Merge branch 'master' into for-next

7 files changed, 1033 insertions, 649 deletions

diff --git a/arch/mips/cavium-octeon/Kconfig b/arch/mips/cavium-octeon/Kconfig
index 47323ca452dc..caae22858163 100644
--- a/arch/mips/cavium-octeon/Kconfig
+++ b/arch/mips/cavium-octeon/Kconfig
@@ -3,6 +3,17 @@ config CAVIUM_OCTEON_SPECIFIC_OPTIONS
 	depends on CPU_CAVIUM_OCTEON
 	default "y"
 
+config CAVIUM_CN63XXP1
+	bool "Enable CN63XXP1 errata worarounds"
+	depends on CAVIUM_OCTEON_SPECIFIC_OPTIONS
+	default "n"
+	help
+	  The CN63XXP1 chip requires build time workarounds to
+	  function reliably, select this option to enable them.  These
+	  workarounds will cause a slight decrease in performance on
+	  non-CN63XXP1 hardware, so it is recommended to select "n"
+	  unless it is known the workarounds are needed.
+
 config CAVIUM_OCTEON_2ND_KERNEL
 	bool "Build the kernel to be used as a 2nd kernel on the same chip"
 	depends on CAVIUM_OCTEON_SPECIFIC_OPTIONS
@@ -87,3 +98,15 @@ config ARCH_SPARSEMEM_ENABLE
 config CAVIUM_OCTEON_HELPER
 	def_bool y
 	depends on OCTEON_ETHERNET || PCI
+
+config IOMMU_HELPER
+	bool
+
+config NEED_SG_DMA_LENGTH
+	bool
+
+config SWIOTLB
+	def_bool y
+	depends on CPU_CAVIUM_OCTEON
+	select IOMMU_HELPER
+	select NEED_SG_DMA_LENGTH
diff --git a/arch/mips/cavium-octeon/csrc-octeon.c b/arch/mips/cavium-octeon/csrc-octeon.c
index b6847c8e0ddd..26bf71130bf8 100644
--- a/arch/mips/cavium-octeon/csrc-octeon.c
+++ b/arch/mips/cavium-octeon/csrc-octeon.c
@@ -4,14 +4,18 @@
  * for more details.
  *
  * Copyright (C) 2007 by Ralf Baechle
+ * Copyright (C) 2009, 2010 Cavium Networks, Inc.
  */
 #include <linux/clocksource.h>
 #include <linux/init.h>
+#include <linux/smp.h>
 
+#include <asm/cpu-info.h>
 #include <asm/time.h>
 
 #include <asm/octeon/octeon.h>
 #include <asm/octeon/cvmx-ipd-defs.h>
+#include <asm/octeon/cvmx-mio-defs.h>
 
 /*
  * Set the current core's cvmcount counter to the value of the
@@ -19,11 +23,23 @@
  * on-line.  This allows for a read from a local cpu register to
  * access a synchronized counter.
  *
+ * On CPU_CAVIUM_OCTEON2 the IPD_CLK_COUNT is scaled by rdiv/sdiv.
  */
 void octeon_init_cvmcount(void)
 {
 	unsigned long flags;
 	unsigned loops = 2;
+	u64 f = 0;
+	u64 rdiv = 0;
+	u64 sdiv = 0;
+	if (current_cpu_type() == CPU_CAVIUM_OCTEON2) {
+		union cvmx_mio_rst_boot rst_boot;
+		rst_boot.u64 = cvmx_read_csr(CVMX_MIO_RST_BOOT);
+		rdiv = rst_boot.s.c_mul;	/* CPU clock */
+		sdiv = rst_boot.s.pnr_mul;	/* I/O clock */
+		f = (0x8000000000000000ull / sdiv) * 2;
+	}
+
 
 	/* Clobber loops so GCC will not unroll the following while loop. */
 	asm("" : "+r" (loops));
@@ -33,8 +49,20 @@ void octeon_init_cvmcount(void)
 	 * Loop several times so we are executing from the cache,
 	 * which should give more deterministic timing.
 	 */
-	while (loops--)
-		write_c0_cvmcount(cvmx_read_csr(CVMX_IPD_CLK_COUNT));
+	while (loops--) {
+		u64 ipd_clk_count = cvmx_read_csr(CVMX_IPD_CLK_COUNT);
+		if (rdiv != 0) {
+			ipd_clk_count *= rdiv;
+			if (f != 0) {
+				asm("dmultu\t%[cnt],%[f]\n\t"
+				    "mfhi\t%[cnt]"
+				    : [cnt] "+r" (ipd_clk_count),
+				      [f] "=r" (f)
+				    : : "hi", "lo");
+			}
+		}
+		write_c0_cvmcount(ipd_clk_count);
+	}
 	local_irq_restore(flags);
 }
 
@@ -77,7 +105,7 @@ unsigned long long notrace sched_clock(void)
 void __init plat_time_init(void)
 {
 	clocksource_mips.rating = 300;
-	clocksource_set_clock(&clocksource_mips, mips_hpt_frequency);
+	clocksource_set_clock(&clocksource_mips, octeon_get_clock_rate());
 	clocksource_register(&clocksource_mips);
 }
 
diff --git a/arch/mips/cavium-octeon/dma-octeon.c b/arch/mips/cavium-octeon/dma-octeon.c
index d22b5a2d64f4..1abb66caaa1d 100644
--- a/arch/mips/cavium-octeon/dma-octeon.c
+++ b/arch/mips/cavium-octeon/dma-octeon.c
@@ -8,335 +8,342 @@
  * Copyright (C) 2005 Ilya A. Volynets-Evenbakh <ilya@total-knowledge.com>
  * swiped from i386, and cloned for MIPS by Geert, polished by Ralf.
  * IP32 changes by Ilya.
- * Cavium Networks: Create new dma setup for Cavium Networks Octeon based on
- * the kernels original.
+ * Copyright (C) 2010 Cavium Networks, Inc.
  */
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/module.h>
-#include <linux/string.h>
 #include <linux/dma-mapping.h>
-#include <linux/platform_device.h>
 #include <linux/scatterlist.h>
+#include <linux/bootmem.h>
+#include <linux/swiotlb.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/mm.h>
 
-#include <linux/cache.h>
-#include <linux/io.h>
+#include <asm/bootinfo.h>
 
 #include <asm/octeon/octeon.h>
+
+#ifdef CONFIG_PCI
+#include <asm/octeon/pci-octeon.h>
 #include <asm/octeon/cvmx-npi-defs.h>
 #include <asm/octeon/cvmx-pci-defs.h>
 
-#include <dma-coherence.h>
+static dma_addr_t octeon_hole_phys_to_dma(phys_addr_t paddr)
+{
+	if (paddr >= CVMX_PCIE_BAR1_PHYS_BASE && paddr < (CVMX_PCIE_BAR1_PHYS_BASE + CVMX_PCIE_BAR1_PHYS_SIZE))
+		return paddr - CVMX_PCIE_BAR1_PHYS_BASE + CVMX_PCIE_BAR1_RC_BASE;
+	else
+		return paddr;
+}
 
-#ifdef CONFIG_PCI
-#include <asm/octeon/pci-octeon.h>
-#endif
+static phys_addr_t octeon_hole_dma_to_phys(dma_addr_t daddr)
+{
+	if (daddr >= CVMX_PCIE_BAR1_RC_BASE)
+		return daddr + CVMX_PCIE_BAR1_PHYS_BASE - CVMX_PCIE_BAR1_RC_BASE;
+	else
+		return daddr;
+}
+
+static dma_addr_t octeon_gen1_phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	if (paddr >= 0x410000000ull && paddr < 0x420000000ull)
+		paddr -= 0x400000000ull;
+	return octeon_hole_phys_to_dma(paddr);
+}
 
-#define BAR2_PCI_ADDRESS 0x8000000000ul
+static phys_addr_t octeon_gen1_dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	daddr = octeon_hole_dma_to_phys(daddr);
 
-struct bar1_index_state {
-	int16_t ref_count;	/* Number of PCI mappings using this index */
-	uint16_t address_bits;	/* Upper bits of physical address. This is
-				   shifted 22 bits */
-};
+	if (daddr >= 0x10000000ull && daddr < 0x20000000ull)
+		daddr += 0x400000000ull;
 
-#ifdef CONFIG_PCI
-static DEFINE_RAW_SPINLOCK(bar1_lock);
-static struct bar1_index_state bar1_state[32];
-#endif
+	return daddr;
+}
 
-dma_addr_t octeon_map_dma_mem(struct device *dev, void *ptr, size_t size)
+static dma_addr_t octeon_big_phys_to_dma(struct device *dev, phys_addr_t paddr)
 {
-#ifndef CONFIG_PCI
-	/* Without PCI/PCIe this function can be called for Octeon internal
-	   devices such as USB. These devices all support 64bit addressing */
+	if (paddr >= 0x410000000ull && paddr < 0x420000000ull)
+		paddr -= 0x400000000ull;
+
+	/* Anything in the BAR1 hole or above goes via BAR2 */
+	if (paddr >= 0xf0000000ull)
+		paddr = OCTEON_BAR2_PCI_ADDRESS + paddr;
+
+	return paddr;
+}
+
+static phys_addr_t octeon_big_dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	if (daddr >= OCTEON_BAR2_PCI_ADDRESS)
+		daddr -= OCTEON_BAR2_PCI_ADDRESS;
+
+	if (daddr >= 0x10000000ull && daddr < 0x20000000ull)
+		daddr += 0x400000000ull;
+	return daddr;
+}
+
+static dma_addr_t octeon_small_phys_to_dma(struct device *dev,
+					   phys_addr_t paddr)
+{
+	if (paddr >= 0x410000000ull && paddr < 0x420000000ull)
+		paddr -= 0x400000000ull;
+
+	/* Anything not in the BAR1 range goes via BAR2 */
+	if (paddr >= octeon_bar1_pci_phys && paddr < octeon_bar1_pci_phys + 0x8000000ull)
+		paddr = paddr - octeon_bar1_pci_phys;
+	else
+		paddr = OCTEON_BAR2_PCI_ADDRESS + paddr;
+
+	return paddr;
+}
+
+static phys_addr_t octeon_small_dma_to_phys(struct device *dev,
+					    dma_addr_t daddr)
+{
+	if (daddr >= OCTEON_BAR2_PCI_ADDRESS)
+		daddr -= OCTEON_BAR2_PCI_ADDRESS;
+	else
+		daddr += octeon_bar1_pci_phys;
+
+	if (daddr >= 0x10000000ull && daddr < 0x20000000ull)
+		daddr += 0x400000000ull;
+	return daddr;
+}
+
+#endif /* CONFIG_PCI */
+
+static dma_addr_t octeon_dma_map_page(struct device *dev, struct page *page,
+	unsigned long offset, size_t size, enum dma_data_direction direction,
+	struct dma_attrs *attrs)
+{
+	dma_addr_t daddr = swiotlb_map_page(dev, page, offset, size,
+					    direction, attrs);
 	mb();
-	return virt_to_phys(ptr);
-#else
-	unsigned long flags;
-	uint64_t dma_mask;
-	int64_t start_index;
-	dma_addr_t result = -1;
-	uint64_t physical = virt_to_phys(ptr);
-	int64_t index;
 
+	return daddr;
+}
+
+static int octeon_dma_map_sg(struct device *dev, struct scatterlist *sg,
+	int nents, enum dma_data_direction direction, struct dma_attrs *attrs)
+{
+	int r = swiotlb_map_sg_attrs(dev, sg, nents, direction, attrs);
 	mb();
-	/*
-	 * Use the DMA masks to determine the allowed memory
-	 * region. For us it doesn't limit the actual memory, just the
-	 * address visible over PCI.  Devices with limits need to use
-	 * lower indexed Bar1 entries.
-	 */
-	if (dev) {
-		dma_mask = dev->coherent_dma_mask;
-		if (dev->dma_mask)
-			dma_mask = *dev->dma_mask;
-	} else {
-		dma_mask = 0xfffffffful;
-	}
+	return r;
+}
 
