diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index f481780..db3b1ab 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -843,6 +843,7 @@
  TASKLET:          0          0          0        290
    SCHED:      27035      26983      26971      26746
  HRTIMER:          0          0          0          0
+     RCU:       1678       1769       2178       2250
 
 
 1.3 IDE devices in /proc/ide
diff --git a/MAINTAINERS b/MAINTAINERS
index 6c59eb9..b12b8c1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6463,7 +6463,7 @@
 L:	linux-usb@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid.git
 S:	Maintained
-F:	Documentation/usb/hiddev.txt
+F:	Documentation/hid/hiddev.txt
 F:	drivers/hid/usbhid/
 
 USB ISP116X DRIVER
@@ -7007,6 +7007,13 @@
 S:	Maintained
 F:	drivers/platform/x86
 
+X86 MCE INFRASTRUCTURE
+M:	Tony Luck <tony.luck@intel.com>
+M:	Borislav Petkov <bp@amd64.org>
+L:	linux-edac@vger.kernel.org
+S:	Maintained
+F:	arch/x86/kernel/cpu/mcheck/*
+
 XEN HYPERVISOR INTERFACE
 M:	Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
 M:	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
diff --git a/arch/arm/mach-omap1/Makefile b/arch/arm/mach-omap1/Makefile
index af98117..5b114d1 100644
--- a/arch/arm/mach-omap1/Makefile
+++ b/arch/arm/mach-omap1/Makefile
@@ -4,14 +4,14 @@
 
 # Common support
 obj-y := io.o id.o sram.o time.o irq.o mux.o flash.o serial.o devices.o dma.o
-obj-y += clock.o clock_data.o opp_data.o reset.o
+obj-y += clock.o clock_data.o opp_data.o reset.o pm_bus.o
 
 obj-$(CONFIG_OMAP_MCBSP) += mcbsp.o
 
 obj-$(CONFIG_OMAP_32K_TIMER)	+= timer32k.o
 
 # Power Management
-obj-$(CONFIG_PM) += pm.o sleep.o pm_bus.o
+obj-$(CONFIG_PM) += pm.o sleep.o
 
 # DSP
 obj-$(CONFIG_OMAP_MBOX_FWK)	+= mailbox_mach.o
diff --git a/arch/arm/mach-omap1/pm_bus.c b/arch/arm/mach-omap1/pm_bus.c
index fe31d93..334fb88 100644
--- a/arch/arm/mach-omap1/pm_bus.c
+++ b/arch/arm/mach-omap1/pm_bus.c
@@ -56,9 +56,13 @@
 		USE_PLATFORM_PM_SLEEP_OPS
 	},
 };
+#define OMAP1_PWR_DOMAIN (&default_power_domain)
+#else
+#define OMAP1_PWR_DOMAIN NULL
+#endif /* CONFIG_PM_RUNTIME */
 
 static struct pm_clk_notifier_block platform_bus_notifier = {
-	.pwr_domain = &default_power_domain,
+	.pwr_domain = OMAP1_PWR_DOMAIN,
 	.con_ids = { "ick", "fck", NULL, },
 };
 
@@ -72,4 +76,4 @@
 	return 0;
 }
 core_initcall(omap1_pm_runtime_init);
-#endif /* CONFIG_PM_RUNTIME */
+
diff --git a/arch/arm/mach-omap2/board-omap3pandora.c b/arch/arm/mach-omap2/board-omap3pandora.c
index 2a0bb48..23f71d4 100644
--- a/arch/arm/mach-omap2/board-omap3pandora.c
+++ b/arch/arm/mach-omap2/board-omap3pandora.c
@@ -84,7 +84,8 @@
 
 static struct omap_nand_platform_data pandora_nand_data = {
 	.cs		= 0,
-	.devsize	= 1,	/* '0' for 8-bit, '1' for 16-bit device */
+	.devsize	= NAND_BUSWIDTH_16,
+	.xfer_type	= NAND_OMAP_PREFETCH_DMA,
 	.parts		= omap3pandora_nand_partitions,
 	.nr_parts	= ARRAY_SIZE(omap3pandora_nand_partitions),
 };
diff --git a/arch/arm/mach-omap2/pm-debug.c b/arch/arm/mach-omap2/pm-debug.c
index a5a83b3..e01da45 100644
--- a/arch/arm/mach-omap2/pm-debug.c
+++ b/arch/arm/mach-omap2/pm-debug.c
@@ -189,7 +189,7 @@
 
 static int pm_dbg_init_done;
 
-static int __init pm_dbg_init(void);
+static int pm_dbg_init(void);
 
 enum {
 	DEBUG_FILE_COUNTERS = 0,
@@ -595,7 +595,7 @@
 
 DEFINE_SIMPLE_ATTRIBUTE(pm_dbg_option_fops, option_get, option_set, "%llu\n");
 
-static int __init pm_dbg_init(void)
+static int pm_dbg_init(void)
 {
 	int i;
 	struct dentry *d;
diff --git a/arch/arm/mach-pxa/spitz_pm.c b/arch/arm/mach-pxa/spitz_pm.c
index 7fe7406..094279a 100644
--- a/arch/arm/mach-pxa/spitz_pm.c
+++ b/arch/arm/mach-pxa/spitz_pm.c
@@ -14,6 +14,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/delay.h>
+#include <linux/gpio.h>
 #include <linux/interrupt.h>
 #include <linux/platform_device.h>
 #include <linux/apm-emulation.h>
diff --git a/arch/arm/plat-omap/omap_device.c b/arch/arm/plat-omap/omap_device.c
index a37b8eb..49fc0df 100644
--- a/arch/arm/plat-omap/omap_device.c
+++ b/arch/arm/plat-omap/omap_device.c
@@ -84,6 +84,7 @@
 #include <linux/io.h>
 #include <linux/clk.h>
 #include <linux/clkdev.h>
+#include <linux/pm_runtime.h>
 
 #include <plat/omap_device.h>
 #include <plat/omap_hwmod.h>
@@ -539,20 +540,34 @@
 static int _od_runtime_suspend(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
+	int ret;
 
-	return omap_device_idle(pdev);
+	ret = pm_generic_runtime_suspend(dev);
+
+	if (!ret)
+		omap_device_idle(pdev);
+
+	return ret;
+}
+
+static int _od_runtime_idle(struct device *dev)
+{
+	return pm_generic_runtime_idle(dev);
 }
 
 static int _od_runtime_resume(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 
-	return omap_device_enable(pdev);
+	omap_device_enable(pdev);
+
+	return pm_generic_runtime_resume(dev);
 }
 
 static struct dev_power_domain omap_device_power_domain = {
 	.ops = {
 		.runtime_suspend = _od_runtime_suspend,
+		.runtime_idle = _od_runtime_idle,
 		.runtime_resume = _od_runtime_resume,
 		USE_PLATFORM_PM_SLEEP_OPS
 	}
diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h
index 19ae14b..0cd3800 100644
--- a/arch/x86/include/asm/memblock.h
+++ b/arch/x86/include/asm/memblock.h
@@ -4,7 +4,6 @@
 #define ARCH_DISCARD_MEMBLOCK
 
 u64 memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align);
-void memblock_x86_to_bootmem(u64 start, u64 end);
 
 void memblock_x86_reserve_range(u64 start, u64 end, char *name);
 void memblock_x86_free_range(u64 start, u64 end);
@@ -19,5 +18,6 @@
 u64 memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align);
 u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit);
 u64 memblock_x86_memory_in_range(u64 addr, u64 limit);
+bool memblock_x86_check_reserved_size(u64 *addrp, u64 *sizep, u64 align);
 
 #endif
diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h
index 31d84ac..a518c0a 100644
--- a/arch/x86/include/asm/pvclock.h
+++ b/arch/x86/include/asm/pvclock.h
@@ -22,6 +22,8 @@
 	u64 product;
 #ifdef __i386__
 	u32 tmp1, tmp2;
+#else
+	ulong tmp;
 #endif
 
 	if (shift < 0)
@@ -42,8 +44,11 @@
 		: "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) );
 #elif defined(__x86_64__)
 	__asm__ (
-		"mul %%rdx ; shrd $32,%%rdx,%%rax"
-		: "=a" (product) : "0" (delta), "d" ((u64)mul_frac) );
+		"mul %[mul_frac] ; shrd $32, %[hi], %[lo]"
+		: [lo]"=a"(product),
+		  [hi]"=d"(tmp)
+		: "0"(delta),
+		  [mul_frac]"rm"((u64)mul_frac));
 #else
 #error implement me!
 #endif
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index bd14bb4..aee3862 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -565,7 +565,7 @@
 
 static bool mapping_level_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t large_gfn)
 {
-	return gfn_to_memslot_dirty_bitmap(vcpu, large_gfn, true);
+	return !gfn_to_memslot_dirty_bitmap(vcpu, large_gfn, true);
 }
 
 static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn)
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 6c4dc01..9d03ad4 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -121,7 +121,7 @@
 				    gva_t addr, u32 access)
 {
 	pt_element_t pte;
-	pt_element_t __user *ptep_user;
+	pt_element_t __user *uninitialized_var(ptep_user);
 	gfn_t table_gfn;
 	unsigned index, pt_access, uninitialized_var(pte_access);
 	gpa_t pte_gpa;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 4c3fa0f..d48ec60 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2047,7 +2047,8 @@
 					unsigned long cr0,
 					struct kvm_vcpu *vcpu)
 {
-	vmx_decache_cr3(vcpu);
+	if (!test_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail))
+		vmx_decache_cr3(vcpu);
 	if (!(cr0 & X86_CR0_PG)) {
 		/* From paging/starting to nonpaging */
 		vmcs_write32(CPU_BASED_VM_EXEC_CONTROL,
diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c
index aa11693..992da5e 100644
--- a/arch/x86/mm/memblock.c
+++ b/arch/x86/mm/memblock.c
@@ -8,7 +8,7 @@
 #include <linux/range.h>
 
 /* Check for already reserved areas */
-static bool __init check_with_memblock_reserved_size(u64 *addrp, u64 *sizep, u64 align)
+bool __init memblock_x86_check_reserved_size(u64 *addrp, u64 *sizep, u64 align)
 {
 	struct memblock_region *r;
 	u64 addr = *addrp, last;
@@ -59,7 +59,7 @@
 		if (addr >= ei_last)
 			continue;
 		*sizep = ei_last - addr;
-		while (check_with_memblock_reserved_size(&addr, sizep, align))
+		while (memblock_x86_check_reserved_size(&addr, sizep, align))
 			;
 
 		if (*sizep)
diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
index 0d3a4fa..474356b 100644
--- a/arch/x86/platform/efi/efi.c
+++ b/arch/x86/platform/efi/efi.c
@@ -310,14 +310,31 @@
 
 	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
 		efi_memory_desc_t *md = p;
-		unsigned long long start = md->phys_addr;
-		unsigned long long size = md->num_pages << EFI_PAGE_SHIFT;
+		u64 start = md->phys_addr;
+		u64 size = md->num_pages << EFI_PAGE_SHIFT;
 
 		if (md->type != EFI_BOOT_SERVICES_CODE &&
 		    md->type != EFI_BOOT_SERVICES_DATA)
 			continue;
-
-		memblock_x86_reserve_range(start, start + size, "EFI Boot");
+		/* Only reserve where possible:
+		 * - Not within any already allocated areas
+		 * - Not over any memory area (really needed, if above?)
+		 * - Not within any part of the kernel
+		 * - Not the bios reserved area
+		*/
+		if ((start+size >= virt_to_phys(_text)
+				&& start <= virt_to_phys(_end)) ||
+			!e820_all_mapped(start, start+size, E820_RAM) ||
+			memblock_x86_check_reserved_size(&start, &size,
+							1<<EFI_PAGE_SHIFT)) {
+			/* Could not reserve, skip it */
+			md->num_pages = 0;
+			memblock_dbg(PFX "Could not reserve boot range "
+					"[0x%010llx-0x%010llx]\n",
+						start, start+size-1);
+		} else
+			memblock_x86_reserve_range(start, start+size,
+							"EFI Boot");
 	}
 }
 
@@ -334,6 +351,10 @@
 		    md->type != EFI_BOOT_SERVICES_DATA)
 			continue;
 
+		/* Could not reserve boot area */
+		if (!size)
+			continue;
+
 		free_bootmem_late(start, size);
 	}
 }
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index dd7b88f..5525163 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1033,6 +1033,13 @@
 	xen_reboot(SHUTDOWN_poweroff);
 }
 
+static void xen_machine_power_off(void)
+{
+	if (pm_power_off)
+		pm_power_off();
+	xen_reboot(SHUTDOWN_poweroff);
+}
+
 static void xen_crash_shutdown(struct pt_regs *regs)
 {
 	xen_reboot(SHUTDOWN_crash);
@@ -1058,7 +1065,7 @@
 static const struct machine_ops xen_machine_ops __initconst = {
 	.restart = xen_restart,
 	.halt = xen_machine_halt,
-	.power_off = xen_machine_halt,
+	.power_off = xen_machine_power_off,
 	.shutdown = xen_machine_halt,
 	.crash_shutdown = xen_crash_shutdown,
 	.emergency_restart = xen_emergency_restart,
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index dc708dc..673e968 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -59,6 +59,7 @@
 #include <asm/page.h>
 #include <asm/init.h>
 #include <asm/pat.h>
+#include <asm/smp.h>
 
 #include <asm/xen/hypercall.h>
 #include <asm/xen/hypervisor.h>
@@ -1231,7 +1232,7 @@
 {
 	struct {
 		struct mmuext_op op;
-		DECLARE_BITMAP(mask, NR_CPUS);
+		DECLARE_BITMAP(mask, num_processors);
 	} *args;
 	struct multicall_space mcs;
 
@@ -1599,6 +1600,11 @@
 		for (pteidx = 0; pteidx < PTRS_PER_PTE; pteidx++, pfn++) {
 			pte_t pte;
 
+#ifdef CONFIG_X86_32
+			if (pfn > max_pfn_mapped)
+				max_pfn_mapped = pfn;
+#endif
+
 			if (!pte_none(pte_page[pteidx]))
 				continue;
 
@@ -1766,7 +1772,9 @@
 	initial_kernel_pmd =
 		extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE);
 
-	max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->mfn_list));
+	max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->pt_base) +
+				  xen_start_info->nr_pt_frames * PAGE_SIZE +
+				  512*1024);
 
 	kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd);
 	memcpy(initial_kernel_pmd, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD);
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index be1a464..60aeeb5 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -227,11 +227,7 @@
 
 	memcpy(map_raw, map, sizeof(map));
 	e820.nr_map = 0;
-#ifdef CONFIG_X86_32
 	xen_extra_mem_start = mem_end;
-#else
-	xen_extra_mem_start = max((1ULL << 32), mem_end);
-#endif
 	for (i = 0; i < memmap.nr_entries; i++) {
 		unsigned long long end;
 
@@ -266,6 +262,12 @@
 		if (map[i].size > 0)
 			e820_add_region(map[i].addr, map[i].size, map[i].type);
 	}
+	/* Align the balloon area so that max_low_pfn does not get set
+	 * to be at the _end_ of the PCI gap at the far end (fee01000).
+	 * Note that xen_extra_mem_start gets set in the loop above to be
+	 * past the last E820 region. */
+	if (xen_initial_domain() && (xen_extra_mem_start < (1ULL<<32)))
+		xen_extra_mem_start = (1ULL<<32);
 
 	/*
 	 * In domU, the ISA region is normal, usable memory, but we
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 41038c0..b4533a8 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -205,11 +205,18 @@
 static void __init xen_smp_prepare_cpus(unsigned int max_cpus)
 {
 	unsigned cpu;
+	unsigned int i;
 
 	xen_init_lock_cpu(0);
 
 	smp_store_cpu_info(0);
 	cpu_data(0).x86_max_cores = 1;
+
+	for_each_possible_cpu(i) {
+		zalloc_cpumask_var(&per_cpu(cpu_sibling_map, i), GFP_KERNEL);
+		zalloc_cpumask_var(&per_cpu(cpu_core_map, i), GFP_KERNEL);
+		zalloc_cpumask_var(&per_cpu(cpu_llc_shared_map, i), GFP_KERNEL);
+	}
 	set_cpu_sibling_map(0);
 
 	if (xen_smp_intr_init(0))
diff --git a/drivers/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c
index 853f92d..faf7c52 100644
--- a/drivers/cpufreq/cpufreq_stats.c
+++ b/drivers/cpufreq/cpufreq_stats.c
@@ -298,11 +298,13 @@
 	old_index = stat->last_index;
 	new_index = freq_table_get_index(stat, freq->new);
 
-	cpufreq_stats_update(freq->cpu);
-	if (old_index == new_index)
+	/* We can't do stat->time_in_state[-1]= .. */
+	if (old_index == -1 || new_index == -1)
 		return 0;
 
-	if (old_index == -1 || new_index == -1)
+	cpufreq_stats_update(freq->cpu);
+
+	if (old_index == new_index)
 		return 0;
 
 	spin_lock(&cpufreq_stats_lock);
diff --git a/drivers/cpufreq/powernow-k8.c b/drivers/cpufreq/powernow-k8.c
index 83479b6..bce576d 100644
--- a/drivers/cpufreq/powernow-k8.c
+++ b/drivers/cpufreq/powernow-k8.c
@@ -1079,6 +1079,9 @@
 	}
 
 	res = transition_fid_vid(data, fid, vid);
+	if (res)
+		return res;
+
 	freqs.new = find_khz_freq_from_fid(data->currfid);
 
 	for_each_cpu(i, data->available_cores) {
@@ -1101,7 +1104,8 @@
 	/* get MSR index for hardware pstate transition */
 	pstate = index & HW_PSTATE_MASK;
 	if (pstate > data->max_hw_pstate)
-		return 0;
+		return -EINVAL;
+
 	freqs.old = find_khz_freq_from_pstate(data->powernow_table,
 			data->currpstate);
 	freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate);
diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 01f74a8..35bebde 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -469,8 +469,9 @@
 					+ OMAP24XX_GPIO_CLEARWKUENA);
 		}
 	}
-	/* This part needs to be executed always for OMAP34xx */
-	if (cpu_is_omap34xx() || (bank->non_wakeup_gpios & gpio_bit)) {
+	/* This part needs to be executed always for OMAP{34xx, 44xx} */
+	if (cpu_is_omap34xx() || cpu_is_omap44xx() ||
+			(bank->non_wakeup_gpios & gpio_bit)) {
 		/*
 		 * Log the edge gpio and manually trigger the IRQ
 		 * after resume if the input level changes
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 0a9357c..0929219 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -184,9 +184,9 @@
 
 bad:
 	if (raw_edid) {
-		DRM_ERROR("Raw EDID:\n");
+		printk(KERN_ERR "Raw EDID:\n");
 		print_hex_dump_bytes(KERN_ERR, DUMP_PREFIX_NONE, raw_edid, EDID_LENGTH);
-		printk("\n");
+		printk(KERN_ERR "\n");
 	}
 	return 0;
 }
@@ -258,6 +258,17 @@
 	return ret == 2 ? 0 : -1;
 }
 
+static bool drm_edid_is_zero(u8 *in_edid, int length)
+{
+	int i;
+	u32 *raw_edid = (u32 *)in_edid;
+
+	for (i = 0; i < length / 4; i++)
+		if (*(raw_edid + i) != 0)
+			return false;
+	return true;
+}
+
 static u8 *
 drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 {
@@ -273,6 +284,10 @@
 			goto out;
 		if (drm_edid_block_valid(block))
 			break;
+		if (i == 0 && drm_edid_is_zero(block, EDID_LENGTH)) {
+			connector->null_edid_counter++;
+			goto carp;
+		}
 	}
 	if (i == 4)
 		goto carp;
diff --git a/drivers/gpu/drm/drm_ioc32.c b/drivers/gpu/drm/drm_ioc32.c
index d61d185..4a058c7 100644
--- a/drivers/gpu/drm/drm_ioc32.c
+++ b/drivers/gpu/drm/drm_ioc32.c
@@ -28,6 +28,7 @@
  * IN THE SOFTWARE.
  */
 #include <linux/compat.h>
+#include <linux/ratelimit.h>
 
 #include "drmP.h"
 #include "drm_core.h"
@@ -253,10 +254,10 @@
 		return -EFAULT;
 
 	m32.handle = (unsigned long)handle;
-	if (m32.handle != (unsigned long)handle && printk_ratelimit())
-		printk(KERN_ERR "compat_drm_addmap truncated handle"
-		       " %p for type %d offset %x\n",
-		       handle, m32.type, m32.offset);
+	if (m32.handle != (unsigned long)handle)
+		printk_ratelimited(KERN_ERR "compat_drm_addmap truncated handle"
+				   " %p for type %d offset %x\n",
+				   handle, m32.type, m32.offset);
 
 	if (copy_to_user(argp, &m32, sizeof(m32)))
 		return -EFAULT;
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index e1aee4f..b6a19cb 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -251,7 +251,7 @@
 }
 
