192 files changed, 1957 insertions, 978 deletions
diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
index 4e547296831d..52312cb5dbe2 100644
--- a/arch/arc/include/asm/processor.h
+++ b/arch/arc/include/asm/processor.h
@@ -47,9 +47,6 @@ struct thread_struct {
 /* Forward declaration, a strange C thing */
 struct task_struct;
-/* Return saved PC of a blocked thread  */
-unsigned long thread_saved_pc(struct task_struct *t);
 #define task_pt_regs(p) \
        ((struct pt_regs *)(THREAD_SIZE + (void *)task_stack_page(p)) - 1)
@@ -72,18 +69,21 @@ unsigned long thread_saved_pc(struct task_struct *t);
 #define release_segments(mm)        do { } while (0)
 #define KSTK_EIP(tsk)   (task_pt_regs(tsk)->ret)
+#define KSTK_ESP(tsk)   (task_pt_regs(tsk)->sp)
 /*
 * Where abouts of Task's sp, fp, blink when it was last seen in kernel mode.
 * Look in process.c for details of kernel stack layout
 */
-#define KSTK_ESP(tsk)   (tsk->thread.ksp)
+#define TSK_K_ESP(tsk)          (tsk->thread.ksp)
-#define KSTK_REG(tsk, off)      (*((unsigned int *)(KSTK_ESP(tsk) + \
+#define TSK_K_REG(tsk, off)     (*((unsigned int *)(TSK_K_ESP(tsk) + \
                                        sizeof(struct callee_regs) + off)))
-#define KSTK_BLINK(tsk) KSTK_REG(tsk, 4)
+#define TSK_K_BLINK(tsk)        TSK_K_REG(tsk, 4)
-#define KSTK_FP(tsk)    KSTK_REG(tsk, 0)
+#define TSK_K_FP(tsk)           TSK_K_REG(tsk, 0)
+#define thread_saved_pc(tsk)    TSK_K_BLINK(tsk)
 extern void start_thread(struct pt_regs * regs, unsigned long pc,
                         unsigned long usp);
diff --git a/arch/arc/include/asm/stacktrace.h b/arch/arc/include/asm/stacktrace.h
new file mode 100644
index 000000000000..b29b6064ea14
--- /dev/null
+++ b/arch/arc/include/asm/stacktrace.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright (C) 2014-15 Synopsys, Inc. (www.synopsys.com)
+ * Copyright (C) 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_STACKTRACE_H
+#define __ASM_STACKTRACE_H
+#include <linux/sched.h>
+/**
+ * arc_unwind_core - Unwind the kernel mode stack for an execution context
+ * @tsk:                NULL for current task, specific task otherwise
+ * @regs:               pt_regs used to seed the unwinder {SP, FP, BLINK, PC}
+ *                      If NULL, use pt_regs of @tsk (if !NULL) otherwise
+ *                      use the current values of {SP, FP, BLINK, PC}
+ * @consumer_fn:        Callback invoked for each frame unwound
+ *                      Returns 0 to continue unwinding, -1 to stop
+ * @arg:                Arg to callback
+ *
+ * Returns the address of first function in stack
+ *
+ * Semantics:
+ *  - synchronous unwinding (e.g. dump_stack): @tsk  NULL, @regs  NULL
+ *  - Asynchronous unwinding of sleeping task: @tsk !NULL, @regs  NULL
+ *  - Asynchronous unwinding of intr/excp etc: @tsk !NULL, @regs !NULL
+ */
+notrace noinline unsigned int arc_unwind_core(
+        struct task_struct *tsk, struct pt_regs *regs,
+        int (*consumer_fn) (unsigned int, void *),
+        void *arg);
+#endif /* __ASM_STACKTRACE_H */
diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c
index fdd89715d2d3..98c00a2d4dd9 100644
--- a/arch/arc/kernel/process.c
+++ b/arch/arc/kernel/process.c
@@ -192,29 +192,6 @@ int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu)
        return 0;
 }
-/*
- * API: expected by schedular Code: If thread is sleeping where is that.
- * What is this good for? it will be always the scheduler or ret_from_fork.
- * So we hard code that anyways.
- */
-unsigned long thread_saved_pc(struct task_struct *t)
-{
-        struct pt_regs *regs = task_pt_regs(t);
-        unsigned long blink = 0;
-        /*
-         * If the thread being queried for in not itself calling this, then it
-         * implies it is not executing, which in turn implies it is sleeping,
-         * which in turn implies it got switched OUT by the schedular.
-         * In that case, it's kernel mode blink can reliably retrieved as per
-         * the picture above (right above pt_regs).
-         */
-        if (t != current && t->state != TASK_RUNNING)
-                blink = *((unsigned int *)regs - 1);
-        return blink;
-}
 int elf_check_arch(const struct elf32_hdr *x)
 {
        unsigned int eflags;
diff --git a/arch/arc/kernel/stacktrace.c b/arch/arc/kernel/stacktrace.c
index 9ce47cfe2303..92320d6f737c 100644
--- a/arch/arc/kernel/stacktrace.c
+++ b/arch/arc/kernel/stacktrace.c
@@ -43,6 +43,10 @@ static void seed_unwind_frame_info(struct task_struct *tsk,
                                   struct pt_regs *regs,
                                   struct unwind_frame_info *frame_info)
 {
+        /*
+         * synchronous unwinding (e.g. dump_stack)
+         *  - uses current values of SP and friends
+         */
        if (tsk == NULL && regs == NULL) {
                unsigned long fp, sp, blink, ret;
                frame_info->task = current;
@@ -61,12 +65,17 @@ static void seed_unwind_frame_info(struct task_struct *tsk,
                frame_info->regs.r63 = ret;
                frame_info->call_frame = 0;
        } else if (regs == NULL) {
+                /*
+                 * Asynchronous unwinding of sleeping task
+                 *  - Gets SP etc from task's pt_regs (saved bottom of kernel
+                 *    mode stack of task)
+                 */
                frame_info->task = tsk;
-                frame_info->regs.r27 = KSTK_FP(tsk);
+                frame_info->regs.r27 = TSK_K_FP(tsk);
-                frame_info->regs.r28 = KSTK_ESP(tsk);
+                frame_info->regs.r28 = TSK_K_ESP(tsk);
-                frame_info->regs.r31 = KSTK_BLINK(tsk);
+                frame_info->regs.r31 = TSK_K_BLINK(tsk);
                frame_info->regs.r63 = (unsigned int)__switch_to;
                /* In the prologue of __switch_to, first FP is saved on stack
@@ -83,6 +92,10 @@ static void seed_unwind_frame_info(struct task_struct *tsk,
                frame_info->call_frame = 0;
        } else {
+                /*
+                 * Asynchronous unwinding of intr/exception
+                 *  - Just uses the pt_regs passed
+                 */
                frame_info->task = tsk;
                frame_info->regs.r27 = regs->fp;
@@ -95,7 +108,7 @@ static void seed_unwind_frame_info(struct task_struct *tsk,
 #endif
-static noinline unsigned int
+notrace noinline unsigned int
 arc_unwind_core(struct task_struct *tsk, struct pt_regs *regs,
                int (*consumer_fn) (unsigned int, void *), void *arg)
 {
diff --git a/arch/arc/kernel/unaligned.c b/arch/arc/kernel/unaligned.c
index 7ff5b5c183bb..74db59b6f392 100644
--- a/arch/arc/kernel/unaligned.c
+++ b/arch/arc/kernel/unaligned.c
@@ -12,6 +12,7 @@
 */
 #include <linux/types.h>
+#include <linux/perf_event.h>
 #include <linux/ptrace.h>
 #include <linux/uaccess.h>
 #include <asm/disasm.h>
@@ -253,6 +254,7 @@ int misaligned_fixup(unsigned long address, struct pt_regs *regs,
                }
        }
+        perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS, 1, regs, address);
        return 0;
 fault:
diff --git a/arch/arc/mm/fault.c b/arch/arc/mm/fault.c
index 563cb27e37f5..6a2e006cbcce 100644
--- a/arch/arc/mm/fault.c
+++ b/arch/arc/mm/fault.c
@@ -14,6 +14,7 @@
 #include <linux/ptrace.h>
 #include <linux/uaccess.h>
 #include <linux/kdebug.h>
+#include <linux/perf_event.h>
 #include <asm/pgalloc.h>
 #include <asm/mmu.h>
@@ -139,13 +140,20 @@ good_area:
                        return;
        }
+        perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
        if (likely(!(fault & VM_FAULT_ERROR))) {
                if (flags & FAULT_FLAG_ALLOW_RETRY) {
                        /* To avoid updating stats twice for retry case */
-                        if (fault & VM_FAULT_MAJOR)
+                        if (fault & VM_FAULT_MAJOR) {
                                tsk->maj_flt++;
-                        else
+                                perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
+                                              regs, address);
+                        } else {
                                tsk->min_flt++;
+                                perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
+                                              regs, address);
+                        }
                        if (fault & VM_FAULT_RETRY) {
                                flags &= ~FAULT_FLAG_ALLOW_RETRY;
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 7f99cd652203..eb7bb511f853 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -150,6 +150,7 @@ machine-$(CONFIG_ARCH_BERLIN)		+= berlin
 machine-$(CONFIG_ARCH_CLPS711X)         += clps711x
 machine-$(CONFIG_ARCH_CNS3XXX)          += cns3xxx
 machine-$(CONFIG_ARCH_DAVINCI)          += davinci
+machine-$(CONFIG_ARCH_DIGICOLOR)        += digicolor
 machine-$(CONFIG_ARCH_DOVE)             += dove
 machine-$(CONFIG_ARCH_EBSA110)          += ebsa110
 machine-$(CONFIG_ARCH_EFM32)            += efm32
diff --git a/arch/arm/boot/dts/am335x-bone-common.dtsi b/arch/arm/boot/dts/am335x-bone-common.dtsi
index 2c6248d9a9ef..c3255e0c90aa 100644
--- a/arch/arm/boot/dts/am335x-bone-common.dtsi
+++ b/arch/arm/boot/dts/am335x-bone-common.dtsi
@@ -301,3 +301,11 @@
        cd-gpios = <&gpio0 6 GPIO_ACTIVE_HIGH>;
        cd-inverted;
 };
+&aes {
+        status = "okay";
+};
+&sham {
+        status = "okay";
+};
diff --git a/arch/arm/boot/dts/am335x-bone.dts b/arch/arm/boot/dts/am335x-bone.dts
index 83d40f7655e5..6b8493720424 100644
--- a/arch/arm/boot/dts/am335x-bone.dts
+++ b/arch/arm/boot/dts/am335x-bone.dts
@@ -24,11 +24,3 @@
 &mmc1 {
        vmmc-supply = <&ldo3_reg>;
 };
-&sham {
-        status = "okay";
-};
-&aes {
-        status = "okay";
-};
diff --git a/arch/arm/boot/dts/am335x-lxm.dts b/arch/arm/boot/dts/am335x-lxm.dts
index 7266a00aab2e..5c5667a3624d 100644
--- a/arch/arm/boot/dts/am335x-lxm.dts
+++ b/arch/arm/boot/dts/am335x-lxm.dts
@@ -328,6 +328,10 @@
        dual_emac_res_vlan = <3>;
 };
+&phy_sel {
+        rmii-clock-ext;
+};
 &mac {
        pinctrl-names = "default", "sleep";
        pinctrl-0 = <&cpsw_default>;
diff --git a/arch/arm/boot/dts/am33xx-clocks.dtsi b/arch/arm/boot/dts/am33xx-clocks.dtsi
index 712edce7d6fb..071b56aa0c7e 100644
--- a/arch/arm/boot/dts/am33xx-clocks.dtsi
+++ b/arch/arm/boot/dts/am33xx-clocks.dtsi
@@ -99,7 +99,7 @@
        ehrpwm0_tbclk: ehrpwm0_tbclk@44e10664 {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <0>;
                reg = <0x0664>;
        };
@@ -107,7 +107,7 @@
        ehrpwm1_tbclk: ehrpwm1_tbclk@44e10664 {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <1>;
                reg = <0x0664>;
        };
@@ -115,7 +115,7 @@
        ehrpwm2_tbclk: ehrpwm2_tbclk@44e10664 {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <2>;
                reg = <0x0664>;
        };
diff --git a/arch/arm/boot/dts/am43xx-clocks.dtsi b/arch/arm/boot/dts/am43xx-clocks.dtsi
index c7dc9dab93a4..cfb49686ab6a 100644
--- a/arch/arm/boot/dts/am43xx-clocks.dtsi
+++ b/arch/arm/boot/dts/am43xx-clocks.dtsi
@@ -107,7 +107,7 @@
        ehrpwm0_tbclk: ehrpwm0_tbclk {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <0>;
                reg = <0x0664>;
        };
@@ -115,7 +115,7 @@
        ehrpwm1_tbclk: ehrpwm1_tbclk {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <1>;
                reg = <0x0664>;
        };
@@ -123,7 +123,7 @@
        ehrpwm2_tbclk: ehrpwm2_tbclk {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <2>;
                reg = <0x0664>;
        };
@@ -131,7 +131,7 @@
        ehrpwm3_tbclk: ehrpwm3_tbclk {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <4>;
                reg = <0x0664>;
        };
@@ -139,7 +139,7 @@
        ehrpwm4_tbclk: ehrpwm4_tbclk {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <5>;
                reg = <0x0664>;
        };
@@ -147,7 +147,7 @@
        ehrpwm5_tbclk: ehrpwm5_tbclk {
                #clock-cells = <0>;
                compatible = "ti,gate-clock";
-                clocks = <&dpll_per_m2_ck>;
+                clocks = <&l4ls_gclk>;
                ti,bit-shift = <6>;
                reg = <0x0664>;
        };
diff --git a/arch/arm/boot/dts/at91sam9260.dtsi b/arch/arm/boot/dts/at91sam9260.dtsi
index fff0ee69aab4..e7f0a4ae271c 100644
--- a/arch/arm/boot/dts/at91sam9260.dtsi
+++ b/arch/arm/boot/dts/at91sam9260.dtsi
@@ -494,12 +494,12 @@
                                        pinctrl_usart3_rts: usart3_rts-0 {
                                                atmel,pins =
-                                                        <AT91_PIOB 8 AT91_PERIPH_B AT91_PINCTRL_NONE>;  /* PC8 periph B */
+                                                        <AT91_PIOC 8 AT91_PERIPH_B AT91_PINCTRL_NONE>;
                                        };
                                        pinctrl_usart3_cts: usart3_cts-0 {
                                                atmel,pins =
-                                                        <AT91_PIOB 10 AT91_PERIPH_B AT91_PINCTRL_NONE>; /* PC10 periph B */
+                                                        <AT91_PIOC 10 AT91_PERIPH_B AT91_PINCTRL_NONE>;
                                        };
                                };
@@ -853,7 +853,7 @@
                        };
                        usb1: gadget@fffa4000 {
-                                compatible = "atmel,at91rm9200-udc";
+                                compatible = "atmel,at91sam9260-udc";
                                reg = <0xfffa4000 0x4000>;
                                interrupts = <10 IRQ_TYPE_LEVEL_HIGH 2>;
                                clocks = <&udc_clk>, <&udpck>;
@@ -976,7 +976,6 @@
                                atmel,watchdog-type = "hardware";
                                atmel,reset-type = "all";
                                atmel,dbg-halt;
-                                atmel,idle-halt;
                                status = "disabled";
                        };
diff --git a/arch/arm/boot/dts/at91sam9261.dtsi b/arch/arm/boot/dts/at91sam9261.dtsi
index e247b0b5fdab..d55fdf2487ef 100644
--- a/arch/arm/boot/dts/at91sam9261.dtsi
+++ b/arch/arm/boot/dts/at91sam9261.dtsi
@@ -124,11 +124,12 @@
                        };
                        usb1: gadget@fffa4000 {
-                                compatible = "atmel,at91rm9200-udc";
+                                compatible = "atmel,at91sam9261-udc";
                                reg = <0xfffa4000 0x4000>;
                                interrupts = <10 IRQ_TYPE_LEVEL_HIGH 2>;
-                                clocks = <&usb>, <&udc_clk>, <&udpck>;
+                                clocks = <&udc_clk>, <&udpck>;
-                                clock-names = "usb_clk", "udc_clk", "udpck";
+                                clock-names = "pclk", "hclk";
+                                atmel,matrix = <&matrix>;
                                status = "disabled";
                        };
@@ -262,7 +263,7 @@
                        };
                        matrix: matrix@ffffee00 {
-                                compatible = "atmel,at91sam9260-bus-matrix";
+                                compatible = "atmel,at91sam9260-bus-matrix", "syscon";
                                reg = <0xffffee00 0x200>;
                        };
diff --git a/arch/arm/boot/dts/at91sam9263.dtsi b/arch/arm/boot/dts/at91sam9263.dtsi
index 1f67bb4c144e..fce301c4e9d6 100644
--- a/arch/arm/boot/dts/at91sam9263.dtsi
+++ b/arch/arm/boot/dts/at91sam9263.dtsi
@@ -69,7 +69,7 @@
        sram1: sram@00500000 {
                compatible = "mmio-sram";
-                reg = <0x00300000 0x4000>;
+                reg = <0x00500000 0x4000>;
        };
        ahb {
@@ -856,7 +856,7 @@
                        };
                        usb1: gadget@fff78000 {
-                                compatible = "atmel,at91rm9200-udc";
+                                compatible = "atmel,at91sam9263-udc";
                                reg = <0xfff78000 0x4000>;
                                interrupts = <24 IRQ_TYPE_LEVEL_HIGH 2>;
                                clocks = <&udc_clk>, <&udpck>;
@@ -905,7 +905,6 @@
                                atmel,watchdog-type = "hardware";
                                atmel,reset-type = "all";
                                atmel,dbg-halt;
-                                atmel,idle-halt;
                                status = "disabled";
                        };
diff --git a/arch/arm/boot/dts/at91sam9g45.dtsi b/arch/arm/boot/dts/at91sam9g45.dtsi
index ee80aa9c0759..488af63d5174 100644
--- a/arch/arm/boot/dts/at91sam9g45.dtsi
+++ b/arch/arm/boot/dts/at91sam9g45.dtsi
@@ -1116,7 +1116,6 @@
                                atmel,watchdog-type = "hardware";
                                atmel,reset-type = "all";
                                atmel,dbg-halt;
-                                atmel,idle-halt;
                                status = "disabled";
                        };
@@ -1301,7 +1300,7 @@
                        compatible = "atmel,at91sam9g45-ehci", "usb-ehci";
                        reg = <0x00800000 0x100000>;
                        interrupts = <22 IRQ_TYPE_LEVEL_HIGH 2>;
-                        clocks = <&usb>, <&uhphs_clk>, <&uhphs_clk>, <&uhpck>;
+                        clocks = <&utmi>, <&uhphs_clk>, <&uhphs_clk>, <&uhpck>;
                        clock-names = "usb_clk", "ehci_clk", "hclk", "uhpck";
                        status = "disabled";
                };
diff --git a/arch/arm/boot/dts/at91sam9n12.dtsi b/arch/arm/boot/dts/at91sam9n12.dtsi
index c2666a7cb5b1..0c53a375ba99 100644
--- a/arch/arm/boot/dts/at91sam9n12.dtsi
+++ b/arch/arm/boot/dts/at91sam9n12.dtsi
@@ -894,7 +894,6 @@
                                atmel,watchdog-type = "hardware";
                                atmel,reset-type = "all";
                                atmel,dbg-halt;
-                                atmel,idle-halt;
                                status = "disabled";
                        };
diff --git a/arch/arm/boot/dts/at91sam9x5.dtsi b/arch/arm/boot/dts/at91sam9x5.dtsi
index 818dabdd8c0e..d221179d0f1a 100644
--- a/arch/arm/boot/dts/at91sam9x5.dtsi
+++ b/arch/arm/boot/dts/at91sam9x5.dtsi
@@ -1066,7 +1066,7 @@
                                reg = <0x00500000 0x80000
                                       0xf803c000 0x400>;
                                interrupts = <23 IRQ_TYPE_LEVEL_HIGH 0>;
-                                clocks = <&usb>, <&udphs_clk>;
+                                clocks = <&utmi>, <&udphs_clk>;
                                clock-names = "hclk", "pclk";
                                status = "disabled";
@@ -1130,7 +1130,6 @@
                                atmel,watchdog-type = "hardware";
                                atmel,reset-type = "all";
                                atmel,dbg-halt;
-                                atmel,idle-halt;
                                status = "disabled";
                        };
@@ -1186,7 +1185,7 @@
                        compatible = "atmel,at91sam9g45-ehci", "usb-ehci";
                        reg = <0x00700000 0x100000>;
                        interrupts = <22 IRQ_TYPE_LEVEL_HIGH 2>;
-                        clocks = <&usb>, <&uhphs_clk>, <&uhpck>;
+                        clocks = <&utmi>, <&uhphs_clk>, <&uhpck>;
                        clock-names = "usb_clk", "ehci_clk", "uhpck";
                        status = "disabled";
                };
diff --git a/arch/arm/boot/dts/dra7-evm.dts b/arch/arm/boot/dts/dra7-evm.dts
index 3290a96ba586..7563d7ce01bb 100644
--- a/arch/arm/boot/dts/dra7-evm.dts
+++ b/arch/arm/boot/dts/dra7-evm.dts
@@ -263,17 +263,15 @@
        dcan1_pins_default: dcan1_pins_default {
                pinctrl-single,pins = <
-                        0x3d0   (PIN_OUTPUT | MUX_MODE0) /* dcan1_tx */
+                        0x3d0   (PIN_OUTPUT_PULLUP | MUX_MODE0) /* dcan1_tx */
-                        0x3d4   (MUX_MODE15)            /* dcan1_rx.off */
+                        0x418   (PULL_UP | MUX_MODE1) /* wakeup0.dcan1_rx */
-                        0x418   (PULL_DIS | MUX_MODE1) /* wakeup0.dcan1_rx */
                >;
        };
        dcan1_pins_sleep: dcan1_pins_sleep {
                pinctrl-single,pins = <
-                        0x3d0   (MUX_MODE15)    /* dcan1_tx.off */
+                        0x3d0   (MUX_MODE15 | PULL_UP)  /* dcan1_tx.off */
-                        0x3d4   (MUX_MODE15)    /* dcan1_rx.off */
+                        0x418   (MUX_MODE15 | PULL_UP)  /* wakeup0.off */
-                        0x418   (MUX_MODE15)    /* wakeup0.off */
                >;
        };
 };
diff --git a/arch/arm/boot/dts/dra72-evm.dts b/arch/arm/boot/dts/dra72-evm.dts
index e0264d0bf7b9..40ed539ce474 100644
--- a/arch/arm/boot/dts/dra72-evm.dts
+++ b/arch/arm/boot/dts/dra72-evm.dts
@@ -119,17 +119,15 @@
        dcan1_pins_default: dcan1_pins_default {
                pinctrl-single,pins = <
-                        0x3d0   (PIN_OUTPUT | MUX_MODE0) /* dcan1_tx */
+                        0x3d0   (PIN_OUTPUT_PULLUP | MUX_MODE0) /* dcan1_tx */
-                        0x3d4   (MUX_MODE15)            /* dcan1_rx.off */
+                        0x418   (PULL_UP | MUX_MODE1)   /* wakeup0.dcan1_rx */
-                        0x418   (PULL_DIS | MUX_MODE1) /* wakeup0.dcan1_rx */
                >;
        };
        dcan1_pins_sleep: dcan1_pins_sleep {
                pinctrl-single,pins = <
-                        0x3d0   (MUX_MODE15)    /* dcan1_tx.off */
+                        0x3d0   (MUX_MODE15 | PULL_UP)  /* dcan1_tx.off */
-                        0x3d4   (MUX_MODE15)    /* dcan1_rx.off */
+                        0x418   (MUX_MODE15 | PULL_UP)  /* wakeup0.off */
-                        0x418   (MUX_MODE15)    /* wakeup0.off */
                >;
        };
diff --git a/arch/arm/boot/dts/dra7xx-clocks.dtsi b/arch/arm/boot/dts/dra7xx-clocks.dtsi
index 4bdcbd61ce47..99b09a44e269 100644
--- a/arch/arm/boot/dts/dra7xx-clocks.dtsi
+++ b/arch/arm/boot/dts/dra7xx-clocks.dtsi
@@ -243,10 +243,18 @@
                ti,invert-autoidle-bit;
        };
+        dpll_core_byp_mux: dpll_core_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                ti,bit-shift = <23>;
+                reg = <0x012c>;
+        };
        dpll_core_ck: dpll_core_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-core-clock";
-                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                clocks = <&sys_clkin1>, <&dpll_core_byp_mux>;
                reg = <0x0120>, <0x0124>, <0x012c>, <0x0128>;
        };
@@ -309,10 +317,18 @@
                clock-div = <1>;
        };
+        dpll_dsp_byp_mux: dpll_dsp_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&dsp_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x0240>;
+        };
        dpll_dsp_ck: dpll_dsp_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&dsp_dpll_hs_clk_div>;
+                clocks = <&sys_clkin1>, <&dpll_dsp_byp_mux>;
                reg = <0x0234>, <0x0238>, <0x0240>, <0x023c>;
        };
@@ -335,10 +351,18 @@
                clock-div = <1>;
        };
+        dpll_iva_byp_mux: dpll_iva_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&iva_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x01ac>;
+        };
        dpll_iva_ck: dpll_iva_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&iva_dpll_hs_clk_div>;
+                clocks = <&sys_clkin1>, <&dpll_iva_byp_mux>;
                reg = <0x01a0>, <0x01a4>, <0x01ac>, <0x01a8>;
        };
@@ -361,10 +385,18 @@
                clock-div = <1>;
        };
+        dpll_gpu_byp_mux: dpll_gpu_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                ti,bit-shift = <23>;
+                reg = <0x02e4>;
+        };
        dpll_gpu_ck: dpll_gpu_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                clocks = <&sys_clkin1>, <&dpll_gpu_byp_mux>;
                reg = <0x02d8>, <0x02dc>, <0x02e4>, <0x02e0>;
        };
@@ -398,10 +430,18 @@
                clock-div = <1>;
        };
+        dpll_ddr_byp_mux: dpll_ddr_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                ti,bit-shift = <23>;
+                reg = <0x021c>;
+        };
        dpll_ddr_ck: dpll_ddr_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                clocks = <&sys_clkin1>, <&dpll_ddr_byp_mux>;
                reg = <0x0210>, <0x0214>, <0x021c>, <0x0218>;
        };
@@ -416,10 +456,18 @@
                ti,invert-autoidle-bit;
        };
+        dpll_gmac_byp_mux: dpll_gmac_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                ti,bit-shift = <23>;
+                reg = <0x02b4>;
+        };
        dpll_gmac_ck: dpll_gmac_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&dpll_abe_m3x2_ck>;
+                clocks = <&sys_clkin1>, <&dpll_gmac_byp_mux>;
                reg = <0x02a8>, <0x02ac>, <0x02b4>, <0x02b0>;
        };
@@ -482,10 +530,18 @@
                clock-div = <1>;
        };
+        dpll_eve_byp_mux: dpll_eve_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&eve_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x0290>;
+        };
        dpll_eve_ck: dpll_eve_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&eve_dpll_hs_clk_div>;
+                clocks = <&sys_clkin1>, <&dpll_eve_byp_mux>;
                reg = <0x0284>, <0x0288>, <0x0290>, <0x028c>;
        };
@@ -1249,10 +1305,18 @@
                clock-div = <1>;
        };
+        dpll_per_byp_mux: dpll_per_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&per_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x014c>;
+        };
        dpll_per_ck: dpll_per_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin1>, <&per_dpll_hs_clk_div>;
+                clocks = <&sys_clkin1>, <&dpll_per_byp_mux>;
                reg = <0x0140>, <0x0144>, <0x014c>, <0x0148>;
        };
@@ -1275,10 +1339,18 @@
                clock-div = <1>;
        };
+        dpll_usb_byp_mux: dpll_usb_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin1>, <&usb_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x018c>;
+        };
        dpll_usb_ck: dpll_usb_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-j-type-clock";
-                clocks = <&sys_clkin1>, <&usb_dpll_hs_clk_div>;
+                clocks = <&sys_clkin1>, <&dpll_usb_byp_mux>;
                reg = <0x0180>, <0x0184>, <0x018c>, <0x0188>;
        };
diff --git a/arch/arm/boot/dts/exynos3250.dtsi b/arch/arm/boot/dts/exynos3250.dtsi
index 277b48b0b6f9..ac6b0ae42caf 100644
--- a/arch/arm/boot/dts/exynos3250.dtsi
+++ b/arch/arm/boot/dts/exynos3250.dtsi
@@ -18,6 +18,7 @@
 */
 #include "skeleton.dtsi"
+#include "exynos4-cpu-thermal.dtsi"
 #include <dt-bindings/clock/exynos3250.h>
 / {
@@ -193,6 +194,7 @@
                        interrupts = <0 216 0>;
                        clocks = <&cmu CLK_TMU_APBIF>;
                        clock-names = "tmu_apbif";
+                        #include "exynos4412-tmu-sensor-conf.dtsi"
                        status = "disabled";
                };
diff --git a/arch/arm/boot/dts/exynos4-cpu-thermal.dtsi b/arch/arm/boot/dts/exynos4-cpu-thermal.dtsi
new file mode 100644
index 000000000000..735cb2f10817
--- /dev/null
+++ b/arch/arm/boot/dts/exynos4-cpu-thermal.dtsi
@@ -0,0 +1,52 @@
+/*
+ * Device tree sources for Exynos4 thermal zone
+ *
+ * Copyright (c) 2014 Lukasz Majewski <l.majewski@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <dt-bindings/thermal/thermal.h>
+/ {
+thermal-zones {
+        cpu_thermal: cpu-thermal {
+                thermal-sensors = <&tmu 0>;
+                polling-delay-passive = <0>;
+                polling-delay = <0>;
+                trips {
+                        cpu_alert0: cpu-alert-0 {
+                                temperature = <70000>; /* millicelsius */
+                                hysteresis = <10000>; /* millicelsius */
+                                type = "active";
+                        };
+                        cpu_alert1: cpu-alert-1 {
+                                temperature = <95000>; /* millicelsius */
+                                hysteresis = <10000>; /* millicelsius */
+                                type = "active";
+                        };
+                        cpu_alert2: cpu-alert-2 {
+                                temperature = <110000>; /* millicelsius */
+                                hysteresis = <10000>; /* millicelsius */
+                                type = "active";
+                        };
+                        cpu_crit0: cpu-crit-0 {
+                                temperature = <120000>; /* millicelsius */
+                                hysteresis = <0>; /* millicelsius */
+                                type = "critical";
+                        };
+                };
+                cooling-maps {
+                        map0 {
+                                trip = <&cpu_alert0>;
+                        };
+                        map1 {
+                                trip = <&cpu_alert1>;
+                        };
+                };
+        };
+};
+};
diff --git a/arch/arm/boot/dts/exynos4.dtsi b/arch/arm/boot/dts/exynos4.dtsi
index 76173cacd450..77ea547768f4 100644
--- a/arch/arm/boot/dts/exynos4.dtsi
+++ b/arch/arm/boot/dts/exynos4.dtsi
@@ -38,6 +38,7 @@
                i2c5 = &i2c_5;
                i2c6 = &i2c_6;
                i2c7 = &i2c_7;
+                i2c8 = &i2c_8;
                csis0 = &csis_0;
                csis1 = &csis_1;
                fimc0 = &fimc_0;
@@ -104,6 +105,7 @@
                compatible = "samsung,exynos4210-pd";
                reg = <0x10023C20 0x20>;
                #power-domain-cells = <0>;
+                power-domains = <&pd_lcd0>;
        };
        pd_cam: cam-power-domain@10023C00 {
@@ -554,6 +556,22 @@
                status = "disabled";
        };
+        i2c_8: i2c@138E0000 {
+                #address-cells = <1>;
+                #size-cells = <0>;
+                compatible = "samsung,s3c2440-hdmiphy-i2c";
+                reg = <0x138E0000 0x100>;
+                interrupts = <0 93 0>;
+                clocks = <&clock CLK_I2C_HDMI>;
+                clock-names = "i2c";
+                status = "disabled";
+                hdmi_i2c_phy: hdmiphy@38 {
+                        compatible = "exynos4210-hdmiphy";
+                        reg = <0x38>;
