Merge branches 'linus' and 'irq/urgent' into irq/core

Reason: Get the upstream and urgent fixes before applying more complex
changes.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Documentation/devicetree/bindings/ata/apm-xgene.txt

@@ -24,6 +24,7 @@ Required properties:
   * "sata-phy" for the SATA 6.0Gbps PHY
 
 Optional properties:
+- dma-coherent		: Present if dma operations are coherent
 - status		: Shall be "ok" if enabled or "disabled" if disabled.
 			  Default is "ok".
 
@@ -55,6 +56,7 @@ Example:
 			      <0x0 0x1f22e000 0x0 0x1000>,
 			      <0x0 0x1f227000 0x0 0x1000>;
 			interrupts = <0x0 0x87 0x4>;
+			dma-coherent;
 			status = "ok";
 			clocks = <&sataclk 0>;
 			phys = <&phy2 0>;
@@ -69,6 +71,7 @@ Example:
 			      <0x0 0x1f23e000 0x0 0x1000>,
 			      <0x0 0x1f237000 0x0 0x1000>;
 			interrupts = <0x0 0x88 0x4>;
+			dma-coherent;
 			status = "ok";
 			clocks = <&sataclk 0>;
 			phys = <&phy3 0>;

Documentation/devicetree/bindings/net/arc_emac.txt

@@ -4,11 +4,15 @@ Required properties:
 - compatible: Should be "snps,arc-emac"
 - reg: Address and length of the register set for the device
 - interrupts: Should contain the EMAC interrupts
-- clock-frequency: CPU frequency. It is needed to calculate and set polling
-period of EMAC.
 - max-speed: see ethernet.txt file in the same directory.
 - phy: see ethernet.txt file in the same directory.
 
+Clock handling:
+The clock frequency is needed to calculate and set polling period of EMAC.
+It must be provided by one of:
+- clock-frequency: CPU frequency.
+- clocks: reference to the clock supplying the EMAC.
+
 Child nodes of the driver are the individual PHY devices connected to the
 MDIO bus. They must have a "reg" property given the PHY address on the MDIO bus.
 
@@ -19,7 +23,11 @@ Examples:
 		reg = <0xc0fc2000 0x3c>;
 		interrupts = <6>;
 		mac-address = [ 00 11 22 33 44 55 ];
+
 		clock-frequency = <80000000>;
+		/* or */
+		clocks = <&emac_clock>;
+
 		max-speed = <100>;
 		phy = <&phy0>;
 

Documentation/input/elantech.txt

@@ -504,9 +504,12 @@ byte 5:
 * reg_10
 
    bit   7   6   5   4   3   2   1   0
-         0   0   0   0   0   0   0   A
+         0   0   0   0   R   F   T   A
 
          A: 1 = enable absolute tracking
+         T: 1 = enable two finger mode auto correct
+         F: 1 = disable ABS Position Filter
+         R: 1 = enable real hardware resolution
 
 6.2 Native absolute mode 6 byte packet format
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Documentation/networking/scaling.txt

@@ -429,7 +429,7 @@ RPS and RFS were introduced in kernel 2.6.35. XPS was incorporated into
 (therbert@google.com)
 
 Accelerated RFS was introduced in 2.6.35. Original patches were
-submitted by Ben Hutchings (bhutchings@solarflare.com)
+submitted by Ben Hutchings (bwh@kernel.org)
 
 Authors:
 Tom Herbert (therbert@google.com)

MAINTAINERS

@@ -3556,7 +3556,7 @@ F:	include/scsi/libfcoe.h
 F:	include/uapi/scsi/fc/
 
 FILE LOCKING (flock() and fcntl()/lockf())
-M:	Jeff Layton <jlayton@redhat.com>
+M:	Jeff Layton <jlayton@poochiereds.net>
 M:	J. Bruce Fields <bfields@fieldses.org>
 L:	linux-fsdevel@vger.kernel.org
 S:	Maintained
@@ -4818,6 +4818,14 @@ L:	linux-kernel@vger.kernel.org
 S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
 F:	kernel/irq/
+
+IRQCHIP DRIVERS
+M:	Thomas Gleixner <tglx@linutronix.de>
+M:	Jason Cooper <jason@lakedaemon.net>
+L:	linux-kernel@vger.kernel.org
+S:	Maintained
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
+T:	git git://git.infradead.org/users/jcooper/linux.git irqchip/core
 F:	drivers/irqchip/
 
 IRQ DOMAINS (IRQ NUMBER MAPPING LIBRARY)
@@ -7288,7 +7296,6 @@ F:	drivers/video/aty/aty128fb.c
 RALINK RT2X00 WIRELESS LAN DRIVER
 P:	rt2x00 project
 M:	Ivo van Doorn <IvDoorn@gmail.com>
-M:	Gertjan van Wingerde <gwingerde@gmail.com>
 M:	Helmut Schaa <helmut.schaa@googlemail.com>
 L:	linux-wireless@vger.kernel.org
 L:	users@rt2x00.serialmonkey.com (moderated for non-subscribers)
@@ -7304,7 +7311,7 @@ F:	Documentation/blockdev/ramdisk.txt
 F:	drivers/block/brd.c
 
 RANDOM NUMBER DRIVER
-M:	Theodore Ts'o" <tytso@mit.edu>
+M:	"Theodore Ts'o" <tytso@mit.edu>
 S:	Maintained
 F:	drivers/char/random.c
 
@@ -7685,7 +7692,6 @@ F:	drivers/clk/samsung/
 SAMSUNG SXGBE DRIVERS
 M:	Byungho An <bh74.an@samsung.com>
 M:	Girish K S <ks.giri@samsung.com>
-M:	Siva Reddy Kallam <siva.kallam@samsung.com>
 M:	Vipul Pandya <vipul.pandya@samsung.com>
 S:	Supported
 L:	netdev@vger.kernel.org
@@ -9962,7 +9968,7 @@ F:	drivers/net/hamradio/*scc.c
 F:	drivers/net/hamradio/z8530.h
 
 ZBUD COMPRESSED PAGE ALLOCATOR
-M:	Seth Jennings <sjenning@linux.vnet.ibm.com>
+M:	Seth Jennings <sjennings@variantweb.net>
 L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/zbud.c
@@ -10007,7 +10013,7 @@ F:	mm/zsmalloc.c
 F:	include/linux/zsmalloc.h
 
 ZSWAP COMPRESSED SWAP CACHING
-M:	Seth Jennings <sjenning@linux.vnet.ibm.com>
+M:	Seth Jennings <sjennings@variantweb.net>
 L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/zswap.c

Makefile

@@ -1,7 +1,7 @@
 VERSION = 3
 PATCHLEVEL = 15
 SUBLEVEL = 0
-EXTRAVERSION = -rc3
+EXTRAVERSION = -rc5
 NAME = Shuffling Zombie Juror
 
 # *DOCUMENTATION*

arch/arm64/boot/dts/apm-storm.dtsi

@@ -307,6 +307,7 @@
 			      <0x0 0x1f21e000 0x0 0x1000>,
 			      <0x0 0x1f217000 0x0 0x1000>;
 			interrupts = <0x0 0x86 0x4>;
+			dma-coherent;
 			status = "disabled";
 			clocks = <&sata01clk 0>;
 			phys = <&phy1 0>;
@@ -321,6 +322,7 @@
 			      <0x0 0x1f22e000 0x0 0x1000>,
 			      <0x0 0x1f227000 0x0 0x1000>;
 			interrupts = <0x0 0x87 0x4>;
+			dma-coherent;
 			status = "ok";
 			clocks = <&sata23clk 0>;
 			phys = <&phy2 0>;
@@ -334,6 +336,7 @@
 			      <0x0 0x1f23d000 0x0 0x1000>,
 			      <0x0 0x1f23e000 0x0 0x1000>;
 			interrupts = <0x0 0x88 0x4>;
+			dma-coherent;
 			status = "ok";
 			clocks = <&sata45clk 0>;
 			phys = <&phy3 0>;

arch/arm64/kernel/early_printk.c

@@ -143,10 +143,8 @@ static int __init setup_early_printk(char *buf)
 	}
 	/* no options parsing yet */
 
-	if (paddr) {
-		set_fixmap_io(FIX_EARLYCON_MEM_BASE, paddr);
-		early_base = (void __iomem *)fix_to_virt(FIX_EARLYCON_MEM_BASE);
-	}
+	if (paddr)
+		early_base = (void __iomem *)set_fixmap_offset_io(FIX_EARLYCON_MEM_BASE, paddr);
 
 	printch = match->printch;
 	early_console = &early_console_dev;

arch/arm64/kernel/setup.c

@@ -396,7 +396,7 @@ static int __init arm64_device_init(void)
 	of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
 	return 0;
 }
-arch_initcall(arm64_device_init);
+arch_initcall_sync(arm64_device_init);
 
 static DEFINE_PER_CPU(struct cpu, cpu_data);
 

arch/arm64/mm/dma-mapping.c

@@ -22,8 +22,11 @@
 #include <linux/slab.h>
 #include <linux/dma-mapping.h>
 #include <linux/dma-contiguous.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
+#include <linux/amba/bus.h>
 
 #include <asm/cacheflush.h>
 
@@ -305,17 +308,45 @@ struct dma_map_ops coherent_swiotlb_dma_ops = {
 };
 EXPORT_SYMBOL(coherent_swiotlb_dma_ops);
 
+static int dma_bus_notifier(struct notifier_block *nb,
+			    unsigned long event, void *_dev)
+{
+	struct device *dev = _dev;
+
+	if (event != BUS_NOTIFY_ADD_DEVICE)
+		return NOTIFY_DONE;
+
+	if (of_property_read_bool(dev->of_node, "dma-coherent"))
+		set_dma_ops(dev, &coherent_swiotlb_dma_ops);
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block platform_bus_nb = {
+	.notifier_call = dma_bus_notifier,
+};
+
+static struct notifier_block amba_bus_nb = {
+	.notifier_call = dma_bus_notifier,
+};
+
 extern int swiotlb_late_init_with_default_size(size_t default_size);
 
 static int __init swiotlb_late_init(void)
 {
 	size_t swiotlb_size = min(SZ_64M, MAX_ORDER_NR_PAGES << PAGE_SHIFT);
 
-	dma_ops = &coherent_swiotlb_dma_ops;
+	/*
+	 * These must be registered before of_platform_populate().
+	 */
+	bus_register_notifier(&platform_bus_type, &platform_bus_nb);
+	bus_register_notifier(&amba_bustype, &amba_bus_nb);
+
+	dma_ops = &noncoherent_swiotlb_dma_ops;
 
 	return swiotlb_late_init_with_default_size(swiotlb_size);
 }
-subsys_initcall(swiotlb_late_init);
+arch_initcall(swiotlb_late_init);
 
 #define PREALLOC_DMA_DEBUG_ENTRIES	4096
 

arch/arm64/mm/mmu.c

@@ -374,6 +374,9 @@ int kern_addr_valid(unsigned long addr)
 	if (pmd_none(*pmd))
 		return 0;
 
+	if (pmd_sect(*pmd))
+		return pfn_valid(pmd_pfn(*pmd));
+
 	pte = pte_offset_kernel(pmd, addr);
 	if (pte_none(*pte))
 		return 0;

arch/sparc/include/asm/pgtable_64.h

@@ -71,6 +71,23 @@
 
 #include <linux/sched.h>
 
+extern unsigned long sparc64_valid_addr_bitmap[];
+
+/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */
+static inline bool __kern_addr_valid(unsigned long paddr)
+{
+	if ((paddr >> MAX_PHYS_ADDRESS_BITS) != 0UL)
+		return false;
+	return test_bit(paddr >> ILOG2_4MB, sparc64_valid_addr_bitmap);
+}
+
+static inline bool kern_addr_valid(unsigned long addr)
+{
+	unsigned long paddr = __pa(addr);
+
+	return __kern_addr_valid(paddr);
+}
+
 /* Entries per page directory level. */
 #define PTRS_PER_PTE	(1UL << (PAGE_SHIFT-3))
 #define PTRS_PER_PMD	(1UL << PMD_BITS)
@@ -79,9 +96,12 @@
 /* Kernel has a separate 44bit address space. */
 #define FIRST_USER_ADDRESS	0
 
-#define pte_ERROR(e)	__builtin_trap()
-#define pmd_ERROR(e)	__builtin_trap()
-#define pgd_ERROR(e)	__builtin_trap()
+#define pmd_ERROR(e)							\
+	pr_err("%s:%d: bad pmd %p(%016lx) seen at (%pS)\n",		\
+	       __FILE__, __LINE__, &(e), pmd_val(e), __builtin_return_address(0))
+#define pgd_ERROR(e)							\
+	pr_err("%s:%d: bad pgd %p(%016lx) seen at (%pS)\n",		\
+	       __FILE__, __LINE__, &(e), pgd_val(e), __builtin_return_address(0))
 
 #endif /* !(__ASSEMBLY__) */
 
@@ -258,8 +278,8 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot)
 {
 	unsigned long mask, tmp;
 
-	/* SUN4U: 0x600307ffffffecb8 (negated == 0x9ffcf80000001347)
-	 * SUN4V: 0x30ffffffffffee17 (negated == 0xcf000000000011e8)
+	/* SUN4U: 0x630107ffffffec38 (negated == 0x9cfef800000013c7)
+	 * SUN4V: 0x33ffffffffffee07 (negated == 0xcc000000000011f8)
 	 *
 	 * Even if we use negation tricks the result is still a 6
 	 * instruction sequence, so don't try to play fancy and just
@@ -289,10 +309,10 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot)
 	"	.previous\n"
 	: "=r" (mask), "=r" (tmp)
 	: "i" (_PAGE_PADDR_4U | _PAGE_MODIFIED_4U | _PAGE_ACCESSED_4U |
-	       _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U | _PAGE_PRESENT_4U |
+	       _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U |
 	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4U),
 	  "i" (_PAGE_PADDR_4V | _PAGE_MODIFIED_4V | _PAGE_ACCESSED_4V |
-	       _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V | _PAGE_PRESENT_4V |
+	       _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V |
 	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4V));
 
 	return __pte((pte_val(pte) & mask) | (pgprot_val(prot) & ~mask));
@@ -633,7 +653,7 @@ static inline unsigned long pmd_large(pmd_t pmd)
 {
 	pte_t pte = __pte(pmd_val(pmd));
 
-	return (pte_val(pte) & _PAGE_PMD_HUGE) && pte_present(pte);
+	return pte_val(pte) & _PAGE_PMD_HUGE;
 }
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
@@ -719,20 +739,6 @@ static inline pmd_t pmd_mkwrite(pmd_t pmd)
 	return __pmd(pte_val(pte));
 }
 
-static inline pmd_t pmd_mknotpresent(pmd_t pmd)
-{
-	unsigned long mask;
-
-	if (tlb_type == hypervisor)
-		mask = _PAGE_PRESENT_4V;
-	else
-		mask = _PAGE_PRESENT_4U;
-
-	pmd_val(pmd) &= ~mask;
-
-	return pmd;
-}
-
 static inline pmd_t pmd_mksplitting(pmd_t pmd)
 {
 	pte_t pte = __pte(pmd_val(pmd));
@@ -757,6 +763,20 @@ static inline int pmd_present(pmd_t pmd)
 
 #define pmd_none(pmd)			(!pmd_val(pmd))
 
+/* pmd_bad() is only called on non-trans-huge PMDs.  Our encoding is
+ * very simple, it's just the physical address.  PTE tables are of
+ * size PAGE_SIZE so make sure the sub-PAGE_SIZE bits are clear and
+ * the top bits outside of the range of any physical address size we
+ * support are clear as well.  We also validate the physical itself.
+ */
+#define pmd_bad(pmd)			((pmd_val(pmd) & ~PAGE_MASK) || \
+					 !__kern_addr_valid(pmd_val(pmd)))
+
+#define pud_none(pud)			(!pud_val(pud))
+
+#define pud_bad(pud)			((pud_val(pud) & ~PAGE_MASK) || \
+					 !__kern_addr_valid(pud_val(pud)))
+
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 extern void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 		       pmd_t *pmdp, pmd_t pmd);
@@ -790,10 +810,7 @@ static inline unsigned long __pmd_page(pmd_t pmd)
 #define pud_page_vaddr(pud)		\
 	((unsigned long) __va(pud_val(pud)))
 #define pud_page(pud) 			virt_to_page((void *)pud_page_vaddr(pud))
-#define pmd_bad(pmd)			(0)
 #define pmd_clear(pmdp)			(pmd_val(*(pmdp)) = 0UL)
-#define pud_none(pud)			(!pud_val(pud))
-#define pud_bad(pud)			(0)
 #define pud_present(pud)		(pud_val(pud) != 0U)
 #define pud_clear(pudp)			(pud_val(*(pudp)) = 0UL)
 
@@ -893,6 +910,10 @@ extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t *);
 extern void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr,
 				 pmd_t *pmd);
 
+#define __HAVE_ARCH_PMDP_INVALIDATE
+extern void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
+			    pmd_t *pmdp);
+
 #define __HAVE_ARCH_PGTABLE_DEPOSIT
 extern void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
 				       pgtable_t pgtable);
@@ -919,18 +940,6 @@ extern unsigned long pte_file(pte_t);
 extern pte_t pgoff_to_pte(unsigned long);
 #define PTE_FILE_MAX_BITS	(64UL - PAGE_SHIFT - 1UL)
 
-extern unsigned long sparc64_valid_addr_bitmap[];
-
-/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */
-static inline bool kern_addr_valid(unsigned long addr)
-{
-	unsigned long paddr = __pa(addr);
-
-	if ((paddr >> 41UL) != 0UL)
-		return false;
-	return test_bit(paddr >> 22, sparc64_valid_addr_bitmap);
-}
-
 extern int page_in_phys_avail(unsigned long paddr);
 
 /*

arch/sparc/include/asm/tsb.h

@@ -171,7 +171,8 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end;
 	andcc		REG1, REG2, %g0;		\
 	be,pt		%xcc, 700f;			\
 	 sethi		%hi(4 * 1024 * 1024), REG2;	\
-	andn		REG1, REG2, REG1;		\
+	brgez,pn	REG1, FAIL_LABEL;		\
+	 andn		REG1, REG2, REG1;		\
 	and		VADDR, REG2, REG2;		\
 	brlz,pt		REG1, PTE_LABEL;		\
 	 or		REG1, REG2, REG1;		\

arch/sparc/kernel/head_64.S

@@ -282,8 +282,8 @@ sun4v_chip_type:
 	stx	%l2, [%l4 + 0x0]
 	ldx	[%sp + 2047 + 128 + 0x50], %l3	! physaddr low
 	/* 4MB align */
-	srlx	%l3, 22, %l3
-	sllx	%l3, 22, %l3
+	srlx	%l3, ILOG2_4MB, %l3
+	sllx	%l3, ILOG2_4MB, %l3
 	stx	%l3, [%l4 + 0x8]
 
