Merge tag 'kvm-s390-next-4.9-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD

KVM: s390: features and fixes for 4.9

- lazy enablement of runtime instrumentation
- up to 255 CPUs for nested guests
- rework of machine check deliver
- cleanups/fixes

Documentation/conf.py

@@ -131,7 +131,7 @@ pygments_style = 'sphinx'
 todo_include_todos = False
 
 primary_domain = 'C'
-highlight_language = 'C'
+highlight_language = 'guess'
 
 # -- Options for HTML output ----------------------------------------------
 

Documentation/hwmon/ftsteutates

@@ -19,5 +19,5 @@ enhancements. It can monitor up to 4 voltages, 16 temperatures and
 implemented in this driver.
 
 Specification of the chip can be found here:
-ftp:///pub/Mainboard-OEM-Sales/Services/Software&Tools/Linux_SystemMonitoring&Watchdog&GPIO/BMC-Teutates_Specification_V1.21.pdf
-ftp:///pub/Mainboard-OEM-Sales/Services/Software&Tools/Linux_SystemMonitoring&Watchdog&GPIO/Fujitsu_mainboards-1-Sensors_HowTo-en-US.pdf
+ftp://ftp.ts.fujitsu.com/pub/Mainboard-OEM-Sales/Services/Software&Tools/Linux_SystemMonitoring&Watchdog&GPIO/BMC-Teutates_Specification_V1.21.pdf
+ftp://ftp.ts.fujitsu.com/pub/Mainboard-OEM-Sales/Services/Software&Tools/Linux_SystemMonitoring&Watchdog&GPIO/Fujitsu_mainboards-1-Sensors_HowTo-en-US.pdf

Documentation/kernel-documentation.rst

@@ -366,8 +366,6 @@ Domain`_ references.
 Cross-referencing from reStructuredText
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. highlight:: none
-
 To cross-reference the functions and types defined in the kernel-doc comments
 from reStructuredText documents, please use the `Sphinx C Domain`_
 references. For example::
@@ -390,8 +388,6 @@ For further details, please refer to the `Sphinx C Domain`_ documentation.
 Function documentation
 ----------------------
 
-.. highlight:: c
-
 The general format of a function and function-like macro kernel-doc comment is::
 
   /**
@@ -572,8 +568,6 @@ DocBook XML [DEPRECATED]
 Converting DocBook to Sphinx
 ----------------------------
 
-.. highlight:: none
-
 Over time, we expect all of the documents under ``Documentation/DocBook`` to be
 converted to Sphinx and reStructuredText. For most DocBook XML documents, a good
 enough solution is to use the simple ``Documentation/sphinx/tmplcvt`` script,

Documentation/power/basic-pm-debugging.txt

@@ -164,7 +164,32 @@ load n/2 modules more and try again.
 Again, if you find the offending module(s), it(they) must be unloaded every time
 before hibernation, and please report the problem with it(them).
 
-c) Advanced debugging
+c) Using the "test_resume" hibernation option
+
+/sys/power/disk generally tells the kernel what to do after creating a
+hibernation image.  One of the available options is "test_resume" which
+causes the just created image to be used for immediate restoration.  Namely,
+after doing:
+
+# echo test_resume > /sys/power/disk
+# echo disk > /sys/power/state
+
+a hibernation image will be created and a resume from it will be triggered
+immediately without involving the platform firmware in any way.
+
+That test can be used to check if failures to resume from hibernation are
+related to bad interactions with the platform firmware.  That is, if the above
+works every time, but resume from actual hibernation does not work or is
+unreliable, the platform firmware may be responsible for the failures.
+
+On architectures and platforms that support using different kernels to restore
+hibernation images (that is, the kernel used to read the image from storage and
+load it into memory is different from the one included in the image) or support
+kernel address space randomization, it also can be used to check if failures
+to resume may be related to the differences between the restore and image
+kernels.
+
+d) Advanced debugging
 
 In case that hibernation does not work on your system even in the minimal
 configuration and compiling more drivers as modules is not practical or some

Documentation/power/interface.txt

@@ -1,75 +1,76 @@
-Power Management Interface
-
-
-The power management subsystem provides a unified sysfs interface to 
-userspace, regardless of what architecture or platform one is
-running. The interface exists in /sys/power/ directory (assuming sysfs
-is mounted at /sys). 
-
-/sys/power/state controls system power state. Reading from this file
-returns what states are supported, which is hard-coded to 'freeze',
-'standby' (Power-On Suspend), 'mem' (Suspend-to-RAM), and 'disk'
-(Suspend-to-Disk). 
-
-Writing to this file one of those strings causes the system to
-transition into that state. Please see the file
-Documentation/power/states.txt for a description of each of those
-states.
-
-
-/sys/power/disk controls the operating mode of the suspend-to-disk
-mechanism. Suspend-to-disk can be handled in several ways. We have a
-few options for putting the system to sleep - using the platform driver
-(e.g. ACPI or other suspend_ops), powering off the system or rebooting the
-system (for testing).
-
-Additionally, /sys/power/disk can be used to turn on one of the two testing
-modes of the suspend-to-disk mechanism: 'testproc' or 'test'.  If the
-suspend-to-disk mechanism is in the 'testproc' mode, writing 'disk' to
-/sys/power/state will cause the kernel to disable nonboot CPUs and freeze
-tasks, wait for 5 seconds, unfreeze tasks and enable nonboot CPUs.  If it is
-in the 'test' mode, writing 'disk' to /sys/power/state will cause the kernel
-to disable nonboot CPUs and freeze tasks, shrink memory, suspend devices, wait
-for 5 seconds, resume devices, unfreeze tasks and enable nonboot CPUs.  Then,
-we are able to look in the log messages and work out, for example, which code
-is being slow and which device drivers are misbehaving.
-
-Reading from this file will display all supported modes and the currently
-selected one in brackets, for example
-
-	[shutdown] reboot test testproc
-
-Writing to this file will accept one of
-
-       'platform' (only if the platform supports it)
-       'shutdown'
-       'reboot'
-       'testproc'
-       'test'
-
-/sys/power/image_size controls the size of the image created by
-the suspend-to-disk mechanism.  It can be written a string
-representing a non-negative integer that will be used as an upper
-limit of the image size, in bytes.  The suspend-to-disk mechanism will
-do its best to ensure the image size will not exceed that number.  However,
-if this turns out to be impossible, it will try to suspend anyway using the
-smallest image possible.  In particular, if "0" is written to this file, the
-suspend image will be as small as possible.
-
-Reading from this file will display the current image size limit, which
-is set to 2/5 of available RAM by default.
-
-/sys/power/pm_trace controls the code which saves the last PM event point in
-the RTC across reboots, so that you can debug a machine that just hangs
-during suspend (or more commonly, during resume).  Namely, the RTC is only
-used to save the last PM event point if this file contains '1'.  Initially it
-contains '0' which may be changed to '1' by writing a string representing a
-nonzero integer into it.
-
-To use this debugging feature you should attempt to suspend the machine, then
-reboot it and run
-
-	dmesg -s 1000000 | grep 'hash matches'
-
-CAUTION: Using it will cause your machine's real-time (CMOS) clock to be
-set to a random invalid time after a resume.
+Power Management Interface for System Sleep
+
+Copyright (c) 2016 Intel Corp., Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+
+The power management subsystem provides userspace with a unified sysfs interface
+for system sleep regardless of the underlying system architecture or platform.
+The interface is located in the /sys/power/ directory (assuming that sysfs is
+mounted at /sys).
+
+/sys/power/state is the system sleep state control file.
+
+Reading from it returns a list of supported sleep states, encoded as:
+
+'freeze' (Suspend-to-Idle)
+'standby' (Power-On Suspend)
+'mem' (Suspend-to-RAM)
+'disk' (Suspend-to-Disk)
+
+Suspend-to-Idle is always supported.  Suspend-to-Disk is always supported
+too as long the kernel has been configured to support hibernation at all
+(ie. CONFIG_HIBERNATION is set in the kernel configuration file).  Support
+for Suspend-to-RAM and Power-On Suspend depends on the capabilities of the
+platform.
+
+If one of the strings listed in /sys/power/state is written to it, the system
+will attempt to transition into the corresponding sleep state.  Refer to
+Documentation/power/states.txt for a description of each of those states.
+
+/sys/power/disk controls the operating mode of hibernation (Suspend-to-Disk).
+Specifically, it tells the kernel what to do after creating a hibernation image.
+
+Reading from it returns a list of supported options encoded as:
+
+'platform' (put the system into sleep using a platform-provided method)
+'shutdown' (shut the system down)
+'reboot' (reboot the system)
+'suspend' (trigger a Suspend-to-RAM transition)
+'test_resume' (resume-after-hibernation test mode)
+
+The currently selected option is printed in square brackets.
+
+The 'platform' option is only available if the platform provides a special
+mechanism to put the system to sleep after creating a hibernation image (ACPI
+does that, for example).  The 'suspend' option is available if Suspend-to-RAM
+is supported.  Refer to Documentation/power/basic_pm_debugging.txt for the
+description of the 'test_resume' option.
+
+To select an option, write the string representing it to /sys/power/disk.
+
+/sys/power/image_size controls the size of hibernation images.
+
+It can be written a string representing a non-negative integer that will be
+used as a best-effort upper limit of the image size, in bytes.  The hibernation
+core will do its best to ensure that the image size will not exceed that number.
+However, if that turns out to be impossible to achieve, a hibernation image will
+still be created and its size will be as small as possible.  In particular,
+writing '0' to this file will enforce hibernation images to be as small as
+possible.
+
+Reading from this file returns the current image size limit, which is set to
+around 2/5 of available RAM by default.
+
+/sys/power/pm_trace controls the PM trace mechanism saving the last suspend
+or resume event point in the RTC across reboots.
+
+It helps to debug hard lockups or reboots due to device driver failures that
+occur during system suspend or resume (which is more common) more effectively.
+
+If /sys/power/pm_trace contains '1', the fingerprint of each suspend/resume
+event point in turn will be stored in the RTC memory (overwriting the actual
+RTC information), so it will survive a system crash if one occurs right after
+storing it and it can be used later to identify the driver that caused the crash
+to happen (see Documentation/power/s2ram.txt for more information).
+
+Initially it contains '0' which may be changed to '1' by writing a string
+representing a nonzero integer into it.

Documentation/sphinx-static/theme_overrides.css

@@ -42,11 +42,12 @@
     caption a.headerlink { opacity: 0; }
     caption a.headerlink:hover { opacity: 1; }
 
-    /* inline literal: drop the borderbox and red color */
+    /* inline literal: drop the borderbox, padding and red color */
 
     code, .rst-content tt, .rst-content code {
         color: inherit;
         border: none;
+        padding: unset;
         background: inherit;
         font-size: 85%;
     }

MAINTAINERS

@@ -4525,6 +4525,12 @@ L:	linux-edac@vger.kernel.org
 S:	Maintained
 F:	drivers/edac/sb_edac.c
 
+EDAC-SKYLAKE
+M:	Tony Luck <tony.luck@intel.com>
+L:	linux-edac@vger.kernel.org
+S:	Maintained
+F:	drivers/edac/skx_edac.c
+
 EDAC-XGENE
 APPLIED MICRO (APM) X-GENE SOC EDAC
 M:     Loc Ho <lho@apm.com>

Makefile

@@ -1,7 +1,7 @@
 VERSION = 4
 PATCHLEVEL = 8
 SUBLEVEL = 0
-EXTRAVERSION = -rc2
+EXTRAVERSION = -rc3
 NAME = Psychotic Stoned Sheep
 
 # *DOCUMENTATION*

arch/arm/kernel/entry-armv.S

@@ -295,6 +295,7 @@ __und_svc_fault:
 	bl	__und_fault
 
 __und_svc_finish:
+	get_thread_info tsk
 	ldr	r5, [sp, #S_PSR]		@ Get SVC cpsr
 	svc_exit r5				@ return from exception
  UNWIND(.fnend		)

arch/arm/mach-imx/gpc.c

@@ -271,6 +271,12 @@ static int __init imx_gpc_init(struct device_node *node,
 	for (i = 0; i < IMR_NUM; i++)
 		writel_relaxed(~0, gpc_base + GPC_IMR1 + i * 4);
 
+	/*
+	 * Clear the OF_POPULATED flag set in of_irq_init so that
+	 * later the GPC power domain driver will not be skipped.
+	 */
+	of_node_clear_flag(node, OF_POPULATED);
+
 	return 0;
 }
 IRQCHIP_DECLARE(imx_gpc, "fsl,imx6q-gpc", imx_gpc_init);

arch/arm/mm/mmu.c

@@ -728,7 +728,8 @@ static void *__init late_alloc(unsigned long sz)
 {
 	void *ptr = (void *)__get_free_pages(PGALLOC_GFP, get_order(sz));
 
-	BUG_ON(!ptr);
+	if (!ptr || !pgtable_page_ctor(virt_to_page(ptr)))
+		BUG();
 	return ptr;
 }
 
@@ -1155,10 +1156,19 @@ void __init sanity_check_meminfo(void)
 {
 	phys_addr_t memblock_limit = 0;
 	int highmem = 0;
-	phys_addr_t vmalloc_limit = __pa(vmalloc_min - 1) + 1;
+	u64 vmalloc_limit;
 	struct memblock_region *reg;
 	bool should_use_highmem = false;
 
+	/*
+	 * Let's use our own (unoptimized) equivalent of __pa() that is
+	 * not affected by wrap-arounds when sizeof(phys_addr_t) == 4.
+	 * The result is used as the upper bound on physical memory address
+	 * and may itself be outside the valid range for which phys_addr_t
+	 * and therefore __pa() is defined.
+	 */
+	vmalloc_limit = (u64)(uintptr_t)vmalloc_min - PAGE_OFFSET + PHYS_OFFSET;
+
 	for_each_memblock(memory, reg) {
 		phys_addr_t block_start = reg->base;
 		phys_addr_t block_end = reg->base + reg->size;
@@ -1183,10 +1193,11 @@ void __init sanity_check_meminfo(void)
 			if (reg->size > size_limit) {
 				phys_addr_t overlap_size = reg->size - size_limit;
 