+                };
+        };
        spi_0: spi@13920000 {
                compatible = "samsung,exynos4210-spi";
                reg = <0x13920000 0x100>;
@@ -663,6 +681,33 @@
                status = "disabled";
        };
+        tmu: tmu@100C0000 {
+                #include "exynos4412-tmu-sensor-conf.dtsi"
+        };
+        hdmi: hdmi@12D00000 {
+                compatible = "samsung,exynos4210-hdmi";
+                reg = <0x12D00000 0x70000>;
+                interrupts = <0 92 0>;
+                clock-names = "hdmi", "sclk_hdmi", "sclk_pixel", "sclk_hdmiphy",
+                        "mout_hdmi";
+                clocks = <&clock CLK_HDMI>, <&clock CLK_SCLK_HDMI>,
+                        <&clock CLK_SCLK_PIXEL>, <&clock CLK_SCLK_HDMIPHY>,
+                        <&clock CLK_MOUT_HDMI>;
+                phy = <&hdmi_i2c_phy>;
+                power-domains = <&pd_tv>;
+                samsung,syscon-phandle = <&pmu_system_controller>;
+                status = "disabled";
+        };
+        mixer: mixer@12C10000 {
+                compatible = "samsung,exynos4210-mixer";
+                interrupts = <0 91 0>;
+                reg = <0x12C10000 0x2100>, <0x12c00000 0x300>;
+                power-domains = <&pd_tv>;
+                status = "disabled";
+        };
        ppmu_dmc0: ppmu_dmc0@106a0000 {
                compatible = "samsung,exynos-ppmu";
                reg = <0x106a0000 0x2000>;
diff --git a/arch/arm/boot/dts/exynos4210-trats.dts b/arch/arm/boot/dts/exynos4210-trats.dts
index 3d6652a4b6cb..32c5fd8f6269 100644
--- a/arch/arm/boot/dts/exynos4210-trats.dts
+++ b/arch/arm/boot/dts/exynos4210-trats.dts
@@ -426,6 +426,25 @@
                status = "okay";
        };
+        tmu@100C0000 {
+                status = "okay";
+        };
+        thermal-zones {
+                cpu_thermal: cpu-thermal {
+                        cooling-maps {
+                                map0 {
+                                     /* Corresponds to 800MHz at freq_table */
+                                     cooling-device = <&cpu0 2 2>;
+                                };
+                                map1 {
+                                     /* Corresponds to 200MHz at freq_table */
+                                     cooling-device = <&cpu0 4 4>;
+                               };
+                       };
+                };
+        };
        camera {
                pinctrl-names = "default";
                pinctrl-0 = <>;
diff --git a/arch/arm/boot/dts/exynos4210-universal_c210.dts b/arch/arm/boot/dts/exynos4210-universal_c210.dts
index b57e6b82ea20..d4f2b11319dd 100644
--- a/arch/arm/boot/dts/exynos4210-universal_c210.dts
+++ b/arch/arm/boot/dts/exynos4210-universal_c210.dts
@@ -505,6 +505,63 @@
                        assigned-clock-rates = <0>, <160000000>;
                };
        };
+        hdmi_en: voltage-regulator-hdmi-5v {
+                compatible = "regulator-fixed";
+                regulator-name = "HDMI_5V";
+                regulator-min-microvolt = <5000000>;
+                regulator-max-microvolt = <5000000>;
+                gpio = <&gpe0 1 0>;
+                enable-active-high;
+        };
+        hdmi_ddc: i2c-ddc {
+                compatible = "i2c-gpio";
+                gpios = <&gpe4 2 0 &gpe4 3 0>;
+                i2c-gpio,delay-us = <100>;
+                #address-cells = <1>;
+                #size-cells = <0>;
+                pinctrl-0 = <&i2c_ddc_bus>;
+                pinctrl-names = "default";
+                status = "okay";
+        };
+        mixer@12C10000 {
+                status = "okay";
+        };
+        hdmi@12D00000 {
+                hpd-gpio = <&gpx3 7 0>;
+                pinctrl-names = "default";
+                pinctrl-0 = <&hdmi_hpd>;
+                hdmi-en-supply = <&hdmi_en>;
+                vdd-supply = <&ldo3_reg>;
+                vdd_osc-supply = <&ldo4_reg>;
+                vdd_pll-supply = <&ldo3_reg>;
+                ddc = <&hdmi_ddc>;
+                status = "okay";
+        };
+        i2c@138E0000 {
+                status = "okay";
+        };
+};
+&pinctrl_1 {
+        hdmi_hpd: hdmi-hpd {
+                samsung,pins = "gpx3-7";
+                samsung,pin-pud = <0>;
+        };
+};
+&pinctrl_0 {
+        i2c_ddc_bus: i2c-ddc-bus {
+                samsung,pins = "gpe4-2", "gpe4-3";
+                samsung,pin-function = <2>;
+                samsung,pin-pud = <3>;
+                samsung,pin-drv = <0>;
+        };
 };
 &mdma1 {
diff --git a/arch/arm/boot/dts/exynos4210.dtsi b/arch/arm/boot/dts/exynos4210.dtsi
index 67c832c9dcf1..be89f83f70e7 100644
--- a/arch/arm/boot/dts/exynos4210.dtsi
+++ b/arch/arm/boot/dts/exynos4210.dtsi
@@ -21,6 +21,7 @@
 #include "exynos4.dtsi"
 #include "exynos4210-pinctrl.dtsi"
+#include "exynos4-cpu-thermal.dtsi"
 / {
        compatible = "samsung,exynos4210", "samsung,exynos4";
@@ -35,10 +36,13 @@
                #address-cells = <1>;
                #size-cells = <0>;
-                cpu@900 {
+                cpu0: cpu@900 {
                        device_type = "cpu";
                        compatible = "arm,cortex-a9";
                        reg = <0x900>;
+                        cooling-min-level = <4>;
+                        cooling-max-level = <2>;
+                        #cooling-cells = <2>; /* min followed by max */
                };
                cpu@901 {
@@ -153,16 +157,38 @@
                reg = <0x03860000 0x1000>;
        };
-        tmu@100C0000 {
+        tmu: tmu@100C0000 {
                compatible = "samsung,exynos4210-tmu";
                interrupt-parent = <&combiner>;
                reg = <0x100C0000 0x100>;
                interrupts = <2 4>;
                clocks = <&clock CLK_TMU_APBIF>;
                clock-names = "tmu_apbif";
+                samsung,tmu_gain = <15>;
+                samsung,tmu_reference_voltage = <7>;
                status = "disabled";
        };
+        thermal-zones {
+                cpu_thermal: cpu-thermal {
+                        polling-delay-passive = <0>;
+                        polling-delay = <0>;
+                        thermal-sensors = <&tmu 0>;
+                        trips {
+                              cpu_alert0: cpu-alert-0 {
+                                      temperature = <85000>; /* millicelsius */
+                              };
+                              cpu_alert1: cpu-alert-1 {
+                                      temperature = <100000>; /* millicelsius */
+                              };
+                              cpu_alert2: cpu-alert-2 {
+                                      temperature = <110000>; /* millicelsius */
+                              };
+                        };
+                };
+        };
        g2d@12800000 {
                compatible = "samsung,s5pv210-g2d";
                reg = <0x12800000 0x1000>;
@@ -203,6 +229,14 @@
                };
        };
+        mixer: mixer@12C10000 {
+                clock-names = "mixer", "hdmi", "sclk_hdmi", "vp", "mout_mixer",
+                        "sclk_mixer";
+                clocks = <&clock CLK_MIXER>, <&clock CLK_HDMI>,
+                        <&clock CLK_SCLK_HDMI>, <&clock CLK_VP>,
+                        <&clock CLK_MOUT_MIXER>, <&clock CLK_SCLK_MIXER>;
+        };
        ppmu_lcd1: ppmu_lcd1@12240000 {
                compatible = "samsung,exynos-ppmu";
                reg = <0x12240000 0x2000>;
diff --git a/arch/arm/boot/dts/exynos4212.dtsi b/arch/arm/boot/dts/exynos4212.dtsi
index dd0a43ec56da..5be03288f1ee 100644
--- a/arch/arm/boot/dts/exynos4212.dtsi
+++ b/arch/arm/boot/dts/exynos4212.dtsi
@@ -26,10 +26,13 @@
                #address-cells = <1>;
                #size-cells = <0>;
-                cpu@A00 {
+                cpu0: cpu@A00 {
                        device_type = "cpu";
                        compatible = "arm,cortex-a9";
                        reg = <0xA00>;
+                        cooling-min-level = <13>;
+                        cooling-max-level = <7>;
+                        #cooling-cells = <2>; /* min followed by max */
                };
                cpu@A01 {
diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
index de80b5bba204..adb4f6a97a1d 100644
--- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
+++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
@@ -249,6 +249,20 @@
                                        regulator-always-on;
                                };
+                                ldo8_reg: ldo@8 {
+                                        regulator-compatible = "LDO8";
+                                        regulator-name = "VDD10_HDMI_1.0V";
+                                        regulator-min-microvolt = <1000000>;
+                                        regulator-max-microvolt = <1000000>;
+                                };
+                                ldo10_reg: ldo@10 {
+                                        regulator-compatible = "LDO10";
+                                        regulator-name = "VDDQ_MIPIHSI_1.8V";
+                                        regulator-min-microvolt = <1800000>;
+                                        regulator-max-microvolt = <1800000>;
+                                };
                                ldo11_reg: LDO11 {
                                        regulator-name = "VDD18_ABB1_1.8V";
                                        regulator-min-microvolt = <1800000>;
@@ -411,6 +425,51 @@
        ehci: ehci@12580000 {
                status = "okay";
        };
+        tmu@100C0000 {
+                vtmu-supply = <&ldo10_reg>;
+                status = "okay";
+        };
+        thermal-zones {
+                cpu_thermal: cpu-thermal {
+                        cooling-maps {
+                                map0 {
+                                     /* Corresponds to 800MHz at freq_table */
+                                     cooling-device = <&cpu0 7 7>;
+                                };
+                                map1 {
+                                     /* Corresponds to 200MHz at freq_table */
+                                     cooling-device = <&cpu0 13 13>;
+                               };
+                       };
+                };
+        };
+        mixer: mixer@12C10000 {
+                status = "okay";
+        };
+        hdmi@12D00000 {
+                hpd-gpio = <&gpx3 7 0>;
+                pinctrl-names = "default";
+                pinctrl-0 = <&hdmi_hpd>;
+                vdd-supply = <&ldo8_reg>;
+                vdd_osc-supply = <&ldo10_reg>;
+                vdd_pll-supply = <&ldo8_reg>;
+                ddc = <&hdmi_ddc>;
+                status = "okay";
+        };
+        hdmi_ddc: i2c@13880000 {
+                status = "okay";
+                pinctrl-names = "default";
+                pinctrl-0 = <&i2c2_bus>;
+        };
+        i2c@138E0000 {
+                status = "okay";
+        };
 };
 &pinctrl_1 {
@@ -425,4 +484,9 @@
                samsung,pin-pud = <0>;
                samsung,pin-drv = <0>;
        };
+        hdmi_hpd: hdmi-hpd {
+                samsung,pins = "gpx3-7";
+                samsung,pin-pud = <1>;
+        };
 };
diff --git a/arch/arm/boot/dts/exynos4412-tmu-sensor-conf.dtsi b/arch/arm/boot/dts/exynos4412-tmu-sensor-conf.dtsi
new file mode 100644
index 000000000000..e3f7934d19d0
--- /dev/null
+++ b/arch/arm/boot/dts/exynos4412-tmu-sensor-conf.dtsi
@@ -0,0 +1,24 @@
+/*
+ * Device tree sources for Exynos4412 TMU sensor configuration
+ *
+ * Copyright (c) 2014 Lukasz Majewski <l.majewski@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <dt-bindings/thermal/thermal_exynos.h>
+#thermal-sensor-cells = <0>;
+samsung,tmu_gain = <8>;
+samsung,tmu_reference_voltage = <16>;
+samsung,tmu_noise_cancel_mode = <4>;
+samsung,tmu_efuse_value = <55>;
+samsung,tmu_min_efuse_value = <40>;
+samsung,tmu_max_efuse_value = <100>;
+samsung,tmu_first_point_trim = <25>;
+samsung,tmu_second_point_trim = <85>;
+samsung,tmu_default_temp_offset = <50>;
+samsung,tmu_cal_type = <TYPE_ONE_POINT_TRIMMING>;
diff --git a/arch/arm/boot/dts/exynos4412-trats2.dts b/arch/arm/boot/dts/exynos4412-trats2.dts
index 21f748083586..173ffa479ad3 100644
--- a/arch/arm/boot/dts/exynos4412-trats2.dts
+++ b/arch/arm/boot/dts/exynos4412-trats2.dts
@@ -927,6 +927,21 @@
                pulldown-ohm = <100000>; /* 100K */
                io-channels = <&adc 2>;  /* Battery temperature */
        };
+        thermal-zones {
+                cpu_thermal: cpu-thermal {
+                        cooling-maps {
+                                map0 {
+                                     /* Corresponds to 800MHz at freq_table */
+                                     cooling-device = <&cpu0 7 7>;
+                                };
+                                map1 {
+                                     /* Corresponds to 200MHz at freq_table */
+                                     cooling-device = <&cpu0 13 13>;
+                               };
+                       };
+                };
+        };
 };
 &pmu_system_controller {
diff --git a/arch/arm/boot/dts/exynos4412.dtsi b/arch/arm/boot/dts/exynos4412.dtsi
index 0f6ec93bb1d8..68ad43b391ae 100644
--- a/arch/arm/boot/dts/exynos4412.dtsi
+++ b/arch/arm/boot/dts/exynos4412.dtsi
@@ -26,10 +26,13 @@
                #address-cells = <1>;
                #size-cells = <0>;
-                cpu@A00 {
+                cpu0: cpu@A00 {
                        device_type = "cpu";
                        compatible = "arm,cortex-a9";
                        reg = <0xA00>;
+                        cooling-min-level = <13>;
+                        cooling-max-level = <7>;
+                        #cooling-cells = <2>; /* min followed by max */
                };
                cpu@A01 {
diff --git a/arch/arm/boot/dts/exynos4x12.dtsi b/arch/arm/boot/dts/exynos4x12.dtsi
index f5e0ae780d6c..6a6abe14fd9b 100644
--- a/arch/arm/boot/dts/exynos4x12.dtsi
+++ b/arch/arm/boot/dts/exynos4x12.dtsi
@@ -19,6 +19,7 @@
 #include "exynos4.dtsi"
 #include "exynos4x12-pinctrl.dtsi"
+#include "exynos4-cpu-thermal.dtsi"
 / {
        aliases {
@@ -297,4 +298,15 @@
                clock-names = "tmu_apbif";
                status = "disabled";
        };
+        hdmi: hdmi@12D00000 {
+                compatible = "samsung,exynos4212-hdmi";
+        };
+        mixer: mixer@12C10000 {
+                compatible = "samsung,exynos4212-mixer";
+                clock-names = "mixer", "hdmi", "sclk_hdmi", "vp";
+                clocks = <&clock CLK_MIXER>, <&clock CLK_HDMI>,
+                         <&clock CLK_SCLK_HDMI>, <&clock CLK_VP>;
+        };
 };
diff --git a/arch/arm/boot/dts/exynos5250.dtsi b/arch/arm/boot/dts/exynos5250.dtsi
index 9bb1b0b738f5..adbde1adad95 100644
--- a/arch/arm/boot/dts/exynos5250.dtsi
+++ b/arch/arm/boot/dts/exynos5250.dtsi
@@ -20,7 +20,7 @@
 #include <dt-bindings/clock/exynos5250.h>
 #include "exynos5.dtsi"
 #include "exynos5250-pinctrl.dtsi"
+#include "exynos4-cpu-thermal.dtsi"
 #include <dt-bindings/clock/exynos-audss-clk.h>
 / {
@@ -58,11 +58,14 @@
                #address-cells = <1>;
                #size-cells = <0>;
-                cpu@0 {
+                cpu0: cpu@0 {
                        device_type = "cpu";
                        compatible = "arm,cortex-a15";
                        reg = <0>;
                        clock-frequency = <1700000000>;
+                        cooling-min-level = <15>;
+                        cooling-max-level = <9>;
+                        #cooling-cells = <2>; /* min followed by max */
                };
                cpu@1 {
                        device_type = "cpu";
@@ -102,6 +105,12 @@
                #power-domain-cells = <0>;
        };
+        pd_disp1: disp1-power-domain@100440A0 {
+                compatible = "samsung,exynos4210-pd";
+                reg = <0x100440A0 0x20>;
+                #power-domain-cells = <0>;
+        };
        clock: clock-controller@10010000 {
                compatible = "samsung,exynos5250-clock";
                reg = <0x10010000 0x30000>;
@@ -235,12 +244,32 @@
                status = "disabled";
        };
-        tmu@10060000 {
+        tmu: tmu@10060000 {
                compatible = "samsung,exynos5250-tmu";
                reg = <0x10060000 0x100>;
                interrupts = <0 65 0>;
                clocks = <&clock CLK_TMU>;
                clock-names = "tmu_apbif";
+                #include "exynos4412-tmu-sensor-conf.dtsi"
+        };
+        thermal-zones {
+                cpu_thermal: cpu-thermal {
+                        polling-delay-passive = <0>;
+                        polling-delay = <0>;
+                        thermal-sensors = <&tmu 0>;
+                        cooling-maps {
+                                map0 {
+                                     /* Corresponds to 800MHz at freq_table */
+                                     cooling-device = <&cpu0 9 9>;
+                                };
+                                map1 {
+                                     /* Corresponds to 200MHz at freq_table */
+                                     cooling-device = <&cpu0 15 15>;
+                               };
+                       };
+                };
        };
        serial@12C00000 {
@@ -719,6 +748,7 @@
        hdmi: hdmi {
                compatible = "samsung,exynos4212-hdmi";
                reg = <0x14530000 0x70000>;
+                power-domains = <&pd_disp1>;
                interrupts = <0 95 0>;
                clocks = <&clock CLK_HDMI>, <&clock CLK_SCLK_HDMI>,
                         <&clock CLK_SCLK_PIXEL>, <&clock CLK_SCLK_HDMIPHY>,
@@ -731,9 +761,11 @@
        mixer {
                compatible = "samsung,exynos5250-mixer";
                reg = <0x14450000 0x10000>;
+                power-domains = <&pd_disp1>;
                interrupts = <0 94 0>;
-                clocks = <&clock CLK_MIXER>, <&clock CLK_SCLK_HDMI>;
+                clocks = <&clock CLK_MIXER>, <&clock CLK_HDMI>,
-                clock-names = "mixer", "sclk_hdmi";
+                         <&clock CLK_SCLK_HDMI>;
+                clock-names = "mixer", "hdmi", "sclk_hdmi";
        };
        dp_phy: video-phy@10040720 {
@@ -743,6 +775,7 @@
        };
        dp: dp-controller@145B0000 {
+                power-domains = <&pd_disp1>;
                clocks = <&clock CLK_DP>;
                clock-names = "dp";
                phys = <&dp_phy>;
@@ -750,6 +783,7 @@
        };
        fimd: fimd@14400000 {
+                power-domains = <&pd_disp1>;
                clocks = <&clock CLK_SCLK_FIMD1>, <&clock CLK_FIMD1>;
                clock-names = "sclk_fimd", "fimd";
        };
diff --git a/arch/arm/boot/dts/exynos5420-trip-points.dtsi b/arch/arm/boot/dts/exynos5420-trip-points.dtsi
new file mode 100644
index 000000000000..5d31fc140823
--- /dev/null
+++ b/arch/arm/boot/dts/exynos5420-trip-points.dtsi
@@ -0,0 +1,35 @@
+/*
+ * Device tree sources for default Exynos5420 thermal zone definition
+ *
+ * Copyright (c) 2014 Lukasz Majewski <l.majewski@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+polling-delay-passive = <0>;
+polling-delay = <0>;
+trips {
+        cpu-alert-0 {
+                temperature = <85000>; /* millicelsius */
+                hysteresis = <10000>; /* millicelsius */
+                type = "active";
+        };
+        cpu-alert-1 {
+                temperature = <103000>; /* millicelsius */
+                hysteresis = <10000>; /* millicelsius */
+                type = "active";
+        };
+        cpu-alert-2 {
+                temperature = <110000>; /* millicelsius */
+                hysteresis = <10000>; /* millicelsius */
+                type = "active";
+        };
+        cpu-crit-0 {
+                temperature = <1200000>; /* millicelsius */
+                hysteresis = <0>; /* millicelsius */
+                type = "critical";
+        };
+};
diff --git a/arch/arm/boot/dts/exynos5420.dtsi b/arch/arm/boot/dts/exynos5420.dtsi
index 9dc2e9773b30..c0e98cf3514f 100644
--- a/arch/arm/boot/dts/exynos5420.dtsi
+++ b/arch/arm/boot/dts/exynos5420.dtsi
@@ -740,8 +740,9 @@
                compatible = "samsung,exynos5420-mixer";
                reg = <0x14450000 0x10000>;
                interrupts = <0 94 0>;
-                clocks = <&clock CLK_MIXER>, <&clock CLK_SCLK_HDMI>;
+                clocks = <&clock CLK_MIXER>, <&clock CLK_HDMI>,
-                clock-names = "mixer", "sclk_hdmi";
+                         <&clock CLK_SCLK_HDMI>;
+                clock-names = "mixer", "hdmi", "sclk_hdmi";
                power-domains = <&disp_pd>;
        };
@@ -782,6 +783,7 @@
                interrupts = <0 65 0>;
                clocks = <&clock CLK_TMU>;
                clock-names = "tmu_apbif";
+                #include "exynos4412-tmu-sensor-conf.dtsi"
        };
        tmu_cpu1: tmu@10064000 {
@@ -790,6 +792,7 @@
                interrupts = <0 183 0>;
                clocks = <&clock CLK_TMU>;
                clock-names = "tmu_apbif";
+                #include "exynos4412-tmu-sensor-conf.dtsi"
        };
        tmu_cpu2: tmu@10068000 {
@@ -798,6 +801,7 @@
                interrupts = <0 184 0>;
                clocks = <&clock CLK_TMU>, <&clock CLK_TMU>;
                clock-names = "tmu_apbif", "tmu_triminfo_apbif";
+                #include "exynos4412-tmu-sensor-conf.dtsi"
        };
        tmu_cpu3: tmu@1006c000 {
@@ -806,6 +810,7 @@
                interrupts = <0 185 0>;
                clocks = <&clock CLK_TMU>, <&clock CLK_TMU_GPU>;
                clock-names = "tmu_apbif", "tmu_triminfo_apbif";
+                #include "exynos4412-tmu-sensor-conf.dtsi"
        };
        tmu_gpu: tmu@100a0000 {
@@ -814,6 +819,30 @@
                interrupts = <0 215 0>;
                clocks = <&clock CLK_TMU_GPU>, <&clock CLK_TMU>;
                clock-names = "tmu_apbif", "tmu_triminfo_apbif";
+                #include "exynos4412-tmu-sensor-conf.dtsi"
+        };
+        thermal-zones {
+                cpu0_thermal: cpu0-thermal {
+                        thermal-sensors = <&tmu_cpu0>;
+                        #include "exynos5420-trip-points.dtsi"
+                };
+                cpu1_thermal: cpu1-thermal {
+                       thermal-sensors = <&tmu_cpu1>;
+                       #include "exynos5420-trip-points.dtsi"
+                };
+                cpu2_thermal: cpu2-thermal {
+                       thermal-sensors = <&tmu_cpu2>;
+                       #include "exynos5420-trip-points.dtsi"
+                };
+                cpu3_thermal: cpu3-thermal {
+                       thermal-sensors = <&tmu_cpu3>;
+                       #include "exynos5420-trip-points.dtsi"
+                };
+                gpu_thermal: gpu-thermal {
+                       thermal-sensors = <&tmu_gpu>;
+                       #include "exynos5420-trip-points.dtsi"
+                };
        };
        watchdog: watchdog@101D0000 {
diff --git a/arch/arm/boot/dts/exynos5440-tmu-sensor-conf.dtsi b/arch/arm/boot/dts/exynos5440-tmu-sensor-conf.dtsi
new file mode 100644
index 000000000000..7b2fba0ae92b
--- /dev/null
+++ b/arch/arm/boot/dts/exynos5440-tmu-sensor-conf.dtsi
@@ -0,0 +1,24 @@
+/*
+ * Device tree sources for Exynos5440 TMU sensor configuration
+ *
+ * Copyright (c) 2014 Lukasz Majewski <l.majewski@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <dt-bindings/thermal/thermal_exynos.h>
+#thermal-sensor-cells = <0>;
+samsung,tmu_gain = <5>;
+samsung,tmu_reference_voltage = <16>;
+samsung,tmu_noise_cancel_mode = <4>;
+samsung,tmu_efuse_value = <0x5d2d>;
+samsung,tmu_min_efuse_value = <16>;
+samsung,tmu_max_efuse_value = <76>;
+samsung,tmu_first_point_trim = <25>;
+samsung,tmu_second_point_trim = <70>;
+samsung,tmu_default_temp_offset = <25>;
+samsung,tmu_cal_type = <TYPE_ONE_POINT_TRIMMING>;
diff --git a/arch/arm/boot/dts/exynos5440-trip-points.dtsi b/arch/arm/boot/dts/exynos5440-trip-points.dtsi
new file mode 100644
index 000000000000..48adfa8f4300
--- /dev/null
+++ b/arch/arm/boot/dts/exynos5440-trip-points.dtsi
@@ -0,0 +1,25 @@
+/*
+ * Device tree sources for default Exynos5440 thermal zone definition
+ *
+ * Copyright (c) 2014 Lukasz Majewski <l.majewski@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+polling-delay-passive = <0>;
+polling-delay = <0>;
+trips {
+        cpu-alert-0 {
+                temperature = <100000>; /* millicelsius */
+                hysteresis = <0>; /* millicelsius */
+                type = "active";
+        };
+        cpu-crit-0 {
+                temperature = <1050000>; /* millicelsius */
+                hysteresis = <0>; /* millicelsius */
+                type = "critical";
+        };
+};
diff --git a/arch/arm/boot/dts/exynos5440.dtsi b/arch/arm/boot/dts/exynos5440.dtsi
index 8f3373cd7b87..59d9416b3b03 100644
--- a/arch/arm/boot/dts/exynos5440.dtsi
+++ b/arch/arm/boot/dts/exynos5440.dtsi
@@ -219,6 +219,7 @@
                interrupts = <0 58 0>;
                clocks = <&clock CLK_B_125>;
                clock-names = "tmu_apbif";
+                #include "exynos5440-tmu-sensor-conf.dtsi"
        };
        tmuctrl_1: tmuctrl@16011C {
@@ -227,6 +228,7 @@
                interrupts = <0 58 0>;
                clocks = <&clock CLK_B_125>;
                clock-names = "tmu_apbif";
+                #include "exynos5440-tmu-sensor-conf.dtsi"
        };
        tmuctrl_2: tmuctrl@160120 {
@@ -235,6 +237,22 @@
                interrupts = <0 58 0>;
                clocks = <&clock CLK_B_125>;
                clock-names = "tmu_apbif";
+                #include "exynos5440-tmu-sensor-conf.dtsi"
+        };
+        thermal-zones {
+                cpu0_thermal: cpu0-thermal {
+                        thermal-sensors = <&tmuctrl_0>;
+                        #include "exynos5440-trip-points.dtsi"
+                };
+                cpu1_thermal: cpu1-thermal {
+                       thermal-sensors = <&tmuctrl_1>;
+                       #include "exynos5440-trip-points.dtsi"
+                };
+                cpu2_thermal: cpu2-thermal {
+                       thermal-sensors = <&tmuctrl_2>;
+                       #include "exynos5440-trip-points.dtsi"
+                };
        };
        sata@210000 {
diff --git a/arch/arm/boot/dts/imx6qdl-sabresd.dtsi b/arch/arm/boot/dts/imx6qdl-sabresd.dtsi
index f1cd2147421d..a626e6dd8022 100644
--- a/arch/arm/boot/dts/imx6qdl-sabresd.dtsi
+++ b/arch/arm/boot/dts/imx6qdl-sabresd.dtsi
@@ -35,6 +35,7 @@
                        regulator-max-microvolt = <5000000>;
                        gpio = <&gpio3 22 0>;
                        enable-active-high;
+                        vin-supply = <&swbst_reg>;
                };
                reg_usb_h1_vbus: regulator@1 {
@@ -45,6 +46,7 @@
                        regulator-max-microvolt = <5000000>;
                        gpio = <&gpio1 29 0>;
                        enable-active-high;
+                        vin-supply = <&swbst_reg>;
                };
                reg_audio: regulator@2 {
diff --git a/arch/arm/boot/dts/imx6sl-evk.dts b/arch/arm/boot/dts/imx6sl-evk.dts
index fda4932faefd..945887d3fdb3 100644
--- a/arch/arm/boot/dts/imx6sl-evk.dts
+++ b/arch/arm/boot/dts/imx6sl-evk.dts
@@ -52,6 +52,7 @@
                        regulator-max-microvolt = <5000000>;
                        gpio = <&gpio4 0 0>;
                        enable-active-high;
+                        vin-supply = <&swbst_reg>;
                };
                reg_usb_otg2_vbus: regulator@1 {
@@ -62,6 +63,7 @@
                        regulator-max-microvolt = <5000000>;
                        gpio = <&gpio4 2 0>;
                        enable-active-high;
+                        vin-supply = <&swbst_reg>;
                };
                reg_aud3v: regulator@2 {
diff --git a/arch/arm/boot/dts/omap5-core-thermal.dtsi b/arch/arm/boot/dts/omap5-core-thermal.dtsi
index 19212ac6eef0..de8a3d456cf7 100644
--- a/arch/arm/boot/dts/omap5-core-thermal.dtsi
+++ b/arch/arm/boot/dts/omap5-core-thermal.dtsi
@@ -13,7 +13,7 @@
 core_thermal: core_thermal {
        polling-delay-passive = <250>; /* milliseconds */
-        polling-delay = <1000>; /* milliseconds */
+        polling-delay = <500>; /* milliseconds */
                        /* sensor       ID */
        thermal-sensors = <&bandgap     2>;
diff --git a/arch/arm/boot/dts/omap5-gpu-thermal.dtsi b/arch/arm/boot/dts/omap5-gpu-thermal.dtsi
index 1b87aca88b77..bc3090f2e84b 100644
--- a/arch/arm/boot/dts/omap5-gpu-thermal.dtsi
+++ b/arch/arm/boot/dts/omap5-gpu-thermal.dtsi
@@ -13,7 +13,7 @@
 gpu_thermal: gpu_thermal {
        polling-delay-passive = <250>; /* milliseconds */
-        polling-delay = <1000>; /* milliseconds */
+        polling-delay = <500>; /* milliseconds */
                        /* sensor       ID */
        thermal-sensors = <&bandgap     1>;
diff --git a/arch/arm/boot/dts/omap5.dtsi b/arch/arm/boot/dts/omap5.dtsi
index ddff674bd05e..4a485b63a141 100644
--- a/arch/arm/boot/dts/omap5.dtsi
+++ b/arch/arm/boot/dts/omap5.dtsi
@@ -1079,4 +1079,8 @@
        };
 };
+&cpu_thermal {
+        polling-delay = <500>; /* milliseconds */
+};
 /include/ "omap54xx-clocks.dtsi"
diff --git a/arch/arm/boot/dts/omap54xx-clocks.dtsi b/arch/arm/boot/dts/omap54xx-clocks.dtsi
index 58c27466f012..83b425fb3ac2 100644
--- a/arch/arm/boot/dts/omap54xx-clocks.dtsi
+++ b/arch/arm/boot/dts/omap54xx-clocks.dtsi
@@ -167,10 +167,18 @@
                ti,index-starts-at-one;
        };
+        dpll_core_byp_mux: dpll_core_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin>, <&dpll_abe_m3x2_ck>;
+                ti,bit-shift = <23>;
+                reg = <0x012c>;
+        };
        dpll_core_ck: dpll_core_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-core-clock";
-                clocks = <&sys_clkin>, <&dpll_abe_m3x2_ck>;
+                clocks = <&sys_clkin>, <&dpll_core_byp_mux>;
                reg = <0x0120>, <0x0124>, <0x012c>, <0x0128>;
        };
@@ -294,10 +302,18 @@
                clock-div = <1>;
        };
+        dpll_iva_byp_mux: dpll_iva_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin>, <&iva_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x01ac>;
+        };
        dpll_iva_ck: dpll_iva_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin>, <&iva_dpll_hs_clk_div>;
+                clocks = <&sys_clkin>, <&dpll_iva_byp_mux>;