 	/* Leave service as-is, "call-method" */

arch/sparc/kernel/ktlb.S

@@ -277,7 +277,7 @@ kvmap_dtlb_load:
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
 kvmap_vmemmap:
 	sub		%g4, %g5, %g5
-	srlx		%g5, 22, %g5
+	srlx		%g5, ILOG2_4MB, %g5
 	sethi		%hi(vmemmap_table), %g1
 	sllx		%g5, 3, %g5
 	or		%g1, %lo(vmemmap_table), %g1

arch/sparc/kernel/nmi.c

@@ -68,27 +68,16 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 static void die_nmi(const char *str, struct pt_regs *regs, int do_panic)
 {
+	int this_cpu = smp_processor_id();
+
 	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0,
 		       pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP)
 		return;
 
-	console_verbose();
-	bust_spinlocks(1);
-
-	printk(KERN_EMERG "%s", str);
-	printk(" on CPU%d, ip %08lx, registers:\n",
-	       smp_processor_id(), regs->tpc);
-	show_regs(regs);
-	dump_stack();
-
-	bust_spinlocks(0);
-
 	if (do_panic || panic_on_oops)
-		panic("Non maskable interrupt");
-
-	nmi_exit();
-	local_irq_enable();
-	do_exit(SIGBUS);
+		panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu);
+	else
+		WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu);
 }
 
 notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs)

arch/sparc/kernel/smp_64.c

@@ -149,7 +149,7 @@ void cpu_panic(void)
 #define NUM_ROUNDS	64	/* magic value */
 #define NUM_ITERS	5	/* likewise */
 
-static DEFINE_SPINLOCK(itc_sync_lock);
+static DEFINE_RAW_SPINLOCK(itc_sync_lock);
 static unsigned long go[SLAVE + 1];
 
 #define DEBUG_TICK_SYNC	0
@@ -257,7 +257,7 @@ static void smp_synchronize_one_tick(int cpu)
 	go[MASTER] = 0;
 	membar_safe("#StoreLoad");
 
-	spin_lock_irqsave(&itc_sync_lock, flags);
+	raw_spin_lock_irqsave(&itc_sync_lock, flags);
 	{
 		for (i = 0; i < NUM_ROUNDS*NUM_ITERS; i++) {
 			while (!go[MASTER])
@@ -268,7 +268,7 @@ static void smp_synchronize_one_tick(int cpu)
 			membar_safe("#StoreLoad");
 		}
 	}
-	spin_unlock_irqrestore(&itc_sync_lock, flags);
+	raw_spin_unlock_irqrestore(&itc_sync_lock, flags);
 }
 
 #if defined(CONFIG_SUN_LDOMS) && defined(CONFIG_HOTPLUG_CPU)

arch/sparc/kernel/sys32.S

@@ -44,7 +44,7 @@ SIGN1(sys32_timer_settime, compat_sys_timer_settime, %o1)
 SIGN1(sys32_io_submit, compat_sys_io_submit, %o1)
 SIGN1(sys32_mq_open, compat_sys_mq_open, %o1)
 SIGN1(sys32_select, compat_sys_select, %o0)
-SIGN3(sys32_futex, compat_sys_futex, %o1, %o2, %o5)
+SIGN1(sys32_futex, compat_sys_futex, %o1)
 SIGN1(sys32_recvfrom, compat_sys_recvfrom, %o0)
 SIGN1(sys32_recvmsg, compat_sys_recvmsg, %o0)
 SIGN1(sys32_sendmsg, compat_sys_sendmsg, %o0)

arch/sparc/kernel/unaligned_64.c

@@ -166,17 +166,23 @@ static unsigned long *fetch_reg_addr(unsigned int reg, struct pt_regs *regs)
 unsigned long compute_effective_address(struct pt_regs *regs,
 					unsigned int insn, unsigned int rd)
 {
+	int from_kernel = (regs->tstate & TSTATE_PRIV) != 0;
 	unsigned int rs1 = (insn >> 14) & 0x1f;
 	unsigned int rs2 = insn & 0x1f;
-	int from_kernel = (regs->tstate & TSTATE_PRIV) != 0;
+	unsigned long addr;
 
 	if (insn & 0x2000) {
 		maybe_flush_windows(rs1, 0, rd, from_kernel);
-		return (fetch_reg(rs1, regs) + sign_extend_imm13(insn));
+		addr = (fetch_reg(rs1, regs) + sign_extend_imm13(insn));
 	} else {
 		maybe_flush_windows(rs1, rs2, rd, from_kernel);
-		return (fetch_reg(rs1, regs) + fetch_reg(rs2, regs));
+		addr = (fetch_reg(rs1, regs) + fetch_reg(rs2, regs));
 	}
+
+	if (!from_kernel && test_thread_flag(TIF_32BIT))
+		addr &= 0xffffffff;
+
+	return addr;
 }
 
 /* This is just to make gcc think die_if_kernel does return... */

arch/sparc/mm/fault_64.c

@@ -96,38 +96,51 @@ static unsigned int get_user_insn(unsigned long tpc)
 	pte_t *ptep, pte;
 	unsigned long pa;
 	u32 insn = 0;
-	unsigned long pstate;
 
-	if (pgd_none(*pgdp))
-		goto outret;
+	if (pgd_none(*pgdp) || unlikely(pgd_bad(*pgdp)))
+		goto out;
 	pudp = pud_offset(pgdp, tpc);
-	if (pud_none(*pudp))
-		goto outret;
-	pmdp = pmd_offset(pudp, tpc);
-	if (pmd_none(*pmdp))
-		goto outret;
+	if (pud_none(*pudp) || unlikely(pud_bad(*pudp)))
+		goto out;
 
 	/* This disables preemption for us as well. */
-	__asm__ __volatile__("rdpr %%pstate, %0" : "=r" (pstate));
-	__asm__ __volatile__("wrpr %0, %1, %%pstate"
-				: : "r" (pstate), "i" (PSTATE_IE));
-	ptep = pte_offset_map(pmdp, tpc);
-	pte = *ptep;
-	if (!pte_present(pte))
-		goto out;
+	local_irq_disable();
 
-	pa  = (pte_pfn(pte) << PAGE_SHIFT);
-	pa += (tpc & ~PAGE_MASK);
+	pmdp = pmd_offset(pudp, tpc);
+	if (pmd_none(*pmdp) || unlikely(pmd_bad(*pmdp)))
+		goto out_irq_enable;
+
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	if (pmd_trans_huge(*pmdp)) {
+		if (pmd_trans_splitting(*pmdp))
+			goto out_irq_enable;
 
-	/* Use phys bypass so we don't pollute dtlb/dcache. */
-	__asm__ __volatile__("lduwa [%1] %2, %0"
-			     : "=r" (insn)
-			     : "r" (pa), "i" (ASI_PHYS_USE_EC));
+		pa  = pmd_pfn(*pmdp) << PAGE_SHIFT;
+		pa += tpc & ~HPAGE_MASK;
 
+		/* Use phys bypass so we don't pollute dtlb/dcache. */
+		__asm__ __volatile__("lduwa [%1] %2, %0"
+				     : "=r" (insn)
+				     : "r" (pa), "i" (ASI_PHYS_USE_EC));
+	} else
+#endif
+	{
+		ptep = pte_offset_map(pmdp, tpc);
+		pte = *ptep;
+		if (pte_present(pte)) {
+			pa  = (pte_pfn(pte) << PAGE_SHIFT);
+			pa += (tpc & ~PAGE_MASK);
+
+			/* Use phys bypass so we don't pollute dtlb/dcache. */
+			__asm__ __volatile__("lduwa [%1] %2, %0"
+					     : "=r" (insn)
+					     : "r" (pa), "i" (ASI_PHYS_USE_EC));
+		}
+		pte_unmap(ptep);
+	}
+out_irq_enable:
+	local_irq_enable();
 out:
-	pte_unmap(ptep);
-	__asm__ __volatile__("wrpr %0, 0x0, %%pstate" : : "r" (pstate));
-outret:
 	return insn;
 }
 
@@ -153,7 +166,8 @@ show_signal_msg(struct pt_regs *regs, int sig, int code,
 }
 
 static void do_fault_siginfo(int code, int sig, struct pt_regs *regs,
-			     unsigned int insn, int fault_code)
+			     unsigned long fault_addr, unsigned int insn,
+			     int fault_code)
 {
 	unsigned long addr;
 	siginfo_t info;
@@ -161,10 +175,18 @@ static void do_fault_siginfo(int code, int sig, struct pt_regs *regs,
 	info.si_code = code;
 	info.si_signo = sig;
 	info.si_errno = 0;
-	if (fault_code & FAULT_CODE_ITLB)
+	if (fault_code & FAULT_CODE_ITLB) {
 		addr = regs->tpc;
-	else
-		addr = compute_effective_address(regs, insn, 0);
+	} else {
+		/* If we were able to probe the faulting instruction, use it
+		 * to compute a precise fault address.  Otherwise use the fault
+		 * time provided address which may only have page granularity.
+		 */
+		if (insn)
+			addr = compute_effective_address(regs, insn, 0);
+		else
+			addr = fault_addr;
+	}
 	info.si_addr = (void __user *) addr;
 	info.si_trapno = 0;
 
@@ -239,7 +261,7 @@ static void __kprobes do_kernel_fault(struct pt_regs *regs, int si_code,
 		/* The si_code was set to make clear whether
 		 * this was a SEGV_MAPERR or SEGV_ACCERR fault.
 		 */
-		do_fault_siginfo(si_code, SIGSEGV, regs, insn, fault_code);
+		do_fault_siginfo(si_code, SIGSEGV, regs, address, insn, fault_code);
 		return;
 	}
 
@@ -525,7 +547,7 @@ do_sigbus:
 	 * Send a sigbus, regardless of whether we were in kernel
 	 * or user mode.
 	 */
-	do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, insn, fault_code);
+	do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, address, insn, fault_code);
 
 	/* Kernel mode? Handle exceptions or die */
 	if (regs->tstate & TSTATE_PRIV)

arch/sparc/mm/gup.c

@@ -73,7 +73,7 @@ static int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 	struct page *head, *page, *tail;
 	int refs;
 
-	if (!pmd_large(pmd))
+	if (!(pmd_val(pmd) & _PAGE_VALID))
 		return 0;
 
 	if (write && !pmd_write(pmd))

arch/sparc/mm/init_64.c

@@ -588,7 +588,7 @@ static void __init remap_kernel(void)
 	int i, tlb_ent = sparc64_highest_locked_tlbent();
 
 	tte_vaddr = (unsigned long) KERNBASE;
-	phys_page = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
+	phys_page = (prom_boot_mapping_phys_low >> ILOG2_4MB) << ILOG2_4MB;
 	tte_data = kern_large_tte(phys_page);
 
 	kern_locked_tte_data = tte_data;
@@ -1881,7 +1881,7 @@ void __init paging_init(void)
 
 	BUILD_BUG_ON(NR_CPUS > 4096);
 
-	kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
+	kern_base = (prom_boot_mapping_phys_low >> ILOG2_4MB) << ILOG2_4MB;
 	kern_size = (unsigned long)&_end - (unsigned long)KERNBASE;
 
 	/* Invalidate both kernel TSBs.  */
@@ -1937,7 +1937,7 @@ void __init paging_init(void)
 	shift = kern_base + PAGE_OFFSET - ((unsigned long)KERNBASE);
 
 	real_end = (unsigned long)_end;
-	num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << 22);
+	num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << ILOG2_4MB);
 	printk("Kernel: Using %d locked TLB entries for main kernel image.\n",
 	       num_kernel_image_mappings);
 
@@ -2094,7 +2094,7 @@ static void __init setup_valid_addr_bitmap_from_pavail(unsigned long *bitmap)
 
 				if (new_start <= old_start &&
 				    new_end >= (old_start + PAGE_SIZE)) {
-					set_bit(old_start >> 22, bitmap);
+					set_bit(old_start >> ILOG2_4MB, bitmap);
 					goto do_next_page;
 				}
 			}
@@ -2143,7 +2143,7 @@ void __init mem_init(void)
 	addr = PAGE_OFFSET + kern_base;
 	last = PAGE_ALIGN(kern_size) + addr;
 	while (addr < last) {
-		set_bit(__pa(addr) >> 22, sparc64_valid_addr_bitmap);
+		set_bit(__pa(addr) >> ILOG2_4MB, sparc64_valid_addr_bitmap);
 		addr += PAGE_SIZE;
 	}
 
@@ -2267,7 +2267,7 @@ int __meminit vmemmap_populate(unsigned long vstart, unsigned long vend,
 		void *block;
 
 		if (!(*vmem_pp & _PAGE_VALID)) {
-			block = vmemmap_alloc_block(1UL << 22, node);
+			block = vmemmap_alloc_block(1UL << ILOG2_4MB, node);
 			if (!block)
 				return -ENOMEM;
 

arch/sparc/mm/tlb.c

@@ -134,7 +134,7 @@ no_cache_flush:
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 static void tlb_batch_pmd_scan(struct mm_struct *mm, unsigned long vaddr,
-			       pmd_t pmd, bool exec)
+			       pmd_t pmd)
 {
 	unsigned long end;
 	pte_t *pte;
@@ -142,8 +142,11 @@ static void tlb_batch_pmd_scan(struct mm_struct *mm, unsigned long vaddr,
 	pte = pte_offset_map(&pmd, vaddr);
 	end = vaddr + HPAGE_SIZE;
 	while (vaddr < end) {
-		if (pte_val(*pte) & _PAGE_VALID)
+		if (pte_val(*pte) & _PAGE_VALID) {
+			bool exec = pte_exec(*pte);
+
 			tlb_batch_add_one(mm, vaddr, exec);
+		}
 		pte++;
 		vaddr += PAGE_SIZE;
 	}
@@ -177,19 +180,30 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 	}
 
 	if (!pmd_none(orig)) {
-		pte_t orig_pte = __pte(pmd_val(orig));
-		bool exec = pte_exec(orig_pte);
-
 		addr &= HPAGE_MASK;
 		if (pmd_trans_huge(orig)) {
+			pte_t orig_pte = __pte(pmd_val(orig));
+			bool exec = pte_exec(orig_pte);
+
 			tlb_batch_add_one(mm, addr, exec);
 			tlb_batch_add_one(mm, addr + REAL_HPAGE_SIZE, exec);
 		} else {
-			tlb_batch_pmd_scan(mm, addr, orig, exec);
+			tlb_batch_pmd_scan(mm, addr, orig);
 		}
 	}
 }
 
+void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
+		     pmd_t *pmdp)
+{
+	pmd_t entry = *pmdp;
+
+	pmd_val(entry) &= ~_PAGE_VALID;
+
+	set_pmd_at(vma->vm_mm, address, pmdp, entry);
+	flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+}
+
 void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
 				pgtable_t pgtable)
 {

arch/x86/Makefile

@@ -79,6 +79,7 @@ else
         UTS_MACHINE := x86_64
         CHECKFLAGS += -D__x86_64__ -m64
 
+        biarch := -m64
         KBUILD_AFLAGS += -m64
         KBUILD_CFLAGS += -m64
 

arch/x86/boot/Makefile

@@ -71,7 +71,7 @@ $(obj)/vmlinux.bin: $(obj)/compressed/vmlinux FORCE
 
 SETUP_OBJS = $(addprefix $(obj)/,$(setup-y))
 
-sed-voffset := -e 's/^\([0-9a-fA-F]*\) . \(_text\|_end\)$$/\#define VO_\2 0x\1/p'
+sed-voffset := -e 's/^\([0-9a-fA-F]*\) [ABCDGRSTVW] \(_text\|_end\)$$/\#define VO_\2 0x\1/p'
 
 quiet_cmd_voffset = VOFFSET $@
       cmd_voffset = $(NM) $< | sed -n $(sed-voffset) > $@
@@ -80,7 +80,7 @@ targets += voffset.h
 $(obj)/voffset.h: vmlinux FORCE
 	$(call if_changed,voffset)
 
-sed-zoffset := -e 's/^\([0-9a-fA-F]*\) . \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p'
+sed-zoffset := -e 's/^\([0-9a-fA-F]*\) [ABCDGRSTVW] \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p'
 
 quiet_cmd_zoffset = ZOFFSET $@
       cmd_zoffset = $(NM) $< | sed -n $(sed-zoffset) > $@

arch/x86/boot/compressed/misc.c

@@ -354,7 +354,7 @@ static void parse_elf(void *output)
 	free(phdrs);
 }
 
-asmlinkage void *decompress_kernel(void *rmode, memptr heap,
+asmlinkage __visible void *decompress_kernel(void *rmode, memptr heap,
 				  unsigned char *input_data,
 				  unsigned long input_len,
 				  unsigned char *output,

arch/x86/include/asm/hpet.h

@@ -63,6 +63,7 @@
 /* hpet memory map physical address */
 extern unsigned long hpet_address;
 extern unsigned long force_hpet_address;
+extern int boot_hpet_disable;
 extern u8 hpet_blockid;
 extern int hpet_force_user;
 extern u8 hpet_msi_disable;

arch/x86/include/uapi/asm/msr-index.h

@@ -384,7 +384,7 @@
 #define MSR_IA32_MISC_ENABLE_MWAIT_BIT			18
 #define MSR_IA32_MISC_ENABLE_MWAIT			(1ULL << MSR_IA32_MISC_ENABLE_MWAIT_BIT)
 #define MSR_IA32_MISC_ENABLE_LIMIT_CPUID_BIT		22
-#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID		(1ULL << MSR_IA32_MISC_ENABLE_LIMIT_CPUID_BIT);
+#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID		(1ULL << MSR_IA32_MISC_ENABLE_LIMIT_CPUID_BIT)
 #define MSR_IA32_MISC_ENABLE_XTPR_DISABLE_BIT		23
 #define MSR_IA32_MISC_ENABLE_XTPR_DISABLE		(1ULL << MSR_IA32_MISC_ENABLE_XTPR_DISABLE_BIT)
 #define MSR_IA32_MISC_ENABLE_XD_DISABLE_BIT		34

arch/x86/kernel/acpi/sleep.c

@@ -31,7 +31,7 @@ static char temp_stack[4096];
  *
  * Wrapper around acpi_enter_sleep_state() to be called by assmebly.
  */
-acpi_status asmlinkage x86_acpi_enter_sleep_state(u8 state)
+acpi_status asmlinkage __visible x86_acpi_enter_sleep_state(u8 state)
 {
 	return acpi_enter_sleep_state(state);
 }

arch/x86/kernel/apic/io_apic.c

@@ -2189,7 +2189,7 @@ void send_cleanup_vector(struct irq_cfg *cfg)
 	cfg->move_in_progress = 0;
 }
 