-				pr_notice("Truncating RAM at %pa-%pa to -%pa",
-					  &block_start, &block_end, &vmalloc_limit);
-				memblock_remove(vmalloc_limit, overlap_size);
+				pr_notice("Truncating RAM at %pa-%pa",
+					  &block_start, &block_end);
 				block_end = vmalloc_limit;
+				pr_cont(" to -%pa", &block_end);
+				memblock_remove(vmalloc_limit, overlap_size);
 				should_use_highmem = true;
 			}
 		}

arch/arm64/kernel/sleep.S

@@ -101,12 +101,20 @@ ENTRY(cpu_resume)
 	bl	el2_setup		// if in EL2 drop to EL1 cleanly
 	/* enable the MMU early - so we can access sleep_save_stash by va */
 	adr_l	lr, __enable_mmu	/* __cpu_setup will return here */
-	ldr	x27, =_cpu_resume	/* __enable_mmu will branch here */
+	adr_l	x27, _resume_switched	/* __enable_mmu will branch here */
 	adrp	x25, idmap_pg_dir
 	adrp	x26, swapper_pg_dir
 	b	__cpu_setup
 ENDPROC(cpu_resume)
 
+	.pushsection	".idmap.text", "ax"
+_resume_switched:
+	ldr	x8, =_cpu_resume
+	br	x8
+ENDPROC(_resume_switched)
+	.ltorg
+	.popsection
+
 ENTRY(_cpu_resume)
 	mrs	x1, mpidr_el1
 	adrp	x8, mpidr_hash

arch/arm64/mm/dump.c

@@ -242,7 +242,7 @@ static void note_page(struct pg_state *st, unsigned long addr, unsigned level,
 
 static void walk_pte(struct pg_state *st, pmd_t *pmd, unsigned long start)
 {
-	pte_t *pte = pte_offset_kernel(pmd, 0);
+	pte_t *pte = pte_offset_kernel(pmd, 0UL);
 	unsigned long addr;
 	unsigned i;
 
@@ -254,7 +254,7 @@ static void walk_pte(struct pg_state *st, pmd_t *pmd, unsigned long start)
 
 static void walk_pmd(struct pg_state *st, pud_t *pud, unsigned long start)
 {
-	pmd_t *pmd = pmd_offset(pud, 0);
+	pmd_t *pmd = pmd_offset(pud, 0UL);
 	unsigned long addr;
 	unsigned i;
 
@@ -271,7 +271,7 @@ static void walk_pmd(struct pg_state *st, pud_t *pud, unsigned long start)
 
 static void walk_pud(struct pg_state *st, pgd_t *pgd, unsigned long start)
 {
-	pud_t *pud = pud_offset(pgd, 0);
+	pud_t *pud = pud_offset(pgd, 0UL);
 	unsigned long addr;
 	unsigned i;
 

arch/arm64/mm/numa.c

@@ -23,6 +23,8 @@
 #include <linux/module.h>
 #include <linux/of.h>
 
+#include <asm/acpi.h>
+
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
 nodemask_t numa_nodes_parsed __initdata;

arch/parisc/include/uapi/asm/errno.h

@@ -97,10 +97,10 @@
 #define	ENOTCONN	235	/* Transport endpoint is not connected */
 #define	ESHUTDOWN	236	/* Cannot send after transport endpoint shutdown */
 #define	ETOOMANYREFS	237	/* Too many references: cannot splice */
-#define EREFUSED	ECONNREFUSED	/* for HP's NFS apparently */
 #define	ETIMEDOUT	238	/* Connection timed out */
 #define	ECONNREFUSED	239	/* Connection refused */
-#define EREMOTERELEASE	240	/* Remote peer released connection */
+#define	EREFUSED	ECONNREFUSED	/* for HP's NFS apparently */
+#define	EREMOTERELEASE	240	/* Remote peer released connection */
 #define	EHOSTDOWN	241	/* Host is down */
 #define	EHOSTUNREACH	242	/* No route to host */
 

arch/parisc/kernel/processor.c

@@ -51,8 +51,6 @@ EXPORT_SYMBOL(_parisc_requires_coherency);
 
 DEFINE_PER_CPU(struct cpuinfo_parisc, cpu_data);
 
-extern int update_cr16_clocksource(void);	/* from time.c */
-
 /*
 **  	PARISC CPU driver - claim "device" and initialize CPU data structures.
 **
@@ -228,12 +226,6 @@ static int processor_probe(struct parisc_device *dev)
 	}
 #endif
 
-	/* If we've registered more than one cpu,
-	 * we'll use the jiffies clocksource since cr16
-	 * is not synchronized between CPUs.
-	 */
-	update_cr16_clocksource();
-
 	return 0;
 }
 

arch/parisc/kernel/time.c

@@ -221,18 +221,6 @@ static struct clocksource clocksource_cr16 = {
 	.flags			= CLOCK_SOURCE_IS_CONTINUOUS,
 };
 
-int update_cr16_clocksource(void)
-{
-	/* since the cr16 cycle counters are not synchronized across CPUs,
-	   we'll check if we should switch to a safe clocksource: */
-	if (clocksource_cr16.rating != 0 && num_online_cpus() > 1) {
-		clocksource_change_rating(&clocksource_cr16, 0);
-		return 1;
-	}
-
-	return 0;
-}
-
 void __init start_cpu_itimer(void)
 {
 	unsigned int cpu = smp_processor_id();

arch/s390/include/asm/facilities_src.h

@@ -55,4 +55,28 @@ static struct facility_def facility_defs[] = {
 			-1 /* END */
 		}
 	},
+	{
+		.name = "FACILITIES_KVM",
+		.bits = (int[]){
+			0,  /* N3 instructions */
+			1,  /* z/Arch mode installed */
+			2,  /* z/Arch mode active */
+			3,  /* DAT-enhancement */
+			4,  /* idte segment table */
+			5,  /* idte region table */
+			6,  /* ASN-and-LX reuse */
+			7,  /* stfle */
+			8,  /* enhanced-DAT 1 */
+			9,  /* sense-running-status */
+			10, /* conditional sske */
+			13, /* ipte-range */
+			14, /* nonquiescing key-setting */
+			73, /* transactional execution */
+			75, /* access-exception-fetch/store indication */
+			76, /* msa extension 3 */
+			77, /* msa extension 4 */
+			78, /* enhanced-DAT 2 */
+			-1  /* END */
+		}
+	},
 };

arch/s390/include/asm/kvm_host.h

@@ -28,7 +28,7 @@
 
 #define KVM_S390_BSCA_CPU_SLOTS 64
 #define KVM_S390_ESCA_CPU_SLOTS 248
-#define KVM_MAX_VCPUS KVM_S390_ESCA_CPU_SLOTS
+#define KVM_MAX_VCPUS 255
 #define KVM_USER_MEM_SLOTS 32
 
 /*

arch/s390/kernel/asm-offsets.c

@@ -125,6 +125,7 @@ int main(void)
 	OFFSET(__LC_STFL_FAC_LIST, lowcore, stfl_fac_list);
 	OFFSET(__LC_STFLE_FAC_LIST, lowcore, stfle_fac_list);
 	OFFSET(__LC_MCCK_CODE, lowcore, mcck_interruption_code);
+	OFFSET(__LC_EXT_DAMAGE_CODE, lowcore, external_damage_code);
 	OFFSET(__LC_MCCK_FAIL_STOR_ADDR, lowcore, failing_storage_address);
 	OFFSET(__LC_LAST_BREAK, lowcore, breaking_event_addr);
 	OFFSET(__LC_RST_OLD_PSW, lowcore, restart_old_psw);

arch/s390/kvm/gaccess.c

@@ -495,6 +495,18 @@ static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva,
 	tec = (struct trans_exc_code_bits *)&pgm->trans_exc_code;
 
 	switch (code) {
+	case PGM_PROTECTION:
+		switch (prot) {
+		case PROT_TYPE_ALC:
+			tec->b60 = 1;
+			/* FALL THROUGH */
+		case PROT_TYPE_DAT:
+			tec->b61 = 1;
+			break;
+		default: /* LA and KEYC set b61 to 0, other params undefined */
+			return code;
+		}
+		/* FALL THROUGH */
 	case PGM_ASCE_TYPE:
 	case PGM_PAGE_TRANSLATION:
 	case PGM_REGION_FIRST_TRANS:
@@ -504,8 +516,7 @@ static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva,
 		/*
 		 * op_access_id only applies to MOVE_PAGE -> set bit 61
 		 * exc_access_id has to be set to 0 for some instructions. Both
-		 * cases have to be handled by the caller. We can always store
-		 * exc_access_id, as it is undefined for non-ar cases.
+		 * cases have to be handled by the caller.
 		 */
 		tec->addr = gva >> PAGE_SHIFT;
 		tec->fsi = mode == GACC_STORE ? FSI_STORE : FSI_FETCH;
@@ -516,25 +527,13 @@ static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva,
 	case PGM_ASTE_VALIDITY:
 	case PGM_ASTE_SEQUENCE:
 	case PGM_EXTENDED_AUTHORITY:
+		/*
+		 * We can always store exc_access_id, as it is
+		 * undefined for non-ar cases. It is undefined for
+		 * most DAT protection exceptions.
+		 */
 		pgm->exc_access_id = ar;
 		break;
-	case PGM_PROTECTION:
-		switch (prot) {
-		case PROT_TYPE_ALC:
-			tec->b60 = 1;
-			/* FALL THROUGH */
-		case PROT_TYPE_DAT:
-			tec->b61 = 1;
-			tec->addr = gva >> PAGE_SHIFT;
-			tec->fsi = mode == GACC_STORE ? FSI_STORE : FSI_FETCH;
-			tec->as = psw_bits(vcpu->arch.sie_block->gpsw).as;
-			/* exc_access_id is undefined for most cases */
-			pgm->exc_access_id = ar;
-			break;
-		default: /* LA and KEYC set b61 to 0, other params undefined */
-			break;
-		}
-		break;
 	}
 	return code;
 }

arch/s390/kvm/guestdbg.c

@@ -206,7 +206,7 @@ static int __import_wp_info(struct kvm_vcpu *vcpu,
 int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu,
 			    struct kvm_guest_debug *dbg)
 {
-	int ret = 0, nr_wp = 0, nr_bp = 0, i, size;
+	int ret = 0, nr_wp = 0, nr_bp = 0, i;
 	struct kvm_hw_breakpoint *bp_data = NULL;
 	struct kvm_hw_wp_info_arch *wp_info = NULL;
 	struct kvm_hw_bp_info_arch *bp_info = NULL;
@@ -216,17 +216,10 @@ int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu,
 	else if (dbg->arch.nr_hw_bp > MAX_BP_COUNT)
 		return -EINVAL;
 
-	size = dbg->arch.nr_hw_bp * sizeof(struct kvm_hw_breakpoint);
-	bp_data = kmalloc(size, GFP_KERNEL);
-	if (!bp_data) {
-		ret = -ENOMEM;
-		goto error;
-	}
-
-	if (copy_from_user(bp_data, dbg->arch.hw_bp, size)) {
-		ret = -EFAULT;
-		goto error;
-	}
+	bp_data = memdup_user(dbg->arch.hw_bp,
+			      sizeof(*bp_data) * dbg->arch.nr_hw_bp);
+	if (IS_ERR(bp_data))
+		return PTR_ERR(bp_data);
 
 	for (i = 0; i < dbg->arch.nr_hw_bp; i++) {
 		switch (bp_data[i].type) {
@@ -241,17 +234,19 @@ int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu,
 		}
 	}
 
-	size = nr_wp * sizeof(struct kvm_hw_wp_info_arch);
-	if (size > 0) {
-		wp_info = kmalloc(size, GFP_KERNEL);
+	if (nr_wp > 0) {
+		wp_info = kmalloc_array(nr_wp,
+					sizeof(*wp_info),
+					GFP_KERNEL);
 		if (!wp_info) {
 			ret = -ENOMEM;
 			goto error;
 		}
 	}
-	size = nr_bp * sizeof(struct kvm_hw_bp_info_arch);
-	if (size > 0) {
-		bp_info = kmalloc(size, GFP_KERNEL);
+	if (nr_bp > 0) {
+		bp_info = kmalloc_array(nr_bp,
+					sizeof(*bp_info),
+					GFP_KERNEL);
 		if (!bp_info) {
 			ret = -ENOMEM;
 			goto error;
@@ -382,14 +377,20 @@ void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu)
 	vcpu->guest_debug &= ~KVM_GUESTDBG_EXIT_PENDING;
 }
 
+#define PER_CODE_MASK		(PER_EVENT_MASK >> 24)
+#define PER_CODE_BRANCH		(PER_EVENT_BRANCH >> 24)
+#define PER_CODE_IFETCH		(PER_EVENT_IFETCH >> 24)
+#define PER_CODE_STORE		(PER_EVENT_STORE >> 24)
+#define PER_CODE_STORE_REAL	(PER_EVENT_STORE_REAL >> 24)
+
 #define per_bp_event(code) \
-			(code & (PER_EVENT_IFETCH | PER_EVENT_BRANCH))
+			(code & (PER_CODE_IFETCH | PER_CODE_BRANCH))
 #define per_write_wp_event(code) \
-			(code & (PER_EVENT_STORE | PER_EVENT_STORE_REAL))
+			(code & (PER_CODE_STORE | PER_CODE_STORE_REAL))
 
 static int debug_exit_required(struct kvm_vcpu *vcpu)
 {
-	u32 perc = (vcpu->arch.sie_block->perc << 24);
+	u8 perc = vcpu->arch.sie_block->perc;
 	struct kvm_debug_exit_arch *debug_exit = &vcpu->run->debug.arch;
 	struct kvm_hw_wp_info_arch *wp_info = NULL;
 	struct kvm_hw_bp_info_arch *bp_info = NULL;
@@ -444,7 +445,7 @@ int kvm_s390_handle_per_ifetch_icpt(struct kvm_vcpu *vcpu)
 	const u8 ilen = kvm_s390_get_ilen(vcpu);
 	struct kvm_s390_pgm_info pgm_info = {
 		.code = PGM_PER,
-		.per_code = PER_EVENT_IFETCH >> 24,
+		.per_code = PER_CODE_IFETCH,
 		.per_address = __rewind_psw(vcpu->arch.sie_block->gpsw, ilen),
 	};
 