                reg = <0x01a0>, <0x01a4>, <0x01ac>, <0x01a8>;
        };
@@ -599,10 +615,19 @@
        };
 };
 &cm_core_clocks {
+        dpll_per_byp_mux: dpll_per_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin>, <&per_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x014c>;
+        };
        dpll_per_ck: dpll_per_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-clock";
-                clocks = <&sys_clkin>, <&per_dpll_hs_clk_div>;
+                clocks = <&sys_clkin>, <&dpll_per_byp_mux>;
                reg = <0x0140>, <0x0144>, <0x014c>, <0x0148>;
        };
@@ -714,10 +739,18 @@
                ti,index-starts-at-one;
        };
+        dpll_usb_byp_mux: dpll_usb_byp_mux {
+                #clock-cells = <0>;
+                compatible = "ti,mux-clock";
+                clocks = <&sys_clkin>, <&usb_dpll_hs_clk_div>;
+                ti,bit-shift = <23>;
+                reg = <0x018c>;
+        };
        dpll_usb_ck: dpll_usb_ck {
                #clock-cells = <0>;
                compatible = "ti,omap4-dpll-j-type-clock";
-                clocks = <&sys_clkin>, <&usb_dpll_hs_clk_div>;
+                clocks = <&sys_clkin>, <&dpll_usb_byp_mux>;
                reg = <0x0180>, <0x0184>, <0x018c>, <0x0188>;
        };
diff --git a/arch/arm/boot/dts/sama5d3.dtsi b/arch/arm/boot/dts/sama5d3.dtsi
index 261311bdf65b..367af53c1b84 100644
--- a/arch/arm/boot/dts/sama5d3.dtsi
+++ b/arch/arm/boot/dts/sama5d3.dtsi
@@ -1248,7 +1248,6 @@
                                atmel,watchdog-type = "hardware";
                                atmel,reset-type = "all";
                                atmel,dbg-halt;
-                                atmel,idle-halt;
                                status = "disabled";
                        };
@@ -1416,7 +1415,7 @@
                        compatible = "atmel,at91sam9g45-ehci", "usb-ehci";
                        reg = <0x00700000 0x100000>;
                        interrupts = <32 IRQ_TYPE_LEVEL_HIGH 2>;
-                        clocks = <&usb>, <&uhphs_clk>, <&uhpck>;
+                        clocks = <&utmi>, <&uhphs_clk>, <&uhpck>;
                        clock-names = "usb_clk", "ehci_clk", "uhpck";
                        status = "disabled";
                };
diff --git a/arch/arm/boot/dts/sama5d4.dtsi b/arch/arm/boot/dts/sama5d4.dtsi
index d986b41b9654..4303874889c6 100644
--- a/arch/arm/boot/dts/sama5d4.dtsi
+++ b/arch/arm/boot/dts/sama5d4.dtsi
@@ -66,6 +66,7 @@
                gpio4 = &pioE;
                tcb0 = &tcb0;
                tcb1 = &tcb1;
+                i2c0 = &i2c0;
                i2c2 = &i2c2;
        };
        cpus {
@@ -259,7 +260,7 @@
                        compatible = "atmel,at91sam9g45-ehci", "usb-ehci";
                        reg = <0x00600000 0x100000>;
                        interrupts = <46 IRQ_TYPE_LEVEL_HIGH 2>;
-                        clocks = <&usb>, <&uhphs_clk>, <&uhpck>;
+                        clocks = <&utmi>, <&uhphs_clk>, <&uhpck>;
                        clock-names = "usb_clk", "ehci_clk", "uhpck";
                        status = "disabled";
                };
@@ -461,8 +462,8 @@
                                        lcdck: lcdck {
                                                #clock-cells = <0>;
-                                                reg = <4>;
+                                                reg = <3>;
-                                                clocks = <&smd>;
+                                                clocks = <&mck>;
                                        };
                                        smdck: smdck {
@@ -770,7 +771,7 @@
                                                reg = <50>;
                                        };
-                                        lcd_clk: lcd_clk {
+                                        lcdc_clk: lcdc_clk {
                                                #clock-cells = <0>;
                                                reg = <51>;
                                        };
diff --git a/arch/arm/boot/dts/socfpga.dtsi b/arch/arm/boot/dts/socfpga.dtsi
index 252c3d1bda50..9d8760956752 100644
--- a/arch/arm/boot/dts/socfpga.dtsi
+++ b/arch/arm/boot/dts/socfpga.dtsi
@@ -713,6 +713,9 @@
                        reg-shift = <2>;
                        reg-io-width = <4>;
                        clocks = <&l4_sp_clk>;
+                        dmas = <&pdma 28>,
+                               <&pdma 29>;
+                        dma-names = "tx", "rx";
                };
                uart1: serial1@ffc03000 {
@@ -722,6 +725,9 @@
                        reg-shift = <2>;
                        reg-io-width = <4>;
                        clocks = <&l4_sp_clk>;
+                        dmas = <&pdma 30>,
+                               <&pdma 31>;
+                        dma-names = "tx", "rx";
                };
                rst: rstmgr@ffd05000 {
diff --git a/arch/arm/configs/at91_dt_defconfig b/arch/arm/configs/at91_dt_defconfig
index f2670f638e97..811e72bbe642 100644
--- a/arch/arm/configs/at91_dt_defconfig
+++ b/arch/arm/configs/at91_dt_defconfig
@@ -70,6 +70,7 @@ CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 # CONFIG_SCSI_LOWLEVEL is not set
 CONFIG_NETDEVICES=y
+CONFIG_ARM_AT91_ETHER=y
 CONFIG_MACB=y
 # CONFIG_NET_VENDOR_BROADCOM is not set
 CONFIG_DM9000=y
diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
index b7e6b6fba5e0..06075b6d2463 100644
--- a/arch/arm/configs/multi_v7_defconfig
+++ b/arch/arm/configs/multi_v7_defconfig
@@ -99,7 +99,7 @@ CONFIG_PCI_RCAR_GEN2=y
 CONFIG_PCI_RCAR_GEN2_PCIE=y
 CONFIG_PCIEPORTBUS=y
 CONFIG_SMP=y
-CONFIG_NR_CPUS=8
+CONFIG_NR_CPUS=16
 CONFIG_HIGHPTE=y
 CONFIG_CMA=y
 CONFIG_ARM_APPENDED_DTB=y
diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
index a097cffa1231..8e108599e1af 100644
--- a/arch/arm/configs/omap2plus_defconfig
+++ b/arch/arm/configs/omap2plus_defconfig
@@ -377,6 +377,7 @@ CONFIG_PWM_TWL=m
 CONFIG_PWM_TWL_LED=m
 CONFIG_OMAP_USB2=m
 CONFIG_TI_PIPE3=y
+CONFIG_TWL4030_USB=m
 CONFIG_EXT2_FS=y
 CONFIG_EXT3_FS=y
 # CONFIG_EXT3_FS_XATTR is not set
diff --git a/arch/arm/configs/sama5_defconfig b/arch/arm/configs/sama5_defconfig
index 41d856effe6c..510c747c65b4 100644
--- a/arch/arm/configs/sama5_defconfig
+++ b/arch/arm/configs/sama5_defconfig
@@ -3,8 +3,6 @@
 CONFIG_SYSVIPC=y
 CONFIG_IRQ_DOMAIN_DEBUG=y
 CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED=y
-CONFIG_SYSFS_DEPRECATED_V2=y
 CONFIG_BLK_DEV_INITRD=y
 CONFIG_EMBEDDED=y
 CONFIG_SLAB=y
diff --git a/arch/arm/configs/sunxi_defconfig b/arch/arm/configs/sunxi_defconfig
index 38840a812924..8f6a5702b696 100644
--- a/arch/arm/configs/sunxi_defconfig
+++ b/arch/arm/configs/sunxi_defconfig
@@ -4,6 +4,7 @@ CONFIG_BLK_DEV_INITRD=y
 CONFIG_PERF_EVENTS=y
 CONFIG_ARCH_SUNXI=y
 CONFIG_SMP=y
+CONFIG_NR_CPUS=8
 CONFIG_AEABI=y
 CONFIG_HIGHMEM=y
 CONFIG_HIGHPTE=y
diff --git a/arch/arm/configs/vexpress_defconfig b/arch/arm/configs/vexpress_defconfig
index f489fdaa19b8..37fe607a4ede 100644
--- a/arch/arm/configs/vexpress_defconfig
+++ b/arch/arm/configs/vexpress_defconfig
@@ -118,8 +118,8 @@ CONFIG_HID_ZEROPLUS=y
 CONFIG_USB=y
 CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
 CONFIG_USB_MON=y
-CONFIG_USB_ISP1760_HCD=y
 CONFIG_USB_STORAGE=y
+CONFIG_USB_ISP1760=y
 CONFIG_MMC=y
 CONFIG_MMC_ARMMMCI=y
 CONFIG_NEW_LEDS=y
diff --git a/arch/arm/crypto/aesbs-core.S_shipped b/arch/arm/crypto/aesbs-core.S_shipped
index 71e5fc7cfb18..1d1800f71c5b 100644
--- a/arch/arm/crypto/aesbs-core.S_shipped
+++ b/arch/arm/crypto/aesbs-core.S_shipped
@@ -58,14 +58,18 @@
 # define VFP_ABI_FRAME  0
 # define BSAES_ASM_EXTENDED_KEY
 # define XTS_CHAIN_TWEAK
-# define __ARM_ARCH__   7
+# define __ARM_ARCH__ __LINUX_ARM_ARCH__
+# define __ARM_MAX_ARCH__ 7
 #endif
 #ifdef __thumb__
 # define adrl adr
 #endif
-#if __ARM_ARCH__>=7
+#if __ARM_MAX_ARCH__>=7
+.arch   armv7-a
+.fpu    neon
 .text
 .syntax unified         @ ARMv7-capable assembler is expected to handle this
 #ifdef __thumb2__
@@ -74,8 +78,6 @@
 .code   32
 #endif
-.fpu    neon
 .type   _bsaes_decrypt8,%function
 .align  4
 _bsaes_decrypt8:
@@ -2095,9 +2097,11 @@ bsaes_xts_decrypt:
        vld1.8  {q8}, [r0]                      @ initial tweak
        adr     r2, .Lxts_magic
+#ifndef XTS_CHAIN_TWEAK
        tst     r9, #0xf                        @ if not multiple of 16
        it      ne                              @ Thumb2 thing, sanity check in ARM
        subne   r9, #0x10                       @ subtract another 16 bytes
+#endif
        subs    r9, #0x80
        blo     .Lxts_dec_short
diff --git a/arch/arm/crypto/bsaes-armv7.pl b/arch/arm/crypto/bsaes-armv7.pl
index be068db960ee..a4d3856e7d24 100644
--- a/arch/arm/crypto/bsaes-armv7.pl
+++ b/arch/arm/crypto/bsaes-armv7.pl
@@ -701,14 +701,18 @@ $code.=<<___;
 # define VFP_ABI_FRAME  0
 # define BSAES_ASM_EXTENDED_KEY
 # define XTS_CHAIN_TWEAK
-# define __ARM_ARCH__   7
+# define __ARM_ARCH__ __LINUX_ARM_ARCH__
+# define __ARM_MAX_ARCH__ 7
 #endif
 #ifdef __thumb__
 # define adrl adr
 #endif
-#if __ARM_ARCH__>=7
+#if __ARM_MAX_ARCH__>=7
+.arch   armv7-a
+.fpu    neon
 .text
 .syntax unified         @ ARMv7-capable assembler is expected to handle this
 #ifdef __thumb2__
@@ -717,8 +721,6 @@ $code.=<<___;
 .code   32
 #endif
-.fpu    neon
 .type   _bsaes_decrypt8,%function
 .align  4
 _bsaes_decrypt8:
@@ -2076,9 +2078,11 @@ bsaes_xts_decrypt:
        vld1.8  {@XMM[8]}, [r0]                 @ initial tweak
        adr     $magic, .Lxts_magic
+#ifndef XTS_CHAIN_TWEAK
        tst     $len, #0xf                      @ if not multiple of 16
        it      ne                              @ Thumb2 thing, sanity check in ARM
        subne   $len, #0x10                     @ subtract another 16 bytes
+#endif
        subs    $len, #0x80
        blo     .Lxts_dec_short
diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
index 37ca2a4c6f09..4cf48c3aca13 100644
--- a/arch/arm/include/asm/kvm_mmu.h
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -149,29 +149,28 @@ static inline bool kvm_s2pmd_readonly(pmd_t *pmd)
        (__boundary - 1 < (end) - 1)? __boundary: (end);                \
 })
+#define kvm_pgd_index(addr)                     pgd_index(addr)
 static inline bool kvm_page_empty(void *ptr)
 {
        struct page *ptr_page = virt_to_page(ptr);
        return page_count(ptr_page) == 1;
 }
 #define kvm_pte_table_empty(kvm, ptep) kvm_page_empty(ptep)
 #define kvm_pmd_table_empty(kvm, pmdp) kvm_page_empty(pmdp)
 #define kvm_pud_table_empty(kvm, pudp) (0)
 #define KVM_PREALLOC_LEVEL      0
-static inline int kvm_prealloc_hwpgd(struct kvm *kvm, pgd_t *pgd)
+static inline void *kvm_get_hwpgd(struct kvm *kvm)
 {
-        return 0;
+        return kvm->arch.pgd;
 }
-static inline void kvm_free_hwpgd(struct kvm *kvm) { }
+static inline unsigned int kvm_get_hwpgd_size(void)
-static inline void *kvm_get_hwpgd(struct kvm *kvm)
 {
-        return kvm->arch.pgd;
+        return PTRS_PER_S2_PGD * sizeof(pgd_t);
 }
 struct kvm;
@@ -207,7 +206,7 @@ static inline void __coherent_cache_guest_page(struct kvm_vcpu *vcpu, pfn_t pfn,
        bool need_flush = !vcpu_has_cache_enabled(vcpu) || ipa_uncached;
-        VM_BUG_ON(size & PAGE_MASK);
+        VM_BUG_ON(size & ~PAGE_MASK);
        if (!need_flush && !icache_is_pipt())
                goto vipt_cache;
diff --git a/arch/arm/include/debug/at91.S b/arch/arm/include/debug/at91.S
index 80a6501b4d50..c3c45e628e33 100644
--- a/arch/arm/include/debug/at91.S
+++ b/arch/arm/include/debug/at91.S
@@ -18,8 +18,11 @@
 #define AT91_DBGU 0xfc00c000 /* SAMA5D4_BASE_USART3 */
 #endif
-/* Keep in sync with mach-at91/include/mach/hardware.h */
+#ifdef CONFIG_MMU
 #define AT91_IO_P2V(x) ((x) - 0x01000000)
+#else
+#define AT91_IO_P2V(x) (x)
+#endif
 #define AT91_DBGU_SR            (0x14)  /* Status Register */
 #define AT91_DBGU_THR           (0x1c)  /* Transmitter Holding Register */
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index e55408e96559..1d60bebea4b8 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -246,12 +246,9 @@ static int __get_cpu_architecture(void)
                if (cpu_arch)
                        cpu_arch += CPU_ARCH_ARMv3;
        } else if ((read_cpuid_id() & 0x000f0000) == 0x000f0000) {
-                unsigned int mmfr0;
                /* Revised CPUID format. Read the Memory Model Feature
                 * Register 0 and check for VMSAv7 or PMSAv7 */
-                asm("mrc        p15, 0, %0, c0, c1, 4"
+                unsigned int mmfr0 = read_cpuid_ext(CPUID_EXT_MMFR0);
-                    : "=r" (mmfr0));
                if ((mmfr0 & 0x0000000f) >= 0x00000003 ||
                    (mmfr0 & 0x000000f0) >= 0x00000030)
                        cpu_arch = CPU_ARCH_ARMv7;
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index 07e7eb1d7ab6..5560f74f9eee 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -540,7 +540,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
                vcpu->mode = OUTSIDE_GUEST_MODE;
                kvm_guest_exit();
-                trace_kvm_exit(*vcpu_pc(vcpu));
+                trace_kvm_exit(kvm_vcpu_trap_get_class(vcpu), *vcpu_pc(vcpu));
                /*
                 * We may have taken a host interrupt in HYP mode (ie
                 * while executing the guest). This interrupt is still
diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c
index 3e6859bc3e11..5656d79c5a44 100644
--- a/arch/arm/kvm/mmu.c
+++ b/arch/arm/kvm/mmu.c
@@ -290,7 +290,7 @@ static void unmap_range(struct kvm *kvm, pgd_t *pgdp,
        phys_addr_t addr = start, end = start + size;
        phys_addr_t next;
-        pgd = pgdp + pgd_index(addr);
+        pgd = pgdp + kvm_pgd_index(addr);
        do {
                next = kvm_pgd_addr_end(addr, end);
                if (!pgd_none(*pgd))
@@ -355,7 +355,7 @@ static void stage2_flush_memslot(struct kvm *kvm,
        phys_addr_t next;
        pgd_t *pgd;
-        pgd = kvm->arch.pgd + pgd_index(addr);
+        pgd = kvm->arch.pgd + kvm_pgd_index(addr);
        do {
                next = kvm_pgd_addr_end(addr, end);
                stage2_flush_puds(kvm, pgd, addr, next);
@@ -632,6 +632,20 @@ int create_hyp_io_mappings(void *from, void *to, phys_addr_t phys_addr)
                                     __phys_to_pfn(phys_addr), PAGE_HYP_DEVICE);
 }
+/* Free the HW pgd, one page at a time */
+static void kvm_free_hwpgd(void *hwpgd)
+{
+        free_pages_exact(hwpgd, kvm_get_hwpgd_size());
+}
+/* Allocate the HW PGD, making sure that each page gets its own refcount */
+static void *kvm_alloc_hwpgd(void)
+{
+        unsigned int size = kvm_get_hwpgd_size();
+        return alloc_pages_exact(size, GFP_KERNEL | __GFP_ZERO);
+}
 /**
 * kvm_alloc_stage2_pgd - allocate level-1 table for stage-2 translation.
 * @kvm:        The KVM struct pointer for the VM.
@@ -645,15 +659,31 @@ int create_hyp_io_mappings(void *from, void *to, phys_addr_t phys_addr)
 */
 int kvm_alloc_stage2_pgd(struct kvm *kvm)
 {
-        int ret;
        pgd_t *pgd;
+        void *hwpgd;
        if (kvm->arch.pgd != NULL) {
                kvm_err("kvm_arch already initialized?\n");
                return -EINVAL;
        }
+        hwpgd = kvm_alloc_hwpgd();
+        if (!hwpgd)
+                return -ENOMEM;
+        /* When the kernel uses more levels of page tables than the
+         * guest, we allocate a fake PGD and pre-populate it to point
+         * to the next-level page table, which will be the real
+         * initial page table pointed to by the VTTBR.
+         *
+         * When KVM_PREALLOC_LEVEL==2, we allocate a single page for
+         * the PMD and the kernel will use folded pud.
+         * When KVM_PREALLOC_LEVEL==1, we allocate 2 consecutive PUD
+         * pages.
+         */
        if (KVM_PREALLOC_LEVEL > 0) {
+                int i;
                /*
                 * Allocate fake pgd for the page table manipulation macros to
                 * work.  This is not used by the hardware and we have no
@@ -661,30 +691,32 @@ int kvm_alloc_stage2_pgd(struct kvm *kvm)
                 */
                pgd = (pgd_t *)kmalloc(PTRS_PER_S2_PGD * sizeof(pgd_t),
                                       GFP_KERNEL | __GFP_ZERO);
+                if (!pgd) {
+                        kvm_free_hwpgd(hwpgd);
+                        return -ENOMEM;
+                }
+                /* Plug the HW PGD into the fake one. */
+                for (i = 0; i < PTRS_PER_S2_PGD; i++) {
+                        if (KVM_PREALLOC_LEVEL == 1)
+                                pgd_populate(NULL, pgd + i,
+                                             (pud_t *)hwpgd + i * PTRS_PER_PUD);
+                        else if (KVM_PREALLOC_LEVEL == 2)
+                                pud_populate(NULL, pud_offset(pgd, 0) + i,
+                                             (pmd_t *)hwpgd + i * PTRS_PER_PMD);
+                }
        } else {
                /*
                 * Allocate actual first-level Stage-2 page table used by the
                 * hardware for Stage-2 page table walks.
                 */
-                pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, S2_PGD_ORDER);
+                pgd = (pgd_t *)hwpgd;
        }
-        if (!pgd)
-                return -ENOMEM;
-        ret = kvm_prealloc_hwpgd(kvm, pgd);
-        if (ret)
-                goto out_err;
        kvm_clean_pgd(pgd);
        kvm->arch.pgd = pgd;
        return 0;
-out_err:
-        if (KVM_PREALLOC_LEVEL > 0)
-                kfree(pgd);
-        else
-                free_pages((unsigned long)pgd, S2_PGD_ORDER);
-        return ret;
 }
 /**
@@ -785,11 +817,10 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
                return;
        unmap_stage2_range(kvm, 0, KVM_PHYS_SIZE);
-        kvm_free_hwpgd(kvm);
+        kvm_free_hwpgd(kvm_get_hwpgd(kvm));
        if (KVM_PREALLOC_LEVEL > 0)
                kfree(kvm->arch.pgd);
-        else
-                free_pages((unsigned long)kvm->arch.pgd, S2_PGD_ORDER);
        kvm->arch.pgd = NULL;
 }
@@ -799,7 +830,7 @@ static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache
        pgd_t *pgd;
        pud_t *pud;
-        pgd = kvm->arch.pgd + pgd_index(addr);
+        pgd = kvm->arch.pgd + kvm_pgd_index(addr);
        if (WARN_ON(pgd_none(*pgd))) {
                if (!cache)
                        return NULL;
@@ -1089,7 +1120,7 @@ static void stage2_wp_range(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
        pgd_t *pgd;
        phys_addr_t next;
-        pgd = kvm->arch.pgd + pgd_index(addr);
+        pgd = kvm->arch.pgd + kvm_pgd_index(addr);
        do {
                /*
                 * Release kvm_mmu_lock periodically if the memory region is
diff --git a/arch/arm/kvm/trace.h b/arch/arm/kvm/trace.h
index 881874b1a036..6817664b46b8 100644
--- a/arch/arm/kvm/trace.h
+++ b/arch/arm/kvm/trace.h
@@ -25,18 +25,22 @@ TRACE_EVENT(kvm_entry,
 );
 TRACE_EVENT(kvm_exit,
-        TP_PROTO(unsigned long vcpu_pc),
+        TP_PROTO(unsigned int exit_reason, unsigned long vcpu_pc),
-        TP_ARGS(vcpu_pc),
+        TP_ARGS(exit_reason, vcpu_pc),
        TP_STRUCT__entry(
+                __field(        unsigned int,   exit_reason     )
                __field(        unsigned long,  vcpu_pc         )
        ),
        TP_fast_assign(
+                __entry->exit_reason            = exit_reason;
                __entry->vcpu_pc                = vcpu_pc;
        ),
-        TP_printk("PC: 0x%08lx", __entry->vcpu_pc)
+        TP_printk("HSR_EC: 0x%04x, PC: 0x%08lx",
+                  __entry->exit_reason,
+                  __entry->vcpu_pc)
 );
 TRACE_EVENT(kvm_guest_fault,
diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c
index 5e34fb143309..aa4116e9452f 100644
--- a/arch/arm/mach-at91/pm.c
+++ b/arch/arm/mach-at91/pm.c
@@ -270,37 +270,35 @@ static void __init at91_pm_sram_init(void)
        phys_addr_t sram_pbase;
        unsigned long sram_base;
        struct device_node *node;
-        struct platform_device *pdev;
+        struct platform_device *pdev = NULL;
-        node = of_find_compatible_node(NULL, NULL, "mmio-sram");
+        for_each_compatible_node(node, NULL, "mmio-sram") {
-        if (!node) {
+                pdev = of_find_device_by_node(node);
-                pr_warn("%s: failed to find sram node!\n", __func__);
+                if (pdev) {
-                return;
+                        of_node_put(node);
+                        break;
+                }
        }
-        pdev = of_find_device_by_node(node);
        if (!pdev) {
                pr_warn("%s: failed to find sram device!\n", __func__);
-                goto put_node;
+                return;
        }
        sram_pool = dev_get_gen_pool(&pdev->dev);
        if (!sram_pool) {
                pr_warn("%s: sram pool unavailable!\n", __func__);
-                goto put_node;
+                return;
        }
        sram_base = gen_pool_alloc(sram_pool, at91_slow_clock_sz);
        if (!sram_base) {
                pr_warn("%s: unable to alloc ocram!\n", __func__);
-                goto put_node;
+                return;
        }
        sram_pbase = gen_pool_virt_to_phys(sram_pool, sram_base);
        slow_clock = __arm_ioremap_exec(sram_pbase, at91_slow_clock_sz, false);
-put_node:
-        of_node_put(node);
 }
 #endif
diff --git a/arch/arm/mach-at91/pm.h b/arch/arm/mach-at91/pm.h
index d2c89963af2d..86c0aa819d25 100644
--- a/arch/arm/mach-at91/pm.h
+++ b/arch/arm/mach-at91/pm.h
@@ -44,7 +44,7 @@ static inline void at91rm9200_standby(void)
                "    mcr    p15, 0, %0, c7, c0, 4\n\t"
                "    str    %5, [%1, %2]"
                :
-                : "r" (0), "r" (AT91_BASE_SYS), "r" (AT91RM9200_SDRAMC_LPR),
+                : "r" (0), "r" (at91_ramc_base[0]), "r" (AT91RM9200_SDRAMC_LPR),
                  "r" (1), "r" (AT91RM9200_SDRAMC_SRR),
                  "r" (lpr));
 }
diff --git a/arch/arm/mach-at91/pm_slowclock.S b/arch/arm/mach-at91/pm_slowclock.S
index 556151e85ec4..931f0e302c03 100644
--- a/arch/arm/mach-at91/pm_slowclock.S
+++ b/arch/arm/mach-at91/pm_slowclock.S
@@ -25,11 +25,6 @@
 */
 #undef SLOWDOWN_MASTER_CLOCK
-#define MCKRDY_TIMEOUT          1000
-#define MOSCRDY_TIMEOUT         1000
-#define PLLALOCK_TIMEOUT        1000
-#define PLLBLOCK_TIMEOUT        1000
 pmc     .req    r0
 sdramc  .req    r1
 ramc1   .req    r2
@@ -41,60 +36,42 @@ tmp2	.req	r5
 * Wait until master clock is ready (after switching master clock source)
 */
        .macro wait_mckrdy
-        mov     tmp2, #MCKRDY_TIMEOUT
+1:      ldr     tmp1, [pmc, #AT91_PMC_SR]
-1:      sub     tmp2, tmp2, #1
-        cmp     tmp2, #0
-        beq     2f
-        ldr     tmp1, [pmc, #AT91_PMC_SR]
        tst     tmp1, #AT91_PMC_MCKRDY
        beq     1b
-2:
        .endm
 /*
 * Wait until master oscillator has stabilized.
 */
        .macro wait_moscrdy
-        mov     tmp2, #MOSCRDY_TIMEOUT
+1:      ldr     tmp1, [pmc, #AT91_PMC_SR]
-1:      sub     tmp2, tmp2, #1
-        cmp     tmp2, #0
-        beq     2f
-        ldr     tmp1, [pmc, #AT91_PMC_SR]
        tst     tmp1, #AT91_PMC_MOSCS
        beq     1b
-2:
        .endm
 /*
 * Wait until PLLA has locked.
 */
        .macro wait_pllalock
-        mov     tmp2, #PLLALOCK_TIMEOUT
+1:      ldr     tmp1, [pmc, #AT91_PMC_SR]
-1:      sub     tmp2, tmp2, #1
-        cmp     tmp2, #0
-        beq     2f
-        ldr     tmp1, [pmc, #AT91_PMC_SR]
        tst     tmp1, #AT91_PMC_LOCKA
        beq     1b
-2:
        .endm
 /*
 * Wait until PLLB has locked.
 */
        .macro wait_pllblock
-        mov     tmp2, #PLLBLOCK_TIMEOUT
+1:      ldr     tmp1, [pmc, #AT91_PMC_SR]
-1:      sub     tmp2, tmp2, #1
-        cmp     tmp2, #0
-        beq     2f
-        ldr     tmp1, [pmc, #AT91_PMC_SR]
        tst     tmp1, #AT91_PMC_LOCKB
        beq     1b
-2:
        .endm
        .text
+        .arm
 /* void at91_slow_clock(void __iomem *pmc, void __iomem *sdramc,
 *                      void __iomem *ramc1, int memctrl)
 */
@@ -134,6 +111,16 @@ ddr_sr_enable:
        cmp     memctrl, #AT91_MEMCTRL_DDRSDR
        bne     sdr_sr_enable
+        /* LPDDR1 --> force DDR2 mode during self-refresh */
+        ldr     tmp1, [sdramc, #AT91_DDRSDRC_MDR]
+        str     tmp1, .saved_sam9_mdr
+        bic     tmp1, tmp1, #~AT91_DDRSDRC_MD
+        cmp     tmp1, #AT91_DDRSDRC_MD_LOW_POWER_DDR
+        ldreq   tmp1, [sdramc, #AT91_DDRSDRC_MDR]
+        biceq   tmp1, tmp1, #AT91_DDRSDRC_MD
+        orreq   tmp1, tmp1, #AT91_DDRSDRC_MD_DDR2
+        streq   tmp1, [sdramc, #AT91_DDRSDRC_MDR]
        /* prepare for DDRAM self-refresh mode */
        ldr     tmp1, [sdramc, #AT91_DDRSDRC_LPR]
        str     tmp1, .saved_sam9_lpr
@@ -142,14 +129,26 @@ ddr_sr_enable:
        /* figure out if we use the second ram controller */
        cmp     ramc1, #0
-        ldrne   tmp2, [ramc1, #AT91_DDRSDRC_LPR]
+        beq     ddr_no_2nd_ctrl
-        strne   tmp2, .saved_sam9_lpr1
-        bicne   tmp2, #AT91_DDRSDRC_LPCB
+        ldr     tmp2, [ramc1, #AT91_DDRSDRC_MDR]
-        orrne   tmp2, #AT91_DDRSDRC_LPCB_SELF_REFRESH
+        str     tmp2, .saved_sam9_mdr1
+        bic     tmp2, tmp2, #~AT91_DDRSDRC_MD
+        cmp     tmp2, #AT91_DDRSDRC_MD_LOW_POWER_DDR
+        ldreq   tmp2, [ramc1, #AT91_DDRSDRC_MDR]
+        biceq   tmp2, tmp2, #AT91_DDRSDRC_MD
+        orreq   tmp2, tmp2, #AT91_DDRSDRC_MD_DDR2
+        streq   tmp2, [ramc1, #AT91_DDRSDRC_MDR]
+        ldr     tmp2, [ramc1, #AT91_DDRSDRC_LPR]
+        str     tmp2, .saved_sam9_lpr1
+        bic     tmp2, #AT91_DDRSDRC_LPCB
+        orr     tmp2, #AT91_DDRSDRC_LPCB_SELF_REFRESH
        /* Enable DDRAM self-refresh mode */
+        str     tmp2, [ramc1, #AT91_DDRSDRC_LPR]
+ddr_no_2nd_ctrl:
        str     tmp1, [sdramc, #AT91_DDRSDRC_LPR]
-        strne   tmp2, [ramc1, #AT91_DDRSDRC_LPR]
        b       sdr_sr_done
@@ -208,6 +207,7 @@ sdr_sr_done:
        /* Turn off the main oscillator */
        ldr     tmp1, [pmc, #AT91_CKGR_MOR]
        bic     tmp1, tmp1, #AT91_PMC_MOSCEN
+        orr     tmp1, tmp1, #AT91_PMC_KEY
        str     tmp1, [pmc, #AT91_CKGR_MOR]
        /* Wait for interrupt */
@@ -216,6 +216,7 @@ sdr_sr_done:
        /* Turn on the main oscillator */
        ldr     tmp1, [pmc, #AT91_CKGR_MOR]
        orr     tmp1, tmp1, #AT91_PMC_MOSCEN
+        orr     tmp1, tmp1, #AT91_PMC_KEY
        str     tmp1, [pmc, #AT91_CKGR_MOR]
        wait_moscrdy
@@ -280,12 +281,17 @@ sdr_sr_done:
         */
        cmp     memctrl, #AT91_MEMCTRL_DDRSDR
        bne     sdr_en_restore
+        /* Restore MDR in case of LPDDR1 */
+        ldr     tmp1, .saved_sam9_mdr
+        str     tmp1, [sdramc, #AT91_DDRSDRC_MDR]
        /* Restore LPR on AT91 with DDRAM */
        ldr     tmp1, .saved_sam9_lpr
        str     tmp1, [sdramc, #AT91_DDRSDRC_LPR]
        /* if we use the second ram controller */
        cmp     ramc1, #0
+        ldrne   tmp2, .saved_sam9_mdr1
+        strne   tmp2, [ramc1, #AT91_DDRSDRC_MDR]
        ldrne   tmp2, .saved_sam9_lpr1
        strne   tmp2, [ramc1, #AT91_DDRSDRC_LPR]
@@ -319,5 +325,11 @@ ram_restored:
 .saved_sam9_lpr1:
        .word 0
+.saved_sam9_mdr:
+        .word 0
+.saved_sam9_mdr1:
+        .word 0
 ENTRY(at91_slow_clock_sz)
        .word .-at91_slow_clock
diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c
index 3f32c47a6d74..d2e9f12d12f1 100644
--- a/arch/arm/mach-exynos/platsmp.c
+++ b/arch/arm/mach-exynos/platsmp.c
@@ -126,8 +126,7 @@ static inline void platform_do_lowpower(unsigned int cpu, int *spurious)
 */
 void exynos_cpu_power_down(int cpu)
 {
-        if (cpu == 0 && (of_machine_is_compatible("samsung,exynos5420") ||
+        if (cpu == 0 && (soc_is_exynos5420() || soc_is_exynos5800())) {
-                of_machine_is_compatible("samsung,exynos5800"))) {
                /*
                 * Bypass power down for CPU0 during suspend. Check for
                 * the SYS_PWR_REG value to decide if we are suspending
diff --git a/arch/arm/mach-exynos/pm_domains.c b/arch/arm/mach-exynos/pm_domains.c
index 20f267121b3e..37266a826437 100644
--- a/arch/arm/mach-exynos/pm_domains.c
+++ b/arch/arm/mach-exynos/pm_domains.c
@@ -161,6 +161,34 @@ no_clk:
                of_genpd_add_provider_simple(np, &pd->pd);
        }
+        /* Assign the child power domains to their parents */
+        for_each_compatible_node(np, NULL, "samsung,exynos4210-pd") {
+                struct generic_pm_domain *child_domain, *parent_domain;
+                struct of_phandle_args args;
+                args.np = np;
+                args.args_count = 0;
+                child_domain = of_genpd_get_from_provider(&args);
+                if (!child_domain)
+                        continue;
+                if (of_parse_phandle_with_args(np, "power-domains",
+                                         "#power-domain-cells", 0, &args) != 0)
+                        continue;
+                parent_domain = of_genpd_get_from_provider(&args);
+                if (!parent_domain)
+                        continue;
+                if (pm_genpd_add_subdomain(parent_domain, child_domain))
+                        pr_warn("%s failed to add subdomain: %s\n",
+                                parent_domain->name, child_domain->name);
+                else
+                        pr_info("%s has as child subdomain: %s.\n",
+                                parent_domain->name, child_domain->name);
+                of_node_put(np);
+        }
        return 0;
 }
 arch_initcall(exynos4_pm_init_power_domain);
diff --git a/arch/arm/mach-exynos/suspend.c b/arch/arm/mach-exynos/suspend.c
index 52e2b1a2fddb..318d127df147 100644
--- a/arch/arm/mach-exynos/suspend.c
+++ b/arch/arm/mach-exynos/suspend.c
@@ -87,8 +87,8 @@ static unsigned int exynos_pmu_spare3;
 static u32 exynos_irqwake_intmask = 0xffffffff;
 static const struct exynos_wkup_irq exynos3250_wkup_irq[] = {
-        { 73, BIT(1) }, /* RTC alarm */
+        { 105, BIT(1) }, /* RTC alarm */
-        { 74, BIT(2) }, /* RTC tick */
+        { 106, BIT(2) }, /* RTC tick */
        { /* sentinel */ },
 };
diff --git a/arch/arm/mach-imx/mach-imx6q.c b/arch/arm/mach-imx/mach-imx6q.c
index 4ad6e473cf83..9de3412af406 100644
--- a/arch/arm/mach-imx/mach-imx6q.c
+++ b/arch/arm/mach-imx/mach-imx6q.c
@@ -211,8 +211,9 @@ static void __init imx6q_1588_init(void)
         * set bit IOMUXC_GPR1[21].  Or the PTP clock must be from pad
         * (external OSC), and we need to clear the bit.