-asmlinkage void smp_irq_move_cleanup_interrupt(void)
+asmlinkage __visible void smp_irq_move_cleanup_interrupt(void)
 {
 	unsigned vector, me;
 

arch/x86/kernel/cpu/mcheck/therm_throt.c

@@ -429,14 +429,14 @@ static inline void __smp_thermal_interrupt(void)
 	smp_thermal_vector();
 }
 
-asmlinkage void smp_thermal_interrupt(struct pt_regs *regs)
+asmlinkage __visible void smp_thermal_interrupt(struct pt_regs *regs)
 {
 	entering_irq();
 	__smp_thermal_interrupt();
 	exiting_ack_irq();
 }
 
-asmlinkage void smp_trace_thermal_interrupt(struct pt_regs *regs)
+asmlinkage __visible void smp_trace_thermal_interrupt(struct pt_regs *regs)
 {
 	entering_irq();
 	trace_thermal_apic_entry(THERMAL_APIC_VECTOR);

arch/x86/kernel/cpu/mcheck/threshold.c

@@ -24,14 +24,14 @@ static inline void __smp_threshold_interrupt(void)
 	mce_threshold_vector();
 }
 
-asmlinkage void smp_threshold_interrupt(void)
+asmlinkage __visible void smp_threshold_interrupt(void)
 {
 	entering_irq();
 	__smp_threshold_interrupt();
 	exiting_ack_irq();
 }
 
-asmlinkage void smp_trace_threshold_interrupt(void)
+asmlinkage __visible void smp_trace_threshold_interrupt(void)
 {
 	entering_irq();
 	trace_threshold_apic_entry(THRESHOLD_APIC_VECTOR);

arch/x86/kernel/early-quirks.c

@@ -17,6 +17,7 @@
 #include <asm/dma.h>
 #include <asm/io_apic.h>
 #include <asm/apic.h>
+#include <asm/hpet.h>
 #include <asm/iommu.h>
 #include <asm/gart.h>
 #include <asm/irq_remapping.h>
@@ -530,6 +531,15 @@ static void __init intel_graphics_stolen(int num, int slot, int func)
 	}
 }
 
+static void __init force_disable_hpet(int num, int slot, int func)
+{
+#ifdef CONFIG_HPET_TIMER
+	boot_hpet_disable = 1;
+	pr_info("x86/hpet: Will disable the HPET for this platform because it's not reliable\n");
+#endif
+}
+
+
 #define QFLAG_APPLY_ONCE 	0x1
 #define QFLAG_APPLIED		0x2
 #define QFLAG_DONE		(QFLAG_APPLY_ONCE|QFLAG_APPLIED)
@@ -567,6 +577,12 @@ static struct chipset early_qrk[] __initdata = {
 	  PCI_BASE_CLASS_BRIDGE, 0, intel_remapping_check },
 	{ PCI_VENDOR_ID_INTEL, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA, PCI_ANY_ID,
 	  QFLAG_APPLY_ONCE, intel_graphics_stolen },
+	/*
+	 * HPET on current version of Baytrail platform has accuracy
+	 * problems, disable it for now:
+	 */
+	{ PCI_VENDOR_ID_INTEL, 0x0f00,
+		PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet},
 	{}
 };
 

arch/x86/kernel/head32.c

@@ -29,7 +29,7 @@ static void __init i386_default_early_setup(void)
 	reserve_ebda_region();
 }
 
-asmlinkage void __init i386_start_kernel(void)
+asmlinkage __visible void __init i386_start_kernel(void)
 {
 	sanitize_boot_params(&boot_params);
 

arch/x86/kernel/head64.c

@@ -137,7 +137,7 @@ static void __init copy_bootdata(char *real_mode_data)
 	}
 }
 
-asmlinkage void __init x86_64_start_kernel(char * real_mode_data)
+asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data)
 {
 	int i;
 

arch/x86/kernel/hpet.c

@@ -88,7 +88,7 @@ static inline void hpet_clear_mapping(void)
 /*
  * HPET command line enable / disable
  */
-static int boot_hpet_disable;
+int boot_hpet_disable;
 int hpet_force_user;
 static int hpet_verbose;
 

arch/x86/kernel/process_64.c

@@ -52,7 +52,7 @@
 
 asmlinkage extern void ret_from_fork(void);
 
-asmlinkage DEFINE_PER_CPU(unsigned long, old_rsp);
+__visible DEFINE_PER_CPU(unsigned long, old_rsp);
 
 /* Prints also some state that isn't saved in the pt_regs */
 void __show_regs(struct pt_regs *regs, int all)

arch/x86/kernel/reboot.c

@@ -191,6 +191,16 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = {
 		},
 	},
 
+	/* Certec */
+	{       /* Handle problems with rebooting on Certec BPC600 */
+		.callback = set_pci_reboot,
+		.ident = "Certec BPC600",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Certec"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "BPC600"),
+		},
+	},
+
 	/* Dell */
 	{	/* Handle problems with rebooting on Dell DXP061 */
 		.callback = set_bios_reboot,

arch/x86/kernel/smp.c

@@ -168,7 +168,7 @@ static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs)
  * this function calls the 'stop' function on all other CPUs in the system.
  */
 
-asmlinkage void smp_reboot_interrupt(void)
+asmlinkage __visible void smp_reboot_interrupt(void)
 {
 	ack_APIC_irq();
 	irq_enter();

arch/x86/kernel/traps.c

@@ -357,7 +357,7 @@ exit:
  * for scheduling or signal handling. The actual stack switch is done in
  * entry.S
  */
-asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
+asmlinkage __visible __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 {
 	struct pt_regs *regs = eregs;
 	/* Did already sync */
@@ -601,11 +601,11 @@ do_spurious_interrupt_bug(struct pt_regs *regs, long error_code)
 #endif
 }
 
-asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void)
+asmlinkage __visible void __attribute__((weak)) smp_thermal_interrupt(void)
 {
 }
 
-asmlinkage void __attribute__((weak)) smp_threshold_interrupt(void)
+asmlinkage __visible void __attribute__((weak)) smp_threshold_interrupt(void)
 {
 }
 

arch/x86/kernel/vsmp_64.c

@@ -36,7 +36,7 @@ static int irq_routing_comply = 1;
  * and vice versa.
  */
 
-asmlinkage unsigned long vsmp_save_fl(void)
+asmlinkage __visible unsigned long vsmp_save_fl(void)
 {
 	unsigned long flags = native_save_fl();
 
@@ -56,7 +56,7 @@ __visible void vsmp_restore_fl(unsigned long flags)
 }
 PV_CALLEE_SAVE_REGS_THUNK(vsmp_restore_fl);
 
-asmlinkage void vsmp_irq_disable(void)
+asmlinkage __visible void vsmp_irq_disable(void)
 {
 	unsigned long flags = native_save_fl();
 
@@ -64,7 +64,7 @@ asmlinkage void vsmp_irq_disable(void)
 }
 PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_disable);
 
-asmlinkage void vsmp_irq_enable(void)
+asmlinkage __visible void vsmp_irq_enable(void)
 {
 	unsigned long flags = native_save_fl();
 

arch/x86/kernel/vsyscall_gtod.c

@@ -43,7 +43,7 @@ void update_vsyscall(struct timekeeper *tk)
 	vdata->monotonic_time_sec	= tk->xtime_sec
 					+ tk->wall_to_monotonic.tv_sec;
 	vdata->monotonic_time_snsec	= tk->xtime_nsec
-					+ (tk->wall_to_monotonic.tv_nsec
+					+ ((u64)tk->wall_to_monotonic.tv_nsec
 						<< tk->shift);
 	while (vdata->monotonic_time_snsec >=
 					(((u64)NSEC_PER_SEC) << tk->shift)) {

arch/x86/kvm/x86.c

@@ -280,7 +280,7 @@ int kvm_set_apic_base(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 }
 EXPORT_SYMBOL_GPL(kvm_set_apic_base);
 
-asmlinkage void kvm_spurious_fault(void)
+asmlinkage __visible void kvm_spurious_fault(void)
 {
 	/* Fault while not rebooting.  We want the trace. */
 	BUG();

arch/x86/lguest/boot.c

@@ -233,13 +233,13 @@ static void lguest_end_context_switch(struct task_struct *next)
  * flags word contains all kind of stuff, but in practice Linux only cares
  * about the interrupt flag.  Our "save_flags()" just returns that.
  */
-asmlinkage unsigned long lguest_save_fl(void)
+asmlinkage __visible unsigned long lguest_save_fl(void)
 {
 	return lguest_data.irq_enabled;
 }
 
 /* Interrupts go off... */
-asmlinkage void lguest_irq_disable(void)
+asmlinkage __visible void lguest_irq_disable(void)
 {
 	lguest_data.irq_enabled = 0;
 }

arch/x86/lib/msr.c

@@ -76,7 +76,7 @@ static inline int __flip_bit(u32 msr, u8 bit, bool set)
 	if (m1.q == m.q)
 		return 0;
 
-	err = msr_write(msr, &m);
+	err = msr_write(msr, &m1);
 	if (err)
 		return err;
 

arch/x86/math-emu/errors.c

@@ -302,7 +302,7 @@ static struct {
 	      0x242  in div_Xsig.S
  */
 
-asmlinkage void FPU_exception(int n)
+asmlinkage __visible void FPU_exception(int n)
 {
 	int i, int_type;
 
@@ -492,7 +492,7 @@ int real_2op_NaN(FPU_REG const *b, u_char tagb,
 
 /* Invalid arith operation on Valid registers */
 /* Returns < 0 if the exception is unmasked */
-asmlinkage int arith_invalid(int deststnr)
+asmlinkage __visible int arith_invalid(int deststnr)
 {
 
 	EXCEPTION(EX_Invalid);
@@ -507,7 +507,7 @@ asmlinkage int arith_invalid(int deststnr)
 }
 
 /* Divide a finite number by zero */
-asmlinkage int FPU_divide_by_zero(int deststnr, u_char sign)
+asmlinkage __visible int FPU_divide_by_zero(int deststnr, u_char sign)
 {
 	FPU_REG *dest = &st(deststnr);
 	int tag = TAG_Valid;
@@ -539,7 +539,7 @@ int set_precision_flag(int flags)
 }
 
 /* This may be called often, so keep it lean */
-asmlinkage void set_precision_flag_up(void)
+asmlinkage __visible void set_precision_flag_up(void)
 {
 	if (control_word & CW_Precision)
 		partial_status |= (SW_Precision | SW_C1);	/* The masked response */
@@ -548,7 +548,7 @@ asmlinkage void set_precision_flag_up(void)
 }
 
 /* This may be called often, so keep it lean */
-asmlinkage void set_precision_flag_down(void)
+asmlinkage __visible void set_precision_flag_down(void)
 {
 	if (control_word & CW_Precision) {	/* The masked response */
 		partial_status &= ~SW_C1;
@@ -557,7 +557,7 @@ asmlinkage void set_precision_flag_down(void)
 		EXCEPTION(EX_Precision);
 }
 
-asmlinkage int denormal_operand(void)
+asmlinkage __visible int denormal_operand(void)
 {
 	if (control_word & CW_Denormal) {	/* The masked response */
 		partial_status |= SW_Denorm_Op;
@@ -568,7 +568,7 @@ asmlinkage int denormal_operand(void)
 	}
 }
 
-asmlinkage int arith_overflow(FPU_REG *dest)
+asmlinkage __visible int arith_overflow(FPU_REG *dest)
 {
 	int tag = TAG_Valid;
 
@@ -596,7 +596,7 @@ asmlinkage int arith_overflow(FPU_REG *dest)
 
 }
 
-asmlinkage int arith_underflow(FPU_REG *dest)
+asmlinkage __visible int arith_underflow(FPU_REG *dest)
 {
 	int tag = TAG_Valid;
 

arch/x86/platform/efi/early_printk.c

@@ -14,48 +14,92 @@
 
 static const struct font_desc *font;
 static u32 efi_x, efi_y;
+static void *efi_fb;
+static bool early_efi_keep;
 
-static __init void early_efi_clear_scanline(unsigned int y)
+/*
+ * efi earlyprintk need use early_ioremap to map the framebuffer.
+ * But early_ioremap is not usable for earlyprintk=efi,keep, ioremap should
+ * be used instead. ioremap will be available after paging_init() which is
+ * earlier than initcall callbacks. Thus adding this early initcall function
+ * early_efi_map_fb to map the whole efi framebuffer.
+ */
+static __init int early_efi_map_fb(void)
 {
-	unsigned long base, *dst;
-	u16 len;
+	unsigned long base, size;
+
+	if (!early_efi_keep)
+		return 0;
 
 	base = boot_params.screen_info.lfb_base;
-	len = boot_params.screen_info.lfb_linelength;
+	size = boot_params.screen_info.lfb_size;
+	efi_fb = ioremap(base, size);
+
+	return efi_fb ? 0 : -ENOMEM;
+}
+early_initcall(early_efi_map_fb);
+
+/*
+ * early_efi_map maps efi framebuffer region [start, start + len -1]
+ * In case earlyprintk=efi,keep we have the whole framebuffer mapped already
+ * so just return the offset efi_fb + start.
+ */
+static __init_refok void *early_efi_map(unsigned long start, unsigned long len)
+{
+	unsigned long base;
+
+	base = boot_params.screen_info.lfb_base;
+
+	if (efi_fb)
+		return (efi_fb + start);
+	else
+		return early_ioremap(base + start, len);
+}
 
-	dst = early_ioremap(base + y*len, len);
+static __init_refok void early_efi_unmap(void *addr, unsigned long len)
+{
+	if (!efi_fb)
+		early_iounmap(addr, len);
+}
+
+static void early_efi_clear_scanline(unsigned int y)
+{
+	unsigned long *dst;
+	u16 len;
+
+	len = boot_params.screen_info.lfb_linelength;
+	dst = early_efi_map(y*len, len);
 	if (!dst)
 		return;
 
 	memset(dst, 0, len);
-	early_iounmap(dst, len);
+	early_efi_unmap(dst, len);
 }
 
-static __init void early_efi_scroll_up(void)
+static void early_efi_scroll_up(void)
 {
-	unsigned long base, *dst, *src;
+	unsigned long *dst, *src;
 	u16 len;
 	u32 i, height;
 
-	base = boot_params.screen_info.lfb_base;
 	len = boot_params.screen_info.lfb_linelength;
 	height = boot_params.screen_info.lfb_height;
 
 	for (i = 0; i < height - font->height; i++) {
-		dst = early_ioremap(base + i*len, len);
+		dst = early_efi_map(i*len, len);
 		if (!dst)
 			return;
 
-		src = early_ioremap(base + (i + font->height) * len, len);
+		src = early_efi_map((i + font->height) * len, len);
 		if (!src) {
-			early_iounmap(dst, len);
+			early_efi_unmap(dst, len);
 			return;
 		}
 
 		memmove(dst, src, len);
 
-		early_iounmap(src, len);
-		early_iounmap(dst, len);
+		early_efi_unmap(src, len);
+		early_efi_unmap(dst, len);
 	}
 }
 
@@ -79,16 +123,14 @@ static void early_efi_write_char(u32 *dst, unsigned char c, unsigned int h)
 	}
 }
 
-static __init void
+static void
 early_efi_write(struct console *con, const char *str, unsigned int num)
 {
 	struct screen_info *si;
-	unsigned long base;
 	unsigned int len;
 	const char *s;
 	void *dst;
 
-	base = boot_params.screen_info.lfb_base;
 	si = &boot_params.screen_info;
 	len = si->lfb_linelength;
 
@@ -109,7 +151,7 @@ early_efi_write(struct console *con, const char *str, unsigned int num)
 		for (h = 0; h < font->height; h++) {
 			unsigned int n, x;
 
-			dst = early_ioremap(base + (efi_y + h) * len, len);
+			dst = early_efi_map((efi_y + h) * len, len);
 			if (!dst)
 				return;
 
@@ -123,7 +165,7 @@ early_efi_write(struct console *con, const char *str, unsigned int num)
 				s++;
 			}
 
-			early_iounmap(dst, len);
+			early_efi_unmap(dst, len);
 		}
 
 		num -= count;
@@ -179,6 +221,9 @@ static __init int early_efi_setup(struct console *con, char *options)
 	for (i = 0; i < (yres - efi_y) / font->height; i++)
 		early_efi_scroll_up();
 
+	/* early_console_register will unset CON_BOOT in case ,keep */
+	if (!(con->flags & CON_BOOT))
+		early_efi_keep = true;
 	return 0;
 }
 

arch/x86/platform/olpc/olpc-xo1-pm.c

@@ -75,7 +75,7 @@ static int xo1_power_state_enter(suspend_state_t pm_state)
 	return 0;
 }
 
-asmlinkage int xo1_do_sleep(u8 sleep_state)
+asmlinkage __visible int xo1_do_sleep(u8 sleep_state)
 {
 	void *pgd_addr = __va(read_cr3());
 

arch/x86/power/hibernate_64.c

@@ -23,7 +23,7 @@
 extern __visible const void __nosave_begin, __nosave_end;
 
 /* Defined in hibernate_asm_64.S */
-extern asmlinkage int restore_image(void);
+extern asmlinkage __visible int restore_image(void);
 
 /*
  * Address to jump to in the last phase of restore in order to get to the image

arch/x86/xen/enlighten.c

@@ -1515,7 +1515,7 @@ static void __init xen_pvh_early_guest_init(void)
 }
 
 /* First C function to be called on Xen boot */
-asmlinkage void __init xen_start_kernel(void)
+asmlinkage __visible void __init xen_start_kernel(void)
 {
 	struct physdev_set_iopl set_iopl;
 	int rc;

arch/x86/xen/irq.c

@@ -23,7 +23,7 @@ void xen_force_evtchn_callback(void)
 	(void)HYPERVISOR_xen_version(0, NULL);
 }
 
-asmlinkage unsigned long xen_save_fl(void)
+asmlinkage __visible unsigned long xen_save_fl(void)
 {
 	struct vcpu_info *vcpu;
 	unsigned long flags;
@@ -63,7 +63,7 @@ __visible void xen_restore_fl(unsigned long flags)
 }
 PV_CALLEE_SAVE_REGS_THUNK(xen_restore_fl);
 
-asmlinkage void xen_irq_disable(void)
+asmlinkage __visible void xen_irq_disable(void)
 {
 	/* There's a one instruction preempt window here.  We need to
 	   make sure we're don't switch CPUs between getting the vcpu
@@ -74,7 +74,7 @@ asmlinkage void xen_irq_disable(void)
 }
 PV_CALLEE_SAVE_REGS_THUNK(xen_irq_disable);
 
-asmlinkage void xen_irq_enable(void)
+asmlinkage __visible void xen_irq_enable(void)
 {
 	struct vcpu_info *vcpu;
 

arch/xtensa/Kconfig

@@ -14,6 +14,7 @@ config XTENSA
 	select GENERIC_PCI_IOMAP
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select ARCH_WANT_OPTIONAL_GPIOLIB
+	select BUILDTIME_EXTABLE_SORT
 	select CLONE_BACKWARDS
 	select IRQ_DOMAIN
 	select HAVE_OPROFILE
@@ -189,6 +190,24 @@ config INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX
 
 	  If in doubt, say Y.
 