 
-int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
+static int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
 {
 	if ((p->busnum >> 8) != drm_get_pci_domain(dev) ||
 	    (p->busnum & 0xff) != dev->pdev->bus->number ||
@@ -292,6 +292,7 @@
 	.get_name = drm_pci_get_name,
 	.set_busid = drm_pci_set_busid,
 	.set_unique = drm_pci_set_unique,
+	.irq_by_busid = drm_pci_irq_by_busid,
 	.agp_init = drm_pci_agp_init,
 };
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index b9fafe3..9e34a1a 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1740,6 +1740,16 @@
 		INIT_WORK(&dev_priv->rps_work, gen6_pm_rps_work);
 
 	I915_WRITE(HWSTAM, 0xeffe);
+	if (IS_GEN6(dev)) {
+		/* Workaround stalls observed on Sandy Bridge GPUs by
+		 * making the blitter command streamer generate a
+		 * write to the Hardware Status Page for
+		 * MI_USER_INTERRUPT.  This appears to serialize the
+		 * previous seqno write out before the interrupt
+		 * happens.
+		 */
+		I915_WRITE(GEN6_BLITTER_HWSTAM, ~GEN6_BLITTER_USER_INTERRUPT);
+	}
 
 	/* XXX hotplug from PCH */
 
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index d3b903b..d98cee6 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -401,8 +401,7 @@
 		bus->reg0 = i | GMBUS_RATE_100KHZ;
 
 		/* XXX force bit banging until GMBUS is fully debugged */
-		if (IS_GEN2(dev))
-			bus->force_bit = intel_gpio_create(dev_priv, i);
+		bus->force_bit = intel_gpio_create(dev_priv, i);
 	}
 
 	intel_i2c_reset(dev_priv->dev);
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index 49611e2..1b50ad8 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -1200,6 +1200,7 @@
 #define EXTERNAL_ENCODER_ACTION_V3_ENCODER_BLANKING_OFF   0x10
 #define EXTERNAL_ENCODER_ACTION_V3_ENCODER_BLANKING       0x11
 #define EXTERNAL_ENCODER_ACTION_V3_DACLOAD_DETECTION      0x12
+#define EXTERNAL_ENCODER_ACTION_V3_DDC_SETUP              0x14
 
 // ucConfig
 #define EXTERNAL_ENCODER_CONFIG_V3_DPLINKRATE_MASK				0x03
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 84a69e7..9541995 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -671,6 +671,13 @@
 								DISPPLL_CONFIG_DUAL_LINK;
 					}
 				}
+				if (radeon_encoder_is_dp_bridge(encoder)) {
+					struct drm_encoder *ext_encoder = radeon_atom_get_external_encoder(encoder);
+					struct radeon_encoder *ext_radeon_encoder = to_radeon_encoder(ext_encoder);
+					args.v3.sInput.ucExtTransmitterID = ext_radeon_encoder->encoder_id;
+				} else
+					args.v3.sInput.ucExtTransmitterID = 0;
+
 				atom_execute_table(rdev->mode_info.atom_context,
 						   index, (uint32_t *)&args);
 				adjusted_clock = le32_to_cpu(args.v3.sOutput.ulDispPllFreq) * 10;
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 86157b1..7e3d96e 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -88,7 +88,8 @@
 /* get temperature in millidegrees */
 int evergreen_get_temp(struct radeon_device *rdev)
 {
-	u32 temp, toffset, actual_temp = 0;
+	u32 temp, toffset;
+	int actual_temp = 0;
 
 	if (rdev->family == CHIP_JUNIPER) {
 		toffset = (RREG32(CG_THERMAL_CTRL) & TOFFSET_MASK) >>
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 9bd162f..b244962 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -938,6 +938,13 @@
 int radeon_asic_init(struct radeon_device *rdev)
 {
 	radeon_register_accessor_init(rdev);
+
+	/* set the number of crtcs */
+	if (rdev->flags & RADEON_SINGLE_CRTC)
+		rdev->num_crtc = 1;
+	else
+		rdev->num_crtc = 2;
+
 	switch (rdev->family) {
 	case CHIP_R100:
 	case CHIP_RV100:
@@ -1017,6 +1024,11 @@
 	case CHIP_JUNIPER:
 	case CHIP_CYPRESS:
 	case CHIP_HEMLOCK:
+		/* set num crtcs */
+		if (rdev->family == CHIP_CEDAR)
+			rdev->num_crtc = 4;
+		else
+			rdev->num_crtc = 6;
 		rdev->asic = &evergreen_asic;
 		break;
 	case CHIP_PALM:
@@ -1027,10 +1039,17 @@
 	case CHIP_BARTS:
 	case CHIP_TURKS:
 	case CHIP_CAICOS:
+		/* set num crtcs */
+		if (rdev->family == CHIP_CAICOS)
+			rdev->num_crtc = 4;
+		else
+			rdev->num_crtc = 6;
 		rdev->asic = &btc_asic;
 		break;
 	case CHIP_CAYMAN:
 		rdev->asic = &cayman_asic;
+		/* set num crtcs */
+		rdev->num_crtc = 6;
 		break;
 	default:
 		/* FIXME: not supported yet */
@@ -1042,18 +1061,6 @@
 		rdev->asic->set_memory_clock = NULL;
 	}
 
-	/* set the number of crtcs */
-	if (rdev->flags & RADEON_SINGLE_CRTC)
-		rdev->num_crtc = 1;
-	else {
-		if (ASIC_IS_DCE41(rdev))
-			rdev->num_crtc = 2;
-		else if (ASIC_IS_DCE4(rdev))
-			rdev->num_crtc = 6;
-		else
-			rdev->num_crtc = 2;
-	}
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 797c8bc..e459467 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -1553,9 +1553,12 @@
 			   (rdev->pdev->subsystem_device == 0x4a48)) {
 			/* Mac X800 */
 			rdev->mode_info.connector_table = CT_MAC_X800;
-		} else if (of_machine_is_compatible("PowerMac7,2") ||
-			   of_machine_is_compatible("PowerMac7,3")) {
-			/* Mac G5 9600 */
+		} else if ((of_machine_is_compatible("PowerMac7,2") ||
+			    of_machine_is_compatible("PowerMac7,3")) &&
+			   (rdev->pdev->device == 0x4150) &&
+			   (rdev->pdev->subsystem_vendor == 0x1002) &&
+			   (rdev->pdev->subsystem_device == 0x4150)) {
+			/* Mac G5 tower 9600 */
 			rdev->mode_info.connector_table = CT_MAC_G5_9600;
 		} else
 #endif /* CONFIG_PPC_PMAC */
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 9c2929c..cbfca3a 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -44,6 +44,8 @@
 radeon_legacy_backlight_init(struct radeon_encoder *radeon_encoder,
 			     struct drm_connector *drm_connector);
 
+bool radeon_connector_encoder_is_dp_bridge(struct drm_connector *connector);
+
 void radeon_connector_hotplug(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
@@ -836,6 +838,13 @@
 		if (!radeon_connector->edid) {
 			DRM_ERROR("%s: probed a monitor but no|invalid EDID\n",
 					drm_get_connector_name(connector));
+			/* rs690 seems to have a problem with connectors not existing and always
+			 * return a block of 0's. If we see this just stop polling on this output */
+			if ((rdev->family == CHIP_RS690 || rdev->family == CHIP_RS740) && radeon_connector->base.null_edid_counter) {
+				ret = connector_status_disconnected;
+				DRM_ERROR("%s: detected RS690 floating bus bug, stopping ddc detect\n", drm_get_connector_name(connector));
+				radeon_connector->ddc_bus = NULL;
+			}
 		} else {
 			radeon_connector->use_digital = !!(radeon_connector->edid->input & DRM_EDID_INPUT_DIGITAL);
 
@@ -1063,10 +1072,11 @@
 {
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	struct radeon_connector_atom_dig *radeon_dig_connector = radeon_connector->con_priv;
+	struct drm_encoder *encoder = radeon_best_single_encoder(connector);
 	int ret;
 
-	if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
-		struct drm_encoder *encoder;
+	if ((connector->connector_type == DRM_MODE_CONNECTOR_eDP) ||
+	    (connector->connector_type == DRM_MODE_CONNECTOR_LVDS)) {
 		struct drm_display_mode *mode;
 
 		if (!radeon_dig_connector->edp_on)
@@ -1078,7 +1088,6 @@
 						     ATOM_TRANSMITTER_ACTION_POWER_OFF);
 
 		if (ret > 0) {
-			encoder = radeon_best_single_encoder(connector);
 			if (encoder) {
 				radeon_fixup_lvds_native_mode(encoder, connector);
 				/* add scaled modes */
@@ -1102,8 +1111,14 @@
 			/* add scaled modes */
 			radeon_add_common_modes(encoder, connector);
 		}
-	} else
+	} else {
+		/* need to setup ddc on the bridge */
+		if (radeon_connector_encoder_is_dp_bridge(connector)) {
+			if (encoder)
+				radeon_atom_ext_encoder_setup_ddc(encoder);
+		}
 		ret = radeon_ddc_get_modes(radeon_connector);
+	}
 
 	return ret;
 }
@@ -1187,14 +1202,15 @@
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	enum drm_connector_status ret = connector_status_disconnected;
 	struct radeon_connector_atom_dig *radeon_dig_connector = radeon_connector->con_priv;
+	struct drm_encoder *encoder = radeon_best_single_encoder(connector);
 
 	if (radeon_connector->edid) {
 		kfree(radeon_connector->edid);
 		radeon_connector->edid = NULL;
 	}
 
-	if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
-		struct drm_encoder *encoder = radeon_best_single_encoder(connector);
+	if ((connector->connector_type == DRM_MODE_CONNECTOR_eDP) ||
+	    (connector->connector_type == DRM_MODE_CONNECTOR_LVDS)) {
 		if (encoder) {
 			struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 			struct drm_display_mode *native_mode = &radeon_encoder->native_mode;
@@ -1214,6 +1230,11 @@
 			atombios_set_edp_panel_power(connector,
 						     ATOM_TRANSMITTER_ACTION_POWER_OFF);
 	} else {
+		/* need to setup ddc on the bridge */
+		if (radeon_connector_encoder_is_dp_bridge(connector)) {
+			if (encoder)
+				radeon_atom_ext_encoder_setup_ddc(encoder);
+		}
 		radeon_dig_connector->dp_sink_type = radeon_dp_getsinktype(radeon_connector);
 		if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) {
 			ret = connector_status_connected;
@@ -1228,6 +1249,16 @@
 					ret = connector_status_connected;
 			}
 		}
+
+		if ((ret == connector_status_disconnected) &&
+		    radeon_connector->dac_load_detect) {
+			struct drm_encoder *encoder = radeon_best_single_encoder(connector);
+			struct drm_encoder_helper_funcs *encoder_funcs;
+			if (encoder) {
+				encoder_funcs = encoder->helper_private;
+				ret = encoder_funcs->detect(encoder, connector);
+			}
+		}
 	}
 
 	radeon_connector_update_scratch_regs(connector, ret);
@@ -1242,7 +1273,8 @@
 
 	/* XXX check mode bandwidth */
 
-	if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
+	if ((connector->connector_type == DRM_MODE_CONNECTOR_eDP) ||
+	    (connector->connector_type == DRM_MODE_CONNECTOR_LVDS)) {
 		struct drm_encoder *encoder = radeon_best_single_encoder(connector);
 
 		if ((mode->hdisplay < 320) || (mode->vdisplay < 240))
@@ -1252,7 +1284,7 @@
 			struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 			struct drm_display_mode *native_mode = &radeon_encoder->native_mode;
 
-		/* AVIVO hardware supports downscaling modes larger than the panel
+			/* AVIVO hardware supports downscaling modes larger than the panel
 			 * to the panel size, but I'm not sure this is desirable.
 			 */
 			if ((mode->hdisplay > native_mode->hdisplay) ||
@@ -1401,6 +1433,10 @@
 		default:
 			connector->interlace_allowed = true;
 			connector->doublescan_allowed = true;
+			radeon_connector->dac_load_detect = true;
+			drm_connector_attach_property(&radeon_connector->base,
+						      rdev->mode_info.load_detect_property,
+						      1);
 			break;
 		case DRM_MODE_CONNECTOR_DVII:
 		case DRM_MODE_CONNECTOR_DVID:
@@ -1422,6 +1458,12 @@
 				connector->doublescan_allowed = true;
 			else
 				connector->doublescan_allowed = false;
+			if (connector_type == DRM_MODE_CONNECTOR_DVII) {
+				radeon_connector->dac_load_detect = true;
+				drm_connector_attach_property(&radeon_connector->base,
+							      rdev->mode_info.load_detect_property,
+							      1);
+			}
 			break;
 		case DRM_MODE_CONNECTOR_LVDS:
 		case DRM_MODE_CONNECTOR_eDP:
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index e680501..7cfaa7e 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -215,6 +215,8 @@
 		return r;
 	}
 
+	/* clear wb memory */
+	memset((char *)rdev->wb.wb, 0, RADEON_GPU_PAGE_SIZE);
 	/* disable event_write fences */
 	rdev->wb.use_event = false;
 	/* disabled via module param */
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index 03f124d..f55b64c 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -367,7 +367,8 @@
 	}
 
 	if (ASIC_IS_DCE3(rdev) &&
-	    (radeon_encoder->active_device & (ATOM_DEVICE_DFP_SUPPORT | ATOM_DEVICE_LCD_SUPPORT))) {
+	    ((radeon_encoder->active_device & (ATOM_DEVICE_DFP_SUPPORT | ATOM_DEVICE_LCD_SUPPORT)) ||
+	     radeon_encoder_is_dp_bridge(encoder))) {
 		struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 		radeon_dp_set_link_config(connector, mode);
 	}
@@ -660,21 +661,16 @@
 	if (radeon_encoder_is_dp_bridge(encoder))
 		return ATOM_ENCODER_MODE_DP;
 
+	/* DVO is always DVO */
+	if (radeon_encoder->encoder_id == ATOM_ENCODER_MODE_DVO)
+		return ATOM_ENCODER_MODE_DVO;
+
 	connector = radeon_get_connector_for_encoder(encoder);
-	if (!connector) {
-		switch (radeon_encoder->encoder_id) {
-		case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
-		case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1:
-		case ENCODER_OBJECT_ID_INTERNAL_UNIPHY2:
-		case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_LVTMA:
-		case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1:
-			return ATOM_ENCODER_MODE_DVI;
-		case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1:
-		case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2:
-		default:
-			return ATOM_ENCODER_MODE_CRT;
-		}
-	}
+	/* if we don't have an active device yet, just use one of
+	 * the connectors tied to the encoder.
+	 */
+	if (!connector)
+		connector = radeon_get_connector_for_encoder_init(encoder);
 	radeon_connector = to_radeon_connector(connector);
 
 	switch (connector->connector_type) {
@@ -1526,26 +1522,29 @@
 	}
 
 	if (ext_encoder) {
-		int action;
-
 		switch (mode) {
 		case DRM_MODE_DPMS_ON:
 		default:
-			if (ASIC_IS_DCE41(rdev))
-				action = EXTERNAL_ENCODER_ACTION_V3_ENABLE_OUTPUT;
-			else
-				action = ATOM_ENABLE;
+			if (ASIC_IS_DCE41(rdev)) {
+				atombios_external_encoder_setup(encoder, ext_encoder,
+								EXTERNAL_ENCODER_ACTION_V3_ENABLE_OUTPUT);
+				atombios_external_encoder_setup(encoder, ext_encoder,
+								EXTERNAL_ENCODER_ACTION_V3_ENCODER_BLANKING_OFF);
+			} else
+				atombios_external_encoder_setup(encoder, ext_encoder, ATOM_ENABLE);
 			break;
 		case DRM_MODE_DPMS_STANDBY:
 		case DRM_MODE_DPMS_SUSPEND:
 		case DRM_MODE_DPMS_OFF:
-			if (ASIC_IS_DCE41(rdev))
-				action = EXTERNAL_ENCODER_ACTION_V3_DISABLE_OUTPUT;
-			else
-				action = ATOM_DISABLE;
+			if (ASIC_IS_DCE41(rdev)) {
+				atombios_external_encoder_setup(encoder, ext_encoder,
+								EXTERNAL_ENCODER_ACTION_V3_ENCODER_BLANKING);
+				atombios_external_encoder_setup(encoder, ext_encoder,
+								EXTERNAL_ENCODER_ACTION_V3_DISABLE_OUTPUT);
+			} else
+				atombios_external_encoder_setup(encoder, ext_encoder, ATOM_DISABLE);
 			break;
 		}
-		atombios_external_encoder_setup(encoder, ext_encoder, action);
 	}
 
 	radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
@@ -2004,6 +2003,65 @@
 	return connector_status_disconnected;
 }
 
+static enum drm_connector_status
+radeon_atom_dig_detect(struct drm_encoder *encoder, struct drm_connector *connector)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+	struct drm_encoder *ext_encoder = radeon_atom_get_external_encoder(encoder);
+	u32 bios_0_scratch;
+
+	if (!ASIC_IS_DCE4(rdev))
+		return connector_status_unknown;
+
+	if (!ext_encoder)
+		return connector_status_unknown;
+
+	if ((radeon_connector->devices & ATOM_DEVICE_CRT_SUPPORT) == 0)
+		return connector_status_unknown;
+
+	/* load detect on the dp bridge */
+	atombios_external_encoder_setup(encoder, ext_encoder,
+					EXTERNAL_ENCODER_ACTION_V3_DACLOAD_DETECTION);
+
+	bios_0_scratch = RREG32(R600_BIOS_0_SCRATCH);
+
+	DRM_DEBUG_KMS("Bios 0 scratch %x %08x\n", bios_0_scratch, radeon_encoder->devices);
+	if (radeon_connector->devices & ATOM_DEVICE_CRT1_SUPPORT) {
+		if (bios_0_scratch & ATOM_S0_CRT1_MASK)
+			return connector_status_connected;
+	}
+	if (radeon_connector->devices & ATOM_DEVICE_CRT2_SUPPORT) {
+		if (bios_0_scratch & ATOM_S0_CRT2_MASK)
+			return connector_status_connected;
+	}
+	if (radeon_connector->devices & ATOM_DEVICE_CV_SUPPORT) {
+		if (bios_0_scratch & (ATOM_S0_CV_MASK|ATOM_S0_CV_MASK_A))
+			return connector_status_connected;
+	}
+	if (radeon_connector->devices & ATOM_DEVICE_TV1_SUPPORT) {
+		if (bios_0_scratch & (ATOM_S0_TV1_COMPOSITE | ATOM_S0_TV1_COMPOSITE_A))
+			return connector_status_connected; /* CTV */
+		else if (bios_0_scratch & (ATOM_S0_TV1_SVIDEO | ATOM_S0_TV1_SVIDEO_A))
+			return connector_status_connected; /* STV */
+	}
+	return connector_status_disconnected;
+}
+
+void
+radeon_atom_ext_encoder_setup_ddc(struct drm_encoder *encoder)
+{
+	struct drm_encoder *ext_encoder = radeon_atom_get_external_encoder(encoder);
+
+	if (ext_encoder)
+		/* ddc_setup on the dp bridge */
+		atombios_external_encoder_setup(encoder, ext_encoder,
+						EXTERNAL_ENCODER_ACTION_V3_DDC_SETUP);
+
+}
+
 static void radeon_atom_encoder_prepare(struct drm_encoder *encoder)
 {
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
@@ -2167,7 +2225,7 @@
 	.mode_set = radeon_atom_encoder_mode_set,
 	.commit = radeon_atom_encoder_commit,
 	.disable = radeon_atom_encoder_disable,
-	/* no detect for TMDS/LVDS yet */
+	.detect = radeon_atom_dig_detect,
 };
 
 static const struct drm_encoder_helper_funcs radeon_atom_dac_helper_funcs = {
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 1f82294..021d2b6 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -40,6 +40,35 @@
 #include "radeon.h"
 #include "radeon_trace.h"
 
+static void radeon_fence_write(struct radeon_device *rdev, u32 seq)
+{
+	if (rdev->wb.enabled) {
+		u32 scratch_index;
+		if (rdev->wb.use_event)
+			scratch_index = R600_WB_EVENT_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
+		else
+			scratch_index = RADEON_WB_SCRATCH_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
+		rdev->wb.wb[scratch_index/4] = cpu_to_le32(seq);;
+	} else
+		WREG32(rdev->fence_drv.scratch_reg, seq);
+}
+
+static u32 radeon_fence_read(struct radeon_device *rdev)
+{
+	u32 seq;
+
+	if (rdev->wb.enabled) {
+		u32 scratch_index;
+		if (rdev->wb.use_event)
+			scratch_index = R600_WB_EVENT_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
+		else
+			scratch_index = RADEON_WB_SCRATCH_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
+		seq = le32_to_cpu(rdev->wb.wb[scratch_index/4]);
+	} else
+		seq = RREG32(rdev->fence_drv.scratch_reg);
+	return seq;
+}
+
 int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence)
 {
 	unsigned long irq_flags;
@@ -50,12 +79,12 @@
 		return 0;
 	}
 	fence->seq = atomic_add_return(1, &rdev->fence_drv.seq);
-	if (!rdev->cp.ready) {
+	if (!rdev->cp.ready)
 		/* FIXME: cp is not running assume everythings is done right
 		 * away
 		 */
-		WREG32(rdev->fence_drv.scratch_reg, fence->seq);
-	} else
+		radeon_fence_write(rdev, fence->seq);
+	else
 		radeon_fence_ring_emit(rdev, fence);
 
 	trace_radeon_fence_emit(rdev->ddev, fence->seq);
@@ -73,15 +102,7 @@
 	bool wake = false;
 	unsigned long cjiffies;
 