         */
-        clksel = ptp_clk == enet_ref ? IMX6Q_GPR1_ENET_CLK_SEL_ANATOP :
+        clksel = clk_is_match(ptp_clk, enet_ref) ?
-                                       IMX6Q_GPR1_ENET_CLK_SEL_PAD;
+                                IMX6Q_GPR1_ENET_CLK_SEL_ANATOP :
+                                IMX6Q_GPR1_ENET_CLK_SEL_PAD;
        gpr = syscon_regmap_lookup_by_compatible("fsl,imx6q-iomuxc-gpr");
        if (!IS_ERR(gpr))
                regmap_update_bits(gpr, IOMUXC_GPR1,
diff --git a/arch/arm/mach-msm/board-halibut.c b/arch/arm/mach-msm/board-halibut.c
index 61bfe584a9d7..fc832040c6e9 100644
--- a/arch/arm/mach-msm/board-halibut.c
+++ b/arch/arm/mach-msm/board-halibut.c
@@ -20,6 +20,7 @@
 #include <linux/input.h>
 #include <linux/io.h>
 #include <linux/delay.h>
+#include <linux/smc91x.h>
 #include <mach/hardware.h>
 #include <asm/mach-types.h>
@@ -46,15 +47,20 @@ static struct resource smc91x_resources[] = {
        [1] = {
                .start  = MSM_GPIO_TO_INT(49),
                .end    = MSM_GPIO_TO_INT(49),
-                .flags  = IORESOURCE_IRQ,
+                .flags  = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
        },
 };
+static struct smc91x_platdata smc91x_platdata = {
+        .flags = SMC91X_USE_16BIT | SMC91X_NOWAIT,
+};
 static struct platform_device smc91x_device = {
        .name           = "smc91x",
        .id             = 0,
        .num_resources  = ARRAY_SIZE(smc91x_resources),
        .resource       = smc91x_resources,
+        .dev.platform_data = &smc91x_platdata,
 };
 static struct platform_device *devices[] __initdata = {
diff --git a/arch/arm/mach-msm/board-qsd8x50.c b/arch/arm/mach-msm/board-qsd8x50.c
index 4c748616ef47..10016a3bc698 100644
--- a/arch/arm/mach-msm/board-qsd8x50.c
+++ b/arch/arm/mach-msm/board-qsd8x50.c
@@ -22,6 +22,7 @@
 #include <linux/usb/msm_hsusb.h>
 #include <linux/err.h>
 #include <linux/clkdev.h>
+#include <linux/smc91x.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
@@ -49,15 +50,20 @@ static struct resource smc91x_resources[] = {
                .flags = IORESOURCE_MEM,
        },
        [1] = {
-                .flags = IORESOURCE_IRQ,
+                .flags  = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
        },
 };
+static struct smc91x_platdata smc91x_platdata = {
+        .flags = SMC91X_USE_16BIT | SMC91X_NOWAIT,
+};
 static struct platform_device smc91x_device = {
        .name           = "smc91x",
        .id             = 0,
        .num_resources  = ARRAY_SIZE(smc91x_resources),
        .resource       = smc91x_resources,
+        .dev.platform_data = &smc91x_platdata,
 };
 static int __init msm_init_smc91x(void)
diff --git a/arch/arm/mach-omap2/omap_hwmod.c b/arch/arm/mach-omap2/omap_hwmod.c
index 92afb723dcfc..355b08936871 100644
--- a/arch/arm/mach-omap2/omap_hwmod.c
+++ b/arch/arm/mach-omap2/omap_hwmod.c
@@ -1692,16 +1692,15 @@ static int _deassert_hardreset(struct omap_hwmod *oh, const char *name)
        if (ret == -EBUSY)
                pr_warn("omap_hwmod: %s: failed to hardreset\n", oh->name);
-        if (!ret) {
+        if (oh->clkdm) {
                /*
                 * Set the clockdomain to HW_AUTO, assuming that the
                 * previous state was HW_AUTO.
                 */
-                if (oh->clkdm && hwsup)
+                if (hwsup)
                        clkdm_allow_idle(oh->clkdm);
-        } else {
-                if (oh->clkdm)
+                clkdm_hwmod_disable(oh->clkdm, oh);
-                        clkdm_hwmod_disable(oh->clkdm, oh);
        }
        return ret;
@@ -2698,6 +2697,7 @@ static int __init _register(struct omap_hwmod *oh)
        INIT_LIST_HEAD(&oh->master_ports);
        INIT_LIST_HEAD(&oh->slave_ports);
        spin_lock_init(&oh->_lock);
+        lockdep_set_class(&oh->_lock, &oh->hwmod_key);
        oh->_state = _HWMOD_STATE_REGISTERED;
diff --git a/arch/arm/mach-omap2/omap_hwmod.h b/arch/arm/mach-omap2/omap_hwmod.h
index 9d4bec6ee742..9611c91d9b82 100644
--- a/arch/arm/mach-omap2/omap_hwmod.h
+++ b/arch/arm/mach-omap2/omap_hwmod.h
@@ -674,6 +674,7 @@ struct omap_hwmod {
        u32                             _sysc_cache;
        void __iomem                    *_mpu_rt_va;
        spinlock_t                      _lock;
+        struct lock_class_key           hwmod_key; /* unique lock class */
        struct list_head                node;
        struct omap_hwmod_ocp_if        *_mpu_port;
        unsigned int                    (*xlate_irq)(unsigned int);
diff --git a/arch/arm/mach-omap2/omap_hwmod_7xx_data.c b/arch/arm/mach-omap2/omap_hwmod_7xx_data.c
index e8692e7675b8..16fe7a1b7a35 100644
--- a/arch/arm/mach-omap2/omap_hwmod_7xx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_7xx_data.c
@@ -1466,55 +1466,18 @@ static struct omap_hwmod dra7xx_ocp2scp3_hwmod = {
 *
 */
-static struct omap_hwmod_class dra7xx_pcie_hwmod_class = {
+static struct omap_hwmod_class dra7xx_pciess_hwmod_class = {
        .name   = "pcie",
 };
 /* pcie1 */
-static struct omap_hwmod dra7xx_pcie1_hwmod = {
+static struct omap_hwmod dra7xx_pciess1_hwmod = {
        .name           = "pcie1",
-        .class          = &dra7xx_pcie_hwmod_class,
+        .class          = &dra7xx_pciess_hwmod_class,
        .clkdm_name     = "pcie_clkdm",
        .main_clk       = "l4_root_clk_div",
        .prcm = {
                .omap4 = {
-                        .clkctrl_offs   = DRA7XX_CM_PCIE_CLKSTCTRL_OFFSET,
-                        .modulemode     = MODULEMODE_SWCTRL,
-                },
-        },
-};
-/* pcie2 */
-static struct omap_hwmod dra7xx_pcie2_hwmod = {
-        .name           = "pcie2",
-        .class          = &dra7xx_pcie_hwmod_class,
-        .clkdm_name     = "pcie_clkdm",
-        .main_clk       = "l4_root_clk_div",
-        .prcm = {
-                .omap4 = {
-                        .clkctrl_offs = DRA7XX_CM_PCIE_CLKSTCTRL_OFFSET,
-                        .modulemode   = MODULEMODE_SWCTRL,
-                },
-        },
-};
-/*
- * 'PCIE PHY' class
- *
- */
-static struct omap_hwmod_class dra7xx_pcie_phy_hwmod_class = {
-        .name   = "pcie-phy",
-};
-/* pcie1 phy */
-static struct omap_hwmod dra7xx_pcie1_phy_hwmod = {
-        .name           = "pcie1-phy",
-        .class          = &dra7xx_pcie_phy_hwmod_class,
-        .clkdm_name     = "l3init_clkdm",
-        .main_clk       = "l4_root_clk_div",
-        .prcm = {
-                .omap4 = {
                        .clkctrl_offs = DRA7XX_CM_L3INIT_PCIESS1_CLKCTRL_OFFSET,
                        .context_offs = DRA7XX_RM_L3INIT_PCIESS1_CONTEXT_OFFSET,
                        .modulemode   = MODULEMODE_SWCTRL,
@@ -1522,11 +1485,11 @@ static struct omap_hwmod dra7xx_pcie1_phy_hwmod = {
        },
 };
-/* pcie2 phy */
+/* pcie2 */
-static struct omap_hwmod dra7xx_pcie2_phy_hwmod = {
+static struct omap_hwmod dra7xx_pciess2_hwmod = {
-        .name           = "pcie2-phy",
+        .name           = "pcie2",
-        .class          = &dra7xx_pcie_phy_hwmod_class,
+        .class          = &dra7xx_pciess_hwmod_class,
-        .clkdm_name     = "l3init_clkdm",
+        .clkdm_name     = "pcie_clkdm",
        .main_clk       = "l4_root_clk_div",
        .prcm = {
                .omap4 = {
@@ -2877,50 +2840,34 @@ static struct omap_hwmod_ocp_if dra7xx_l4_cfg__ocp2scp3 = {
        .user           = OCP_USER_MPU | OCP_USER_SDMA,
 };
-/* l3_main_1 -> pcie1 */
+/* l3_main_1 -> pciess1 */
-static struct omap_hwmod_ocp_if dra7xx_l3_main_1__pcie1 = {
+static struct omap_hwmod_ocp_if dra7xx_l3_main_1__pciess1 = {
        .master         = &dra7xx_l3_main_1_hwmod,
-        .slave          = &dra7xx_pcie1_hwmod,
+        .slave          = &dra7xx_pciess1_hwmod,
        .clk            = "l3_iclk_div",
        .user           = OCP_USER_MPU | OCP_USER_SDMA,
 };
-/* l4_cfg -> pcie1 */
+/* l4_cfg -> pciess1 */
-static struct omap_hwmod_ocp_if dra7xx_l4_cfg__pcie1 = {
+static struct omap_hwmod_ocp_if dra7xx_l4_cfg__pciess1 = {
        .master         = &dra7xx_l4_cfg_hwmod,
-        .slave          = &dra7xx_pcie1_hwmod,
+        .slave          = &dra7xx_pciess1_hwmod,
        .clk            = "l4_root_clk_div",
        .user           = OCP_USER_MPU | OCP_USER_SDMA,
 };
-/* l3_main_1 -> pcie2 */
+/* l3_main_1 -> pciess2 */
-static struct omap_hwmod_ocp_if dra7xx_l3_main_1__pcie2 = {
+static struct omap_hwmod_ocp_if dra7xx_l3_main_1__pciess2 = {
        .master         = &dra7xx_l3_main_1_hwmod,
-        .slave          = &dra7xx_pcie2_hwmod,
+        .slave          = &dra7xx_pciess2_hwmod,
        .clk            = "l3_iclk_div",
        .user           = OCP_USER_MPU | OCP_USER_SDMA,
 };
-/* l4_cfg -> pcie2 */
+/* l4_cfg -> pciess2 */
-static struct omap_hwmod_ocp_if dra7xx_l4_cfg__pcie2 = {
+static struct omap_hwmod_ocp_if dra7xx_l4_cfg__pciess2 = {
-        .master         = &dra7xx_l4_cfg_hwmod,
-        .slave          = &dra7xx_pcie2_hwmod,
-        .clk            = "l4_root_clk_div",
-        .user           = OCP_USER_MPU | OCP_USER_SDMA,
-};
-/* l4_cfg -> pcie1 phy */
-static struct omap_hwmod_ocp_if dra7xx_l4_cfg__pcie1_phy = {
-        .master         = &dra7xx_l4_cfg_hwmod,
-        .slave          = &dra7xx_pcie1_phy_hwmod,
-        .clk            = "l4_root_clk_div",
-        .user           = OCP_USER_MPU | OCP_USER_SDMA,
-};
-/* l4_cfg -> pcie2 phy */
-static struct omap_hwmod_ocp_if dra7xx_l4_cfg__pcie2_phy = {
        .master         = &dra7xx_l4_cfg_hwmod,
-        .slave          = &dra7xx_pcie2_phy_hwmod,
+        .slave          = &dra7xx_pciess2_hwmod,
        .clk            = "l4_root_clk_div",
        .user           = OCP_USER_MPU | OCP_USER_SDMA,
 };
@@ -3327,12 +3274,10 @@ static struct omap_hwmod_ocp_if *dra7xx_hwmod_ocp_ifs[] __initdata = {
        &dra7xx_l4_cfg__mpu,
        &dra7xx_l4_cfg__ocp2scp1,
        &dra7xx_l4_cfg__ocp2scp3,
-        &dra7xx_l3_main_1__pcie1,
+        &dra7xx_l3_main_1__pciess1,
-        &dra7xx_l4_cfg__pcie1,
+        &dra7xx_l4_cfg__pciess1,
-        &dra7xx_l3_main_1__pcie2,
+        &dra7xx_l3_main_1__pciess2,
-        &dra7xx_l4_cfg__pcie2,
+        &dra7xx_l4_cfg__pciess2,
-        &dra7xx_l4_cfg__pcie1_phy,
-        &dra7xx_l4_cfg__pcie2_phy,
        &dra7xx_l3_main_1__qspi,
        &dra7xx_l4_per3__rtcss,
        &dra7xx_l4_cfg__sata,
diff --git a/arch/arm/mach-omap2/pdata-quirks.c b/arch/arm/mach-omap2/pdata-quirks.c
index 190fa43e7479..e642b079e9f3 100644
--- a/arch/arm/mach-omap2/pdata-quirks.c
+++ b/arch/arm/mach-omap2/pdata-quirks.c
@@ -173,6 +173,7 @@ static void __init omap3_igep0030_rev_g_legacy_init(void)
 static void __init omap3_evm_legacy_init(void)
 {
+        hsmmc2_internal_input_clk();
        legacy_init_wl12xx(WL12XX_REFCLOCK_38, 0, 149);
 }
diff --git a/arch/arm/mach-omap2/prm44xx.c b/arch/arm/mach-omap2/prm44xx.c
index a08a617a6c11..d6d6bc39e05c 100644
--- a/arch/arm/mach-omap2/prm44xx.c
+++ b/arch/arm/mach-omap2/prm44xx.c
@@ -252,10 +252,10 @@ static void omap44xx_prm_save_and_clear_irqen(u32 *saved_mask)
 {
        saved_mask[0] =
                omap4_prm_read_inst_reg(OMAP4430_PRM_OCP_SOCKET_INST,
-                                        OMAP4_PRM_IRQSTATUS_MPU_OFFSET);
+                                        OMAP4_PRM_IRQENABLE_MPU_OFFSET);
        saved_mask[1] =
                omap4_prm_read_inst_reg(OMAP4430_PRM_OCP_SOCKET_INST,
-                                        OMAP4_PRM_IRQSTATUS_MPU_2_OFFSET);
+                                        OMAP4_PRM_IRQENABLE_MPU_2_OFFSET);
        omap4_prm_write_inst_reg(0, OMAP4430_PRM_OCP_SOCKET_INST,
                                 OMAP4_PRM_IRQENABLE_MPU_OFFSET);
diff --git a/arch/arm/mach-pxa/idp.c b/arch/arm/mach-pxa/idp.c
index 343c4e3a7c5d..f6d02e4cbcda 100644
--- a/arch/arm/mach-pxa/idp.c
+++ b/arch/arm/mach-pxa/idp.c
@@ -36,6 +36,7 @@
 #include <linux/platform_data/video-pxafb.h>
 #include <mach/bitfield.h>
 #include <linux/platform_data/mmc-pxamci.h>
+#include <linux/smc91x.h>
 #include "generic.h"
 #include "devices.h"
@@ -81,11 +82,16 @@ static struct resource smc91x_resources[] = {
        }
 };
+static struct smc91x_platdata smc91x_platdata = {
+        .flags = SMC91X_USE_32BIT | SMC91X_USE_DMA | SMC91X_NOWAIT,
+};
 static struct platform_device smc91x_device = {
        .name           = "smc91x",
        .id             = 0,
        .num_resources  = ARRAY_SIZE(smc91x_resources),
        .resource       = smc91x_resources,
+        .dev.platform_data = &smc91x_platdata,
 };
 static void idp_backlight_power(int on)
diff --git a/arch/arm/mach-pxa/lpd270.c b/arch/arm/mach-pxa/lpd270.c
index ad777b353bd5..eaee2c20b189 100644
--- a/arch/arm/mach-pxa/lpd270.c
+++ b/arch/arm/mach-pxa/lpd270.c
@@ -24,6 +24,7 @@
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
 #include <linux/pwm_backlight.h>
+#include <linux/smc91x.h>
 #include <asm/types.h>
 #include <asm/setup.h>
@@ -189,15 +190,20 @@ static struct resource smc91x_resources[] = {
        [1] = {
                .start  = LPD270_ETHERNET_IRQ,
                .end    = LPD270_ETHERNET_IRQ,
-                .flags  = IORESOURCE_IRQ,
+                .flags  = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHEDGE,
        },
 };
+struct smc91x_platdata smc91x_platdata = {
+        .flags = SMC91X_USE_16BIT | SMC91X_NOWAIT,
+};
 static struct platform_device smc91x_device = {
        .name           = "smc91x",
        .id             = 0,
        .num_resources  = ARRAY_SIZE(smc91x_resources),
        .resource       = smc91x_resources,
+        .dev.platform_data = &smc91x_platdata,
 };
 static struct resource lpd270_flash_resources[] = {
diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c
index 850e506926df..c309593abdb2 100644
--- a/arch/arm/mach-realview/core.c
+++ b/arch/arm/mach-realview/core.c
@@ -28,6 +28,7 @@
 #include <linux/platform_data/video-clcd-versatile.h>
 #include <linux/io.h>
 #include <linux/smsc911x.h>
+#include <linux/smc91x.h>
 #include <linux/ata_platform.h>
 #include <linux/amba/mmci.h>
 #include <linux/gfp.h>
@@ -94,6 +95,10 @@ static struct smsc911x_platform_config smsc911x_config = {
        .phy_interface  = PHY_INTERFACE_MODE_MII,
 };
+static struct smc91x_platdata smc91x_platdata = {
+        .flags = SMC91X_USE_32BIT | SMC91X_NOWAIT,
+};
 static struct platform_device realview_eth_device = {
        .name           = "smsc911x",
        .id             = 0,
@@ -107,6 +112,8 @@ int realview_eth_register(const char *name, struct resource *res)
        realview_eth_device.resource = res;
        if (strcmp(realview_eth_device.name, "smsc911x") == 0)
                realview_eth_device.dev.platform_data = &smsc911x_config;
+        else
+                realview_eth_device.dev.platform_data = &smc91x_platdata;
        return platform_device_register(&realview_eth_device);
 }
diff --git a/arch/arm/mach-realview/realview_eb.c b/arch/arm/mach-realview/realview_eb.c
index 64c88d657f9e..b3869cbbcc68 100644
--- a/arch/arm/mach-realview/realview_eb.c
+++ b/arch/arm/mach-realview/realview_eb.c
@@ -234,7 +234,7 @@ static struct resource realview_eb_eth_resources[] = {
        [1] = {
                .start          = IRQ_EB_ETH,
                .end            = IRQ_EB_ETH,
-                .flags          = IORESOURCE_IRQ,
+                .flags          = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHEDGE,
        },
 };
diff --git a/arch/arm/mach-sa1100/neponset.c b/arch/arm/mach-sa1100/neponset.c
index 169262e3040d..af868d258e66 100644
--- a/arch/arm/mach-sa1100/neponset.c
+++ b/arch/arm/mach-sa1100/neponset.c
@@ -12,6 +12,7 @@
 #include <linux/pm.h>
 #include <linux/serial_core.h>
 #include <linux/slab.h>
+#include <linux/smc91x.h>
 #include <asm/mach-types.h>
 #include <asm/mach/map.h>
@@ -258,12 +259,17 @@ static int neponset_probe(struct platform_device *dev)
                        0x02000000, "smc91x-attrib"),
                { .flags = IORESOURCE_IRQ },
        };
+        struct smc91x_platdata smc91x_platdata = {
+                .flags = SMC91X_USE_8BIT | SMC91X_IO_SHIFT_2 | SMC91X_NOWAIT,
+        };
        struct platform_device_info smc91x_devinfo = {
                .parent = &dev->dev,
                .name = "smc91x",
                .id = 0,
                .res = smc91x_resources,
                .num_res = ARRAY_SIZE(smc91x_resources),
+                .data = &smc91x_platdata,
+                .size_data = sizeof(smc91x_platdata),
        };
        int ret, irq;
diff --git a/arch/arm/mach-sa1100/pleb.c b/arch/arm/mach-sa1100/pleb.c
index 091261878eff..1525d7b5f1b7 100644
--- a/arch/arm/mach-sa1100/pleb.c
+++ b/arch/arm/mach-sa1100/pleb.c
@@ -11,6 +11,7 @@
 #include <linux/irq.h>
 #include <linux/io.h>
 #include <linux/mtd/partitions.h>
+#include <linux/smc91x.h>
 #include <mach/hardware.h>
 #include <asm/setup.h>
@@ -43,12 +44,18 @@ static struct resource smc91x_resources[] = {
 #endif
 };
+static struct smc91x_platdata smc91x_platdata = {
+        .flags = SMC91X_USE_16BIT | SMC91X_NOWAIT,
+};
 static struct platform_device smc91x_device = {
        .name           = "smc91x",
        .id             = 0,
        .num_resources  = ARRAY_SIZE(smc91x_resources),
        .resource       = smc91x_resources,
+        .dev = {
+                .platform_data  = &smc91x_platdata,
+        },
 };
 static struct platform_device *devices[] __initdata = {
diff --git a/arch/arm/mach-socfpga/core.h b/arch/arm/mach-socfpga/core.h
index 483cb467bf65..a0f3b1cd497c 100644
--- a/arch/arm/mach-socfpga/core.h
+++ b/arch/arm/mach-socfpga/core.h
@@ -45,6 +45,6 @@ extern char secondary_trampoline, secondary_trampoline_end;
 extern unsigned long socfpga_cpu1start_addr;
-#define SOCFPGA_SCU_VIRT_BASE   0xfffec000
+#define SOCFPGA_SCU_VIRT_BASE   0xfee00000
 #endif
diff --git a/arch/arm/mach-socfpga/socfpga.c b/arch/arm/mach-socfpga/socfpga.c
index 383d61e138af..f5e597c207b9 100644
--- a/arch/arm/mach-socfpga/socfpga.c
+++ b/arch/arm/mach-socfpga/socfpga.c
@@ -23,6 +23,7 @@
 #include <asm/hardware/cache-l2x0.h>
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
+#include <asm/cacheflush.h>
 #include "core.h"
@@ -73,6 +74,10 @@ void __init socfpga_sysmgr_init(void)
                        (u32 *) &socfpga_cpu1start_addr))
                pr_err("SMP: Need cpu1-start-addr in device tree.\n");
+        /* Ensure that socfpga_cpu1start_addr is visible to other CPUs */
+        smp_wmb();
+        sync_cache_w(&socfpga_cpu1start_addr);
        sys_manager_base_addr = of_iomap(np, 0);
        np = of_find_compatible_node(NULL, NULL, "altr,rst-mgr");
diff --git a/arch/arm/mach-sti/board-dt.c b/arch/arm/mach-sti/board-dt.c
index b067390cef4e..b373acade338 100644
--- a/arch/arm/mach-sti/board-dt.c
+++ b/arch/arm/mach-sti/board-dt.c
@@ -18,6 +18,7 @@ static const char *stih41x_dt_match[] __initdata = {
        "st,stih415",
        "st,stih416",
        "st,stih407",
+        "st,stih410",
        "st,stih418",
        NULL
 };
diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c
index c6c7696b8db9..8f15f70622a6 100644
--- a/arch/arm/mm/cache-l2x0.c
+++ b/arch/arm/mm/cache-l2x0.c
@@ -1131,23 +1131,22 @@ static void __init l2c310_of_parse(const struct device_node *np,
        }
        ret = l2x0_cache_size_of_parse(np, aux_val, aux_mask, &assoc, SZ_512K);
-        if (ret)
+        if (!ret) {
-                return;
+                switch (assoc) {
+                case 16:
-        switch (assoc) {
+                        *aux_val &= ~L2X0_AUX_CTRL_ASSOC_MASK;
-        case 16:
+                        *aux_val |= L310_AUX_CTRL_ASSOCIATIVITY_16;
-                *aux_val &= ~L2X0_AUX_CTRL_ASSOC_MASK;
+                        *aux_mask &= ~L2X0_AUX_CTRL_ASSOC_MASK;
-                *aux_val |= L310_AUX_CTRL_ASSOCIATIVITY_16;
+                        break;
-                *aux_mask &= ~L2X0_AUX_CTRL_ASSOC_MASK;
+                case 8:
-                break;
+                        *aux_val &= ~L2X0_AUX_CTRL_ASSOC_MASK;
-        case 8:
+                        *aux_mask &= ~L2X0_AUX_CTRL_ASSOC_MASK;
-                *aux_val &= ~L2X0_AUX_CTRL_ASSOC_MASK;
+                        break;
-                *aux_mask &= ~L2X0_AUX_CTRL_ASSOC_MASK;
+                default:
-                break;
+                        pr_err("L2C-310 OF cache associativity %d invalid, only 8 or 16 permitted\n",
-        default:
+                               assoc);
-                pr_err("L2C-310 OF cache associativity %d invalid, only 8 or 16 permitted\n",
+                        break;
-                       assoc);
+                }
-                break;
        }
        prefetch = l2x0_saved_regs.prefetch_ctrl;
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 170a116d1b29..c27447653903 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -171,7 +171,7 @@ static int __dma_supported(struct device *dev, u64 mask, bool warn)
         */
        if (sizeof(mask) != sizeof(dma_addr_t) &&
            mask > (dma_addr_t)~0 &&
-            dma_to_pfn(dev, ~0) < max_pfn) {
+            dma_to_pfn(dev, ~0) < max_pfn - 1) {
                if (warn) {
                        dev_warn(dev, "Coherent DMA mask %#llx is larger than dma_addr_t allows\n",
                                 mask);
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index a982dc3190df..6333d9c17875 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -552,6 +552,7 @@ do_DataAbort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
        pr_alert("Unhandled fault: %s (0x%03x) at 0x%08lx\n",
                inf->name, fsr, addr);
+        show_pte(current->mm, addr);
        info.si_signo = inf->sig;
        info.si_errno = 0;
diff --git a/arch/arm/mm/pageattr.c b/arch/arm/mm/pageattr.c
index 004e35cdcfff..cf30daff8932 100644
--- a/arch/arm/mm/pageattr.c
+++ b/arch/arm/mm/pageattr.c
@@ -49,7 +49,10 @@ static int change_memory_common(unsigned long addr, int numpages,
                WARN_ON_ONCE(1);
        }
-        if (!is_module_address(start) || !is_module_address(end - 1))
+        if (start < MODULES_VADDR || start >= MODULES_END)
+                return -EINVAL;
+        if (end < MODULES_VADDR || start >= MODULES_END)
                return -EINVAL;
        data.set_mask = set_mask;
diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi
index f1ad9c2ab2e9..a857794432d6 100644
--- a/arch/arm64/boot/dts/apm/apm-storm.dtsi
+++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi
@@ -622,7 +622,7 @@
                };
                sgenet0: ethernet@1f210000 {
-                        compatible = "apm,xgene-enet";
+                        compatible = "apm,xgene1-sgenet";
                        status = "disabled";
                        reg = <0x0 0x1f210000 0x0 0xd100>,
                              <0x0 0x1f200000 0x0 0Xc300>,
@@ -636,7 +636,7 @@
                };
                xgenet: ethernet@1f610000 {
-                        compatible = "apm,xgene-enet";
+                        compatible = "apm,xgene1-xgenet";
                        status = "disabled";
                        reg = <0x0 0x1f610000 0x0 0xd100>,
                              <0x0 0x1f600000 0x0 0Xc300>,
diff --git a/arch/arm64/include/asm/cmpxchg.h b/arch/arm64/include/asm/cmpxchg.h
index cb9593079f29..d8c25b7b18fb 100644
--- a/arch/arm64/include/asm/cmpxchg.h
+++ b/arch/arm64/include/asm/cmpxchg.h
@@ -246,14 +246,30 @@ static inline unsigned long __cmpxchg_mb(volatile void *ptr, unsigned long old,
        __ret; \
 })
-#define this_cpu_cmpxchg_1(ptr, o, n) cmpxchg_local(raw_cpu_ptr(&(ptr)), o, n)
+#define _protect_cmpxchg_local(pcp, o, n)                       \
-#define this_cpu_cmpxchg_2(ptr, o, n) cmpxchg_local(raw_cpu_ptr(&(ptr)), o, n)
+({                                                              \
-#define this_cpu_cmpxchg_4(ptr, o, n) cmpxchg_local(raw_cpu_ptr(&(ptr)), o, n)
+        typeof(*raw_cpu_ptr(&(pcp))) __ret;                     \
-#define this_cpu_cmpxchg_8(ptr, o, n) cmpxchg_local(raw_cpu_ptr(&(ptr)), o, n)
+        preempt_disable();                                      \
+        __ret = cmpxchg_local(raw_cpu_ptr(&(pcp)), o, n);       \
-#define this_cpu_cmpxchg_double_8(ptr1, ptr2, o1, o2, n1, n2) \
+        preempt_enable();                                       \
-        cmpxchg_double_local(raw_cpu_ptr(&(ptr1)), raw_cpu_ptr(&(ptr2)), \
+        __ret;                                                  \
-                                o1, o2, n1, n2)
+})
+#define this_cpu_cmpxchg_1(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
+#define this_cpu_cmpxchg_2(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
+#define this_cpu_cmpxchg_4(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
+#define this_cpu_cmpxchg_8(ptr, o, n) _protect_cmpxchg_local(ptr, o, n)
+#define this_cpu_cmpxchg_double_8(ptr1, ptr2, o1, o2, n1, n2)           \
+({                                                                      \
+        int __ret;                                                      \
+        preempt_disable();                                              \
+        __ret = cmpxchg_double_local(   raw_cpu_ptr(&(ptr1)),           \
+                                        raw_cpu_ptr(&(ptr2)),           \
+                                        o1, o2, n1, n2);                \
+        preempt_enable();                                               \
+        __ret;                                                          \
+})
 #define cmpxchg64(ptr,o,n)              cmpxchg((ptr),(o),(n))
 #define cmpxchg64_local(ptr,o,n)        cmpxchg_local((ptr),(o),(n))
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 94674eb7e7bb..54bb4ba97441 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -129,6 +129,9 @@
 * 40 bits wide (T0SZ = 24).  Systems with a PARange smaller than 40 bits are
 * not known to exist and will break with this configuration.
 *
+ * VTCR_EL2.PS is extracted from ID_AA64MMFR0_EL1.PARange at boot time
+ * (see hyp-init.S).
+ *
 * Note that when using 4K pages, we concatenate two first level page tables
 * together.
 *
@@ -138,7 +141,6 @@
 #ifdef CONFIG_ARM64_64K_PAGES
 /*
 * Stage2 translation configuration:
- * 40bits output (PS = 2)
 * 40bits input  (T0SZ = 24)
 * 64kB pages (TG0 = 1)
 * 2 level page tables (SL = 1)
@@ -150,7 +152,6 @@
 #else
 /*
 * Stage2 translation configuration:
- * 40bits output (PS = 2)
 * 40bits input  (T0SZ = 24)
 * 4kB pages (TG0 = 0)
 * 3 level page tables (SL = 1)
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index 6458b5373142..bbfb600fa822 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -158,6 +158,8 @@ static inline bool kvm_s2pmd_readonly(pmd_t *pmd)
 #define PTRS_PER_S2_PGD         (1 << PTRS_PER_S2_PGD_SHIFT)
 #define S2_PGD_ORDER            get_order(PTRS_PER_S2_PGD * sizeof(pgd_t))
+#define kvm_pgd_index(addr)     (((addr) >> PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1))
 /*
 * If we are concatenating first level stage-2 page tables, we would have less
 * than or equal to 16 pointers in the fake PGD, because that's what the
@@ -171,43 +173,6 @@ static inline bool kvm_s2pmd_readonly(pmd_t *pmd)
 #define KVM_PREALLOC_LEVEL      (0)
 #endif
-/**
- * kvm_prealloc_hwpgd - allocate inital table for VTTBR
- * @kvm:        The KVM struct pointer for the VM.