+config HIGHMEM
+	bool "High Memory Support"
+	help
+	  Linux can use the full amount of RAM in the system by
+	  default. However, the default MMUv2 setup only maps the
+	  lowermost 128 MB of memory linearly to the areas starting
+	  at 0xd0000000 (cached) and 0xd8000000 (uncached).
+	  When there are more than 128 MB memory in the system not
+	  all of it can be "permanently mapped" by the kernel.
+	  The physical memory that's not permanently mapped is called
+	  "high memory".
+
+	  If you are compiling a kernel which will never run on a
+	  machine with more than 128 MB total physical RAM, answer
+	  N here.
+
+	  If unsure, say Y.
+
 endmenu
 
 config XTENSA_CALIBRATE_CCOUNT
@@ -224,7 +243,6 @@ choice
 
 config XTENSA_PLATFORM_ISS
 	bool "ISS"
-	depends on TTY
 	select XTENSA_CALIBRATE_CCOUNT
 	select SERIAL_CONSOLE
 	help

arch/xtensa/boot/dts/kc705.dts

@@ -0,0 +1,11 @@
+/dts-v1/;
+/include/ "xtfpga.dtsi"
+/include/ "xtfpga-flash-128m.dtsi"
+
+/ {
+	compatible = "cdns,xtensa-kc705";
+	memory@0 {
+		device_type = "memory";
+		reg = <0x00000000 0x08000000>;
+	};
+};

arch/xtensa/boot/dts/xtfpga-flash-128m.dtsi

@@ -0,0 +1,28 @@
+/ {
+	soc {
+		flash: flash@00000000 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "cfi-flash";
+			reg = <0x00000000 0x08000000>;
+			bank-width = <2>;
+			device-width = <2>;
+			partition@0x0 {
+				label = "data";
+				reg = <0x00000000 0x06000000>;
+			};
+			partition@0x6000000 {
+				label = "boot loader area";
+				reg = <0x06000000 0x00800000>;
+			};
+			partition@0x6800000 {
+				label = "kernel image";
+				reg = <0x06800000 0x017e0000>;
+			};
+			partition@0x7fe0000 {
+				label = "boot environment";
+				reg = <0x07fe0000 0x00020000>;
+			};
+		};
+        };
+};

arch/xtensa/boot/dts/xtfpga-flash-16m.dtsi

@@ -1,26 +1,28 @@
 / {
-	flash: flash@f8000000 {
-		#address-cells = <1>;
-		#size-cells = <1>;
-		compatible = "cfi-flash";
-		reg = <0xf8000000 0x01000000>;
-		bank-width = <2>;
-		device-width = <2>;
-		partition@0x0 {
-			label = "boot loader area";
-			reg = <0x00000000 0x00400000>;
+	soc {
+		flash: flash@08000000 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "cfi-flash";
+			reg = <0x08000000 0x01000000>;
+			bank-width = <2>;
+			device-width = <2>;
+			partition@0x0 {
+				label = "boot loader area";
+				reg = <0x00000000 0x00400000>;
+			};
+			partition@0x400000 {
+				label = "kernel image";
+				reg = <0x00400000 0x00600000>;
+			};
+			partition@0xa00000 {
+				label = "data";
+				reg = <0x00a00000 0x005e0000>;
+			};
+			partition@0xfe0000 {
+				label = "boot environment";
+				reg = <0x00fe0000 0x00020000>;
+			};
 		};
-		partition@0x400000 {
-			label = "kernel image";
-			reg = <0x00400000 0x00600000>;
-		};
-		partition@0xa00000 {
-			label = "data";
-			reg = <0x00a00000 0x005e0000>;
-		};
-		partition@0xfe0000 {
-			label = "boot environment";
-			reg = <0x00fe0000 0x00020000>;
-		};
-        };
+	};
 };

arch/xtensa/boot/dts/xtfpga-flash-4m.dtsi

@@ -1,18 +1,20 @@
 / {
-	flash: flash@f8000000 {
-		#address-cells = <1>;
-		#size-cells = <1>;
-		compatible = "cfi-flash";
-		reg = <0xf8000000 0x00400000>;
-		bank-width = <2>;
-		device-width = <2>;
-		partition@0x0 {
-			label = "boot loader area";
-			reg = <0x00000000 0x003f0000>;
+	soc {
+		flash: flash@08000000 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "cfi-flash";
+			reg = <0x08000000 0x00400000>;
+			bank-width = <2>;
+			device-width = <2>;
+			partition@0x0 {
+				label = "boot loader area";
+				reg = <0x00000000 0x003f0000>;
+			};
+			partition@0x3f0000 {
+				label = "boot environment";
+				reg = <0x003f0000 0x00010000>;
+			};
 		};
-		partition@0x3f0000 {
-			label = "boot environment";
-			reg = <0x003f0000 0x00010000>;
-		};
-        };
+	};
 };

arch/xtensa/boot/dts/xtfpga.dtsi

@@ -42,21 +42,28 @@
 		};
 	};
 
-	serial0: serial@fd050020 {
-		device_type = "serial";
-		compatible = "ns16550a";
-		no-loopback-test;
-		reg = <0xfd050020 0x20>;
-		reg-shift = <2>;
-		interrupts = <0 1>; /* external irq 0 */
-		clocks = <&osc>;
-	};
+	soc {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "simple-bus";
+		ranges = <0x00000000 0xf0000000 0x10000000>;
 
-	enet0: ethoc@fd030000 {
-		compatible = "opencores,ethoc";
-		reg = <0xfd030000 0x4000 0xfd800000 0x4000>;
-		interrupts = <1 1>; /* external irq 1 */
-		local-mac-address = [00 50 c2 13 6f 00];
-		clocks = <&osc>;
+		serial0: serial@0d050020 {
+			device_type = "serial";
+			compatible = "ns16550a";
+			no-loopback-test;
+			reg = <0x0d050020 0x20>;
+			reg-shift = <2>;
+			interrupts = <0 1>; /* external irq 0 */
+			clocks = <&osc>;
+		};
+
+		enet0: ethoc@0d030000 {
+			compatible = "opencores,ethoc";
+			reg = <0x0d030000 0x4000 0x0d800000 0x4000>;
+			interrupts = <1 1>; /* external irq 1 */
+			local-mac-address = [00 50 c2 13 6f 00];
+			clocks = <&osc>;
+		};
 	};
 };

arch/xtensa/include/asm/bootparam.h

@@ -37,23 +37,14 @@ typedef struct bp_tag {
 	unsigned long data[0];	/* data */
 } bp_tag_t;
 
-typedef struct meminfo {
+struct bp_meminfo {
 	unsigned long type;
 	unsigned long start;
 	unsigned long end;
-} meminfo_t;
-
-#define SYSMEM_BANKS_MAX 5
+};
 
 #define MEMORY_TYPE_CONVENTIONAL	0x1000
 #define MEMORY_TYPE_NONE		0x2000
 
-typedef struct sysmem_info {
-	int nr_banks;
-	meminfo_t bank[SYSMEM_BANKS_MAX];
-} sysmem_info_t;
-
-extern sysmem_info_t sysmem;
-
 #endif
 #endif

arch/xtensa/include/asm/fixmap.h

@@ -0,0 +1,58 @@
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1998 Ingo Molnar
+ *
+ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ */
+
+#ifndef _ASM_FIXMAP_H
+#define _ASM_FIXMAP_H
+
+#include <asm/pgtable.h>
+#ifdef CONFIG_HIGHMEM
+#include <linux/threads.h>
+#include <asm/kmap_types.h>
+#endif
+
+/*
+ * Here we define all the compile-time 'special' virtual
+ * addresses. The point is to have a constant address at
+ * compile time, but to set the physical address only
+ * in the boot process. We allocate these special  addresses
+ * from the end of the consistent memory region backwards.
+ * Also this lets us do fail-safe vmalloc(), we
+ * can guarantee that these special addresses and
+ * vmalloc()-ed addresses never overlap.
+ *
+ * these 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages. (or larger if used with an increment
+ * higher than 1) use fixmap_set(idx,phys) to associate
+ * physical memory with fixmap indices.
+ */
+enum fixed_addresses {
+#ifdef CONFIG_HIGHMEM
+	/* reserved pte's for temporary kernel mappings */
+	FIX_KMAP_BEGIN,
+	FIX_KMAP_END = FIX_KMAP_BEGIN + (KM_TYPE_NR * NR_CPUS) - 1,
+#endif
+	__end_of_fixed_addresses
+};
+
+#define FIXADDR_TOP     (VMALLOC_START - PAGE_SIZE)
+#define FIXADDR_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_START	((FIXADDR_TOP - FIXADDR_SIZE) & PMD_MASK)
+
+#include <asm-generic/fixmap.h>
+
+#define kmap_get_fixmap_pte(vaddr) \
+	pte_offset_kernel( \
+		pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)), (vaddr)), \
+		(vaddr) \
+	)
+
+#endif

arch/xtensa/include/asm/highmem.h

@@ -6,11 +6,54 @@
  * this archive for more details.
  *
  * Copyright (C) 2003 - 2005 Tensilica Inc.
+ * Copyright (C) 2014 Cadence Design Systems Inc.
  */
 
 #ifndef _XTENSA_HIGHMEM_H
 #define _XTENSA_HIGHMEM_H
 
-extern void flush_cache_kmaps(void);
+#include <asm/cacheflush.h>
+#include <asm/fixmap.h>
+#include <asm/kmap_types.h>
+#include <asm/pgtable.h>
+
+#define PKMAP_BASE		(FIXADDR_START - PMD_SIZE)
+#define LAST_PKMAP		PTRS_PER_PTE
+#define LAST_PKMAP_MASK		(LAST_PKMAP - 1)
+#define PKMAP_NR(virt)		(((virt) - PKMAP_BASE) >> PAGE_SHIFT)
+#define PKMAP_ADDR(nr)		(PKMAP_BASE + ((nr) << PAGE_SHIFT))
+
+#define kmap_prot		PAGE_KERNEL
+
+extern pte_t *pkmap_page_table;
+
+void *kmap_high(struct page *page);
+void kunmap_high(struct page *page);
+
+static inline void *kmap(struct page *page)
+{
+	BUG_ON(in_interrupt());
+	if (!PageHighMem(page))
+		return page_address(page);
+	return kmap_high(page);
+}
+
+static inline void kunmap(struct page *page)
+{
+	BUG_ON(in_interrupt());
+	if (!PageHighMem(page))
+		return;
+	kunmap_high(page);
+}
+
+static inline void flush_cache_kmaps(void)
+{
+	flush_cache_all();
+}
+
+void *kmap_atomic(struct page *page);
+void __kunmap_atomic(void *kvaddr);
+
+void kmap_init(void);
 
 #endif

arch/xtensa/include/asm/pgtable.h

@@ -310,6 +310,10 @@ set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pteval)
 	update_pte(ptep, pteval);
 }
 
+static inline void set_pte(pte_t *ptep, pte_t pteval)
+{
+	update_pte(ptep, pteval);
+}
 
 static inline void
 set_pmd(pmd_t *pmdp, pmd_t pmdval)

arch/xtensa/include/asm/sysmem.h

@@ -0,0 +1,38 @@
+/*
+ * sysmem-related prototypes.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2014 Cadence Design Systems Inc.
+ */
+
+#ifndef _XTENSA_SYSMEM_H
+#define _XTENSA_SYSMEM_H
+
+#define SYSMEM_BANKS_MAX 31
+
+struct meminfo {
+	unsigned long start;
+	unsigned long end;
+};
+
+/*
+ * Bank array is sorted by .start.
+ * Banks don't overlap and there's at least one page gap
+ * between adjacent bank entries.
+ */
+struct sysmem_info {
+	int nr_banks;
+	struct meminfo bank[SYSMEM_BANKS_MAX];
+};
+
+extern struct sysmem_info sysmem;
+
+int add_sysmem_bank(unsigned long start, unsigned long end);
+int mem_reserve(unsigned long, unsigned long, int);
+void bootmem_init(void);
+void zones_init(void);
+
+#endif /* _XTENSA_SYSMEM_H */

arch/xtensa/include/asm/tlbflush.h

@@ -36,6 +36,7 @@ void local_flush_tlb_page(struct vm_area_struct *vma,
 		unsigned long page);
 void local_flush_tlb_range(struct vm_area_struct *vma,
 		unsigned long start, unsigned long end);
+void local_flush_tlb_kernel_range(unsigned long start, unsigned long end);
 
 #ifdef CONFIG_SMP
 
@@ -44,12 +45,7 @@ void flush_tlb_mm(struct mm_struct *);
 void flush_tlb_page(struct vm_area_struct *, unsigned long);
 void flush_tlb_range(struct vm_area_struct *, unsigned long,
 		unsigned long);
-
-static inline void flush_tlb_kernel_range(unsigned long start,
-		unsigned long end)
-{
-	flush_tlb_all();
-}
+void flush_tlb_kernel_range(unsigned long start, unsigned long end);
 
 #else /* !CONFIG_SMP */
 
@@ -58,7 +54,8 @@ static inline void flush_tlb_kernel_range(unsigned long start,
 #define flush_tlb_page(vma, page)	   local_flush_tlb_page(vma, page)
 #define flush_tlb_range(vma, vmaddr, end)  local_flush_tlb_range(vma, vmaddr, \
 								 end)
-#define flush_tlb_kernel_range(start, end) local_flush_tlb_all()
+#define flush_tlb_kernel_range(start, end) local_flush_tlb_kernel_range(start, \
+									end)
 
 #endif /* CONFIG_SMP */
 

arch/xtensa/kernel/setup.c

@@ -50,6 +50,7 @@
 #include <asm/param.h>
 #include <asm/traps.h>
 #include <asm/smp.h>
+#include <asm/sysmem.h>
 
 #include <platform/hardware.h>
 
@@ -88,12 +89,6 @@ static char __initdata command_line[COMMAND_LINE_SIZE];
 static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE;
 #endif
 
-sysmem_info_t __initdata sysmem;
-
-extern int mem_reserve(unsigned long, unsigned long, int);
-extern void bootmem_init(void);
-extern void zones_init(void);
-
 /*
  * Boot parameter parsing.
  *
@@ -113,31 +108,14 @@ typedef struct tagtable {
 
 /* parse current tag */
 
-static int __init add_sysmem_bank(unsigned long type, unsigned long start,
-		unsigned long end)
-{
-	if (sysmem.nr_banks >= SYSMEM_BANKS_MAX) {
-		printk(KERN_WARNING
-				"Ignoring memory bank 0x%08lx size %ldKB\n",
-				start, end - start);
-		return -EINVAL;
-	}
-	sysmem.bank[sysmem.nr_banks].type  = type;
-	sysmem.bank[sysmem.nr_banks].start = PAGE_ALIGN(start);
-	sysmem.bank[sysmem.nr_banks].end   = end & PAGE_MASK;
-	sysmem.nr_banks++;
-
-	return 0;
-}
-
 static int __init parse_tag_mem(const bp_tag_t *tag)
 {
-	meminfo_t *mi = (meminfo_t *)(tag->data);
+	struct bp_meminfo *mi = (struct bp_meminfo *)(tag->data);
 
 	if (mi->type != MEMORY_TYPE_CONVENTIONAL)
 		return -1;
 
-	return add_sysmem_bank(mi->type, mi->start, mi->end);
+	return add_sysmem_bank(mi->start, mi->end);
 }
 
 __tagtable(BP_TAG_MEMORY, parse_tag_mem);
@@ -146,8 +124,8 @@ __tagtable(BP_TAG_MEMORY, parse_tag_mem);
 
 static int __init parse_tag_initrd(const bp_tag_t* tag)
 {
-	meminfo_t* mi;
-	mi = (meminfo_t*)(tag->data);
+	struct bp_meminfo *mi = (struct bp_meminfo *)(tag->data);
+
 	initrd_start = (unsigned long)__va(mi->start);
 	initrd_end = (unsigned long)__va(mi->end);
 
@@ -255,7 +233,7 @@ void __init early_init_dt_add_memory_arch(u64 base, u64 size)
 		return;
 
 	size &= PAGE_MASK;
-	add_sysmem_bank(MEMORY_TYPE_CONVENTIONAL, base, base + size);
+	add_sysmem_bank(base, base + size);
 }
 
 void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align)
@@ -292,8 +270,6 @@ device_initcall(xtensa_device_probe);
 
 void __init init_arch(bp_tag_t *bp_start)
 {
-	sysmem.nr_banks = 0;
-
 	/* Parse boot parameters */
 
 	if (bp_start)
@@ -304,10 +280,9 @@ void __init init_arch(bp_tag_t *bp_start)
 #endif
 
 	if (sysmem.nr_banks == 0) {
-		sysmem.nr_banks = 1;
-		sysmem.bank[0].start = PLATFORM_DEFAULT_MEM_START;
-		sysmem.bank[0].end = PLATFORM_DEFAULT_MEM_START
-				     + PLATFORM_DEFAULT_MEM_SIZE;
+		add_sysmem_bank(PLATFORM_DEFAULT_MEM_START,
+				PLATFORM_DEFAULT_MEM_START +
+				PLATFORM_DEFAULT_MEM_SIZE);
 	}
 
 #ifdef CONFIG_CMDLINE_BOOL
@@ -487,7 +462,7 @@ void __init setup_arch(char **cmdline_p)
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (initrd_start < initrd_end) {
 		initrd_is_mapped = mem_reserve(__pa(initrd_start),
-					       __pa(initrd_end), 0);
+					       __pa(initrd_end), 0) == 0;
 		initrd_below_start_ok = 1;
 	} else {
 		initrd_start = 0;
@@ -532,6 +507,7 @@ void __init setup_arch(char **cmdline_p)
 		    __pa(&_Level6InterruptVector_text_end), 0);
 #endif
 