-	/*
-	 * Platform devices, such as the internal USB, skip all
-	 * translation and use Octeon physical addresses directly.
-	 */
-	if (!dev || dev->bus == &platform_bus_type)
-		return physical;
+static void octeon_dma_sync_single_for_device(struct device *dev,
+	dma_addr_t dma_handle, size_t size, enum dma_data_direction direction)
+{
+	swiotlb_sync_single_for_device(dev, dma_handle, size, direction);
+	mb();
+}
 
-	switch (octeon_dma_bar_type) {
-	case OCTEON_DMA_BAR_TYPE_PCIE:
-		if (unlikely(physical < (16ul << 10)))
-			panic("dma_map_single: Not allowed to map first 16KB."
-			      " It interferes with BAR0 special area\n");
-		else if ((physical + size >= (256ul << 20)) &&
-			 (physical < (512ul << 20)))
-			panic("dma_map_single: Not allowed to map bootbus\n");
-		else if ((physical + size >= 0x400000000ull) &&
-			 physical < 0x410000000ull)
-			panic("dma_map_single: "
-			      "Attempt to map illegal memory address 0x%llx\n",
-			      physical);
-		else if (physical >= 0x420000000ull)
-			panic("dma_map_single: "
-			      "Attempt to map illegal memory address 0x%llx\n",
-			      physical);
-		else if (physical >= CVMX_PCIE_BAR1_PHYS_BASE &&
-			 physical + size < (CVMX_PCIE_BAR1_PHYS_BASE + CVMX_PCIE_BAR1_PHYS_SIZE)) {
-			result = physical - CVMX_PCIE_BAR1_PHYS_BASE + CVMX_PCIE_BAR1_RC_BASE;
-
-			if (((result+size-1) & dma_mask) != result+size-1)
-				panic("dma_map_single: Attempt to map address 0x%llx-0x%llx, which can't be accessed according to the dma mask 0x%llx\n",
-				      physical, physical+size-1, dma_mask);
-			goto done;
-		}
-
-		/* The 2nd 256MB is mapped at 256<<20 instead of 0x410000000 */
-		if ((physical >= 0x410000000ull) && physical < 0x420000000ull)
-			result = physical - 0x400000000ull;
-		else
-			result = physical;
-		if (((result+size-1) & dma_mask) != result+size-1)
-			panic("dma_map_single: Attempt to map address "
-			      "0x%llx-0x%llx, which can't be accessed "
-			      "according to the dma mask 0x%llx\n",
-			      physical, physical+size-1, dma_mask);
-		goto done;
+static void octeon_dma_sync_sg_for_device(struct device *dev,
+	struct scatterlist *sg, int nelems, enum dma_data_direction direction)
+{
+	swiotlb_sync_sg_for_device(dev, sg, nelems, direction);
+	mb();
+}
 
-	case OCTEON_DMA_BAR_TYPE_BIG:
-#ifdef CONFIG_64BIT
-		/* If the device supports 64bit addressing, then use BAR2 */
-		if (dma_mask > BAR2_PCI_ADDRESS) {
-			result = physical + BAR2_PCI_ADDRESS;
-			goto done;
-		}
-#endif
-		if (unlikely(physical < (4ul << 10))) {
-			panic("dma_map_single: Not allowed to map first 4KB. "
-			      "It interferes with BAR0 special area\n");
-		} else if (physical < (256ul << 20)) {
-			if (unlikely(physical + size > (256ul << 20)))
-				panic("dma_map_single: Requested memory spans "
-				      "Bar0 0:256MB and bootbus\n");
-			result = physical;
-			goto done;
-		} else if (unlikely(physical < (512ul << 20))) {
-			panic("dma_map_single: Not allowed to map bootbus\n");
-		} else if (physical < (2ul << 30)) {
-			if (unlikely(physical + size > (2ul << 30)))
-				panic("dma_map_single: Requested memory spans "
-				      "Bar0 512MB:2GB and BAR1\n");
-			result = physical;
-			goto done;
-		} else if (physical < (2ul << 30) + (128 << 20)) {
-			/* Fall through */
-		} else if (physical <
-			   (4ul << 30) - (OCTEON_PCI_BAR1_HOLE_SIZE << 20)) {
-			if (unlikely
-			    (physical + size >
-			     (4ul << 30) - (OCTEON_PCI_BAR1_HOLE_SIZE << 20)))
-				panic("dma_map_single: Requested memory "
-				      "extends past Bar1 (4GB-%luMB)\n",
-				      OCTEON_PCI_BAR1_HOLE_SIZE);
-			result = physical;
-			goto done;
-		} else if ((physical >= 0x410000000ull) &&
-			   (physical < 0x420000000ull)) {
-			if (unlikely(physical + size > 0x420000000ull))
-				panic("dma_map_single: Requested memory spans "
-				      "non existant memory\n");
-			/* BAR0 fixed mapping 256MB:512MB ->
-			 * 16GB+256MB:16GB+512MB */
-			result = physical - 0x400000000ull;
-			goto done;
-		} else {
-			/* Continued below switch statement */
-		}
-		break;
+static void *octeon_dma_alloc_coherent(struct device *dev, size_t size,
+	dma_addr_t *dma_handle, gfp_t gfp)
+{
+	void *ret;
 
-	case OCTEON_DMA_BAR_TYPE_SMALL:
-#ifdef CONFIG_64BIT
-		/* If the device supports 64bit addressing, then use BAR2 */
-		if (dma_mask > BAR2_PCI_ADDRESS) {
-			result = physical + BAR2_PCI_ADDRESS;
-			goto done;
-		}
+	if (dma_alloc_from_coherent(dev, size, dma_handle, &ret))
+		return ret;
+
+	/* ignore region specifiers */
+	gfp &= ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM);
+
+#ifdef CONFIG_ZONE_DMA
+	if (dev == NULL)
+		gfp |= __GFP_DMA;
+	else if (dev->coherent_dma_mask <= DMA_BIT_MASK(24))
+		gfp |= __GFP_DMA;
+	else
 #endif
-		/* Continued below switch statement */
-		break;
+#ifdef CONFIG_ZONE_DMA32
+	     if (dev->coherent_dma_mask <= DMA_BIT_MASK(32))
+		gfp |= __GFP_DMA32;
+	else
+#endif
+		;
 
-	default:
-		panic("dma_map_single: Invalid octeon_dma_bar_type\n");
-	}
+	/* Don't invoke OOM killer */
+	gfp |= __GFP_NORETRY;
 
-	/* Don't allow mapping to span multiple Bar entries. The hardware guys
-	   won't guarantee that DMA across boards work */
-	if (unlikely((physical >> 22) != ((physical + size - 1) >> 22)))
-		panic("dma_map_single: "
-		      "Requested memory spans more than one Bar1 entry\n");
+	ret = swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
 
-	if (octeon_dma_bar_type == OCTEON_DMA_BAR_TYPE_BIG)
-		start_index = 31;
-	else if (unlikely(dma_mask < (1ul << 27)))
-		start_index = (dma_mask >> 22);
-	else
-		start_index = 31;
-
-	/* Only one processor can access the Bar register at once */
-	raw_spin_lock_irqsave(&bar1_lock, flags);
-
-	/* Look through Bar1 for existing mapping that will work */
-	for (index = start_index; index >= 0; index--) {
-		if ((bar1_state[index].address_bits == physical >> 22) &&
-		    (bar1_state[index].ref_count)) {
-			/* An existing mapping will work, use it */
-			bar1_state[index].ref_count++;
-			if (unlikely(bar1_state[index].ref_count < 0))
-				panic("dma_map_single: "
-				      "Bar1[%d] reference count overflowed\n",
-				      (int) index);
-			result = (index << 22) | (physical & ((1 << 22) - 1));
-			/* Large BAR1 is offset at 2GB */
-			if (octeon_dma_bar_type == OCTEON_DMA_BAR_TYPE_BIG)
-				result += 2ul << 30;
-			goto done_unlock;
-		}
-	}
+	mb();
 
-	/* No existing mappings, look for a free entry */
-	for (index = start_index; index >= 0; index--) {
-		if (unlikely(bar1_state[index].ref_count == 0)) {
-			union cvmx_pci_bar1_indexx bar1_index;
-			/* We have a free entry, use it */
-			bar1_state[index].ref_count = 1;
-			bar1_state[index].address_bits = physical >> 22;
-			bar1_index.u32 = 0;
-			/* Address bits[35:22] sent to L2C */
-			bar1_index.s.addr_idx = physical >> 22;
-			/* Don't put PCI accesses in L2. */
-			bar1_index.s.ca = 1;
-			/* Endian Swap Mode */
-			bar1_index.s.end_swp = 1;
-			/* Set '1' when the selected address range is valid. */
-			bar1_index.s.addr_v = 1;
-			octeon_npi_write32(CVMX_NPI_PCI_BAR1_INDEXX(index),
-					   bar1_index.u32);
-			/* An existing mapping will work, use it */
-			result = (index << 22) | (physical & ((1 << 22) - 1));
-			/* Large BAR1 is offset at 2GB */
-			if (octeon_dma_bar_type == OCTEON_DMA_BAR_TYPE_BIG)
-				result += 2ul << 30;
-			goto done_unlock;
-		}
-	}
+	return ret;
+}
 
-	pr_err("dma_map_single: "
-	       "Can't find empty BAR1 index for physical mapping 0x%llx\n",
-	       (unsigned long long) physical);
+static void octeon_dma_free_coherent(struct device *dev, size_t size,
+	void *vaddr, dma_addr_t dma_handle)
+{
+	int order = get_order(size);
 
-done_unlock:
-	raw_spin_unlock_irqrestore(&bar1_lock, flags);
-done:
-	pr_debug("dma_map_single 0x%llx->0x%llx\n", physical, result);
-	return result;
-#endif
+	if (dma_release_from_coherent(dev, order, vaddr))
+		return;
+
+	swiotlb_free_coherent(dev, size, vaddr, dma_handle);
 }
 
-void octeon_unmap_dma_mem(struct device *dev, dma_addr_t dma_addr)
+static dma_addr_t octeon_unity_phys_to_dma(struct device *dev, phys_addr_t paddr)
 {
-#ifndef CONFIG_PCI
-	/*
-	 * Without PCI/PCIe this function can be called for Octeon internal
-	 * devices such as USB. These devices all support 64bit addressing.
-	 */
-	return;
-#else
-	unsigned long flags;
-	uint64_t index;
+	return paddr;
+}
 