- * @pgd:        The kernel pseudo pgd
- *
- * When the kernel uses more levels of page tables than the guest, we allocate
- * a fake PGD and pre-populate it to point to the next-level page table, which
- * will be the real initial page table pointed to by the VTTBR.
- *
- * When KVM_PREALLOC_LEVEL==2, we allocate a single page for the PMD and
- * the kernel will use folded pud.  When KVM_PREALLOC_LEVEL==1, we
- * allocate 2 consecutive PUD pages.
- */
-static inline int kvm_prealloc_hwpgd(struct kvm *kvm, pgd_t *pgd)
-{
-        unsigned int i;
-        unsigned long hwpgd;
-        if (KVM_PREALLOC_LEVEL == 0)
-                return 0;
-        hwpgd = __get_free_pages(GFP_KERNEL | __GFP_ZERO, PTRS_PER_S2_PGD_SHIFT);
-        if (!hwpgd)
-                return -ENOMEM;
-        for (i = 0; i < PTRS_PER_S2_PGD; i++) {
-                if (KVM_PREALLOC_LEVEL == 1)
-                        pgd_populate(NULL, pgd + i,
-                                     (pud_t *)hwpgd + i * PTRS_PER_PUD);
-                else if (KVM_PREALLOC_LEVEL == 2)
-                        pud_populate(NULL, pud_offset(pgd, 0) + i,
-                                     (pmd_t *)hwpgd + i * PTRS_PER_PMD);
-        }
-        return 0;
-}
 static inline void *kvm_get_hwpgd(struct kvm *kvm)
 {
        pgd_t *pgd = kvm->arch.pgd;
@@ -224,12 +189,11 @@ static inline void *kvm_get_hwpgd(struct kvm *kvm)
        return pmd_offset(pud, 0);
 }
-static inline void kvm_free_hwpgd(struct kvm *kvm)
+static inline unsigned int kvm_get_hwpgd_size(void)
 {
-        if (KVM_PREALLOC_LEVEL > 0) {
+        if (KVM_PREALLOC_LEVEL > 0)
-                unsigned long hwpgd = (unsigned long)kvm_get_hwpgd(kvm);
+                return PTRS_PER_S2_PGD * PAGE_SIZE;
-                free_pages(hwpgd, PTRS_PER_S2_PGD_SHIFT);
+        return PTRS_PER_S2_PGD * sizeof(pgd_t);
-        }
 }
 static inline bool kvm_page_empty(void *ptr)
diff --git a/arch/arm64/include/asm/mmu_context.h b/arch/arm64/include/asm/mmu_context.h
index a9eee33dfa62..101a42bde728 100644
--- a/arch/arm64/include/asm/mmu_context.h
+++ b/arch/arm64/include/asm/mmu_context.h
@@ -151,6 +151,15 @@ switch_mm(struct mm_struct *prev, struct mm_struct *next,
 {
        unsigned int cpu = smp_processor_id();
+        /*
+         * init_mm.pgd does not contain any user mappings and it is always
+         * active for kernel addresses in TTBR1. Just set the reserved TTBR0.
+         */
+        if (next == &init_mm) {
+                cpu_set_reserved_ttbr0();
+                return;
+        }
        if (!cpumask_test_and_set_cpu(cpu, mm_cpumask(next)) || prev != next)
                check_and_switch_context(next, tsk);
 }
diff --git a/arch/arm64/include/asm/percpu.h b/arch/arm64/include/asm/percpu.h
index 09da25bc596f..4fde8c1df97f 100644
--- a/arch/arm64/include/asm/percpu.h
+++ b/arch/arm64/include/asm/percpu.h
@@ -204,25 +204,47 @@ static inline unsigned long __percpu_xchg(void *ptr, unsigned long val,
        return ret;
 }
+#define _percpu_read(pcp)                                               \
+({                                                                      \
+        typeof(pcp) __retval;                                           \
+        preempt_disable();                                              \
+        __retval = (typeof(pcp))__percpu_read(raw_cpu_ptr(&(pcp)),      \
+                                              sizeof(pcp));             \
+        preempt_enable();                                               \
+        __retval;                                                       \
+})
+#define _percpu_write(pcp, val)                                         \
+do {                                                                    \
+        preempt_disable();                                              \
+        __percpu_write(raw_cpu_ptr(&(pcp)), (unsigned long)(val),       \
+                                sizeof(pcp));                           \
+        preempt_enable();                                               \
+} while(0)                                                              \
+#define _pcp_protect(operation, pcp, val)                       \
+({                                                              \
+        typeof(pcp) __retval;                                   \
+        preempt_disable();                                      \
+        __retval = (typeof(pcp))operation(raw_cpu_ptr(&(pcp)),  \
+                                          (val), sizeof(pcp));  \
+        preempt_enable();                                       \
+        __retval;                                               \
+})
 #define _percpu_add(pcp, val) \
-        __percpu_add(raw_cpu_ptr(&(pcp)), val, sizeof(pcp))
+        _pcp_protect(__percpu_add, pcp, val)
-#define _percpu_add_return(pcp, val) (typeof(pcp)) (_percpu_add(pcp, val))
+#define _percpu_add_return(pcp, val) _percpu_add(pcp, val)
 #define _percpu_and(pcp, val) \
-        __percpu_and(raw_cpu_ptr(&(pcp)), val, sizeof(pcp))
+        _pcp_protect(__percpu_and, pcp, val)
 #define _percpu_or(pcp, val) \
-        __percpu_or(raw_cpu_ptr(&(pcp)), val, sizeof(pcp))
+        _pcp_protect(__percpu_or, pcp, val)
-#define _percpu_read(pcp) (typeof(pcp)) \
-        (__percpu_read(raw_cpu_ptr(&(pcp)), sizeof(pcp)))
-#define _percpu_write(pcp, val) \
-        __percpu_write(raw_cpu_ptr(&(pcp)), (unsigned long)(val), sizeof(pcp))
 #define _percpu_xchg(pcp, val) (typeof(pcp)) \
-        (__percpu_xchg(raw_cpu_ptr(&(pcp)), (unsigned long)(val), sizeof(pcp)))
+        _pcp_protect(__percpu_xchg, pcp, (unsigned long)(val))
 #define this_cpu_add_1(pcp, val) _percpu_add(pcp, val)
 #define this_cpu_add_2(pcp, val) _percpu_add(pcp, val)
diff --git a/arch/arm64/include/asm/proc-fns.h b/arch/arm64/include/asm/proc-fns.h
index 9a8fd84f8fb2..941c375616e2 100644
--- a/arch/arm64/include/asm/proc-fns.h
+++ b/arch/arm64/include/asm/proc-fns.h
@@ -39,7 +39,11 @@ extern u64 cpu_do_resume(phys_addr_t ptr, u64 idmap_ttbr);
 #include <asm/memory.h>
-#define cpu_switch_mm(pgd,mm) cpu_do_switch_mm(virt_to_phys(pgd),mm)
+#define cpu_switch_mm(pgd,mm)                           \
+do {                                                    \
+        BUG_ON(pgd == swapper_pg_dir);                  \
+        cpu_do_switch_mm(virt_to_phys(pgd),mm);         \
+} while (0)
 #define cpu_get_pgd()                                   \
 ({                                                      \
diff --git a/arch/arm64/include/asm/tlb.h b/arch/arm64/include/asm/tlb.h
index c028fe37456f..53d9c354219f 100644
--- a/arch/arm64/include/asm/tlb.h
+++ b/arch/arm64/include/asm/tlb.h
@@ -48,6 +48,7 @@ static inline void tlb_flush(struct mmu_gather *tlb)
 static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
                                  unsigned long addr)
 {
+        __flush_tlb_pgtable(tlb->mm, addr);
        pgtable_page_dtor(pte);
        tlb_remove_entry(tlb, pte);
 }
@@ -56,6 +57,7 @@ static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
 static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp,
                                  unsigned long addr)
 {
+        __flush_tlb_pgtable(tlb->mm, addr);
        tlb_remove_entry(tlb, virt_to_page(pmdp));
 }
 #endif
@@ -64,6 +66,7 @@ static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp,
 static inline void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pudp,
                                  unsigned long addr)
 {
+        __flush_tlb_pgtable(tlb->mm, addr);
        tlb_remove_entry(tlb, virt_to_page(pudp));
 }
 #endif
diff --git a/arch/arm64/include/asm/tlbflush.h b/arch/arm64/include/asm/tlbflush.h
index 4abe9b945f77..c3bb05b98616 100644
--- a/arch/arm64/include/asm/tlbflush.h
+++ b/arch/arm64/include/asm/tlbflush.h
@@ -144,6 +144,19 @@ static inline void flush_tlb_kernel_range(unsigned long start, unsigned long end
 }
 /*
+ * Used to invalidate the TLB (walk caches) corresponding to intermediate page
+ * table levels (pgd/pud/pmd).
+ */
+static inline void __flush_tlb_pgtable(struct mm_struct *mm,
+                                       unsigned long uaddr)
+{
+        unsigned long addr = uaddr >> 12 | ((unsigned long)ASID(mm) << 48);
+        dsb(ishst);
+        asm("tlbi       vae1is, %0" : : "r" (addr));
+        dsb(ish);
+}
+/*
 * On AArch64, the cache coherency is handled via the set_pte_at() function.
 */
 static inline void update_mmu_cache(struct vm_area_struct *vma,
diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c
index b42c7b480e1e..ab21e0d58278 100644
--- a/arch/arm64/kernel/efi.c
+++ b/arch/arm64/kernel/efi.c
@@ -337,7 +337,11 @@ core_initcall(arm64_dmi_init);
 static void efi_set_pgd(struct mm_struct *mm)
 {
-        cpu_switch_mm(mm->pgd, mm);
+        if (mm == &init_mm)
+                cpu_set_reserved_ttbr0();
+        else
+                cpu_switch_mm(mm->pgd, mm);
        flush_tlb_all();
        if (icache_is_aivivt())
                __flush_icache_all();
@@ -354,3 +358,12 @@ void efi_virtmap_unload(void)
        efi_set_pgd(current->active_mm);
        preempt_enable();
 }
+/*
+ * UpdateCapsule() depends on the system being shutdown via
+ * ResetSystem().
+ */
+bool efi_poweroff_required(void)
+{
+        return efi_enabled(EFI_RUNTIME_SERVICES);
+}
diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S
index 8ce88e08c030..07f930540f4a 100644
--- a/arch/arm64/kernel/head.S
+++ b/arch/arm64/kernel/head.S
@@ -585,8 +585,8 @@ ENDPROC(set_cpu_boot_mode_flag)
 * zeroing of .bss would clobber it.
 */
        .pushsection    .data..cacheline_aligned
-ENTRY(__boot_cpu_mode)
        .align  L1_CACHE_SHIFT
+ENTRY(__boot_cpu_mode)
        .long   BOOT_CPU_MODE_EL2
        .long   0
        .popsection
diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c
index fde9923af859..c6b1f3b96f45 100644
--- a/arch/arm64/kernel/process.c
+++ b/arch/arm64/kernel/process.c
@@ -21,6 +21,7 @@
 #include <stdarg.h>
 #include <linux/compat.h>
+#include <linux/efi.h>
 #include <linux/export.h>
 #include <linux/sched.h>
 #include <linux/kernel.h>
@@ -150,6 +151,13 @@ void machine_restart(char *cmd)
        local_irq_disable();
        smp_send_stop();
+        /*
+         * UpdateCapsule() depends on the system being reset via
+         * ResetSystem().
+         */
+        if (efi_enabled(EFI_RUNTIME_SERVICES))
+                efi_reboot(reboot_mode, NULL);
        /* Now call the architecture specific reboot code. */
        if (arm_pm_restart)
                arm_pm_restart(reboot_mode, cmd);
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 58e0c2bdde04..ef7d112f5ce0 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c
@@ -51,7 +51,7 @@ static int __init early_coherent_pool(char *p)
 }
 early_param("coherent_pool", early_coherent_pool);
-static void *__alloc_from_pool(size_t size, struct page **ret_page)
+static void *__alloc_from_pool(size_t size, struct page **ret_page, gfp_t flags)
 {
        unsigned long val;
        void *ptr = NULL;
@@ -67,6 +67,8 @@ static void *__alloc_from_pool(size_t size, struct page **ret_page)
                *ret_page = phys_to_page(phys);
                ptr = (void *)val;
+                if (flags & __GFP_ZERO)
+                        memset(ptr, 0, size);
        }
        return ptr;
@@ -101,6 +103,7 @@ static void *__dma_alloc_coherent(struct device *dev, size_t size,
                flags |= GFP_DMA;
        if (IS_ENABLED(CONFIG_DMA_CMA) && (flags & __GFP_WAIT)) {
                struct page *page;
+                void *addr;
                size = PAGE_ALIGN(size);
                page = dma_alloc_from_contiguous(dev, size >> PAGE_SHIFT,
@@ -109,7 +112,10 @@ static void *__dma_alloc_coherent(struct device *dev, size_t size,
                        return NULL;
                *dma_handle = phys_to_dma(dev, page_to_phys(page));
-                return page_address(page);
+                addr = page_address(page);
+                if (flags & __GFP_ZERO)
+                        memset(addr, 0, size);
+                return addr;
        } else {
                return swiotlb_alloc_coherent(dev, size, dma_handle, flags);
        }
@@ -146,7 +152,7 @@ static void *__dma_alloc(struct device *dev, size_t size,
        if (!coherent && !(flags & __GFP_WAIT)) {
                struct page *page = NULL;
-                void *addr = __alloc_from_pool(size, &page);
+                void *addr = __alloc_from_pool(size, &page, flags);
                if (addr)
                        *dma_handle = phys_to_dma(dev, page_to_phys(page));
diff --git a/arch/arm64/mm/pageattr.c b/arch/arm64/mm/pageattr.c
index bb0ea94c4ba1..1d3ec3ddd84b 100644
--- a/arch/arm64/mm/pageattr.c
+++ b/arch/arm64/mm/pageattr.c
@@ -51,7 +51,10 @@ static int change_memory_common(unsigned long addr, int numpages,
                WARN_ON_ONCE(1);
        }
-        if (!is_module_address(start) || !is_module_address(end - 1))
+        if (start < MODULES_VADDR || start >= MODULES_END)
+                return -EINVAL;
+        if (end < MODULES_VADDR || end >= MODULES_END)
                return -EINVAL;
        data.set_mask = set_mask;
diff --git a/arch/c6x/include/asm/pgtable.h b/arch/c6x/include/asm/pgtable.h
index 78d4483ba40c..ec4db6df5e0d 100644
--- a/arch/c6x/include/asm/pgtable.h
+++ b/arch/c6x/include/asm/pgtable.h
@@ -67,6 +67,11 @@ extern unsigned long empty_zero_page;
 */
 #define pgtable_cache_init()   do { } while (0)
+/*
+ * c6x is !MMU, so define the simpliest implementation
+ */
+#define pgprot_writecombine pgprot_noncached
 #include <asm-generic/pgtable.h>
 #endif /* _ASM_C6X_PGTABLE_H */
diff --git a/arch/metag/include/asm/io.h b/arch/metag/include/asm/io.h
index 9359e5048442..d5779b0ec573 100644
--- a/arch/metag/include/asm/io.h
+++ b/arch/metag/include/asm/io.h
@@ -2,6 +2,7 @@
 #define _ASM_METAG_IO_H
 #include <linux/types.h>
+#include <asm/pgtable-bits.h>
 #define IO_SPACE_LIMIT  0
diff --git a/arch/metag/include/asm/pgtable-bits.h b/arch/metag/include/asm/pgtable-bits.h
new file mode 100644
index 000000000000..25ba6729f496
--- /dev/null
+++ b/arch/metag/include/asm/pgtable-bits.h
@@ -0,0 +1,104 @@
+/*
+ * Meta page table definitions.
+ */
+#ifndef _METAG_PGTABLE_BITS_H
+#define _METAG_PGTABLE_BITS_H
+#include <asm/metag_mem.h>
+/*
+ * Definitions for MMU descriptors
+ *
+ * These are the hardware bits in the MMCU pte entries.
+ * Derived from the Meta toolkit headers.
+ */
+#define _PAGE_PRESENT           MMCU_ENTRY_VAL_BIT
+#define _PAGE_WRITE             MMCU_ENTRY_WR_BIT
+#define _PAGE_PRIV              MMCU_ENTRY_PRIV_BIT
+/* Write combine bit - this can cause writes to occur out of order */
+#define _PAGE_WR_COMBINE        MMCU_ENTRY_WRC_BIT
+/* Sys coherent bit - this bit is never used by Linux */
+#define _PAGE_SYS_COHERENT      MMCU_ENTRY_SYS_BIT
+#define _PAGE_ALWAYS_ZERO_1     0x020
+#define _PAGE_CACHE_CTRL0       0x040
+#define _PAGE_CACHE_CTRL1       0x080
+#define _PAGE_ALWAYS_ZERO_2     0x100
+#define _PAGE_ALWAYS_ZERO_3     0x200
+#define _PAGE_ALWAYS_ZERO_4     0x400
+#define _PAGE_ALWAYS_ZERO_5     0x800
+/* These are software bits that we stuff into the gaps in the hardware
+ * pte entries that are not used.  Note, these DO get stored in the actual
+ * hardware, but the hardware just does not use them.
+ */
+#define _PAGE_ACCESSED          _PAGE_ALWAYS_ZERO_1
+#define _PAGE_DIRTY             _PAGE_ALWAYS_ZERO_2
+/* Pages owned, and protected by, the kernel. */
+#define _PAGE_KERNEL            _PAGE_PRIV
+/* No cacheing of this page */
+#define _PAGE_CACHE_WIN0        (MMCU_CWIN_UNCACHED << MMCU_ENTRY_CWIN_S)
+/* burst cacheing - good for data streaming */
+#define _PAGE_CACHE_WIN1        (MMCU_CWIN_BURST << MMCU_ENTRY_CWIN_S)
+/* One cache way per thread */
+#define _PAGE_CACHE_WIN2        (MMCU_CWIN_C1SET << MMCU_ENTRY_CWIN_S)
+/* Full on cacheing */
+#define _PAGE_CACHE_WIN3        (MMCU_CWIN_CACHED << MMCU_ENTRY_CWIN_S)
+#define _PAGE_CACHEABLE         (_PAGE_CACHE_WIN3 | _PAGE_WR_COMBINE)
+/* which bits are used for cache control ... */
+#define _PAGE_CACHE_MASK        (_PAGE_CACHE_CTRL0 | _PAGE_CACHE_CTRL1 | \
+                                 _PAGE_WR_COMBINE)
+/* This is a mask of the bits that pte_modify is allowed to change. */
+#define _PAGE_CHG_MASK          (PAGE_MASK)
+#define _PAGE_SZ_SHIFT          1
+#define _PAGE_SZ_4K             (0x0)
+#define _PAGE_SZ_8K             (0x1 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_16K            (0x2 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_32K            (0x3 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_64K            (0x4 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_128K           (0x5 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_256K           (0x6 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_512K           (0x7 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_1M             (0x8 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_2M             (0x9 << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_4M             (0xa << _PAGE_SZ_SHIFT)
+#define _PAGE_SZ_MASK           (0xf << _PAGE_SZ_SHIFT)
+#if defined(CONFIG_PAGE_SIZE_4K)
+#define _PAGE_SZ                (_PAGE_SZ_4K)
+#elif defined(CONFIG_PAGE_SIZE_8K)
+#define _PAGE_SZ                (_PAGE_SZ_8K)
+#elif defined(CONFIG_PAGE_SIZE_16K)
+#define _PAGE_SZ                (_PAGE_SZ_16K)
+#endif
+#define _PAGE_TABLE             (_PAGE_SZ | _PAGE_PRESENT)
+#if defined(CONFIG_HUGETLB_PAGE_SIZE_8K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_8K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_16K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_16K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_32K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_32K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_64K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_128K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_128K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_256K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K)
+# define _PAGE_SZHUGE           (_PAGE_SZ_512K)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_1M)
+# define _PAGE_SZHUGE           (_PAGE_SZ_1M)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_2M)
+# define _PAGE_SZHUGE           (_PAGE_SZ_2M)
+#elif defined(CONFIG_HUGETLB_PAGE_SIZE_4M)
+# define _PAGE_SZHUGE           (_PAGE_SZ_4M)
+#endif
+#endif /* _METAG_PGTABLE_BITS_H */
diff --git a/arch/metag/include/asm/pgtable.h b/arch/metag/include/asm/pgtable.h
index d0604c0a8702..ffa3a3a2ecad 100644
--- a/arch/metag/include/asm/pgtable.h
+++ b/arch/metag/include/asm/pgtable.h
@@ -5,6 +5,7 @@
 #ifndef _METAG_PGTABLE_H
 #define _METAG_PGTABLE_H
+#include <asm/pgtable-bits.h>
 #include <asm-generic/pgtable-nopmd.h>
 /* Invalid regions on Meta: 0x00000000-0x001FFFFF and 0xFFFF0000-0xFFFFFFFF */
@@ -21,100 +22,6 @@
 #endif
 /*
- * Definitions for MMU descriptors
- *
- * These are the hardware bits in the MMCU pte entries.
- * Derived from the Meta toolkit headers.
- */
-#define _PAGE_PRESENT           MMCU_ENTRY_VAL_BIT
-#define _PAGE_WRITE             MMCU_ENTRY_WR_BIT
-#define _PAGE_PRIV              MMCU_ENTRY_PRIV_BIT
-/* Write combine bit - this can cause writes to occur out of order */
-#define _PAGE_WR_COMBINE        MMCU_ENTRY_WRC_BIT
-/* Sys coherent bit - this bit is never used by Linux */
-#define _PAGE_SYS_COHERENT      MMCU_ENTRY_SYS_BIT
-#define _PAGE_ALWAYS_ZERO_1     0x020
-#define _PAGE_CACHE_CTRL0       0x040
-#define _PAGE_CACHE_CTRL1       0x080
-#define _PAGE_ALWAYS_ZERO_2     0x100
-#define _PAGE_ALWAYS_ZERO_3     0x200
-#define _PAGE_ALWAYS_ZERO_4     0x400
-#define _PAGE_ALWAYS_ZERO_5     0x800
-/* These are software bits that we stuff into the gaps in the hardware
- * pte entries that are not used.  Note, these DO get stored in the actual
- * hardware, but the hardware just does not use them.
- */
-#define _PAGE_ACCESSED          _PAGE_ALWAYS_ZERO_1
-#define _PAGE_DIRTY             _PAGE_ALWAYS_ZERO_2
-/* Pages owned, and protected by, the kernel. */
-#define _PAGE_KERNEL            _PAGE_PRIV
-/* No cacheing of this page */
-#define _PAGE_CACHE_WIN0        (MMCU_CWIN_UNCACHED << MMCU_ENTRY_CWIN_S)
-/* burst cacheing - good for data streaming */
-#define _PAGE_CACHE_WIN1        (MMCU_CWIN_BURST << MMCU_ENTRY_CWIN_S)
-/* One cache way per thread */
-#define _PAGE_CACHE_WIN2        (MMCU_CWIN_C1SET << MMCU_ENTRY_CWIN_S)
-/* Full on cacheing */
-#define _PAGE_CACHE_WIN3        (MMCU_CWIN_CACHED << MMCU_ENTRY_CWIN_S)
-#define _PAGE_CACHEABLE         (_PAGE_CACHE_WIN3 | _PAGE_WR_COMBINE)
-/* which bits are used for cache control ... */
-#define _PAGE_CACHE_MASK        (_PAGE_CACHE_CTRL0 | _PAGE_CACHE_CTRL1 | \
-                                 _PAGE_WR_COMBINE)
-/* This is a mask of the bits that pte_modify is allowed to change. */
-#define _PAGE_CHG_MASK          (PAGE_MASK)
-#define _PAGE_SZ_SHIFT          1
-#define _PAGE_SZ_4K             (0x0)
-#define _PAGE_SZ_8K             (0x1 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_16K            (0x2 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_32K            (0x3 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_64K            (0x4 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_128K           (0x5 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_256K           (0x6 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_512K           (0x7 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_1M             (0x8 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_2M             (0x9 << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_4M             (0xa << _PAGE_SZ_SHIFT)
-#define _PAGE_SZ_MASK           (0xf << _PAGE_SZ_SHIFT)
-#if defined(CONFIG_PAGE_SIZE_4K)
-#define _PAGE_SZ                (_PAGE_SZ_4K)
-#elif defined(CONFIG_PAGE_SIZE_8K)
-#define _PAGE_SZ                (_PAGE_SZ_8K)
-#elif defined(CONFIG_PAGE_SIZE_16K)
-#define _PAGE_SZ                (_PAGE_SZ_16K)
-#endif
-#define _PAGE_TABLE             (_PAGE_SZ | _PAGE_PRESENT)
-#if defined(CONFIG_HUGETLB_PAGE_SIZE_8K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_8K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_16K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_16K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_32K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_32K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_64K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_128K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_128K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_256K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K)
-# define _PAGE_SZHUGE           (_PAGE_SZ_512K)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_1M)
-# define _PAGE_SZHUGE           (_PAGE_SZ_1M)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_2M)
-# define _PAGE_SZHUGE           (_PAGE_SZ_2M)
-#elif defined(CONFIG_HUGETLB_PAGE_SIZE_4M)
-# define _PAGE_SZHUGE           (_PAGE_SZ_4M)
-#endif
-/*
 * The Linux memory management assumes a three-level page table setup. On
 * Meta, we use that, but "fold" the mid level into the top-level page
 * table.
diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S
index 0536bc021cc6..ef548510b951 100644
--- a/arch/microblaze/kernel/entry.S
+++ b/arch/microblaze/kernel/entry.S
@@ -348,8 +348,9 @@ C_ENTRY(_user_exception):
 * The LP register should point to the location where the called function
 * should return.  [note that MAKE_SYS_CALL uses label 1] */
        /* See if the system call number is valid */
+        blti    r12, 5f
        addi    r11, r12, -__NR_syscalls;
-        bgei    r11,5f;
+        bgei    r11, 5f;
        /* Figure out which function to use for this system call.  */
        /* Note Microblaze barrel shift is optional, so don't rely on it */
        add     r12, r12, r12;                  /* convert num -> ptr */
@@ -375,7 +376,7 @@ C_ENTRY(_user_exception):
        /* The syscall number is invalid, return an error.  */
 5:
-        rtsd    r15, 8;         /* looks like a normal subroutine return */
+        braid   ret_from_trap
        addi    r3, r0, -ENOSYS;
 /* Entry point used to return from a syscall/trap */
@@ -411,7 +412,7 @@ C_ENTRY(ret_from_trap):
        bri     1b
        /* Maybe handle a signal */
-5:      
+5:
        andi    r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME;
        beqi    r11, 4f;                /* Signals to handle, handle them */
diff --git a/arch/mips/kvm/tlb.c b/arch/mips/kvm/tlb.c
index bbcd82242059..b6beb0e07b1b 100644
--- a/arch/mips/kvm/tlb.c
+++ b/arch/mips/kvm/tlb.c
@@ -216,6 +216,7 @@ int kvm_mips_host_tlb_write(struct kvm_vcpu *vcpu, unsigned long entryhi,
        if (idx > current_cpu_data.tlbsize) {
                kvm_err("%s: Invalid Index: %d\n", __func__, idx);
                kvm_mips_dump_host_tlbs();
+                local_irq_restore(flags);
                return -1;
        }
diff --git a/arch/mips/kvm/trace.h b/arch/mips/kvm/trace.h
index c1388d40663b..bd6437f67dc0 100644
--- a/arch/mips/kvm/trace.h
+++ b/arch/mips/kvm/trace.h
@@ -24,18 +24,18 @@ TRACE_EVENT(kvm_exit,
            TP_PROTO(struct kvm_vcpu *vcpu, unsigned int reason),
            TP_ARGS(vcpu, reason),
            TP_STRUCT__entry(
-                        __field(struct kvm_vcpu *, vcpu)
+                        __field(unsigned long, pc)
                        __field(unsigned int, reason)
            ),
            TP_fast_assign(
-                        __entry->vcpu = vcpu;
+                        __entry->pc = vcpu->arch.pc;
                        __entry->reason = reason;
            ),
            TP_printk("[%s]PC: 0x%08lx",
                      kvm_mips_exit_types_str[__entry->reason],
-                      __entry->vcpu->arch.pc)
+                      __entry->pc)
 );
 #endif /* _TRACE_KVM_H */
diff --git a/arch/nios2/include/asm/ptrace.h b/arch/nios2/include/asm/ptrace.h
index 20fb1cf2dab6..642462144872 100644
--- a/arch/nios2/include/asm/ptrace.h
+++ b/arch/nios2/include/asm/ptrace.h
@@ -15,7 +15,54 @@
 #include <uapi/asm/ptrace.h>
+/* This struct defines the way the registers are stored on the
+   stack during a system call.  */
 #ifndef __ASSEMBLY__
+struct pt_regs {
+        unsigned long  r8;      /* r8-r15 Caller-saved GP registers */
+        unsigned long  r9;
+        unsigned long  r10;
+        unsigned long  r11;
+        unsigned long  r12;
+        unsigned long  r13;
+        unsigned long  r14;
+        unsigned long  r15;
+        unsigned long  r1;      /* Assembler temporary */
+        unsigned long  r2;      /* Retval LS 32bits */
+        unsigned long  r3;      /* Retval MS 32bits */
+        unsigned long  r4;      /* r4-r7 Register arguments */
+        unsigned long  r5;
+        unsigned long  r6;
+        unsigned long  r7;
+        unsigned long  orig_r2; /* Copy of r2 ?? */
+        unsigned long  ra;      /* Return address */
+        unsigned long  fp;      /* Frame pointer */
+        unsigned long  sp;      /* Stack pointer */
+        unsigned long  gp;      /* Global pointer */
+        unsigned long  estatus;
+        unsigned long  ea;      /* Exception return address (pc) */
+        unsigned long  orig_r7;
+};
+/*
+ * This is the extended stack used by signal handlers and the context
+ * switcher: it's pushed after the normal "struct pt_regs".
+ */
+struct switch_stack {
+        unsigned long  r16;     /* r16-r23 Callee-saved GP registers */
+        unsigned long  r17;
+        unsigned long  r18;
+        unsigned long  r19;
+        unsigned long  r20;
+        unsigned long  r21;
+        unsigned long  r22;
+        unsigned long  r23;
+        unsigned long  fp;
+        unsigned long  gp;
+        unsigned long  ra;
+};
 #define user_mode(regs) (((regs)->estatus & ESTATUS_EU))
 #define instruction_pointer(regs)       ((regs)->ra)
diff --git a/arch/nios2/include/asm/ucontext.h b/arch/nios2/include/asm/ucontext.h
deleted file mode 100644
index 2c87614b0f6e..000000000000
--- a/arch/nios2/include/asm/ucontext.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * Copyright (C) 2010 Tobias Klauser <tklauser@distanz.ch>
- * Copyright (C) 2004 Microtronix Datacom Ltd
- *
- * 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.
- */
-#ifndef _ASM_NIOS2_UCONTEXT_H
-#define _ASM_NIOS2_UCONTEXT_H
-typedef int greg_t;
-#define NGREG 32
-typedef greg_t gregset_t[NGREG];
-struct mcontext {
-        int version;
-        gregset_t gregs;
-};
-#define MCONTEXT_VERSION 2
-struct ucontext {
-        unsigned long     uc_flags;
-        struct ucontext  *uc_link;
-        stack_t           uc_stack;
-        struct mcontext   uc_mcontext;
-        sigset_t          uc_sigmask;   /* mask last for extensibility */
-};
-#endif
diff --git a/arch/nios2/include/uapi/asm/Kbuild b/arch/nios2/include/uapi/asm/Kbuild
index 4f07ca3f8d10..e0bb972a50d7 100644
--- a/arch/nios2/include/uapi/asm/Kbuild
+++ b/arch/nios2/include/uapi/asm/Kbuild
@@ -1,4 +1,5 @@
 include include/uapi/asm-generic/Kbuild.asm
 header-y += elf.h
-header-y += ucontext.h
+generic-y += ucontext.h
diff --git a/arch/nios2/include/uapi/asm/elf.h b/arch/nios2/include/uapi/asm/elf.h
index a5b91ae5cf56..6f06d3b2949e 100644
--- a/arch/nios2/include/uapi/asm/elf.h
+++ b/arch/nios2/include/uapi/asm/elf.h
@@ -50,9 +50,7 @@
 typedef unsigned long elf_greg_t;
-#define ELF_NGREG       \
+#define ELF_NGREG               49
-        ((sizeof(struct pt_regs) + sizeof(struct switch_stack)) /       \
-                sizeof(elf_greg_t))
 typedef elf_greg_t elf_gregset_t[ELF_NGREG];
 typedef unsigned long elf_fpregset_t;
diff --git a/arch/nios2/include/uapi/asm/ptrace.h b/arch/nios2/include/uapi/asm/ptrace.h
index e83a7c9d1c36..71a330597adf 100644
--- a/arch/nios2/include/uapi/asm/ptrace.h
+++ b/arch/nios2/include/uapi/asm/ptrace.h
@@ -67,53 +67,9 @@
 #define NUM_PTRACE_REG (PTR_TLBMISC + 1)
-/* this struct defines the way the registers are stored on the
+/* User structures for general purpose registers.  */
-   stack during a system call.