-	if (rdev->wb.enabled) {
-		u32 scratch_index;
-		if (rdev->wb.use_event)
-			scratch_index = R600_WB_EVENT_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
-		else
-			scratch_index = RADEON_WB_SCRATCH_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
-		seq = le32_to_cpu(rdev->wb.wb[scratch_index/4]);
-	} else
-		seq = RREG32(rdev->fence_drv.scratch_reg);
+	seq = radeon_fence_read(rdev);
 	if (seq != rdev->fence_drv.last_seq) {
 		rdev->fence_drv.last_seq = seq;
 		rdev->fence_drv.last_jiffies = jiffies;
@@ -251,7 +272,7 @@
 			r = radeon_gpu_reset(rdev);
 			if (r)
 				return r;
-			WREG32(rdev->fence_drv.scratch_reg, fence->seq);
+			radeon_fence_write(rdev, fence->seq);
 			rdev->gpu_lockup = false;
 		}
 		timeout = RADEON_FENCE_JIFFIES_TIMEOUT;
@@ -351,7 +372,7 @@
 		write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
 		return r;
 	}
-	WREG32(rdev->fence_drv.scratch_reg, 0);
+	radeon_fence_write(rdev, 0);
 	atomic_set(&rdev->fence_drv.seq, 0);
 	INIT_LIST_HEAD(&rdev->fence_drv.created);
 	INIT_LIST_HEAD(&rdev->fence_drv.emited);
@@ -391,7 +412,7 @@
 	struct radeon_fence *fence;
 
 	seq_printf(m, "Last signaled fence 0x%08X\n",
-		   RREG32(rdev->fence_drv.scratch_reg));
+		   radeon_fence_read(rdev));
 	if (!list_empty(&rdev->fence_drv.emited)) {
 		   fence = list_entry(rdev->fence_drv.emited.prev,
 				      struct radeon_fence, list);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 977a341..6df4e3c 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -483,6 +483,8 @@
 extern void atombios_dig_transmitter_setup(struct drm_encoder *encoder,
 					   int action, uint8_t lane_num,
 					   uint8_t lane_set);
+extern void radeon_atom_ext_encoder_setup_ddc(struct drm_encoder *encoder);
+extern struct drm_encoder *radeon_atom_get_external_encoder(struct drm_encoder *encoder);
 extern int radeon_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 				u8 write_byte, u8 *read_byte);
 
diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
index 67d2a75..36ca465 100644
--- a/drivers/hid/Kconfig
+++ b/drivers/hid/Kconfig
@@ -305,6 +305,7 @@
 	  - 3M PCT touch screens
 	  - ActionStar dual touch panels
 	  - Cando dual touch panels
+	  - Chunghwa panels
 	  - CVTouch panels
 	  - Cypress TrueTouch panels
 	  - Elo TouchSystems IntelliTouch Plus panels
diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index c957c4b..f7440e8 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c
@@ -1359,6 +1359,7 @@
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CHERRY, USB_DEVICE_ID_CHERRY_CYMOTION_SOLAR) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CHICONY, USB_DEVICE_ID_CHICONY_TACTICAL_PAD) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CHICONY, USB_DEVICE_ID_CHICONY_WIRELESS) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_CHUNGHWAT, USB_DEVICE_ID_CHUNGHWAT_MULTITOUCH) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CREATIVELABS, USB_DEVICE_ID_PRODIKEYS_PCMIDI) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CVTOUCH, USB_DEVICE_ID_CVTOUCH_SCREEN) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_BARCODE_1) },
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 0b374a6..aecb5a4 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -173,6 +173,9 @@
 #define USB_DEVICE_ID_CHICONY_MULTI_TOUCH	0xb19d
 #define USB_DEVICE_ID_CHICONY_WIRELESS	0x0618
 
+#define USB_VENDOR_ID_CHUNGHWAT		0x2247
+#define USB_DEVICE_ID_CHUNGHWAT_MULTITOUCH	0x0001
+
 #define USB_VENDOR_ID_CIDC		0x1677
 
 #define USB_VENDOR_ID_CMEDIA		0x0d8c
@@ -622,6 +625,7 @@
 #define USB_VENDOR_ID_UCLOGIC		0x5543
 #define USB_DEVICE_ID_UCLOGIC_TABLET_PF1209	0x0042
 #define USB_DEVICE_ID_UCLOGIC_TABLET_KNA5	0x6001
+#define USB_DEVICE_ID_UCLOGIC_TABLET_TWA60	0x0064
 #define USB_DEVICE_ID_UCLOGIC_TABLET_WP4030U	0x0003
 #define USB_DEVICE_ID_UCLOGIC_TABLET_WP5540U	0x0004
 #define USB_DEVICE_ID_UCLOGIC_TABLET_WP8060U	0x0005
diff --git a/drivers/hid/hid-magicmouse.c b/drivers/hid/hid-magicmouse.c
index a5eda4c..0ec91c1 100644
--- a/drivers/hid/hid-magicmouse.c
+++ b/drivers/hid/hid-magicmouse.c
@@ -501,17 +501,9 @@
 	}
 	report->size = 6;
 
-	/*
-	 * The device reponds with 'invalid report id' when feature
-	 * report switching it into multitouch mode is sent to it.
-	 *
-	 * This results in -EIO from the _raw low-level transport callback,
-	 * but there seems to be no other way of switching the mode.
-	 * Thus the super-ugly hacky success check below.
-	 */
 	ret = hdev->hid_output_raw_report(hdev, feature, sizeof(feature),
 			HID_FEATURE_REPORT);
-	if (ret != -EIO) {
+	if (ret != sizeof(feature)) {
 		hid_err(hdev, "unable to request touch data (%d)\n", ret);
 		goto err_stop_hw;
 	}
diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c
index ecd4d2d..0b2dcd0 100644
--- a/drivers/hid/hid-multitouch.c
+++ b/drivers/hid/hid-multitouch.c
@@ -64,6 +64,7 @@
 	struct mt_class *mtclass;	/* our mt device class */
 	unsigned last_field_index;	/* last field index of the report */
 	unsigned last_slot_field;	/* the last field of a slot */
+	int last_mt_collection;	/* last known mt-related collection */
 	__s8 inputmode;		/* InputMode HID feature, -1 if non-existent */
 	__u8 num_received;	/* how many contacts we received */
 	__u8 num_expected;	/* expected last contact index */
@@ -225,8 +226,10 @@
 				cls->sn_move);
 			/* touchscreen emulation */
 			set_abs(hi->input, ABS_X, field, cls->sn_move);
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_GD_Y:
 			if (quirks & MT_QUIRK_EGALAX_XYZ_FIXUP)
@@ -237,8 +240,10 @@
 				cls->sn_move);
 			/* touchscreen emulation */
 			set_abs(hi->input, ABS_Y, field, cls->sn_move);
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		}
 		return 0;
@@ -246,31 +251,40 @@
 	case HID_UP_DIGITIZER:
 		switch (usage->hid) {
 		case HID_DG_INRANGE:
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_DG_CONFIDENCE:
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_DG_TIPSWITCH:
 			hid_map_usage(hi, usage, bit, max, EV_KEY, BTN_TOUCH);
 			input_set_capability(hi->input, EV_KEY, BTN_TOUCH);
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_DG_CONTACTID:
 			input_mt_init_slots(hi->input, td->maxcontacts);
 			td->last_slot_field = usage->hid;
 			td->last_field_index = field->index;
+			td->last_mt_collection = usage->collection_index;
 			return 1;
 		case HID_DG_WIDTH:
 			hid_map_usage(hi, usage, bit, max,
 					EV_ABS, ABS_MT_TOUCH_MAJOR);
 			set_abs(hi->input, ABS_MT_TOUCH_MAJOR, field,
 				cls->sn_width);
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_DG_HEIGHT:
 			hid_map_usage(hi, usage, bit, max,
@@ -279,8 +293,10 @@
 				cls->sn_height);
 			input_set_abs_params(hi->input,
 					ABS_MT_ORIENTATION, 0, 1, 0, 0);
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_DG_TIPPRESSURE:
 			if (quirks & MT_QUIRK_EGALAX_XYZ_FIXUP)
@@ -292,16 +308,20 @@
 			/* touchscreen emulation */
 			set_abs(hi->input, ABS_PRESSURE, field,
 				cls->sn_pressure);
-			td->last_slot_field = usage->hid;
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index) {
+				td->last_slot_field = usage->hid;
+				td->last_field_index = field->index;
+			}
 			return 1;
 		case HID_DG_CONTACTCOUNT:
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index)
+				td->last_field_index = field->index;
 			return 1;
 		case HID_DG_CONTACTMAX:
 			/* we don't set td->last_slot_field as contactcount and
 			 * contact max are global to the report */
-			td->last_field_index = field->index;
+			if (td->last_mt_collection == usage->collection_index)
+				td->last_field_index = field->index;
 			return -1;
 		}
 		/* let hid-input decide for the others */
@@ -516,6 +536,7 @@
 	}
 	td->mtclass = mtclass;
 	td->inputmode = -1;
+	td->last_mt_collection = -1;
 	hid_set_drvdata(hdev, td);
 
 	ret = hid_parse(hdev);
@@ -593,6 +614,11 @@
 		HID_USB_DEVICE(USB_VENDOR_ID_CANDO,
 			USB_DEVICE_ID_CANDO_MULTI_TOUCH_15_6) },
 
+	/* Chunghwa Telecom touch panels */
+	{  .driver_data = MT_CLS_DEFAULT,
+		HID_USB_DEVICE(USB_VENDOR_ID_CHUNGHWAT,
+			USB_DEVICE_ID_CHUNGHWAT_MULTITOUCH) },
+
 	/* CVTouch panels */
 	{ .driver_data = MT_CLS_DEFAULT,
 		HID_USB_DEVICE(USB_VENDOR_ID_CVTOUCH,
diff --git a/drivers/hid/usbhid/hid-quirks.c b/drivers/hid/usbhid/hid-quirks.c
index 0e30b14..621959d 100644
--- a/drivers/hid/usbhid/hid-quirks.c
+++ b/drivers/hid/usbhid/hid-quirks.c
@@ -74,6 +74,7 @@
 	{ USB_VENDOR_ID_UCLOGIC, USB_DEVICE_ID_UCLOGIC_TABLET_PF1209, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_UCLOGIC, USB_DEVICE_ID_UCLOGIC_TABLET_WP4030U, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_UCLOGIC, USB_DEVICE_ID_UCLOGIC_TABLET_KNA5, HID_QUIRK_MULTI_INPUT },
+	{ USB_VENDOR_ID_UCLOGIC, USB_DEVICE_ID_UCLOGIC_TABLET_TWA60, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_UCLOGIC, USB_DEVICE_ID_UCLOGIC_TABLET_WP5540U, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_UCLOGIC, USB_DEVICE_ID_UCLOGIC_TABLET_WP8060U, HID_QUIRK_MULTI_INPUT },
 	{ USB_VENDOR_ID_WALTOP, USB_DEVICE_ID_WALTOP_MEDIA_TABLET_10_6_INCH, HID_QUIRK_MULTI_INPUT },
diff --git a/drivers/hid/usbhid/hiddev.c b/drivers/hid/usbhid/hiddev.c
index ff3c644..7c1188b 100644
--- a/drivers/hid/usbhid/hiddev.c
+++ b/drivers/hid/usbhid/hiddev.c
@@ -248,12 +248,15 @@
 			usbhid_close(list->hiddev->hid);
 			usbhid_put_power(list->hiddev->hid);
 		} else {
+			mutex_unlock(&list->hiddev->existancelock);
 			kfree(list->hiddev);
+			kfree(list);
+			return 0;
 		}
 	}
 
-	kfree(list);
 	mutex_unlock(&list->hiddev->existancelock);
+	kfree(list);
 
 	return 0;
 }
@@ -923,10 +926,11 @@
 	usb_deregister_dev(usbhid->intf, &hiddev_class);
 
 	if (hiddev->open) {
+		mutex_unlock(&hiddev->existancelock);
 		usbhid_close(hiddev->hid);
 		wake_up_interruptible(&hiddev->wait);
 	} else {
+		mutex_unlock(&hiddev->existancelock);
 		kfree(hiddev);
 	}
-	mutex_unlock(&hiddev->existancelock);
 }
diff --git a/drivers/hwmon/asus_atk0110.c b/drivers/hwmon/asus_atk0110.c
index b5e8920..dcb78a7 100644
--- a/drivers/hwmon/asus_atk0110.c
+++ b/drivers/hwmon/asus_atk0110.c
@@ -268,6 +268,7 @@
 static void atk_init_attribute(struct device_attribute *attr, char *name,
 		sysfs_show_func show)
 {
+	sysfs_attr_init(&attr->attr);
 	attr->attr.name = name;
 	attr->attr.mode = 0444;
 	attr->show = show;
@@ -1188,19 +1189,15 @@
 	int err;
 
 	list_for_each_entry(s, &data->sensor_list, list) {
-		sysfs_attr_init(&s->input_attr.attr);
 		err = device_create_file(data->hwmon_dev, &s->input_attr);
 		if (err)
 			return err;
-		sysfs_attr_init(&s->label_attr.attr);
 		err = device_create_file(data->hwmon_dev, &s->label_attr);
 		if (err)
 			return err;
-		sysfs_attr_init(&s->limit1_attr.attr);
 		err = device_create_file(data->hwmon_dev, &s->limit1_attr);
 		if (err)
 			return err;
-		sysfs_attr_init(&s->limit2_attr.attr);
 		err = device_create_file(data->hwmon_dev, &s->limit2_attr);
 		if (err)
 			return err;
diff --git a/drivers/hwmon/coretemp.c b/drivers/hwmon/coretemp.c
index 85e9379..0070d54 100644
--- a/drivers/hwmon/coretemp.c
+++ b/drivers/hwmon/coretemp.c
@@ -97,9 +97,7 @@
 struct pdev_entry {
 	struct list_head list;
 	struct platform_device *pdev;
-	unsigned int cpu;
 	u16 phys_proc_id;
-	u16 cpu_core_id;
 };
 
 static LIST_HEAD(pdev_list);
@@ -653,9 +651,7 @@
 	}
 
 	pdev_entry->pdev = pdev;
-	pdev_entry->cpu = cpu;
 	pdev_entry->phys_proc_id = TO_PHYS_ID(cpu);
-	pdev_entry->cpu_core_id = TO_CORE_ID(cpu);
 
 	list_add_tail(&pdev_entry->list, &pdev_list);
 	mutex_unlock(&pdev_list_mutex);
diff --git a/drivers/hwmon/ibmaem.c b/drivers/hwmon/ibmaem.c
index 537409d..1a409c5 100644
--- a/drivers/hwmon/ibmaem.c
+++ b/drivers/hwmon/ibmaem.c
@@ -947,6 +947,7 @@
 
 	/* Set up read-only sensors */
 	while (ro->label) {
+		sysfs_attr_init(&sensors->dev_attr.attr);
 		sensors->dev_attr.attr.name = ro->label;
 		sensors->dev_attr.attr.mode = S_IRUGO;
 		sensors->dev_attr.show = ro->show;
@@ -963,6 +964,7 @@
 
 	/* Set up read-write sensors */
 	while (rw->label) {
+		sysfs_attr_init(&sensors->dev_attr.attr);
 		sensors->dev_attr.attr.name = rw->label;
 		sensors->dev_attr.attr.mode = S_IRUGO | S_IWUSR;
 		sensors->dev_attr.show = rw->show;
diff --git a/drivers/hwmon/ibmpex.c b/drivers/hwmon/ibmpex.c
index 06d4eaf..41dbf81 100644
--- a/drivers/hwmon/ibmpex.c
+++ b/drivers/hwmon/ibmpex.c
@@ -358,6 +358,7 @@
 	else if (type == POWER_SENSOR)
 		sprintf(n, power_sensor_name_templates[func], "power", counter);
 
+	sysfs_attr_init(&data->sensors[sensor].attr[func].dev_attr.attr);
 	data->sensors[sensor].attr[func].dev_attr.attr.name = n;
 	data->sensors[sensor].attr[func].dev_attr.attr.mode = S_IRUGO;
 	data->sensors[sensor].attr[func].dev_attr.show = ibmpex_show_sensor;
diff --git a/drivers/hwmon/s3c-hwmon.c b/drivers/hwmon/s3c-hwmon.c
index 92b42db..b39f52e 100644
--- a/drivers/hwmon/s3c-hwmon.c
+++ b/drivers/hwmon/s3c-hwmon.c
@@ -232,6 +232,7 @@
 
 	attr = &attrs->in;
 	attr->index = channel;
+	sysfs_attr_init(&attr->dev_attr.attr);
 	attr->dev_attr.attr.name  = attrs->in_name;
 	attr->dev_attr.attr.mode  = S_IRUGO;
 	attr->dev_attr.show = s3c_hwmon_ch_show;
@@ -249,6 +250,7 @@
 
 		attr = &attrs->label;
 		attr->index = channel;
+		sysfs_attr_init(&attr->dev_attr.attr);
 		attr->dev_attr.attr.name  = attrs->label_name;
 		attr->dev_attr.attr.mode  = S_IRUGO;
 		attr->dev_attr.show = s3c_hwmon_label_show;
diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index be0921e..4cf2534 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -111,7 +111,8 @@
 
 	rcu_read_unlock();
 
-	wake_up_interruptible(&evdev->wait);
+	if (type == EV_SYN && code == SYN_REPORT)
+		wake_up_interruptible(&evdev->wait);
 }
 
 static int evdev_fasync(int fd, struct file *file, int on)
diff --git a/drivers/input/input.c b/drivers/input/input.c
index 75e11c7..da38d97 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -1756,7 +1756,7 @@
 	} else if (test_bit(ABS_MT_TRACKING_ID, dev->absbit)) {
 		mt_slots = dev->absinfo[ABS_MT_TRACKING_ID].maximum -
 			   dev->absinfo[ABS_MT_TRACKING_ID].minimum + 1,
-		clamp(mt_slots, 2, 32);
+		mt_slots = clamp(mt_slots, 2, 32);
 	} else if (test_bit(ABS_MT_POSITION_X, dev->absbit)) {
 		mt_slots = 2;
 	} else {
diff --git a/drivers/input/keyboard/omap-keypad.c b/drivers/input/keyboard/omap-keypad.c
index f23a743..33d0bdc 100644
--- a/drivers/input/keyboard/omap-keypad.c
+++ b/drivers/input/keyboard/omap-keypad.c
@@ -209,6 +209,7 @@
 #endif
 		}
 	}
+	input_sync(omap_kp_data->input);
 	memcpy(keypad_state, new_state, sizeof(keypad_state));
 
 	if (key_down) {
diff --git a/drivers/input/keyboard/sh_keysc.c b/drivers/input/keyboard/sh_keysc.c
index 834cf98..6876700 100644
--- a/drivers/input/keyboard/sh_keysc.c
+++ b/drivers/input/keyboard/sh_keysc.c
@@ -32,7 +32,7 @@
 	[SH_KEYSC_MODE_3] = { 2, 4, 7 },
 	[SH_KEYSC_MODE_4] = { 3, 6, 6 },
 	[SH_KEYSC_MODE_5] = { 4, 6, 7 },
-	[SH_KEYSC_MODE_6] = { 5, 7, 7 },
+	[SH_KEYSC_MODE_6] = { 5, 8, 8 },
 };
 
 struct sh_keysc_priv {
diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c
index 257e033..0110b5a 100644
--- a/drivers/input/mousedev.c
+++ b/drivers/input/mousedev.c
@@ -187,7 +187,7 @@
 		if (size == 0)
 			size = xres ? : 1;
 
-		clamp(value, min, max);
+		value = clamp(value, min, max);
 
 		mousedev->packet.x = ((value - min) * xres) / size;
 		mousedev->packet.abs_event = 1;
@@ -201,7 +201,7 @@
 		if (size == 0)
 			size = yres ? : 1;
 
-		clamp(value, min, max);
+		value = clamp(value, min, max);
 
 		mousedev->packet.y = yres - ((value - min) * yres) / size;
 		mousedev->packet.abs_event = 1;
diff --git a/drivers/pcmcia/pxa2xx_vpac270.c b/drivers/pcmcia/pxa2xx_vpac270.c
index 435002d..712baab 100644
--- a/drivers/pcmcia/pxa2xx_vpac270.c
+++ b/drivers/pcmcia/pxa2xx_vpac270.c
@@ -11,6 +11,7 @@
  *
  */
 