+	parse_early_param();
 	bootmem_init();
 
 	unflatten_and_copy_device_tree();

arch/xtensa/kernel/smp.c

@@ -496,6 +496,21 @@ void flush_tlb_range(struct vm_area_struct *vma,
 	on_each_cpu(ipi_flush_tlb_range, &fd, 1);
 }
 
+static void ipi_flush_tlb_kernel_range(void *arg)
+{
+	struct flush_data *fd = arg;
+	local_flush_tlb_kernel_range(fd->addr1, fd->addr2);
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	struct flush_data fd = {
+		.addr1 = start,
+		.addr2 = end,
+	};
+	on_each_cpu(ipi_flush_tlb_kernel_range, &fd, 1);
+}
+
 /* Cache flush functions */
 
 static void ipi_flush_cache_all(void *arg)

arch/xtensa/kernel/xtensa_ksyms.c

@@ -20,6 +20,7 @@
 #include <linux/in6.h>
 
 #include <asm/uaccess.h>
+#include <asm/cacheflush.h>
 #include <asm/checksum.h>
 #include <asm/dma.h>
 #include <asm/io.h>
@@ -105,6 +106,7 @@ EXPORT_SYMBOL(csum_partial_copy_generic);
  * Architecture-specific symbols
  */
 EXPORT_SYMBOL(__xtensa_copy_user);
+EXPORT_SYMBOL(__invalidate_icache_range);
 
 /*
  * Kernel hacking ...
@@ -127,3 +129,8 @@ EXPORT_SYMBOL(common_exception_return);
 #ifdef CONFIG_FUNCTION_TRACER
 EXPORT_SYMBOL(_mcount);
 #endif
+
+EXPORT_SYMBOL(__invalidate_dcache_range);
+#if XCHAL_DCACHE_IS_WRITEBACK
+EXPORT_SYMBOL(__flush_dcache_range);
+#endif

arch/xtensa/mm/Makefile

@@ -4,3 +4,4 @@
 
 obj-y			:= init.o cache.o misc.o
 obj-$(CONFIG_MMU)	+= fault.o mmu.o tlb.o
+obj-$(CONFIG_HIGHMEM)	+= highmem.o

arch/xtensa/mm/cache.c

@@ -59,6 +59,10 @@
  *
  */
 
+#if (DCACHE_WAY_SIZE > PAGE_SIZE) && defined(CONFIG_HIGHMEM)
+#error "HIGHMEM is not supported on cores with aliasing cache."
+#endif
+
 #if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK
 
 /*
@@ -179,10 +183,11 @@ update_mmu_cache(struct vm_area_struct * vma, unsigned long addr, pte_t *ptep)
 #else
 	if (!PageReserved(page) && !test_bit(PG_arch_1, &page->flags)
 	    && (vma->vm_flags & VM_EXEC) != 0) {
-	    	unsigned long paddr = (unsigned long) page_address(page);
+		unsigned long paddr = (unsigned long)kmap_atomic(page);
 		__flush_dcache_page(paddr);
 		__invalidate_icache_page(paddr);
 		set_bit(PG_arch_1, &page->flags);
+		kunmap_atomic((void *)paddr);
 	}
 #endif
 }

arch/xtensa/mm/highmem.c

@@ -0,0 +1,72 @@
+/*
+ * High memory support for Xtensa architecture
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of
+ * this archive for more details.
+ *
+ * Copyright (C) 2014 Cadence Design Systems Inc.
+ */
+
+#include <linux/export.h>
+#include <linux/highmem.h>
+#include <asm/tlbflush.h>
+
+static pte_t *kmap_pte;
+
+void *kmap_atomic(struct page *page)
+{
+	enum fixed_addresses idx;
+	unsigned long vaddr;
+	int type;
+
+	pagefault_disable();
+	if (!PageHighMem(page))
+		return page_address(page);
+
+	type = kmap_atomic_idx_push();
+	idx = type + KM_TYPE_NR * smp_processor_id();
+	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+#ifdef CONFIG_DEBUG_HIGHMEM
+	BUG_ON(!pte_none(*(kmap_pte - idx)));
+#endif
+	set_pte(kmap_pte - idx, mk_pte(page, PAGE_KERNEL_EXEC));
+
+	return (void *)vaddr;
+}
+EXPORT_SYMBOL(kmap_atomic);
+
+void __kunmap_atomic(void *kvaddr)
+{
+	int idx, type;
+
+	if (kvaddr >= (void *)FIXADDR_START &&
+	    kvaddr < (void *)FIXADDR_TOP) {
+		type = kmap_atomic_idx();
+		idx = type + KM_TYPE_NR * smp_processor_id();
+
+		/*
+		 * Force other mappings to Oops if they'll try to access this
+		 * pte without first remap it.  Keeping stale mappings around
+		 * is a bad idea also, in case the page changes cacheability
+		 * attributes or becomes a protected page in a hypervisor.
+		 */
+		pte_clear(&init_mm, kvaddr, kmap_pte - idx);
+		local_flush_tlb_kernel_range((unsigned long)kvaddr,
+					     (unsigned long)kvaddr + PAGE_SIZE);
+
+		kmap_atomic_idx_pop();
+	}
+
+	pagefault_enable();
+}
+EXPORT_SYMBOL(__kunmap_atomic);
+
+void __init kmap_init(void)
+{
+	unsigned long kmap_vstart;
+
+	/* cache the first kmap pte */
+	kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
+	kmap_pte = kmap_get_fixmap_pte(kmap_vstart);
+}

arch/xtensa/mm/init.c

@@ -8,6 +8,7 @@
  * for more details.
  *
  * Copyright (C) 2001 - 2005 Tensilica Inc.
+ * Copyright (C) 2014 Cadence Design Systems Inc.
  *
  * Chris Zankel	<chris@zankel.net>
  * Joe Taylor	<joe@tensilica.com, joetylr@yahoo.com>
@@ -19,6 +20,7 @@
 #include <linux/errno.h>
 #include <linux/bootmem.h>
 #include <linux/gfp.h>
+#include <linux/highmem.h>
 #include <linux/swap.h>
 #include <linux/mman.h>
 #include <linux/nodemask.h>
@@ -27,11 +29,133 @@
 #include <asm/bootparam.h>
 #include <asm/page.h>
 #include <asm/sections.h>
+#include <asm/sysmem.h>
+
+struct sysmem_info sysmem __initdata;
+
+static void __init sysmem_dump(void)
+{
+	unsigned i;
+
+	pr_debug("Sysmem:\n");
+	for (i = 0; i < sysmem.nr_banks; ++i)
+		pr_debug("  0x%08lx - 0x%08lx (%ldK)\n",
+			 sysmem.bank[i].start, sysmem.bank[i].end,
+			 (sysmem.bank[i].end - sysmem.bank[i].start) >> 10);
+}
+
+/*
+ * Find bank with maximal .start such that bank.start <= start
+ */
+static inline struct meminfo * __init find_bank(unsigned long start)
+{
+	unsigned i;
+	struct meminfo *it = NULL;
+
+	for (i = 0; i < sysmem.nr_banks; ++i)
+		if (sysmem.bank[i].start <= start)
+			it = sysmem.bank + i;
+		else
+			break;
+	return it;
+}
+
+/*
+ * Move all memory banks starting at 'from' to a new place at 'to',
+ * adjust nr_banks accordingly.
+ * Both 'from' and 'to' must be inside the sysmem.bank.
+ *
+ * Returns: 0 (success), -ENOMEM (not enough space in the sysmem.bank).
+ */
+static int __init move_banks(struct meminfo *to, struct meminfo *from)
+{
+	unsigned n = sysmem.nr_banks - (from - sysmem.bank);
+
+	if (to > from && to - from + sysmem.nr_banks > SYSMEM_BANKS_MAX)
+		return -ENOMEM;
+	if (to != from)
+		memmove(to, from, n * sizeof(struct meminfo));
+	sysmem.nr_banks += to - from;
+	return 0;
+}
+
+/*
+ * Add new bank to sysmem. Resulting sysmem is the union of bytes of the
+ * original sysmem and the new bank.
+ *
+ * Returns: 0 (success), < 0 (error)
+ */
+int __init add_sysmem_bank(unsigned long start, unsigned long end)
+{
+	unsigned i;
+	struct meminfo *it = NULL;
+	unsigned long sz;
+	unsigned long bank_sz = 0;
+
+	if (start == end ||
+	    (start < end) != (PAGE_ALIGN(start) < (end & PAGE_MASK))) {
+		pr_warn("Ignoring small memory bank 0x%08lx size: %ld bytes\n",
+			start, end - start);
+		return -EINVAL;
+	}
+
+	start = PAGE_ALIGN(start);
+	end &= PAGE_MASK;
+	sz = end - start;
+
+	it = find_bank(start);
+
+	if (it)
+		bank_sz = it->end - it->start;
+
+	if (it && bank_sz >= start - it->start) {
+		if (end - it->start > bank_sz)
+			it->end = end;
+		else
+			return 0;
+	} else {
+		if (!it)
+			it = sysmem.bank;
+		else
+			++it;
+
+		if (it - sysmem.bank < sysmem.nr_banks &&
+		    it->start - start <= sz) {
+			it->start = start;
+			if (it->end - it->start < sz)
+				it->end = end;
+			else
+				return 0;
+		} else {
+			if (move_banks(it + 1, it) < 0) {
+				pr_warn("Ignoring memory bank 0x%08lx size %ld bytes\n",
+					start, end - start);
+				return -EINVAL;
+			}
+			it->start = start;
+			it->end = end;
+			return 0;
+		}
+	}
+	sz = it->end - it->start;
+	for (i = it + 1 - sysmem.bank; i < sysmem.nr_banks; ++i)
+		if (sysmem.bank[i].start - it->start <= sz) {
+			if (sz < sysmem.bank[i].end - it->start)
+				it->end = sysmem.bank[i].end;
+		} else {
+			break;
+		}
+
+	move_banks(it + 1, sysmem.bank + i);
+	return 0;
+}
 
 /*
  * mem_reserve(start, end, must_exist)
  *
  * Reserve some memory from the memory pool.
+ * If must_exist is set and a part of the region being reserved does not exist
+ * memory map is not altered.
  *
  * Parameters:
  *  start	Start of region,
@@ -39,53 +163,69 @@
  *  must_exist	Must exist in memory pool.
  *
  * Returns:
- *  0 (memory area couldn't be mapped)
- * -1 (success)
+ *  0 (success)
+ *  < 0 (error)
  */
 
 int __init mem_reserve(unsigned long start, unsigned long end, int must_exist)
 {
-	int i;
-
-	if (start == end)
-		return 0;
+	struct meminfo *it;
+	struct meminfo *rm = NULL;
+	unsigned long sz;
+	unsigned long bank_sz = 0;
 
 	start = start & PAGE_MASK;
 	end = PAGE_ALIGN(end);
+	sz = end - start;
+	if (!sz)
+		return -EINVAL;
 
-	for (i = 0; i < sysmem.nr_banks; i++)
-		if (start < sysmem.bank[i].end
-		    && end >= sysmem.bank[i].start)
-			break;
+	it = find_bank(start);
+
+	if (it)
+		bank_sz = it->end - it->start;
 
-	if (i == sysmem.nr_banks) {
-		if (must_exist)
-			printk (KERN_WARNING "mem_reserve: [0x%0lx, 0x%0lx) "
-				"not in any region!\n", start, end);
-		return 0;
+	if ((!it || end - it->start > bank_sz) && must_exist) {
+		pr_warn("mem_reserve: [0x%0lx, 0x%0lx) not in any region!\n",
+			start, end);
+		return -EINVAL;
 	}
 
-	if (start > sysmem.bank[i].start) {
-		if (end < sysmem.bank[i].end) {
-			/* split entry */
-			if (sysmem.nr_banks >= SYSMEM_BANKS_MAX)
-				panic("meminfo overflow\n");
-			sysmem.bank[sysmem.nr_banks].start = end;
-			sysmem.bank[sysmem.nr_banks].end = sysmem.bank[i].end;
-			sysmem.nr_banks++;
+	if (it && start - it->start < bank_sz) {
+		if (start == it->start) {
+			if (end - it->start < bank_sz) {
+				it->start = end;
+				return 0;
+			} else {
+				rm = it;
+			}
+		} else {
+			it->end = start;
+			if (end - it->start < bank_sz)
+				return add_sysmem_bank(end,
+						       it->start + bank_sz);
+			++it;
 		}
-		sysmem.bank[i].end = start;
+	}
 
-	} else if (end < sysmem.bank[i].end) {
-		sysmem.bank[i].start = end;
+	if (!it)
+		it = sysmem.bank;
 
-	} else {
-		/* remove entry */
-		sysmem.nr_banks--;
-		sysmem.bank[i].start = sysmem.bank[sysmem.nr_banks].start;
-		sysmem.bank[i].end   = sysmem.bank[sysmem.nr_banks].end;
+	for (; it < sysmem.bank + sysmem.nr_banks; ++it) {
+		if (it->end - start <= sz) {
+			if (!rm)
+				rm = it;
+		} else {
+			if (it->start - start < sz)
+				it->start = end;
+			break;
+		}
 	}
-	return -1;
+
+	if (rm)
+		move_banks(rm, it);
+
+	return 0;
 }
 
 
@@ -99,6 +239,7 @@ void __init bootmem_init(void)
 	unsigned long bootmap_start, bootmap_size;
 	int i;
 
+	sysmem_dump();
 	max_low_pfn = max_pfn = 0;
 	min_low_pfn = ~0;
 
@@ -156,19 +297,13 @@ void __init bootmem_init(void)
 
 void __init zones_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES];
-	int i;
-
 	/* All pages are DMA-able, so we put them all in the DMA zone. */
-
-	zones_size[ZONE_DMA] = max_low_pfn - ARCH_PFN_OFFSET;
-	for (i = 1; i < MAX_NR_ZONES; i++)
-		zones_size[i] = 0;
-
+	unsigned long zones_size[MAX_NR_ZONES] = {
+		[ZONE_DMA] = max_low_pfn - ARCH_PFN_OFFSET,
 #ifdef CONFIG_HIGHMEM
-	zones_size[ZONE_HIGHMEM] = max_pfn - max_low_pfn;
+		[ZONE_HIGHMEM] = max_pfn - max_low_pfn,
 #endif
-
+	};
 	free_area_init_node(0, zones_size, ARCH_PFN_OFFSET, NULL);
 }
 
@@ -178,16 +313,38 @@ void __init zones_init(void)
 
 void __init mem_init(void)
 {
-	max_mapnr = max_low_pfn - ARCH_PFN_OFFSET;
-	high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT);
-
 #ifdef CONFIG_HIGHMEM
-#error HIGHGMEM not implemented in init.c
+	unsigned long tmp;
+
+	reset_all_zones_managed_pages();
+	for (tmp = max_low_pfn; tmp < max_pfn; tmp++)
+		free_highmem_page(pfn_to_page(tmp));
 #endif
 
+	max_mapnr = max_pfn - ARCH_PFN_OFFSET;
+	high_memory = (void *)__va(max_low_pfn << PAGE_SHIFT);
+
 	free_all_bootmem();
 
 	mem_init_print_info(NULL);
+	pr_info("virtual kernel memory layout:\n"
+#ifdef CONFIG_HIGHMEM
+		"    pkmap   : 0x%08lx - 0x%08lx  (%5lu kB)\n"
+		"    fixmap  : 0x%08lx - 0x%08lx  (%5lu kB)\n"
+#endif
+		"    vmalloc : 0x%08x - 0x%08x  (%5u MB)\n"
+		"    lowmem  : 0x%08x - 0x%08lx  (%5lu MB)\n",
+#ifdef CONFIG_HIGHMEM
+		PKMAP_BASE, PKMAP_BASE + LAST_PKMAP * PAGE_SIZE,
+		(LAST_PKMAP*PAGE_SIZE) >> 10,
+		FIXADDR_START, FIXADDR_TOP,
+		(FIXADDR_TOP - FIXADDR_START) >> 10,
+#endif
+		VMALLOC_START, VMALLOC_END,
+		(VMALLOC_END - VMALLOC_START) >> 20,
+		PAGE_OFFSET, PAGE_OFFSET +
+		(max_low_pfn - min_low_pfn) * PAGE_SIZE,
+		((max_low_pfn - min_low_pfn) * PAGE_SIZE) >> 20);
 }
 
 #ifdef CONFIG_BLK_DEV_INITRD
@@ -204,3 +361,53 @@ void free_initmem(void)
 {
 	free_initmem_default(-1);
 }
+
+static void __init parse_memmap_one(char *p)
+{
+	char *oldp;
+	unsigned long start_at, mem_size;
+
+	if (!p)
+		return;
+
+	oldp = p;
+	mem_size = memparse(p, &p);
+	if (p == oldp)
+		return;
+
+	switch (*p) {
+	case '@':
+		start_at = memparse(p + 1, &p);
+		add_sysmem_bank(start_at, start_at + mem_size);
+		break;
+
+	case '$':
+		start_at = memparse(p + 1, &p);
+		mem_reserve(start_at, start_at + mem_size, 0);
+		break;
+
+	case 0:
+		mem_reserve(mem_size, 0, 0);
+		break;
+
+	default:
+		pr_warn("Unrecognized memmap syntax: %s\n", p);
+		break;
+	}
+}
+
+static int __init parse_memmap_opt(char *str)
+{
+	while (str) {
+		char *k = strchr(str, ',');
+
+		if (k)
+			*k++ = 0;
+
+		parse_memmap_one(str);
+		str = k;
+	}
+
+	return 0;
+}
+early_param("memmap", parse_memmap_opt);

arch/xtensa/mm/mmu.c

@@ -3,6 +3,7 @@
  *
  * Extracted from init.c
  */
+#include <linux/bootmem.h>
 #include <linux/percpu.h>
 #include <linux/init.h>
 #include <linux/string.h>
@@ -16,9 +17,44 @@
 #include <asm/initialize_mmu.h>
 #include <asm/io.h>
 