@@ -458,33 +459,33 @@ int kvm_s390_handle_per_ifetch_icpt(struct kvm_vcpu *vcpu)
 
 static void filter_guest_per_event(struct kvm_vcpu *vcpu)
 {
-	u32 perc = vcpu->arch.sie_block->perc << 24;
+	const u8 perc = vcpu->arch.sie_block->perc;
 	u64 peraddr = vcpu->arch.sie_block->peraddr;
 	u64 addr = vcpu->arch.sie_block->gpsw.addr;
 	u64 cr9 = vcpu->arch.sie_block->gcr[9];
 	u64 cr10 = vcpu->arch.sie_block->gcr[10];
 	u64 cr11 = vcpu->arch.sie_block->gcr[11];
 	/* filter all events, demanded by the guest */
-	u32 guest_perc = perc & cr9 & PER_EVENT_MASK;
+	u8 guest_perc = perc & (cr9 >> 24) & PER_CODE_MASK;
 
 	if (!guest_per_enabled(vcpu))
 		guest_perc = 0;
 
 	/* filter "successful-branching" events */
-	if (guest_perc & PER_EVENT_BRANCH &&
+	if (guest_perc & PER_CODE_BRANCH &&
 	    cr9 & PER_CONTROL_BRANCH_ADDRESS &&
 	    !in_addr_range(addr, cr10, cr11))
-		guest_perc &= ~PER_EVENT_BRANCH;
+		guest_perc &= ~PER_CODE_BRANCH;
 
 	/* filter "instruction-fetching" events */
-	if (guest_perc & PER_EVENT_IFETCH &&
+	if (guest_perc & PER_CODE_IFETCH &&
 	    !in_addr_range(peraddr, cr10, cr11))
-		guest_perc &= ~PER_EVENT_IFETCH;
+		guest_perc &= ~PER_CODE_IFETCH;
 
 	/* All other PER events will be given to the guest */
 	/* TODO: Check altered address/address space */
 
-	vcpu->arch.sie_block->perc = guest_perc >> 24;
+	vcpu->arch.sie_block->perc = guest_perc;
 
 	if (!guest_perc)
 		vcpu->arch.sie_block->iprcc &= ~PGM_PER;

arch/s390/kvm/intercept.c

@@ -29,6 +29,7 @@ static const intercept_handler_t instruction_handlers[256] = {
 	[0x01] = kvm_s390_handle_01,
 	[0x82] = kvm_s390_handle_lpsw,
 	[0x83] = kvm_s390_handle_diag,
+	[0xaa] = kvm_s390_handle_aa,
 	[0xae] = kvm_s390_handle_sigp,
 	[0xb2] = kvm_s390_handle_b2,
 	[0xb6] = kvm_s390_handle_stctl,

arch/s390/kvm/interrupt.c

@@ -24,6 +24,8 @@
 #include <asm/sclp.h>
 #include <asm/isc.h>
 #include <asm/gmap.h>
+#include <asm/switch_to.h>
+#include <asm/nmi.h>
 #include "kvm-s390.h"
 #include "gaccess.h"
 #include "trace-s390.h"
@@ -40,6 +42,7 @@ static int sca_ext_call_pending(struct kvm_vcpu *vcpu, int *src_id)
 	if (!(atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_ECALL_PEND))
 		return 0;
 
+	BUG_ON(!kvm_s390_use_sca_entries());
 	read_lock(&vcpu->kvm->arch.sca_lock);
 	if (vcpu->kvm->arch.use_esca) {
 		struct esca_block *sca = vcpu->kvm->arch.sca;
@@ -68,6 +71,7 @@ static int sca_inject_ext_call(struct kvm_vcpu *vcpu, int src_id)
 {
 	int expect, rc;
 
+	BUG_ON(!kvm_s390_use_sca_entries());
 	read_lock(&vcpu->kvm->arch.sca_lock);
 	if (vcpu->kvm->arch.use_esca) {
 		struct esca_block *sca = vcpu->kvm->arch.sca;
@@ -109,6 +113,8 @@ static void sca_clear_ext_call(struct kvm_vcpu *vcpu)
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
 	int rc, expect;
 
+	if (!kvm_s390_use_sca_entries())
+		return;
 	atomic_andnot(CPUSTAT_ECALL_PEND, li->cpuflags);
 	read_lock(&vcpu->kvm->arch.sca_lock);
 	if (vcpu->kvm->arch.use_esca) {
@@ -400,12 +406,78 @@ static int __must_check __deliver_pfault_init(struct kvm_vcpu *vcpu)
 	return rc ? -EFAULT : 0;
 }
 
+static int __write_machine_check(struct kvm_vcpu *vcpu,
+				 struct kvm_s390_mchk_info *mchk)
+{
+	unsigned long ext_sa_addr;
+	freg_t fprs[NUM_FPRS];
+	union mci mci;
+	int rc;
+
+	mci.val = mchk->mcic;
+	/* take care of lazy register loading via vcpu load/put */
+	save_fpu_regs();
+	save_access_regs(vcpu->run->s.regs.acrs);
+
+	/* Extended save area */
+	rc = read_guest_lc(vcpu, __LC_VX_SAVE_AREA_ADDR, &ext_sa_addr,
+			    sizeof(unsigned long));
+	/* Only bits 0-53 are used for address formation */
+	ext_sa_addr &= ~0x3ffUL;
+	if (!rc && mci.vr && ext_sa_addr && test_kvm_facility(vcpu->kvm, 129)) {
+		if (write_guest_abs(vcpu, ext_sa_addr, vcpu->run->s.regs.vrs,
+				    512))
+			mci.vr = 0;
+	} else {
+		mci.vr = 0;
+	}
+
+	/* General interruption information */
+	rc |= put_guest_lc(vcpu, 1, (u8 __user *) __LC_AR_MODE_ID);
+	rc |= write_guest_lc(vcpu, __LC_MCK_OLD_PSW,
+			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
+	rc |= read_guest_lc(vcpu, __LC_MCK_NEW_PSW,
+			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
+	rc |= put_guest_lc(vcpu, mci.val, (u64 __user *) __LC_MCCK_CODE);
+
+	/* Register-save areas */
+	if (MACHINE_HAS_VX) {
+		convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
+		rc |= write_guest_lc(vcpu, __LC_FPREGS_SAVE_AREA, fprs, 128);
+	} else {
+		rc |= write_guest_lc(vcpu, __LC_FPREGS_SAVE_AREA,
+				     vcpu->run->s.regs.fprs, 128);
+	}
+	rc |= write_guest_lc(vcpu, __LC_GPREGS_SAVE_AREA,
+			     vcpu->run->s.regs.gprs, 128);
+	rc |= put_guest_lc(vcpu, current->thread.fpu.fpc,
+			   (u32 __user *) __LC_FP_CREG_SAVE_AREA);
+	rc |= put_guest_lc(vcpu, vcpu->arch.sie_block->todpr,
+			   (u32 __user *) __LC_TOD_PROGREG_SAVE_AREA);
+	rc |= put_guest_lc(vcpu, kvm_s390_get_cpu_timer(vcpu),
+			   (u64 __user *) __LC_CPU_TIMER_SAVE_AREA);
+	rc |= put_guest_lc(vcpu, vcpu->arch.sie_block->ckc >> 8,
+			   (u64 __user *) __LC_CLOCK_COMP_SAVE_AREA);
+	rc |= write_guest_lc(vcpu, __LC_AREGS_SAVE_AREA,
+			     &vcpu->run->s.regs.acrs, 64);
+	rc |= write_guest_lc(vcpu, __LC_CREGS_SAVE_AREA,
+			     &vcpu->arch.sie_block->gcr, 128);
+
+	/* Extended interruption information */
+	rc |= put_guest_lc(vcpu, mchk->ext_damage_code,
+			   (u32 __user *) __LC_EXT_DAMAGE_CODE);
+	rc |= put_guest_lc(vcpu, mchk->failing_storage_address,
+			   (u64 __user *) __LC_MCCK_FAIL_STOR_ADDR);
+	rc |= write_guest_lc(vcpu, __LC_PSW_SAVE_AREA, &mchk->fixed_logout,
+			     sizeof(mchk->fixed_logout));
+	return rc ? -EFAULT : 0;
+}
+
 static int __must_check __deliver_machine_check(struct kvm_vcpu *vcpu)
 {
 	struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int;
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
 	struct kvm_s390_mchk_info mchk = {};
-	unsigned long adtl_status_addr;
 	int deliver = 0;
 	int rc = 0;
 
@@ -446,29 +518,9 @@ static int __must_check __deliver_machine_check(struct kvm_vcpu *vcpu)
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
 						 KVM_S390_MCHK,
 						 mchk.cr14, mchk.mcic);
-
-		rc  = kvm_s390_vcpu_store_status(vcpu,
-						 KVM_S390_STORE_STATUS_PREFIXED);
-		rc |= read_guest_lc(vcpu, __LC_VX_SAVE_AREA_ADDR,
-				    &adtl_status_addr,
-				    sizeof(unsigned long));
-		rc |= kvm_s390_vcpu_store_adtl_status(vcpu,
-						      adtl_status_addr);
-		rc |= put_guest_lc(vcpu, mchk.mcic,
-				   (u64 __user *) __LC_MCCK_CODE);
-		rc |= put_guest_lc(vcpu, mchk.failing_storage_address,
-				   (u64 __user *) __LC_MCCK_FAIL_STOR_ADDR);
-		rc |= write_guest_lc(vcpu, __LC_PSW_SAVE_AREA,
-				     &mchk.fixed_logout,
-				     sizeof(mchk.fixed_logout));
-		rc |= write_guest_lc(vcpu, __LC_MCK_OLD_PSW,
-				     &vcpu->arch.sie_block->gpsw,
-				     sizeof(psw_t));
-		rc |= read_guest_lc(vcpu, __LC_MCK_NEW_PSW,
-				    &vcpu->arch.sie_block->gpsw,
-				    sizeof(psw_t));
+		rc = __write_machine_check(vcpu, &mchk);
 	}
-	return rc ? -EFAULT : 0;
+	return rc;
 }
 
 static int __must_check __deliver_restart(struct kvm_vcpu *vcpu)

arch/s390/kvm/kvm-s390.c

@@ -132,10 +132,7 @@ module_param(nested, int, S_IRUGO);
 MODULE_PARM_DESC(nested, "Nested virtualization support");
 
 /* upper facilities limit for kvm */
-unsigned long kvm_s390_fac_list_mask[16] = {
-	0xffe6000000000000UL,
-	0x005e000000000000UL,
-};
+unsigned long kvm_s390_fac_list_mask[16] = { FACILITIES_KVM };
 
 unsigned long kvm_s390_fac_list_mask_size(void)
 {
@@ -376,7 +373,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_NR_VCPUS:
 	case KVM_CAP_MAX_VCPUS:
 		r = KVM_S390_BSCA_CPU_SLOTS;
-		if (sclp.has_esca && sclp.has_64bscao)
+		if (!kvm_s390_use_sca_entries())
+			r = KVM_MAX_VCPUS;
+		else if (sclp.has_esca && sclp.has_64bscao)
 			r = KVM_S390_ESCA_CPU_SLOTS;
 		break;
 	case KVM_CAP_NR_MEMSLOTS:
@@ -1553,6 +1552,8 @@ static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
 
 static void sca_del_vcpu(struct kvm_vcpu *vcpu)
 {
+	if (!kvm_s390_use_sca_entries())
+		return;
 	read_lock(&vcpu->kvm->arch.sca_lock);
 	if (vcpu->kvm->arch.use_esca) {
 		struct esca_block *sca = vcpu->kvm->arch.sca;
@@ -1570,6 +1571,13 @@ static void sca_del_vcpu(struct kvm_vcpu *vcpu)
 
 static void sca_add_vcpu(struct kvm_vcpu *vcpu)
 {
+	if (!kvm_s390_use_sca_entries()) {
+		struct bsca_block *sca = vcpu->kvm->arch.sca;
+
+		/* we still need the basic sca for the ipte control */
+		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
+		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
+	}
 	read_lock(&vcpu->kvm->arch.sca_lock);
 	if (vcpu->kvm->arch.use_esca) {
 		struct esca_block *sca = vcpu->kvm->arch.sca;
@@ -1650,6 +1658,11 @@ static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
 {
 	int rc;
 