+#include <linux/gpio.h>
 #include <linux/module.h>
 #include <linux/platform_device.h>
 
diff --git a/drivers/spi/spi_bfin5xx.c b/drivers/spi/spi_bfin5xx.c
index f706dba..cc880c9 100644
--- a/drivers/spi/spi_bfin5xx.c
+++ b/drivers/spi/spi_bfin5xx.c
@@ -681,13 +681,14 @@
 	drv_data->cs_change = transfer->cs_change;
 
 	/* Bits per word setup */
-	bits_per_word = transfer->bits_per_word ? : message->spi->bits_per_word;
-	if ((bits_per_word > 0) && (bits_per_word % 16 == 0)) {
+	bits_per_word = transfer->bits_per_word ? :
+		message->spi->bits_per_word ? : 8;
+	if (bits_per_word % 16 == 0) {
 		drv_data->n_bytes = bits_per_word/8;
 		drv_data->len = (transfer->len) >> 1;
 		cr_width = BIT_CTL_WORDSIZE;
 		drv_data->ops = &bfin_bfin_spi_transfer_ops_u16;
-	} else if ((bits_per_word > 0) && (bits_per_word % 8 == 0)) {
+	} else if (bits_per_word % 8 == 0) {
 		drv_data->n_bytes = bits_per_word/8;
 		drv_data->len = transfer->len;
 		cr_width = 0;
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 553da68..30df85d 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -395,9 +395,9 @@
 static void xen_irq_init(unsigned irq)
 {
 	struct irq_info *info;
+#ifdef CONFIG_SMP
 	struct irq_desc *desc = irq_to_desc(irq);
 
-#ifdef CONFIG_SMP
 	/* By default all event channels notify CPU#0. */
 	cpumask_copy(desc->irq_data.affinity, cpumask_of(0));
 #endif
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 378b5b4..3006287 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -967,6 +967,12 @@
 	struct srcu_struct subvol_srcu;
 
 	spinlock_t trans_lock;
+	/*
+	 * the reloc mutex goes with the trans lock, it is taken
+	 * during commit to protect us from the relocation code
+	 */
+	struct mutex reloc_mutex;
+
 	struct list_head trans_list;
 	struct list_head hashers;
 	struct list_head dead_roots;
@@ -1172,6 +1178,14 @@
 	u32 type;
 
 	u64 highest_objectid;
+
+	/* btrfs_record_root_in_trans is a multi-step process,
+	 * and it can race with the balancing code.   But the
+	 * race is very small, and only the first time the root
+	 * is added to each transaction.  So in_trans_setup
+	 * is used to tell us when more checks are required
+	 */
+	unsigned long in_trans_setup;
 	int ref_cows;
 	int track_dirty;
 	int in_radix;
@@ -1181,7 +1195,6 @@
 	struct btrfs_key defrag_max;
 	int defrag_running;
 	char *name;
-	int in_sysfs;
 
 	/* the dirty list is only used by non-reference counted roots */
 	struct list_head dirty_list;
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 6462c29..f1cbd02 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -297,7 +297,6 @@
 		item->data_len = data_len;
 		item->ins_or_del = 0;
 		item->bytes_reserved = 0;
-		item->block_rsv = NULL;
 		item->delayed_node = NULL;
 		atomic_set(&item->refs, 1);
 	}
@@ -593,10 +592,8 @@
 
 	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
 	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
-	if (!ret) {
+	if (!ret)
 		item->bytes_reserved = num_bytes;
-		item->block_rsv = dst_rsv;
-	}
 
 	return ret;
 }
@@ -604,10 +601,13 @@
 static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
 						struct btrfs_delayed_item *item)
 {
+	struct btrfs_block_rsv *rsv;
+
 	if (!item->bytes_reserved)
 		return;
 
-	btrfs_block_rsv_release(root, item->block_rsv,
+	rsv = &root->fs_info->global_block_rsv;
+	btrfs_block_rsv_release(root, rsv,
 				item->bytes_reserved);
 }
 
@@ -1014,6 +1014,7 @@
 	struct btrfs_delayed_root *delayed_root;
 	struct btrfs_delayed_node *curr_node, *prev_node;
 	struct btrfs_path *path;
+	struct btrfs_block_rsv *block_rsv;
 	int ret = 0;
 
 	path = btrfs_alloc_path();
@@ -1021,6 +1022,9 @@
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	block_rsv = trans->block_rsv;
+	trans->block_rsv = &root->fs_info->global_block_rsv;
+
 	delayed_root = btrfs_get_delayed_root(root);
 
 	curr_node = btrfs_first_delayed_node(delayed_root);
@@ -1045,6 +1049,7 @@
 	}
 
 	btrfs_free_path(path);
+	trans->block_rsv = block_rsv;
 	return ret;
 }
 
@@ -1052,6 +1057,7 @@
 					      struct btrfs_delayed_node *node)
 {
 	struct btrfs_path *path;
+	struct btrfs_block_rsv *block_rsv;
 	int ret;
 
 	path = btrfs_alloc_path();
@@ -1059,6 +1065,9 @@
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	block_rsv = trans->block_rsv;
+	trans->block_rsv = &node->root->fs_info->global_block_rsv;
+
 	ret = btrfs_insert_delayed_items(trans, path, node->root, node);
 	if (!ret)
 		ret = btrfs_delete_delayed_items(trans, path, node->root, node);
@@ -1066,6 +1075,7 @@
 		ret = btrfs_update_delayed_inode(trans, node->root, path, node);
 	btrfs_free_path(path);
 
+	trans->block_rsv = block_rsv;
 	return ret;
 }
 
@@ -1116,6 +1126,7 @@
 	struct btrfs_path *path;
 	struct btrfs_delayed_node *delayed_node = NULL;
 	struct btrfs_root *root;
+	struct btrfs_block_rsv *block_rsv;
 	unsigned long nr = 0;
 	int need_requeue = 0;
 	int ret;
@@ -1134,6 +1145,9 @@
 	if (IS_ERR(trans))
 		goto free_path;
 
+	block_rsv = trans->block_rsv;
+	trans->block_rsv = &root->fs_info->global_block_rsv;
+
 	ret = btrfs_insert_delayed_items(trans, path, root, delayed_node);
 	if (!ret)
 		ret = btrfs_delete_delayed_items(trans, path, root,
@@ -1176,6 +1190,7 @@
 
 	nr = trans->blocks_used;
 
+	trans->block_rsv = block_rsv;
 	btrfs_end_transaction_dmeta(trans, root);
 	__btrfs_btree_balance_dirty(root, nr);
 free_path:
@@ -1222,6 +1237,13 @@
 	return 0;
 }
 
+void btrfs_assert_delayed_root_empty(struct btrfs_root *root)
+{
+	struct btrfs_delayed_root *delayed_root;
+	delayed_root = btrfs_get_delayed_root(root);
+	WARN_ON(btrfs_first_delayed_node(delayed_root));
+}
+
 void btrfs_balance_delayed_items(struct btrfs_root *root)
 {
 	struct btrfs_delayed_root *delayed_root;
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index eb7d240..d1a6a29 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -75,7 +75,6 @@
 	struct list_head tree_list;	/* used for batch insert/delete items */
 	struct list_head readdir_list;	/* used for readdir items */
 	u64 bytes_reserved;
-	struct btrfs_block_rsv *block_rsv;
 	struct btrfs_delayed_node *delayed_node;
 	atomic_t refs;
 	int ins_or_del;
@@ -138,4 +137,8 @@
 /* for init */
 int __init btrfs_delayed_inode_init(void);
 void btrfs_delayed_inode_exit(void);
+
+/* for debugging */
+void btrfs_assert_delayed_root_empty(struct btrfs_root *root);
+
 #endif
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 9f68c68..1ac8db5d 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1044,7 +1044,6 @@
 	root->last_trans = 0;
 	root->highest_objectid = 0;
 	root->name = NULL;
-	root->in_sysfs = 0;
 	root->inode_tree = RB_ROOT;
 	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
 	root->block_rsv = NULL;
@@ -1300,19 +1299,21 @@
 		return root;
 
 	root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS);
-	if (!root->free_ino_ctl)
-		goto fail;
 	root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned),
 					GFP_NOFS);
-	if (!root->free_ino_pinned)
+	if (!root->free_ino_pinned || !root->free_ino_ctl) {
+		ret = -ENOMEM;
 		goto fail;
+	}
 
 	btrfs_init_free_ino_ctl(root);
 	mutex_init(&root->fs_commit_mutex);
 	spin_lock_init(&root->cache_lock);
 	init_waitqueue_head(&root->cache_wait);
 
-	set_anon_super(&root->anon_super, NULL);
+	ret = set_anon_super(&root->anon_super, NULL);
+	if (ret)
+		goto fail;
 
 	if (btrfs_root_refs(&root->root_item) == 0) {
 		ret = -ENOENT;
@@ -1618,6 +1619,7 @@
 	spin_lock_init(&fs_info->fs_roots_radix_lock);
 	spin_lock_init(&fs_info->delayed_iput_lock);
 	spin_lock_init(&fs_info->defrag_inodes_lock);
+	mutex_init(&fs_info->reloc_mutex);
 
 	init_completion(&fs_info->kobj_unregister);
 	fs_info->tree_root = tree_root;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index b42efc2..1f61bf5 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3314,10 +3314,6 @@
 	if (reserved == 0)
 		return 0;
 
-	/* nothing to shrink - nothing to reclaim */
-	if (root->fs_info->delalloc_bytes == 0)
-		return 0;
-
 	max_reclaim = min(reserved, to_reclaim);
 
 	while (loops < 1024) {
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 751ddf8..0a9b10c 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3076,6 +3076,7 @@
 	ret = btrfs_update_inode(trans, root, dir);
 	BUG_ON(ret);
 
+	btrfs_free_path(path);
 	return 0;
 }
 
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index b793d11..a3c4751 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -482,8 +482,10 @@
 	ret = btrfs_snap_reserve_metadata(trans, pending_snapshot);
 	BUG_ON(ret);
 
+	spin_lock(&root->fs_info->trans_lock);
 	list_add(&pending_snapshot->list,
 		 &trans->transaction->pending_snapshots);
+	spin_unlock(&root->fs_info->trans_lock);
 	if (async_transid) {
 		*async_transid = trans->transid;
 		ret = btrfs_commit_transaction_async(trans,
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index b1ef27c..5e0a3dc 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1368,7 +1368,7 @@
 	int ret;
 
 	if (!root->reloc_root)
-		return 0;
+		goto out;
 
 	reloc_root = root->reloc_root;
 	root_item = &reloc_root->root_item;
@@ -1390,6 +1390,8 @@
 	ret = btrfs_update_root(trans, root->fs_info->tree_root,
 				&reloc_root->root_key, root_item);
 	BUG_ON(ret);
+
+out:
 	return 0;
 }
 
@@ -2142,10 +2144,11 @@
 	u64 num_bytes = 0;
 	int ret;
 
-	spin_lock(&root->fs_info->trans_lock);
+	mutex_lock(&root->fs_info->reloc_mutex);
 	rc->merging_rsv_size += root->nodesize * (BTRFS_MAX_LEVEL - 1) * 2;
 	rc->merging_rsv_size += rc->nodes_relocated * 2;
-	spin_unlock(&root->fs_info->trans_lock);
+	mutex_unlock(&root->fs_info->reloc_mutex);
+
 again:
 	if (!err) {
 		num_bytes = rc->merging_rsv_size;
@@ -2214,9 +2217,16 @@
 	int ret;
 again:
 	root = rc->extent_root;
-	spin_lock(&root->fs_info->trans_lock);
+
+	/*
+	 * this serializes us with btrfs_record_root_in_transaction,
+	 * we have to make sure nobody is in the middle of
+	 * adding their roots to the list while we are
+	 * doing this splice
+	 */
+	mutex_lock(&root->fs_info->reloc_mutex);
 	list_splice_init(&rc->reloc_roots, &reloc_roots);
-	spin_unlock(&root->fs_info->trans_lock);
+	mutex_unlock(&root->fs_info->reloc_mutex);
 
 	while (!list_empty(&reloc_roots)) {
 		found = 1;
@@ -3590,17 +3600,19 @@
 static void set_reloc_control(struct reloc_control *rc)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	spin_lock(&fs_info->trans_lock);
+
+	mutex_lock(&fs_info->reloc_mutex);
 	fs_info->reloc_ctl = rc;
-	spin_unlock(&fs_info->trans_lock);
+	mutex_unlock(&fs_info->reloc_mutex);
 }
 
 static void unset_reloc_control(struct reloc_control *rc)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	spin_lock(&fs_info->trans_lock);
+
+	mutex_lock(&fs_info->reloc_mutex);
 	fs_info->reloc_ctl = NULL;
-	spin_unlock(&fs_info->trans_lock);
+	mutex_unlock(&fs_info->reloc_mutex);
 }
 
 static int check_extent_flags(u64 flags)
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index c3c223a..daac9ae 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -28,152 +28,6 @@
 #include "disk-io.h"
 #include "transaction.h"
 
-static ssize_t root_blocks_used_show(struct btrfs_root *root, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)btrfs_root_used(&root->root_item));
-}
-
-static ssize_t root_block_limit_show(struct btrfs_root *root, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)btrfs_root_limit(&root->root_item));
-}
-
-static ssize_t super_blocks_used_show(struct btrfs_fs_info *fs, char *buf)
-{
-
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)btrfs_super_bytes_used(&fs->super_copy));
-}
-
-static ssize_t super_total_blocks_show(struct btrfs_fs_info *fs, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)btrfs_super_total_bytes(&fs->super_copy));
-}
-
-static ssize_t super_blocksize_show(struct btrfs_fs_info *fs, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%llu\n",
-		(unsigned long long)btrfs_super_sectorsize(&fs->super_copy));
-}
-
-/* this is for root attrs (subvols/snapshots) */
-struct btrfs_root_attr {
-	struct attribute attr;
-	ssize_t (*show)(struct btrfs_root *, char *);
-	ssize_t (*store)(struct btrfs_root *, const char *, size_t);
-};
-
-#define ROOT_ATTR(name, mode, show, store) \
-static struct btrfs_root_attr btrfs_root_attr_##name = __ATTR(name, mode, \
-							      show, store)
-
-ROOT_ATTR(blocks_used,	0444,	root_blocks_used_show,	NULL);
-ROOT_ATTR(block_limit,	0644,	root_block_limit_show,	NULL);
-
-static struct attribute *btrfs_root_attrs[] = {
-	&btrfs_root_attr_blocks_used.attr,
-	&btrfs_root_attr_block_limit.attr,
-	NULL,
-};
-
-/* this is for super attrs (actual full fs) */
-struct btrfs_super_attr {
-	struct attribute attr;
-	ssize_t (*show)(struct btrfs_fs_info *, char *);
-	ssize_t (*store)(struct btrfs_fs_info *, const char *, size_t);
-};
-
-#define SUPER_ATTR(name, mode, show, store) \
-static struct btrfs_super_attr btrfs_super_attr_##name = __ATTR(name, mode, \
-								show, store)
-
-SUPER_ATTR(blocks_used,		0444,	super_blocks_used_show,		NULL);
-SUPER_ATTR(total_blocks,	0444,	super_total_blocks_show,	NULL);
-SUPER_ATTR(blocksize,		0444,	super_blocksize_show,		NULL);
-
-static struct attribute *btrfs_super_attrs[] = {
-	&btrfs_super_attr_blocks_used.attr,
-	&btrfs_super_attr_total_blocks.attr,
-	&btrfs_super_attr_blocksize.attr,
-	NULL,
-};
-
-static ssize_t btrfs_super_attr_show(struct kobject *kobj,
-				    struct attribute *attr, char *buf)
-{
-	struct btrfs_fs_info *fs = container_of(kobj, struct btrfs_fs_info,
-						super_kobj);
-	struct btrfs_super_attr *a = container_of(attr,
-						  struct btrfs_super_attr,
-						  attr);
-
-	return a->show ? a->show(fs, buf) : 0;
-}
-
-static ssize_t btrfs_super_attr_store(struct kobject *kobj,
-				     struct attribute *attr,
-				     const char *buf, size_t len)
-{
-	struct btrfs_fs_info *fs = container_of(kobj, struct btrfs_fs_info,
-						super_kobj);
-	struct btrfs_super_attr *a = container_of(attr,
-						  struct btrfs_super_attr,
-						  attr);
-
-	return a->store ? a->store(fs, buf, len) : 0;
-}
-
-static ssize_t btrfs_root_attr_show(struct kobject *kobj,
-				    struct attribute *attr, char *buf)
-{
-	struct btrfs_root *root = container_of(kobj, struct btrfs_root,
-						root_kobj);
-	struct btrfs_root_attr *a = container_of(attr,
-						 struct btrfs_root_attr,
-						 attr);
-
-	return a->show ? a->show(root, buf) : 0;
-}
-
-static ssize_t btrfs_root_attr_store(struct kobject *kobj,
-				     struct attribute *attr,
-				     const char *buf, size_t len)
-{
-	struct btrfs_root *root = container_of(kobj, struct btrfs_root,
-						root_kobj);
-	struct btrfs_root_attr *a = container_of(attr,
-						 struct btrfs_root_attr,
-						 attr);
-	return a->store ? a->store(root, buf, len) : 0;
-}
-
-static void btrfs_super_release(struct kobject *kobj)
-{
-	struct btrfs_fs_info *fs = container_of(kobj, struct btrfs_fs_info,
-						super_kobj);
-	complete(&fs->kobj_unregister);
-}
-
-static void btrfs_root_release(struct kobject *kobj)
-{
-	struct btrfs_root *root = container_of(kobj, struct btrfs_root,
-						root_kobj);
-	complete(&root->kobj_unregister);
-}
-
-static const struct sysfs_ops btrfs_super_attr_ops = {
-	.show	= btrfs_super_attr_show,
-	.store	= btrfs_super_attr_store,
-};
-
-static const struct sysfs_ops btrfs_root_attr_ops = {
-	.show	= btrfs_root_attr_show,
-	.store	= btrfs_root_attr_store,
-};
-
 /* /sys/fs/btrfs/ entry */
 static struct kset *btrfs_kset;
 
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 2b3590b..51dcec8 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -126,28 +126,85 @@
  * to make sure the old root from before we joined the transaction is deleted
  * when the transaction commits
  */
-int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
+static int record_root_in_trans(struct btrfs_trans_handle *trans,
 			       struct btrfs_root *root)
 {
 	if (root->ref_cows && root->last_trans < trans->transid) {
 		WARN_ON(root == root->fs_info->extent_root);
 		WARN_ON(root->commit_root != root->node);
 
+		/*
+		 * see below for in_trans_setup usage rules
+		 * we have the reloc mutex held now, so there
+		 * is only one writer in this function
+		 */
+		root->in_trans_setup = 1;
+
+		/* make sure readers find in_trans_setup before
+		 * they find our root->last_trans update
+		 */
+		smp_wmb();
+
 		spin_lock(&root->fs_info->fs_roots_radix_lock);
 		if (root->last_trans == trans->transid) {
 			spin_unlock(&root->fs_info->fs_roots_radix_lock);
 			return 0;
 		}
-		root->last_trans = trans->transid;
 		radix_tree_tag_set(&root->fs_info->fs_roots_radix,
 			   (unsigned long)root->root_key.objectid,
 			   BTRFS_ROOT_TRANS_TAG);
 		spin_unlock(&root->fs_info->fs_roots_radix_lock);
+		root->last_trans = trans->transid;
+
+		/* this is pretty tricky.  We don't want to
+		 * take the relocation lock in btrfs_record_root_in_trans
+		 * unless we're really doing the first setup for this root in
+		 * this transaction.
+		 *
+		 * Normally we'd use root->last_trans as a flag to decide
+		 * if we want to take the expensive mutex.
+		 *
+		 * But, we have to set root->last_trans before we
+		 * init the relocation root, otherwise, we trip over warnings
+		 * in ctree.c.  The solution used here is to flag ourselves
+		 * with root->in_trans_setup.  When this is 1, we're still
+		 * fixing up the reloc trees and everyone must wait.
+		 *
+		 * When this is zero, they can trust root->last_trans and fly
+		 * through btrfs_record_root_in_trans without having to take the
+		 * lock.  smp_wmb() makes sure that all the writes above are
+		 * done before we pop in the zero below
+		 */
 		btrfs_init_reloc_root(trans, root);
+		smp_wmb();
+		root->in_trans_setup = 0;
 	}
 	return 0;
 }
 