+struct user_pt_regs {
+        __u32           regs[49];
-   There is a fake_regs in setup.c that has to match pt_regs.*/
-struct pt_regs {
-        unsigned long  r8;              /* r8-r15 Caller-saved GP registers */
-        unsigned long  r9;
-        unsigned long  r10;
-        unsigned long  r11;
-        unsigned long  r12;
-        unsigned long  r13;
-        unsigned long  r14;
-        unsigned long  r15;
-        unsigned long  r1;              /* Assembler temporary */
-        unsigned long  r2;              /* Retval LS 32bits */
-        unsigned long  r3;              /* Retval MS 32bits */
-        unsigned long  r4;              /* r4-r7 Register arguments */
-        unsigned long  r5;
-        unsigned long  r6;
-        unsigned long  r7;
-        unsigned long  orig_r2;         /* Copy of r2 ?? */
-        unsigned long  ra;              /* Return address */
-        unsigned long  fp;              /* Frame pointer */
-        unsigned long  sp;              /* Stack pointer */
-        unsigned long  gp;              /* Global pointer */
-        unsigned long  estatus;
-        unsigned long  ea;              /* Exception return address (pc) */
-        unsigned long  orig_r7;
-};
-/*
- * This is the extended stack used by signal handlers and the context
- * switcher: it's pushed after the normal "struct pt_regs".
- */
-struct switch_stack {
-        unsigned long  r16;             /* r16-r23 Callee-saved GP registers */
-        unsigned long  r17;
-        unsigned long  r18;
-        unsigned long  r19;
-        unsigned long  r20;
-        unsigned long  r21;
-        unsigned long  r22;
-        unsigned long  r23;
-        unsigned long  fp;
-        unsigned long  gp;
-        unsigned long  ra;
 };
 #endif /* __ASSEMBLY__ */
diff --git a/arch/nios2/include/uapi/asm/sigcontext.h b/arch/nios2/include/uapi/asm/sigcontext.h
index 7b8bb41867d4..b67944a50927 100644
--- a/arch/nios2/include/uapi/asm/sigcontext.h
+++ b/arch/nios2/include/uapi/asm/sigcontext.h
@@ -15,14 +15,16 @@
 * details.
 */
-#ifndef _ASM_NIOS2_SIGCONTEXT_H
+#ifndef _UAPI__ASM_SIGCONTEXT_H
-#define _ASM_NIOS2_SIGCONTEXT_H
+#define _UAPI__ASM_SIGCONTEXT_H
-#include <asm/ptrace.h>
+#include <linux/types.h>
+#define MCONTEXT_VERSION 2
 struct sigcontext {
-        struct pt_regs regs;
+        int version;
-        unsigned long  sc_mask; /* old sigmask */
+        unsigned long gregs[32];
 };
 #endif
diff --git a/arch/nios2/kernel/signal.c b/arch/nios2/kernel/signal.c
index 2d0ea25be171..dda41e4fe707 100644
--- a/arch/nios2/kernel/signal.c
+++ b/arch/nios2/kernel/signal.c
@@ -39,7 +39,7 @@ static inline int rt_restore_ucontext(struct pt_regs *regs,
                                        struct ucontext *uc, int *pr2)
 {
        int temp;
-        greg_t *gregs = uc->uc_mcontext.gregs;
+        unsigned long *gregs = uc->uc_mcontext.gregs;
        int err;
        /* Always make any pending restarted system calls return -EINTR */
@@ -127,7 +127,7 @@ badframe:
 static inline int rt_setup_ucontext(struct ucontext *uc, struct pt_regs *regs)
 {
        struct switch_stack *sw = (struct switch_stack *)regs - 1;
-        greg_t *gregs = uc->uc_mcontext.gregs;
+        unsigned long *gregs = uc->uc_mcontext.gregs;
        int err = 0;
        err |= __put_user(MCONTEXT_VERSION, &uc->uc_mcontext.version);
diff --git a/arch/nios2/mm/fault.c b/arch/nios2/mm/fault.c
index 0d231adfe576..0c9b6afe69e9 100644
--- a/arch/nios2/mm/fault.c
+++ b/arch/nios2/mm/fault.c
@@ -126,7 +126,6 @@ good_area:
                break;
        }
-survive:
        /*
         * If for any reason at all we couldn't handle the fault,
         * make sure we exit gracefully rather than endlessly redo
@@ -220,11 +219,6 @@ no_context:
 */
 out_of_memory:
        up_read(&mm->mmap_sem);
-        if (is_global_init(tsk)) {
-                yield();
-                down_read(&mm->mmap_sem);
-                goto survive;
-        }
        if (!user_mode(regs))
                goto no_context;
        pagefault_out_of_memory();
diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h
index 9cfa3706a1b8..f1ea5972f6ec 100644
--- a/arch/powerpc/include/asm/iommu.h
+++ b/arch/powerpc/include/asm/iommu.h
@@ -113,6 +113,7 @@ extern void iommu_register_group(struct iommu_table *tbl,
                                 int pci_domain_number, unsigned long pe_num);
 extern int iommu_add_device(struct device *dev);
 extern void iommu_del_device(struct device *dev);
+extern int __init tce_iommu_bus_notifier_init(void);
 #else
 static inline void iommu_register_group(struct iommu_table *tbl,
                                        int pci_domain_number,
@@ -128,6 +129,11 @@ static inline int iommu_add_device(struct device *dev)
 static inline void iommu_del_device(struct device *dev)
 {
 }
+static inline int __init tce_iommu_bus_notifier_init(void)
+{
+        return 0;
+}
 #endif /* !CONFIG_IOMMU_API */
 static inline void set_iommu_table_base_and_group(struct device *dev,
diff --git a/arch/powerpc/include/asm/irq_work.h b/arch/powerpc/include/asm/irq_work.h
new file mode 100644
index 000000000000..744fd54de374
--- /dev/null
+++ b/arch/powerpc/include/asm/irq_work.h
@@ -0,0 +1,9 @@
+#ifndef _ASM_POWERPC_IRQ_WORK_H
+#define _ASM_POWERPC_IRQ_WORK_H
+static inline bool arch_irq_work_has_interrupt(void)
+{
+        return true;
+}
+#endif /* _ASM_POWERPC_IRQ_WORK_H */
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index 03cd858a401c..4cbe23af400a 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -153,6 +153,7 @@
 #define PPC_INST_MFSPR_PVR_MASK         0xfc1fffff
 #define PPC_INST_MFTMR                  0x7c0002dc
 #define PPC_INST_MSGSND                 0x7c00019c
+#define PPC_INST_MSGCLR                 0x7c0001dc
 #define PPC_INST_MSGSNDP                0x7c00011c
 #define PPC_INST_MTTMR                  0x7c0003dc
 #define PPC_INST_NOP                    0x60000000
@@ -309,6 +310,8 @@
                                        ___PPC_RB(b) | __PPC_EH(eh))
 #define PPC_MSGSND(b)           stringify_in_c(.long PPC_INST_MSGSND | \
                                        ___PPC_RB(b))
+#define PPC_MSGCLR(b)           stringify_in_c(.long PPC_INST_MSGCLR | \
+                                        ___PPC_RB(b))
 #define PPC_MSGSNDP(b)          stringify_in_c(.long PPC_INST_MSGSNDP | \
                                        ___PPC_RB(b))
 #define PPC_POPCNTB(a, s)       stringify_in_c(.long PPC_INST_POPCNTB | \
diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h
index 1c874fb533bb..af56b5c6c81a 100644
--- a/arch/powerpc/include/asm/reg.h
+++ b/arch/powerpc/include/asm/reg.h
@@ -608,13 +608,16 @@
 #define   SRR1_ISI_N_OR_G       0x10000000 /* ISI: Access is no-exec or G */
 #define   SRR1_ISI_PROT         0x08000000 /* ISI: Other protection fault */
 #define   SRR1_WAKEMASK         0x00380000 /* reason for wakeup */
+#define   SRR1_WAKEMASK_P8      0x003c0000 /* reason for wakeup on POWER8 */
 #define   SRR1_WAKESYSERR       0x00300000 /* System error */
 #define   SRR1_WAKEEE           0x00200000 /* External interrupt */
 #define   SRR1_WAKEMT           0x00280000 /* mtctrl */
 #define   SRR1_WAKEHMI          0x00280000 /* Hypervisor maintenance */
 #define   SRR1_WAKEDEC          0x00180000 /* Decrementer interrupt */
+#define   SRR1_WAKEDBELL        0x00140000 /* Privileged doorbell on P8 */
 #define   SRR1_WAKETHERM        0x00100000 /* Thermal management interrupt */
 #define   SRR1_WAKERESET        0x00100000 /* System reset */
+#define   SRR1_WAKEHDBELL       0x000c0000 /* Hypervisor doorbell on P8 */
 #define   SRR1_WAKESTATE        0x00030000 /* Powersave exit mask [46:47] */
 #define   SRR1_WS_DEEPEST       0x00030000 /* Some resources not maintained,
                                          * may not be recoverable */
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index f337666768a7..f83046878336 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -437,6 +437,26 @@ static struct cpu_spec __initdata cpu_specs[] = {
                .machine_check_early    = __machine_check_early_realmode_p8,
                .platform               = "power8",
        },
+        {       /* Power8NVL */
+                .pvr_mask               = 0xffff0000,
+                .pvr_value              = 0x004c0000,
+                .cpu_name               = "POWER8NVL (raw)",
+                .cpu_features           = CPU_FTRS_POWER8,
+                .cpu_user_features      = COMMON_USER_POWER8,
+                .cpu_user_features2     = COMMON_USER2_POWER8,
+                .mmu_features           = MMU_FTRS_POWER8,
+                .icache_bsize           = 128,
+                .dcache_bsize           = 128,
+                .num_pmcs               = 6,
+                .pmc_type               = PPC_PMC_IBM,
+                .oprofile_cpu_type      = "ppc64/power8",
+                .oprofile_type          = PPC_OPROFILE_INVALID,
+                .cpu_setup              = __setup_cpu_power8,
+                .cpu_restore            = __restore_cpu_power8,
+                .flush_tlb              = __flush_tlb_power8,
+                .machine_check_early    = __machine_check_early_realmode_p8,
+                .platform               = "power8",
+        },
        {       /* Power8 DD1: Does not support doorbell IPIs */
                .pvr_mask               = 0xffffff00,
                .pvr_value              = 0x004d0100,
diff --git a/arch/powerpc/kernel/dbell.c b/arch/powerpc/kernel/dbell.c
index f4217819cc31..2128f3a96c32 100644
--- a/arch/powerpc/kernel/dbell.c
+++ b/arch/powerpc/kernel/dbell.c
@@ -17,6 +17,7 @@
 #include <asm/dbell.h>
 #include <asm/irq_regs.h>
+#include <asm/kvm_ppc.h>
 #ifdef CONFIG_SMP
 void doorbell_setup_this_cpu(void)
@@ -41,6 +42,7 @@ void doorbell_exception(struct pt_regs *regs)
        may_hard_irq_enable();
+        kvmppc_set_host_ipi(smp_processor_id(), 0);
        __this_cpu_inc(irq_stat.doorbell_irqs);
        smp_ipi_demux();
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index c2df8150bd7a..9519e6bdc6d7 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1408,7 +1408,7 @@ machine_check_handle_early:
        bne     9f                      /* continue in V mode if we are. */
 5:
-#ifdef CONFIG_KVM_BOOK3S_64_HV
+#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
        /*
         * We are coming from kernel context. Check if we are coming from
         * guest. if yes, then we can continue. We will fall through
diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index 5d3968c4d799..b054f33ab1fb 100644
--- a/arch/powerpc/kernel/iommu.c
+++ b/arch/powerpc/kernel/iommu.c
@@ -1175,4 +1175,30 @@ void iommu_del_device(struct device *dev)
 }
 EXPORT_SYMBOL_GPL(iommu_del_device);
+static int tce_iommu_bus_notifier(struct notifier_block *nb,
+                unsigned long action, void *data)
+{
+        struct device *dev = data;
+        switch (action) {
+        case BUS_NOTIFY_ADD_DEVICE:
+                return iommu_add_device(dev);
+        case BUS_NOTIFY_DEL_DEVICE:
+                if (dev->iommu_group)
+                        iommu_del_device(dev);
+                return 0;
+        default:
+                return 0;
+        }
+}
+static struct notifier_block tce_iommu_bus_nb = {
+        .notifier_call = tce_iommu_bus_notifier,
+};
+int __init tce_iommu_bus_notifier_init(void)
+{
+        bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
+        return 0;
+}
 #endif /* CONFIG_IOMMU_API */
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 6e19afa35a15..ec9ec2058d2d 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -541,8 +541,8 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle)
        if (smp_ops->give_timebase)
                smp_ops->give_timebase();
-        /* Wait until cpu puts itself in the online map */
+        /* Wait until cpu puts itself in the online & active maps */
-        while (!cpu_online(cpu))
+        while (!cpu_online(cpu) || !cpu_active(cpu))
                cpu_relax();
        return 0;
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index e69142f4af08..54323d6b5166 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -836,30 +836,4 @@ void __init pnv_pci_init(void)
 #endif
 }
-static int tce_iommu_bus_notifier(struct notifier_block *nb,
-                unsigned long action, void *data)
-{
-        struct device *dev = data;
-        switch (action) {
-        case BUS_NOTIFY_ADD_DEVICE:
-                return iommu_add_device(dev);
-        case BUS_NOTIFY_DEL_DEVICE:
-                if (dev->iommu_group)
-                        iommu_del_device(dev);
-                return 0;
-        default:
-                return 0;
-        }
-}
-static struct notifier_block tce_iommu_bus_nb = {
-        .notifier_call = tce_iommu_bus_notifier,
-};
-static int __init tce_iommu_bus_notifier_init(void)
-{
-        bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb);
-        return 0;
-}
 machine_subsys_initcall_sync(powernv, tce_iommu_bus_notifier_init);
diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c
index fc34025ef822..38a45088f633 100644
--- a/arch/powerpc/platforms/powernv/smp.c
+++ b/arch/powerpc/platforms/powernv/smp.c
@@ -33,6 +33,8 @@
 #include <asm/runlatch.h>
 #include <asm/code-patching.h>
 #include <asm/dbell.h>
+#include <asm/kvm_ppc.h>
+#include <asm/ppc-opcode.h>
 #include "powernv.h"
@@ -149,7 +151,7 @@ static int pnv_smp_cpu_disable(void)
 static void pnv_smp_cpu_kill_self(void)
 {
        unsigned int cpu;
-        unsigned long srr1;
+        unsigned long srr1, wmask;
        u32 idle_states;
        /* Standard hot unplug procedure */
@@ -161,6 +163,10 @@ static void pnv_smp_cpu_kill_self(void)
        generic_set_cpu_dead(cpu);
        smp_wmb();
+        wmask = SRR1_WAKEMASK;
+        if (cpu_has_feature(CPU_FTR_ARCH_207S))
+                wmask = SRR1_WAKEMASK_P8;
        idle_states = pnv_get_supported_cpuidle_states();
        /* We don't want to take decrementer interrupts while we are offline,
         * so clear LPCR:PECE1. We keep PECE2 enabled.
@@ -191,10 +197,14 @@ static void pnv_smp_cpu_kill_self(void)
                 * having finished executing in a KVM guest, then srr1
                 * contains 0.
                 */
-                if ((srr1 & SRR1_WAKEMASK) == SRR1_WAKEEE) {
+                if ((srr1 & wmask) == SRR1_WAKEEE) {
                        icp_native_flush_interrupt();
                        local_paca->irq_happened &= PACA_IRQ_HARD_DIS;
                        smp_mb();
+                } else if ((srr1 & wmask) == SRR1_WAKEHDBELL) {
+                        unsigned long msg = PPC_DBELL_TYPE(PPC_DBELL_SERVER);
+                        asm volatile(PPC_MSGCLR(%0) : : "r" (msg));
+                        kvmppc_set_host_ipi(cpu, 0);
                }
                if (cpu_core_split_required())
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index 1d3d52dc3ff3..7803a19adb31 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -1340,3 +1340,5 @@ static int __init disable_multitce(char *str)
 }
 __setup("multitce=", disable_multitce);
+machine_subsys_initcall_sync(pseries, tce_iommu_bus_notifier_init);
diff --git a/arch/powerpc/platforms/pseries/mobility.c b/arch/powerpc/platforms/pseries/mobility.c
index 90cf3dcbd9f2..8f35d525cede 100644
--- a/arch/powerpc/platforms/pseries/mobility.c
+++ b/arch/powerpc/platforms/pseries/mobility.c
@@ -25,10 +25,10 @@
 static struct kobject *mobility_kobj;
 struct update_props_workarea {
-        u32 phandle;
+        __be32 phandle;
-        u32 state;
+        __be32 state;
-        u64 reserved;
+        __be64 reserved;
-        u32 nprops;
+        __be32 nprops;
 } __packed;
 #define NODE_ACTION_MASK        0xff000000
@@ -54,11 +54,11 @@ static int mobility_rtas_call(int token, char *buf, s32 scope)
        return rc;
 }
-static int delete_dt_node(u32 phandle)
+static int delete_dt_node(__be32 phandle)
 {
        struct device_node *dn;
-        dn = of_find_node_by_phandle(phandle);
+        dn = of_find_node_by_phandle(be32_to_cpu(phandle));
        if (!dn)
                return -ENOENT;
@@ -127,7 +127,7 @@ static int update_dt_property(struct device_node *dn, struct property **prop,
        return 0;
 }
-static int update_dt_node(u32 phandle, s32 scope)
+static int update_dt_node(__be32 phandle, s32 scope)
 {
        struct update_props_workarea *upwa;
        struct device_node *dn;
@@ -136,6 +136,7 @@ static int update_dt_node(u32 phandle, s32 scope)
        char *prop_data;
        char *rtas_buf;
        int update_properties_token;
+        u32 nprops;
        u32 vd;
        update_properties_token = rtas_token("ibm,update-properties");
@@ -146,7 +147,7 @@ static int update_dt_node(u32 phandle, s32 scope)
        if (!rtas_buf)
                return -ENOMEM;
-        dn = of_find_node_by_phandle(phandle);
+        dn = of_find_node_by_phandle(be32_to_cpu(phandle));
        if (!dn) {
                kfree(rtas_buf);
                return -ENOENT;
@@ -162,6 +163,7 @@ static int update_dt_node(u32 phandle, s32 scope)
                        break;
                prop_data = rtas_buf + sizeof(*upwa);
+                nprops = be32_to_cpu(upwa->nprops);
                /* On the first call to ibm,update-properties for a node the
                 * the first property value descriptor contains an empty
@@ -170,17 +172,17 @@ static int update_dt_node(u32 phandle, s32 scope)
                 */
                if (*prop_data == 0) {
                        prop_data++;
-                        vd = *(u32 *)prop_data;
+                        vd = be32_to_cpu(*(__be32 *)prop_data);
                        prop_data += vd + sizeof(vd);
-                        upwa->nprops--;
+                        nprops--;
                }
-                for (i = 0; i < upwa->nprops; i++) {
+                for (i = 0; i < nprops; i++) {
                        char *prop_name;
                        prop_name = prop_data;
                        prop_data += strlen(prop_name) + 1;
-                        vd = *(u32 *)prop_data;
+                        vd = be32_to_cpu(*(__be32 *)prop_data);
                        prop_data += sizeof(vd);
                        switch (vd) {
@@ -212,13 +214,13 @@ static int update_dt_node(u32 phandle, s32 scope)
        return 0;
 }
-static int add_dt_node(u32 parent_phandle, u32 drc_index)
+static int add_dt_node(__be32 parent_phandle, __be32 drc_index)
 {
        struct device_node *dn;
        struct device_node *parent_dn;
        int rc;
-        parent_dn = of_find_node_by_phandle(parent_phandle);
+        parent_dn = of_find_node_by_phandle(be32_to_cpu(parent_phandle));
        if (!parent_dn)
                return -ENOENT;
@@ -237,7 +239,7 @@ static int add_dt_node(u32 parent_phandle, u32 drc_index)
 int pseries_devicetree_update(s32 scope)
 {
        char *rtas_buf;
-        u32 *data;
+        __be32 *data;
        int update_nodes_token;
        int rc;
@@ -254,17 +256,17 @@ int pseries_devicetree_update(s32 scope)
                if (rc && rc != 1)
                        break;
-                data = (u32 *)rtas_buf + 4;
+                data = (__be32 *)rtas_buf + 4;
-                while (*data & NODE_ACTION_MASK) {
+                while (be32_to_cpu(*data) & NODE_ACTION_MASK) {
                        int i;
-                        u32 action = *data & NODE_ACTION_MASK;
+                        u32 action = be32_to_cpu(*data) & NODE_ACTION_MASK;
-                        int node_count = *data & NODE_COUNT_MASK;
+                        u32 node_count = be32_to_cpu(*data) & NODE_COUNT_MASK;
                        data++;
                        for (i = 0; i < node_count; i++) {
-                                u32 phandle = *data++;
+                                __be32 phandle = *data++;
-                                u32 drc_index;
+                                __be32 drc_index;
                                switch (action) {
                                case DELETE_DT_NODE:
diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h
index c9df40b5c0ac..c9c875d9ed31 100644
--- a/arch/s390/include/asm/elf.h
+++ b/arch/s390/include/asm/elf.h
@@ -211,7 +211,7 @@ do {								\
 extern unsigned long mmap_rnd_mask;
-#define STACK_RND_MASK  (mmap_rnd_mask)
+#define STACK_RND_MASK  (test_thread_flag(TIF_31BIT) ? 0x7ff : mmap_rnd_mask)
 #define ARCH_DLINFO                                                         \
 do {                                                                        \
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index d84559e31f32..f407bbf5ee94 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -515,15 +515,15 @@ struct s390_io_adapter {
 #define S390_ARCH_FAC_MASK_SIZE_U64 \
        (S390_ARCH_FAC_MASK_SIZE_BYTE / sizeof(u64))
-struct s390_model_fac {
+struct kvm_s390_fac {
-        /* facilities used in SIE context */
+        /* facility list requested by guest */
-        __u64 sie[S390_ARCH_FAC_LIST_SIZE_U64];
+        __u64 list[S390_ARCH_FAC_LIST_SIZE_U64];
-        /* subset enabled by kvm */
+        /* facility mask supported by kvm & hosting machine */
-        __u64 kvm[S390_ARCH_FAC_LIST_SIZE_U64];
+        __u64 mask[S390_ARCH_FAC_LIST_SIZE_U64];
 };
 struct kvm_s390_cpu_model {
-        struct s390_model_fac *fac;
+        struct kvm_s390_fac *fac;
        struct cpuid cpu_id;
        unsigned short ibc;
 };
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index f49b71954654..8fb3802f8fad 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -62,6 +62,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 {
        int cpu = smp_processor_id();
+        S390_lowcore.user_asce = next->context.asce_bits | __pa(next->pgd);
        if (prev == next)
                return;
        if (MACHINE_HAS_TLB_LC)
@@ -73,7 +74,6 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
        atomic_dec(&prev->context.attach_count);
        if (MACHINE_HAS_TLB_LC)
                cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask);
-        S390_lowcore.user_asce = next->context.asce_bits | __pa(next->pgd);
 }
 #define finish_arch_post_lock_switch finish_arch_post_lock_switch
diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h
index 7b2ac6e44166..53eacbd4f09b 100644
--- a/arch/s390/include/asm/page.h
+++ b/arch/s390/include/asm/page.h
@@ -37,16 +37,7 @@ static inline void storage_key_init_range(unsigned long start, unsigned long end
 #endif
 }
-static inline void clear_page(void *page)
+#define clear_page(page)        memset((page), 0, PAGE_SIZE)
-{
-        register unsigned long reg1 asm ("1") = 0;
-        register void *reg2 asm ("2") = page;
-        register unsigned long reg3 asm ("3") = 4096;
-        asm volatile(
-                "       mvcl    2,0"
-                : "+d" (reg2), "+d" (reg3) : "d" (reg1)
-                : "memory", "cc");
-}
 /*
 * copy_page uses the mvcl instruction with 0xb0 padding byte in order to
diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c
index 82c19899574f..6c79f1b44fe7 100644
--- a/arch/s390/kernel/ftrace.c
+++ b/arch/s390/kernel/ftrace.c
@@ -57,6 +57,44 @@
 unsigned long ftrace_plt;
+static inline void ftrace_generate_orig_insn(struct ftrace_insn *insn)
+{
+#ifdef CC_USING_HOTPATCH
+        /* brcl 0,0 */
+        insn->opc = 0xc004;
+        insn->disp = 0;
+#else
+        /* stg r14,8(r15) */
+        insn->opc = 0xe3e0;
+        insn->disp = 0xf0080024;
+#endif
+}
+static inline int is_kprobe_on_ftrace(struct ftrace_insn *insn)
+{
+#ifdef CONFIG_KPROBES
+        if (insn->opc == BREAKPOINT_INSTRUCTION)
+                return 1;
+#endif
+        return 0;
+}
+static inline void ftrace_generate_kprobe_nop_insn(struct ftrace_insn *insn)
+{
+#ifdef CONFIG_KPROBES
+        insn->opc = BREAKPOINT_INSTRUCTION;
+        insn->disp = KPROBE_ON_FTRACE_NOP;
+#endif
+}
+static inline void ftrace_generate_kprobe_call_insn(struct ftrace_insn *insn)
+{
+#ifdef CONFIG_KPROBES
+        insn->opc = BREAKPOINT_INSTRUCTION;
+        insn->disp = KPROBE_ON_FTRACE_CALL;
+#endif
+}
 int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
                       unsigned long addr)
 {
@@ -72,16 +110,9 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
                return -EFAULT;
        if (addr == MCOUNT_ADDR) {
                /* Initial code replacement */
-#ifdef CC_USING_HOTPATCH
+                ftrace_generate_orig_insn(&orig);
-                /* We expect to see brcl 0,0 */
-                ftrace_generate_nop_insn(&orig);
-#else
-                /* We expect to see stg r14,8(r15) */
-                orig.opc = 0xe3e0;
-                orig.disp = 0xf0080024;
-#endif
                ftrace_generate_nop_insn(&new);
-        } else if (old.opc == BREAKPOINT_INSTRUCTION) {
+        } else if (is_kprobe_on_ftrace(&old)) {
                /*
                 * If we find a breakpoint instruction, a kprobe has been
                 * placed at the beginning of the function. We write the
@@ -89,9 +120,8 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
                 * bytes of the original instruction so that the kprobes
                 * handler can execute a nop, if it reaches this breakpoint.
                 */
-                new.opc = orig.opc = BREAKPOINT_INSTRUCTION;
+                ftrace_generate_kprobe_call_insn(&orig);
-                orig.disp = KPROBE_ON_FTRACE_CALL;
+                ftrace_generate_kprobe_nop_insn(&new);
-                new.disp = KPROBE_ON_FTRACE_NOP;
        } else {
                /* Replace ftrace call with a nop. */
                ftrace_generate_call_insn(&orig, rec->ip);
@@ -111,7 +141,7 @@ int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
        if (probe_kernel_read(&old, (void *) rec->ip, sizeof(old)))
                return -EFAULT;
-        if (old.opc == BREAKPOINT_INSTRUCTION) {
+        if (is_kprobe_on_ftrace(&old)) {
                /*
                 * If we find a breakpoint instruction, a kprobe has been
                 * placed at the beginning of the function. We write the
@@ -119,9 +149,8 @@ int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
                 * bytes of the original instruction so that the kprobes
                 * handler can execute a brasl if it reaches this breakpoint.
                 */
-                new.opc = orig.opc = BREAKPOINT_INSTRUCTION;
+                ftrace_generate_kprobe_nop_insn(&orig);
-                orig.disp = KPROBE_ON_FTRACE_NOP;
+                ftrace_generate_kprobe_call_insn(&new);
-                new.disp = KPROBE_ON_FTRACE_CALL;
        } else {
                /* Replace nop with an ftrace call. */
                ftrace_generate_nop_insn(&orig);
diff --git a/arch/s390/kernel/jump_label.c b/arch/s390/kernel/jump_label.c
index cb2d51e779df..830066f936c8 100644
--- a/arch/s390/kernel/jump_label.c
+++ b/arch/s390/kernel/jump_label.c
@@ -36,16 +36,20 @@ static void jump_label_make_branch(struct jump_entry *entry, struct insn *insn)
        insn->offset = (entry->target - entry->code) >> 1;
 }
-static void jump_label_bug(struct jump_entry *entry, struct insn *insn)
+static void jump_label_bug(struct jump_entry *entry, struct insn *expected,
+                           struct insn *new)
 {
        unsigned char *ipc = (unsigned char *)entry->code;
-        unsigned char *ipe = (unsigned char *)insn;
+        unsigned char *ipe = (unsigned char *)expected;
+        unsigned char *ipn = (unsigned char *)new;
        pr_emerg("Jump label code mismatch at %pS [%p]\n", ipc, ipc);
        pr_emerg("Found:    %02x %02x %02x %02x %02x %02x\n",
                 ipc[0], ipc[1], ipc[2], ipc[3], ipc[4], ipc[5]);
        pr_emerg("Expected: %02x %02x %02x %02x %02x %02x\n",
                 ipe[0], ipe[1], ipe[2], ipe[3], ipe[4], ipe[5]);
+        pr_emerg("New:      %02x %02x %02x %02x %02x %02x\n",
+                 ipn[0], ipn[1], ipn[2], ipn[3], ipn[4], ipn[5]);
        panic("Corrupted kernel text");
 }
@@ -69,10 +73,10 @@ static void __jump_label_transform(struct jump_entry *entry,
        }
        if (init) {
                if (memcmp((void *)entry->code, &orignop, sizeof(orignop)))
-                        jump_label_bug(entry, &old);
+                        jump_label_bug(entry, &orignop, &new);
        } else {
                if (memcmp((void *)entry->code, &old, sizeof(old)))
-                        jump_label_bug(entry, &old);
+                        jump_label_bug(entry, &old, &new);
        }
        probe_kernel_write((void *)entry->code, &new, sizeof(new));
 }
diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c
index 36154a2f1814..2ca95862e336 100644
--- a/arch/s390/kernel/module.c
+++ b/arch/s390/kernel/module.c
@@ -436,6 +436,7 @@ int module_finalize(const Elf_Ehdr *hdr,
                    const Elf_Shdr *sechdrs,
                    struct module *me)
 {
+        jump_label_apply_nops(me);
        vfree(me->arch.syminfo);
        me->arch.syminfo = NULL;
        return 0;
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index c3f8d157cb0d..e6a1578fc000 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1415,7 +1415,7 @@ CPUMF_EVENT_ATTR(SF, SF_CYCLES_BASIC_DIAG, PERF_EVENT_CPUM_SF_DIAG);
 static struct attribute *cpumsf_pmu_events_attr[] = {
        CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC),
-        CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC_DIAG),
+        NULL,
        NULL,
 };
@@ -1606,8 +1606,11 @@ static int __init init_cpum_sampling_pmu(void)
                return -EINVAL;
        }
-        if (si.ad)
+        if (si.ad) {
                sfb_set_limits(CPUM_SF_MIN_SDB, CPUM_SF_MAX_SDB);
+                cpumsf_pmu_events_attr[1] =
+                        CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC_DIAG);
+        }
        sfdbg = debug_register(KMSG_COMPONENT, 2, 1, 80);
        if (!sfdbg)
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index 26108232fcaa..dc488e13b7e3 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -18,7 +18,7 @@
 static DEFINE_PER_CPU(struct cpuid, cpu_id);
-void cpu_relax(void)
+void notrace cpu_relax(void)
 {
        if (!smp_cpu_mtid && MACHINE_HAS_DIAG44)
                asm volatile("diag 0,0,0x44");
diff --git a/arch/s390/kernel/swsusp_asm64.S b/arch/s390/kernel/swsusp_asm64.S
index 6b09fdffbd2f..ca6294645dd3 100644
--- a/arch/s390/kernel/swsusp_asm64.S
+++ b/arch/s390/kernel/swsusp_asm64.S
@@ -177,6 +177,17 @@ restart_entry:
        lhi     %r1,1
        sigp    %r1,%r0,SIGP_SET_ARCHITECTURE
        sam64
+#ifdef CONFIG_SMP
+        larl    %r1,smp_cpu_mt_shift
+        icm     %r1,15,0(%r1)
+        jz      smt_done
+        llgfr   %r1,%r1
+smt_loop:
+        sigp    %r1,%r0,SIGP_SET_MULTI_THREADING
+        brc     8,smt_done                      /* accepted */
+        brc     2,smt_loop                      /* busy, try again */
+smt_done:
+#endif
        larl    %r1,.Lnew_pgm_check_psw
        lpswe   0(%r1)
 pgm_check_entry:
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0c3623927563..19e17bd7aec0 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -165,7 +165,6 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
        case KVM_CAP_ONE_REG:
        case KVM_CAP_ENABLE_CAP:
        case KVM_CAP_S390_CSS_SUPPORT:
-        case KVM_CAP_IRQFD:
        case KVM_CAP_IOEVENTFD:
        case KVM_CAP_DEVICE_CTRL:
        case KVM_CAP_ENABLE_CAP_VM:
@@ -522,7 +521,7 @@ static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
                memcpy(&kvm->arch.model.cpu_id, &proc->cpuid,
                       sizeof(struct cpuid));
                kvm->arch.model.ibc = proc->ibc;
-                memcpy(kvm->arch.model.fac->kvm, proc->fac_list,
+                memcpy(kvm->arch.model.fac->list, proc->fac_list,
                       S390_ARCH_FAC_LIST_SIZE_BYTE);
        } else
                ret = -EFAULT;
@@ -556,7 +555,7 @@ static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr)
        }
        memcpy(&proc->cpuid, &kvm->arch.model.cpu_id, sizeof(struct cpuid));
        proc->ibc = kvm->arch.model.ibc;
-        memcpy(&proc->fac_list, kvm->arch.model.fac->kvm, S390_ARCH_FAC_LIST_SIZE_BYTE);
+        memcpy(&proc->fac_list, kvm->arch.model.fac->list, S390_ARCH_FAC_LIST_SIZE_BYTE);
        if (copy_to_user((void __user *)attr->addr, proc, sizeof(*proc)))
                ret = -EFAULT;
        kfree(proc);
@@ -576,10 +575,10 @@ static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr)
        }
        get_cpu_id((struct cpuid *) &mach->cpuid);
        mach->ibc = sclp_get_ibc();
-        memcpy(&mach->fac_mask, kvm_s390_fac_list_mask,
+        memcpy(&mach->fac_mask, kvm->arch.model.fac->mask,
-               kvm_s390_fac_list_mask_size() * sizeof(u64));
+               S390_ARCH_FAC_LIST_SIZE_BYTE);
        memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
-               S390_ARCH_FAC_LIST_SIZE_U64);
+               S390_ARCH_FAC_LIST_SIZE_BYTE);
        if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
                ret = -EFAULT;
        kfree(mach);
@@ -778,15 +777,18 @@ long kvm_arch_vm_ioctl(struct file *filp,
 static int kvm_s390_query_ap_config(u8 *config)
 {
        u32 fcn_code = 0x04000000UL;
-        u32 cc;
+        u32 cc = 0;
+        memset(config, 0, 128);
        asm volatile(
                "lgr 0,%1\n"
                "lgr 2,%2\n"
                ".long 0xb2af0000\n"            /* PQAP(QCI) */
-                "ipm %0\n"
+                "0: ipm %0\n"
                "srl %0,28\n"
-                : "=r" (cc)
+                "1:\n"
+                EX_TABLE(0b, 1b)
+                : "+r" (cc)
                : "r" (fcn_code), "r" (config)
                : "cc", "0", "2", "memory"
        );
@@ -839,9 +841,13 @@ static int kvm_s390_crypto_init(struct kvm *kvm)
        kvm_s390_set_crycb_format(kvm);
-        /* Disable AES/DEA protected key functions by default */
+        /* Enable AES/DEA protected key functions by default */
-        kvm->arch.crypto.aes_kw = 0;
+        kvm->arch.crypto.aes_kw = 1;
-        kvm->arch.crypto.dea_kw = 0;
+        kvm->arch.crypto.dea_kw = 1;
+        get_random_bytes(kvm->arch.crypto.crycb->aes_wrapping_key_mask,
+                         sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
+        get_random_bytes(kvm->arch.crypto.crycb->dea_wrapping_key_mask,
+                         sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
        return 0;
 }
@@ -886,40 +892,29 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
        /*
         * The architectural maximum amount of facilities is 16 kbit. To store
         * this amount, 2 kbyte of memory is required. Thus we need a full
-         * page to hold the active copy (arch.model.fac->sie) and the current
+         * page to hold the guest facility list (arch.model.fac->list) and the
-         * facilities set (arch.model.fac->kvm). Its address size has to be
+         * facility mask (arch.model.fac->mask). Its address size has to be
         * 31 bits and word aligned.
