->dev, "%s:%d texture invalid format %d\n",
 			 __func__, __LINE__, format);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index b2b944bcd05a..0245ae6c204e 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -536,7 +536,7 @@
 #define IH_RB_WPTR_ADDR_LO                                0x3e14
 #define IH_CNTL                                           0x3e18
 #       define ENABLE_INTR                                (1 << 0)
-#       define IH_MC_SWAP(x)                              ((x) << 2)
+#       define IH_MC_SWAP(x)                              ((x) << 1)
 #       define IH_MC_SWAP_NONE                            0
 #       define IH_MC_SWAP_16BIT                           1
 #       define IH_MC_SWAP_32BIT                           2
@@ -1309,6 +1309,9 @@
 #define     V_038004_FMT_BC3                           0x00000033
 #define     V_038004_FMT_BC4                           0x00000034
 #define     V_038004_FMT_BC5                           0x00000035
+#define     V_038004_FMT_BC6                           0x00000036
+#define     V_038004_FMT_BC7                           0x00000037
+#define     V_038004_FMT_32_AS_32_32_32_32             0x00000038
 #define R_038010_SQ_TEX_RESOURCE_WORD4_0             0x038010
 #define   S_038010_FORMAT_COMP_X(x)                    (((x) & 0x3) << 0)
 #define   G_038010_FORMAT_COMP_X(x)                    (((x) >> 0) & 0x3)
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index ba643b576054..ef0e0e016914 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -165,6 +165,7 @@ struct radeon_clock {
 	uint32_t default_sclk;
 	uint32_t default_dispclk;
 	uint32_t dp_extclk;
+	uint32_t max_pixel_clock;
 };
 
 /*
@@ -178,6 +179,7 @@ void radeon_pm_resume(struct radeon_device *rdev);
 void radeon_combios_get_power_modes(struct radeon_device *rdev);
 void radeon_atombios_get_power_modes(struct radeon_device *rdev);
 void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 voltage_type);
+int radeon_atom_get_max_vddc(struct radeon_device *rdev, u16 *voltage);
 void rs690_pm_info(struct radeon_device *rdev);
 extern int rv6xx_get_temp(struct radeon_device *rdev);
 extern int rv770_get_temp(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index d948265db87e..b2449629537d 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -906,9 +906,9 @@ static struct radeon_asic cayman_asic = {
 	.get_vblank_counter = &evergreen_get_vblank_counter,
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
-	.copy_blit = NULL,
-	.copy_dma = NULL,
-	.copy = NULL,
+	.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 = &radeon_atom_get_memory_clock,
@@ -938,6 +938,13 @@ static struct radeon_asic cayman_asic = {
 int radeon_asic_init(struct radeon_device *rdev)
 {
 	radeon_register_accessor_init(rdev);
+
+	/* set the number of crtcs */
+	if (rdev->flags & RADEON_SINGLE_CRTC)
+		rdev->num_crtc = 1;
+	else
+		rdev->num_crtc = 2;
+
 	switch (rdev->family) {
 	case CHIP_R100:
 	case CHIP_RV100:
@@ -1017,18 +1024,32 @@ int radeon_asic_init(struct radeon_device *rdev)
 	case CHIP_JUNIPER:
 	case CHIP_CYPRESS:
 	case CHIP_HEMLOCK:
+		/* set num crtcs */
+		if (rdev->family == CHIP_CEDAR)
+			rdev->num_crtc = 4;
+		else
+			rdev->num_crtc = 6;
 		rdev->asic = &evergreen_asic;
 		break;
 	case CHIP_PALM:
+	case CHIP_SUMO:
+	case CHIP_SUMO2:
 		rdev->asic = &sumo_asic;
 		break;
 	case CHIP_BARTS:
 	case CHIP_TURKS:
 	case CHIP_CAICOS:
+		/* set num crtcs */
+		if (rdev->family == CHIP_CAICOS)
+			rdev->num_crtc = 4;
+		else
+			rdev->num_crtc = 6;
 		rdev->asic = &btc_asic;
 		break;
 	case CHIP_CAYMAN:
 		rdev->asic = &cayman_asic;
+		/* set num crtcs */
+		rdev->num_crtc = 6;
 		break;
 	default:
 		/* FIXME: not supported yet */
@@ -1040,18 +1061,6 @@ int radeon_asic_init(struct radeon_device *rdev)
 		rdev->asic->set_memory_clock = NULL;
 	}
 