+
+int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root)
+{
+	if (!root->ref_cows)
+		return 0;
+
+	/*
+	 * see record_root_in_trans for comments about in_trans_setup usage
+	 * and barriers
+	 */
+	smp_rmb();
+	if (root->last_trans == trans->transid &&
+	    !root->in_trans_setup)
+		return 0;
+
+	mutex_lock(&root->fs_info->reloc_mutex);
+	record_root_in_trans(trans, root);
+	mutex_unlock(&root->fs_info->reloc_mutex);
+
+	return 0;
+}
+
 /* wait for commit against the current transaction to become unblocked
  * when this is done, it is safe to start a new transaction, but the current
  * transaction might not be fully on disk.
@@ -882,7 +939,7 @@
 	parent = dget_parent(dentry);
 	parent_inode = parent->d_inode;
 	parent_root = BTRFS_I(parent_inode)->root;
-	btrfs_record_root_in_trans(trans, parent_root);
+	record_root_in_trans(trans, parent_root);
 
 	/*
 	 * insert the directory item
@@ -900,7 +957,16 @@
 	ret = btrfs_update_inode(trans, parent_root, parent_inode);
 	BUG_ON(ret);
 
-	btrfs_record_root_in_trans(trans, root);
+	/*
+	 * pull in the delayed directory update
+	 * and the delayed inode item
+	 * otherwise we corrupt the FS during
+	 * snapshot
+	 */
+	ret = btrfs_run_delayed_items(trans, root);
+	BUG_ON(ret);
+
+	record_root_in_trans(trans, root);
 	btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
 	memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));
 	btrfs_check_and_init_root_item(new_root_item);
@@ -961,14 +1027,6 @@
 	int ret;
 
 	list_for_each_entry(pending, head, list) {
-		/*
-		 * We must deal with the delayed items before creating
-		 * snapshots, or we will create a snapthot with inconsistent
-		 * information.
-		*/
-		ret = btrfs_run_delayed_items(trans, fs_info->fs_root);
-		BUG_ON(ret);
-
 		ret = create_pending_snapshot(trans, fs_info, pending);
 		BUG_ON(ret);
 	}
@@ -1241,21 +1299,42 @@
 			schedule_timeout(1);
 
 		finish_wait(&cur_trans->writer_wait, &wait);
-		spin_lock(&root->fs_info->trans_lock);
-		root->fs_info->trans_no_join = 1;
-		spin_unlock(&root->fs_info->trans_lock);
 	} while (atomic_read(&cur_trans->num_writers) > 1 ||
 		 (should_grow && cur_trans->num_joined != joined));
 
-	ret = create_pending_snapshots(trans, root->fs_info);
-	BUG_ON(ret);
+	/*
+	 * Ok now we need to make sure to block out any other joins while we
+	 * commit the transaction.  We could have started a join before setting
+	 * no_join so make sure to wait for num_writers to == 1 again.
+	 */
+	spin_lock(&root->fs_info->trans_lock);
+	root->fs_info->trans_no_join = 1;
+	spin_unlock(&root->fs_info->trans_lock);
+	wait_event(cur_trans->writer_wait,
+		   atomic_read(&cur_trans->num_writers) == 1);
+
+	/*
+	 * the reloc mutex makes sure that we stop
+	 * the balancing code from coming in and moving
+	 * extents around in the middle of the commit
+	 */
+	mutex_lock(&root->fs_info->reloc_mutex);
 
 	ret = btrfs_run_delayed_items(trans, root);
 	BUG_ON(ret);
 
+	ret = create_pending_snapshots(trans, root->fs_info);
+	BUG_ON(ret);
+
 	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
 	BUG_ON(ret);
 
+	/*
+	 * make sure none of the code above managed to slip in a
+	 * delayed item
+	 */
+	btrfs_assert_delayed_root_empty(root);
+
 	WARN_ON(cur_trans != trans->transaction);
 
 	btrfs_scrub_pause(root);
@@ -1312,6 +1391,7 @@
 	root->fs_info->running_transaction = NULL;
 	root->fs_info->trans_no_join = 0;
 	spin_unlock(&root->fs_info->trans_lock);
+	mutex_unlock(&root->fs_info->reloc_mutex);
 
 	wake_up(&root->fs_info->transaction_wait);
 
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 592396c..4ce8a9f 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -3177,7 +3177,7 @@
 		tmp_key.offset = (u64)-1;
 
 		wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key);
-		BUG_ON(!wc.replay_dest);
+		BUG_ON(IS_ERR_OR_NULL(wc.replay_dest));
 
 		wc.replay_dest->log_root = log;
 		btrfs_record_root_in_trans(trans, wc.replay_dest);
diff --git a/fs/exec.c b/fs/exec.c
index 97e0d52..6075a1e 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1996,7 +1996,7 @@
  * is a special value that we use to trap recursive
  * core dumps
  */
-static int umh_pipe_setup(struct subprocess_info *info)
+static int umh_pipe_setup(struct subprocess_info *info, struct cred *new)
 {
 	struct file *rp, *wp;
 	struct fdtable *fdt;
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 3db5ba4..b3cc858 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -974,7 +974,7 @@
 out_no_read:
 	printk(KERN_WARNING "%s: bread failed, dev=%s, iso_blknum=%d, block=%d\n",
 		__func__, s->s_id, iso_blknum, block);
-	goto out_freesbi;
+	goto out_freebh;
 out_bad_zone_size:
 	printk(KERN_WARNING "ISOFS: Bad logical zone size %ld\n",
 		sbi->s_log_zone_size);
@@ -989,6 +989,7 @@
 
 out_freebh:
 	brelse(bh);
+	brelse(pri_bh);
 out_freesbi:
 	kfree(opt.iocharset);
 	kfree(sbi);
diff --git a/fs/timerfd.c b/fs/timerfd.c
index f67acbd..dffeb37 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -61,7 +61,9 @@
 
 /*
  * Called when the clock was set to cancel the timers in the cancel
- * list.
+ * list. This will wake up processes waiting on these timers. The
+ * wake-up requires ctx->ticks to be non zero, therefore we increment
+ * it before calling wake_up_locked().
  */
 void timerfd_clock_was_set(void)
 {
@@ -76,6 +78,7 @@
 		spin_lock_irqsave(&ctx->wqh.lock, flags);
 		if (ctx->moffs.tv64 != moffs.tv64) {
 			ctx->moffs.tv64 = KTIME_MAX;
+			ctx->ticks++;
 			wake_up_locked(&ctx->wqh);
 		}
 		spin_unlock_irqrestore(&ctx->wqh.lock, flags);
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index f4213ba..7f782af 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -131,19 +131,34 @@
 {
 	struct inode		*inode = file->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_trans	*tp;
 	int			error = 0;
 	int			log_flushed = 0;
 
 	trace_xfs_file_fsync(ip);
 
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+	if (XFS_FORCED_SHUTDOWN(mp))
 		return -XFS_ERROR(EIO);
 
 	xfs_iflags_clear(ip, XFS_ITRUNCATED);
 
 	xfs_ioend_wait(ip);
 
+	if (mp->m_flags & XFS_MOUNT_BARRIER) {
+		/*
+		 * If we have an RT and/or log subvolume we need to make sure
+		 * to flush the write cache the device used for file data
+		 * first.  This is to ensure newly written file data make
+		 * it to disk before logging the new inode size in case of
+		 * an extending write.
+		 */
+		if (XFS_IS_REALTIME_INODE(ip))
+			xfs_blkdev_issue_flush(mp->m_rtdev_targp);
+		else if (mp->m_logdev_targp != mp->m_ddev_targp)
+			xfs_blkdev_issue_flush(mp->m_ddev_targp);
+	}
+
 	/*
 	 * We always need to make sure that the required inode state is safe on
 	 * disk.  The inode might be clean but we still might need to force the
@@ -175,9 +190,9 @@
 		 * updates.  The sync transaction will also force the log.
 		 */
 		xfs_iunlock(ip, XFS_ILOCK_SHARED);
-		tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
+		tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
 		error = xfs_trans_reserve(tp, 0,
-				XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
+				XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
 		if (error) {
 			xfs_trans_cancel(tp, 0);
 			return -error;
@@ -209,28 +224,25 @@
 		 * force the log.
 		 */
 		if (xfs_ipincount(ip)) {
-			error = _xfs_log_force_lsn(ip->i_mount,
+			error = _xfs_log_force_lsn(mp,
 					ip->i_itemp->ili_last_lsn,
 					XFS_LOG_SYNC, &log_flushed);
 		}
 		xfs_iunlock(ip, XFS_ILOCK_SHARED);
 	}
 
-	if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) {
-		/*
-		 * If the log write didn't issue an ordered tag we need
-		 * to flush the disk cache for the data device now.
-		 */
-		if (!log_flushed)
-			xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
-
-		/*
-		 * If this inode is on the RT dev we need to flush that
-		 * cache as well.
-		 */
-		if (XFS_IS_REALTIME_INODE(ip))
-			xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
-	}
+	/*
+	 * If we only have a single device, and the log force about was
+	 * a no-op we might have to flush the data device cache here.
+	 * This can only happen for fdatasync/O_DSYNC if we were overwriting
+	 * an already allocated file and thus do not have any metadata to
+	 * commit.
+	 */
+	if ((mp->m_flags & XFS_MOUNT_BARRIER) &&
+	    mp->m_logdev_targp == mp->m_ddev_targp &&
+	    !XFS_IS_REALTIME_INODE(ip) &&
+	    !log_flushed)
+		xfs_blkdev_issue_flush(mp->m_ddev_targp);
 
 	return -error;
 }
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index dd21784..d44d92c 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -182,7 +182,7 @@
 	if (IS_POSIXACL(dir)) {
 		default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT);
 		if (IS_ERR(default_acl))
-			return -PTR_ERR(default_acl);
+			return PTR_ERR(default_acl);
 
 		if (!default_acl)
 			mode &= ~current_umask();
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 1e3a7ce..a1a881e 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -627,68 +627,6 @@
 		blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
 }
 
-/*
- * Try to write out the superblock using barriers.
- */
-STATIC int
-xfs_barrier_test(
-	xfs_mount_t	*mp)
-{
-	xfs_buf_t	*sbp = xfs_getsb(mp, 0);
-	int		error;
-
-	XFS_BUF_UNDONE(sbp);
-	XFS_BUF_UNREAD(sbp);
-	XFS_BUF_UNDELAYWRITE(sbp);
-	XFS_BUF_WRITE(sbp);
-	XFS_BUF_UNASYNC(sbp);
-	XFS_BUF_ORDERED(sbp);
-
-	xfsbdstrat(mp, sbp);
-	error = xfs_buf_iowait(sbp);
-
-	/*
-	 * Clear all the flags we set and possible error state in the
-	 * buffer.  We only did the write to try out whether barriers
-	 * worked and shouldn't leave any traces in the superblock
-	 * buffer.
-	 */
-	XFS_BUF_DONE(sbp);
-	XFS_BUF_ERROR(sbp, 0);
-	XFS_BUF_UNORDERED(sbp);
-
-	xfs_buf_relse(sbp);
-	return error;
-}
-
-STATIC void
-xfs_mountfs_check_barriers(xfs_mount_t *mp)
-{
-	int error;
-
-	if (mp->m_logdev_targp != mp->m_ddev_targp) {
-		xfs_notice(mp,
-		  "Disabling barriers, not supported with external log device");
-		mp->m_flags &= ~XFS_MOUNT_BARRIER;
-		return;
-	}
-
-	if (xfs_readonly_buftarg(mp->m_ddev_targp)) {
-		xfs_notice(mp,
-			"Disabling barriers, underlying device is readonly");
-		mp->m_flags &= ~XFS_MOUNT_BARRIER;
-		return;
-	}
-
-	error = xfs_barrier_test(mp);
-	if (error) {
-		xfs_notice(mp,
-			"Disabling barriers, trial barrier write failed");
-		mp->m_flags &= ~XFS_MOUNT_BARRIER;
-		return;
-	}
-}
-
 void
 xfs_blkdev_issue_flush(
 	xfs_buftarg_t		*buftarg)
@@ -1240,14 +1178,6 @@
 		switch (token) {
 		case Opt_barrier:
 			mp->m_flags |= XFS_MOUNT_BARRIER;
-
-			/*
-			 * Test if barriers are actually working if we can,
-			 * else delay this check until the filesystem is
-			 * marked writeable.
-			 */
-			if (!(mp->m_flags & XFS_MOUNT_RDONLY))
-				xfs_mountfs_check_barriers(mp);
 			break;
 		case Opt_nobarrier:
 			mp->m_flags &= ~XFS_MOUNT_BARRIER;
@@ -1282,8 +1212,6 @@
 	/* ro -> rw */
 	if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) {
 		mp->m_flags &= ~XFS_MOUNT_RDONLY;
-		if (mp->m_flags & XFS_MOUNT_BARRIER)
-			xfs_mountfs_check_barriers(mp);
 
 		/*
 		 * If this is the first remount to writeable state we
@@ -1465,9 +1393,6 @@
 	if (error)
 		goto out_free_sb;
 
-	if (mp->m_flags & XFS_MOUNT_BARRIER)
-		xfs_mountfs_check_barriers(mp);
-
 	error = xfs_filestream_mount(mp);
 	if (error)
 		goto out_free_sb;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 2119302..41d5b8f 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -1372,8 +1372,17 @@
 	XFS_BUF_ASYNC(bp);
 	bp->b_flags |= XBF_LOG_BUFFER;
 
-	if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
+	if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) {
+		/*
+		 * If we have an external log device, flush the data device
+		 * before flushing the log to make sure all meta data
+		 * written back from the AIL actually made it to disk
+		 * before writing out the new log tail LSN in the log buffer.
+		 */
+		if (log->l_mp->m_logdev_targp != log->l_mp->m_ddev_targp)
+			xfs_blkdev_issue_flush(log->l_mp->m_ddev_targp);
 		XFS_BUF_ORDERED(bp);
+	}
 
 	ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
 	ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize);
diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h
index fcdcb5d..d494001 100644
--- a/include/asm-generic/gpio.h
+++ b/include/asm-generic/gpio.h
@@ -170,16 +170,6 @@
 
 extern int __gpio_to_irq(unsigned gpio);
 
-#define GPIOF_DIR_OUT	(0 << 0)
-#define GPIOF_DIR_IN	(1 << 0)
-
-#define GPIOF_INIT_LOW	(0 << 1)
-#define GPIOF_INIT_HIGH	(1 << 1)
-
-#define GPIOF_IN		(GPIOF_DIR_IN)
-#define GPIOF_OUT_INIT_LOW	(GPIOF_DIR_OUT | GPIOF_INIT_LOW)
-#define GPIOF_OUT_INIT_HIGH	(GPIOF_DIR_OUT | GPIOF_INIT_HIGH)
-
 /**
  * struct gpio - a structure describing a GPIO with configuration
  * @gpio:	the GPIO number
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 9573e0c..33d12f8 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -520,6 +520,8 @@
 	uint32_t encoder_ids[DRM_CONNECTOR_MAX_ENCODER];
 	uint32_t force_encoder_id;
 	struct drm_encoder *encoder; /* currently active encoder */
+
+	int null_edid_counter; /* needed to workaround some HW bugs where we get all 0s */
 };
 
 /**
diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index d4646b4..18a1baf 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -188,6 +188,7 @@
 #ifdef CONFIG_CLOCKSOURCE_WATCHDOG
 	/* Watchdog related data, used by the framework */
 	struct list_head wd_list;
+	cycle_t cs_last;
 	cycle_t wd_last;
 #endif
 } ____cacheline_aligned;
diff --git a/include/linux/gpio.h b/include/linux/gpio.h
index 32d47e7..17b5a0d 100644
--- a/include/linux/gpio.h
+++ b/include/linux/gpio.h
@@ -3,6 +3,17 @@
 
 /* see Documentation/gpio.txt */
 
+/* make these flag values available regardless of GPIO kconfig options */
+#define GPIOF_DIR_OUT	(0 << 0)
+#define GPIOF_DIR_IN	(1 << 0)
+
+#define GPIOF_INIT_LOW	(0 << 1)
+#define GPIOF_INIT_HIGH	(1 << 1)
+
+#define GPIOF_IN		(GPIOF_DIR_IN)
+#define GPIOF_OUT_INIT_LOW	(GPIOF_DIR_OUT | GPIOF_INIT_LOW)
+#define GPIOF_OUT_INIT_HIGH	(GPIOF_DIR_OUT | GPIOF_INIT_HIGH)
+
 #ifdef CONFIG_GENERIC_GPIO
 #include <asm/gpio.h>
 
diff --git a/include/linux/input/sh_keysc.h b/include/linux/input/sh_keysc.h
index 649dc7f..5d253cd 100644
--- a/include/linux/input/sh_keysc.h
+++ b/include/linux/input/sh_keysc.h
@@ -1,7 +1,7 @@
 #ifndef __SH_KEYSC_H__
 #define __SH_KEYSC_H__
 
-#define SH_KEYSC_MAXKEYS 49
+#define SH_KEYSC_MAXKEYS 64
 
 struct sh_keysc_info {
 	enum { SH_KEYSC_MODE_1, SH_KEYSC_MODE_2, SH_KEYSC_MODE_3,
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 6c12989..f6efed0 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -414,6 +414,7 @@
 	TASKLET_SOFTIRQ,
 	SCHED_SOFTIRQ,
 	HRTIMER_SOFTIRQ,
+	RCU_SOFTIRQ,    /* Preferable RCU should always be the last softirq */
 
 	NR_SOFTIRQS
 };
diff --git a/include/linux/kmod.h b/include/linux/kmod.h
index d4a5c84..0da38cf 100644
--- a/include/linux/kmod.h
+++ b/include/linux/kmod.h
@@ -45,7 +45,7 @@
 #endif
 
 
-struct key;
+struct cred;
 struct file;
 
 enum umh_wait {
@@ -62,7 +62,7 @@
 	char **envp;
 	enum umh_wait wait;
 	int retval;
-	int (*init)(struct subprocess_info *info);
+	int (*init)(struct subprocess_info *info, struct cred *new);
 	void (*cleanup)(struct subprocess_info *info);
 	void *data;
 };
@@ -73,7 +73,7 @@
 
 /* Set various pieces of state into the subprocess_info structure */
 void call_usermodehelper_setfns(struct subprocess_info *info,
-		    int (*init)(struct subprocess_info *info),
+		    int (*init)(struct subprocess_info *info, struct cred *new),
 		    void (*cleanup)(struct subprocess_info *info),
 		    void *data);
 
@@ -87,7 +87,7 @@
 static inline int
 call_usermodehelper_fns(char *path, char **argv, char **envp,
 			enum umh_wait wait,
-			int (*init)(struct subprocess_info *info),
+			int (*init)(struct subprocess_info *info, struct cred *new),
 			void (*cleanup)(struct subprocess_info *), void *data)
 {
 	struct subprocess_info *info;
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 7ad824d..8cc38d3 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -85,12 +85,15 @@
  * Generic and arch helpers
  */
 #ifdef CONFIG_USE_GENERIC_SMP_HELPERS
+void __init call_function_init(void);
 void generic_smp_call_function_single_interrupt(void);
 void generic_smp_call_function_interrupt(void);
 void ipi_call_lock(void);
 void ipi_call_unlock(void);
 void ipi_call_lock_irq(void);
 void ipi_call_unlock_irq(void);
+#else
+static inline void call_function_init(void) { }
 #endif
 
 /*
@@ -134,7 +137,7 @@
 #define smp_prepare_boot_cpu()			do {} while (0)
 #define smp_call_function_many(mask, func, info, wait) \
 			(up_smp_call_function(func, info))
-static inline void init_call_single_data(void) { }
+static inline void call_function_init(void) { }
 
 static inline int
 smp_call_function_any(const struct cpumask *mask, smp_call_func_t func,
diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h
index ae045ca..1c09820 100644
--- a/include/trace/events/irq.h
+++ b/include/trace/events/irq.h
@@ -20,7 +20,8 @@
 			 softirq_name(BLOCK_IOPOLL),	\
 			 softirq_name(TASKLET),		\
 			 softirq_name(SCHED),		\
-			 softirq_name(HRTIMER))
+			 softirq_name(HRTIMER),		\
+			 softirq_name(RCU))
 
 /**
  * irq_handler_entry - called immediately before the irq action handler
diff --git a/init/main.c b/init/main.c
index cafba67..d7211fa 100644
--- a/init/main.c
+++ b/init/main.c
@@ -542,6 +542,7 @@
 	timekeeping_init();
 	time_init();
 	profile_init();
+	call_function_init();
 	if (!irqs_disabled())
 		printk(KERN_CRIT "start_kernel(): bug: interrupts were "
 				 "enabled early\n");
diff --git a/kernel/kmod.c b/kernel/kmod.c
index ad6a81c..47613df 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -156,12 +156,6 @@
 	 */
 	set_user_nice(current, 0);
 
-	if (sub_info->init) {
-		retval = sub_info->init(sub_info);
-		if (retval)
-			goto fail;
-	}
-
 	retval = -ENOMEM;
 	new = prepare_kernel_cred(current);
 	if (!new)
@@ -173,6 +167,14 @@
 					     new->cap_inheritable);
 	spin_unlock(&umh_sysctl_lock);
 