+	if (!kvm_s390_use_sca_entries()) {
+		if (id < KVM_MAX_VCPUS)
+			return true;
+		return false;
+	}
 	if (id < KVM_S390_BSCA_CPU_SLOTS)
 		return true;
 	if (!sclp.has_esca || !sclp.has_64bscao)
@@ -1938,8 +1951,6 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 		vcpu->arch.sie_block->eca |= 1;
 	if (sclp.has_sigpif)
 		vcpu->arch.sie_block->eca |= 0x10000000U;
-	if (test_kvm_facility(vcpu->kvm, 64))
-		vcpu->arch.sie_block->ecb3 |= 0x01;
 	if (test_kvm_facility(vcpu->kvm, 129)) {
 		vcpu->arch.sie_block->eca |= 0x00020000;
 		vcpu->arch.sie_block->ecd |= 0x20000000;
@@ -2694,6 +2705,19 @@ static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
 			kvm_clear_async_pf_completion_queue(vcpu);
 	}
+	/*
+	 * If userspace sets the riccb (e.g. after migration) to a valid state,
+	 * we should enable RI here instead of doing the lazy enablement.
+	 */
+	if ((kvm_run->kvm_dirty_regs & KVM_SYNC_RICCB) &&
+	    test_kvm_facility(vcpu->kvm, 64)) {
+		struct runtime_instr_cb *riccb =
+			(struct runtime_instr_cb *) &kvm_run->s.regs.riccb;
+
+		if (riccb->valid)
+			vcpu->arch.sie_block->ecb3 |= 0x01;
+	}
+
 	kvm_run->kvm_dirty_regs = 0;
 }
 
@@ -2837,38 +2861,6 @@ int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr)
 	return kvm_s390_store_status_unloaded(vcpu, addr);
 }
 
-/*
- * store additional status at address
- */
-int kvm_s390_store_adtl_status_unloaded(struct kvm_vcpu *vcpu,
-					unsigned long gpa)
-{
-	/* Only bits 0-53 are used for address formation */
-	if (!(gpa & ~0x3ff))
-		return 0;
-
-	return write_guest_abs(vcpu, gpa & ~0x3ff,
-			       (void *)&vcpu->run->s.regs.vrs, 512);
-}
-
-int kvm_s390_vcpu_store_adtl_status(struct kvm_vcpu *vcpu, unsigned long addr)
-{
-	if (!test_kvm_facility(vcpu->kvm, 129))
-		return 0;
-
-	/*
-	 * The guest VXRS are in the host VXRs due to the lazy
-	 * copying in vcpu load/put. We can simply call save_fpu_regs()
-	 * to save the current register state because we are in the
-	 * middle of a load/put cycle.
-	 *
-	 * Let's update our copies before we save it into the save area.
-	 */
-	save_fpu_regs();
-
-	return kvm_s390_store_adtl_status_unloaded(vcpu, addr);
-}
-
 static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
 {
 	kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);

arch/s390/kvm/kvm-s390.h

@@ -20,6 +20,7 @@
 #include <linux/kvm_host.h>
 #include <asm/facility.h>
 #include <asm/processor.h>
+#include <asm/sclp.h>
 
 typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu);
 
@@ -245,6 +246,7 @@ static inline void kvm_s390_retry_instr(struct kvm_vcpu *vcpu)
 
 /* implemented in priv.c */
 int is_valid_psw(psw_t *psw);
+int kvm_s390_handle_aa(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_b2(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_e5(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_01(struct kvm_vcpu *vcpu);
@@ -273,10 +275,7 @@ int handle_sthyi(struct kvm_vcpu *vcpu);
 void kvm_s390_set_tod_clock(struct kvm *kvm, u64 tod);
 long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable);
 int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long addr);
-int kvm_s390_store_adtl_status_unloaded(struct kvm_vcpu *vcpu,
-					unsigned long addr);
 int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr);
-int kvm_s390_vcpu_store_adtl_status(struct kvm_vcpu *vcpu, unsigned long addr);
 void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu);
 void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu);
 void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu);
@@ -389,4 +388,13 @@ static inline union ipte_control *kvm_s390_get_ipte_control(struct kvm *kvm)
 
 	return &sca->ipte_control;
 }
+static inline int kvm_s390_use_sca_entries(void)
+{
+	/*
+	 * Without SIGP interpretation, only SRS interpretation (if available)
+	 * might use the entries. By not setting the entries and keeping them
+	 * invalid, hardware will not access them but intercept.
+	 */
+	return sclp.has_sigpif;
+}
 #endif

arch/s390/kvm/priv.c

@@ -32,6 +32,24 @@
 #include "kvm-s390.h"
 #include "trace.h"
 
+static int handle_ri(struct kvm_vcpu *vcpu)
+{
+	if (test_kvm_facility(vcpu->kvm, 64)) {
+		vcpu->arch.sie_block->ecb3 |= 0x01;
+		kvm_s390_retry_instr(vcpu);
+		return 0;
+	} else
+		return kvm_s390_inject_program_int(vcpu, PGM_OPERATION);
+}
+
+int kvm_s390_handle_aa(struct kvm_vcpu *vcpu)
+{
+	if ((vcpu->arch.sie_block->ipa & 0xf) <= 4)
+		return handle_ri(vcpu);
+	else
+		return -EOPNOTSUPP;
+}
+
 /* Handle SCK (SET CLOCK) interception */
 static int handle_set_clock(struct kvm_vcpu *vcpu)
 {
@@ -1093,6 +1111,9 @@ static int handle_stctg(struct kvm_vcpu *vcpu)
 static const intercept_handler_t eb_handlers[256] = {
 	[0x2f] = handle_lctlg,
 	[0x25] = handle_stctg,
+	[0x60] = handle_ri,
+	[0x61] = handle_ri,
+	[0x62] = handle_ri,
 };
 
 int kvm_s390_handle_eb(struct kvm_vcpu *vcpu)

arch/x86/kernel/cpu/microcode/amd.c

@@ -355,6 +355,7 @@ void load_ucode_amd_ap(void)
 	unsigned int cpu = smp_processor_id();
 	struct equiv_cpu_entry *eq;
 	struct microcode_amd *mc;
+	u8 *cont = container;
 	u32 rev, eax;
 	u16 eq_id;
 
@@ -371,8 +372,11 @@ void load_ucode_amd_ap(void)
 	if (check_current_patch_level(&rev, false))
 		return;
 
+	/* Add CONFIG_RANDOMIZE_MEMORY offset. */
+	cont += PAGE_OFFSET - __PAGE_OFFSET_BASE;
+
 	eax = cpuid_eax(0x00000001);
-	eq  = (struct equiv_cpu_entry *)(container + CONTAINER_HDR_SZ);
+	eq  = (struct equiv_cpu_entry *)(cont + CONTAINER_HDR_SZ);
 
 	eq_id = find_equiv_id(eq, eax);
 	if (!eq_id)
@@ -434,6 +438,9 @@ int __init save_microcode_in_initrd_amd(void)
 	else
 		container = cont_va;
 
+	/* Add CONFIG_RANDOMIZE_MEMORY offset. */
+	container += PAGE_OFFSET - __PAGE_OFFSET_BASE;
+
 	eax   = cpuid_eax(0x00000001);
 	eax   = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff);
 

arch/x86/kernel/smpboot.c

@@ -100,10 +100,11 @@ EXPORT_PER_CPU_SYMBOL(cpu_info);
 /* Logical package management. We might want to allocate that dynamically */
 static int *physical_to_logical_pkg __read_mostly;
 static unsigned long *physical_package_map __read_mostly;;
-static unsigned long *logical_package_map  __read_mostly;
 static unsigned int max_physical_pkg_id __read_mostly;
 unsigned int __max_logical_packages __read_mostly;
 EXPORT_SYMBOL(__max_logical_packages);
+static unsigned int logical_packages __read_mostly;
+static bool logical_packages_frozen __read_mostly;
 
 /* Maximum number of SMT threads on any online core */
 int __max_smt_threads __read_mostly;
@@ -277,14 +278,14 @@ int topology_update_package_map(unsigned int apicid, unsigned int cpu)
 	if (test_and_set_bit(pkg, physical_package_map))
 		goto found;
 
-	new = find_first_zero_bit(logical_package_map, __max_logical_packages);
-	if (new >= __max_logical_packages) {
+	if (logical_packages_frozen) {
 		physical_to_logical_pkg[pkg] = -1;
-		pr_warn("APIC(%x) Package %u exceeds logical package map\n",
+		pr_warn("APIC(%x) Package %u exceeds logical package max\n",
 			apicid, pkg);
 		return -ENOSPC;
 	}
-	set_bit(new, logical_package_map);
+
+	new = logical_packages++;
 	pr_info("APIC(%x) Converting physical %u to logical package %u\n",
 		apicid, pkg, new);
 	physical_to_logical_pkg[pkg] = new;
@@ -341,6 +342,7 @@ static void __init smp_init_package_map(void)
 	}
 
 	__max_logical_packages = DIV_ROUND_UP(total_cpus, ncpus);
+	logical_packages = 0;
 
 	/*
 	 * Possibly larger than what we need as the number of apic ids per
@@ -352,10 +354,6 @@ static void __init smp_init_package_map(void)
 	memset(physical_to_logical_pkg, 0xff, size);
 	size = BITS_TO_LONGS(max_physical_pkg_id) * sizeof(unsigned long);
 	physical_package_map = kzalloc(size, GFP_KERNEL);
-	size = BITS_TO_LONGS(__max_logical_packages) * sizeof(unsigned long);
-	logical_package_map = kzalloc(size, GFP_KERNEL);
-
-	pr_info("Max logical packages: %u\n", __max_logical_packages);
 
 	for_each_present_cpu(cpu) {
 		unsigned int apicid = apic->cpu_present_to_apicid(cpu);
@@ -369,6 +367,15 @@ static void __init smp_init_package_map(void)
 		set_cpu_possible(cpu, false);
 		set_cpu_present(cpu, false);
 	}
+
+	if (logical_packages > __max_logical_packages) {
+		pr_warn("Detected more packages (%u), then computed by BIOS data (%u).\n",
+			logical_packages, __max_logical_packages);
+		logical_packages_frozen = true;
+		__max_logical_packages  = logical_packages;
+	}
+
+	pr_info("Max logical packages: %u\n", __max_logical_packages);
 }
 
 void __init smp_store_boot_cpu_info(void)

arch/x86/power/hibernate_64.c

@@ -113,7 +113,7 @@ static int set_up_temporary_mappings(void)
 			return result;
 	}
 
-	temp_level4_pgt = (unsigned long)pgd - __PAGE_OFFSET;
+	temp_level4_pgt = __pa(pgd);
 	return 0;
 }
 

drivers/clocksource/bcm_kona_timer.c

@@ -66,10 +66,10 @@ static void kona_timer_disable_and_clear(void __iomem *base)
 
 }
 
-static void
+static int
 kona_timer_get_counter(void __iomem *timer_base, uint32_t *msw, uint32_t *lsw)
 {
-	int loop_limit = 4;
+	int loop_limit = 3;
 
 	/*
 	 * Read 64-bit free running counter
@@ -83,18 +83,19 @@ kona_timer_get_counter(void __iomem *timer_base, uint32_t *msw, uint32_t *lsw)
 	 *      if new hi-word is equal to previously read hi-word then stop.
 	 */
 
-	while (--loop_limit) {
+	do {
 		*msw = readl(timer_base + KONA_GPTIMER_STCHI_OFFSET);
 		*lsw = readl(timer_base + KONA_GPTIMER_STCLO_OFFSET);
 		if (*msw == readl(timer_base + KONA_GPTIMER_STCHI_OFFSET))
 			break;
-	}
+	} while (--loop_limit);
 	if (!loop_limit) {
 		pr_err("bcm_kona_timer: getting counter failed.\n");
 		pr_err(" Timer will be impacted\n");
+		return -ETIMEDOUT;
 	}
 
-	return;
+	return 0;
 }
 
 static int kona_timer_set_next_event(unsigned long clc,
@@ -112,8 +113,11 @@ static int kona_timer_set_next_event(unsigned long clc,
 
 	uint32_t lsw, msw;
 	uint32_t reg;
+	int ret;
 
-	kona_timer_get_counter(timers.tmr_regs, &msw, &lsw);
+	ret = kona_timer_get_counter(timers.tmr_regs, &msw, &lsw);
+	if (ret)
+		return ret;
 
 	/* Load the "next" event tick value */
 	writel(lsw + clc, timers.tmr_regs + KONA_GPTIMER_STCM0_OFFSET);

drivers/clocksource/mips-gic-timer.c

@@ -164,7 +164,7 @@ void __init gic_clocksource_init(unsigned int frequency)
 	gic_start_count();
 }
 
-static void __init gic_clocksource_of_init(struct device_node *node)
+static int __init gic_clocksource_of_init(struct device_node *node)
 {
 	struct clk *clk;
 	int ret;

drivers/clocksource/time-armada-370-xp.c

@@ -338,7 +338,6 @@ static int __init armada_xp_timer_init(struct device_node *np)
 	struct clk *clk = of_clk_get_by_name(np, "fixed");
 	int ret;
 
-	clk = of_clk_get(np, 0);
 	if (IS_ERR(clk)) {
 		pr_err("Failed to get clock");
 		return PTR_ERR(clk);

drivers/edac/Kconfig

@@ -251,6 +251,14 @@ config EDAC_SBRIDGE
 	  Support for error detection and correction the Intel
 	  Sandy Bridge, Ivy Bridge and Haswell Integrated Memory Controllers.
 