         */
        kvm->arch.model.fac =
-                (struct s390_model_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+                (struct kvm_s390_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
        if (!kvm->arch.model.fac)
                goto out_nofac;
-        memcpy(kvm->arch.model.fac->kvm, S390_lowcore.stfle_fac_list,
+        /* Populate the facility mask initially. */
-               S390_ARCH_FAC_LIST_SIZE_U64);
+        memcpy(kvm->arch.model.fac->mask, S390_lowcore.stfle_fac_list,
+               S390_ARCH_FAC_LIST_SIZE_BYTE);
-        /*
-         * If this KVM host runs *not* in a LPAR, relax the facility bits
-         * of the kvm facility mask by all missing facilities. This will allow
-         * to determine the right CPU model by means of the remaining facilities.
-         * Live guest migration must prohibit the migration of KVMs running in
-         * a LPAR to non LPAR hosts.
-         */
-        if (!MACHINE_IS_LPAR)
-                for (i = 0; i < kvm_s390_fac_list_mask_size(); i++)
-                        kvm_s390_fac_list_mask[i] &= kvm->arch.model.fac->kvm[i];
-        /*
-         * Apply the kvm facility mask to limit the kvm supported/tolerated
-         * facility list.
-         */
        for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
                if (i < kvm_s390_fac_list_mask_size())
-                        kvm->arch.model.fac->kvm[i] &= kvm_s390_fac_list_mask[i];
+                        kvm->arch.model.fac->mask[i] &= kvm_s390_fac_list_mask[i];
                else
-                        kvm->arch.model.fac->kvm[i] = 0UL;
+                        kvm->arch.model.fac->mask[i] = 0UL;
        }
+        /* Populate the facility list initially. */
+        memcpy(kvm->arch.model.fac->list, kvm->arch.model.fac->mask,
+               S390_ARCH_FAC_LIST_SIZE_BYTE);
        kvm_s390_get_cpu_id(&kvm->arch.model.cpu_id);
        kvm->arch.model.ibc = sclp_get_ibc() & 0x0fff;
@@ -1165,8 +1160,6 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
        mutex_lock(&vcpu->kvm->lock);
        vcpu->arch.cpu_id = vcpu->kvm->arch.model.cpu_id;
-        memcpy(vcpu->kvm->arch.model.fac->sie, vcpu->kvm->arch.model.fac->kvm,
-               S390_ARCH_FAC_LIST_SIZE_BYTE);
        vcpu->arch.sie_block->ibc = vcpu->kvm->arch.model.ibc;
        mutex_unlock(&vcpu->kvm->lock);
@@ -1212,7 +1205,7 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
                vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca;
                set_bit(63 - id, (unsigned long *) &kvm->arch.sca->mcn);
        }
-        vcpu->arch.sie_block->fac = (int) (long) kvm->arch.model.fac->sie;
+        vcpu->arch.sie_block->fac = (int) (long) kvm->arch.model.fac->list;
        spin_lock_init(&vcpu->arch.local_int.lock);
        vcpu->arch.local_int.float_int = &kvm->arch.float_int;
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index 985c2114d7ef..c34109aa552d 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -128,7 +128,8 @@ static inline void kvm_s390_set_psw_cc(struct kvm_vcpu *vcpu, unsigned long cc)
 /* test availability of facility in a kvm intance */
 static inline int test_kvm_facility(struct kvm *kvm, unsigned long nr)
 {
-        return __test_facility(nr, kvm->arch.model.fac->kvm);
+        return __test_facility(nr, kvm->arch.model.fac->mask) &&
+                __test_facility(nr, kvm->arch.model.fac->list);
 }
 /* are cpu states controlled by user space */
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index bdd9b5b17e03..351116939ea2 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -348,7 +348,7 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
         * We need to shift the lower 32 facility bits (bit 0-31) from a u64
         * into a u32 memory representation. They will remain bits 0-31.
         */
-        fac = *vcpu->kvm->arch.model.fac->sie >> 32;
+        fac = *vcpu->kvm->arch.model.fac->list >> 32;
        rc = write_guest_lc(vcpu, offsetof(struct _lowcore, stfl_fac_list),
                            &fac, sizeof(fac));
        if (rc)
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index 753a56731951..f0b85443e060 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -287,7 +287,7 @@ void __iomem *pci_iomap_range(struct pci_dev *pdev,
        addr = ZPCI_IOMAP_ADDR_BASE | ((u64) idx << 48);
        return (void __iomem *) addr + offset;
 }
-EXPORT_SYMBOL_GPL(pci_iomap_range);
+EXPORT_SYMBOL(pci_iomap_range);
 void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 {
@@ -309,7 +309,7 @@ void pci_iounmap(struct pci_dev *pdev, void __iomem *addr)
        }
        spin_unlock(&zpci_iomap_lock);
 }
-EXPORT_SYMBOL_GPL(pci_iounmap);
+EXPORT_SYMBOL(pci_iounmap);
 static int pci_read(struct pci_bus *bus, unsigned int devfn, int where,
                    int size, u32 *val)
@@ -483,9 +483,8 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev)
        airq_iv_free_bit(zpci_aisb_iv, zdev->aisb);
 }
-static void zpci_map_resources(struct zpci_dev *zdev)
+static void zpci_map_resources(struct pci_dev *pdev)
 {
-        struct pci_dev *pdev = zdev->pdev;
        resource_size_t len;
        int i;
@@ -499,9 +498,8 @@ static void zpci_map_resources(struct zpci_dev *zdev)
        }
 }
-static void zpci_unmap_resources(struct zpci_dev *zdev)
+static void zpci_unmap_resources(struct pci_dev *pdev)
 {
-        struct pci_dev *pdev = zdev->pdev;
        resource_size_t len;
        int i;
@@ -651,7 +649,7 @@ int pcibios_add_device(struct pci_dev *pdev)
        zdev->pdev = pdev;
        pdev->dev.groups = zpci_attr_groups;
-        zpci_map_resources(zdev);
+        zpci_map_resources(pdev);
        for (i = 0; i < PCI_BAR_COUNT; i++) {
                res = &pdev->resource[i];
@@ -663,6 +661,11 @@ int pcibios_add_device(struct pci_dev *pdev)
        return 0;
 }
+void pcibios_release_device(struct pci_dev *pdev)
+{
+        zpci_unmap_resources(pdev);
+}
 int pcibios_enable_device(struct pci_dev *pdev, int mask)
 {
        struct zpci_dev *zdev = get_zdev(pdev);
@@ -670,7 +673,6 @@ int pcibios_enable_device(struct pci_dev *pdev, int mask)
        zdev->pdev = pdev;
        zpci_debug_init_device(zdev);
        zpci_fmb_enable_device(zdev);
-        zpci_map_resources(zdev);
        return pci_enable_resources(pdev, mask);
 }
@@ -679,7 +681,6 @@ void pcibios_disable_device(struct pci_dev *pdev)
 {
        struct zpci_dev *zdev = get_zdev(pdev);
-        zpci_unmap_resources(zdev);
        zpci_fmb_disable_device(zdev);
        zpci_debug_exit_device(zdev);
        zdev->pdev = NULL;
@@ -688,7 +689,8 @@ void pcibios_disable_device(struct pci_dev *pdev)
 #ifdef CONFIG_HIBERNATE_CALLBACKS
 static int zpci_restore(struct device *dev)
 {
-        struct zpci_dev *zdev = get_zdev(to_pci_dev(dev));
+        struct pci_dev *pdev = to_pci_dev(dev);
+        struct zpci_dev *zdev = get_zdev(pdev);
        int ret = 0;
        if (zdev->state != ZPCI_FN_STATE_ONLINE)
@@ -698,7 +700,7 @@ static int zpci_restore(struct device *dev)
        if (ret)
                goto out;
-        zpci_map_resources(zdev);
+        zpci_map_resources(pdev);
        zpci_register_ioat(zdev, 0, zdev->start_dma + PAGE_OFFSET,
                           zdev->start_dma + zdev->iommu_size - 1,
                           (u64) zdev->dma_table);
@@ -709,12 +711,14 @@ out:
 static int zpci_freeze(struct device *dev)
 {
-        struct zpci_dev *zdev = get_zdev(to_pci_dev(dev));
+        struct pci_dev *pdev = to_pci_dev(dev);
+        struct zpci_dev *zdev = get_zdev(pdev);
        if (zdev->state != ZPCI_FN_STATE_ONLINE)
                return 0;
        zpci_unregister_ioat(zdev, 0);
+        zpci_unmap_resources(pdev);
        return clp_disable_fh(zdev);
 }
diff --git a/arch/s390/pci/pci_mmio.c b/arch/s390/pci/pci_mmio.c
index 8aa271b3d1ad..b1bb2b72302c 100644
--- a/arch/s390/pci/pci_mmio.c
+++ b/arch/s390/pci/pci_mmio.c
@@ -64,8 +64,7 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr,
        if (copy_from_user(buf, user_buffer, length))
                goto out;
-        memcpy_toio(io_addr, buf, length);
+        ret = zpci_memcpy_toio(io_addr, buf, length);
-        ret = 0;
 out:
        if (buf != local_buf)
                kfree(buf);
@@ -98,16 +97,16 @@ SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr,
                goto out;
        io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK));
-        ret = -EFAULT;
+        if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) {
-        if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE)
+                ret = -EFAULT;
                goto out;
+        }
-        memcpy_fromio(buf, io_addr, length);
+        ret = zpci_memcpy_fromio(buf, io_addr, length);
+        if (ret)
-        if (copy_to_user(user_buffer, buf, length))
                goto out;
+        if (copy_to_user(user_buffer, buf, length))
+                ret = -EFAULT;
-        ret = 0;
 out:
        if (buf != local_buf)
                kfree(buf);
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 96ac69c5eba0..efb00ec75805 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -86,6 +86,9 @@ config ARCH_DEFCONFIG
        default "arch/sparc/configs/sparc32_defconfig" if SPARC32
        default "arch/sparc/configs/sparc64_defconfig" if SPARC64
+config ARCH_PROC_KCORE_TEXT
+        def_bool y
 config IOMMU_HELPER
        bool
        default y if SPARC64
diff --git a/arch/sparc/include/asm/hypervisor.h b/arch/sparc/include/asm/hypervisor.h
index 4f6725ff4c33..f5b6537306f0 100644
--- a/arch/sparc/include/asm/hypervisor.h
+++ b/arch/sparc/include/asm/hypervisor.h
@@ -2957,6 +2957,17 @@ unsigned long sun4v_t5_set_perfreg(unsigned long reg_num,
                                   unsigned long reg_val);
 #endif
+#define HV_FAST_M7_GET_PERFREG  0x43
+#define HV_FAST_M7_SET_PERFREG  0x44
+#ifndef __ASSEMBLY__
+unsigned long sun4v_m7_get_perfreg(unsigned long reg_num,
+                                      unsigned long *reg_val);
+unsigned long sun4v_m7_set_perfreg(unsigned long reg_num,
+                                      unsigned long reg_val);
+#endif
 /* Function numbers for HV_CORE_TRAP.  */
 #define HV_CORE_SET_VER                 0x00
 #define HV_CORE_PUTCHAR                 0x01
@@ -2981,6 +2992,7 @@ unsigned long sun4v_t5_set_perfreg(unsigned long reg_num,
 #define HV_GRP_SDIO                     0x0108
 #define HV_GRP_SDIO_ERR                 0x0109
 #define HV_GRP_REBOOT_DATA              0x0110
+#define HV_GRP_M7_PERF                  0x0114
 #define HV_GRP_NIAG_PERF                0x0200
 #define HV_GRP_FIRE_PERF                0x0201
 #define HV_GRP_N2_CPU                   0x0202
diff --git a/arch/sparc/include/asm/io_64.h b/arch/sparc/include/asm/io_64.h
index 9b672be70dda..50d4840d9aeb 100644
--- a/arch/sparc/include/asm/io_64.h
+++ b/arch/sparc/include/asm/io_64.h
@@ -407,16 +407,16 @@ static inline void iounmap(volatile void __iomem *addr)
 {
 }
-#define ioread8(X)                      readb(X)
+#define ioread8                 readb
-#define ioread16(X)                     readw(X)
+#define ioread16                readw
-#define ioread16be(X)                   __raw_readw(X)
+#define ioread16be              __raw_readw
-#define ioread32(X)                     readl(X)
+#define ioread32                readl
-#define ioread32be(X)                   __raw_readl(X)
+#define ioread32be              __raw_readl
-#define iowrite8(val,X)                 writeb(val,X)
+#define iowrite8                writeb
-#define iowrite16(val,X)                writew(val,X)
+#define iowrite16               writew
-#define iowrite16be(val,X)              __raw_writew(val,X)
+#define iowrite16be             __raw_writew
-#define iowrite32(val,X)                writel(val,X)
+#define iowrite32               writel
-#define iowrite32be(val,X)              __raw_writel(val,X)
+#define iowrite32be             __raw_writel
 /* Create a virtual mapping cookie for an IO port range */
 void __iomem *ioport_map(unsigned long port, unsigned int nr);
diff --git a/arch/sparc/include/asm/starfire.h b/arch/sparc/include/asm/starfire.h
index c100dc27a0a9..176fa0ad19f1 100644
--- a/arch/sparc/include/asm/starfire.h
+++ b/arch/sparc/include/asm/starfire.h
@@ -12,7 +12,6 @@
 extern int this_is_starfire;
 void check_if_starfire(void);
-int starfire_hard_smp_processor_id(void);
 void starfire_hookup(int);
 unsigned int starfire_translate(unsigned long imap, unsigned int upaid);
diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h
index 88d322b67fac..07cc49e541f4 100644
--- a/arch/sparc/kernel/entry.h
+++ b/arch/sparc/kernel/entry.h
@@ -98,11 +98,7 @@ void sun4v_do_mna(struct pt_regs *regs,
 void do_privop(struct pt_regs *regs);
 void do_privact(struct pt_regs *regs);
 void do_cee(struct pt_regs *regs);
-void do_cee_tl1(struct pt_regs *regs);
-void do_dae_tl1(struct pt_regs *regs);
-void do_iae_tl1(struct pt_regs *regs);
 void do_div0_tl1(struct pt_regs *regs);
-void do_fpdis_tl1(struct pt_regs *regs);
 void do_fpieee_tl1(struct pt_regs *regs);
 void do_fpother_tl1(struct pt_regs *regs);
 void do_ill_tl1(struct pt_regs *regs);
diff --git a/arch/sparc/kernel/hvapi.c b/arch/sparc/kernel/hvapi.c
index 5c55145bfbf0..662500fa555f 100644
--- a/arch/sparc/kernel/hvapi.c
+++ b/arch/sparc/kernel/hvapi.c
@@ -48,6 +48,7 @@ static struct api_info api_table[] = {
        { .group = HV_GRP_VT_CPU,                               },
        { .group = HV_GRP_T5_CPU,                               },
        { .group = HV_GRP_DIAG,         .flags = FLAG_PRE_API   },
+        { .group = HV_GRP_M7_PERF,                              },
 };
 static DEFINE_SPINLOCK(hvapi_lock);
diff --git a/arch/sparc/kernel/hvcalls.S b/arch/sparc/kernel/hvcalls.S
index caedf8320416..afbaba52d2f1 100644
--- a/arch/sparc/kernel/hvcalls.S
+++ b/arch/sparc/kernel/hvcalls.S
@@ -837,3 +837,19 @@ ENTRY(sun4v_t5_set_perfreg)
        retl
         nop
 ENDPROC(sun4v_t5_set_perfreg)
+ENTRY(sun4v_m7_get_perfreg)
+        mov     %o1, %o4
+        mov     HV_FAST_M7_GET_PERFREG, %o5
+        ta      HV_FAST_TRAP
+        stx     %o1, [%o4]
+        retl
+        nop
+ENDPROC(sun4v_m7_get_perfreg)
+ENTRY(sun4v_m7_set_perfreg)
+        mov     HV_FAST_M7_SET_PERFREG, %o5
+        ta      HV_FAST_TRAP
+        retl
+        nop
+ENDPROC(sun4v_m7_set_perfreg)
diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c
index 7e967c8018c8..eb978c77c76a 100644
--- a/arch/sparc/kernel/pcr.c
+++ b/arch/sparc/kernel/pcr.c
@@ -217,6 +217,31 @@ static const struct pcr_ops n5_pcr_ops = {
        .pcr_nmi_disable        = PCR_N4_PICNPT,
 };
+static u64 m7_pcr_read(unsigned long reg_num)
+{
+        unsigned long val;
+        (void) sun4v_m7_get_perfreg(reg_num, &val);
+        return val;
+}
+static void m7_pcr_write(unsigned long reg_num, u64 val)
+{
+        (void) sun4v_m7_set_perfreg(reg_num, val);
+}
+static const struct pcr_ops m7_pcr_ops = {
+        .read_pcr               = m7_pcr_read,
+        .write_pcr              = m7_pcr_write,
+        .read_pic               = n4_pic_read,
+        .write_pic              = n4_pic_write,
+        .nmi_picl_value         = n4_picl_value,
+        .pcr_nmi_enable         = (PCR_N4_PICNPT | PCR_N4_STRACE |
+                                   PCR_N4_UTRACE | PCR_N4_TOE |
+                                   (26 << PCR_N4_SL_SHIFT)),
+        .pcr_nmi_disable        = PCR_N4_PICNPT,
+};
 static unsigned long perf_hsvc_group;
 static unsigned long perf_hsvc_major;
@@ -248,6 +273,10 @@ static int __init register_perf_hsvc(void)
                        perf_hsvc_group = HV_GRP_T5_CPU;
                        break;
+                case SUN4V_CHIP_SPARC_M7:
+                        perf_hsvc_group = HV_GRP_M7_PERF;
+                        break;
                default:
                        return -ENODEV;
                }
@@ -293,6 +322,10 @@ static int __init setup_sun4v_pcr_ops(void)
                pcr_ops = &n5_pcr_ops;
                break;
+        case SUN4V_CHIP_SPARC_M7:
+                pcr_ops = &m7_pcr_ops;
+                break;
        default:
                ret = -ENODEV;
                break;
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 46a5e4508752..86eebfa3b158 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -792,6 +792,42 @@ static const struct sparc_pmu niagara4_pmu = {
        .num_pic_regs   = 4,
 };
+static void sparc_m7_write_pmc(int idx, u64 val)
+{
+        u64 pcr;
+        pcr = pcr_ops->read_pcr(idx);
+        /* ensure ov and ntc are reset */
+        pcr &= ~(PCR_N4_OV | PCR_N4_NTC);
+        pcr_ops->write_pic(idx, val & 0xffffffff);
+        pcr_ops->write_pcr(idx, pcr);
+}
+static const struct sparc_pmu sparc_m7_pmu = {
+        .event_map      = niagara4_event_map,
+        .cache_map      = &niagara4_cache_map,
+        .max_events     = ARRAY_SIZE(niagara4_perfmon_event_map),
+        .read_pmc       = sparc_vt_read_pmc,
+        .write_pmc      = sparc_m7_write_pmc,
+        .upper_shift    = 5,
+        .lower_shift    = 5,
+        .event_mask     = 0x7ff,
+        .user_bit       = PCR_N4_UTRACE,
+        .priv_bit       = PCR_N4_STRACE,
+        /* We explicitly don't support hypervisor tracing. */
+        .hv_bit         = 0,
+        .irq_bit        = PCR_N4_TOE,
+        .upper_nop      = 0,
+        .lower_nop      = 0,
+        .flags          = 0,
+        .max_hw_events  = 4,
+        .num_pcrs       = 4,
+        .num_pic_regs   = 4,
+};
 static const struct sparc_pmu *sparc_pmu __read_mostly;
 static u64 event_encoding(u64 event_id, int idx)
@@ -960,6 +996,8 @@ out:
        cpuc->pcr[0] |= cpuc->event[0]->hw.config_base;
 }
+static void sparc_pmu_start(struct perf_event *event, int flags);
 /* On this PMU each PIC has it's own PCR control register.  */
 static void calculate_multiple_pcrs(struct cpu_hw_events *cpuc)
 {
@@ -972,20 +1010,13 @@ static void calculate_multiple_pcrs(struct cpu_hw_events *cpuc)
                struct perf_event *cp = cpuc->event[i];
                struct hw_perf_event *hwc = &cp->hw;
                int idx = hwc->idx;
-                u64 enc;
                if (cpuc->current_idx[i] != PIC_NO_INDEX)
                        continue;
-                sparc_perf_event_set_period(cp, hwc, idx);
                cpuc->current_idx[i] = idx;
-                enc = perf_event_get_enc(cpuc->events[i]);
+                sparc_pmu_start(cp, PERF_EF_RELOAD);
-                cpuc->pcr[idx] &= ~mask_for_index(idx);
-                if (hwc->state & PERF_HES_STOPPED)
-                        cpuc->pcr[idx] |= nop_for_index(idx);
-                else
-                        cpuc->pcr[idx] |= event_encoding(enc, idx);
        }
 out:
        for (i = 0; i < cpuc->n_events; i++) {
@@ -1101,7 +1132,6 @@ static void sparc_pmu_del(struct perf_event *event, int _flags)
        int i;
        local_irq_save(flags);
-        perf_pmu_disable(event->pmu);
        for (i = 0; i < cpuc->n_events; i++) {
                if (event == cpuc->event[i]) {
@@ -1127,7 +1157,6 @@ static void sparc_pmu_del(struct perf_event *event, int _flags)
                }
        }
-        perf_pmu_enable(event->pmu);
        local_irq_restore(flags);
 }
@@ -1361,7 +1390,6 @@ static int sparc_pmu_add(struct perf_event *event, int ef_flags)
        unsigned long flags;
        local_irq_save(flags);
-        perf_pmu_disable(event->pmu);
        n0 = cpuc->n_events;
        if (n0 >= sparc_pmu->max_hw_events)
@@ -1394,7 +1422,6 @@ nocheck:
        ret = 0;
 out:
-        perf_pmu_enable(event->pmu);
        local_irq_restore(flags);
        return ret;
 }
@@ -1667,6 +1694,10 @@ static bool __init supported_pmu(void)
                sparc_pmu = &niagara4_pmu;
                return true;
        }
+        if (!strcmp(sparc_pmu_type, "sparc-m7")) {
+                sparc_pmu = &sparc_m7_pmu;
+                return true;
+        }
        return false;
 }
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index 0be7bf978cb1..46a59643bb1c 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -287,6 +287,8 @@ void arch_trigger_all_cpu_backtrace(bool include_self)
                        printk("             TPC[%lx] O7[%lx] I7[%lx] RPC[%lx]\n",
                               gp->tpc, gp->o7, gp->i7, gp->rpc);
                }
+                touch_nmi_watchdog();
        }
        memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot));
@@ -362,6 +364,8 @@ static void pmu_snapshot_all_cpus(void)
                       (cpu == this_cpu ? '*' : ' '), cpu,
                       pp->pcr[0], pp->pcr[1], pp->pcr[2], pp->pcr[3],
                       pp->pic[0], pp->pic[1], pp->pic[2], pp->pic[3]);
+                touch_nmi_watchdog();
        }
        memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot));
diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c
index da6f1a7fc4db..61139d9924ca 100644
--- a/arch/sparc/kernel/smp_64.c
+++ b/arch/sparc/kernel/smp_64.c
@@ -1406,11 +1406,32 @@ void __irq_entry smp_receive_signal_client(int irq, struct pt_regs *regs)
        scheduler_ipi();
 }
-/* This is a nop because we capture all other cpus
+static void stop_this_cpu(void *dummy)
- * anyways when making the PROM active.
+{
- */
+        prom_stopself();
+}
 void smp_send_stop(void)
 {
+        int cpu;
+        if (tlb_type == hypervisor) {
+                for_each_online_cpu(cpu) {
+                        if (cpu == smp_processor_id())
+                                continue;
+#ifdef CONFIG_SUN_LDOMS
+                        if (ldom_domaining_enabled) {
+                                unsigned long hv_err;
+                                hv_err = sun4v_cpu_stop(cpu);
+                                if (hv_err)
+                                        printk(KERN_ERR "sun4v_cpu_stop() "
+                                               "failed err=%lu\n", hv_err);
+                        } else
+#endif
+                                prom_stopcpu_cpuid(cpu);
+                }
+        } else
+                smp_call_function(stop_this_cpu, NULL, 0);
 }
 /**
diff --git a/arch/sparc/kernel/starfire.c b/arch/sparc/kernel/starfire.c
index 82281a566bb8..167fdfd9c837 100644
--- a/arch/sparc/kernel/starfire.c
+++ b/arch/sparc/kernel/starfire.c
@@ -28,11 +28,6 @@ void check_if_starfire(void)
                this_is_starfire = 1;
 }
-int starfire_hard_smp_processor_id(void)
-{
-        return upa_readl(0x1fff40000d0UL);
-}
 /*
 * Each Starfire board has 32 registers which perform translation
 * and delivery of traditional interrupt packets into the extended
diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c
index c85403d0496c..30e7ddb27a3a 100644
--- a/arch/sparc/kernel/sys_sparc_64.c
+++ b/arch/sparc/kernel/sys_sparc_64.c
@@ -333,7 +333,7 @@ SYSCALL_DEFINE6(sparc_ipc, unsigned int, call, int, first, unsigned long, second
        long err;
        /* No need for backward compatibility. We can start fresh... */
-        if (call <= SEMCTL) {
+        if (call <= SEMTIMEDOP) {
                switch (call) {
                case SEMOP:
                        err = sys_semtimedop(first, ptr,
diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c
index a27651e866e7..0e699745d643 100644
--- a/arch/sparc/kernel/traps_64.c
+++ b/arch/sparc/kernel/traps_64.c
@@ -2427,6 +2427,8 @@ void __noreturn die_if_kernel(char *str, struct pt_regs *regs)
                }
                user_instruction_dump ((unsigned int __user *) regs->tpc);
        }
+        if (panic_on_oops)
+                panic("Fatal exception");
        if (regs->tstate & TSTATE_PRIV)
                do_exit(SIGKILL);
        do_exit(SIGSEGV);
@@ -2564,27 +2566,6 @@ void do_cee(struct pt_regs *regs)
        die_if_kernel("TL0: Cache Error Exception", regs);
 }
-void do_cee_tl1(struct pt_regs *regs)
-{
-        exception_enter();
-        dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
-        die_if_kernel("TL1: Cache Error Exception", regs);
-}
-void do_dae_tl1(struct pt_regs *regs)
-{
-        exception_enter();
-        dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
-        die_if_kernel("TL1: Data Access Exception", regs);
-}
-void do_iae_tl1(struct pt_regs *regs)
-{
-        exception_enter();
-        dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
-        die_if_kernel("TL1: Instruction Access Exception", regs);
-}
 void do_div0_tl1(struct pt_regs *regs)
 {
        exception_enter();
@@ -2592,13 +2573,6 @@ void do_div0_tl1(struct pt_regs *regs)
        die_if_kernel("TL1: DIV0 Exception", regs);
 }
-void do_fpdis_tl1(struct pt_regs *regs)
-{
-        exception_enter();
-        dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
-        die_if_kernel("TL1: FPU Disabled", regs);
-}
 void do_fpieee_tl1(struct pt_regs *regs)
 {
        exception_enter();
diff --git a/arch/sparc/lib/memmove.S b/arch/sparc/lib/memmove.S
index b7f6334e159f..857ad4f8905f 100644
--- a/arch/sparc/lib/memmove.S
+++ b/arch/sparc/lib/memmove.S
@@ -8,9 +8,11 @@
        .text
 ENTRY(memmove) /* o0=dst o1=src o2=len */
-        mov             %o0, %g1
+        brz,pn          %o2, 99f
+         mov            %o0, %g1
        cmp             %o0, %o1
-        bleu,pt         %xcc, memcpy
+        bleu,pt         %xcc, 2f
         add            %o1, %o2, %g7
        cmp             %g7, %o0
        bleu,pt         %xcc, memcpy
@@ -24,7 +26,34 @@ ENTRY(memmove) /* o0=dst o1=src o2=len */
        stb             %g7, [%o0]
        bne,pt          %icc, 1b
         sub            %o0, 1, %o0
+99:
        retl
         mov            %g1, %o0
+        /* We can't just call memcpy for these memmove cases.  On some
+         * chips the memcpy uses cache initializing stores and when dst
+         * and src are close enough, those can clobber the source data
+         * before we've loaded it in.
+         */
+2:      or              %o0, %o1, %g7
+        or              %o2, %g7, %g7
+        andcc           %g7, 0x7, %g0
+        bne,pn          %xcc, 4f
+         nop
+3:      ldx             [%o1], %g7
+        add             %o1, 8, %o1
+        subcc           %o2, 8, %o2
+        add             %o0, 8, %o0
+        bne,pt          %icc, 3b
+         stx            %g7, [%o0 - 0x8]
+        ba,a,pt         %xcc, 99b
+4:      ldub            [%o1], %g7
+        add             %o1, 1, %o1
+        subcc           %o2, 1, %o2
+        add             %o0, 1, %o0
+        bne,pt          %icc, 4b
+         stb            %g7, [%o0 - 0x1]
+        ba,a,pt         %xcc, 99b
 ENDPROC(memmove)
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index 3ea267c53320..4ca0d6ba5ec8 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -2820,7 +2820,7 @@ static int __init report_memory(void)
        return 0;
 }
-device_initcall(report_memory);
+arch_initcall(report_memory);
 #ifdef CONFIG_SMP
 #define do_flush_tlb_kernel_range       smp_flush_tlb_kernel_range
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index c2fb8a87dccb..b7d31ca55187 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -499,6 +499,7 @@ config X86_INTEL_QUARK
        depends on X86_IO_APIC
        select IOSF_MBI
        select INTEL_IMR
+        select COMMON_CLK
        ---help---
          Select to include support for Quark X1000 SoC.