+	if (sub_info->init) {
+		retval = sub_info->init(sub_info, new);
+		if (retval) {
+			abort_creds(new);
+			goto fail;
+		}
+	}
+
 	commit_creds(new);
 
 	retval = kernel_execve(sub_info->path,
@@ -388,7 +390,7 @@
  * context in which call_usermodehelper_exec is called.
  */
 void call_usermodehelper_setfns(struct subprocess_info *info,
-		    int (*init)(struct subprocess_info *info),
+		    int (*init)(struct subprocess_info *info, struct cred *new),
 		    void (*cleanup)(struct subprocess_info *info),
 		    void *data)
 {
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 89419ff..7e59ffb 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -87,6 +87,8 @@
 int rcu_scheduler_active __read_mostly;
 EXPORT_SYMBOL_GPL(rcu_scheduler_active);
 
+#ifdef CONFIG_RCU_BOOST
+
 /*
  * Control variables for per-CPU and per-rcu_node kthreads.  These
  * handle all flavors of RCU.
@@ -98,8 +100,11 @@
 DEFINE_PER_CPU(char, rcu_cpu_has_work);
 static char rcu_kthreads_spawnable;
 
+#endif /* #ifdef CONFIG_RCU_BOOST */
+
 static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu);
-static void invoke_rcu_cpu_kthread(void);
+static void invoke_rcu_core(void);
+static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp);
 
 #define RCU_KTHREAD_PRIO 1	/* RT priority for per-CPU kthreads. */
 
@@ -1088,14 +1093,8 @@
 	int need_report = 0;
 	struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu);
 	struct rcu_node *rnp;
-	struct task_struct *t;
 
-	/* Stop the CPU's kthread. */
-	t = per_cpu(rcu_cpu_kthread_task, cpu);
-	if (t != NULL) {
-		per_cpu(rcu_cpu_kthread_task, cpu) = NULL;
-		kthread_stop(t);
-	}
+	rcu_stop_cpu_kthread(cpu);
 
 	/* Exclude any attempts to start a new grace period. */
 	raw_spin_lock_irqsave(&rsp->onofflock, flags);
@@ -1231,7 +1230,7 @@
 
 	/* Re-raise the RCU softirq if there are callbacks remaining. */
 	if (cpu_has_callbacks_ready_to_invoke(rdp))
-		invoke_rcu_cpu_kthread();
+		invoke_rcu_core();
 }
 
 /*
@@ -1277,7 +1276,7 @@
 	}
 	rcu_preempt_check_callbacks(cpu);
 	if (rcu_pending(cpu))
-		invoke_rcu_cpu_kthread();
+		invoke_rcu_core();
 }
 
 #ifdef CONFIG_SMP
@@ -1442,13 +1441,14 @@
 	}
 
 	/* If there are callbacks ready, invoke them. */
-	rcu_do_batch(rsp, rdp);
+	if (cpu_has_callbacks_ready_to_invoke(rdp))
+		invoke_rcu_callbacks(rsp, rdp);
 }
 
 /*
  * Do softirq processing for the current CPU.
  */
-static void rcu_process_callbacks(void)
+static void rcu_process_callbacks(struct softirq_action *unused)
 {
 	__rcu_process_callbacks(&rcu_sched_state,
 				&__get_cpu_var(rcu_sched_data));
@@ -1465,342 +1465,20 @@
  * the current CPU with interrupts disabled, the rcu_cpu_kthread_task
  * cannot disappear out from under us.
  */
-static void invoke_rcu_cpu_kthread(void)
+static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp)
 {
-	unsigned long flags;
-
-	local_irq_save(flags);
-	__this_cpu_write(rcu_cpu_has_work, 1);
-	if (__this_cpu_read(rcu_cpu_kthread_task) == NULL) {
-		local_irq_restore(flags);
+	if (likely(!rsp->boost)) {
+		rcu_do_batch(rsp, rdp);
 		return;
 	}
-	wake_up_process(__this_cpu_read(rcu_cpu_kthread_task));
-	local_irq_restore(flags);
+	invoke_rcu_callbacks_kthread();
 }
 
-/*
- * Wake up the specified per-rcu_node-structure kthread.
- * Because the per-rcu_node kthreads are immortal, we don't need
- * to do anything to keep them alive.
- */
-static void invoke_rcu_node_kthread(struct rcu_node *rnp)
+static void invoke_rcu_core(void)
 {
-	struct task_struct *t;
-
-	t = rnp->node_kthread_task;
-	if (t != NULL)
-		wake_up_process(t);
+	raise_softirq(RCU_SOFTIRQ);
 }
 
-/*
- * Set the specified CPU's kthread to run RT or not, as specified by
- * the to_rt argument.  The CPU-hotplug locks are held, so the task
- * is not going away.
- */
-static void rcu_cpu_kthread_setrt(int cpu, int to_rt)
-{
-	int policy;
-	struct sched_param sp;
-	struct task_struct *t;
-
-	t = per_cpu(rcu_cpu_kthread_task, cpu);
-	if (t == NULL)
-		return;
-	if (to_rt) {
-		policy = SCHED_FIFO;
-		sp.sched_priority = RCU_KTHREAD_PRIO;
-	} else {
-		policy = SCHED_NORMAL;
-		sp.sched_priority = 0;
-	}
-	sched_setscheduler_nocheck(t, policy, &sp);
-}
-
-/*
- * Timer handler to initiate the waking up of per-CPU kthreads that
- * have yielded the CPU due to excess numbers of RCU callbacks.
- * We wake up the per-rcu_node kthread, which in turn will wake up
- * the booster kthread.
- */
-static void rcu_cpu_kthread_timer(unsigned long arg)
-{
-	struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, arg);
-	struct rcu_node *rnp = rdp->mynode;
-
-	atomic_or(rdp->grpmask, &rnp->wakemask);
-	invoke_rcu_node_kthread(rnp);
-}
-
-/*
- * Drop to non-real-time priority and yield, but only after posting a
- * timer that will cause us to regain our real-time priority if we
- * remain preempted.  Either way, we restore our real-time priority
- * before returning.
- */
-static void rcu_yield(void (*f)(unsigned long), unsigned long arg)
-{
-	struct sched_param sp;
-	struct timer_list yield_timer;
-
-	setup_timer_on_stack(&yield_timer, f, arg);
-	mod_timer(&yield_timer, jiffies + 2);
-	sp.sched_priority = 0;
-	sched_setscheduler_nocheck(current, SCHED_NORMAL, &sp);
-	set_user_nice(current, 19);
-	schedule();
-	sp.sched_priority = RCU_KTHREAD_PRIO;
-	sched_setscheduler_nocheck(current, SCHED_FIFO, &sp);
-	del_timer(&yield_timer);
-}
-
-/*
- * Handle cases where the rcu_cpu_kthread() ends up on the wrong CPU.
- * This can happen while the corresponding CPU is either coming online
- * or going offline.  We cannot wait until the CPU is fully online
- * before starting the kthread, because the various notifier functions
- * can wait for RCU grace periods.  So we park rcu_cpu_kthread() until
- * the corresponding CPU is online.
- *
- * Return 1 if the kthread needs to stop, 0 otherwise.
- *
- * Caller must disable bh.  This function can momentarily enable it.
- */
-static int rcu_cpu_kthread_should_stop(int cpu)
-{
-	while (cpu_is_offline(cpu) ||
-	       !cpumask_equal(&current->cpus_allowed, cpumask_of(cpu)) ||
-	       smp_processor_id() != cpu) {
-		if (kthread_should_stop())
-			return 1;
-		per_cpu(rcu_cpu_kthread_status, cpu) = RCU_KTHREAD_OFFCPU;
-		per_cpu(rcu_cpu_kthread_cpu, cpu) = raw_smp_processor_id();
-		local_bh_enable();
-		schedule_timeout_uninterruptible(1);
-		if (!cpumask_equal(&current->cpus_allowed, cpumask_of(cpu)))
-			set_cpus_allowed_ptr(current, cpumask_of(cpu));
-		local_bh_disable();
-	}
-	per_cpu(rcu_cpu_kthread_cpu, cpu) = cpu;
-	return 0;
-}
-
-/*
- * Per-CPU kernel thread that invokes RCU callbacks.  This replaces the
- * earlier RCU softirq.
- */
-static int rcu_cpu_kthread(void *arg)
-{
-	int cpu = (int)(long)arg;
-	unsigned long flags;
-	int spincnt = 0;
-	unsigned int *statusp = &per_cpu(rcu_cpu_kthread_status, cpu);
-	char work;
-	char *workp = &per_cpu(rcu_cpu_has_work, cpu);
-
-	for (;;) {
-		*statusp = RCU_KTHREAD_WAITING;
-		rcu_wait(*workp != 0 || kthread_should_stop());
-		local_bh_disable();
-		if (rcu_cpu_kthread_should_stop(cpu)) {
-			local_bh_enable();
-			break;
-		}
-		*statusp = RCU_KTHREAD_RUNNING;
-		per_cpu(rcu_cpu_kthread_loops, cpu)++;
-		local_irq_save(flags);
-		work = *workp;
-		*workp = 0;
-		local_irq_restore(flags);
-		if (work)
-			rcu_process_callbacks();
-		local_bh_enable();
-		if (*workp != 0)
-			spincnt++;
-		else
-			spincnt = 0;
-		if (spincnt > 10) {
-			*statusp = RCU_KTHREAD_YIELDING;
-			rcu_yield(rcu_cpu_kthread_timer, (unsigned long)cpu);
-			spincnt = 0;
-		}
-	}
-	*statusp = RCU_KTHREAD_STOPPED;
-	return 0;
-}
-
-/*
- * Spawn a per-CPU kthread, setting up affinity and priority.
- * Because the CPU hotplug lock is held, no other CPU will be attempting
- * to manipulate rcu_cpu_kthread_task.  There might be another CPU
- * attempting to access it during boot, but the locking in kthread_bind()
- * will enforce sufficient ordering.
- */
-static int __cpuinit rcu_spawn_one_cpu_kthread(int cpu)
-{
-	struct sched_param sp;
-	struct task_struct *t;
-
-	if (!rcu_kthreads_spawnable ||
-	    per_cpu(rcu_cpu_kthread_task, cpu) != NULL)
-		return 0;
-	t = kthread_create(rcu_cpu_kthread, (void *)(long)cpu, "rcuc%d", cpu);
-	if (IS_ERR(t))
-		return PTR_ERR(t);
-	kthread_bind(t, cpu);
-	per_cpu(rcu_cpu_kthread_cpu, cpu) = cpu;
-	WARN_ON_ONCE(per_cpu(rcu_cpu_kthread_task, cpu) != NULL);
-	per_cpu(rcu_cpu_kthread_task, cpu) = t;
-	sp.sched_priority = RCU_KTHREAD_PRIO;
-	sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
-	return 0;
-}
-
-/*
- * Per-rcu_node kthread, which is in charge of waking up the per-CPU
- * kthreads when needed.  We ignore requests to wake up kthreads
- * for offline CPUs, which is OK because force_quiescent_state()
- * takes care of this case.
- */
-static int rcu_node_kthread(void *arg)
-{
-	int cpu;
-	unsigned long flags;
-	unsigned long mask;
-	struct rcu_node *rnp = (struct rcu_node *)arg;
-	struct sched_param sp;
-	struct task_struct *t;
-
-	for (;;) {
-		rnp->node_kthread_status = RCU_KTHREAD_WAITING;
-		rcu_wait(atomic_read(&rnp->wakemask) != 0);
-		rnp->node_kthread_status = RCU_KTHREAD_RUNNING;
-		raw_spin_lock_irqsave(&rnp->lock, flags);
-		mask = atomic_xchg(&rnp->wakemask, 0);
-		rcu_initiate_boost(rnp, flags); /* releases rnp->lock. */
-		for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++, mask >>= 1) {
-			if ((mask & 0x1) == 0)
-				continue;
-			preempt_disable();
-			t = per_cpu(rcu_cpu_kthread_task, cpu);
-			if (!cpu_online(cpu) || t == NULL) {
-				preempt_enable();
-				continue;
-			}
-			per_cpu(rcu_cpu_has_work, cpu) = 1;
-			sp.sched_priority = RCU_KTHREAD_PRIO;
-			sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
-			preempt_enable();
-		}
-	}
-	/* NOTREACHED */
-	rnp->node_kthread_status = RCU_KTHREAD_STOPPED;
-	return 0;
-}
-
-/*
- * Set the per-rcu_node kthread's affinity to cover all CPUs that are
- * served by the rcu_node in question.  The CPU hotplug lock is still
- * held, so the value of rnp->qsmaskinit will be stable.
- *
- * We don't include outgoingcpu in the affinity set, use -1 if there is
- * no outgoing CPU.  If there are no CPUs left in the affinity set,
- * this function allows the kthread to execute on any CPU.
- */
-static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu)
-{
-	cpumask_var_t cm;
-	int cpu;
-	unsigned long mask = rnp->qsmaskinit;
-
-	if (rnp->node_kthread_task == NULL)
-		return;
-	if (!alloc_cpumask_var(&cm, GFP_KERNEL))
-		return;
-	cpumask_clear(cm);
-	for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++, mask >>= 1)
-		if ((mask & 0x1) && cpu != outgoingcpu)
-			cpumask_set_cpu(cpu, cm);
-	if (cpumask_weight(cm) == 0) {
-		cpumask_setall(cm);
-		for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++)
-			cpumask_clear_cpu(cpu, cm);
-		WARN_ON_ONCE(cpumask_weight(cm) == 0);
-	}
-	set_cpus_allowed_ptr(rnp->node_kthread_task, cm);
-	rcu_boost_kthread_setaffinity(rnp, cm);
-	free_cpumask_var(cm);
-}
-
-/*
- * Spawn a per-rcu_node kthread, setting priority and affinity.
- * Called during boot before online/offline can happen, or, if
- * during runtime, with the main CPU-hotplug locks held.  So only
- * one of these can be executing at a time.
- */
-static int __cpuinit rcu_spawn_one_node_kthread(struct rcu_state *rsp,
-						struct rcu_node *rnp)
-{
-	unsigned long flags;
-	int rnp_index = rnp - &rsp->node[0];
-	struct sched_param sp;
-	struct task_struct *t;
-
-	if (!rcu_kthreads_spawnable ||
-	    rnp->qsmaskinit == 0)
-		return 0;
-	if (rnp->node_kthread_task == NULL) {
-		t = kthread_create(rcu_node_kthread, (void *)rnp,
-				   "rcun%d", rnp_index);
-		if (IS_ERR(t))
-			return PTR_ERR(t);
-		raw_spin_lock_irqsave(&rnp->lock, flags);
-		rnp->node_kthread_task = t;
-		raw_spin_unlock_irqrestore(&rnp->lock, flags);
-		sp.sched_priority = 99;
-		sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
-	}
-	return rcu_spawn_one_boost_kthread(rsp, rnp, rnp_index);
-}
-
-static void rcu_wake_one_boost_kthread(struct rcu_node *rnp);
-
-/*
- * Spawn all kthreads -- called as soon as the scheduler is running.
- */
-static int __init rcu_spawn_kthreads(void)
-{
-	int cpu;
-	struct rcu_node *rnp;
-	struct task_struct *t;
-
-	rcu_kthreads_spawnable = 1;
-	for_each_possible_cpu(cpu) {
-		per_cpu(rcu_cpu_has_work, cpu) = 0;
-		if (cpu_online(cpu)) {
-			(void)rcu_spawn_one_cpu_kthread(cpu);
-			t = per_cpu(rcu_cpu_kthread_task, cpu);
-			if (t)
-				wake_up_process(t);
-		}
-	}
-	rnp = rcu_get_root(rcu_state);
-	(void)rcu_spawn_one_node_kthread(rcu_state, rnp);
-	if (rnp->node_kthread_task)
-		wake_up_process(rnp->node_kthread_task);
-	if (NUM_RCU_NODES > 1) {
-		rcu_for_each_leaf_node(rcu_state, rnp) {
-			(void)rcu_spawn_one_node_kthread(rcu_state, rnp);
-			t = rnp->node_kthread_task;
-			if (t)
-				wake_up_process(t);
-			rcu_wake_one_boost_kthread(rnp);
-		}
-	}
-	return 0;
-}
-early_initcall(rcu_spawn_kthreads);
-
 static void
 __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu),
 	   struct rcu_state *rsp)
@@ -2207,44 +1885,6 @@
 	rcu_preempt_init_percpu_data(cpu);
 }
 
-static void __cpuinit rcu_prepare_kthreads(int cpu)
-{
-	struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu);
-	struct rcu_node *rnp = rdp->mynode;
-
-	/* Fire up the incoming CPU's kthread and leaf rcu_node kthread. */
-	if (rcu_kthreads_spawnable) {
-		(void)rcu_spawn_one_cpu_kthread(cpu);
-		if (rnp->node_kthread_task == NULL)
-			(void)rcu_spawn_one_node_kthread(rcu_state, rnp);
-	}
-}
-
-/*
- * kthread_create() creates threads in TASK_UNINTERRUPTIBLE state,
- * but the RCU threads are woken on demand, and if demand is low this
- * could be a while triggering the hung task watchdog.
- *
- * In order to avoid this, poke all tasks once the CPU is fully
- * up and running.
- */
-static void __cpuinit rcu_online_kthreads(int cpu)
-{
-	struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu);
-	struct rcu_node *rnp = rdp->mynode;
-	struct task_struct *t;
-
-	t = per_cpu(rcu_cpu_kthread_task, cpu);
-	if (t)
-		wake_up_process(t);
-
-	t = rnp->node_kthread_task;
-	if (t)
-		wake_up_process(t);
-
-	rcu_wake_one_boost_kthread(rnp);
-}
-
 /*
  * Handle CPU online/offline notification events.
  */
@@ -2262,7 +1902,6 @@
 		rcu_prepare_kthreads(cpu);
 		break;
 	case CPU_ONLINE:
-		rcu_online_kthreads(cpu);
 	case CPU_DOWN_FAILED:
 		rcu_node_kthread_setaffinity(rnp, -1);
 		rcu_cpu_kthread_setrt(cpu, 1);
@@ -2410,6 +2049,7 @@
 	rcu_init_one(&rcu_sched_state, &rcu_sched_data);
 	rcu_init_one(&rcu_bh_state, &rcu_bh_data);
 	__rcu_init_preempt();
+	 open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
 
 	/*
 	 * We don't need protection against CPU-hotplug here because
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index 7b9a08b..01b2ccd 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -369,6 +369,7 @@
 						/*  period because */
 						/*  force_quiescent_state() */
 						/*  was running. */
+	u8	boost;				/* Subject to priority boost. */
 	unsigned long gpnum;			/* Current gp number. */
 	unsigned long completed;		/* # of last completed gp. */
 
@@ -426,6 +427,7 @@
 #ifdef CONFIG_HOTPLUG_CPU
 static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp,
 				      unsigned long flags);
+static void rcu_stop_cpu_kthread(int cpu);
 #endif /* #ifdef CONFIG_HOTPLUG_CPU */
 static void rcu_print_detail_task_stall(struct rcu_state *rsp);
 static void rcu_print_task_stall(struct rcu_node *rnp);
@@ -450,11 +452,19 @@
 static void __init __rcu_init_preempt(void);
 static void rcu_needs_cpu_flush(void);
 static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags);
+static void rcu_preempt_boost_start_gp(struct rcu_node *rnp);
+static void invoke_rcu_callbacks_kthread(void);
+#ifdef CONFIG_RCU_BOOST
+static void rcu_preempt_do_callbacks(void);
 static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp,
 					  cpumask_var_t cm);
-static void rcu_preempt_boost_start_gp(struct rcu_node *rnp);
 static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
 						 struct rcu_node *rnp,
 						 int rnp_index);
+static void invoke_rcu_node_kthread(struct rcu_node *rnp);
+static void rcu_yield(void (*f)(unsigned long), unsigned long arg);
+#endif /* #ifdef CONFIG_RCU_BOOST */
+static void rcu_cpu_kthread_setrt(int cpu, int to_rt);
+static void __cpuinit rcu_prepare_kthreads(int cpu);
 
 #endif /* #ifndef RCU_TREE_NONCORE */
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index c8bff30..14dc7dd 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -602,6 +602,15 @@
 				&__get_cpu_var(rcu_preempt_data));
 }
 
+#ifdef CONFIG_RCU_BOOST
+
+static void rcu_preempt_do_callbacks(void)
+{
+	rcu_do_batch(&rcu_preempt_state, &__get_cpu_var(rcu_preempt_data));
+}
+
+#endif /* #ifdef CONFIG_RCU_BOOST */
+
 /*
  * Queue a preemptible-RCU callback for invocation after a grace period.
  */
@@ -1249,6 +1258,23 @@
 }
 