+static phys_addr_t octeon_unity_dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	return daddr;
+}
+
+struct octeon_dma_map_ops {
+	struct dma_map_ops dma_map_ops;
+	dma_addr_t (*phys_to_dma)(struct device *dev, phys_addr_t paddr);
+	phys_addr_t (*dma_to_phys)(struct device *dev, dma_addr_t daddr);
+};
+
+dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	struct octeon_dma_map_ops *ops = container_of(get_dma_ops(dev),
+						      struct octeon_dma_map_ops,
+						      dma_map_ops);
+
+	return ops->phys_to_dma(dev, paddr);
+}
+EXPORT_SYMBOL(phys_to_dma);
+
+phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	struct octeon_dma_map_ops *ops = container_of(get_dma_ops(dev),
+						      struct octeon_dma_map_ops,
+						      dma_map_ops);
+
+	return ops->dma_to_phys(dev, daddr);
+}
+EXPORT_SYMBOL(dma_to_phys);
+
+static struct octeon_dma_map_ops octeon_linear_dma_map_ops = {
+	.dma_map_ops = {
+		.alloc_coherent = octeon_dma_alloc_coherent,
+		.free_coherent = octeon_dma_free_coherent,
+		.map_page = octeon_dma_map_page,
+		.unmap_page = swiotlb_unmap_page,
+		.map_sg = octeon_dma_map_sg,
+		.unmap_sg = swiotlb_unmap_sg_attrs,
+		.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
+		.sync_single_for_device = octeon_dma_sync_single_for_device,
+		.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
+		.sync_sg_for_device = octeon_dma_sync_sg_for_device,
+		.mapping_error = swiotlb_dma_mapping_error,
+		.dma_supported = swiotlb_dma_supported
+	},
+	.phys_to_dma = octeon_unity_phys_to_dma,
+	.dma_to_phys = octeon_unity_dma_to_phys
+};
+
+char *octeon_swiotlb;
+
+void __init plat_swiotlb_setup(void)
+{
+	int i;
+	phys_t max_addr;
+	phys_t addr_size;
+	size_t swiotlbsize;
+	unsigned long swiotlb_nslabs;
+
+	max_addr = 0;
+	addr_size = 0;
+
+	for (i = 0 ; i < boot_mem_map.nr_map; i++) {
+		struct boot_mem_map_entry *e = &boot_mem_map.map[i];
+		if (e->type != BOOT_MEM_RAM)
+			continue;
+
+		/* These addresses map low for PCI. */
+		if (e->addr > 0x410000000ull)
+			continue;
+
+		addr_size += e->size;
+
+		if (max_addr < e->addr + e->size)
+			max_addr = e->addr + e->size;
+
+	}
+
+	swiotlbsize = PAGE_SIZE;
+
+#ifdef CONFIG_PCI
 	/*
-	 * Platform devices, such as the internal USB, skip all
-	 * translation and use Octeon physical addresses directly.
+	 * For OCTEON_DMA_BAR_TYPE_SMALL, size the iotlb at 1/4 memory
+	 * size to a maximum of 64MB
 	 */
-	if (dev->bus == &platform_bus_type)
-		return;
+	if (OCTEON_IS_MODEL(OCTEON_CN31XX)
+	    || OCTEON_IS_MODEL(OCTEON_CN38XX_PASS2)) {
+		swiotlbsize = addr_size / 4;
+		if (swiotlbsize > 64 * (1<<20))
+			swiotlbsize = 64 * (1<<20);
+	} else if (max_addr > 0xf0000000ul) {
+		/*
+		 * Otherwise only allocate a big iotlb if there is
+		 * memory past the BAR1 hole.
+		 */
+		swiotlbsize = 64 * (1<<20);
+	}
+#endif
+	swiotlb_nslabs = swiotlbsize >> IO_TLB_SHIFT;
+	swiotlb_nslabs = ALIGN(swiotlb_nslabs, IO_TLB_SEGSIZE);
+	swiotlbsize = swiotlb_nslabs << IO_TLB_SHIFT;
+
+	octeon_swiotlb = alloc_bootmem_low_pages(swiotlbsize);
 
+	swiotlb_init_with_tbl(octeon_swiotlb, swiotlb_nslabs, 1);
+
+	mips_dma_map_ops = &octeon_linear_dma_map_ops.dma_map_ops;
+}
+
+#ifdef CONFIG_PCI
+static struct octeon_dma_map_ops _octeon_pci_dma_map_ops = {
+	.dma_map_ops = {
+		.alloc_coherent = octeon_dma_alloc_coherent,
+		.free_coherent = octeon_dma_free_coherent,
+		.map_page = octeon_dma_map_page,
+		.unmap_page = swiotlb_unmap_page,
+		.map_sg = octeon_dma_map_sg,
+		.unmap_sg = swiotlb_unmap_sg_attrs,
+		.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
+		.sync_single_for_device = octeon_dma_sync_single_for_device,
+		.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
+		.sync_sg_for_device = octeon_dma_sync_sg_for_device,
+		.mapping_error = swiotlb_dma_mapping_error,
+		.dma_supported = swiotlb_dma_supported
+	},
+};
+
+struct dma_map_ops *octeon_pci_dma_map_ops;
+
+void __init octeon_pci_dma_init(void)
+{
 	switch (octeon_dma_bar_type) {
 	case OCTEON_DMA_BAR_TYPE_PCIE:
-		/* Nothing to do, all mappings are static */
-		goto done;
-
+		_octeon_pci_dma_map_ops.phys_to_dma = octeon_gen1_phys_to_dma;
+		_octeon_pci_dma_map_ops.dma_to_phys = octeon_gen1_dma_to_phys;
+		break;
 	case OCTEON_DMA_BAR_TYPE_BIG:
-#ifdef CONFIG_64BIT
-		/* Nothing to do for addresses using BAR2 */
-		if (dma_addr >= BAR2_PCI_ADDRESS)
-			goto done;
-#endif
-		if (unlikely(dma_addr < (4ul << 10)))
-			panic("dma_unmap_single: Unexpect DMA address 0x%llx\n",
-			      dma_addr);
-		else if (dma_addr < (2ul << 30))
-			/* Nothing to do for addresses using BAR0 */
-			goto done;
-		else if (dma_addr < (2ul << 30) + (128ul << 20))
-			/* Need to unmap, fall through */
-			index = (dma_addr - (2ul << 30)) >> 22;
-		else if (dma_addr <
-			 (4ul << 30) - (OCTEON_PCI_BAR1_HOLE_SIZE << 20))
-			goto done;	/* Nothing to do for the rest of BAR1 */
-		else
-			panic("dma_unmap_single: Unexpect DMA address 0x%llx\n",
-			      dma_addr);
-		/* Continued below switch statement */
+		_octeon_pci_dma_map_ops.phys_to_dma = octeon_big_phys_to_dma;
+		_octeon_pci_dma_map_ops.dma_to_phys = octeon_big_dma_to_phys;
 		break;
-
 	case OCTEON_DMA_BAR_TYPE_SMALL:
-#ifdef CONFIG_64BIT
-		/* Nothing to do for addresses using BAR2 */
-		if (dma_addr >= BAR2_PCI_ADDRESS)
-			goto done;
-#endif
-		index = dma_addr >> 22;
-		/* Continued below switch statement */
+		_octeon_pci_dma_map_ops.phys_to_dma = octeon_small_phys_to_dma;
+		_octeon_pci_dma_map_ops.dma_to_phys = octeon_small_dma_to_phys;
 		break;
-
 	default:
-		panic("dma_unmap_single: Invalid octeon_dma_bar_type\n");
+		BUG();
 	}
-
-	if (unlikely(index > 31))
-		panic("dma_unmap_single: "
-		      "Attempt to unmap an invalid address (0x%llx)\n",
-		      dma_addr);
-
-	raw_spin_lock_irqsave(&bar1_lock, flags);
-	bar1_state[index].ref_count--;
-	if (bar1_state[index].ref_count == 0)
-		octeon_npi_write32(CVMX_NPI_PCI_BAR1_INDEXX(index), 0);
-	else if (unlikely(bar1_state[index].ref_count < 0))
-		panic("dma_unmap_single: Bar1[%u] reference count < 0\n",
-		      (int) index);
-	raw_spin_unlock_irqrestore(&bar1_lock, flags);
-done:
-	pr_debug("dma_unmap_single 0x%llx\n", dma_addr);
-	return;
-#endif
+	octeon_pci_dma_map_ops = &_octeon_pci_dma_map_ops.dma_map_ops;
 }
+#endif /* CONFIG_PCI */
diff --git a/arch/mips/cavium-octeon/executive/cvmx-l2c.c b/arch/mips/cavium-octeon/executive/cvmx-l2c.c
index 6abe56f1e097..d38246e33ddb 100644
--- a/arch/mips/cavium-octeon/executive/cvmx-l2c.c
+++ b/arch/mips/cavium-octeon/executive/cvmx-l2c.c
@@ -4,7 +4,7 @@
  * Contact: support@caviumnetworks.com
  * This file is part of the OCTEON SDK
  *
- * Copyright (c) 2003-2008 Cavium Networks
+ * Copyright (c) 2003-2010 Cavium Networks
  *
  * This file is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License, Version 2, as
@@ -26,8 +26,8 @@
  ***********************license end**************************************/
 
 /*
- * Implementation of the Level 2 Cache (L2C) control, measurement, and
- * debugging facilities.
+ * Implementation of the Level 2 Cache (L2C) control,
+ * measurement, and debugging facilities.
  */
 
 #include <asm/octeon/cvmx.h>
@@ -42,13 +42,7 @@
  * if multiple applications or operating systems are running, then it
  * is up to the user program to coordinate between them.
  */
-static cvmx_spinlock_t cvmx_l2c_spinlock;
-
-static inline int l2_size_half(void)
-{
-	uint64_t val = cvmx_read_csr(CVMX_L2D_FUS3);
-	return !!(val & (1ull << 34));
-}
+cvmx_spinlock_t cvmx_l2c_spinlock;
 
 int cvmx_l2c_get_core_way_partition(uint32_t core)
 {
@@ -58,6 +52,9 @@ int cvmx_l2c_get_core_way_partition(uint32_t core)
 	if (core >= cvmx_octeon_num_cores())
 		return -1;
 
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX))
+		return cvmx_read_csr(CVMX_L2C_WPAR_PPX(core)) & 0xffff;
+
 	/*
 	 * Use the lower two bits of the coreNumber to determine the
 	 * bit offset of the UMSK[] field in the L2C_SPAR register.
@@ -71,17 +68,13 @@ int cvmx_l2c_get_core_way_partition(uint32_t core)
 
 	switch (core & 0xC) {
 	case 0x0:
-		return (cvmx_read_csr(CVMX_L2C_SPAR0) & (0xFF << field)) >>
-			field;
+		return (cvmx_read_csr(CVMX_L2C_SPAR0) & (0xFF << field)) >> field;
 	case 0x4:
-		return (cvmx_read_csr(CVMX_L2C_SPAR1) & (0xFF << field)) >>
-			field;
+		return (cvmx_read_csr(CVMX_L2C_SPAR1) & (0xFF << field)) >> field;
 	case 0x8:
-		return (cvmx_read_csr(CVMX_L2C_SPAR2) & (0xFF << field)) >>
-			field;
+		return (cvmx_read_csr(CVMX_L2C_SPAR2) & (0xFF << field)) >> field;
 	case 0xC:
-		return (cvmx_read_csr(CVMX_L2C_SPAR3) & (0xFF << field)) >>
-			field;
+		return (cvmx_read_csr(CVMX_L2C_SPAR3) & (0xFF << field)) >> field;
 	}
 	return 0;
 }
@@ -95,48 +88,50 @@ int cvmx_l2c_set_core_way_partition(uint32_t core, uint32_t mask)
 
 	mask &= valid_mask;
 
-	/* A UMSK setting which blocks all L2C Ways is an error. */
-	if (mask == valid_mask)
+	/* A UMSK setting which blocks all L2C Ways is an error on some chips */
+	if (mask == valid_mask && !OCTEON_IS_MODEL(OCTEON_CN63XX))
 		return -1;
 
 	/* Validate the core number */
 	if (core >= cvmx_octeon_num_cores())
 		return -1;
 
-	/* Check to make sure current mask & new mask don't block all ways */
-	if (((mask | cvmx_l2c_get_core_way_partition(core)) & valid_mask) ==
-	    valid_mask)
-		return -1;
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+		cvmx_write_csr(CVMX_L2C_WPAR_PPX(core), mask);
+		return 0;
+	}
 
-	/* Use the lower two bits of core to determine the bit offset of the
+	/*
+	 * Use the lower two bits of core to determine the bit offset of the
 	 * UMSK[] field in the L2C_SPAR register.
 	 */
 	field = (core & 0x3) * 8;
 
-	/* Assign the new mask setting to the UMSK[] field in the appropriate
+	/*
+	 * Assign the new mask setting to the UMSK[] field in the appropriate
 	 * L2C_SPAR register based on the core_num.
 	 *
 	 */
 	switch (core & 0xC) {
 	case 0x0:
 		cvmx_write_csr(CVMX_L2C_SPAR0,
-			       (cvmx_read_csr(CVMX_L2C_SPAR0) &
-				~(0xFF << field)) | mask << field);
+			       (cvmx_read_csr(CVMX_L2C_SPAR0) & ~(0xFF << field)) |
+			       mask << field);
 		break;
 	case 0x4:
 		cvmx_write_csr(CVMX_L2C_SPAR1,
-			       (cvmx_read_csr(CVMX_L2C_SPAR1) &
-				~(0xFF << field)) | mask << field);
+			       (cvmx_read_csr(CVMX_L2C_SPAR1) & ~(0xFF << field)) |
+			       mask << field);
 		break;
 	case 0x8:
 		cvmx_write_csr(CVMX_L2C_SPAR2,
-			       (cvmx_read_csr(CVMX_L2C_SPAR2) &
-				~(0xFF << field)) | mask << field);
+			       (cvmx_read_csr(CVMX_L2C_SPAR2) & ~(0xFF << field)) |
+			       mask << field);
 		break;
 	case 0xC:
 		cvmx_write_csr(CVMX_L2C_SPAR3,
-			       (cvmx_read_csr(CVMX_L2C_SPAR3) &
-				~(0xFF << field)) | mask << field);
+			       (cvmx_read_csr(CVMX_L2C_SPAR3) & ~(0xFF << field)) |
+			       mask << field);
 		break;
 	}
 	return 0;
@@ -146,84 +141,137 @@ int cvmx_l2c_set_hw_way_partition(uint32_t mask)
 {
 	uint32_t valid_mask;
 