+config EDAC_SKX
+	tristate "Intel Skylake server Integrated MC"
+	depends on EDAC_MM_EDAC && PCI && X86_64 && X86_MCE_INTEL
+	depends on PCI_MMCONFIG
+	help
+	  Support for error detection and correction the Intel
+	  Skylake server Integrated Memory Controllers.
+
 config EDAC_MPC85XX
 	tristate "Freescale MPC83xx / MPC85xx"
 	depends on EDAC_MM_EDAC && FSL_SOC

drivers/edac/Makefile

@@ -31,6 +31,7 @@ obj-$(CONFIG_EDAC_I5400)		+= i5400_edac.o
 obj-$(CONFIG_EDAC_I7300)		+= i7300_edac.o
 obj-$(CONFIG_EDAC_I7CORE)		+= i7core_edac.o
 obj-$(CONFIG_EDAC_SBRIDGE)		+= sb_edac.o
+obj-$(CONFIG_EDAC_SKX)			+= skx_edac.o
 obj-$(CONFIG_EDAC_E7XXX)		+= e7xxx_edac.o
 obj-$(CONFIG_EDAC_E752X)		+= e752x_edac.o
 obj-$(CONFIG_EDAC_I82443BXGX)		+= i82443bxgx_edac.o

drivers/edac/skx_edac.c

@@ -0,0 +1,1121 @@
+/*
+ * EDAC driver for Intel(R) Xeon(R) Skylake processors
+ * Copyright (c) 2016, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/pci_ids.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/edac.h>
+#include <linux/mmzone.h>
+#include <linux/smp.h>
+#include <linux/bitmap.h>
+#include <linux/math64.h>
+#include <linux/mod_devicetable.h>
+#include <asm/cpu_device_id.h>
+#include <asm/processor.h>
+#include <asm/mce.h>
+
+#include "edac_core.h"
+
+#define SKX_REVISION    " Ver: 1.0 "
+
+/*
+ * Debug macros
+ */
+#define skx_printk(level, fmt, arg...)			\
+	edac_printk(level, "skx", fmt, ##arg)
+
+#define skx_mc_printk(mci, level, fmt, arg...)		\
+	edac_mc_chipset_printk(mci, level, "skx", fmt, ##arg)
+
+/*
+ * Get a bit field at register value <v>, from bit <lo> to bit <hi>
+ */
+#define GET_BITFIELD(v, lo, hi) \
+	(((v) & GENMASK_ULL((hi), (lo))) >> (lo))
+
+static LIST_HEAD(skx_edac_list);
+
+static u64 skx_tolm, skx_tohm;
+
+#define NUM_IMC			2	/* memory controllers per socket */
+#define NUM_CHANNELS		3	/* channels per memory controller */
+#define NUM_DIMMS		2	/* Max DIMMS per channel */
+
+#define	MASK26	0x3FFFFFF		/* Mask for 2^26 */
+#define MASK29	0x1FFFFFFF		/* Mask for 2^29 */
+
+/*
+ * Each cpu socket contains some pci devices that provide global
+ * information, and also some that are local to each of the two
+ * memory controllers on the die.
+ */
+struct skx_dev {
+	struct list_head	list;
+	u8			bus[4];
+	struct pci_dev	*sad_all;
+	struct pci_dev	*util_all;
+	u32	mcroute;
+	struct skx_imc {
+		struct mem_ctl_info *mci;
+		u8	mc;	/* system wide mc# */
+		u8	lmc;	/* socket relative mc# */
+		u8	src_id, node_id;
+		struct skx_channel {
+			struct pci_dev *cdev;
+			struct skx_dimm {
+				u8	close_pg;
+				u8	bank_xor_enable;
+				u8	fine_grain_bank;
+				u8	rowbits;
+				u8	colbits;
+			} dimms[NUM_DIMMS];
+		} chan[NUM_CHANNELS];
+	} imc[NUM_IMC];
+};
+static int skx_num_sockets;
+
+struct skx_pvt {
+	struct skx_imc	*imc;
+};
+
+struct decoded_addr {
+	struct skx_dev *dev;
+	u64	addr;
+	int	socket;
+	int	imc;
+	int	channel;
+	u64	chan_addr;
+	int	sktways;
+	int	chanways;
+	int	dimm;
+	int	rank;
+	int	channel_rank;
+	u64	rank_address;
+	int	row;
+	int	column;
+	int	bank_address;
+	int	bank_group;
+};
+
+static struct skx_dev *get_skx_dev(u8 bus, u8 idx)
+{
+	struct skx_dev *d;
+
+	list_for_each_entry(d, &skx_edac_list, list) {
+		if (d->bus[idx] == bus)
+			return d;
+	}
+
+	return NULL;
+}
+
+enum munittype {
+	CHAN0, CHAN1, CHAN2, SAD_ALL, UTIL_ALL, SAD
+};
+
+struct munit {
+	u16	did;
+	u16	devfn[NUM_IMC];
+	u8	busidx;
+	u8	per_socket;
+	enum munittype mtype;
+};
+
+/*
+ * List of PCI device ids that we need together with some device
+ * number and function numbers to tell which memory controller the
+ * device belongs to.
+ */
+static const struct munit skx_all_munits[] = {
+	{ 0x2054, { }, 1, 1, SAD_ALL },
+	{ 0x2055, { }, 1, 1, UTIL_ALL },
+	{ 0x2040, { PCI_DEVFN(10, 0), PCI_DEVFN(12, 0) }, 2, 2, CHAN0 },
+	{ 0x2044, { PCI_DEVFN(10, 4), PCI_DEVFN(12, 4) }, 2, 2, CHAN1 },
+	{ 0x2048, { PCI_DEVFN(11, 0), PCI_DEVFN(13, 0) }, 2, 2, CHAN2 },
+	{ 0x208e, { }, 1, 0, SAD },
+	{ }
+};
+
+/*
+ * We use the per-socket device 0x2016 to count how many sockets are present,
+ * and to detemine which PCI buses are associated with each socket. Allocate
+ * and build the full list of all the skx_dev structures that we need here.
+ */
+static int get_all_bus_mappings(void)
+{
+	struct pci_dev *pdev, *prev;
+	struct skx_dev *d;
+	u32 reg;
+	int ndev = 0;
+
+	prev = NULL;
+	for (;;) {
+		pdev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x2016, prev);
+		if (!pdev)
+			break;
+		ndev++;
+		d = kzalloc(sizeof(*d), GFP_KERNEL);
+		if (!d) {
+			pci_dev_put(pdev);
+			return -ENOMEM;
+		}
+		pci_read_config_dword(pdev, 0xCC, &reg);
+		d->bus[0] =  GET_BITFIELD(reg, 0, 7);
+		d->bus[1] =  GET_BITFIELD(reg, 8, 15);
+		d->bus[2] =  GET_BITFIELD(reg, 16, 23);
+		d->bus[3] =  GET_BITFIELD(reg, 24, 31);
+		edac_dbg(2, "busses: %x, %x, %x, %x\n",
+			 d->bus[0], d->bus[1], d->bus[2], d->bus[3]);
+		list_add_tail(&d->list, &skx_edac_list);
+		skx_num_sockets++;
+		prev = pdev;
+	}
+
+	return ndev;
+}
+
+static int get_all_munits(const struct munit *m)
+{
+	struct pci_dev *pdev, *prev;
+	struct skx_dev *d;
+	u32 reg;
+	int i = 0, ndev = 0;
+
+	prev = NULL;
+	for (;;) {
+		pdev = pci_get_device(PCI_VENDOR_ID_INTEL, m->did, prev);
+		if (!pdev)
+			break;
+		ndev++;
+		if (m->per_socket == NUM_IMC) {
+			for (i = 0; i < NUM_IMC; i++)
+				if (m->devfn[i] == pdev->devfn)
+					break;
+			if (i == NUM_IMC)
+				goto fail;
+		}
+		d = get_skx_dev(pdev->bus->number, m->busidx);
+		if (!d)
+			goto fail;
+
+		/* Be sure that the device is enabled */
+		if (unlikely(pci_enable_device(pdev) < 0)) {
+			skx_printk(KERN_ERR,
+				"Couldn't enable %04x:%04x\n", PCI_VENDOR_ID_INTEL, m->did);
+			goto fail;
+		}
+
+		switch (m->mtype) {
+		case CHAN0: case CHAN1: case CHAN2:
+			pci_dev_get(pdev);
+			d->imc[i].chan[m->mtype].cdev = pdev;
+			break;
+		case SAD_ALL:
+			pci_dev_get(pdev);
+			d->sad_all = pdev;
+			break;
+		case UTIL_ALL:
+			pci_dev_get(pdev);
+			d->util_all = pdev;
+			break;
+		case SAD:
+			/*
+			 * one of these devices per core, including cores
+			 * that don't exist on this SKU. Ignore any that
+			 * read a route table of zero, make sure all the
+			 * non-zero values match.
+			 */
+			pci_read_config_dword(pdev, 0xB4, &reg);
+			if (reg != 0) {
+				if (d->mcroute == 0)
+					d->mcroute = reg;
+				else if (d->mcroute != reg) {
+					skx_printk(KERN_ERR,
+						"mcroute mismatch\n");
+					goto fail;
+				}
+			}
+			ndev--;
+			break;
+		}
+
+		prev = pdev;
+	}
+
+	return ndev;
+fail:
+	pci_dev_put(pdev);
+	return -ENODEV;
+}
+
+const struct x86_cpu_id skx_cpuids[] = {
+	{ X86_VENDOR_INTEL, 6, 0x55, 0, 0 },	/* Skylake */
+	{ }
+};
+MODULE_DEVICE_TABLE(x86cpu, skx_cpuids);
+
+static u8 get_src_id(struct skx_dev *d)
+{
+	u32 reg;
+
+	pci_read_config_dword(d->util_all, 0xF0, &reg);
+
+	return GET_BITFIELD(reg, 12, 14);
+}
+
+static u8 skx_get_node_id(struct skx_dev *d)
+{
+	u32 reg;
+
+	pci_read_config_dword(d->util_all, 0xF4, &reg);
+
+	return GET_BITFIELD(reg, 0, 2);
+}
+
+static int get_dimm_attr(u32 reg, int lobit, int hibit, int add, int minval,
+			 int maxval, char *name)
+{
+	u32 val = GET_BITFIELD(reg, lobit, hibit);
+
+	if (val < minval || val > maxval) {
+		edac_dbg(2, "bad %s = %d (raw=%x)\n", name, val, reg);
+		return -EINVAL;
+	}
+	return val + add;
+}
+
+#define IS_DIMM_PRESENT(mtr)		GET_BITFIELD((mtr), 15, 15)
+
+#define numrank(reg) get_dimm_attr((reg), 12, 13, 0, 1, 2, "ranks")
+#define numrow(reg) get_dimm_attr((reg), 2, 4, 12, 1, 6, "rows")
+#define numcol(reg) get_dimm_attr((reg), 0, 1, 10, 0, 2, "cols")
+
+static int get_width(u32 mtr)
+{
+	switch (GET_BITFIELD(mtr, 8, 9)) {
+	case 0:
+		return DEV_X4;
+	case 1:
+		return DEV_X8;
+	case 2:
+		return DEV_X16;
+	}
+	return DEV_UNKNOWN;
+}
+
+static int skx_get_hi_lo(void)
+{
+	struct pci_dev *pdev;
+	u32 reg;
+
+	pdev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x2034, NULL);
+	if (!pdev) {
+		edac_dbg(0, "Can't get tolm/tohm\n");
+		return -ENODEV;
+	}
+
+	pci_read_config_dword(pdev, 0xD0, &reg);
+	skx_tolm = reg;
+	pci_read_config_dword(pdev, 0xD4, &reg);
+	skx_tohm = reg;
+	pci_read_config_dword(pdev, 0xD8, &reg);
+	skx_tohm |= (u64)reg << 32;
+
+	pci_dev_put(pdev);
+	edac_dbg(2, "tolm=%llx tohm=%llx\n", skx_tolm, skx_tohm);
+
+	return 0;
+}
+
+static int get_dimm_info(u32 mtr, u32 amap, struct dimm_info *dimm,
+			 struct skx_imc *imc, int chan, int dimmno)
+{
+	int  banks = 16, ranks, rows, cols, npages;
+	u64 size;
+
+	if (!IS_DIMM_PRESENT(mtr))
+		return 0;
+	ranks = numrank(mtr);
+	rows = numrow(mtr);
+	cols = numcol(mtr);
+
+	/*
+	 * Compute size in 8-byte (2^3) words, then shift to MiB (2^20)
+	 */
+	size = ((1ull << (rows + cols + ranks)) * banks) >> (20 - 3);
+	npages = MiB_TO_PAGES(size);
+
+	edac_dbg(0, "mc#%d: channel %d, dimm %d, %lld Mb (%d pages) bank: %d, rank: %d, row: %#x, col: %#x\n",
+		 imc->mc, chan, dimmno, size, npages,
+		 banks, ranks, rows, cols);
+
+	imc->chan[chan].dimms[dimmno].close_pg = GET_BITFIELD(mtr, 0, 0);
+	imc->chan[chan].dimms[dimmno].bank_xor_enable = GET_BITFIELD(mtr, 9, 9);
+	imc->chan[chan].dimms[dimmno].fine_grain_bank = GET_BITFIELD(amap, 0, 0);
+	imc->chan[chan].dimms[dimmno].rowbits = rows;
+	imc->chan[chan].dimms[dimmno].colbits = cols;
+
+	dimm->nr_pages = npages;
+	dimm->grain = 32;
+	dimm->dtype = get_width(mtr);
+	dimm->mtype = MEM_DDR4;
+	dimm->edac_mode = EDAC_SECDED; /* likely better than this */
+	snprintf(dimm->label, sizeof(dimm->label), "CPU_SrcID#%u_MC#%u_Chan#%u_DIMM#%u",
+		 imc->src_id, imc->lmc, chan, dimmno);
+
+	return 1;
+}
+
+#define SKX_GET_MTMTR(dev, reg) \
+	pci_read_config_dword((dev), 0x87c, &reg)
+
+static bool skx_check_ecc(struct pci_dev *pdev)
+{
+	u32 mtmtr;
+
+	SKX_GET_MTMTR(pdev, mtmtr);
+
+	return !!GET_BITFIELD(mtmtr, 2, 2);
+}
+
+static int skx_get_dimm_config(struct mem_ctl_info *mci)
+{
+	struct skx_pvt *pvt = mci->pvt_info;
+	struct skx_imc *imc = pvt->imc;