 /*
+ * Wake up the per-CPU kthread to invoke RCU callbacks.
+ */
+static void invoke_rcu_callbacks_kthread(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	__this_cpu_write(rcu_cpu_has_work, 1);
+	if (__this_cpu_read(rcu_cpu_kthread_task) == NULL) {
+		local_irq_restore(flags);
+		return;
+	}
+	wake_up_process(__this_cpu_read(rcu_cpu_kthread_task));
+	local_irq_restore(flags);
+}
+
+/*
  * Set the affinity of the boost kthread.  The CPU-hotplug locks are
  * held, so no one should be messing with the existence of the boost
  * kthread.
@@ -1288,6 +1314,7 @@
 
 	if (&rcu_preempt_state != rsp)
 		return 0;
+	rsp->boost = 1;
 	if (rnp->boost_kthread_task != NULL)
 		return 0;
 	t = kthread_create(rcu_boost_kthread, (void *)rnp,
@@ -1299,13 +1326,372 @@
 	raw_spin_unlock_irqrestore(&rnp->lock, flags);
 	sp.sched_priority = RCU_KTHREAD_PRIO;
 	sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
+	wake_up_process(t); /* get to TASK_INTERRUPTIBLE quickly. */
 	return 0;
 }
 
-static void __cpuinit rcu_wake_one_boost_kthread(struct rcu_node *rnp)
+#ifdef CONFIG_HOTPLUG_CPU
+
+/*
+ * Stop the RCU's per-CPU kthread when its CPU goes offline,.
+ */
+static void rcu_stop_cpu_kthread(int cpu)
 {
-	if (rnp->boost_kthread_task)
-		wake_up_process(rnp->boost_kthread_task);
+	struct task_struct *t;
+
+	/* Stop the CPU's kthread. */
+	t = per_cpu(rcu_cpu_kthread_task, cpu);
+	if (t != NULL) {
+		per_cpu(rcu_cpu_kthread_task, cpu) = NULL;
+		kthread_stop(t);
+	}
+}
+
+#endif /* #ifdef CONFIG_HOTPLUG_CPU */
+
+static void rcu_kthread_do_work(void)
+{
+	rcu_do_batch(&rcu_sched_state, &__get_cpu_var(rcu_sched_data));
+	rcu_do_batch(&rcu_bh_state, &__get_cpu_var(rcu_bh_data));
+	rcu_preempt_do_callbacks();
+}
+
+/*
+ * Wake up the specified per-rcu_node-structure kthread.
+ * Because the per-rcu_node kthreads are immortal, we don't need
+ * to do anything to keep them alive.
+ */
+static void invoke_rcu_node_kthread(struct rcu_node *rnp)
+{
+	struct task_struct *t;
+
+	t = rnp->node_kthread_task;
+	if (t != NULL)
+		wake_up_process(t);
+}
+
+/*
+ * Set the specified CPU's kthread to run RT or not, as specified by
+ * the to_rt argument.  The CPU-hotplug locks are held, so the task
+ * is not going away.
+ */
+static void rcu_cpu_kthread_setrt(int cpu, int to_rt)
+{
+	int policy;
+	struct sched_param sp;
+	struct task_struct *t;
+
+	t = per_cpu(rcu_cpu_kthread_task, cpu);
+	if (t == NULL)
+		return;
+	if (to_rt) {
+		policy = SCHED_FIFO;
+		sp.sched_priority = RCU_KTHREAD_PRIO;
+	} else {
+		policy = SCHED_NORMAL;
+		sp.sched_priority = 0;
+	}
+	sched_setscheduler_nocheck(t, policy, &sp);
+}
+
+/*
+ * Timer handler to initiate the waking up of per-CPU kthreads that
+ * have yielded the CPU due to excess numbers of RCU callbacks.
+ * We wake up the per-rcu_node kthread, which in turn will wake up
+ * the booster kthread.
+ */
+static void rcu_cpu_kthread_timer(unsigned long arg)
+{
+	struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, arg);
+	struct rcu_node *rnp = rdp->mynode;
+
+	atomic_or(rdp->grpmask, &rnp->wakemask);
+	invoke_rcu_node_kthread(rnp);
+}
+
+/*
+ * Drop to non-real-time priority and yield, but only after posting a
+ * timer that will cause us to regain our real-time priority if we
+ * remain preempted.  Either way, we restore our real-time priority
+ * before returning.
+ */
+static void rcu_yield(void (*f)(unsigned long), unsigned long arg)
+{
+	struct sched_param sp;
+	struct timer_list yield_timer;
+
+	setup_timer_on_stack(&yield_timer, f, arg);
+	mod_timer(&yield_timer, jiffies + 2);
+	sp.sched_priority = 0;
+	sched_setscheduler_nocheck(current, SCHED_NORMAL, &sp);
+	set_user_nice(current, 19);
+	schedule();
+	sp.sched_priority = RCU_KTHREAD_PRIO;
+	sched_setscheduler_nocheck(current, SCHED_FIFO, &sp);
+	del_timer(&yield_timer);
+}
+
+/*
+ * Handle cases where the rcu_cpu_kthread() ends up on the wrong CPU.
+ * This can happen while the corresponding CPU is either coming online
+ * or going offline.  We cannot wait until the CPU is fully online
+ * before starting the kthread, because the various notifier functions
+ * can wait for RCU grace periods.  So we park rcu_cpu_kthread() until
+ * the corresponding CPU is online.
+ *
+ * Return 1 if the kthread needs to stop, 0 otherwise.
+ *
+ * Caller must disable bh.  This function can momentarily enable it.
+ */
+static int rcu_cpu_kthread_should_stop(int cpu)
+{
+	while (cpu_is_offline(cpu) ||
+	       !cpumask_equal(&current->cpus_allowed, cpumask_of(cpu)) ||
+	       smp_processor_id() != cpu) {
+		if (kthread_should_stop())
+			return 1;
+		per_cpu(rcu_cpu_kthread_status, cpu) = RCU_KTHREAD_OFFCPU;
+		per_cpu(rcu_cpu_kthread_cpu, cpu) = raw_smp_processor_id();
+		local_bh_enable();
+		schedule_timeout_uninterruptible(1);
+		if (!cpumask_equal(&current->cpus_allowed, cpumask_of(cpu)))
+			set_cpus_allowed_ptr(current, cpumask_of(cpu));
+		local_bh_disable();
+	}
+	per_cpu(rcu_cpu_kthread_cpu, cpu) = cpu;
+	return 0;
+}
+
+/*
+ * Per-CPU kernel thread that invokes RCU callbacks.  This replaces the
+ * earlier RCU softirq.
+ */
+static int rcu_cpu_kthread(void *arg)
+{
+	int cpu = (int)(long)arg;
+	unsigned long flags;
+	int spincnt = 0;
+	unsigned int *statusp = &per_cpu(rcu_cpu_kthread_status, cpu);
+	char work;
+	char *workp = &per_cpu(rcu_cpu_has_work, cpu);
+
+	for (;;) {
+		*statusp = RCU_KTHREAD_WAITING;
+		rcu_wait(*workp != 0 || kthread_should_stop());
+		local_bh_disable();
+		if (rcu_cpu_kthread_should_stop(cpu)) {
+			local_bh_enable();
+			break;
+		}
+		*statusp = RCU_KTHREAD_RUNNING;
+		per_cpu(rcu_cpu_kthread_loops, cpu)++;
+		local_irq_save(flags);
+		work = *workp;
+		*workp = 0;
+		local_irq_restore(flags);
+		if (work)
+			rcu_kthread_do_work();
+		local_bh_enable();
+		if (*workp != 0)
+			spincnt++;
+		else
+			spincnt = 0;
+		if (spincnt > 10) {
+			*statusp = RCU_KTHREAD_YIELDING;
+			rcu_yield(rcu_cpu_kthread_timer, (unsigned long)cpu);
+			spincnt = 0;
+		}
+	}
+	*statusp = RCU_KTHREAD_STOPPED;
+	return 0;
+}
+
+/*
+ * Spawn a per-CPU kthread, setting up affinity and priority.
+ * Because the CPU hotplug lock is held, no other CPU will be attempting
+ * to manipulate rcu_cpu_kthread_task.  There might be another CPU
+ * attempting to access it during boot, but the locking in kthread_bind()
+ * will enforce sufficient ordering.
+ *
+ * Please note that we cannot simply refuse to wake up the per-CPU
+ * kthread because kthreads are created in TASK_UNINTERRUPTIBLE state,
+ * which can result in softlockup complaints if the task ends up being
+ * idle for more than a couple of minutes.
+ *
+ * However, please note also that we cannot bind the per-CPU kthread to its
+ * CPU until that CPU is fully online.  We also cannot wait until the
+ * CPU is fully online before we create its per-CPU kthread, as this would
+ * deadlock the system when CPU notifiers tried waiting for grace
+ * periods.  So we bind the per-CPU kthread to its CPU only if the CPU
+ * is online.  If its CPU is not yet fully online, then the code in
+ * rcu_cpu_kthread() will wait until it is fully online, and then do
+ * the binding.
+ */
+static int __cpuinit rcu_spawn_one_cpu_kthread(int cpu)
+{
+	struct sched_param sp;
+	struct task_struct *t;
+
+	if (!rcu_kthreads_spawnable ||
+	    per_cpu(rcu_cpu_kthread_task, cpu) != NULL)
+		return 0;
+	t = kthread_create(rcu_cpu_kthread, (void *)(long)cpu, "rcuc%d", cpu);
+	if (IS_ERR(t))
+		return PTR_ERR(t);
+	if (cpu_online(cpu))
+		kthread_bind(t, cpu);
+	per_cpu(rcu_cpu_kthread_cpu, cpu) = cpu;
+	WARN_ON_ONCE(per_cpu(rcu_cpu_kthread_task, cpu) != NULL);
+	sp.sched_priority = RCU_KTHREAD_PRIO;
+	sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
+	per_cpu(rcu_cpu_kthread_task, cpu) = t;
+	wake_up_process(t); /* Get to TASK_INTERRUPTIBLE quickly. */
+	return 0;
+}
+
+/*
+ * Per-rcu_node kthread, which is in charge of waking up the per-CPU
+ * kthreads when needed.  We ignore requests to wake up kthreads
+ * for offline CPUs, which is OK because force_quiescent_state()
+ * takes care of this case.
+ */
+static int rcu_node_kthread(void *arg)
+{
+	int cpu;
+	unsigned long flags;
+	unsigned long mask;
+	struct rcu_node *rnp = (struct rcu_node *)arg;
+	struct sched_param sp;
+	struct task_struct *t;
+
+	for (;;) {
+		rnp->node_kthread_status = RCU_KTHREAD_WAITING;
+		rcu_wait(atomic_read(&rnp->wakemask) != 0);
+		rnp->node_kthread_status = RCU_KTHREAD_RUNNING;
+		raw_spin_lock_irqsave(&rnp->lock, flags);
+		mask = atomic_xchg(&rnp->wakemask, 0);
+		rcu_initiate_boost(rnp, flags); /* releases rnp->lock. */
+		for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++, mask >>= 1) {
+			if ((mask & 0x1) == 0)
+				continue;
+			preempt_disable();
+			t = per_cpu(rcu_cpu_kthread_task, cpu);
+			if (!cpu_online(cpu) || t == NULL) {
+				preempt_enable();
+				continue;
+			}
+			per_cpu(rcu_cpu_has_work, cpu) = 1;
+			sp.sched_priority = RCU_KTHREAD_PRIO;
+			sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
+			preempt_enable();
+		}
+	}
+	/* NOTREACHED */
+	rnp->node_kthread_status = RCU_KTHREAD_STOPPED;
+	return 0;
+}
+
+/*
+ * Set the per-rcu_node kthread's affinity to cover all CPUs that are
+ * served by the rcu_node in question.  The CPU hotplug lock is still
+ * held, so the value of rnp->qsmaskinit will be stable.
+ *
+ * We don't include outgoingcpu in the affinity set, use -1 if there is
+ * no outgoing CPU.  If there are no CPUs left in the affinity set,
+ * this function allows the kthread to execute on any CPU.
+ */
+static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu)
+{
+	cpumask_var_t cm;
+	int cpu;
+	unsigned long mask = rnp->qsmaskinit;
+
+	if (rnp->node_kthread_task == NULL)
+		return;
+	if (!alloc_cpumask_var(&cm, GFP_KERNEL))
+		return;
+	cpumask_clear(cm);
+	for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++, mask >>= 1)
+		if ((mask & 0x1) && cpu != outgoingcpu)
+			cpumask_set_cpu(cpu, cm);
+	if (cpumask_weight(cm) == 0) {
+		cpumask_setall(cm);
+		for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++)
+			cpumask_clear_cpu(cpu, cm);
+		WARN_ON_ONCE(cpumask_weight(cm) == 0);
+	}
+	set_cpus_allowed_ptr(rnp->node_kthread_task, cm);
+	rcu_boost_kthread_setaffinity(rnp, cm);
+	free_cpumask_var(cm);
+}
+
+/*
+ * Spawn a per-rcu_node kthread, setting priority and affinity.
+ * Called during boot before online/offline can happen, or, if
+ * during runtime, with the main CPU-hotplug locks held.  So only
+ * one of these can be executing at a time.
+ */
+static int __cpuinit rcu_spawn_one_node_kthread(struct rcu_state *rsp,
+						struct rcu_node *rnp)
+{
+	unsigned long flags;
+	int rnp_index = rnp - &rsp->node[0];
+	struct sched_param sp;
+	struct task_struct *t;
+
+	if (!rcu_kthreads_spawnable ||
+	    rnp->qsmaskinit == 0)
+		return 0;
+	if (rnp->node_kthread_task == NULL) {
+		t = kthread_create(rcu_node_kthread, (void *)rnp,
+				   "rcun%d", rnp_index);
+		if (IS_ERR(t))
+			return PTR_ERR(t);
+		raw_spin_lock_irqsave(&rnp->lock, flags);
+		rnp->node_kthread_task = t;
+		raw_spin_unlock_irqrestore(&rnp->lock, flags);
+		sp.sched_priority = 99;
+		sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
+		wake_up_process(t); /* get to TASK_INTERRUPTIBLE quickly. */
+	}
+	return rcu_spawn_one_boost_kthread(rsp, rnp, rnp_index);
+}
+
+/*
+ * Spawn all kthreads -- called as soon as the scheduler is running.
+ */
+static int __init rcu_spawn_kthreads(void)
+{
+	int cpu;
+	struct rcu_node *rnp;
+
+	rcu_kthreads_spawnable = 1;
+	for_each_possible_cpu(cpu) {
+		per_cpu(rcu_cpu_has_work, cpu) = 0;
+		if (cpu_online(cpu))
+			(void)rcu_spawn_one_cpu_kthread(cpu);
+	}
+	rnp = rcu_get_root(rcu_state);
+	(void)rcu_spawn_one_node_kthread(rcu_state, rnp);
+	if (NUM_RCU_NODES > 1) {
+		rcu_for_each_leaf_node(rcu_state, rnp)
+			(void)rcu_spawn_one_node_kthread(rcu_state, rnp);
+	}
+	return 0;
+}
+early_initcall(rcu_spawn_kthreads);
+
+static void __cpuinit rcu_prepare_kthreads(int cpu)
+{
+	struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu);
+	struct rcu_node *rnp = rdp->mynode;
+
+	/* Fire up the incoming CPU's kthread and leaf rcu_node kthread. */
+	if (rcu_kthreads_spawnable) {
+		(void)rcu_spawn_one_cpu_kthread(cpu);
+		if (rnp->node_kthread_task == NULL)
+			(void)rcu_spawn_one_node_kthread(rcu_state, rnp);
+	}
 }
 
 #else /* #ifdef CONFIG_RCU_BOOST */
@@ -1315,23 +1701,32 @@
 	raw_spin_unlock_irqrestore(&rnp->lock, flags);
 }
 
-static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp,
-					  cpumask_var_t cm)
+static void invoke_rcu_callbacks_kthread(void)
 {
+	WARN_ON_ONCE(1);
 }
 
 static void rcu_preempt_boost_start_gp(struct rcu_node *rnp)
 {
 }
 
-static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
-						 struct rcu_node *rnp,
-						 int rnp_index)
+#ifdef CONFIG_HOTPLUG_CPU
+
+static void rcu_stop_cpu_kthread(int cpu)
 {
-	return 0;
 }
 
-static void __cpuinit rcu_wake_one_boost_kthread(struct rcu_node *rnp)
+#endif /* #ifdef CONFIG_HOTPLUG_CPU */
+
+static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu)
+{
+}
+
+static void rcu_cpu_kthread_setrt(int cpu, int to_rt)
+{
+}
+
+static void __cpuinit rcu_prepare_kthreads(int cpu)
 {
 }
 
@@ -1509,7 +1904,7 @@
  *
  * Because it is not legal to invoke rcu_process_callbacks() with irqs
  * disabled, we do one pass of force_quiescent_state(), then do a
- * invoke_rcu_cpu_kthread() to cause rcu_process_callbacks() to be invoked
+ * invoke_rcu_core() to cause rcu_process_callbacks() to be invoked
  * later.  The per-cpu rcu_dyntick_drain variable controls the sequencing.
  */
 int rcu_needs_cpu(int cpu)
@@ -1560,7 +1955,7 @@
 
 	/* If RCU callbacks are still pending, RCU still needs this CPU. */
 	if (c)
-		invoke_rcu_cpu_kthread();
+		invoke_rcu_core();
 	return c;
 }
 
diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c
index 9678cc3..4e14487 100644
--- a/kernel/rcutree_trace.c
+++ b/kernel/rcutree_trace.c
@@ -46,6 +46,8 @@
 #define RCU_TREE_NONCORE
 #include "rcutree.h"
 
+#ifdef CONFIG_RCU_BOOST
+
 DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status);
 DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_cpu);
 DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_loops);
@@ -58,6 +60,8 @@
 	return "SRWOY"[kthread_status];
 }
 
+#endif /* #ifdef CONFIG_RCU_BOOST */
+
 static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp)
 {
 	if (!rdp->beenonline)
@@ -76,7 +80,7 @@
 		   rdp->dynticks_fqs);
 #endif /* #ifdef CONFIG_NO_HZ */
 	seq_printf(m, " of=%lu ri=%lu", rdp->offline_fqs, rdp->resched_ipi);
-	seq_printf(m, " ql=%ld qs=%c%c%c%c kt=%d/%c/%d ktl=%x b=%ld",
+	seq_printf(m, " ql=%ld qs=%c%c%c%c",
 		   rdp->qlen,
 		   ".N"[rdp->nxttail[RCU_NEXT_READY_TAIL] !=
 			rdp->nxttail[RCU_NEXT_TAIL]],
@@ -84,13 +88,16 @@
 			rdp->nxttail[RCU_NEXT_READY_TAIL]],
 		   ".W"[rdp->nxttail[RCU_DONE_TAIL] !=
 			rdp->nxttail[RCU_WAIT_TAIL]],
-		   ".D"[&rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]],
+		   ".D"[&rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]]);
+#ifdef CONFIG_RCU_BOOST
+	seq_printf(m, " kt=%d/%c/%d ktl=%x",
 		   per_cpu(rcu_cpu_has_work, rdp->cpu),
 		   convert_kthread_status(per_cpu(rcu_cpu_kthread_status,
 					  rdp->cpu)),
 		   per_cpu(rcu_cpu_kthread_cpu, rdp->cpu),
-		   per_cpu(rcu_cpu_kthread_loops, rdp->cpu) & 0xffff,
-		   rdp->blimit);
+		   per_cpu(rcu_cpu_kthread_loops, rdp->cpu) & 0xffff);
+#endif /* #ifdef CONFIG_RCU_BOOST */
+	seq_printf(m, " b=%ld", rdp->blimit);
 	seq_printf(m, " ci=%lu co=%lu ca=%lu\n",
 		   rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted);
 }
@@ -147,18 +154,21 @@
 		   rdp->dynticks_fqs);
 #endif /* #ifdef CONFIG_NO_HZ */
 	seq_printf(m, ",%lu,%lu", rdp->offline_fqs, rdp->resched_ipi);
-	seq_printf(m, ",%ld,\"%c%c%c%c\",%d,\"%c\",%ld", rdp->qlen,
+	seq_printf(m, ",%ld,\"%c%c%c%c\"", rdp->qlen,
 		   ".N"[rdp->nxttail[RCU_NEXT_READY_TAIL] !=
 			rdp->nxttail[RCU_NEXT_TAIL]],
 		   ".R"[rdp->nxttail[RCU_WAIT_TAIL] !=
 			rdp->nxttail[RCU_NEXT_READY_TAIL]],
 		   ".W"[rdp->nxttail[RCU_DONE_TAIL] !=
 			rdp->nxttail[RCU_WAIT_TAIL]],
-		   ".D"[&rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]],
+		   ".D"[&rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]]);
+#ifdef CONFIG_RCU_BOOST
+	seq_printf(m, ",%d,\"%c\"",
 		   per_cpu(rcu_cpu_has_work, rdp->cpu),
 		   convert_kthread_status(per_cpu(rcu_cpu_kthread_status,
-					  rdp->cpu)),
-		   rdp->blimit);
+					  rdp->cpu)));
+#endif /* #ifdef CONFIG_RCU_BOOST */
+	seq_printf(m, ",%ld", rdp->blimit);
 	seq_printf(m, ",%lu,%lu,%lu\n",
 		   rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted);
 }
@@ -169,7 +179,11 @@
 #ifdef CONFIG_NO_HZ
 	seq_puts(m, "\"dt\",\"dt nesting\",\"dt NMI nesting\",\"df\",");
 #endif /* #ifdef CONFIG_NO_HZ */
-	seq_puts(m, "\"of\",\"ri\",\"ql\",\"b\",\"ci\",\"co\",\"ca\"\n");
+	seq_puts(m, "\"of\",\"ri\",\"ql\",\"qs\"");
+#ifdef CONFIG_RCU_BOOST
+	seq_puts(m, "\"kt\",\"ktl\"");
+#endif /* #ifdef CONFIG_RCU_BOOST */
+	seq_puts(m, ",\"b\",\"ci\",\"co\",\"ca\"\n");
 #ifdef CONFIG_TREE_PREEMPT_RCU
 	seq_puts(m, "\"rcu_preempt:\"\n");
 	PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data_csv, m);
diff --git a/kernel/smp.c b/kernel/smp.c
index 73a1951..fb67dfa 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -74,7 +74,7 @@
 	.notifier_call		= hotplug_cfd,
 };
 