-	valid_mask = 0xff;
-
-	if (OCTEON_IS_MODEL(OCTEON_CN58XX) || OCTEON_IS_MODEL(OCTEON_CN38XX)) {
-		if (l2_size_half())
-			valid_mask = 0xf;
-	} else if (l2_size_half())
-		valid_mask = 0x3;
-
+	valid_mask = (0x1 << cvmx_l2c_get_num_assoc()) - 1;
 	mask &= valid_mask;
 
-	/* A UMSK setting which blocks all L2C Ways is an error. */
-	if (mask == valid_mask)
-		return -1;
-	/* Check to make sure current mask & new mask don't block all ways */
-	if (((mask | cvmx_l2c_get_hw_way_partition()) & valid_mask) ==
-	    valid_mask)
+	/* A UMSK setting which blocks all L2C Ways is an error on some chips */
+	if (mask == valid_mask  && !OCTEON_IS_MODEL(OCTEON_CN63XX))
 		return -1;
 
-	cvmx_write_csr(CVMX_L2C_SPAR4,
-		       (cvmx_read_csr(CVMX_L2C_SPAR4) & ~0xFF) | mask);
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX))
+		cvmx_write_csr(CVMX_L2C_WPAR_IOBX(0), mask);
+	else
+		cvmx_write_csr(CVMX_L2C_SPAR4,
+			       (cvmx_read_csr(CVMX_L2C_SPAR4) & ~0xFF) | mask);
 	return 0;
 }
 
 int cvmx_l2c_get_hw_way_partition(void)
 {
-	return cvmx_read_csr(CVMX_L2C_SPAR4) & (0xFF);
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX))
+		return cvmx_read_csr(CVMX_L2C_WPAR_IOBX(0)) & 0xffff;
+	else
+		return cvmx_read_csr(CVMX_L2C_SPAR4) & (0xFF);
 }
 
 void cvmx_l2c_config_perf(uint32_t counter, enum cvmx_l2c_event event,
 			  uint32_t clear_on_read)
 {
-	union cvmx_l2c_pfctl pfctl;
+	if (OCTEON_IS_MODEL(OCTEON_CN5XXX) || OCTEON_IS_MODEL(OCTEON_CN3XXX)) {
+		union cvmx_l2c_pfctl pfctl;
 
-	pfctl.u64 = cvmx_read_csr(CVMX_L2C_PFCTL);
+		pfctl.u64 = cvmx_read_csr(CVMX_L2C_PFCTL);
 
-	switch (counter) {
-	case 0:
-		pfctl.s.cnt0sel = event;
-		pfctl.s.cnt0ena = 1;
-		if (!cvmx_octeon_is_pass1())
+		switch (counter) {
+		case 0:
+			pfctl.s.cnt0sel = event;
+			pfctl.s.cnt0ena = 1;
 			pfctl.s.cnt0rdclr = clear_on_read;
-		break;
-	case 1:
-		pfctl.s.cnt1sel = event;
-		pfctl.s.cnt1ena = 1;
-		if (!cvmx_octeon_is_pass1())
+			break;
+		case 1:
+			pfctl.s.cnt1sel = event;
+			pfctl.s.cnt1ena = 1;
 			pfctl.s.cnt1rdclr = clear_on_read;
-		break;
-	case 2:
-		pfctl.s.cnt2sel = event;
-		pfctl.s.cnt2ena = 1;
-		if (!cvmx_octeon_is_pass1())
+			break;
+		case 2:
+			pfctl.s.cnt2sel = event;
+			pfctl.s.cnt2ena = 1;
 			pfctl.s.cnt2rdclr = clear_on_read;
-		break;
-	case 3:
-	default:
-		pfctl.s.cnt3sel = event;
-		pfctl.s.cnt3ena = 1;
-		if (!cvmx_octeon_is_pass1())
+			break;
+		case 3:
+		default:
+			pfctl.s.cnt3sel = event;
+			pfctl.s.cnt3ena = 1;
 			pfctl.s.cnt3rdclr = clear_on_read;
-		break;
-	}
+			break;
+		}
 
-	cvmx_write_csr(CVMX_L2C_PFCTL, pfctl.u64);
+		cvmx_write_csr(CVMX_L2C_PFCTL, pfctl.u64);
+	} else {
+		union cvmx_l2c_tadx_prf l2c_tadx_prf;
+		int tad;
+
+		cvmx_dprintf("L2C performance counter events are different for this chip, mapping 'event' to cvmx_l2c_tad_event_t\n");
+		if (clear_on_read)
+			cvmx_dprintf("L2C counters don't support clear on read for this chip\n");
+
+		l2c_tadx_prf.u64 = cvmx_read_csr(CVMX_L2C_TADX_PRF(0));
+
+		switch (counter) {
+		case 0:
+			l2c_tadx_prf.s.cnt0sel = event;
+			break;
+		case 1:
+			l2c_tadx_prf.s.cnt1sel = event;
+			break;
+		case 2:
+			l2c_tadx_prf.s.cnt2sel = event;
+			break;
+		default:
+		case 3:
+			l2c_tadx_prf.s.cnt3sel = event;
+			break;
+		}
+		for (tad = 0; tad < CVMX_L2C_TADS; tad++)
+			cvmx_write_csr(CVMX_L2C_TADX_PRF(tad),
+				       l2c_tadx_prf.u64);
+	}
 }
 
 uint64_t cvmx_l2c_read_perf(uint32_t counter)
 {
 	switch (counter) {
 	case 0:
-		return cvmx_read_csr(CVMX_L2C_PFC0);
+		if (OCTEON_IS_MODEL(OCTEON_CN5XXX) || OCTEON_IS_MODEL(OCTEON_CN3XXX))
+			return cvmx_read_csr(CVMX_L2C_PFC0);
+		else {
+			uint64_t counter = 0;
+			int tad;
+			for (tad = 0; tad < CVMX_L2C_TADS; tad++)
+				counter += cvmx_read_csr(CVMX_L2C_TADX_PFC0(tad));
+			return counter;
+		}
 	case 1:
-		return cvmx_read_csr(CVMX_L2C_PFC1);
+		if (OCTEON_IS_MODEL(OCTEON_CN5XXX) || OCTEON_IS_MODEL(OCTEON_CN3XXX))
+			return cvmx_read_csr(CVMX_L2C_PFC1);
+		else {
+			uint64_t counter = 0;
+			int tad;
+			for (tad = 0; tad < CVMX_L2C_TADS; tad++)
+				counter += cvmx_read_csr(CVMX_L2C_TADX_PFC1(tad));
+			return counter;
+		}
 	case 2:
-		return cvmx_read_csr(CVMX_L2C_PFC2);
+		if (OCTEON_IS_MODEL(OCTEON_CN5XXX) || OCTEON_IS_MODEL(OCTEON_CN3XXX))
+			return cvmx_read_csr(CVMX_L2C_PFC2);
+		else {
+			uint64_t counter = 0;
+			int tad;
+			for (tad = 0; tad < CVMX_L2C_TADS; tad++)
+				counter += cvmx_read_csr(CVMX_L2C_TADX_PFC2(tad));
+			return counter;
+		}
 	case 3:
 	default:
-		return cvmx_read_csr(CVMX_L2C_PFC3);
+		if (OCTEON_IS_MODEL(OCTEON_CN5XXX) || OCTEON_IS_MODEL(OCTEON_CN3XXX))
+			return cvmx_read_csr(CVMX_L2C_PFC3);
+		else {
+			uint64_t counter = 0;
+			int tad;
+			for (tad = 0; tad < CVMX_L2C_TADS; tad++)
+				counter += cvmx_read_csr(CVMX_L2C_TADX_PFC3(tad));
+			return counter;
+		}
 	}
 }
 
@@ -240,7 +288,7 @@ static void fault_in(uint64_t addr, int len)
 	volatile char dummy;
 	/*
 	 * Adjust addr and length so we get all cache lines even for
-	 * small ranges spanning two cache lines
+	 * small ranges spanning two cache lines.
 	 */
 	len += addr & CVMX_CACHE_LINE_MASK;
 	addr &= ~CVMX_CACHE_LINE_MASK;
@@ -259,67 +307,100 @@ static void fault_in(uint64_t addr, int len)
 
 int cvmx_l2c_lock_line(uint64_t addr)
 {
-	int retval = 0;
-	union cvmx_l2c_dbg l2cdbg;
-	union cvmx_l2c_lckbase lckbase;
-	union cvmx_l2c_lckoff lckoff;
-	union cvmx_l2t_err l2t_err;
-	l2cdbg.u64 = 0;
-	lckbase.u64 = 0;
-	lckoff.u64 = 0;
-
-	cvmx_spinlock_lock(&cvmx_l2c_spinlock);
-
-	/* Clear l2t error bits if set */
-	l2t_err.u64 = cvmx_read_csr(CVMX_L2T_ERR);
-	l2t_err.s.lckerr = 1;
-	l2t_err.s.lckerr2 = 1;
-	cvmx_write_csr(CVMX_L2T_ERR, l2t_err.u64);
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+		int shift = CVMX_L2C_TAG_ADDR_ALIAS_SHIFT;
+		uint64_t assoc = cvmx_l2c_get_num_assoc();
+		uint64_t tag = addr >> shift;
+		uint64_t index = CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS, cvmx_l2c_address_to_index(addr) << CVMX_L2C_IDX_ADDR_SHIFT);
+		uint64_t way;
+		union cvmx_l2c_tadx_tag l2c_tadx_tag;
+
+		CVMX_CACHE_LCKL2(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS, addr), 0);
+
+		/* Make sure we were able to lock the line */
+		for (way = 0; way < assoc; way++) {
+			CVMX_CACHE_LTGL2I(index | (way << shift), 0);
+			/* make sure CVMX_L2C_TADX_TAG is updated */
+			CVMX_SYNC;
+			l2c_tadx_tag.u64 = cvmx_read_csr(CVMX_L2C_TADX_TAG(0));
+			if (l2c_tadx_tag.s.valid && l2c_tadx_tag.s.tag == tag)
+				break;
+		}
 
-	addr &= ~CVMX_CACHE_LINE_MASK;
+		/* Check if a valid line is found */
+		if (way >= assoc) {
+			/* cvmx_dprintf("ERROR: cvmx_l2c_lock_line: line not found for locking at 0x%llx address\n", (unsigned long long)addr); */
+			return -1;
+		}
 
-	/* Set this core as debug core */
-	l2cdbg.s.ppnum = cvmx_get_core_num();
-	CVMX_SYNC;
-	cvmx_write_csr(CVMX_L2C_DBG, l2cdbg.u64);
-	cvmx_read_csr(CVMX_L2C_DBG);
-
-	lckoff.s.lck_offset = 0;	/* Only lock 1 line at a time */
-	cvmx_write_csr(CVMX_L2C_LCKOFF, lckoff.u64);
-	cvmx_read_csr(CVMX_L2C_LCKOFF);
-
-	if (((union cvmx_l2c_cfg) (cvmx_read_csr(CVMX_L2C_CFG))).s.idxalias) {
-		int alias_shift =
-		    CVMX_L2C_IDX_ADDR_SHIFT + 2 * CVMX_L2_SET_BITS - 1;
-		uint64_t addr_tmp =
-		    addr ^ (addr & ((1 << alias_shift) - 1)) >>
-		    CVMX_L2_SET_BITS;
-		lckbase.s.lck_base = addr_tmp >> 7;
+		/* Check if lock bit is not set */
+		if (!l2c_tadx_tag.s.lock) {
+			/* cvmx_dprintf("ERROR: cvmx_l2c_lock_line: Not able to lock at 0x%llx address\n", (unsigned long long)addr); */
+			return -1;
+		}
+		return way;
 	} else {
-		lckbase.s.lck_base = addr >> 7;
-	}
+		int retval = 0;
+		union cvmx_l2c_dbg l2cdbg;
+		union cvmx_l2c_lckbase lckbase;
+		union cvmx_l2c_lckoff lckoff;
+		union cvmx_l2t_err l2t_err;
 