+#if defined(CONFIG_HIGHMEM)
+static void * __init init_pmd(unsigned long vaddr)
+{
+	pgd_t *pgd = pgd_offset_k(vaddr);
+	pmd_t *pmd = pmd_offset(pgd, vaddr);
+
+	if (pmd_none(*pmd)) {
+		unsigned i;
+		pte_t *pte = alloc_bootmem_low_pages(PAGE_SIZE);
+
+		for (i = 0; i < 1024; i++)
+			pte_clear(NULL, 0, pte + i);
+
+		set_pmd(pmd, __pmd(((unsigned long)pte) & PAGE_MASK));
+		BUG_ON(pte != pte_offset_kernel(pmd, 0));
+		pr_debug("%s: vaddr: 0x%08lx, pmd: 0x%p, pte: 0x%p\n",
+			 __func__, vaddr, pmd, pte);
+		return pte;
+	} else {
+		return pte_offset_kernel(pmd, 0);
+	}
+}
+
+static void __init fixedrange_init(void)
+{
+	BUILD_BUG_ON(FIXADDR_SIZE > PMD_SIZE);
+	init_pmd(__fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK);
+}
+#endif
+
 void __init paging_init(void)
 {
 	memset(swapper_pg_dir, 0, PAGE_SIZE);
+#ifdef CONFIG_HIGHMEM
+	fixedrange_init();
+	pkmap_page_table = init_pmd(PKMAP_BASE);
+	kmap_init();
+#endif
 }
 
 /*

arch/xtensa/mm/tlb.c

@@ -149,6 +149,21 @@ void local_flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
 	local_irq_restore(flags);
 }
 
+void local_flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	if (end > start && start >= TASK_SIZE && end <= PAGE_OFFSET &&
+	    end - start < _TLB_ENTRIES << PAGE_SHIFT) {
+		start &= PAGE_MASK;
+		while (start < end) {
+			invalidate_itlb_mapping(start);
+			invalidate_dtlb_mapping(start);
+			start += PAGE_SIZE;
+		}
+	} else {
+		local_flush_tlb_all();
+	}
+}
+
 #ifdef CONFIG_DEBUG_TLB_SANITY
 
 static unsigned get_pte_for_vaddr(unsigned vaddr)

arch/xtensa/platforms/iss/Makefile

@@ -4,6 +4,7 @@
 # "prom monitor" library routines under Linux.
 #
 
-obj-y			= console.o setup.o
+obj-y			= setup.o
+obj-$(CONFIG_TTY)	+= console.o
 obj-$(CONFIG_NET)	+= network.o
 obj-$(CONFIG_BLK_DEV_SIMDISK) += simdisk.o

arch/xtensa/platforms/xt2000/setup.c

@@ -92,18 +92,8 @@ void __init platform_setup(char** cmdline)
 
 /* early initialization */
 
-extern sysmem_info_t __initdata sysmem;
-
-void platform_init(bp_tag_t* first)
+void __init platform_init(bp_tag_t *first)
 {
-	/* Set default memory block if not provided by the bootloader. */
-
-	if (sysmem.nr_banks == 0) {
-		sysmem.nr_banks = 1;
-		sysmem.bank[0].start = PLATFORM_DEFAULT_MEM_START;
-		sysmem.bank[0].end = PLATFORM_DEFAULT_MEM_START
-				     + PLATFORM_DEFAULT_MEM_SIZE;
-	}
 }
 
 /* Heartbeat. Let the LED blink. */

crypto/crypto_user.c

@@ -466,7 +466,7 @@ static int crypto_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	type -= CRYPTO_MSG_BASE;
 	link = &crypto_dispatch[type];
 
-	if (!capable(CAP_NET_ADMIN))
+	if (!netlink_capable(skb, CAP_NET_ADMIN))
 		return -EPERM;
 
 	if ((type == (CRYPTO_MSG_GETALG - CRYPTO_MSG_BASE) &&

drivers/block/floppy.c

@@ -3067,7 +3067,10 @@ static int raw_cmd_copyout(int cmd, void __user *param,
 	int ret;
 
 	while (ptr) {
-		ret = copy_to_user(param, ptr, sizeof(*ptr));
+		struct floppy_raw_cmd cmd = *ptr;
+		cmd.next = NULL;
+		cmd.kernel_data = NULL;
+		ret = copy_to_user(param, &cmd, sizeof(cmd));
 		if (ret)
 			return -EFAULT;
 		param += sizeof(struct floppy_raw_cmd);
@@ -3121,10 +3124,11 @@ loop:
 		return -ENOMEM;
 	*rcmd = ptr;
 	ret = copy_from_user(ptr, param, sizeof(*ptr));
-	if (ret)
-		return -EFAULT;
 	ptr->next = NULL;
 	ptr->buffer_length = 0;
+	ptr->kernel_data = NULL;
+	if (ret)
+		return -EFAULT;
 	param += sizeof(struct floppy_raw_cmd);
 	if (ptr->cmd_count > 33)
 			/* the command may now also take up the space
@@ -3140,7 +3144,6 @@ loop:
 	for (i = 0; i < 16; i++)
 		ptr->reply[i] = 0;
 	ptr->resultcode = 0;
-	ptr->kernel_data = NULL;
 
 	if (ptr->flags & (FD_RAW_READ | FD_RAW_WRITE)) {
 		if (ptr->length <= 0)

drivers/bluetooth/ath3k.c

@@ -82,6 +82,7 @@ static const struct usb_device_id ath3k_table[] = {
 	{ USB_DEVICE(0x04CA, 0x3004) },
 	{ USB_DEVICE(0x04CA, 0x3005) },
 	{ USB_DEVICE(0x04CA, 0x3006) },
+	{ USB_DEVICE(0x04CA, 0x3007) },
 	{ USB_DEVICE(0x04CA, 0x3008) },
 	{ USB_DEVICE(0x04CA, 0x300b) },
 	{ USB_DEVICE(0x0930, 0x0219) },
@@ -131,6 +132,7 @@ static const struct usb_device_id ath3k_blist_tbl[] = {
 	{ USB_DEVICE(0x04ca, 0x3004), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3005), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3006), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x04ca, 0x3007), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3008), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x300b), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x0219), .driver_info = BTUSB_ATH3012 },

drivers/bluetooth/btusb.c

@@ -152,6 +152,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x04ca, 0x3004), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3005), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3006), .driver_info = BTUSB_ATH3012 },
+	{ USB_DEVICE(0x04ca, 0x3007), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x3008), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x04ca, 0x300b), .driver_info = BTUSB_ATH3012 },
 	{ USB_DEVICE(0x0930, 0x0219), .driver_info = BTUSB_ATH3012 },
@@ -1485,10 +1486,8 @@ static int btusb_probe(struct usb_interface *intf,
 	if (id->driver_info & BTUSB_BCM92035)
 		hdev->setup = btusb_setup_bcm92035;
 
-	if (id->driver_info & BTUSB_INTEL) {
-		usb_enable_autosuspend(data->udev);
+	if (id->driver_info & BTUSB_INTEL)
 		hdev->setup = btusb_setup_intel;
-	}
 
 	/* Interface numbers are hardcoded in the specification */
 	data->isoc = usb_ifnum_to_if(data->udev, 1);

drivers/char/agp/frontend.c

@@ -730,6 +730,7 @@ static int agpioc_info_wrap(struct agp_file_private *priv, void __user *arg)
 
 	agp_copy_info(agp_bridge, &kerninfo);
 
+	memset(&userinfo, 0, sizeof(userinfo));
 	userinfo.version.major = kerninfo.version.major;
 	userinfo.version.minor = kerninfo.version.minor;
 	userinfo.bridge_id = kerninfo.device->vendor |

drivers/clk/versatile/clk-vexpress-osc.c

@@ -100,6 +100,8 @@ void __init vexpress_osc_of_setup(struct device_node *node)
 	struct clk *clk;
 	u32 range[2];
 
+	vexpress_sysreg_of_early_init();
+
 	osc = kzalloc(sizeof(*osc), GFP_KERNEL);
 	if (!osc)
 		return;

drivers/connector/cn_proc.c

@@ -369,7 +369,7 @@ static void cn_proc_mcast_ctl(struct cn_msg *msg,
 		return;
 
 	/* Can only change if privileged. */
-	if (!capable(CAP_NET_ADMIN)) {
+	if (!__netlink_ns_capable(nsp, &init_user_ns, CAP_NET_ADMIN)) {
 		err = EPERM;
 		goto out;
 	}

drivers/gpu/drm/i915/i915_drv.h

@@ -1954,6 +1954,9 @@ struct drm_i915_cmd_table {
 #define IS_ULT(dev)		(IS_HSW_ULT(dev) || IS_BDW_ULT(dev))
 #define IS_HSW_GT3(dev)		(IS_HASWELL(dev) && \
 				 ((dev)->pdev->device & 0x00F0) == 0x0020)
+/* ULX machines are also considered ULT. */
+#define IS_HSW_ULX(dev)		((dev)->pdev->device == 0x0A0E || \
+				 (dev)->pdev->device == 0x0A1E)
 #define IS_PRELIMINARY_HW(intel_info) ((intel_info)->is_preliminary)
 
 /*

drivers/gpu/drm/i915/i915_gem_gtt.c

@@ -34,25 +34,35 @@ static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv);
 
 bool intel_enable_ppgtt(struct drm_device *dev, bool full)
 {
-	if (i915.enable_ppgtt == 0 || !HAS_ALIASING_PPGTT(dev))
+	if (i915.enable_ppgtt == 0)
 		return false;
 
 	if (i915.enable_ppgtt == 1 && full)
 		return false;
 
+	return true;
+}
+
+static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
+{
+	if (enable_ppgtt == 0 || !HAS_ALIASING_PPGTT(dev))
+		return 0;
+
+	if (enable_ppgtt == 1)
+		return 1;
+
+	if (enable_ppgtt == 2 && HAS_PPGTT(dev))
+		return 2;
+
 #ifdef CONFIG_INTEL_IOMMU
 	/* Disable ppgtt on SNB if VT-d is on. */
 	if (INTEL_INFO(dev)->gen == 6 && intel_iommu_gfx_mapped) {
 		DRM_INFO("Disabling PPGTT because VT-d is on\n");
-		return false;
+		return 0;
 	}
 #endif
 
-	/* Full ppgtt disabled by default for now due to issues. */
-	if (full)
-		return HAS_PPGTT(dev) && (i915.enable_ppgtt == 2);
-	else
-		return HAS_ALIASING_PPGTT(dev);
+	return HAS_ALIASING_PPGTT(dev) ? 1 : 0;
 }
 
 #define GEN6_PPGTT_PD_ENTRIES 512
@@ -2031,6 +2041,14 @@ int i915_gem_gtt_init(struct drm_device *dev)
 		 gtt->base.total >> 20);
 	DRM_DEBUG_DRIVER("GMADR size = %ldM\n", gtt->mappable_end >> 20);
 	DRM_DEBUG_DRIVER("GTT stolen size = %zdM\n", gtt->stolen_size >> 20);
+	/*
+	 * i915.enable_ppgtt is read-only, so do an early pass to validate the
+	 * user's requested state against the hardware/driver capabilities.  We
+	 * do this now so that we can print out any log messages once rather
+	 * than every time we check intel_enable_ppgtt().
+	 */
+	i915.enable_ppgtt = sanitize_enable_ppgtt(dev, i915.enable_ppgtt);
+	DRM_DEBUG_DRIVER("ppgtt mode: %i\n", i915.enable_ppgtt);
 
 	return 0;
 }

drivers/gpu/drm/i915/intel_display.c

@@ -11395,15 +11395,6 @@ void intel_modeset_init(struct drm_device *dev)
 	}
 }
 
-static void
-intel_connector_break_all_links(struct intel_connector *connector)
-{
-	connector->base.dpms = DRM_MODE_DPMS_OFF;
-	connector->base.encoder = NULL;
-	connector->encoder->connectors_active = false;
-	connector->encoder->base.crtc = NULL;
-}
-
 static void intel_enable_pipe_a(struct drm_device *dev)
 {
 	struct intel_connector *connector;
@@ -11485,8 +11476,17 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 			if (connector->encoder->base.crtc != &crtc->base)
 				continue;
 
-			intel_connector_break_all_links(connector);
+			connector->base.dpms = DRM_MODE_DPMS_OFF;
+			connector->base.encoder = NULL;
 		}
+		/* multiple connectors may have the same encoder:
+		 *  handle them and break crtc link separately */
+		list_for_each_entry(connector, &dev->mode_config.connector_list,
+				    base.head)
+			if (connector->encoder->base.crtc == &crtc->base) {
+				connector->encoder->base.crtc = NULL;
+				connector->encoder->connectors_active = false;
+			}
 
 		WARN_ON(crtc->active);
 		crtc->base.enabled = false;
@@ -11568,6 +11568,8 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 				      drm_get_encoder_name(&encoder->base));
 			encoder->disable(encoder);
 		}
+		encoder->base.crtc = NULL;
+		encoder->connectors_active = false;
 
 		/* Inconsistent output/port/pipe state happens presumably due to
 		 * a bug in one of the get_hw_state functions. Or someplace else
@@ -11578,8 +11580,8 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 				    base.head) {
 			if (connector->encoder != encoder)
 				continue;
-
-			intel_connector_break_all_links(connector);
+			connector->base.dpms = DRM_MODE_DPMS_OFF;
+			connector->base.encoder = NULL;
 		}
 	}
 	/* Enabled encoders without active connectors will be fixed in

drivers/gpu/drm/i915/intel_dp.c

@@ -105,7 +105,8 @@ intel_dp_max_link_bw(struct intel_dp *intel_dp)
 	case DP_LINK_BW_2_7:
 		break;
 	case DP_LINK_BW_5_4: /* 1.2 capable displays may advertise higher bw */
-		if ((IS_HASWELL(dev) || INTEL_INFO(dev)->gen >= 8) &&
+		if (((IS_HASWELL(dev) && !IS_HSW_ULX(dev)) ||
+		     INTEL_INFO(dev)->gen >= 8) &&
 		    intel_dp->dpcd[DP_DPCD_REV] >= 0x12)
 			max_link_bw = DP_LINK_BW_5_4;
 		else

drivers/gpu/drm/nouveau/core/engine/graph/ctxgm107.c

@@ -863,7 +863,7 @@ gm107_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 {
 	mmio_data(0x003000, 0x0100, NV_MEM_ACCESS_RW | NV_MEM_ACCESS_SYS);
 	mmio_data(0x008000, 0x0100, NV_MEM_ACCESS_RW | NV_MEM_ACCESS_SYS);
-	mmio_data(0x060000, 0x1000, NV_MEM_ACCESS_RW);
+	mmio_data(0x200000, 0x1000, NV_MEM_ACCESS_RW);
 
 	mmio_list(0x40800c, 0x00000000,  8, 1);
 	mmio_list(0x408010, 0x80000000,  0, 0);
@@ -877,6 +877,8 @@ gm107_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 	mmio_list(0x418e24, 0x00000000,  8, 0);
 	mmio_list(0x418e28, 0x80000030,  0, 0);
 
+	mmio_list(0x4064c8, 0x018002c0,  0, 0);
+
 	mmio_list(0x418810, 0x80000000, 12, 2);
 	mmio_list(0x419848, 0x10000000, 12, 2);
 	mmio_list(0x419c2c, 0x10000000, 12, 2);

drivers/gpu/drm/nouveau/core/subdev/bios/base.c

@@ -168,7 +168,8 @@ nouveau_bios_shadow_prom(struct nouveau_bios *bios)
 	 */
 	i = 16;
 	do {
-		if ((nv_rd32(bios, 0x300000) & 0xffff) == 0xaa55)
+		u32 data = le32_to_cpu(nv_rd32(bios, 0x300000)) & 0xffff;
+		if (data == 0xaa55)
 			break;
 	} while (i--);
 
@@ -176,14 +177,15 @@ nouveau_bios_shadow_prom(struct nouveau_bios *bios)
 		goto out;
 
 	/* read entire bios image to system memory */
-	bios->size = ((nv_rd32(bios, 0x300000) >> 16) & 0xff) * 512;
+	bios->size = (le32_to_cpu(nv_rd32(bios, 0x300000)) >> 16) & 0xff;
+	bios->size = bios->size * 512;
 	if (!bios->size)
 		goto out;
 
 	bios->data = kmalloc(bios->size, GFP_KERNEL);
 	if (bios->data) {
-		for (i = 0; i < bios->size; i+=4)
-			nv_wo32(bios, i, nv_rd32(bios, 0x300000 + i));
+		for (i = 0; i < bios->size; i += 4)
+			((u32 *)bios->data)[i/4] = nv_rd32(bios, 0x300000 + i);
 	}
 
 	/* check the PCI record header */

drivers/gpu/drm/nouveau/nouveau_acpi.c

@@ -389,9 +389,6 @@ bool nouveau_acpi_rom_supported(struct pci_dev *pdev)
 	acpi_status status;
 	acpi_handle dhandle, rom_handle;
 
-	if (!nouveau_dsm_priv.dsm_detected && !nouveau_dsm_priv.optimus_detected)
-		return false;
-
 	dhandle = ACPI_HANDLE(&pdev->dev);
 	if (!dhandle)
 		return false;

drivers/gpu/drm/nouveau/nouveau_display.c

@@ -764,9 +764,9 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	}
 
 	ret = nouveau_page_flip_emit(chan, old_bo, new_bo, s, &fence);
-	mutex_unlock(&chan->cli->mutex);
 	if (ret)
 		goto fail_unreserve;
+	mutex_unlock(&chan->cli->mutex);
 
 	/* Update the crtc struct and cleanup */
 	crtc->primary->fb = fb;

drivers/gpu/drm/radeon/atombios_crtc.c

@@ -1177,27 +1177,43 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 
 		/* Set NUM_BANKS. */
 		if (rdev->family >= CHIP_TAHITI) {
-			unsigned tileb, index, num_banks, tile_split_bytes;
+			unsigned index, num_banks;
 
-			/* Calculate the macrotile mode index. */
-			tile_split_bytes = 64 << tile_split;
-			tileb = 8 * 8 * target_fb->bits_per_pixel / 8;
-			tileb = min(tile_split_bytes, tileb);
+			if (rdev->family >= CHIP_BONAIRE) {
+				unsigned tileb, tile_split_bytes;
 
-			for (index = 0; tileb > 64; index++) {
-				tileb >>= 1;
-			}
+				/* Calculate the macrotile mode index. */
+				tile_split_bytes = 64 << tile_split;
+				tileb = 8 * 8 * target_fb->bits_per_pixel / 8;
+				tileb = min(tile_split_bytes, tileb);
 
-			if (index >= 16) {
-				DRM_ERROR("Wrong screen bpp (%u) or tile split (%u)\n",
-					  target_fb->bits_per_pixel, tile_split);
-				return -EINVAL;
-			}
+				for (index = 0; tileb > 64; index++)
+					tileb >>= 1;
+
+				if (index >= 16) {
+					DRM_ERROR("Wrong screen bpp (%u) or tile split (%u)\n",
+						  target_fb->bits_per_pixel, tile_split);
+					return -EINVAL;
+				}
 