+	struct dimm_info *dimm;
+	int i, j;
+	u32 mtr, amap;
+	int ndimms;
+
+	for (i = 0; i < NUM_CHANNELS; i++) {
+		ndimms = 0;
+		pci_read_config_dword(imc->chan[i].cdev, 0x8C, &amap);
+		for (j = 0; j < NUM_DIMMS; j++) {
+			dimm = EDAC_DIMM_PTR(mci->layers, mci->dimms,
+					     mci->n_layers, i, j, 0);
+			pci_read_config_dword(imc->chan[i].cdev,
+					0x80 + 4*j, &mtr);
+			ndimms += get_dimm_info(mtr, amap, dimm, imc, i, j);
+		}
+		if (ndimms && !skx_check_ecc(imc->chan[0].cdev)) {
+			skx_printk(KERN_ERR, "ECC is disabled on imc %d\n", imc->mc);
+			return -ENODEV;
+		}
+	}
+
+	return 0;
+}
+
+static void skx_unregister_mci(struct skx_imc *imc)
+{
+	struct mem_ctl_info *mci = imc->mci;
+
+	if (!mci)
+		return;
+
+	edac_dbg(0, "MC%d: mci = %p\n", imc->mc, mci);
+
+	/* Remove MC sysfs nodes */
+	edac_mc_del_mc(mci->pdev);
+
+	edac_dbg(1, "%s: free mci struct\n", mci->ctl_name);
+	kfree(mci->ctl_name);
+	edac_mc_free(mci);
+}
+
+static int skx_register_mci(struct skx_imc *imc)
+{
+	struct mem_ctl_info *mci;
+	struct edac_mc_layer layers[2];
+	struct pci_dev *pdev = imc->chan[0].cdev;
+	struct skx_pvt *pvt;
+	int rc;
+
+	/* allocate a new MC control structure */
+	layers[0].type = EDAC_MC_LAYER_CHANNEL;
+	layers[0].size = NUM_CHANNELS;
+	layers[0].is_virt_csrow = false;
+	layers[1].type = EDAC_MC_LAYER_SLOT;
+	layers[1].size = NUM_DIMMS;
+	layers[1].is_virt_csrow = true;
+	mci = edac_mc_alloc(imc->mc, ARRAY_SIZE(layers), layers,
+			    sizeof(struct skx_pvt));
+
+	if (unlikely(!mci))
+		return -ENOMEM;
+
+	edac_dbg(0, "MC#%d: mci = %p\n", imc->mc, mci);
+
+	/* Associate skx_dev and mci for future usage */
+	imc->mci = mci;
+	pvt = mci->pvt_info;
+	pvt->imc = imc;
+
+	mci->ctl_name = kasprintf(GFP_KERNEL, "Skylake Socket#%d IMC#%d",
+				  imc->node_id, imc->lmc);
+	mci->mtype_cap = MEM_FLAG_DDR4;
+	mci->edac_ctl_cap = EDAC_FLAG_NONE;
+	mci->edac_cap = EDAC_FLAG_NONE;
+	mci->mod_name = "skx_edac.c";
+	mci->dev_name = pci_name(imc->chan[0].cdev);
+	mci->mod_ver = SKX_REVISION;
+	mci->ctl_page_to_phys = NULL;
+
+	rc = skx_get_dimm_config(mci);
+	if (rc < 0)
+		goto fail;
+
+	/* record ptr to the generic device */
+	mci->pdev = &pdev->dev;
+
+	/* add this new MC control structure to EDAC's list of MCs */
+	if (unlikely(edac_mc_add_mc(mci))) {
+		edac_dbg(0, "MC: failed edac_mc_add_mc()\n");
+		rc = -EINVAL;
+		goto fail;
+	}
+
+	return 0;
+
+fail:
+	kfree(mci->ctl_name);
+	edac_mc_free(mci);
+	imc->mci = NULL;
+	return rc;
+}
+
+#define	SKX_MAX_SAD 24
+
+#define SKX_GET_SAD(d, i, reg)	\
+	pci_read_config_dword((d)->sad_all, 0x60 + 8 * (i), &reg)
+#define SKX_GET_ILV(d, i, reg)	\
+	pci_read_config_dword((d)->sad_all, 0x64 + 8 * (i), &reg)
+
+#define	SKX_SAD_MOD3MODE(sad)	GET_BITFIELD((sad), 30, 31)
+#define	SKX_SAD_MOD3(sad)	GET_BITFIELD((sad), 27, 27)
+#define SKX_SAD_LIMIT(sad)	(((u64)GET_BITFIELD((sad), 7, 26) << 26) | MASK26)
+#define	SKX_SAD_MOD3ASMOD2(sad)	GET_BITFIELD((sad), 5, 6)
+#define	SKX_SAD_ATTR(sad)	GET_BITFIELD((sad), 3, 4)
+#define	SKX_SAD_INTERLEAVE(sad)	GET_BITFIELD((sad), 1, 2)
+#define SKX_SAD_ENABLE(sad)	GET_BITFIELD((sad), 0, 0)
+
+#define SKX_ILV_REMOTE(tgt)	(((tgt) & 8) == 0)
+#define SKX_ILV_TARGET(tgt)	((tgt) & 7)
+
+static bool skx_sad_decode(struct decoded_addr *res)
+{
+	struct skx_dev *d = list_first_entry(&skx_edac_list, typeof(*d), list);
+	u64 addr = res->addr;
+	int i, idx, tgt, lchan, shift;
+	u32 sad, ilv;
+	u64 limit, prev_limit;
+	int remote = 0;
+
+	/* Simple sanity check for I/O space or out of range */
+	if (addr >= skx_tohm || (addr >= skx_tolm && addr < BIT_ULL(32))) {
+		edac_dbg(0, "Address %llx out of range\n", addr);
+		return false;
+	}
+
+restart:
+	prev_limit = 0;
+	for (i = 0; i < SKX_MAX_SAD; i++) {
+		SKX_GET_SAD(d, i, sad);
+		limit = SKX_SAD_LIMIT(sad);
+		if (SKX_SAD_ENABLE(sad)) {
+			if (addr >= prev_limit && addr <= limit)
+				goto sad_found;
+		}
+		prev_limit = limit + 1;
+	}
+	edac_dbg(0, "No SAD entry for %llx\n", addr);
+	return false;
+
+sad_found:
+	SKX_GET_ILV(d, i, ilv);
+
+	switch (SKX_SAD_INTERLEAVE(sad)) {
+	case 0:
+		idx = GET_BITFIELD(addr, 6, 8);
+		break;
+	case 1:
+		idx = GET_BITFIELD(addr, 8, 10);
+		break;
+	case 2:
+		idx = GET_BITFIELD(addr, 12, 14);
+		break;
+	case 3:
+		idx = GET_BITFIELD(addr, 30, 32);
+		break;
+	}
+
+	tgt = GET_BITFIELD(ilv, 4 * idx, 4 * idx + 3);
+
+	/* If point to another node, find it and start over */
+	if (SKX_ILV_REMOTE(tgt)) {
+		if (remote) {
+			edac_dbg(0, "Double remote!\n");
+			return false;
+		}
+		remote = 1;
+		list_for_each_entry(d, &skx_edac_list, list) {
+			if (d->imc[0].src_id == SKX_ILV_TARGET(tgt))
+				goto restart;
+		}
+		edac_dbg(0, "Can't find node %d\n", SKX_ILV_TARGET(tgt));
+		return false;
+	}
+
+	if (SKX_SAD_MOD3(sad) == 0)
+		lchan = SKX_ILV_TARGET(tgt);
+	else {
+		switch (SKX_SAD_MOD3MODE(sad)) {
+		case 0:
+			shift = 6;
+			break;
+		case 1:
+			shift = 8;
+			break;
+		case 2:
+			shift = 12;
+			break;
+		default:
+			edac_dbg(0, "illegal mod3mode\n");
+			return false;
+		}
+		switch (SKX_SAD_MOD3ASMOD2(sad)) {
+		case 0:
+			lchan = (addr >> shift) % 3;
+			break;
+		case 1:
+			lchan = (addr >> shift) % 2;
+			break;
+		case 2:
+			lchan = (addr >> shift) % 2;
+			lchan = (lchan << 1) | ~lchan;
+			break;
+		case 3:
+			lchan = ((addr >> shift) % 2) << 1;
+			break;
+		}
+		lchan = (lchan << 1) | (SKX_ILV_TARGET(tgt) & 1);
+	}
+
+	res->dev = d;
+	res->socket = d->imc[0].src_id;
+	res->imc = GET_BITFIELD(d->mcroute, lchan * 3, lchan * 3 + 2);
+	res->channel = GET_BITFIELD(d->mcroute, lchan * 2 + 18, lchan * 2 + 19);
+
+	edac_dbg(2, "%llx: socket=%d imc=%d channel=%d\n",
+		 res->addr, res->socket, res->imc, res->channel);
+	return true;
+}
+
+#define	SKX_MAX_TAD 8
+
+#define SKX_GET_TADBASE(d, mc, i, reg)			\
+	pci_read_config_dword((d)->imc[mc].chan[0].cdev, 0x850 + 4 * (i), &reg)
+#define SKX_GET_TADWAYNESS(d, mc, i, reg)		\
+	pci_read_config_dword((d)->imc[mc].chan[0].cdev, 0x880 + 4 * (i), &reg)
+#define SKX_GET_TADCHNILVOFFSET(d, mc, ch, i, reg)	\
+	pci_read_config_dword((d)->imc[mc].chan[ch].cdev, 0x90 + 4 * (i), &reg)
+
+#define	SKX_TAD_BASE(b)		((u64)GET_BITFIELD((b), 12, 31) << 26)
+#define SKX_TAD_SKT_GRAN(b)	GET_BITFIELD((b), 4, 5)
+#define SKX_TAD_CHN_GRAN(b)	GET_BITFIELD((b), 6, 7)
+#define	SKX_TAD_LIMIT(b)	(((u64)GET_BITFIELD((b), 12, 31) << 26) | MASK26)
+#define	SKX_TAD_OFFSET(b)	((u64)GET_BITFIELD((b), 4, 23) << 26)
+#define	SKX_TAD_SKTWAYS(b)	(1 << GET_BITFIELD((b), 10, 11))
+#define	SKX_TAD_CHNWAYS(b)	(GET_BITFIELD((b), 8, 9) + 1)
+
+/* which bit used for both socket and channel interleave */
+static int skx_granularity[] = { 6, 8, 12, 30 };
+
+static u64 skx_do_interleave(u64 addr, int shift, int ways, u64 lowbits)
+{
+	addr >>= shift;
+	addr /= ways;
+	addr <<= shift;
+
+	return addr | (lowbits & ((1ull << shift) - 1));
+}
+
+static bool skx_tad_decode(struct decoded_addr *res)
+{
+	int i;
+	u32 base, wayness, chnilvoffset;
+	int skt_interleave_bit, chn_interleave_bit;
+	u64 channel_addr;
+
+	for (i = 0; i < SKX_MAX_TAD; i++) {
+		SKX_GET_TADBASE(res->dev, res->imc, i, base);
+		SKX_GET_TADWAYNESS(res->dev, res->imc, i, wayness);
+		if (SKX_TAD_BASE(base) <= res->addr && res->addr <= SKX_TAD_LIMIT(wayness))
+			goto tad_found;
+	}
+	edac_dbg(0, "No TAD entry for %llx\n", res->addr);
+	return false;
+
+tad_found:
+	res->sktways = SKX_TAD_SKTWAYS(wayness);
+	res->chanways = SKX_TAD_CHNWAYS(wayness);
+	skt_interleave_bit = skx_granularity[SKX_TAD_SKT_GRAN(base)];
+	chn_interleave_bit = skx_granularity[SKX_TAD_CHN_GRAN(base)];
+
+	SKX_GET_TADCHNILVOFFSET(res->dev, res->imc, res->channel, i, chnilvoffset);
+	channel_addr = res->addr - SKX_TAD_OFFSET(chnilvoffset);
+
+	if (res->chanways == 3 && skt_interleave_bit > chn_interleave_bit) {
+		/* Must handle channel first, then socket */
+		channel_addr = skx_do_interleave(channel_addr, chn_interleave_bit,
+						 res->chanways, channel_addr);
+		channel_addr = skx_do_interleave(channel_addr, skt_interleave_bit,
+						 res->sktways, channel_addr);
+	} else {
+		/* Handle socket then channel. Preserve low bits from original address */
+		channel_addr = skx_do_interleave(channel_addr, skt_interleave_bit,
+						 res->sktways, res->addr);
+		channel_addr = skx_do_interleave(channel_addr, chn_interleave_bit,
+						 res->chanways, res->addr);
+	}
+
+	res->chan_addr = channel_addr;
+
+	edac_dbg(2, "%llx: chan_addr=%llx sktways=%d chanways=%d\n",
+		 res->addr, res->chan_addr, res->sktways, res->chanways);
+	return true;
+}
+
+#define SKX_MAX_RIR 4
+
+#define SKX_GET_RIRWAYNESS(d, mc, ch, i, reg)		\
+	pci_read_config_dword((d)->imc[mc].chan[ch].cdev,	\
+			      0x108 + 4 * (i), &reg)
+#define SKX_GET_RIRILV(d, mc, ch, idx, i, reg)		\
+	pci_read_config_dword((d)->imc[mc].chan[ch].cdev,	\
+			      0x120 + 16 * idx + 4 * (i), &reg)
+
+#define	SKX_RIR_VALID(b) GET_BITFIELD((b), 31, 31)
+#define	SKX_RIR_LIMIT(b) (((u64)GET_BITFIELD((b), 1, 11) << 29) | MASK29)
+#define	SKX_RIR_WAYS(b) (1 << GET_BITFIELD((b), 28, 29))
+#define	SKX_RIR_CHAN_RANK(b) GET_BITFIELD((b), 16, 19)
+#define	SKX_RIR_OFFSET(b) ((u64)(GET_BITFIELD((b), 2, 15) << 26))
+
+static bool skx_rir_decode(struct decoded_addr *res)
+{
+	int i, idx, chan_rank;
+	int shift;
+	u32 rirway, rirlv;
+	u64 rank_addr, prev_limit = 0, limit;
+
+	if (res->dev->imc[res->imc].chan[res->channel].dimms[0].close_pg)
+		shift = 6;
+	else
+		shift = 13;
+
+	for (i = 0; i < SKX_MAX_RIR; i++) {
+		SKX_GET_RIRWAYNESS(res->dev, res->imc, res->channel, i, rirway);
+		limit = SKX_RIR_LIMIT(rirway);
+		if (SKX_RIR_VALID(rirway)) {
+			if (prev_limit <= res->chan_addr &&
+			    res->chan_addr <= limit)
+				goto rir_found;
+		}
+		prev_limit = limit;
+	}
+	edac_dbg(0, "No RIR entry for %llx\n", res->addr);
+	return false;
+
+rir_found:
+	rank_addr = res->chan_addr >> shift;
+	rank_addr /= SKX_RIR_WAYS(rirway);
+	rank_addr <<= shift;