-	lckbase.s.lck_ena = 1;
-	cvmx_write_csr(CVMX_L2C_LCKBASE, lckbase.u64);
-	cvmx_read_csr(CVMX_L2C_LCKBASE);	/* Make sure it gets there */
+		cvmx_spinlock_lock(&cvmx_l2c_spinlock);
 
-	fault_in(addr, CVMX_CACHE_LINE_SIZE);
+		l2cdbg.u64 = 0;
+		lckbase.u64 = 0;
+		lckoff.u64 = 0;
 
-	lckbase.s.lck_ena = 0;
-	cvmx_write_csr(CVMX_L2C_LCKBASE, lckbase.u64);
-	cvmx_read_csr(CVMX_L2C_LCKBASE);	/* Make sure it gets there */
+		/* Clear l2t error bits if set */
+		l2t_err.u64 = cvmx_read_csr(CVMX_L2T_ERR);
+		l2t_err.s.lckerr = 1;
+		l2t_err.s.lckerr2 = 1;
+		cvmx_write_csr(CVMX_L2T_ERR, l2t_err.u64);
 
-	/* Stop being debug core */
-	cvmx_write_csr(CVMX_L2C_DBG, 0);
-	cvmx_read_csr(CVMX_L2C_DBG);
+		addr &= ~CVMX_CACHE_LINE_MASK;
 
-	l2t_err.u64 = cvmx_read_csr(CVMX_L2T_ERR);
-	if (l2t_err.s.lckerr || l2t_err.s.lckerr2)
-		retval = 1;	/* We were unable to lock the line */
+		/* Set this core as debug core */
+		l2cdbg.s.ppnum = cvmx_get_core_num();
+		CVMX_SYNC;
+		cvmx_write_csr(CVMX_L2C_DBG, l2cdbg.u64);
+		cvmx_read_csr(CVMX_L2C_DBG);
+
+		lckoff.s.lck_offset = 0; /* Only lock 1 line at a time */
+		cvmx_write_csr(CVMX_L2C_LCKOFF, lckoff.u64);
+		cvmx_read_csr(CVMX_L2C_LCKOFF);
+
+		if (((union cvmx_l2c_cfg)(cvmx_read_csr(CVMX_L2C_CFG))).s.idxalias) {
+			int alias_shift = CVMX_L2C_IDX_ADDR_SHIFT + 2 * CVMX_L2_SET_BITS - 1;
+			uint64_t addr_tmp = addr ^ (addr & ((1 << alias_shift) - 1)) >> CVMX_L2_SET_BITS;
+			lckbase.s.lck_base = addr_tmp >> 7;
+		} else {
+			lckbase.s.lck_base = addr >> 7;
+		}
 
-	cvmx_spinlock_unlock(&cvmx_l2c_spinlock);
+		lckbase.s.lck_ena = 1;
+		cvmx_write_csr(CVMX_L2C_LCKBASE, lckbase.u64);
+		/* Make sure it gets there */
+		cvmx_read_csr(CVMX_L2C_LCKBASE);
 
-	return retval;
+		fault_in(addr, CVMX_CACHE_LINE_SIZE);
+
+		lckbase.s.lck_ena = 0;
+		cvmx_write_csr(CVMX_L2C_LCKBASE, lckbase.u64);
+		/* Make sure it gets there */
+		cvmx_read_csr(CVMX_L2C_LCKBASE);
+
+		/* Stop being debug core */
+		cvmx_write_csr(CVMX_L2C_DBG, 0);
+		cvmx_read_csr(CVMX_L2C_DBG);
+
+		l2t_err.u64 = cvmx_read_csr(CVMX_L2T_ERR);
+		if (l2t_err.s.lckerr || l2t_err.s.lckerr2)
+			retval = 1;  /* We were unable to lock the line */
+
+		cvmx_spinlock_unlock(&cvmx_l2c_spinlock);
+		return retval;
+	}
 }
 
 int cvmx_l2c_lock_mem_region(uint64_t start, uint64_t len)
@@ -336,7 +417,6 @@ int cvmx_l2c_lock_mem_region(uint64_t start, uint64_t len)
 		start += CVMX_CACHE_LINE_SIZE;
 		len -= CVMX_CACHE_LINE_SIZE;
 	}
-
 	return retval;
 }
 
@@ -344,80 +424,73 @@ void cvmx_l2c_flush(void)
 {
 	uint64_t assoc, set;
 	uint64_t n_assoc, n_set;
-	union cvmx_l2c_dbg l2cdbg;
-
-	cvmx_spinlock_lock(&cvmx_l2c_spinlock);
 
-	l2cdbg.u64 = 0;
-	if (!OCTEON_IS_MODEL(OCTEON_CN30XX))
-		l2cdbg.s.ppnum = cvmx_get_core_num();
-	l2cdbg.s.finv = 1;
-	n_set = CVMX_L2_SETS;
-	n_assoc = l2_size_half() ? (CVMX_L2_ASSOC / 2) : CVMX_L2_ASSOC;
-	for (set = 0; set < n_set; set++) {
-		for (assoc = 0; assoc < n_assoc; assoc++) {
-			l2cdbg.s.set = assoc;
-			/* Enter debug mode, and make sure all other
-			 ** writes complete before we enter debug
-			 ** mode */
-			CVMX_SYNCW;
-			cvmx_write_csr(CVMX_L2C_DBG, l2cdbg.u64);
-			cvmx_read_csr(CVMX_L2C_DBG);
-
-			CVMX_PREPARE_FOR_STORE(CVMX_ADD_SEG
-					       (CVMX_MIPS_SPACE_XKPHYS,
-						set * CVMX_CACHE_LINE_SIZE), 0);
-			CVMX_SYNCW;	/* Push STF out to L2 */
-			/* Exit debug mode */
-			CVMX_SYNC;
-			cvmx_write_csr(CVMX_L2C_DBG, 0);
-			cvmx_read_csr(CVMX_L2C_DBG);
+	n_set = cvmx_l2c_get_num_sets();
+	n_assoc = cvmx_l2c_get_num_assoc();
+
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+		uint64_t address;
+		/* These may look like constants, but they aren't... */
+		int assoc_shift = CVMX_L2C_TAG_ADDR_ALIAS_SHIFT;
+		int set_shift = CVMX_L2C_IDX_ADDR_SHIFT;
+		for (set = 0; set < n_set; set++) {
+			for (assoc = 0; assoc < n_assoc; assoc++) {
+				address = CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						       (assoc << assoc_shift) |	(set << set_shift));
+				CVMX_CACHE_WBIL2I(address, 0);
+			}
 		}
+	} else {
+		for (set = 0; set < n_set; set++)
+			for (assoc = 0; assoc < n_assoc; assoc++)
+				cvmx_l2c_flush_line(assoc, set);
 	}
-
-	cvmx_spinlock_unlock(&cvmx_l2c_spinlock);
 }
 
+
 int cvmx_l2c_unlock_line(uint64_t address)
 {
-	int assoc;
-	union cvmx_l2c_tag tag;
-	union cvmx_l2c_dbg l2cdbg;
-	uint32_t tag_addr;
 
-	uint32_t index = cvmx_l2c_address_to_index(address);
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+		int assoc;
+		union cvmx_l2c_tag tag;
+		uint32_t tag_addr;
+		uint32_t index = cvmx_l2c_address_to_index(address);
+
+		tag_addr = ((address >> CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) & ((1 << CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) - 1));
+
+		/*
+		 * For 63XX, we can flush a line by using the physical
+		 * address directly, so finding the cache line used by
+		 * the address is only required to provide the proper
+		 * return value for the function.
+		 */
+		for (assoc = 0; assoc < CVMX_L2_ASSOC; assoc++) {
+			tag = cvmx_l2c_get_tag(assoc, index);
+
+			if (tag.s.V && (tag.s.addr == tag_addr)) {
+				CVMX_CACHE_WBIL2(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS, address), 0);
+				return tag.s.L;
+			}
+		}
+	} else {
+		int assoc;
+		union cvmx_l2c_tag tag;
+		uint32_t tag_addr;
 
-	cvmx_spinlock_lock(&cvmx_l2c_spinlock);
-	/* Compute portion of address that is stored in tag */
-	tag_addr =
-	    ((address >> CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) &
-	     ((1 << CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) - 1));
-	for (assoc = 0; assoc < CVMX_L2_ASSOC; assoc++) {
-		tag = cvmx_get_l2c_tag(assoc, index);
+		uint32_t index = cvmx_l2c_address_to_index(address);
 
-		if (tag.s.V && (tag.s.addr == tag_addr)) {
-			l2cdbg.u64 = 0;
-			l2cdbg.s.ppnum = cvmx_get_core_num();
-			l2cdbg.s.set = assoc;
-			l2cdbg.s.finv = 1;
+		/* Compute portion of address that is stored in tag */
+		tag_addr = ((address >> CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) & ((1 << CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) - 1));
+		for (assoc = 0; assoc < CVMX_L2_ASSOC; assoc++) {
+			tag = cvmx_l2c_get_tag(assoc, index);
 
-			CVMX_SYNC;
-			/* Enter debug mode */
-			cvmx_write_csr(CVMX_L2C_DBG, l2cdbg.u64);
-			cvmx_read_csr(CVMX_L2C_DBG);
-
-			CVMX_PREPARE_FOR_STORE(CVMX_ADD_SEG
-					       (CVMX_MIPS_SPACE_XKPHYS,
-						address), 0);
-			CVMX_SYNC;
-			/* Exit debug mode */
-			cvmx_write_csr(CVMX_L2C_DBG, 0);
-			cvmx_read_csr(CVMX_L2C_DBG);
-			cvmx_spinlock_unlock(&cvmx_l2c_spinlock);
-			return tag.s.L;
+			if (tag.s.V && (tag.s.addr == tag_addr)) {
+				cvmx_l2c_flush_line(assoc, index);
+				return tag.s.L;
+			}
 		}
 	}
-	cvmx_spinlock_unlock(&cvmx_l2c_spinlock);
 	return 0;
 }
 