-			if (rdev->family >= CHIP_BONAIRE)
 				num_banks = (rdev->config.cik.macrotile_mode_array[index] >> 6) & 0x3;
-			else
+			} else {
+				switch (target_fb->bits_per_pixel) {
+				case 8:
+					index = 10;
+					break;
+				case 16:
+					index = SI_TILE_MODE_COLOR_2D_SCANOUT_16BPP;
+					break;
+				default:
+				case 32:
+					index = SI_TILE_MODE_COLOR_2D_SCANOUT_32BPP;
+					break;
+				}
+
 				num_banks = (rdev->config.si.tile_mode_array[index] >> 20) & 0x3;
+			}
+
 			fb_format |= EVERGREEN_GRPH_NUM_BANKS(num_banks);
 		} else {
 			/* NI and older. */
@@ -1720,8 +1736,9 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc)
 		}
 		/* otherwise, pick one of the plls */
 		if ((rdev->family == CHIP_KAVERI) ||
-		    (rdev->family == CHIP_KABINI)) {
-			/* KB/KV has PPLL1 and PPLL2 */
+		    (rdev->family == CHIP_KABINI) ||
+		    (rdev->family == CHIP_MULLINS)) {
+			/* KB/KV/ML has PPLL1 and PPLL2 */
 			pll_in_use = radeon_get_pll_use_mask(crtc);
 			if (!(pll_in_use & (1 << ATOM_PPLL2)))
 				return ATOM_PPLL2;
@@ -1885,6 +1902,9 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc,
 	    (ATOM_DEVICE_TV_SUPPORT | ATOM_DEVICE_CV_SUPPORT))
 		is_tvcv = true;
 
+	if (!radeon_crtc->adjusted_clock)
+		return -EINVAL;
+
 	atombios_crtc_set_pll(crtc, adjusted_mode);
 
 	if (ASIC_IS_DCE4(rdev))

drivers/gpu/drm/radeon/atombios_dp.c

@@ -366,11 +366,11 @@ static void radeon_dp_probe_oui(struct radeon_connector *radeon_connector)
 	if (!(dig_connector->dpcd[DP_DOWN_STREAM_PORT_COUNT] & DP_OUI_SUPPORT))
 		return;
 
-	if (drm_dp_dpcd_read(&radeon_connector->ddc_bus->aux, DP_SINK_OUI, buf, 3))
+	if (drm_dp_dpcd_read(&radeon_connector->ddc_bus->aux, DP_SINK_OUI, buf, 3) == 3)
 		DRM_DEBUG_KMS("Sink OUI: %02hx%02hx%02hx\n",
 			      buf[0], buf[1], buf[2]);
 
-	if (drm_dp_dpcd_read(&radeon_connector->ddc_bus->aux, DP_BRANCH_OUI, buf, 3))
+	if (drm_dp_dpcd_read(&radeon_connector->ddc_bus->aux, DP_BRANCH_OUI, buf, 3) == 3)
 		DRM_DEBUG_KMS("Branch OUI: %02hx%02hx%02hx\n",
 			      buf[0], buf[1], buf[2]);
 }
@@ -419,21 +419,23 @@ int radeon_dp_get_panel_mode(struct drm_encoder *encoder,
 
 	if (dp_bridge != ENCODER_OBJECT_ID_NONE) {
 		/* DP bridge chips */
-		drm_dp_dpcd_readb(&radeon_connector->ddc_bus->aux,
-				  DP_EDP_CONFIGURATION_CAP, &tmp);
-		if (tmp & 1)
-			panel_mode = DP_PANEL_MODE_INTERNAL_DP2_MODE;
-		else if ((dp_bridge == ENCODER_OBJECT_ID_NUTMEG) ||
-			 (dp_bridge == ENCODER_OBJECT_ID_TRAVIS))
-			panel_mode = DP_PANEL_MODE_INTERNAL_DP1_MODE;
-		else
-			panel_mode = DP_PANEL_MODE_EXTERNAL_DP_MODE;
+		if (drm_dp_dpcd_readb(&radeon_connector->ddc_bus->aux,
+				      DP_EDP_CONFIGURATION_CAP, &tmp) == 1) {
+			if (tmp & 1)
+				panel_mode = DP_PANEL_MODE_INTERNAL_DP2_MODE;
+			else if ((dp_bridge == ENCODER_OBJECT_ID_NUTMEG) ||
+				 (dp_bridge == ENCODER_OBJECT_ID_TRAVIS))
+				panel_mode = DP_PANEL_MODE_INTERNAL_DP1_MODE;
+			else
+				panel_mode = DP_PANEL_MODE_EXTERNAL_DP_MODE;
+		}
 	} else if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
 		/* eDP */
-		drm_dp_dpcd_readb(&radeon_connector->ddc_bus->aux,
-				  DP_EDP_CONFIGURATION_CAP, &tmp);
-		if (tmp & 1)
-			panel_mode = DP_PANEL_MODE_INTERNAL_DP2_MODE;
+		if (drm_dp_dpcd_readb(&radeon_connector->ddc_bus->aux,
+				      DP_EDP_CONFIGURATION_CAP, &tmp) == 1) {
+			if (tmp & 1)
+				panel_mode = DP_PANEL_MODE_INTERNAL_DP2_MODE;
+		}
 	}
 
 	return panel_mode;
@@ -809,11 +811,15 @@ void radeon_dp_link_train(struct drm_encoder *encoder,
 	else
 		dp_info.enc_id |= ATOM_DP_CONFIG_LINK_A;
 
-	drm_dp_dpcd_readb(&radeon_connector->ddc_bus->aux, DP_MAX_LANE_COUNT, &tmp);
-	if (ASIC_IS_DCE5(rdev) && (tmp & DP_TPS3_SUPPORTED))
-		dp_info.tp3_supported = true;
-	else
+	if (drm_dp_dpcd_readb(&radeon_connector->ddc_bus->aux, DP_MAX_LANE_COUNT, &tmp)
+	    == 1) {
+		if (ASIC_IS_DCE5(rdev) && (tmp & DP_TPS3_SUPPORTED))
+			dp_info.tp3_supported = true;
+		else
+			dp_info.tp3_supported = false;
+	} else {
 		dp_info.tp3_supported = false;
+	}
 
 	memcpy(dp_info.dpcd, dig_connector->dpcd, DP_RECEIVER_CAP_SIZE);
 	dp_info.rdev = rdev;

drivers/gpu/drm/radeon/cik.c

@@ -63,6 +63,12 @@ MODULE_FIRMWARE("radeon/KABINI_ce.bin");
 MODULE_FIRMWARE("radeon/KABINI_mec.bin");
 MODULE_FIRMWARE("radeon/KABINI_rlc.bin");
 MODULE_FIRMWARE("radeon/KABINI_sdma.bin");
+MODULE_FIRMWARE("radeon/MULLINS_pfp.bin");
+MODULE_FIRMWARE("radeon/MULLINS_me.bin");
+MODULE_FIRMWARE("radeon/MULLINS_ce.bin");
+MODULE_FIRMWARE("radeon/MULLINS_mec.bin");
+MODULE_FIRMWARE("radeon/MULLINS_rlc.bin");
+MODULE_FIRMWARE("radeon/MULLINS_sdma.bin");
 
 extern int r600_ih_ring_alloc(struct radeon_device *rdev);
 extern void r600_ih_ring_fini(struct radeon_device *rdev);
@@ -1473,6 +1479,43 @@ static const u32 hawaii_mgcg_cgcg_init[] =
 	0xd80c, 0xff000ff0, 0x00000100
 };
 
+static const u32 godavari_golden_registers[] =
+{
+	0x55e4, 0xff607fff, 0xfc000100,
+	0x6ed8, 0x00010101, 0x00010000,
+	0x9830, 0xffffffff, 0x00000000,
+	0x98302, 0xf00fffff, 0x00000400,
+	0x6130, 0xffffffff, 0x00010000,
+	0x5bb0, 0x000000f0, 0x00000070,
+	0x5bc0, 0xf0311fff, 0x80300000,
+	0x98f8, 0x73773777, 0x12010001,
+	0x98fc, 0xffffffff, 0x00000010,
+	0x8030, 0x00001f0f, 0x0000100a,
+	0x2f48, 0x73773777, 0x12010001,
+	0x2408, 0x000fffff, 0x000c007f,
+	0x8a14, 0xf000003f, 0x00000007,
+	0x8b24, 0xffffffff, 0x00ff0fff,
+	0x30a04, 0x0000ff0f, 0x00000000,
+	0x28a4c, 0x07ffffff, 0x06000000,
+	0x4d8, 0x00000fff, 0x00000100,
+	0xd014, 0x00010000, 0x00810001,
+	0xd814, 0x00010000, 0x00810001,
+	0x3e78, 0x00000001, 0x00000002,
+	0xc768, 0x00000008, 0x00000008,
+	0xc770, 0x00000f00, 0x00000800,
+	0xc774, 0x00000f00, 0x00000800,
+	0xc798, 0x00ffffff, 0x00ff7fbf,
+	0xc79c, 0x00ffffff, 0x00ff7faf,
+	0x8c00, 0x000000ff, 0x00000001,
+	0x214f8, 0x01ff01ff, 0x00000002,
+	0x21498, 0x007ff800, 0x00200000,
+	0x2015c, 0xffffffff, 0x00000f40,
+	0x88c4, 0x001f3ae3, 0x00000082,
+	0x88d4, 0x0000001f, 0x00000010,
+	0x30934, 0xffffffff, 0x00000000
+};
+
+
 static void cik_init_golden_registers(struct radeon_device *rdev)
 {
 	switch (rdev->family) {
@@ -1504,6 +1547,20 @@ static void cik_init_golden_registers(struct radeon_device *rdev)
 						 kalindi_golden_spm_registers,
 						 (const u32)ARRAY_SIZE(kalindi_golden_spm_registers));
 		break;
+	case CHIP_MULLINS:
+		radeon_program_register_sequence(rdev,
+						 kalindi_mgcg_cgcg_init,
+						 (const u32)ARRAY_SIZE(kalindi_mgcg_cgcg_init));
+		radeon_program_register_sequence(rdev,
+						 godavari_golden_registers,
+						 (const u32)ARRAY_SIZE(godavari_golden_registers));
+		radeon_program_register_sequence(rdev,
+						 kalindi_golden_common_registers,
+						 (const u32)ARRAY_SIZE(kalindi_golden_common_registers));
+		radeon_program_register_sequence(rdev,
+						 kalindi_golden_spm_registers,
+						 (const u32)ARRAY_SIZE(kalindi_golden_spm_registers));
+		break;
 	case CHIP_KAVERI:
 		radeon_program_register_sequence(rdev,
 						 spectre_mgcg_cgcg_init,
@@ -1834,6 +1891,15 @@ static int cik_init_microcode(struct radeon_device *rdev)
 		rlc_req_size = KB_RLC_UCODE_SIZE * 4;
 		sdma_req_size = CIK_SDMA_UCODE_SIZE * 4;
 		break;
+	case CHIP_MULLINS:
+		chip_name = "MULLINS";
+		pfp_req_size = CIK_PFP_UCODE_SIZE * 4;
+		me_req_size = CIK_ME_UCODE_SIZE * 4;
+		ce_req_size = CIK_CE_UCODE_SIZE * 4;
+		mec_req_size = CIK_MEC_UCODE_SIZE * 4;
+		rlc_req_size = ML_RLC_UCODE_SIZE * 4;
+		sdma_req_size = CIK_SDMA_UCODE_SIZE * 4;
+		break;
 	default: BUG();
 	}
 
@@ -3272,6 +3338,7 @@ static void cik_gpu_init(struct radeon_device *rdev)
 		gb_addr_config = BONAIRE_GB_ADDR_CONFIG_GOLDEN;
 		break;
 	case CHIP_KABINI:
+	case CHIP_MULLINS:
 	default:
 		rdev->config.cik.max_shader_engines = 1;
 		rdev->config.cik.max_tile_pipes = 2;
@@ -3702,6 +3769,7 @@ int cik_copy_cpdma(struct radeon_device *rdev,
 	r = radeon_fence_emit(rdev, fence, ring->idx);
 	if (r) {
 		radeon_ring_unlock_undo(rdev, ring);
+		radeon_semaphore_free(rdev, &sem, NULL);
 		return r;
 	}
 
@@ -5800,6 +5868,9 @@ static int cik_rlc_resume(struct radeon_device *rdev)
 	case CHIP_KABINI:
 		size = KB_RLC_UCODE_SIZE;
 		break;
+	case CHIP_MULLINS:
+		size = ML_RLC_UCODE_SIZE;
+		break;
 	}
 
 	cik_rlc_stop(rdev);
@@ -6548,6 +6619,7 @@ void cik_get_csb_buffer(struct radeon_device *rdev, volatile u32 *buffer)
 		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
 	case CHIP_KABINI:
+	case CHIP_MULLINS:
 		buffer[count++] = cpu_to_le32(0x00000000); /* XXX */
 		buffer[count++] = cpu_to_le32(0x00000000);
 		break;
@@ -6693,6 +6765,19 @@ static void cik_disable_interrupt_state(struct radeon_device *rdev)
 		WREG32(LB_INTERRUPT_MASK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0);
 		WREG32(LB_INTERRUPT_MASK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0);
 	}
+	/* pflip */
+	if (rdev->num_crtc >= 2) {
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
+	}
+	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);
+	}
 
 	/* dac hotplug */
 	WREG32(DAC_AUTODETECT_INT_CONTROL, 0);
@@ -7049,6 +7134,25 @@ int cik_irq_set(struct radeon_device *rdev)
 		WREG32(LB_INTERRUPT_MASK + EVERGREEN_CRTC5_REGISTER_OFFSET, crtc6);
 	}
 
+	if (rdev->num_crtc >= 2) {
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+	}
+	if (rdev->num_crtc >= 4) {
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+	}
+	if (rdev->num_crtc >= 6) {
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+	}
+
 	WREG32(DC_HPD1_INT_CONTROL, hpd1);
 	WREG32(DC_HPD2_INT_CONTROL, hpd2);
 	WREG32(DC_HPD3_INT_CONTROL, hpd3);
@@ -7085,6 +7189,29 @@ static inline void cik_irq_ack(struct radeon_device *rdev)
 	rdev->irq.stat_regs.cik.disp_int_cont5 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE5);
 	rdev->irq.stat_regs.cik.disp_int_cont6 = RREG32(DISP_INTERRUPT_STATUS_CONTINUE6);
 
+	rdev->irq.stat_regs.cik.d1grph_int = RREG32(GRPH_INT_STATUS +
+		EVERGREEN_CRTC0_REGISTER_OFFSET);
+	rdev->irq.stat_regs.cik.d2grph_int = RREG32(GRPH_INT_STATUS +
+		EVERGREEN_CRTC1_REGISTER_OFFSET);
+	if (rdev->num_crtc >= 4) {
+		rdev->irq.stat_regs.cik.d3grph_int = RREG32(GRPH_INT_STATUS +
+			EVERGREEN_CRTC2_REGISTER_OFFSET);
+		rdev->irq.stat_regs.cik.d4grph_int = RREG32(GRPH_INT_STATUS +
+			EVERGREEN_CRTC3_REGISTER_OFFSET);
+	}
+	if (rdev->num_crtc >= 6) {
+		rdev->irq.stat_regs.cik.d5grph_int = RREG32(GRPH_INT_STATUS +
+			EVERGREEN_CRTC4_REGISTER_OFFSET);
+		rdev->irq.stat_regs.cik.d6grph_int = RREG32(GRPH_INT_STATUS +
+			EVERGREEN_CRTC5_REGISTER_OFFSET);
+	}
+
+	if (rdev->irq.stat_regs.cik.d1grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_CLEAR);
+	if (rdev->irq.stat_regs.cik.d2grph_int & GRPH_PFLIP_INT_OCCURRED)
+		WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_CLEAR);
 	if (rdev->irq.stat_regs.cik.disp_int & LB_D1_VBLANK_INTERRUPT)
 		WREG32(LB_VBLANK_STATUS + EVERGREEN_CRTC0_REGISTER_OFFSET, VBLANK_ACK);
 	if (rdev->irq.stat_regs.cik.disp_int & LB_D1_VLINE_INTERRUPT)
@@ -7095,6 +7222,12 @@ static inline void cik_irq_ack(struct radeon_device *rdev)
 		WREG32(LB_VLINE_STATUS + EVERGREEN_CRTC1_REGISTER_OFFSET, VLINE_ACK);
 
 	if (rdev->num_crtc >= 4) {
+		if (rdev->irq.stat_regs.cik.d3grph_int & GRPH_PFLIP_INT_OCCURRED)
+			WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET,
+			       GRPH_PFLIP_INT_CLEAR);
+		if (rdev->irq.stat_regs.cik.d4grph_int & GRPH_PFLIP_INT_OCCURRED)
+			WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC3_REGISTER_OFFSET,
+			       GRPH_PFLIP_INT_CLEAR);
 		if (rdev->irq.stat_regs.cik.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT)
 			WREG32(LB_VBLANK_STATUS + EVERGREEN_CRTC2_REGISTER_OFFSET, VBLANK_ACK);
 		if (rdev->irq.stat_regs.cik.disp_int_cont2 & LB_D3_VLINE_INTERRUPT)
@@ -7106,6 +7239,12 @@ static inline void cik_irq_ack(struct radeon_device *rdev)
 	}
 
 	if (rdev->num_crtc >= 6) {
+		if (rdev->irq.stat_regs.cik.d5grph_int & GRPH_PFLIP_INT_OCCURRED)
+			WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET,
+			       GRPH_PFLIP_INT_CLEAR);
+		if (rdev->irq.stat_regs.cik.d6grph_int & GRPH_PFLIP_INT_OCCURRED)
+			WREG32(GRPH_INT_STATUS + EVERGREEN_CRTC5_REGISTER_OFFSET,
+			       GRPH_PFLIP_INT_CLEAR);
 		if (rdev->irq.stat_regs.cik.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT)
 			WREG32(LB_VBLANK_STATUS + EVERGREEN_CRTC4_REGISTER_OFFSET, VBLANK_ACK);
 		if (rdev->irq.stat_regs.cik.disp_int_cont4 & LB_D5_VLINE_INTERRUPT)
@@ -7457,6 +7596,15 @@ restart_ih:
 				break;
 			}
 			break;
+		case 8: /* D1 page flip */
+		case 10: /* D2 page flip */
+		case 12: /* D3 page flip */
+		case 14: /* D4 page flip */
+		case 16: /* D5 page flip */
+		case 18: /* D6 page flip */
+			DRM_DEBUG("IH: D%d flip\n", ((src_id - 8) >> 1) + 1);
+			radeon_crtc_handle_flip(rdev, (src_id - 8) >> 1);
+			break;
 		case 42: /* HPD hotplug */
 			switch (src_data) {
 			case 0:

drivers/gpu/drm/radeon/cik_sdma.c

@@ -562,6 +562,7 @@ int cik_copy_dma(struct radeon_device *rdev,
 	r = radeon_fence_emit(rdev, fence, ring->idx);
 	if (r) {
 		radeon_ring_unlock_undo(rdev, ring);
+		radeon_semaphore_free(rdev, &sem, NULL);
 		return r;
 	}
 

drivers/gpu/drm/radeon/cikd.h

@@ -888,6 +888,15 @@
 #       define DC_HPD6_RX_INTERRUPT                     (1 << 18)
 #define DISP_INTERRUPT_STATUS_CONTINUE6                 0x6780
 