-	/* set the number of crtcs */
-	if (rdev->flags & RADEON_SINGLE_CRTC)
-		rdev->num_crtc = 1;
-	else {
-		if (ASIC_IS_DCE41(rdev))
-			rdev->num_crtc = 2;
-		else if (ASIC_IS_DCE4(rdev))
-			rdev->num_crtc = 6;
-		else
-			rdev->num_crtc = 2;
-	}
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 90dfb2b8cf03..bf2b61584cdb 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -1246,6 +1246,10 @@ bool radeon_atom_get_clock_info(struct drm_device *dev)
 		}
 		*dcpll = *p1pll;
 
+		rdev->clock.max_pixel_clock = le16_to_cpu(firmware_info->info.usMaxPixelClock);
+		if (rdev->clock.max_pixel_clock == 0)
+			rdev->clock.max_pixel_clock = 40000;
+
 		return true;
 	}
 
@@ -2316,6 +2320,14 @@ static bool radeon_atombios_parse_pplib_clock_info(struct radeon_device *rdev,
 			le16_to_cpu(clock_info->r600.usVDDC);
 	}
 
+	/* patch up vddc if necessary */
+	if (rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage == 0xff01) {
+		u16 vddc;
+
+		if (radeon_atom_get_max_vddc(rdev, &vddc) == 0)
+			rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage = vddc;
+	}
+
 	if (rdev->flags & RADEON_IS_IGP) {
 		/* skip invalid modes */
 		if (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0)
@@ -2603,6 +2615,10 @@ void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 v
 	if (!atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev, &crev))
 		return;
 
+	/* 0xff01 is a flag rather then an actual voltage */
+	if (voltage_level == 0xff01)
+		return;
+
 	switch (crev) {
 	case 1:
 		args.v1.ucVoltageType = voltage_type;
@@ -2622,7 +2638,35 @@ void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 v
 	atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 }
 
+int radeon_atom_get_max_vddc(struct radeon_device *rdev,
+			     u16 *voltage)
+{
+	union set_voltage args;
+	int index = GetIndexIntoMasterTable(COMMAND, SetVoltage);
+	u8 frev, crev;
+
+	if (!atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev, &crev))
+		return -EINVAL;
 
+	switch (crev) {
+	case 1:
+		return -EINVAL;
+	case 2:
+		args.v2.ucVoltageType = SET_VOLTAGE_GET_MAX_VOLTAGE;
+		args.v2.ucVoltageMode = 0;
+		args.v2.usVoltageLevel = 0;
+
+		atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+
+		*voltage = le16_to_cpu(args.v2.usVoltageLevel);
+		break;
+	default:
+		DRM_ERROR("Unknown table version %d, %d\n", frev, crev);
+		return -EINVAL;
+	}
+
+	return 0;
+}
 
 void radeon_atom_initialize_bios_scratch_regs(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index 1aba85cad1a8..229a20f10e2b 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -104,7 +104,7 @@ static bool radeon_read_bios(struct radeon_device *rdev)
 static bool radeon_atrm_get_bios(struct radeon_device *rdev)
 {
 	int ret;
-	int size = 64 * 1024;
+	int size = 256 * 1024;
 	int i;
 
 	if (!radeon_atrm_supported(rdev->pdev))
@@ -331,7 +331,7 @@ static bool avivo_read_disabled_bios(struct radeon_device *rdev)
 