@@ -445,48 +518,49 @@ union __cvmx_l2c_tag {
 	uint64_t u64;
 	struct cvmx_l2c_tag_cn50xx {
 		uint64_t reserved:40;
-		uint64_t V:1;	/* Line valid */
-		uint64_t D:1;	/* Line dirty */
-		uint64_t L:1;	/* Line locked */
-		uint64_t U:1;	/* Use, LRU eviction */
+		uint64_t V:1;		/* Line valid */
+		uint64_t D:1;		/* Line dirty */
+		uint64_t L:1;		/* Line locked */
+		uint64_t U:1;		/* Use, LRU eviction */
 		uint64_t addr:20;	/* Phys mem addr (33..14) */
 	} cn50xx;
 	struct cvmx_l2c_tag_cn30xx {
 		uint64_t reserved:41;
-		uint64_t V:1;	/* Line valid */
-		uint64_t D:1;	/* Line dirty */
-		uint64_t L:1;	/* Line locked */
-		uint64_t U:1;	/* Use, LRU eviction */
+		uint64_t V:1;		/* Line valid */
+		uint64_t D:1;		/* Line dirty */
+		uint64_t L:1;		/* Line locked */
+		uint64_t U:1;		/* Use, LRU eviction */
 		uint64_t addr:19;	/* Phys mem addr (33..15) */
 	} cn30xx;
 	struct cvmx_l2c_tag_cn31xx {
 		uint64_t reserved:42;
-		uint64_t V:1;	/* Line valid */
-		uint64_t D:1;	/* Line dirty */
-		uint64_t L:1;	/* Line locked */
-		uint64_t U:1;	/* Use, LRU eviction */
+		uint64_t V:1;		/* Line valid */
+		uint64_t D:1;		/* Line dirty */
+		uint64_t L:1;		/* Line locked */
+		uint64_t U:1;		/* Use, LRU eviction */
 		uint64_t addr:18;	/* Phys mem addr (33..16) */
 	} cn31xx;
 	struct cvmx_l2c_tag_cn38xx {
 		uint64_t reserved:43;
-		uint64_t V:1;	/* Line valid */
-		uint64_t D:1;	/* Line dirty */
-		uint64_t L:1;	/* Line locked */
-		uint64_t U:1;	/* Use, LRU eviction */
+		uint64_t V:1;		/* Line valid */
+		uint64_t D:1;		/* Line dirty */
+		uint64_t L:1;		/* Line locked */
+		uint64_t U:1;		/* Use, LRU eviction */
 		uint64_t addr:17;	/* Phys mem addr (33..17) */
 	} cn38xx;
 	struct cvmx_l2c_tag_cn58xx {
 		uint64_t reserved:44;
-		uint64_t V:1;	/* Line valid */
-		uint64_t D:1;	/* Line dirty */
-		uint64_t L:1;	/* Line locked */
-		uint64_t U:1;	/* Use, LRU eviction */
+		uint64_t V:1;		/* Line valid */
+		uint64_t D:1;		/* Line dirty */
+		uint64_t L:1;		/* Line locked */
+		uint64_t U:1;		/* Use, LRU eviction */
 		uint64_t addr:16;	/* Phys mem addr (33..18) */
 	} cn58xx;
 	struct cvmx_l2c_tag_cn58xx cn56xx;	/* 2048 sets */
 	struct cvmx_l2c_tag_cn31xx cn52xx;	/* 512 sets */
 };
 
+
 /**
  * @INTERNAL
  * Function to read a L2C tag.  This code make the current core
@@ -503,7 +577,7 @@ union __cvmx_l2c_tag {
 static union __cvmx_l2c_tag __read_l2_tag(uint64_t assoc, uint64_t index)
 {
 
-	uint64_t debug_tag_addr = (((1ULL << 63) | (index << 7)) + 96);
+	uint64_t debug_tag_addr = CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS, (index << 7) + 96);
 	uint64_t core = cvmx_get_core_num();
 	union __cvmx_l2c_tag tag_val;
 	uint64_t dbg_addr = CVMX_L2C_DBG;
@@ -512,12 +586,15 @@ static union __cvmx_l2c_tag __read_l2_tag(uint64_t assoc, uint64_t index)
 	union cvmx_l2c_dbg debug_val;
 	debug_val.u64 = 0;
 	/*
-	 * For low core count parts, the core number is always small enough
-	 * to stay in the correct field and not set any reserved bits.
+	 * For low core count parts, the core number is always small
+	 * enough to stay in the correct field and not set any
+	 * reserved bits.
 	 */
 	debug_val.s.ppnum = core;
 	debug_val.s.l2t = 1;
 	debug_val.s.set = assoc;
+
+	local_irq_save(flags);
 	/*
 	 * Make sure core is quiet (no prefetches, etc.) before
 	 * entering debug mode.
@@ -526,112 +603,139 @@ static union __cvmx_l2c_tag __read_l2_tag(uint64_t assoc, uint64_t index)
 	/* Flush L1 to make sure debug load misses L1 */
 	CVMX_DCACHE_INVALIDATE;
 
-	local_irq_save(flags);
-
 	/*
 	 * The following must be done in assembly as when in debug
 	 * mode all data loads from L2 return special debug data, not
-	 * normal memory contents.  Also, interrupts must be
-	 * disabled, since if an interrupt occurs while in debug mode
-	 * the ISR will get debug data from all its memory reads
-	 * instead of the contents of memory
+	 * normal memory contents.  Also, interrupts must be disabled,
+	 * since if an interrupt occurs while in debug mode the ISR
+	 * will get debug data from all its memory * reads instead of
+	 * the contents of memory.
 	 */
 
-	asm volatile (".set push              \n"
-		"        .set mips64              \n"
-		"        .set noreorder           \n"
-		/* Enter debug mode, wait for store */
-		"        sd    %[dbg_val], 0(%[dbg_addr])  \n"
-		"        ld    $0, 0(%[dbg_addr]) \n"
-		/* Read L2C tag data */
-		"        ld    %[tag_val], 0(%[tag_addr]) \n"
-		/* Exit debug mode, wait for store */
-		"        sd    $0, 0(%[dbg_addr])  \n"
-		"        ld    $0, 0(%[dbg_addr]) \n"
-		/* Invalidate dcache to discard debug data */
-		"        cache 9, 0($0) \n"
-		"        .set pop" :
-		[tag_val] "=r"(tag_val.u64) : [dbg_addr] "r"(dbg_addr),
-		[dbg_val] "r"(debug_val.u64),
-		[tag_addr] "r"(debug_tag_addr) : "memory");
+	asm volatile (
+		".set push\n\t"
+		".set mips64\n\t"
+		".set noreorder\n\t"
+		"sd    %[dbg_val], 0(%[dbg_addr])\n\t"   /* Enter debug mode, wait for store */
+		"ld    $0, 0(%[dbg_addr])\n\t"
+		"ld    %[tag_val], 0(%[tag_addr])\n\t"   /* Read L2C tag data */
+		"sd    $0, 0(%[dbg_addr])\n\t"          /* Exit debug mode, wait for store */
+		"ld    $0, 0(%[dbg_addr])\n\t"
+		"cache 9, 0($0)\n\t"             /* Invalidate dcache to discard debug data */
+		".set pop"
+		: [tag_val] "=r" (tag_val)
+		: [dbg_addr] "r" (dbg_addr), [dbg_val] "r" (debug_val), [tag_addr] "r" (debug_tag_addr)
+		: "memory");
 
 	local_irq_restore(flags);
-	return tag_val;
 
+	return tag_val;
 }
 
+
 union cvmx_l2c_tag cvmx_l2c_get_tag(uint32_t association, uint32_t index)
 {
-	union __cvmx_l2c_tag tmp_tag;
 	union cvmx_l2c_tag tag;
 	tag.u64 = 0;
 
 	if ((int)association >= cvmx_l2c_get_num_assoc()) {
-		cvmx_dprintf
-		    ("ERROR: cvmx_get_l2c_tag association out of range\n");
+		cvmx_dprintf("ERROR: cvmx_l2c_get_tag association out of range\n");
 		return tag;
 	}
 	if ((int)index >= cvmx_l2c_get_num_sets()) {
-		cvmx_dprintf("ERROR: cvmx_get_l2c_tag "
-			     "index out of range (arg: %d, max: %d\n",
-		     index, cvmx_l2c_get_num_sets());
+		cvmx_dprintf("ERROR: cvmx_l2c_get_tag index out of range (arg: %d, max: %d)\n",
+			     (int)index, cvmx_l2c_get_num_sets());
 		return tag;
 	}
-	/* __read_l2_tag is intended for internal use only */
-	tmp_tag = __read_l2_tag(association, index);
-
-	/*
-	 * Convert all tag structure types to generic version, as it
-	 * can represent all models.
-	 */
-	if (OCTEON_IS_MODEL(OCTEON_CN58XX) || OCTEON_IS_MODEL(OCTEON_CN56XX)) {
-		tag.s.V = tmp_tag.cn58xx.V;
-		tag.s.D = tmp_tag.cn58xx.D;
-		tag.s.L = tmp_tag.cn58xx.L;
-		tag.s.U = tmp_tag.cn58xx.U;
-		tag.s.addr = tmp_tag.cn58xx.addr;
-	} else if (OCTEON_IS_MODEL(OCTEON_CN38XX)) {
-		tag.s.V = tmp_tag.cn38xx.V;
-		tag.s.D = tmp_tag.cn38xx.D;
-		tag.s.L = tmp_tag.cn38xx.L;
-		tag.s.U = tmp_tag.cn38xx.U;
-		tag.s.addr = tmp_tag.cn38xx.addr;
-	} else if (OCTEON_IS_MODEL(OCTEON_CN31XX)
-		   || OCTEON_IS_MODEL(OCTEON_CN52XX)) {
-		tag.s.V = tmp_tag.cn31xx.V;
-		tag.s.D = tmp_tag.cn31xx.D;
-		tag.s.L = tmp_tag.cn31xx.L;
-		tag.s.U = tmp_tag.cn31xx.U;
-		tag.s.addr = tmp_tag.cn31xx.addr;
-	} else if (OCTEON_IS_MODEL(OCTEON_CN30XX)) {
-		tag.s.V = tmp_tag.cn30xx.V;
-		tag.s.D = tmp_tag.cn30xx.D;
-		tag.s.L = tmp_tag.cn30xx.L;
-		tag.s.U = tmp_tag.cn30xx.U;
-		tag.s.addr = tmp_tag.cn30xx.addr;
-	} else if (OCTEON_IS_MODEL(OCTEON_CN50XX)) {
-		tag.s.V = tmp_tag.cn50xx.V;
-		tag.s.D = tmp_tag.cn50xx.D;
-		tag.s.L = tmp_tag.cn50xx.L;
-		tag.s.U = tmp_tag.cn50xx.U;
-		tag.s.addr = tmp_tag.cn50xx.addr;
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+		union cvmx_l2c_tadx_tag l2c_tadx_tag;
+		uint64_t address = CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						(association << CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) |
+						(index << CVMX_L2C_IDX_ADDR_SHIFT));
+		/*
+		 * Use L2 cache Index load tag cache instruction, as
+		 * hardware loads the virtual tag for the L2 cache
+		 * block with the contents of L2C_TAD0_TAG
+		 * register.
+		 */
+		CVMX_CACHE_LTGL2I(address, 0);
+		CVMX_SYNC;   /* make sure CVMX_L2C_TADX_TAG is updated */
+		l2c_tadx_tag.u64 = cvmx_read_csr(CVMX_L2C_TADX_TAG(0));
+
+		tag.s.V     = l2c_tadx_tag.s.valid;
+		tag.s.D     = l2c_tadx_tag.s.dirty;
+		tag.s.L     = l2c_tadx_tag.s.lock;
+		tag.s.U     = l2c_tadx_tag.s.use;
+		tag.s.addr  = l2c_tadx_tag.s.tag;
 	} else {
-		cvmx_dprintf("Unsupported OCTEON Model in %s\n", __func__);
+		union __cvmx_l2c_tag tmp_tag;
+		/* __read_l2_tag is intended for internal use only */
+		tmp_tag = __read_l2_tag(association, index);
+
+		/*
+		 * Convert all tag structure types to generic version,
+		 * as it can represent all models.
+		 */
+		if (OCTEON_IS_MODEL(OCTEON_CN58XX) || OCTEON_IS_MODEL(OCTEON_CN56XX)) {
+			tag.s.V    = tmp_tag.cn58xx.V;
+			tag.s.D    = tmp_tag.cn58xx.D;
+			tag.s.L    = tmp_tag.cn58xx.L;
+			tag.s.U    = tmp_tag.cn58xx.U;
+			tag.s.addr = tmp_tag.cn58xx.addr;
+		} else if (OCTEON_IS_MODEL(OCTEON_CN38XX)) {
+			tag.s.V    = tmp_tag.cn38xx.V;
+			tag.s.D    = tmp_tag.cn38xx.D;
+			tag.s.L    = tmp_tag.cn38xx.L;
+			tag.s.U    = tmp_tag.cn38xx.U;
+			tag.s.addr = tmp_tag.cn38xx.addr;
+		} else if (OCTEON_IS_MODEL(OCTEON_CN31XX) || OCTEON_IS_MODEL(OCTEON_CN52XX)) {
+			tag.s.V    = tmp_tag.cn31xx.V;
+			tag.s.D    = tmp_tag.cn31xx.D;
+			tag.s.L    = tmp_tag.cn31xx.L;
+			tag.s.U    = tmp_tag.cn31xx.U;
+			tag.s.addr = tmp_tag.cn31xx.addr;
+		} else if (OCTEON_IS_MODEL(OCTEON_CN30XX)) {
+			tag.s.V    = tmp_tag.cn30xx.V;
+			tag.s.D    = tmp_tag.cn30xx.D;
+			tag.s.L    = tmp_tag.cn30xx.L;
+			tag.s.U    = tmp_tag.cn30xx.U;
+			tag.s.addr = tmp_tag.cn30xx.addr;
+		} else if (OCTEON_IS_MODEL(OCTEON_CN50XX)) {
+			tag.s.V    = tmp_tag.cn50xx.V;
+			tag.s.D    = tmp_tag.cn50xx.D;
+			tag.s.L    = tmp_tag.cn50xx.L;
+			tag.s.U    = tmp_tag.cn50xx.U;
+			tag.s.addr = tmp_tag.cn50xx.addr;
+		} else {
+			cvmx_dprintf("Unsupported OCTEON Model in %s\n", __func__);
+		}
 	}
-
 	return tag;
 }
 
 uint32_t cvmx_l2c_address_to_index(uint64_t addr)
 {
 	uint64_t idx = addr >> CVMX_L2C_IDX_ADDR_SHIFT;
-	union cvmx_l2c_cfg l2c_cfg;
-	l2c_cfg.u64 = cvmx_read_csr(CVMX_L2C_CFG);
+	int indxalias = 0;
 