+/* 0x6858, 0x7458, 0x10058, 0x10c58, 0x11858, 0x12458 */
+#define GRPH_INT_STATUS                                 0x6858
+#       define GRPH_PFLIP_INT_OCCURRED                  (1 << 0)
+#       define GRPH_PFLIP_INT_CLEAR                     (1 << 8)
+/* 0x685c, 0x745c, 0x1005c, 0x10c5c, 0x1185c, 0x1245c */
+#define GRPH_INT_CONTROL                                0x685c
+#       define GRPH_PFLIP_INT_MASK                      (1 << 0)
+#       define GRPH_PFLIP_INT_TYPE                      (1 << 8)
+
 #define	DAC_AUTODETECT_INT_CONTROL			0x67c8
 
 #define DC_HPD1_INT_STATUS                              0x601c

drivers/gpu/drm/radeon/evergreen.c

@@ -4371,7 +4371,6 @@ int evergreen_irq_set(struct radeon_device *rdev)
 	u32 crtc1 = 0, crtc2 = 0, crtc3 = 0, crtc4 = 0, crtc5 = 0, crtc6 = 0;
 	u32 hpd1, hpd2, hpd3, hpd4, hpd5, hpd6;
 	u32 grbm_int_cntl = 0;
-	u32 grph1 = 0, grph2 = 0, grph3 = 0, grph4 = 0, grph5 = 0, grph6 = 0;
 	u32 afmt1 = 0, afmt2 = 0, afmt3 = 0, afmt4 = 0, afmt5 = 0, afmt6 = 0;
 	u32 dma_cntl, dma_cntl1 = 0;
 	u32 thermal_int = 0;
@@ -4554,15 +4553,21 @@ int evergreen_irq_set(struct radeon_device *rdev)
 		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_CRTC0_REGISTER_OFFSET,
+	       GRPH_PFLIP_INT_MASK);
+	WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET,
+	       GRPH_PFLIP_INT_MASK);
 	if (rdev->num_crtc >= 4) {
-		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, grph3);
-		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, grph4);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
 	}
 	if (rdev->num_crtc >= 6) {
-		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, grph5);
-		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, grph6);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
+		WREG32(GRPH_INT_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET,
+		       GRPH_PFLIP_INT_MASK);
 	}
 
 	WREG32(DC_HPD1_INT_CONTROL, hpd1);
@@ -4951,6 +4956,15 @@ restart_ih:
 				break;
 			}
 			break;
+		case 8: /* D1 page flip */
+		case 10: /* D2 page flip */
+		case 12: /* D3 page flip */
+		case 14: /* D4 page flip */
+		case 16: /* D5 page flip */
+		case 18: /* D6 page flip */
+			DRM_DEBUG("IH: D%d flip\n", ((src_id - 8) >> 1) + 1);
+			radeon_crtc_handle_flip(rdev, (src_id - 8) >> 1);
+			break;
 		case 42: /* HPD hotplug */
 			switch (src_data) {
 			case 0:

drivers/gpu/drm/radeon/evergreen_dma.c

@@ -151,6 +151,7 @@ int evergreen_copy_dma(struct radeon_device *rdev,
 	r = radeon_fence_emit(rdev, fence, ring->idx);
 	if (r) {
 		radeon_ring_unlock_undo(rdev, ring);
+		radeon_semaphore_free(rdev, &sem, NULL);
 		return r;
 	}
 

drivers/gpu/drm/radeon/kv_dpm.c

@@ -546,6 +546,52 @@ static int kv_set_divider_value(struct radeon_device *rdev,
 	return 0;
 }
 
+static u32 kv_convert_vid2_to_vid7(struct radeon_device *rdev,
+				   struct sumo_vid_mapping_table *vid_mapping_table,
+				   u32 vid_2bit)
+{
+	struct radeon_clock_voltage_dependency_table *vddc_sclk_table =
+		&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+	u32 i;
+
+	if (vddc_sclk_table && vddc_sclk_table->count) {
+		if (vid_2bit < vddc_sclk_table->count)
+			return vddc_sclk_table->entries[vid_2bit].v;
+		else
+			return vddc_sclk_table->entries[vddc_sclk_table->count - 1].v;
+	} else {
+		for (i = 0; i < vid_mapping_table->num_entries; i++) {
+			if (vid_mapping_table->entries[i].vid_2bit == vid_2bit)
+				return vid_mapping_table->entries[i].vid_7bit;
+		}
+		return vid_mapping_table->entries[vid_mapping_table->num_entries - 1].vid_7bit;
+	}
+}
+
+static u32 kv_convert_vid7_to_vid2(struct radeon_device *rdev,
+				   struct sumo_vid_mapping_table *vid_mapping_table,
+				   u32 vid_7bit)
+{
+	struct radeon_clock_voltage_dependency_table *vddc_sclk_table =
+		&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
+	u32 i;
+
+	if (vddc_sclk_table && vddc_sclk_table->count) {
+		for (i = 0; i < vddc_sclk_table->count; i++) {
+			if (vddc_sclk_table->entries[i].v == vid_7bit)
+				return i;
+		}
+		return vddc_sclk_table->count - 1;
+	} else {
+		for (i = 0; i < vid_mapping_table->num_entries; i++) {
+			if (vid_mapping_table->entries[i].vid_7bit == vid_7bit)
+				return vid_mapping_table->entries[i].vid_2bit;
+		}
+
+		return vid_mapping_table->entries[vid_mapping_table->num_entries - 1].vid_2bit;
+	}
+}
+
 static u16 kv_convert_8bit_index_to_voltage(struct radeon_device *rdev,
 					    u16 voltage)
 {
@@ -556,9 +602,9 @@ static u16 kv_convert_2bit_index_to_voltage(struct radeon_device *rdev,
 					    u32 vid_2bit)
 {
 	struct kv_power_info *pi = kv_get_pi(rdev);
-	u32 vid_8bit = sumo_convert_vid2_to_vid7(rdev,
-						 &pi->sys_info.vid_mapping_table,
-						 vid_2bit);
+	u32 vid_8bit = kv_convert_vid2_to_vid7(rdev,
+					       &pi->sys_info.vid_mapping_table,
+					       vid_2bit);
 
 	return kv_convert_8bit_index_to_voltage(rdev, (u16)vid_8bit);
 }
@@ -639,7 +685,7 @@ static int kv_force_lowest_valid(struct radeon_device *rdev)
 
 static int kv_unforce_levels(struct radeon_device *rdev)
 {
-	if (rdev->family == CHIP_KABINI)
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS)
 		return kv_notify_message_to_smu(rdev, PPSMC_MSG_NoForcedLevel);
 	else
 		return kv_set_enabled_levels(rdev);
@@ -1362,13 +1408,20 @@ static int kv_update_uvd_dpm(struct radeon_device *rdev, bool gate)
 	struct radeon_uvd_clock_voltage_dependency_table *table =
 		&rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table;
 	int ret;
+	u32 mask;
 
 	if (!gate) {
-		if (!pi->caps_uvd_dpm || table->count || pi->caps_stable_p_state)
+		if (table->count)
 			pi->uvd_boot_level = table->count - 1;
 		else
 			pi->uvd_boot_level = 0;
 
+		if (!pi->caps_uvd_dpm || pi->caps_stable_p_state) {
+			mask = 1 << pi->uvd_boot_level;
+		} else {
+			mask = 0x1f;
+		}
+
 		ret = kv_copy_bytes_to_smc(rdev,
 					   pi->dpm_table_start +
 					   offsetof(SMU7_Fusion_DpmTable, UvdBootLevel),
@@ -1377,11 +1430,9 @@ static int kv_update_uvd_dpm(struct radeon_device *rdev, bool gate)
 		if (ret)
 			return ret;
 
-		if (!pi->caps_uvd_dpm ||
-		    pi->caps_stable_p_state)
-			kv_send_msg_to_smc_with_parameter(rdev,
-							  PPSMC_MSG_UVDDPM_SetEnabledMask,
-							  (1 << pi->uvd_boot_level));
+		kv_send_msg_to_smc_with_parameter(rdev,
+						  PPSMC_MSG_UVDDPM_SetEnabledMask,
+						  mask);
 	}
 
 	return kv_enable_uvd_dpm(rdev, !gate);
@@ -1617,7 +1668,7 @@ static void kv_dpm_powergate_acp(struct radeon_device *rdev, bool gate)
 	if (pi->acp_power_gated == gate)
 		return;
 
-	if (rdev->family == CHIP_KABINI)
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS)
 		return;
 
 	pi->acp_power_gated = gate;
@@ -1786,7 +1837,7 @@ int kv_dpm_set_power_state(struct radeon_device *rdev)
 		}
 	}
 
-	if (rdev->family == CHIP_KABINI) {
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS) {
 		if (pi->enable_dpm) {
 			kv_set_valid_clock_range(rdev, new_ps);
 			kv_update_dfs_bypass_settings(rdev, new_ps);
@@ -1812,6 +1863,8 @@ int kv_dpm_set_power_state(struct radeon_device *rdev)
 				return ret;
 			}
 			kv_update_sclk_t(rdev);
+			if (rdev->family == CHIP_MULLINS)
+				kv_enable_nb_dpm(rdev);
 		}
 	} else {
 		if (pi->enable_dpm) {
@@ -1862,7 +1915,7 @@ void kv_dpm_reset_asic(struct radeon_device *rdev)
 {
 	struct kv_power_info *pi = kv_get_pi(rdev);
 
-	if (rdev->family == CHIP_KABINI) {
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS) {
 		kv_force_lowest_valid(rdev);
 		kv_init_graphics_levels(rdev);
 		kv_program_bootup_state(rdev);
@@ -1901,14 +1954,41 @@ static void kv_construct_max_power_limits_table(struct radeon_device *rdev,
 static void kv_patch_voltage_values(struct radeon_device *rdev)
 {
 	int i;
-	struct radeon_uvd_clock_voltage_dependency_table *table =
+	struct radeon_uvd_clock_voltage_dependency_table *uvd_table =
 		&rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table;
+	struct radeon_vce_clock_voltage_dependency_table *vce_table =
+		&rdev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+	struct radeon_clock_voltage_dependency_table *samu_table =
+		&rdev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table;
+	struct radeon_clock_voltage_dependency_table *acp_table =
+		&rdev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table;
 
-	if (table->count) {
-		for (i = 0; i < table->count; i++)
-			table->entries[i].v =
+	if (uvd_table->count) {
+		for (i = 0; i < uvd_table->count; i++)
+			uvd_table->entries[i].v =
 				kv_convert_8bit_index_to_voltage(rdev,
-								 table->entries[i].v);
+								 uvd_table->entries[i].v);
+	}
+
+	if (vce_table->count) {
+		for (i = 0; i < vce_table->count; i++)
+			vce_table->entries[i].v =
+				kv_convert_8bit_index_to_voltage(rdev,
+								 vce_table->entries[i].v);
+	}
+
+	if (samu_table->count) {
+		for (i = 0; i < samu_table->count; i++)
+			samu_table->entries[i].v =
+				kv_convert_8bit_index_to_voltage(rdev,
+								 samu_table->entries[i].v);
+	}
+
+	if (acp_table->count) {
+		for (i = 0; i < acp_table->count; i++)
+			acp_table->entries[i].v =
+				kv_convert_8bit_index_to_voltage(rdev,
+								 acp_table->entries[i].v);
 	}
 
 }
@@ -1941,7 +2021,7 @@ static int kv_force_dpm_highest(struct radeon_device *rdev)
 			break;
 	}
 
-	if (rdev->family == CHIP_KABINI)
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS)
 		return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i);
 	else
 		return kv_set_enabled_level(rdev, i);
@@ -1961,7 +2041,7 @@ static int kv_force_dpm_lowest(struct radeon_device *rdev)
 			break;
 	}
 
-	if (rdev->family == CHIP_KABINI)
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS)
 		return kv_send_msg_to_smc_with_parameter(rdev, PPSMC_MSG_DPM_ForceState, i);
 	else
 		return kv_set_enabled_level(rdev, i);
@@ -2118,7 +2198,7 @@ static void kv_apply_state_adjust_rules(struct radeon_device *rdev,
 	else
 		pi->battery_state = false;
 
-	if (rdev->family == CHIP_KABINI) {
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS) {
 		ps->dpm0_pg_nb_ps_lo = 0x1;
 		ps->dpm0_pg_nb_ps_hi = 0x0;
 		ps->dpmx_nb_ps_lo = 0x1;
@@ -2179,7 +2259,7 @@ static int kv_calculate_nbps_level_settings(struct radeon_device *rdev)
 	if (pi->lowest_valid > pi->highest_valid)
 		return -EINVAL;
 
-	if (rdev->family == CHIP_KABINI) {
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS) {
 		for (i = pi->lowest_valid; i <= pi->highest_valid; i++) {
 			pi->graphics_level[i].GnbSlow = 1;
 			pi->graphics_level[i].ForceNbPs1 = 0;
@@ -2253,9 +2333,9 @@ static void kv_init_graphics_levels(struct radeon_device *rdev)
 				break;
 
 			kv_set_divider_value(rdev, i, table->entries[i].clk);
-			vid_2bit = sumo_convert_vid7_to_vid2(rdev,
-							     &pi->sys_info.vid_mapping_table,
-							     table->entries[i].v);
+			vid_2bit = kv_convert_vid7_to_vid2(rdev,
+							   &pi->sys_info.vid_mapping_table,
+							   table->entries[i].v);
 			kv_set_vid(rdev, i, vid_2bit);
 			kv_set_at(rdev, i, pi->at[i]);
 			kv_dpm_power_level_enabled_for_throttle(rdev, i, true);
@@ -2324,7 +2404,7 @@ static void kv_program_nbps_index_settings(struct radeon_device *rdev,
 	struct kv_power_info *pi = kv_get_pi(rdev);
 	u32 nbdpmconfig1;
 
-	if (rdev->family == CHIP_KABINI)
+	if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS)
 		return;
 
 	if (pi->sys_info.nb_dpm_enable) {
@@ -2631,9 +2711,6 @@ int kv_dpm_init(struct radeon_device *rdev)
 
         pi->sram_end = SMC_RAM_END;
 
-	if (rdev->family == CHIP_KABINI)
-		pi->high_voltage_t = 4001;
-
 	pi->enable_nb_dpm = true;
 
 	pi->caps_power_containment = true;

drivers/gpu/drm/radeon/r600.c

@@ -2839,6 +2839,7 @@ int r600_copy_cpdma(struct radeon_device *rdev,
 	r = radeon_fence_emit(rdev, fence, ring->idx);
 	if (r) {
 		radeon_ring_unlock_undo(rdev, ring);
+		radeon_semaphore_free(rdev, &sem, NULL);
 		return r;
 	}
 
@@ -3505,7 +3506,6 @@ int r600_irq_set(struct radeon_device *rdev)
 	u32 hpd1, hpd2, hpd3, hpd4 = 0, hpd5 = 0, hpd6 = 0;
 	u32 grbm_int_cntl = 0;
 	u32 hdmi0, hdmi1;
-	u32 d1grph = 0, d2grph = 0;
 	u32 dma_cntl;
 	u32 thermal_int = 0;
 
@@ -3614,8 +3614,8 @@ int r600_irq_set(struct radeon_device *rdev)
 	WREG32(CP_INT_CNTL, cp_int_cntl);
 	WREG32(DMA_CNTL, dma_cntl);
 	WREG32(DxMODE_INT_MASK, mode_int);
-	WREG32(D1GRPH_INTERRUPT_CONTROL, d1grph);
-	WREG32(D2GRPH_INTERRUPT_CONTROL, d2grph);
+	WREG32(D1GRPH_INTERRUPT_CONTROL, DxGRPH_PFLIP_INT_MASK);
+	WREG32(D2GRPH_INTERRUPT_CONTROL, DxGRPH_PFLIP_INT_MASK);
 	WREG32(GRBM_INT_CNTL, grbm_int_cntl);
 	if (ASIC_IS_DCE3(rdev)) {
 		WREG32(DC_HPD1_INT_CONTROL, hpd1);
@@ -3918,6 +3918,14 @@ restart_ih:
 				break;
 			}
 			break;
+		case 9: /* D1 pflip */
+			DRM_DEBUG("IH: D1 flip\n");
+			radeon_crtc_handle_flip(rdev, 0);
+			break;
+		case 11: /* D2 pflip */
+			DRM_DEBUG("IH: D2 flip\n");
+			radeon_crtc_handle_flip(rdev, 1);
+			break;
 		case 19: /* HPD/DAC hotplug */
 			switch (src_data) {
 			case 0:

drivers/gpu/drm/radeon/r600_dma.c

@@ -489,6 +489,7 @@ int r600_copy_dma(struct radeon_device *rdev,
 	r = radeon_fence_emit(rdev, fence, ring->idx);
 	if (r) {
 		radeon_ring_unlock_undo(rdev, ring);
+		radeon_semaphore_free(rdev, &sem, NULL);
 		return r;
 	}
 

drivers/gpu/drm/radeon/radeon.h

@@ -730,6 +730,12 @@ struct cik_irq_stat_regs {
 	u32 disp_int_cont4;
 	u32 disp_int_cont5;
 	u32 disp_int_cont6;
+	u32 d1grph_int;
+	u32 d2grph_int;
+	u32 d3grph_int;
+	u32 d4grph_int;
+	u32 d5grph_int;
+	u32 d6grph_int;
 };
 
 union radeon_irq_stat_regs {