 	seprom_cntl1 = RREG32(RADEON_SEPROM_CNTL1);
 	viph_control = RREG32(RADEON_VIPH_CONTROL);
-	bus_cntl = RREG32(RADEON_BUS_CNTL);
+	bus_cntl = RREG32(RV370_BUS_CNTL);
 	d1vga_control = RREG32(AVIVO_D1VGA_CONTROL);
 	d2vga_control = RREG32(AVIVO_D2VGA_CONTROL);
 	vga_render_control = RREG32(AVIVO_VGA_RENDER_CONTROL);
@@ -350,7 +350,7 @@ static bool avivo_read_disabled_bios(struct radeon_device *rdev)
 	WREG32(RADEON_VIPH_CONTROL, (viph_control & ~RADEON_VIPH_EN));
 
 	/* enable the rom */
-	WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM));
+	WREG32(RV370_BUS_CNTL, (bus_cntl & ~RV370_BUS_BIOS_DIS_ROM));
 
 	/* Disable VGA mode */
 	WREG32(AVIVO_D1VGA_CONTROL,
@@ -367,7 +367,7 @@ static bool avivo_read_disabled_bios(struct radeon_device *rdev)
 	/* restore regs */
 	WREG32(RADEON_SEPROM_CNTL1, seprom_cntl1);
 	WREG32(RADEON_VIPH_CONTROL, viph_control);
-	WREG32(RADEON_BUS_CNTL, bus_cntl);
+	WREG32(RV370_BUS_CNTL, bus_cntl);
 	WREG32(AVIVO_D1VGA_CONTROL, d1vga_control);
 	WREG32(AVIVO_D2VGA_CONTROL, d2vga_control);
 	WREG32(AVIVO_VGA_RENDER_CONTROL, vga_render_control);
@@ -390,7 +390,10 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev)
 
 	seprom_cntl1 = RREG32(RADEON_SEPROM_CNTL1);
 	viph_control = RREG32(RADEON_VIPH_CONTROL);
-	bus_cntl = RREG32(RADEON_BUS_CNTL);
+	if (rdev->flags & RADEON_IS_PCIE)
+		bus_cntl = RREG32(RV370_BUS_CNTL);
+	else
+		bus_cntl = RREG32(RADEON_BUS_CNTL);
 	crtc_gen_cntl = RREG32(RADEON_CRTC_GEN_CNTL);
 	crtc2_gen_cntl = 0;
 	crtc_ext_cntl = RREG32(RADEON_CRTC_EXT_CNTL);
@@ -412,7 +415,10 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev)
 	WREG32(RADEON_VIPH_CONTROL, (viph_control & ~RADEON_VIPH_EN));
 
 	/* enable the rom */
-	WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM));
+	if (rdev->flags & RADEON_IS_PCIE)
+		WREG32(RV370_BUS_CNTL, (bus_cntl & ~RV370_BUS_BIOS_DIS_ROM));
+	else
+		WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM));
 
 	/* Turn off mem requests and CRTC for both controllers */
 	WREG32(RADEON_CRTC_GEN_CNTL,
@@ -439,7 +445,10 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev)
 	/* restore regs */
 	WREG32(RADEON_SEPROM_CNTL1, seprom_cntl1);
 	WREG32(RADEON_VIPH_CONTROL, viph_control);
-	WREG32(RADEON_BUS_CNTL, bus_cntl);
+	if (rdev->flags & RADEON_IS_PCIE)
+		WREG32(RV370_BUS_CNTL, bus_cntl);
+	else
+		WREG32(RADEON_BUS_CNTL, bus_cntl);
 	WREG32(RADEON_CRTC_GEN_CNTL, crtc_gen_cntl);
 	if (!(rdev->flags & RADEON_SINGLE_CRTC)) {
 		WREG32(RADEON_CRTC2_GEN_CNTL, crtc2_gen_cntl);
diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c
index 5249af8931e6..2d48e7a1474b 100644
--- a/drivers/gpu/drm/radeon/radeon_clocks.c
+++ b/drivers/gpu/drm/radeon/radeon_clocks.c
@@ -117,7 +117,7 @@ static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
 	p1pll->reference_div = RREG32_PLL(RADEON_PPLL_REF_DIV) & 0x3ff;
 	if (p1pll->reference_div < 2)
 		p1pll->reference_div = 12;
-	p2pll->reference_div = p1pll->reference_div;	
+	p2pll->reference_div = p1pll->reference_div;
 