-static int __cpuinit init_call_single_data(void)
+void __init call_function_init(void)
 {
 	void *cpu = (void *)(long)smp_processor_id();
 	int i;
@@ -88,10 +88,7 @@
 
 	hotplug_cfd(&hotplug_cfd_notifier, CPU_UP_PREPARE, cpu);
 	register_cpu_notifier(&hotplug_cfd_notifier);
-
-	return 0;
 }
-early_initcall(init_call_single_data);
 
 /*
  * csd_lock/csd_unlock used to serialize access to per-cpu csd resources
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 1396017..40cf63d 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -58,7 +58,7 @@
 
 char *softirq_to_name[NR_SOFTIRQS] = {
 	"HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "BLOCK_IOPOLL",
-	"TASKLET", "SCHED", "HRTIMER"
+	"TASKLET", "SCHED", "HRTIMER", "RCU"
 };
 
 /*
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 1c95fd6..e0980f0 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -185,7 +185,6 @@
 static struct timer_list watchdog_timer;
 static DECLARE_WORK(watchdog_work, clocksource_watchdog_work);
 static DEFINE_SPINLOCK(watchdog_lock);
-static cycle_t watchdog_last;
 static int watchdog_running;
 
 static int clocksource_watchdog_kthread(void *data);
@@ -254,11 +253,6 @@
 	if (!watchdog_running)
 		goto out;
 
-	wdnow = watchdog->read(watchdog);
-	wd_nsec = clocksource_cyc2ns((wdnow - watchdog_last) & watchdog->mask,
-				     watchdog->mult, watchdog->shift);
-	watchdog_last = wdnow;
-
 	list_for_each_entry(cs, &watchdog_list, wd_list) {
 
 		/* Clocksource already marked unstable? */
@@ -268,19 +262,28 @@
 			continue;
 		}
 
+		local_irq_disable();
 		csnow = cs->read(cs);
+		wdnow = watchdog->read(watchdog);
+		local_irq_enable();
 
 		/* Clocksource initialized ? */
 		if (!(cs->flags & CLOCK_SOURCE_WATCHDOG)) {
 			cs->flags |= CLOCK_SOURCE_WATCHDOG;
-			cs->wd_last = csnow;
+			cs->wd_last = wdnow;
+			cs->cs_last = csnow;
 			continue;
 		}
 
-		/* Check the deviation from the watchdog clocksource. */
-		cs_nsec = clocksource_cyc2ns((csnow - cs->wd_last) &
+		wd_nsec = clocksource_cyc2ns((wdnow - cs->wd_last) & watchdog->mask,
+					     watchdog->mult, watchdog->shift);
+
+		cs_nsec = clocksource_cyc2ns((csnow - cs->cs_last) &
 					     cs->mask, cs->mult, cs->shift);
-		cs->wd_last = csnow;
+		cs->cs_last = csnow;
+		cs->wd_last = wdnow;
+
+		/* Check the deviation from the watchdog clocksource. */
 		if (abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) {
 			clocksource_unstable(cs, cs_nsec - wd_nsec);
 			continue;
@@ -318,7 +321,6 @@
 		return;
 	init_timer(&watchdog_timer);
 	watchdog_timer.function = clocksource_watchdog;
-	watchdog_last = watchdog->read(watchdog);
 	watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL;
 	add_timer_on(&watchdog_timer, cpumask_first(cpu_online_mask));
 	watchdog_running = 1;
diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c
index dff763b..1f06468 100644
--- a/kernel/trace/trace_printk.c
+++ b/kernel/trace/trace_printk.c
@@ -240,13 +240,10 @@
 	const char **fmt = v;
 	int start_index;
 
-	if (!fmt)
-		fmt = __start___trace_bprintk_fmt + *pos;
-
 	start_index = __stop___trace_bprintk_fmt - __start___trace_bprintk_fmt;
 
 	if (*pos < start_index)
-		return fmt;
+		return __start___trace_bprintk_fmt + *pos;
 
 	return find_next_mod_format(start_index, v, fmt, pos);
 }
diff --git a/mm/rmap.c b/mm/rmap.c
index 0eb463e..27dfd3b 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -112,9 +112,9 @@
 	kmem_cache_free(anon_vma_cachep, anon_vma);
 }
 
-static inline struct anon_vma_chain *anon_vma_chain_alloc(void)
+static inline struct anon_vma_chain *anon_vma_chain_alloc(gfp_t gfp)
 {
-	return kmem_cache_alloc(anon_vma_chain_cachep, GFP_KERNEL);
+	return kmem_cache_alloc(anon_vma_chain_cachep, gfp);
 }
 
 static void anon_vma_chain_free(struct anon_vma_chain *anon_vma_chain)
@@ -159,7 +159,7 @@
 		struct mm_struct *mm = vma->vm_mm;
 		struct anon_vma *allocated;
 
-		avc = anon_vma_chain_alloc();
+		avc = anon_vma_chain_alloc(GFP_KERNEL);
 		if (!avc)
 			goto out_enomem;
 
@@ -200,6 +200,32 @@
 	return -ENOMEM;
 }
 
+/*
+ * This is a useful helper function for locking the anon_vma root as
+ * we traverse the vma->anon_vma_chain, looping over anon_vma's that
+ * have the same vma.
+ *
+ * Such anon_vma's should have the same root, so you'd expect to see
+ * just a single mutex_lock for the whole traversal.
+ */
+static inline struct anon_vma *lock_anon_vma_root(struct anon_vma *root, struct anon_vma *anon_vma)
+{
+	struct anon_vma *new_root = anon_vma->root;
+	if (new_root != root) {
+		if (WARN_ON_ONCE(root))
+			mutex_unlock(&root->mutex);
+		root = new_root;
+		mutex_lock(&root->mutex);
+	}
+	return root;
+}
+
+static inline void unlock_anon_vma_root(struct anon_vma *root)
+{
+	if (root)
+		mutex_unlock(&root->mutex);
+}
+
 static void anon_vma_chain_link(struct vm_area_struct *vma,
 				struct anon_vma_chain *avc,
 				struct anon_vma *anon_vma)
@@ -208,13 +234,11 @@
 	avc->anon_vma = anon_vma;
 	list_add(&avc->same_vma, &vma->anon_vma_chain);
 
-	anon_vma_lock(anon_vma);
 	/*
 	 * It's critical to add new vmas to the tail of the anon_vma,
 	 * see comment in huge_memory.c:__split_huge_page().
 	 */
 	list_add_tail(&avc->same_anon_vma, &anon_vma->head);
-	anon_vma_unlock(anon_vma);
 }
 
 /*
@@ -224,13 +248,24 @@
 int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
 {
 	struct anon_vma_chain *avc, *pavc;
+	struct anon_vma *root = NULL;
 
 	list_for_each_entry_reverse(pavc, &src->anon_vma_chain, same_vma) {
-		avc = anon_vma_chain_alloc();
-		if (!avc)
-			goto enomem_failure;
-		anon_vma_chain_link(dst, avc, pavc->anon_vma);
+		struct anon_vma *anon_vma;
+
+		avc = anon_vma_chain_alloc(GFP_NOWAIT | __GFP_NOWARN);
+		if (unlikely(!avc)) {
+			unlock_anon_vma_root(root);
+			root = NULL;
+			avc = anon_vma_chain_alloc(GFP_KERNEL);
+			if (!avc)
+				goto enomem_failure;
+		}
+		anon_vma = pavc->anon_vma;
+		root = lock_anon_vma_root(root, anon_vma);
+		anon_vma_chain_link(dst, avc, anon_vma);
 	}
+	unlock_anon_vma_root(root);
 	return 0;
 
  enomem_failure:
@@ -263,7 +298,7 @@
 	anon_vma = anon_vma_alloc();
 	if (!anon_vma)
 		goto out_error;
-	avc = anon_vma_chain_alloc();
+	avc = anon_vma_chain_alloc(GFP_KERNEL);
 	if (!avc)
 		goto out_error_free_anon_vma;
 
@@ -280,7 +315,9 @@
 	get_anon_vma(anon_vma->root);
 	/* Mark this anon_vma as the one where our new (COWed) pages go. */
 	vma->anon_vma = anon_vma;
+	anon_vma_lock(anon_vma);
 	anon_vma_chain_link(vma, avc, anon_vma);
+	anon_vma_unlock(anon_vma);
 
 	return 0;
 
@@ -291,36 +328,43 @@
 	return -ENOMEM;
 }
 
-static void anon_vma_unlink(struct anon_vma_chain *anon_vma_chain)
-{
-	struct anon_vma *anon_vma = anon_vma_chain->anon_vma;
-	int empty;
-
-	/* If anon_vma_fork fails, we can get an empty anon_vma_chain. */
-	if (!anon_vma)
-		return;
-
-	anon_vma_lock(anon_vma);
-	list_del(&anon_vma_chain->same_anon_vma);
-
-	/* We must garbage collect the anon_vma if it's empty */
-	empty = list_empty(&anon_vma->head);
-	anon_vma_unlock(anon_vma);
-
-	if (empty)
-		put_anon_vma(anon_vma);
-}
-
 void unlink_anon_vmas(struct vm_area_struct *vma)
 {
 	struct anon_vma_chain *avc, *next;
+	struct anon_vma *root = NULL;
 
 	/*
 	 * Unlink each anon_vma chained to the VMA.  This list is ordered
 	 * from newest to oldest, ensuring the root anon_vma gets freed last.
 	 */
 	list_for_each_entry_safe(avc, next, &vma->anon_vma_chain, same_vma) {
-		anon_vma_unlink(avc);
+		struct anon_vma *anon_vma = avc->anon_vma;
+
+		root = lock_anon_vma_root(root, anon_vma);
+		list_del(&avc->same_anon_vma);
+
+		/*
+		 * Leave empty anon_vmas on the list - we'll need
+		 * to free them outside the lock.
+		 */
+		if (list_empty(&anon_vma->head))
+			continue;
+
+		list_del(&avc->same_vma);
+		anon_vma_chain_free(avc);
+	}
+	unlock_anon_vma_root(root);
+
+	/*
+	 * Iterate the list once more, it now only contains empty and unlinked
+	 * anon_vmas, destroy them. Could not do before due to __put_anon_vma()
+	 * needing to acquire the anon_vma->root->mutex.
+	 */
+	list_for_each_entry_safe(avc, next, &vma->anon_vma_chain, same_vma) {
+		struct anon_vma *anon_vma = avc->anon_vma;
+
+		put_anon_vma(anon_vma);
+
 		list_del(&avc->same_vma);
 		anon_vma_chain_free(avc);
 	}
diff --git a/security/keys/request_key.c b/security/keys/request_key.c
index d31862e..8e319a4 100644
--- a/security/keys/request_key.c
+++ b/security/keys/request_key.c
@@ -71,9 +71,8 @@
  * This is called in context of freshly forked kthread before kernel_execve(),
  * so we can simply install the desired session_keyring at this point.
  */
-static int umh_keys_init(struct subprocess_info *info)
+static int umh_keys_init(struct subprocess_info *info, struct cred *cred)
 {
-	struct cred *cred = (struct cred*)current_cred();
 	struct key *keyring = info->data;
 
 	return install_session_keyring_to_cred(cred, keyring);
diff --git a/sound/firewire/isight.c b/sound/firewire/isight.c
index 86ee16c..4400308 100644
--- a/sound/firewire/isight.c
+++ b/sound/firewire/isight.c
@@ -209,6 +209,7 @@
 		isight->packet_index = -1;
 		return;
 	}
+	fw_iso_context_queue_flush(isight->context);
 
 	if (++index >= QUEUE_LENGTH)
 		index = 0;
diff --git a/sound/pci/emu10k1/emu10k1_main.c b/sound/pci/emu10k1/emu10k1_main.c
index 5e619a8..15f0161 100644
--- a/sound/pci/emu10k1/emu10k1_main.c
+++ b/sound/pci/emu10k1/emu10k1_main.c
@@ -1440,6 +1440,14 @@
 	 .ca0102_chip = 1,
 	 .spk71 = 1,
 	 .emu_model = EMU_MODEL_EMU0404}, /* EMU 0404 */
+	/* EMU0404 PCIe */
+	{.vendor = 0x1102, .device = 0x0008, .subsystem = 0x40051102,
+	 .driver = "Audigy2", .name = "E-mu 0404 PCIe [MAEM8984]",
+	 .id = "EMU0404",
+	 .emu10k2_chip = 1,
+	 .ca0108_chip = 1,
+	 .spk71 = 1,
+	 .emu_model = EMU_MODEL_EMU0404}, /* EMU 0404 PCIe ver_03 */
 	/* Note that all E-mu cards require kernel 2.6 or newer. */
 	{.vendor = 0x1102, .device = 0x0008,
 	 .driver = "Audigy2", .name = "SB Audigy 2 Value [Unknown]",
diff --git a/sound/pci/hda/hda_beep.h b/sound/pci/hda/hda_beep.h
index f1de1ba..55f0647 100644
--- a/sound/pci/hda/hda_beep.h
+++ b/sound/pci/hda/hda_beep.h
@@ -50,7 +50,12 @@
 int snd_hda_attach_beep_device(struct hda_codec *codec, int nid);
 void snd_hda_detach_beep_device(struct hda_codec *codec);
 #else
-#define snd_hda_attach_beep_device(...)		0
-#define snd_hda_detach_beep_device(...)
+static inline int snd_hda_attach_beep_device(struct hda_codec *codec, int nid)
+{
+	return 0;
+}
+static inline void snd_hda_detach_beep_device(struct hda_codec *codec)
+{
+}
 #endif
 #endif
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 43fcfbd..61a774b 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -13316,9 +13316,8 @@
 	struct alc_spec *spec = codec->spec;
 	spec->autocfg.hp_pins[0] = 0x15;
 	spec->autocfg.speaker_pins[0] = 0x14;
-	spec->automute_mixer_nid[0] = 0x0f;
 	spec->automute = 1;
-	spec->automute_mode = ALC_AUTOMUTE_MIXER;
+	spec->automute_mode = ALC_AUTOMUTE_AMP;
 	spec->ext_mic.pin = 0x18;
 	spec->ext_mic.mux_idx = 0;
 	spec->int_mic.pin = 0x12;
diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c
index 605c99e..c952582 100644
--- a/sound/pci/hda/patch_via.c
+++ b/sound/pci/hda/patch_via.c
@@ -832,10 +832,13 @@
 	knew->subdevice = HDA_SUBDEV_NID_FLAG | nid;
 	knew->private_value = nid;
 
-	knew = via_clone_control(spec, &via_hp_mixer[1]);
-	if (knew == NULL)
-		return -ENOMEM;
-	knew->subdevice = side_mute_channel(spec);
+	nid = side_mute_channel(spec);
+	if (nid) {
+		knew = via_clone_control(spec, &via_hp_mixer[1]);
+		if (knew == NULL)
+			return -ENOMEM;
+		knew->subdevice = nid;
+	}
 
 	return 0;
 }
diff --git a/sound/pci/lola/lola.c b/sound/pci/lola/lola.c
index 34b2428..2692e5a 100644
--- a/sound/pci/lola/lola.c
+++ b/sound/pci/lola/lola.c
@@ -445,7 +445,7 @@
 	lola_setup_all_analog_gains(chip, PLAY, false); /* output, update */
 }
 
-static int lola_parse_tree(struct lola *chip)
+static int __devinit lola_parse_tree(struct lola *chip)
 {
 	unsigned int val;
 	int nid, err;
diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c
index 949691a..3f08afc 100644
--- a/sound/pci/rme9652/hdspm.c
+++ b/sound/pci/rme9652/hdspm.c
@@ -521,6 +521,7 @@
 #define HDSPM_DMA_AREA_KILOBYTES (HDSPM_DMA_AREA_BYTES/1024)
 
 /* revisions >= 230 indicate AES32 card */
+#define HDSPM_MADI_OLD_REV	207
 #define HDSPM_MADI_REV		210
 #define HDSPM_RAYDAT_REV	211
 #define HDSPM_AIO_REV		212
@@ -1143,7 +1144,7 @@
 
 		/* if wordclock has synced freq and wordclock is valid */
 		if ((status2 & HDSPM_wcLock) != 0 &&
-				(status & HDSPM_SelSyncRef0) == 0) {
+				(status2 & HDSPM_SelSyncRef0) == 0) {
 
 			rate_bits = status2 & HDSPM_wcFreqMask;
 
@@ -1639,12 +1640,14 @@
 		}
 	}
 	hmidi->pending = 0;
+	spin_unlock_irqrestore(&hmidi->lock, flags);
 
+	spin_lock_irqsave(&hmidi->hdspm->lock, flags);
 	hmidi->hdspm->control_register |= hmidi->ie;
 	hdspm_write(hmidi->hdspm, HDSPM_controlRegister,
 		    hmidi->hdspm->control_register);
+	spin_unlock_irqrestore(&hmidi->hdspm->lock, flags);
 
-	spin_unlock_irqrestore (&hmidi->lock, flags);
 	return snd_hdspm_midi_output_write (hmidi);
 }
 
@@ -6377,6 +6380,7 @@
 
 	switch (hdspm->firmware_rev) {
 	case HDSPM_MADI_REV:
+	case HDSPM_MADI_OLD_REV:
 		hdspm->io_type = MADI;
 		hdspm->card_name = "RME MADI";
 		hdspm->midiPorts = 3;
diff --git a/sound/usb/6fire/firmware.c b/sound/usb/6fire/firmware.c
index a91719d..1e3ae33 100644
--- a/sound/usb/6fire/firmware.c
+++ b/sound/usb/6fire/firmware.c
@@ -270,7 +270,6 @@
 	data = 0x00; /* resume ezusb cpu */
 	ret = usb6fire_fw_ezusb_write(device, 0xa0, 0xe600, &data, 1);
 	if (ret < 0) {
-		release_firmware(fw);
 		snd_printk(KERN_ERR PREFIX "unable to upload ezusb "
 				"firmware %s: end message.\n", fwname);
 		return ret;
diff --git a/sound/usb/6fire/pcm.c b/sound/usb/6fire/pcm.c
index b137b25..d144cdb 100644
--- a/sound/usb/6fire/pcm.c
+++ b/sound/usb/6fire/pcm.c
@@ -395,12 +395,12 @@
 	alsa_rt->hw = pcm_hw;
 
 	if (alsa_sub->stream == SNDRV_PCM_STREAM_PLAYBACK) {
-		if (rt->rate >= 0)
+		if (rt->rate < ARRAY_SIZE(rates))
 			alsa_rt->hw.rates = rates_alsaid[rt->rate];
 		alsa_rt->hw.channels_max = OUT_N_CHANNELS;
 		sub = &rt->playback;
 	} else if (alsa_sub->stream == SNDRV_PCM_STREAM_CAPTURE) {
-		if (rt->rate >= 0)
+		if (rt->rate < ARRAY_SIZE(rates))
 			alsa_rt->hw.rates = rates_alsaid[rt->rate];
 		alsa_rt->hw.channels_max = IN_N_CHANNELS;
 		sub = &rt->capture;
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 032ba63..940257b 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -633,7 +633,7 @@
 
 SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH))
 
-LIBS = -Wl,--whole-archive $(PERFLIBS) -Wl,--no-whole-archive $(EXTLIBS)
+LIBS = -Wl,--whole-archive $(PERFLIBS) -Wl,--no-whole-archive -Wl,--start-group $(EXTLIBS) -Wl,--end-group
 
 ALL_CFLAGS += $(BASIC_CFLAGS)
 ALL_CFLAGS += $(ARCH_CFLAGS)
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index 1e88485..0a7ed5b 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -2187,6 +2187,7 @@
 	{ "TASKLET_SOFTIRQ", 6 },
 	{ "SCHED_SOFTIRQ", 7 },
 	{ "HRTIMER_SOFTIRQ", 8 },
+	{ "RCU_SOFTIRQ", 9 },
 
 	{ "HRTIMER_NORESTART", 0 },
 	{ "HRTIMER_RESTART", 1 },