+	rank_addr |= res->chan_addr & GENMASK_ULL(shift - 1, 0);
+
+	res->rank_address = rank_addr;
+	idx = (res->chan_addr >> shift) % SKX_RIR_WAYS(rirway);
+
+	SKX_GET_RIRILV(res->dev, res->imc, res->channel, idx, i, rirlv);
+	res->rank_address = rank_addr - SKX_RIR_OFFSET(rirlv);
+	chan_rank = SKX_RIR_CHAN_RANK(rirlv);
+	res->channel_rank = chan_rank;
+	res->dimm = chan_rank / 4;
+	res->rank = chan_rank % 4;
+
+	edac_dbg(2, "%llx: dimm=%d rank=%d chan_rank=%d rank_addr=%llx\n",
+		 res->addr, res->dimm, res->rank,
+		 res->channel_rank, res->rank_address);
+	return true;
+}
+
+static u8 skx_close_row[] = {
+	15, 16, 17, 18, 20, 21, 22, 28, 10, 11, 12, 13, 29, 30, 31, 32, 33
+};
+static u8 skx_close_column[] = {
+	3, 4, 5, 14, 19, 23, 24, 25, 26, 27
+};
+static u8 skx_open_row[] = {
+	14, 15, 16, 20, 28, 21, 22, 23, 24, 25, 26, 27, 29, 30, 31, 32, 33
+};
+static u8 skx_open_column[] = {
+	3, 4, 5, 6, 7, 8, 9, 10, 11, 12
+};
+static u8 skx_open_fine_column[] = {
+	3, 4, 5, 7, 8, 9, 10, 11, 12, 13
+};
+
+static int skx_bits(u64 addr, int nbits, u8 *bits)
+{
+	int i, res = 0;
+
+	for (i = 0; i < nbits; i++)
+		res |= ((addr >> bits[i]) & 1) << i;
+	return res;
+}
+
+static int skx_bank_bits(u64 addr, int b0, int b1, int do_xor, int x0, int x1)
+{
+	int ret = GET_BITFIELD(addr, b0, b0) | (GET_BITFIELD(addr, b1, b1) << 1);
+
+	if (do_xor)
+		ret ^= GET_BITFIELD(addr, x0, x0) | (GET_BITFIELD(addr, x1, x1) << 1);
+
+	return ret;
+}
+
+static bool skx_mad_decode(struct decoded_addr *r)
+{
+	struct skx_dimm *dimm = &r->dev->imc[r->imc].chan[r->channel].dimms[r->dimm];
+	int bg0 = dimm->fine_grain_bank ? 6 : 13;
+
+	if (dimm->close_pg) {
+		r->row = skx_bits(r->rank_address, dimm->rowbits, skx_close_row);
+		r->column = skx_bits(r->rank_address, dimm->colbits, skx_close_column);
+		r->column |= 0x400; /* C10 is autoprecharge, always set */
+		r->bank_address = skx_bank_bits(r->rank_address, 8, 9, dimm->bank_xor_enable, 22, 28);
+		r->bank_group = skx_bank_bits(r->rank_address, 6, 7, dimm->bank_xor_enable, 20, 21);
+	} else {
+		r->row = skx_bits(r->rank_address, dimm->rowbits, skx_open_row);
+		if (dimm->fine_grain_bank)
+			r->column = skx_bits(r->rank_address, dimm->colbits, skx_open_fine_column);
+		else
+			r->column = skx_bits(r->rank_address, dimm->colbits, skx_open_column);
+		r->bank_address = skx_bank_bits(r->rank_address, 18, 19, dimm->bank_xor_enable, 22, 23);
+		r->bank_group = skx_bank_bits(r->rank_address, bg0, 17, dimm->bank_xor_enable, 20, 21);
+	}
+	r->row &= (1u << dimm->rowbits) - 1;
+
+	edac_dbg(2, "%llx: row=%x col=%x bank_addr=%d bank_group=%d\n",
+		 r->addr, r->row, r->column, r->bank_address,
+		 r->bank_group);
+	return true;
+}
+
+static bool skx_decode(struct decoded_addr *res)
+{
+
+	return skx_sad_decode(res) && skx_tad_decode(res) &&
+		skx_rir_decode(res) && skx_mad_decode(res);
+}
+
+#ifdef CONFIG_EDAC_DEBUG
+/*
+ * Debug feature. Make /sys/kernel/debug/skx_edac_test/addr.
+ * Write an address to this file to exercise the address decode
+ * logic in this driver.
+ */
+static struct dentry *skx_test;
+static u64 skx_fake_addr;
+
+static int debugfs_u64_set(void *data, u64 val)
+{
+	struct decoded_addr res;
+
+	res.addr = val;
+	skx_decode(&res);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(fops_u64_wo, NULL, debugfs_u64_set, "%llu\n");
+
+static struct dentry *mydebugfs_create(const char *name, umode_t mode,
+				       struct dentry *parent, u64 *value)
+{
+	return debugfs_create_file(name, mode, parent, value, &fops_u64_wo);
+}
+
+static void setup_skx_debug(void)
+{
+	skx_test = debugfs_create_dir("skx_edac_test", NULL);
+	mydebugfs_create("addr", S_IWUSR, skx_test, &skx_fake_addr);
+}
+
+static void teardown_skx_debug(void)
+{
+	debugfs_remove_recursive(skx_test);
+}
+#else
+static void setup_skx_debug(void)
+{
+}
+
+static void teardown_skx_debug(void)
+{
+}
+#endif /*CONFIG_EDAC_DEBUG*/
+
+static void skx_mce_output_error(struct mem_ctl_info *mci,
+				 const struct mce *m,
+				 struct decoded_addr *res)
+{
+	enum hw_event_mc_err_type tp_event;
+	char *type, *optype, msg[256];
+	bool ripv = GET_BITFIELD(m->mcgstatus, 0, 0);
+	bool overflow = GET_BITFIELD(m->status, 62, 62);
+	bool uncorrected_error = GET_BITFIELD(m->status, 61, 61);
+	bool recoverable;
+	u32 core_err_cnt = GET_BITFIELD(m->status, 38, 52);
+	u32 mscod = GET_BITFIELD(m->status, 16, 31);
+	u32 errcode = GET_BITFIELD(m->status, 0, 15);
+	u32 optypenum = GET_BITFIELD(m->status, 4, 6);
+
+	recoverable = GET_BITFIELD(m->status, 56, 56);
+
+	if (uncorrected_error) {
+		if (ripv) {
+			type = "FATAL";
+			tp_event = HW_EVENT_ERR_FATAL;
+		} else {
+			type = "NON_FATAL";
+			tp_event = HW_EVENT_ERR_UNCORRECTED;
+		}
+	} else {
+		type = "CORRECTED";
+		tp_event = HW_EVENT_ERR_CORRECTED;
+	}
+
+	/*
+	 * According with Table 15-9 of the Intel Architecture spec vol 3A,
+	 * memory errors should fit in this mask:
+	 *	000f 0000 1mmm cccc (binary)
+	 * where:
+	 *	f = Correction Report Filtering Bit. If 1, subsequent errors
+	 *	    won't be shown
+	 *	mmm = error type
+	 *	cccc = channel
+	 * If the mask doesn't match, report an error to the parsing logic
+	 */
+	if (!((errcode & 0xef80) == 0x80)) {
+		optype = "Can't parse: it is not a mem";
+	} else {
+		switch (optypenum) {
+		case 0:
+			optype = "generic undef request error";
+			break;
+		case 1:
+			optype = "memory read error";
+			break;
+		case 2:
+			optype = "memory write error";
+			break;
+		case 3:
+			optype = "addr/cmd error";
+			break;
+		case 4:
+			optype = "memory scrubbing error";
+			break;
+		default:
+			optype = "reserved";
+			break;
+		}
+	}
+
+	snprintf(msg, sizeof(msg),
+		 "%s%s err_code:%04x:%04x socket:%d imc:%d rank:%d bg:%d ba:%d row:%x col:%x",
+		 overflow ? " OVERFLOW" : "",
+		 (uncorrected_error && recoverable) ? " recoverable" : "",
+		 mscod, errcode,
+		 res->socket, res->imc, res->rank,
+		 res->bank_group, res->bank_address, res->row, res->column);
+
+	edac_dbg(0, "%s\n", msg);
+
+	/* Call the helper to output message */
+	edac_mc_handle_error(tp_event, mci, core_err_cnt,
+			     m->addr >> PAGE_SHIFT, m->addr & ~PAGE_MASK, 0,
+			     res->channel, res->dimm, -1,
+			     optype, msg);
+}
+
+static int skx_mce_check_error(struct notifier_block *nb, unsigned long val,
+			       void *data)
+{
+	struct mce *mce = (struct mce *)data;
+	struct decoded_addr res;
+	struct mem_ctl_info *mci;
+	char *type;
+
+	if (get_edac_report_status() == EDAC_REPORTING_DISABLED)
+		return NOTIFY_DONE;
+
+	/* ignore unless this is memory related with an address */
+	if ((mce->status & 0xefff) >> 7 != 1 || !(mce->status & MCI_STATUS_ADDRV))
+		return NOTIFY_DONE;
+
+	res.addr = mce->addr;
+	if (!skx_decode(&res))
+		return NOTIFY_DONE;
+	mci = res.dev->imc[res.imc].mci;
+
+	if (mce->mcgstatus & MCG_STATUS_MCIP)
+		type = "Exception";
+	else
+		type = "Event";
+
+	skx_mc_printk(mci, KERN_DEBUG, "HANDLING MCE MEMORY ERROR\n");
+
+	skx_mc_printk(mci, KERN_DEBUG, "CPU %d: Machine Check %s: %Lx "
+			  "Bank %d: %016Lx\n", mce->extcpu, type,
+			  mce->mcgstatus, mce->bank, mce->status);
+	skx_mc_printk(mci, KERN_DEBUG, "TSC %llx ", mce->tsc);
+	skx_mc_printk(mci, KERN_DEBUG, "ADDR %llx ", mce->addr);
+	skx_mc_printk(mci, KERN_DEBUG, "MISC %llx ", mce->misc);
+
+	skx_mc_printk(mci, KERN_DEBUG, "PROCESSOR %u:%x TIME %llu SOCKET "
+			  "%u APIC %x\n", mce->cpuvendor, mce->cpuid,
+			  mce->time, mce->socketid, mce->apicid);
+
+	skx_mce_output_error(mci, mce, &res);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block skx_mce_dec = {
+	.notifier_call = skx_mce_check_error,
+};
+
+static void skx_remove(void)
+{
+	int i, j;
+	struct skx_dev *d, *tmp;
+
+	edac_dbg(0, "\n");
+
+	list_for_each_entry_safe(d, tmp, &skx_edac_list, list) {
+		list_del(&d->list);
+		for (i = 0; i < NUM_IMC; i++) {
+			skx_unregister_mci(&d->imc[i]);
+			for (j = 0; j < NUM_CHANNELS; j++)
+				pci_dev_put(d->imc[i].chan[j].cdev);
+		}
+		pci_dev_put(d->util_all);
+		pci_dev_put(d->sad_all);
+
+		kfree(d);
+	}
+}
+
+/*
+ * skx_init:
+ *	make sure we are running on the correct cpu model
+ *	search for all the devices we need
+ *	check which DIMMs are present.
+ */
+int __init skx_init(void)
+{
+	const struct x86_cpu_id *id;
+	const struct munit *m;
+	int rc = 0, i;
+	u8 mc = 0, src_id, node_id;
+	struct skx_dev *d;
+
+	edac_dbg(2, "\n");
+
+	id = x86_match_cpu(skx_cpuids);
+	if (!id)
+		return -ENODEV;
+
+	rc = skx_get_hi_lo();
+	if (rc)
+		return rc;
+
+	rc = get_all_bus_mappings();
+	if (rc < 0)
+		goto fail;
+	if (rc == 0) {
+		edac_dbg(2, "No memory controllers found\n");
+		return -ENODEV;
+	}
+
+	for (m = skx_all_munits; m->did; m++) {
+		rc = get_all_munits(m);
+		if (rc < 0)
+			goto fail;
+		if (rc != m->per_socket * skx_num_sockets) {
+			edac_dbg(2, "Expected %d, got %d of %x\n",
+				 m->per_socket * skx_num_sockets, rc, m->did);
+			rc = -ENODEV;
+			goto fail;
+		}
+	}
+
+	list_for_each_entry(d, &skx_edac_list, list) {
+		src_id = get_src_id(d);
+		node_id = skx_get_node_id(d);
+		edac_dbg(2, "src_id=%d node_id=%d\n", src_id, node_id);
+		for (i = 0; i < NUM_IMC; i++) {
+			d->imc[i].mc = mc++;
+			d->imc[i].lmc = i;
+			d->imc[i].src_id = src_id;
+			d->imc[i].node_id = node_id;
+			rc = skx_register_mci(&d->imc[i]);
+			if (rc < 0)
+				goto fail;
+		}
+	}
+
+	/* Ensure that the OPSTATE is set correctly for POLL or NMI */
+	opstate_init();
+
+	setup_skx_debug();
+
+	mce_register_decode_chain(&skx_mce_dec);
+
+	return 0;
+fail:
+	skx_remove();
+	return rc;
+}
+
+static void __exit skx_exit(void)
+{
+	edac_dbg(2, "\n");
+	mce_unregister_decode_chain(&skx_mce_dec);
+	skx_remove();
+	teardown_skx_debug();
+}
+
+module_init(skx_init);
+module_exit(skx_exit);
+
+module_param(edac_op_state, int, 0444);
+MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI");
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Tony Luck");
+MODULE_DESCRIPTION("MC Driver for Intel Skylake server processors");