 	/* These aren't in the device-tree */
 	if (rdev->family >= CHIP_R420) {
@@ -139,6 +139,8 @@ static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
 		p2pll->pll_out_min = 12500;
 		p2pll->pll_out_max = 35000;
 	}
+	/* not sure what the max should be in all cases */
+	rdev->clock.max_pixel_clock = 35000;
 
 	spll->reference_freq = mpll->reference_freq = p1pll->reference_freq;
 	spll->reference_div = mpll->reference_div =
@@ -151,7 +153,7 @@ static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
 	else
 		rdev->clock.default_sclk =
 			radeon_legacy_get_engine_clock(rdev);
-			
+
 	val = of_get_property(dp, "ATY,MCLK", NULL);
 	if (val && *val)
 		rdev->clock.default_mclk = (*val) / 10;
@@ -160,7 +162,7 @@ static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
 			radeon_legacy_get_memory_clock(rdev);
 
 	DRM_INFO("Using device-tree clock info\n");
-	
+
 	return true;
 }
 #else
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 5b991f7c6e2a..e4594676a07c 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -866,6 +866,11 @@ bool radeon_combios_get_clock_info(struct drm_device *dev)
 		rdev->clock.default_sclk = sclk;
 		rdev->clock.default_mclk = mclk;
 
+		if (RBIOS32(pll_info + 0x16))
+			rdev->clock.max_pixel_clock = RBIOS32(pll_info + 0x16);
+		else
+			rdev->clock.max_pixel_clock = 35000; /* might need something asic specific */
+
 		return true;
 	}
 	return false;
@@ -1548,10 +1553,12 @@ bool radeon_get_legacy_connector_info_from_table(struct drm_device *dev)
 			   (rdev->pdev->subsystem_device == 0x4a48)) {
 			/* Mac X800 */
 			rdev->mode_info.connector_table = CT_MAC_X800;
-		} else if ((rdev->pdev->device == 0x4150) &&
+		} else if ((of_machine_is_compatible("PowerMac7,2") ||
+			    of_machine_is_compatible("PowerMac7,3")) &&
+			   (rdev->pdev->device == 0x4150) &&
 			   (rdev->pdev->subsystem_vendor == 0x1002) &&
 			   (rdev->pdev->subsystem_device == 0x4150)) {
-			/* Mac G5 9600 */
+			/* Mac G5 tower 9600 */
 			rdev->mode_info.connector_table = CT_MAC_G5_9600;
 		} else
 #endif /* CONFIG_PPC_PMAC */
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index ee1dccb3fec9..9792d4ffdc86 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
author	Linus Torvalds <torvalds@linux-foundation.org>	2022-09-30 09:33:33 -0700
committer	Linus Torvalds <torvalds@linux-foundation.org>	2022-09-30 09:33:33 -0700
commit	7bc6e90d7aa4170039abe80b9f4e8c8e4eb35091 (patch)
tree	14bc5b5b614437063a144a439ab529e1887b003b /arch/arc/plat-sim/git:/git.zx2c4.com/ssh:/git@git.zx2c4.com
parent	Merge tag 'io_uring-6.0-2022-09-29' of git://git.kernel.dk/linux (diff)
parent	Merge tag 'nvme-6.0-2022-09-29' of git://git.infradead.org/nvme into block-6.0 (diff)