-	if (l2c_cfg.s.idxalias) {
-		idx ^=
-		    ((addr & CVMX_L2C_ALIAS_MASK) >>
-		     CVMX_L2C_TAG_ADDR_ALIAS_SHIFT);
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+		union cvmx_l2c_ctl l2c_ctl;
+		l2c_ctl.u64 = cvmx_read_csr(CVMX_L2C_CTL);
+		indxalias = !l2c_ctl.s.disidxalias;
+	} else {
+		union cvmx_l2c_cfg l2c_cfg;
+		l2c_cfg.u64 = cvmx_read_csr(CVMX_L2C_CFG);
+		indxalias = l2c_cfg.s.idxalias;
+	}
+
+	if (indxalias) {
+		if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+			uint32_t a_14_12 = (idx / (CVMX_L2C_MEMBANK_SELECT_SIZE/(1<<CVMX_L2C_IDX_ADDR_SHIFT))) & 0x7;
+			idx ^= idx / cvmx_l2c_get_num_sets();
+			idx ^= a_14_12;
+		} else {
+			idx ^= ((addr & CVMX_L2C_ALIAS_MASK) >> CVMX_L2C_TAG_ADDR_ALIAS_SHIFT);
+		}
 	}
 	idx &= CVMX_L2C_IDX_MASK;
 	return idx;
@@ -652,10 +756,9 @@ int cvmx_l2c_get_set_bits(void)
 	int l2_set_bits;
 	if (OCTEON_IS_MODEL(OCTEON_CN56XX) || OCTEON_IS_MODEL(OCTEON_CN58XX))
 		l2_set_bits = 11;	/* 2048 sets */
-	else if (OCTEON_IS_MODEL(OCTEON_CN38XX))
+	else if (OCTEON_IS_MODEL(OCTEON_CN38XX) || OCTEON_IS_MODEL(OCTEON_CN63XX))
 		l2_set_bits = 10;	/* 1024 sets */
-	else if (OCTEON_IS_MODEL(OCTEON_CN31XX)
-		 || OCTEON_IS_MODEL(OCTEON_CN52XX))
+	else if (OCTEON_IS_MODEL(OCTEON_CN31XX) || OCTEON_IS_MODEL(OCTEON_CN52XX))
 		l2_set_bits = 9;	/* 512 sets */
 	else if (OCTEON_IS_MODEL(OCTEON_CN30XX))
 		l2_set_bits = 8;	/* 256 sets */
@@ -666,7 +769,6 @@ int cvmx_l2c_get_set_bits(void)
 		l2_set_bits = 11;	/* 2048 sets */
 	}
 	return l2_set_bits;
-
 }
 
 /* Return the number of sets in the L2 Cache */
@@ -682,8 +784,11 @@ int cvmx_l2c_get_num_assoc(void)
 	if (OCTEON_IS_MODEL(OCTEON_CN56XX) ||
 	    OCTEON_IS_MODEL(OCTEON_CN52XX) ||
 	    OCTEON_IS_MODEL(OCTEON_CN58XX) ||
-	    OCTEON_IS_MODEL(OCTEON_CN50XX) || OCTEON_IS_MODEL(OCTEON_CN38XX))
+	    OCTEON_IS_MODEL(OCTEON_CN50XX) ||
+	    OCTEON_IS_MODEL(OCTEON_CN38XX))
 		l2_assoc = 8;
+	else if (OCTEON_IS_MODEL(OCTEON_CN63XX))
+		l2_assoc = 16;
 	else if (OCTEON_IS_MODEL(OCTEON_CN31XX) ||
 		 OCTEON_IS_MODEL(OCTEON_CN30XX))
 		l2_assoc = 4;
@@ -693,11 +798,42 @@ int cvmx_l2c_get_num_assoc(void)
 	}
 
 	/* Check to see if part of the cache is disabled */
-	if (cvmx_fuse_read(265))
-		l2_assoc = l2_assoc >> 2;
-	else if (cvmx_fuse_read(264))
-		l2_assoc = l2_assoc >> 1;
-
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+		union cvmx_mio_fus_dat3 mio_fus_dat3;
+
+		mio_fus_dat3.u64 = cvmx_read_csr(CVMX_MIO_FUS_DAT3);
+		/*
+		 * cvmx_mio_fus_dat3.s.l2c_crip fuses map as follows
+		 * <2> will be not used for 63xx
+		 * <1> disables 1/2 ways
+		 * <0> disables 1/4 ways
+		 * They are cumulative, so for 63xx:
+		 * <1> <0>
+		 * 0 0 16-way 2MB cache
+		 * 0 1 12-way 1.5MB cache
+		 * 1 0 8-way 1MB cache
+		 * 1 1 4-way 512KB cache
+		 */
+
+		if (mio_fus_dat3.s.l2c_crip == 3)
+			l2_assoc = 4;
+		else if (mio_fus_dat3.s.l2c_crip == 2)
+			l2_assoc = 8;
+		else if (mio_fus_dat3.s.l2c_crip == 1)
+			l2_assoc = 12;
+	} else {
+		union cvmx_l2d_fus3 val;
+		val.u64 = cvmx_read_csr(CVMX_L2D_FUS3);
+		/*
+		 * Using shifts here, as bit position names are
+		 * different for each model but they all mean the
+		 * same.
+		 */
+		if ((val.u64 >> 35) & 0x1)
+			l2_assoc = l2_assoc >> 2;
+		else if ((val.u64 >> 34) & 0x1)
+			l2_assoc = l2_assoc >> 1;
+	}
 	return l2_assoc;
 }
 
@@ -711,24 +847,54 @@ int cvmx_l2c_get_num_assoc(void)
  */
 void cvmx_l2c_flush_line(uint32_t assoc, uint32_t index)
 {
-	union cvmx_l2c_dbg l2cdbg;
+	/* Check the range of the index. */
+	if (index > (uint32_t)cvmx_l2c_get_num_sets()) {
+		cvmx_dprintf("ERROR: cvmx_l2c_flush_line index out of range.\n");
+		return;
+	}
 
-	l2cdbg.u64 = 0;
-	l2cdbg.s.ppnum = cvmx_get_core_num();
-	l2cdbg.s.finv = 1;
+	/* Check the range of association. */
+	if (assoc > (uint32_t)cvmx_l2c_get_num_assoc()) {
+		cvmx_dprintf("ERROR: cvmx_l2c_flush_line association out of range.\n");
+		return;
+	}
 
-	l2cdbg.s.set = assoc;
-	/*
-	 * Enter debug mode, and make sure all other writes complete
-	 * before we enter debug mode.
-	 */
-	asm volatile ("sync" : : : "memory");
-	cvmx_write_csr(CVMX_L2C_DBG, l2cdbg.u64);
-	cvmx_read_csr(CVMX_L2C_DBG);
-
-	CVMX_PREPARE_FOR_STORE(((1ULL << 63) + (index) * 128), 0);
-	/* Exit debug mode */
-	asm volatile ("sync" : : : "memory");
-	cvmx_write_csr(CVMX_L2C_DBG, 0);
-	cvmx_read_csr(CVMX_L2C_DBG);
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX)) {
+		uint64_t address;
+		/* Create the address based on index and association.
+		 * Bits<20:17> select the way of the cache block involved in
+		 *             the operation
+		 * Bits<16:7> of the effect address select the index
+		 */
+		address = CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+				(assoc << CVMX_L2C_TAG_ADDR_ALIAS_SHIFT) |
+				(index << CVMX_L2C_IDX_ADDR_SHIFT));
+		CVMX_CACHE_WBIL2I(address, 0);
+	} else {
+		union cvmx_l2c_dbg l2cdbg;
+
+		l2cdbg.u64 = 0;
+		if (!OCTEON_IS_MODEL(OCTEON_CN30XX))
+			l2cdbg.s.ppnum = cvmx_get_core_num();
+		l2cdbg.s.finv = 1;
+
+		l2cdbg.s.set = assoc;
+		cvmx_spinlock_lock(&cvmx_l2c_spinlock);
+		/*
+		 * Enter debug mode, and make sure all other writes
+		 * complete before we enter debug mode
+		 */
+		CVMX_SYNC;
+		cvmx_write_csr(CVMX_L2C_DBG, l2cdbg.u64);
+		cvmx_read_csr(CVMX_L2C_DBG);
+
+		CVMX_PREPARE_FOR_STORE(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						    index * CVMX_CACHE_LINE_SIZE),
+				       0);
+		/* Exit debug mode */
+		CVMX_SYNC;
+		cvmx_write_csr(CVMX_L2C_DBG, 0);
+		cvmx_read_csr(CVMX_L2C_DBG);
+		cvmx_spinlock_unlock(&cvmx_l2c_spinlock);
+	}
 }
diff --git a/arch/mips/cavium-octeon/octeon-platform.c b/arch/mips/cavium-octeon/octeon-platform.c
index 62ac30eef5e8..cecaf62aef32 100644
--- a/arch/mips/cavium-octeon/octeon-platform.c
+++ b/arch/mips/cavium-octeon/octeon-platform.c
@@ -3,13 +3,15 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2004-2009 Cavium Networks
+ * Copyright (C) 2004-2010 Cavium Networks
  * Copyright (C) 2008 Wind River Systems
  */
 
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/i2c.h>
+#include <linux/usb.h>
+#include <linux/dma-mapping.h>
 #include <linux/module.h>
 #include <linux/platform_device.h>
 
@@ -198,7 +200,7 @@ static int __init octeon_i2c_device_init(void)
 		num_ports = 1;
 
 	for (port = 0; port < num_ports; port++) {
-		octeon_i2c_data[port].sys_freq = octeon_get_clock_rate();
+		octeon_i2c_data[port].sys_freq = octeon_get_io_clock_rate();
 		/*FIXME: should be examined. At the moment is set for 100Khz */
 		octeon_i2c_data[port].i2c_freq = 100000;
 
@@ -301,6 +303,10 @@ static int __init octeon_mgmt_device_init(void)
 			ret = -ENOMEM;
 			goto out;
 		}
+		/* No DMA restrictions */
+		pd->dev.coherent_dma_mask = DMA_BIT_MASK(64);
+		pd->dev.dma_mask = &pd->dev.coherent_dma_mask;
+
 		switch (port) {
 		case 0:
 			mgmt_port_resource.start = OCTEON_IRQ_MII0;
@@ -332,6 +338,108 @@ out:
 }
 device_initcall(octeon_mgmt_device_init);
 