          Say Y here if you have a Quark based system such as the Arduino
diff --git a/arch/x86/boot/compressed/aslr.c b/arch/x86/boot/compressed/aslr.c
index 7083c16cccba..bb1376381985 100644
--- a/arch/x86/boot/compressed/aslr.c
+++ b/arch/x86/boot/compressed/aslr.c
@@ -14,13 +14,6 @@
 static const char build_str[] = UTS_RELEASE " (" LINUX_COMPILE_BY "@"
                LINUX_COMPILE_HOST ") (" LINUX_COMPILER ") " UTS_VERSION;
-struct kaslr_setup_data {
-        __u64 next;
-        __u32 type;
-        __u32 len;
-        __u8 data[1];
-} kaslr_setup_data;
 #define I8254_PORT_CONTROL      0x43
 #define I8254_PORT_COUNTER0     0x40
 #define I8254_CMD_READBACK      0xC0
@@ -302,29 +295,7 @@ static unsigned long find_random_addr(unsigned long minimum,
        return slots_fetch_random();
 }
-static void add_kaslr_setup_data(struct boot_params *params, __u8 enabled)
+unsigned char *choose_kernel_location(unsigned char *input,
-{
-        struct setup_data *data;
-        kaslr_setup_data.type = SETUP_KASLR;
-        kaslr_setup_data.len = 1;
-        kaslr_setup_data.next = 0;
-        kaslr_setup_data.data[0] = enabled;
-        data = (struct setup_data *)(unsigned long)params->hdr.setup_data;
-        while (data && data->next)
-                data = (struct setup_data *)(unsigned long)data->next;
-        if (data)
-                data->next = (unsigned long)&kaslr_setup_data;
-        else
-                params->hdr.setup_data = (unsigned long)&kaslr_setup_data;
-}
-unsigned char *choose_kernel_location(struct boot_params *params,
-                                      unsigned char *input,
                                      unsigned long input_size,
                                      unsigned char *output,
                                      unsigned long output_size)
@@ -335,17 +306,14 @@ unsigned char *choose_kernel_location(struct boot_params *params,
 #ifdef CONFIG_HIBERNATION
        if (!cmdline_find_option_bool("kaslr")) {
                debug_putstr("KASLR disabled by default...\n");
-                add_kaslr_setup_data(params, 0);
                goto out;
        }
 #else
        if (cmdline_find_option_bool("nokaslr")) {
                debug_putstr("KASLR disabled by cmdline...\n");
-                add_kaslr_setup_data(params, 0);
                goto out;
        }
 #endif
-        add_kaslr_setup_data(params, 1);
        /* Record the various known unsafe memory ranges. */
        mem_avoid_init((unsigned long)input, input_size,
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
index 5903089c818f..a950864a64da 100644
--- a/arch/x86/boot/compressed/misc.c
+++ b/arch/x86/boot/compressed/misc.c
@@ -401,8 +401,7 @@ asmlinkage __visible void *decompress_kernel(void *rmode, memptr heap,
         * the entire decompressed kernel plus relocation table, or the
         * entire decompressed kernel plus .bss and .brk sections.
         */
-        output = choose_kernel_location(real_mode, input_data, input_len,
+        output = choose_kernel_location(input_data, input_len, output,
-                                        output,
                                        output_len > run_size ? output_len
                                                              : run_size);
diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h
index ee3576b2666b..04477d68403f 100644
--- a/arch/x86/boot/compressed/misc.h
+++ b/arch/x86/boot/compressed/misc.h
@@ -57,8 +57,7 @@ int cmdline_find_option_bool(const char *option);
 #if CONFIG_RANDOMIZE_BASE
 /* aslr.c */
-unsigned char *choose_kernel_location(struct boot_params *params,
+unsigned char *choose_kernel_location(unsigned char *input,
-                                      unsigned char *input,
                                      unsigned long input_size,
                                      unsigned char *output,
                                      unsigned long output_size);
@@ -66,8 +65,7 @@ unsigned char *choose_kernel_location(struct boot_params *params,
 bool has_cpuflag(int flag);
 #else
 static inline
-unsigned char *choose_kernel_location(struct boot_params *params,
+unsigned char *choose_kernel_location(unsigned char *input,
-                                      unsigned char *input,
                                      unsigned long input_size,
                                      unsigned char *output,
                                      unsigned long output_size)
diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
index 947c6bf52c33..54f60ab41c63 100644
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -1155,7 +1155,7 @@ static int __driver_rfc4106_decrypt(struct aead_request *req)
                src = kmalloc(req->cryptlen + req->assoclen, GFP_ATOMIC);
                if (!src)
                        return -ENOMEM;
-                assoc = (src + req->cryptlen + auth_tag_len);
+                assoc = (src + req->cryptlen);
                scatterwalk_map_and_copy(src, req->src, 0, req->cryptlen, 0);
                scatterwalk_map_and_copy(assoc, req->assoc, 0,
                        req->assoclen, 0);
@@ -1180,7 +1180,7 @@ static int __driver_rfc4106_decrypt(struct aead_request *req)
                scatterwalk_done(&src_sg_walk, 0, 0);
                scatterwalk_done(&assoc_sg_walk, 0, 0);
        } else {
-                scatterwalk_map_and_copy(dst, req->dst, 0, req->cryptlen, 1);
+                scatterwalk_map_and_copy(dst, req->dst, 0, tempCipherLen, 1);
                kfree(src);
        }
        return retval;
diff --git a/arch/x86/include/asm/fpu-internal.h b/arch/x86/include/asm/fpu-internal.h
index 0dbc08282291..72ba21a8b5fc 100644
--- a/arch/x86/include/asm/fpu-internal.h
+++ b/arch/x86/include/asm/fpu-internal.h
@@ -370,7 +370,7 @@ static inline void drop_fpu(struct task_struct *tsk)
        preempt_disable();
        tsk->thread.fpu_counter = 0;
        __drop_fpu(tsk);
-        clear_used_math();
+        clear_stopped_child_used_math(tsk);
        preempt_enable();
 }
diff --git a/arch/x86/include/asm/page_types.h b/arch/x86/include/asm/page_types.h
index 95e11f79f123..f97fbe3abb67 100644
--- a/arch/x86/include/asm/page_types.h
+++ b/arch/x86/include/asm/page_types.h
@@ -51,8 +51,6 @@ extern int devmem_is_allowed(unsigned long pagenr);
 extern unsigned long max_low_pfn_mapped;
 extern unsigned long max_pfn_mapped;
-extern bool kaslr_enabled;
 static inline phys_addr_t get_max_mapped(void)
 {
        return (phys_addr_t)max_pfn_mapped << PAGE_SHIFT;
diff --git a/arch/x86/include/asm/pci_x86.h b/arch/x86/include/asm/pci_x86.h
index fa1195dae425..164e3f8d3c3d 100644
--- a/arch/x86/include/asm/pci_x86.h
+++ b/arch/x86/include/asm/pci_x86.h
@@ -93,6 +93,8 @@ extern raw_spinlock_t pci_config_lock;
 extern int (*pcibios_enable_irq)(struct pci_dev *dev);
 extern void (*pcibios_disable_irq)(struct pci_dev *dev);
+extern bool mp_should_keep_irq(struct device *dev);
 struct pci_raw_ops {
        int (*read)(unsigned int domain, unsigned int bus, unsigned int devfn,
                                                int reg, int len, u32 *val);
diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h
index 5fa9770035dc..c9a6d68b8d62 100644
--- a/arch/x86/include/asm/xsave.h
+++ b/arch/x86/include/asm/xsave.h
@@ -82,18 +82,15 @@ static inline int xsave_state_booting(struct xsave_struct *fx, u64 mask)
        if (boot_cpu_has(X86_FEATURE_XSAVES))
                asm volatile("1:"XSAVES"\n\t"
                        "2:\n\t"
-                        : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
+                             xstate_fault
+                        : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
                        :   "memory");
        else
                asm volatile("1:"XSAVE"\n\t"
                        "2:\n\t"
-                        : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
+                             xstate_fault
+                        : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
                        :   "memory");
-        asm volatile(xstate_fault
-                     : "0" (0)
-                     : "memory");
        return err;
 }
@@ -112,18 +109,15 @@ static inline int xrstor_state_booting(struct xsave_struct *fx, u64 mask)
        if (boot_cpu_has(X86_FEATURE_XSAVES))
                asm volatile("1:"XRSTORS"\n\t"
                        "2:\n\t"
-                        : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
+                             xstate_fault
+                        : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
                        :   "memory");
        else
                asm volatile("1:"XRSTOR"\n\t"
                        "2:\n\t"
-                        : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
+                             xstate_fault
+                        : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
                        :   "memory");
-        asm volatile(xstate_fault
-                     : "0" (0)
-                     : "memory");
        return err;
 }
@@ -149,9 +143,9 @@ static inline int xsave_state(struct xsave_struct *fx, u64 mask)
         */
        alternative_input_2(
                "1:"XSAVE,
-                "1:"XSAVEOPT,
+                XSAVEOPT,
                X86_FEATURE_XSAVEOPT,
-                "1:"XSAVES,
+                XSAVES,
                X86_FEATURE_XSAVES,
                [fx] "D" (fx), "a" (lmask), "d" (hmask) :
                "memory");
@@ -178,7 +172,7 @@ static inline int xrstor_state(struct xsave_struct *fx, u64 mask)
         */
        alternative_input(
                "1: " XRSTOR,
-                "1: " XRSTORS,
+                XRSTORS,
                X86_FEATURE_XSAVES,
                "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
                : "memory");
diff --git a/arch/x86/include/uapi/asm/bootparam.h b/arch/x86/include/uapi/asm/bootparam.h
index 44e6dd7e36a2..225b0988043a 100644
--- a/arch/x86/include/uapi/asm/bootparam.h
+++ b/arch/x86/include/uapi/asm/bootparam.h
@@ -7,7 +7,6 @@
 #define SETUP_DTB                       2
 #define SETUP_PCI                       3
 #define SETUP_EFI                       4
-#define SETUP_KASLR                     5
 /* ram_size flags */
 #define RAMDISK_IMAGE_START_MASK        0x07FF
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 3d525c6124f6..803b684676ff 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -1338,6 +1338,26 @@ static int __init dmi_ignore_irq0_timer_override(const struct dmi_system_id *d)
 }
 /*
+ * ACPI offers an alternative platform interface model that removes
+ * ACPI hardware requirements for platforms that do not implement
+ * the PC Architecture.
+ *
+ * We initialize the Hardware-reduced ACPI model here:
+ */
+static void __init acpi_reduced_hw_init(void)
+{
+        if (acpi_gbl_reduced_hardware) {
+                /*
+                 * Override x86_init functions and bypass legacy pic
+                 * in Hardware-reduced ACPI mode
+                 */
+                x86_init.timers.timer_init      = x86_init_noop;
+                x86_init.irqs.pre_vector_init   = x86_init_noop;
+                legacy_pic                      = &null_legacy_pic;
+        }
+}
+/*
 * If your system is blacklisted here, but you find that acpi=force
 * works for you, please contact linux-acpi@vger.kernel.org
 */
@@ -1536,6 +1556,11 @@ int __init early_acpi_boot_init(void)
         */
        early_acpi_process_madt();
+        /*
+         * Hardware-reduced ACPI mode initialization:
+         */
+        acpi_reduced_hw_init();
        return 0;
 }
diff --git a/arch/x86/kernel/apic/apic_numachip.c b/arch/x86/kernel/apic/apic_numachip.c
index c2fd21fed002..017149cded07 100644
--- a/arch/x86/kernel/apic/apic_numachip.c
+++ b/arch/x86/kernel/apic/apic_numachip.c
@@ -37,10 +37,12 @@ static const struct apic apic_numachip;
 static unsigned int get_apic_id(unsigned long x)
 {
        unsigned long value;
-        unsigned int id;
+        unsigned int id = (x >> 24) & 0xff;
-        rdmsrl(MSR_FAM10H_NODE_ID, value);
+        if (static_cpu_has_safe(X86_FEATURE_NODEID_MSR)) {
-        id = ((x >> 24) & 0xffU) | ((value << 2) & 0xff00U);
+                rdmsrl(MSR_FAM10H_NODE_ID, value);
+                id |= (value << 2) & 0xff00;
+        }
        return id;
 }
@@ -155,10 +157,18 @@ static int __init numachip_probe(void)
 static void fixup_cpu_id(struct cpuinfo_x86 *c, int node)
 {
-        if (c->phys_proc_id != node) {
+        u64 val;
-                c->phys_proc_id = node;
+        u32 nodes = 1;
-                per_cpu(cpu_llc_id, smp_processor_id()) = node;
+        this_cpu_write(cpu_llc_id, node);
+        /* Account for nodes per socket in multi-core-module processors */
+        if (static_cpu_has_safe(X86_FEATURE_NODEID_MSR)) {
+                rdmsrl(MSR_FAM10H_NODE_ID, val);
+                nodes = ((val >> 3) & 7) + 1;
        }
+        c->phys_proc_id = node / nodes;
 }
 static int __init numachip_system_init(void)
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 10074ad9ebf8..1d74d161687c 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -269,11 +269,14 @@ ENTRY(ret_from_fork)
        testl $3, CS-ARGOFFSET(%rsp)            # from kernel_thread?
        jz   1f
-        testl $_TIF_IA32, TI_flags(%rcx)        # 32-bit compat task needs IRET
+        /*
-        jnz  int_ret_from_sys_call
+         * By the time we get here, we have no idea whether our pt_regs,
+         * ti flags, and ti status came from the 64-bit SYSCALL fast path,
-        RESTORE_TOP_OF_STACK %rdi, -ARGOFFSET
+         * the slow path, or one of the ia32entry paths.
-        jmp ret_from_sys_call                   # go to the SYSRET fastpath
+         * Use int_ret_from_sys_call to return, since it can safely handle
+         * all of the above.
+         */
+        jmp  int_ret_from_sys_call
 1:
        subq $REST_SKIP, %rsp   # leave space for volatiles
diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
index 9bbb9b35c144..d1ac80b72c72 100644
--- a/arch/x86/kernel/module.c
+++ b/arch/x86/kernel/module.c
@@ -47,13 +47,21 @@ do {							\
 #ifdef CONFIG_RANDOMIZE_BASE
 static unsigned long module_load_offset;
+static int randomize_modules = 1;
 /* Mutex protects the module_load_offset. */
 static DEFINE_MUTEX(module_kaslr_mutex);
+static int __init parse_nokaslr(char *p)
+{
+        randomize_modules = 0;
+        return 0;
+}
+early_param("nokaslr", parse_nokaslr);
 static unsigned long int get_module_load_offset(void)
 {
-        if (kaslr_enabled) {
+        if (randomize_modules) {
                mutex_lock(&module_kaslr_mutex);
                /*
                 * Calculate the module_load_offset the first time this
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 98dc9317286e..0a2421cca01f 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -122,8 +122,6 @@
 unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
-bool __read_mostly kaslr_enabled = false;
 #ifdef CONFIG_DMI
 RESERVE_BRK(dmi_alloc, 65536);
 #endif
@@ -427,11 +425,6 @@ static void __init reserve_initrd(void)
 }
 #endif /* CONFIG_BLK_DEV_INITRD */
-static void __init parse_kaslr_setup(u64 pa_data, u32 data_len)
-{
-        kaslr_enabled = (bool)(pa_data + sizeof(struct setup_data));
-}
 static void __init parse_setup_data(void)
 {
        struct setup_data *data;
@@ -457,9 +450,6 @@ static void __init parse_setup_data(void)
                case SETUP_EFI:
                        parse_efi_setup(pa_data, data_len);
                        break;
-                case SETUP_KASLR:
-                        parse_kaslr_setup(pa_data, data_len);
-                        break;
                default:
                        break;
                }
@@ -842,14 +832,10 @@ static void __init trim_low_memory_range(void)
 static int
 dump_kernel_offset(struct notifier_block *self, unsigned long v, void *p)
 {
-        if (kaslr_enabled)
+        pr_emerg("Kernel Offset: 0x%lx from 0x%lx "
-                pr_emerg("Kernel Offset: 0x%lx from 0x%lx (relocation range: 0x%lx-0x%lx)\n",
+                 "(relocation range: 0x%lx-0x%lx)\n",
-                         (unsigned long)&_text - __START_KERNEL,
+                 (unsigned long)&_text - __START_KERNEL, __START_KERNEL,
-                         __START_KERNEL,
+                 __START_KERNEL_map, MODULES_VADDR-1);
-                         __START_KERNEL_map,
-                         MODULES_VADDR-1);
-        else
-                pr_emerg("Kernel Offset: disabled\n");
        return 0;
 }
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 9d2073e2ecc9..4ff5d162ff9f 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -384,7 +384,7 @@ dotraplinkage void do_bounds(struct pt_regs *regs, long error_code)
                goto exit;
        conditional_sti(regs);
-        if (!user_mode(regs))
+        if (!user_mode_vm(regs))
                die("bounds", regs, error_code);
        if (!cpu_feature_enabled(X86_FEATURE_MPX)) {
@@ -637,7 +637,7 @@ dotraplinkage void do_debug(struct pt_regs *regs, long error_code)
         * then it's very likely the result of an icebp/int01 trap.
         * User wants a sigtrap for that.
         */
-        if (!dr6 && user_mode(regs))
+        if (!dr6 && user_mode_vm(regs))
                user_icebp = 1;
        /* Catch kmemcheck conditions first of all! */
diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c
index 34f66e58a896..cdc6cf903078 100644
--- a/arch/x86/kernel/xsave.c
+++ b/arch/x86/kernel/xsave.c
@@ -379,7 +379,7 @@ int __restore_xstate_sig(void __user *buf, void __user *buf_fx, int size)
                 * thread's fpu state, reconstruct fxstate from the fsave
                 * header. Sanitize the copied state etc.
                 */
-                struct xsave_struct *xsave = &tsk->thread.fpu.state->xsave;
+                struct fpu *fpu = &tsk->thread.fpu;
                struct user_i387_ia32_struct env;
                int err = 0;
@@ -393,14 +393,15 @@ int __restore_xstate_sig(void __user *buf, void __user *buf_fx, int size)
                 */
                drop_fpu(tsk);
-                if (__copy_from_user(xsave, buf_fx, state_size) ||
+                if (__copy_from_user(&fpu->state->xsave, buf_fx, state_size) ||
                    __copy_from_user(&env, buf, sizeof(env))) {
+                        fpu_finit(fpu);
                        err = -1;
                } else {
                        sanitize_restored_xstate(tsk, &env, xstate_bv, fx_only);
-                        set_used_math();
                }
+                set_used_math();
                if (use_eager_fpu()) {
                        preempt_disable();
                        math_state_restore();
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index e0b794a84c35..106c01557f2b 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -4950,7 +4950,8 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
                        goto done;
                }
        }
-        ctxt->dst.orig_val = ctxt->dst.val;
+        /* Copy full 64-bit value for CMPXCHG8B.  */
+        ctxt->dst.orig_val64 = ctxt->dst.val64;
 special_insn:
diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c
index cc31f7c06d3d..9541ba34126b 100644
--- a/arch/x86/kvm/i8259.c
+++ b/arch/x86/kvm/i8259.c
@@ -507,6 +507,7 @@ static int picdev_read(struct kvm_pic *s,
                return -EOPNOTSUPP;
        if (len != 1) {
+                memset(val, 0, len);
                pr_pic_unimpl("non byte read\n");
                return 0;
        }
diff --git a/arch/x86/kvm/ioapic.c b/arch/x86/kvm/ioapic.c
index b1947e0f3e10..46d4449772bc 100644
--- a/arch/x86/kvm/ioapic.c
+++ b/arch/x86/kvm/ioapic.c
@@ -422,6 +422,7 @@ static void __kvm_ioapic_update_eoi(struct kvm_vcpu *vcpu,
                        struct kvm_ioapic *ioapic, int vector, int trigger_mode)
 {
        int i;
+        struct kvm_lapic *apic = vcpu->arch.apic;
        for (i = 0; i < IOAPIC_NUM_PINS; i++) {
                union kvm_ioapic_redirect_entry *ent = &ioapic->redirtbl[i];
@@ -443,7 +444,8 @@ static void __kvm_ioapic_update_eoi(struct kvm_vcpu *vcpu,
                kvm_notify_acked_irq(ioapic->kvm, KVM_IRQCHIP_IOAPIC, i);
                spin_lock(&ioapic->lock);
-                if (trigger_mode != IOAPIC_LEVEL_TRIG)
+                if (trigger_mode != IOAPIC_LEVEL_TRIG ||
+                    kvm_apic_get_reg(apic, APIC_SPIV) & APIC_SPIV_DIRECTED_EOI)
                        continue;
                ASSERT(ent->fields.trig_mode == IOAPIC_LEVEL_TRIG);
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index e55b5fc344eb..4ee827d7bf36 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -833,8 +833,7 @@ int kvm_apic_compare_prio(struct kvm_vcpu *vcpu1, struct kvm_vcpu *vcpu2)
 static void kvm_ioapic_send_eoi(struct kvm_lapic *apic, int vector)
 {
-        if (!(kvm_apic_get_reg(apic, APIC_SPIV) & APIC_SPIV_DIRECTED_EOI) &&
+        if (kvm_ioapic_handles_vector(apic->vcpu->kvm, vector)) {
-            kvm_ioapic_handles_vector(apic->vcpu->kvm, vector)) {
                int trigger_mode;
                if (apic_test_vector(vector, apic->regs + APIC_TMR))
                        trigger_mode = IOAPIC_LEVEL_TRIG;
@@ -1572,7 +1571,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu)
                apic_set_reg(apic, APIC_TMR + 0x10 * i, 0);
        }
        apic->irr_pending = kvm_apic_vid_enabled(vcpu->kvm);
-        apic->isr_count = kvm_apic_vid_enabled(vcpu->kvm);
+        apic->isr_count = kvm_x86_ops->hwapic_isr_update ? 1 : 0;
        apic->highest_isr_cache = -1;
        update_divide_count(apic);
        atomic_set(&apic->lapic_timer.pending, 0);
@@ -1782,7 +1781,7 @@ void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu,
        update_divide_count(apic);
        start_apic_timer(apic);
        apic->irr_pending = true;
-        apic->isr_count = kvm_apic_vid_enabled(vcpu->kvm) ?
+        apic->isr_count = kvm_x86_ops->hwapic_isr_update ?
                                1 : count_vectors(apic->regs + APIC_ISR);
        apic->highest_isr_cache = -1;
        if (kvm_x86_ops->hwapic_irr_update)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index d319e0c24758..cc618c882f90 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -3649,11 +3649,6 @@ static void svm_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap)
        return;
 }
-static void svm_hwapic_isr_update(struct kvm *kvm, int isr)
-{
-        return;
-}
 static void svm_sync_pir_to_irr(struct kvm_vcpu *vcpu)
 {
        return;
@@ -4403,7 +4398,6 @@ static struct kvm_x86_ops svm_x86_ops = {
        .set_virtual_x2apic_mode = svm_set_virtual_x2apic_mode,
        .vm_has_apicv = svm_vm_has_apicv,
        .load_eoi_exitmap = svm_load_eoi_exitmap,
-        .hwapic_isr_update = svm_hwapic_isr_update,
        .sync_pir_to_irr = svm_sync_pir_to_irr,
        .set_tss_addr = svm_set_tss_addr,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 14c1a18d206a..ae4f6d35d19c 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2168,7 +2168,10 @@ static void vmx_set_msr_bitmap(struct kvm_vcpu *vcpu)
 {
        unsigned long *msr_bitmap;
-        if (irqchip_in_kernel(vcpu->kvm) && apic_x2apic_mode(vcpu->arch.apic)) {
+        if (is_guest_mode(vcpu))
+                msr_bitmap = vmx_msr_bitmap_nested;
+        else if (irqchip_in_kernel(vcpu->kvm) &&
+                apic_x2apic_mode(vcpu->arch.apic)) {
                if (is_long_mode(vcpu))
                        msr_bitmap = vmx_msr_bitmap_longmode_x2apic;
                else
@@ -2476,8 +2479,7 @@ static void nested_vmx_setup_ctls_msrs(struct vcpu_vmx *vmx)
        if (enable_ept) {
                /* nested EPT: emulate EPT also to L1 */
                vmx->nested.nested_vmx_secondary_ctls_high |=
-                        SECONDARY_EXEC_ENABLE_EPT |
+                        SECONDARY_EXEC_ENABLE_EPT;
-                        SECONDARY_EXEC_UNRESTRICTED_GUEST;
                vmx->nested.nested_vmx_ept_caps = VMX_EPT_PAGE_WALK_4_BIT |
                         VMX_EPTP_WB_BIT | VMX_EPT_2MB_PAGE_BIT |
                         VMX_EPT_INVEPT_BIT;
@@ -2491,6 +2493,10 @@ static void nested_vmx_setup_ctls_msrs(struct vcpu_vmx *vmx)
        } else
                vmx->nested.nested_vmx_ept_caps = 0;
+        if (enable_unrestricted_guest)
+                vmx->nested.nested_vmx_secondary_ctls_high |=
+                        SECONDARY_EXEC_UNRESTRICTED_GUEST;
        /* miscellaneous data */
        rdmsr(MSR_IA32_VMX_MISC,
                vmx->nested.nested_vmx_misc_low,
@@ -4367,6 +4373,18 @@ static int vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu)
        return 0;
 }
+static inline bool kvm_vcpu_trigger_posted_interrupt(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_SMP
+        if (vcpu->mode == IN_GUEST_MODE) {
+                apic->send_IPI_mask(get_cpu_mask(vcpu->cpu),
+                                POSTED_INTR_VECTOR);
+                return true;
+        }
+#endif
+        return false;
+}
 static int vmx_deliver_nested_posted_interrupt(struct kvm_vcpu *vcpu,
                                                int vector)
 {
@@ -4375,9 +4393,7 @@ static int vmx_deliver_nested_posted_interrupt(struct kvm_vcpu *vcpu,
        if (is_guest_mode(vcpu) &&
            vector == vmx->nested.posted_intr_nv) {
                /* the PIR and ON have been set by L1. */
-                if (vcpu->mode == IN_GUEST_MODE)
+                kvm_vcpu_trigger_posted_interrupt(vcpu);
-                        apic->send_IPI_mask(get_cpu_mask(vcpu->cpu),
-                                POSTED_INTR_VECTOR);
                /*
                 * If a posted intr is not recognized by hardware,
                 * we will accomplish it in the next vmentry.
@@ -4409,12 +4425,7 @@ static void vmx_deliver_posted_interrupt(struct kvm_vcpu *vcpu, int vector)
        r = pi_test_and_set_on(&vmx->pi_desc);
        kvm_make_request(KVM_REQ_EVENT, vcpu);
-#ifdef CONFIG_SMP
+        if (r || !kvm_vcpu_trigger_posted_interrupt(vcpu))
-        if (!r && (vcpu->mode == IN_GUEST_MODE))
-                apic->send_IPI_mask(get_cpu_mask(vcpu->cpu),
-                                POSTED_INTR_VECTOR);
-        else
-#endif
                kvm_vcpu_kick(vcpu);
 }
@@ -9213,9 +9224,9 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
        }
        if (cpu_has_vmx_msr_bitmap() &&
-            exec_control & CPU_BASED_USE_MSR_BITMAPS &&
+            exec_control & CPU_BASED_USE_MSR_BITMAPS) {
-            nested_vmx_merge_msr_bitmap(vcpu, vmcs12)) {
+                nested_vmx_merge_msr_bitmap(vcpu, vmcs12);
-                vmcs_write64(MSR_BITMAP, __pa(vmx_msr_bitmap_nested));
+                /* MSR_BITMAP will be set by following vmx_set_efer. */
        } else
                exec_control &= ~CPU_BASED_USE_MSR_BITMAPS;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index d7a300e0147f..0ee725f1896d 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2744,7 +2744,6 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
        case KVM_CAP_USER_NMI:
        case KVM_CAP_REINJECT_CONTROL:
        case KVM_CAP_IRQ_INJECT_STATUS:
-        case KVM_CAP_IRQFD:
        case KVM_CAP_IOEVENTFD:
        case KVM_CAP_IOEVENTFD_NO_LENGTH:
        case KVM_CAP_PIT2:
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 6ac273832f28..e4695985f9de 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -331,7 +331,7 @@ static void probe_pci_root_info(struct pci_root_info *info,
                                struct list_head *list)
 {
        int ret;
-        struct resource_entry *entry;
+        struct resource_entry *entry, *tmp;
        sprintf(info->name, "PCI Bus %04x:%02x", domain, busnum);
        info->bridge = device;
@@ -345,8 +345,13 @@ static void probe_pci_root_info(struct pci_root_info *info,
                dev_dbg(&device->dev,
                        "no IO and memory resources present in _CRS\n");
        else
-                resource_list_for_each_entry(entry, list)
+                resource_list_for_each_entry_safe(entry, tmp, list) {
-                        entry->res->name = info->name;
+                        if ((entry->res->flags & IORESOURCE_WINDOW) == 0 ||
+                            (entry->res->flags & IORESOURCE_DISABLED))
+                                resource_list_destroy_entry(entry);
+                        else
+                                entry->res->name = info->name;
+                }
 }
 struct pci_bus *pci_acpi_scan_root(struct acpi_pci_root *root)
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 3d2612b68694..2fb384724ebb 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -513,31 +513,6 @@ void __init pcibios_set_cache_line_size(void)
        }
 }
-/*
- * Some device drivers assume dev->irq won't change after calling
- * pci_disable_device(). So delay releasing of IRQ resource to driver
- * unbinding time. Otherwise it will break PM subsystem and drivers
- * like xen-pciback etc.
- */
-static int pci_irq_notifier(struct notifier_block *nb, unsigned long action,
-                            void *data)
-{
-        struct pci_dev *dev = to_pci_dev(data);
-        if (action != BUS_NOTIFY_UNBOUND_DRIVER)
-                return NOTIFY_DONE;
-        if (pcibios_disable_irq)
-                pcibios_disable_irq(dev);
-        return NOTIFY_OK;
-}
-static struct notifier_block pci_irq_nb = {
-        .notifier_call = pci_irq_notifier,
-        .priority = INT_MIN,
-};
 int __init pcibios_init(void)
 {
        if (!raw_pci_ops) {
@@ -550,9 +525,6 @@ int __init pcibios_init(void)
        if (pci_bf_sort >= pci_force_bf)
                pci_sort_breadthfirst();
-        bus_register_notifier(&pci_bus_type, &pci_irq_nb);
        return 0;
 }
@@ -711,6 +683,12 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
        return 0;
 }
+void pcibios_disable_device (struct pci_dev *dev)
+{
+        if (!pci_dev_msi_enabled(dev) && pcibios_disable_irq)
+                pcibios_disable_irq(dev);
+}
 int pci_ext_cfg_avail(void)
 {
        if (raw_pci_ext_ops)
diff --git a/arch/x86/pci/intel_mid_pci.c b/arch/x86/pci/intel_mid_pci.c
index efb849323c74..852aa4c92da0 100644
--- a/arch/x86/pci/intel_mid_pci.c
+++ b/arch/x86/pci/intel_mid_pci.c
@@ -234,10 +234,10 @@ static int intel_mid_pci_irq_enable(struct pci_dev *dev)
 static void intel_mid_pci_irq_disable(struct pci_dev *dev)
 {
-        if (dev->irq_managed && dev->irq > 0) {
+        if (!mp_should_keep_irq(&dev->dev) && dev->irq_managed &&
+            dev->irq > 0) {
                mp_unmap_irq(dev->irq);
                dev->irq_managed = 0;
-                dev->irq = 0;
        }
 }
diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c
index e71b3dbd87b8..5dc6ca5e1741 100644
--- a/arch/x86/pci/irq.c
+++ b/arch/x86/pci/irq.c
@@ -1256,9 +1256,22 @@ static int pirq_enable_irq(struct pci_dev *dev)
        return 0;
 }
+bool mp_should_keep_irq(struct device *dev)
+{
+        if (dev->power.is_prepared)
+                return true;
+#ifdef CONFIG_PM
+        if (dev->power.runtime_status == RPM_SUSPENDING)
+                return true;
+#endif
+        return false;
+}
 static void pirq_disable_irq(struct pci_dev *dev)
 {
-        if (io_apic_assign_pci_irqs && dev->irq_managed && dev->irq) {
+        if (io_apic_assign_pci_irqs && !mp_should_keep_irq(&dev->dev) &&
+            dev->irq_managed && dev->irq) {
                mp_unmap_irq(dev->irq);
                dev->irq = 0;
                dev->irq_managed = 0;
diff --git a/arch/x86/vdso/vdso32/sigreturn.S b/arch/x86/vdso/vdso32/sigreturn.S
index 31776d0efc8c..d7ec4e251c0a 100644
--- a/arch/x86/vdso/vdso32/sigreturn.S
+++ b/arch/x86/vdso/vdso32/sigreturn.S
@@ -17,6 +17,7 @@
        .text
        .globl __kernel_sigreturn
        .type __kernel_sigreturn,@function
+        nop /* this guy is needed for .LSTARTFDEDLSI1 below (watch for HACK) */
        ALIGN
 __kernel_sigreturn:
 .LSTART_sigreturn:
diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
index 740ae3026a14..9f93af56a5fc 100644
--- a/arch/x86/xen/p2m.c
+++ b/arch/x86/xen/p2m.c
@@ -563,7 +563,7 @@ static bool alloc_p2m(unsigned long pfn)
                if (p2m_pfn == PFN_DOWN(__pa(p2m_missing)))
                        p2m_init(p2m);
                else
-                        p2m_init_identity(p2m, pfn);
+                        p2m_init_identity(p2m, pfn & ~(P2M_PER_PAGE - 1));
                spin_lock_irqsave(&p2m_update_lock, flags);