+#ifdef CONFIG_USB
+
+static int __init octeon_ehci_device_init(void)
+{
+	struct platform_device *pd;
+	int ret = 0;
+
+	struct resource usb_resources[] = {
+		{
+			.flags	= IORESOURCE_MEM,
+		}, {
+			.flags	= IORESOURCE_IRQ,
+		}
+	};
+
+	/* Only Octeon2 has ehci/ohci */
+	if (!OCTEON_IS_MODEL(OCTEON_CN63XX))
+		return 0;
+
+	if (octeon_is_simulation() || usb_disabled())
+		return 0; /* No USB in the simulator. */
+
+	pd = platform_device_alloc("octeon-ehci", 0);
+	if (!pd) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	usb_resources[0].start = 0x00016F0000000000ULL;
+	usb_resources[0].end = usb_resources[0].start + 0x100;
+
+	usb_resources[1].start = OCTEON_IRQ_USB0;
+	usb_resources[1].end = OCTEON_IRQ_USB0;
+
+	ret = platform_device_add_resources(pd, usb_resources,
+					    ARRAY_SIZE(usb_resources));
+	if (ret)
+		goto fail;
+
+	ret = platform_device_add(pd);
+	if (ret)
+		goto fail;
+
+	return ret;
+fail:
+	platform_device_put(pd);
+out:
+	return ret;
+}
+device_initcall(octeon_ehci_device_init);
+
+static int __init octeon_ohci_device_init(void)
+{
+	struct platform_device *pd;
+	int ret = 0;
+
+	struct resource usb_resources[] = {
+		{
+			.flags	= IORESOURCE_MEM,
+		}, {
+			.flags	= IORESOURCE_IRQ,
+		}
+	};
+
+	/* Only Octeon2 has ehci/ohci */
+	if (!OCTEON_IS_MODEL(OCTEON_CN63XX))
+		return 0;
+
+	if (octeon_is_simulation() || usb_disabled())
+		return 0; /* No USB in the simulator. */
+
+	pd = platform_device_alloc("octeon-ohci", 0);
+	if (!pd) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	usb_resources[0].start = 0x00016F0000000400ULL;
+	usb_resources[0].end = usb_resources[0].start + 0x100;
+
+	usb_resources[1].start = OCTEON_IRQ_USB0;
+	usb_resources[1].end = OCTEON_IRQ_USB0;
+
+	ret = platform_device_add_resources(pd, usb_resources,
+					    ARRAY_SIZE(usb_resources));
+	if (ret)
+		goto fail;
+
+	ret = platform_device_add(pd);
+	if (ret)
+		goto fail;
+
+	return ret;
+fail:
+	platform_device_put(pd);
+out:
+	return ret;
+}
+device_initcall(octeon_ohci_device_init);
+
+#endif /* CONFIG_USB */
+
 MODULE_AUTHOR("David Daney <ddaney@caviumnetworks.com>");
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Platform driver for Octeon SOC");
diff --git a/arch/mips/cavium-octeon/serial.c b/arch/mips/cavium-octeon/serial.c
index 12dbf533b77d..057f0ae88c99 100644
--- a/arch/mips/cavium-octeon/serial.c
+++ b/arch/mips/cavium-octeon/serial.c
@@ -66,7 +66,7 @@ static void __init octeon_uart_set_common(struct plat_serial8250_port *p)
 		/* Make simulator output fast*/
 		p->uartclk = 115200 * 16;
 	else
-		p->uartclk = mips_hpt_frequency;
+		p->uartclk = octeon_get_io_clock_rate();
 	p->serial_in = octeon_serial_in;
 	p->serial_out = octeon_serial_out;
 }
diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c
index 69197cb6c7ea..b0c3686c96dd 100644
--- a/arch/mips/cavium-octeon/setup.c
+++ b/arch/mips/cavium-octeon/setup.c
@@ -33,6 +33,7 @@
 
 #include <asm/octeon/octeon.h>
 #include <asm/octeon/pci-octeon.h>
+#include <asm/octeon/cvmx-mio-defs.h>
 
 #ifdef CONFIG_CAVIUM_DECODE_RSL
 extern void cvmx_interrupt_rsl_decode(void);
@@ -96,12 +97,21 @@ int octeon_is_pci_host(void)
  */
 uint64_t octeon_get_clock_rate(void)
 {
-	if (octeon_is_simulation())
-		octeon_bootinfo->eclock_hz = 6000000;
-	return octeon_bootinfo->eclock_hz;
+	struct cvmx_sysinfo *sysinfo = cvmx_sysinfo_get();
+
+	return sysinfo->cpu_clock_hz;
 }
 EXPORT_SYMBOL(octeon_get_clock_rate);
 
+static u64 octeon_io_clock_rate;
+
+u64 octeon_get_io_clock_rate(void)
+{
+	return octeon_io_clock_rate;
+}
+EXPORT_SYMBOL(octeon_get_io_clock_rate);
+
+
 /**
  * Write to the LCD display connected to the bootbus. This display
  * exists on most Cavium evaluation boards. If it doesn't exist, then
@@ -346,8 +356,18 @@ void octeon_user_io_init(void)
 	cvmmemctl.s.wbfltime = 0;
 	/* R/W If set, do not put Istream in the L2 cache. */
 	cvmmemctl.s.istrnol2 = 0;
-	/* R/W The write buffer threshold. */
-	cvmmemctl.s.wbthresh = 10;
+
+	/*
+	 * R/W The write buffer threshold. As per erratum Core-14752
+	 * for CN63XX, a sc/scd might fail if the write buffer is
+	 * full.  Lowering WBTHRESH greatly lowers the chances of the
+	 * write buffer ever being full and triggering the erratum.
+	 */
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX_PASS1_X))
+		cvmmemctl.s.wbthresh = 4;
+	else
+		cvmmemctl.s.wbthresh = 10;
+
 	/* R/W If set, CVMSEG is available for loads/stores in
 	 * kernel/debug mode. */
 #if CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE > 0
@@ -365,14 +385,13 @@ void octeon_user_io_init(void)
 	 * is max legal value. */
 	cvmmemctl.s.lmemsz = CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE;
 
+	write_c0_cvmmemctl(cvmmemctl.u64);
 
 	if (smp_processor_id() == 0)
 		pr_notice("CVMSEG size: %d cache lines (%d bytes)\n",
 			  CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE,
 			  CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE * 128);
 
-	write_c0_cvmmemctl(cvmmemctl.u64);
-
 	/* Move the performance counter interrupts to IRQ 6 */
 	cvmctl = read_c0_cvmctl();
 	cvmctl &= ~(7 << 7);
@@ -416,6 +435,41 @@ void __init prom_init(void)
 		cvmx_phys_to_ptr(octeon_boot_desc_ptr->cvmx_desc_vaddr);
 	cvmx_bootmem_init(cvmx_phys_to_ptr(octeon_bootinfo->phy_mem_desc_addr));
 
+	sysinfo = cvmx_sysinfo_get();
+	memset(sysinfo, 0, sizeof(*sysinfo));
+	sysinfo->system_dram_size = octeon_bootinfo->dram_size << 20;
+	sysinfo->phy_mem_desc_ptr =
+		cvmx_phys_to_ptr(octeon_bootinfo->phy_mem_desc_addr);
+	sysinfo->core_mask = octeon_bootinfo->core_mask;
+	sysinfo->exception_base_addr = octeon_bootinfo->exception_base_addr;
+	sysinfo->cpu_clock_hz = octeon_bootinfo->eclock_hz;
+	sysinfo->dram_data_rate_hz = octeon_bootinfo->dclock_hz * 2;
+	sysinfo->board_type = octeon_bootinfo->board_type;
+	sysinfo->board_rev_major = octeon_bootinfo->board_rev_major;
+	sysinfo->board_rev_minor = octeon_bootinfo->board_rev_minor;
+	memcpy(sysinfo->mac_addr_base, octeon_bootinfo->mac_addr_base,
+	       sizeof(sysinfo->mac_addr_base));
+	sysinfo->mac_addr_count = octeon_bootinfo->mac_addr_count;
+	memcpy(sysinfo->board_serial_number,
+	       octeon_bootinfo->board_serial_number,
+	       sizeof(sysinfo->board_serial_number));
+	sysinfo->compact_flash_common_base_addr =
+		octeon_bootinfo->compact_flash_common_base_addr;
+	sysinfo->compact_flash_attribute_base_addr =
+		octeon_bootinfo->compact_flash_attribute_base_addr;
+	sysinfo->led_display_base_addr = octeon_bootinfo->led_display_base_addr;
+	sysinfo->dfa_ref_clock_hz = octeon_bootinfo->dfa_ref_clock_hz;
+	sysinfo->bootloader_config_flags = octeon_bootinfo->config_flags;
+
+	if (OCTEON_IS_MODEL(OCTEON_CN6XXX)) {
+		/* I/O clock runs at a different rate than the CPU. */
+		union cvmx_mio_rst_boot rst_boot;
+		rst_boot.u64 = cvmx_read_csr(CVMX_MIO_RST_BOOT);
+		octeon_io_clock_rate = 50000000 * rst_boot.s.pnr_mul;
+	} else {
+		octeon_io_clock_rate = sysinfo->cpu_clock_hz;
+	}
+
 	/*
 	 * Only enable the LED controller if we're running on a CN38XX, CN58XX,
 	 * or CN56XX. The CN30XX and CN31XX don't have an LED controller.
@@ -479,33 +533,6 @@ void __init prom_init(void)
 	}
 #endif
 
-	sysinfo = cvmx_sysinfo_get();
-	memset(sysinfo, 0, sizeof(*sysinfo));
-	sysinfo->system_dram_size = octeon_bootinfo->dram_size << 20;
-	sysinfo->phy_mem_desc_ptr =
-		cvmx_phys_to_ptr(octeon_bootinfo->phy_mem_desc_addr);
-	sysinfo->core_mask = octeon_bootinfo->core_mask;
-	sysinfo->exception_base_addr = octeon_bootinfo->exception_base_addr;
-	sysinfo->cpu_clock_hz = octeon_bootinfo->eclock_hz;
-	sysinfo->dram_data_rate_hz = octeon_bootinfo->dclock_hz * 2;
-	sysinfo->board_type = octeon_bootinfo->board_type;
-	sysinfo->board_rev_major = octeon_bootinfo->board_rev_major;
-	sysinfo->board_rev_minor = octeon_bootinfo->board_rev_minor;
-	memcpy(sysinfo->mac_addr_base, octeon_bootinfo->mac_addr_base,
-	       sizeof(sysinfo->mac_addr_base));
-	sysinfo->mac_addr_count = octeon_bootinfo->mac_addr_count;
-	memcpy(sysinfo->board_serial_number,
-	       octeon_bootinfo->board_serial_number,
-	       sizeof(sysinfo->board_serial_number));
-	sysinfo->compact_flash_common_base_addr =
-		octeon_bootinfo->compact_flash_common_base_addr;
-	sysinfo->compact_flash_attribute_base_addr =
-		octeon_bootinfo->compact_flash_attribute_base_addr;
-	sysinfo->led_display_base_addr = octeon_bootinfo->led_display_base_addr;
-	sysinfo->dfa_ref_clock_hz = octeon_bootinfo->dfa_ref_clock_hz;
-	sysinfo->bootloader_config_flags = octeon_bootinfo->config_flags;
-
-
 	octeon_check_cpu_bist();
 
 	octeon_uart = octeon_get_boot_uart();
@@ -740,6 +767,31 @@ EXPORT_SYMBOL(prom_putchar);
 
 void prom_free_prom_memory(void)
 {
+	if (OCTEON_IS_MODEL(OCTEON_CN63XX_PASS1_X)) {
+		/* Check for presence of Core-14449 fix.  */
+		u32 insn;
+		u32 *foo;
+
+		foo = &insn;
+
+		asm volatile("# before" : : : "memory");
+		prefetch(foo);
+		asm volatile(
+			".set push\n\t"
+			".set noreorder\n\t"
+			"bal 1f\n\t"
+			"nop\n"
+			"1:\tlw %0,-12($31)\n\t"
+			".set pop\n\t"
+			: "=r" (insn) : : "$31", "memory");
+
+		if ((insn >> 26) != 0x33)
+			panic("No PREF instruction at Core-14449 probe point.\n");
+
+		if (((insn >> 16) & 0x1f) != 28)
+			panic("Core-14449 WAR not in place (%04x).\n"
+			      "Please build kernel with proper options (CONFIG_CAVIUM_CN63XXP1).\n", insn);
+	}
 #ifdef CONFIG_CAVIUM_DECODE_RSL
 	cvmx_interrupt_rsl_enable();