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -646,9 +646,9 @@ int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev);
 void amdgpu_gart_table_vram_unpin(struct amdgpu_device *adev);
 int amdgpu_gart_init(struct amdgpu_device *adev);
 void amdgpu_gart_fini(struct amdgpu_device *adev);
-void amdgpu_gart_unbind(struct amdgpu_device *adev, unsigned offset,
+void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset,
 			int pages);
-int amdgpu_gart_bind(struct amdgpu_device *adev, unsigned offset,
+int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset,
 		     int pages, struct page **pagelist,
 		     dma_addr_t *dma_addr, uint32_t flags);
 

drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c

@@ -200,16 +200,7 @@ static int amdgpu_atpx_validate(struct amdgpu_atpx *atpx)
 	atpx->is_hybrid = false;
 	if (valid_bits & ATPX_MS_HYBRID_GFX_SUPPORTED) {
 		printk("ATPX Hybrid Graphics\n");
-#if 1
-		/* This is a temporary hack until the D3 cold support
-		 * makes it upstream.  The ATPX power_control method seems
-		 * to still work on even if the system should be using
-		 * the new standardized hybrid D3 cold ACPI interface.
-		 */
-		atpx->functions.power_cntl = true;
-#else
 		atpx->functions.power_cntl = false;
-#endif
 		atpx->is_hybrid = true;
 	}
 

drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c

@@ -221,7 +221,7 @@ void amdgpu_gart_table_vram_free(struct amdgpu_device *adev)
  * Unbinds the requested pages from the gart page table and
  * replaces them with the dummy page (all asics).
  */
-void amdgpu_gart_unbind(struct amdgpu_device *adev, unsigned offset,
+void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset,
 			int pages)
 {
 	unsigned t;
@@ -268,7 +268,7 @@ void amdgpu_gart_unbind(struct amdgpu_device *adev, unsigned offset,
  * (all asics).
  * Returns 0 for success, -EINVAL for failure.
  */
-int amdgpu_gart_bind(struct amdgpu_device *adev, unsigned offset,
+int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset,
 		     int pages, struct page **pagelist, dma_addr_t *dma_addr,
 		     uint32_t flags)
 {

drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c

@@ -1187,7 +1187,8 @@ int amdgpu_uvd_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		r = 0;
 	}
 
-error:
 	fence_put(fence);
+
+error:
 	return r;
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

@@ -1535,7 +1535,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	r = amd_sched_entity_init(&ring->sched, &vm->entity,
 				  rq, amdgpu_sched_jobs);
 	if (r)
-		return r;
+		goto err;
 
 	vm->page_directory_fence = NULL;
 
@@ -1565,6 +1565,9 @@ error_free_page_directory:
 error_free_sched_entity:
 	amd_sched_entity_fini(&ring->sched, &vm->entity);
 
+err:
+	drm_free_large(vm->page_tables);
+
 	return r;
 }
 

drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c

@@ -184,7 +184,7 @@ u32 __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd,
 							sizeof(u32)) + inx;
 
 	pr_debug("kfd: get kernel queue doorbell\n"
-			 "     doorbell offset   == 0x%08d\n"
+			 "     doorbell offset   == 0x%08X\n"
 			 "     kernel address    == 0x%08lX\n",
 		*doorbell_off, (uintptr_t)(kfd->doorbell_kernel_ptr + inx));
 

drivers/gpu/drm/drm_fb_helper.c

@@ -464,7 +464,7 @@ static bool drm_fb_helper_is_bound(struct drm_fb_helper *fb_helper)
 
 	/* Sometimes user space wants everything disabled, so don't steal the
 	 * display if there's a master. */
-	if (lockless_dereference(dev->master))
+	if (READ_ONCE(dev->master))
 		return false;
 
 	drm_for_each_crtc(crtc, dev) {

drivers/gpu/drm/etnaviv/etnaviv_gpu.c

@@ -1333,8 +1333,6 @@ int etnaviv_gpu_submit(struct etnaviv_gpu *gpu,
 	if (ret < 0)
 		return ret;
 
-	mutex_lock(&gpu->lock);
-
 	/*
 	 * TODO
 	 *
@@ -1348,16 +1346,18 @@ int etnaviv_gpu_submit(struct etnaviv_gpu *gpu,
 	if (unlikely(event == ~0U)) {
 		DRM_ERROR("no free event\n");
 		ret = -EBUSY;
-		goto out_unlock;
+		goto out_pm_put;
 	}
 
 	fence = etnaviv_gpu_fence_alloc(gpu);
 	if (!fence) {
 		event_free(gpu, event);
 		ret = -ENOMEM;
-		goto out_unlock;
+		goto out_pm_put;
 	}
 
+	mutex_lock(&gpu->lock);
+
 	gpu->event[event].fence = fence;
 	submit->fence = fence->seqno;
 	gpu->active_fence = submit->fence;
@@ -1395,9 +1395,9 @@ int etnaviv_gpu_submit(struct etnaviv_gpu *gpu,
 	hangcheck_timer_reset(gpu);
 	ret = 0;
 
-out_unlock:
 	mutex_unlock(&gpu->lock);
 
+out_pm_put:
 	etnaviv_gpu_pm_put(gpu);
 
 	return ret;

drivers/gpu/drm/i915/i915_drv.h

@@ -1854,6 +1854,7 @@ struct drm_i915_private {
 	enum modeset_restore modeset_restore;
 	struct mutex modeset_restore_lock;
 	struct drm_atomic_state *modeset_restore_state;
+	struct drm_modeset_acquire_ctx reset_ctx;
 
 	struct list_head vm_list; /* Global list of all address spaces */
 	struct i915_ggtt ggtt; /* VM representing the global address space */

drivers/gpu/drm/i915/i915_gem.c

@@ -879,9 +879,12 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data,
 	ret = i915_gem_shmem_pread(dev, obj, args, file);
 
 	/* pread for non shmem backed objects */
-	if (ret == -EFAULT || ret == -ENODEV)
+	if (ret == -EFAULT || ret == -ENODEV) {
+		intel_runtime_pm_get(to_i915(dev));
 		ret = i915_gem_gtt_pread(dev, obj, args->size,
 					args->offset, args->data_ptr);
+		intel_runtime_pm_put(to_i915(dev));
+	}
 
 out:
 	drm_gem_object_unreference(&obj->base);
@@ -1306,7 +1309,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
 		 * textures). Fallback to the shmem path in that case. */
 	}
 
-	if (ret == -EFAULT) {
+	if (ret == -EFAULT || ret == -ENOSPC) {
 		if (obj->phys_handle)
 			ret = i915_gem_phys_pwrite(obj, args, file);
 		else if (i915_gem_object_has_struct_page(obj))
@@ -3169,6 +3172,8 @@ static void i915_gem_reset_engine_cleanup(struct intel_engine_cs *engine)
 	}
 
 	intel_ring_init_seqno(engine, engine->last_submitted_seqno);
+
+	engine->i915->gt.active_engines &= ~intel_engine_flag(engine);
 }
 
 void i915_gem_reset(struct drm_device *dev)
@@ -3186,6 +3191,7 @@ void i915_gem_reset(struct drm_device *dev)
 
 	for_each_engine(engine, dev_priv)
 		i915_gem_reset_engine_cleanup(engine);
+	mod_delayed_work(dev_priv->wq, &dev_priv->gt.idle_work, 0);
 
 	i915_gem_context_reset(dev);
 

drivers/gpu/drm/i915/i915_gem_gtt.c

@@ -2873,6 +2873,7 @@ void i915_ggtt_cleanup_hw(struct drm_device *dev)
 		struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
 
 		ppgtt->base.cleanup(&ppgtt->base);
+		kfree(ppgtt);
 	}
 
 	i915_gem_cleanup_stolen(dev);

drivers/gpu/drm/i915/i915_reg.h

@@ -1536,6 +1536,7 @@ enum skl_disp_power_wells {
 #define BALANCE_LEG_MASK(port)		(7<<(8+3*(port)))
 /* Balance leg disable bits */
 #define BALANCE_LEG_DISABLE_SHIFT	23
+#define BALANCE_LEG_DISABLE(port)	(1 << (23 + (port)))
 
 /*
  * Fence registers

drivers/gpu/drm/i915/intel_audio.c

@@ -600,6 +600,8 @@ static void i915_audio_component_codec_wake_override(struct device *dev,
 	if (!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv))
 		return;
 
+	i915_audio_component_get_power(dev);
+
 	/*
 	 * Enable/disable generating the codec wake signal, overriding the
 	 * internal logic to generate the codec wake to controller.
@@ -615,6 +617,8 @@ static void i915_audio_component_codec_wake_override(struct device *dev,
 		I915_WRITE(HSW_AUD_CHICKENBIT, tmp);
 		usleep_range(1000, 1500);
 	}
+
+	i915_audio_component_put_power(dev);
 }
 
 /* Get CDCLK in kHz  */
@@ -648,6 +652,7 @@ static int i915_audio_component_sync_audio_rate(struct device *dev,
 	    !IS_HASWELL(dev_priv))
 		return 0;
 
+	i915_audio_component_get_power(dev);
 	mutex_lock(&dev_priv->av_mutex);
 	/* 1. get the pipe */
 	intel_encoder = dev_priv->dig_port_map[port];
@@ -698,6 +703,7 @@ static int i915_audio_component_sync_audio_rate(struct device *dev,
 
  unlock:
 	mutex_unlock(&dev_priv->av_mutex);
+	i915_audio_component_put_power(dev);
 	return err;
 }
 

drivers/gpu/drm/i915/intel_ddi.c

@@ -145,7 +145,7 @@ static const struct ddi_buf_trans skl_ddi_translations_dp[] = {
 static const struct ddi_buf_trans skl_u_ddi_translations_dp[] = {
 	{ 0x0000201B, 0x000000A2, 0x0 },
 	{ 0x00005012, 0x00000088, 0x0 },
-	{ 0x80007011, 0x000000CD, 0x0 },
+	{ 0x80007011, 0x000000CD, 0x1 },
 	{ 0x80009010, 0x000000C0, 0x1 },
 	{ 0x0000201B, 0x0000009D, 0x0 },
 	{ 0x80005012, 0x000000C0, 0x1 },
@@ -158,7 +158,7 @@ static const struct ddi_buf_trans skl_u_ddi_translations_dp[] = {
 static const struct ddi_buf_trans skl_y_ddi_translations_dp[] = {
 	{ 0x00000018, 0x000000A2, 0x0 },
 	{ 0x00005012, 0x00000088, 0x0 },
-	{ 0x80007011, 0x000000CD, 0x0 },
+	{ 0x80007011, 0x000000CD, 0x3 },
 	{ 0x80009010, 0x000000C0, 0x3 },
 	{ 0x00000018, 0x0000009D, 0x0 },
 	{ 0x80005012, 0x000000C0, 0x3 },
@@ -388,6 +388,40 @@ skl_get_buf_trans_hdmi(struct drm_i915_private *dev_priv, int *n_entries)
 	}
 }
 
+static int intel_ddi_hdmi_level(struct drm_i915_private *dev_priv, enum port port)
+{
+	int n_hdmi_entries;
+	int hdmi_level;
+	int hdmi_default_entry;
+
+	hdmi_level = dev_priv->vbt.ddi_port_info[port].hdmi_level_shift;
+
+	if (IS_BROXTON(dev_priv))
+		return hdmi_level;
+
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) {
+		skl_get_buf_trans_hdmi(dev_priv, &n_hdmi_entries);
+		hdmi_default_entry = 8;
+	} else if (IS_BROADWELL(dev_priv)) {
+		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
+		hdmi_default_entry = 7;
+	} else if (IS_HASWELL(dev_priv)) {
+		n_hdmi_entries = ARRAY_SIZE(hsw_ddi_translations_hdmi);
+		hdmi_default_entry = 6;
+	} else {
+		WARN(1, "ddi translation table missing\n");
+		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
+		hdmi_default_entry = 7;
+	}
+
+	/* Choose a good default if VBT is badly populated */
+	if (hdmi_level == HDMI_LEVEL_SHIFT_UNKNOWN ||
+	    hdmi_level >= n_hdmi_entries)
+		hdmi_level = hdmi_default_entry;
+
+	return hdmi_level;
+}
+
 /*
  * Starting with Haswell, DDI port buffers must be programmed with correct
  * values in advance. The buffer values are different for FDI and DP modes,
@@ -399,7 +433,7 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 iboost_bit = 0;
-	int i, n_hdmi_entries, n_dp_entries, n_edp_entries, hdmi_default_entry,
+	int i, n_hdmi_entries, n_dp_entries, n_edp_entries,
 	    size;
 	int hdmi_level;
 	enum port port;
@@ -410,7 +444,7 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 	const struct ddi_buf_trans *ddi_translations;
 
 	port = intel_ddi_get_encoder_port(encoder);
-	hdmi_level = dev_priv->vbt.ddi_port_info[port].hdmi_level_shift;
+	hdmi_level = intel_ddi_hdmi_level(dev_priv, port);
 
 	if (IS_BROXTON(dev_priv)) {
 		if (encoder->type != INTEL_OUTPUT_HDMI)
@@ -430,7 +464,6 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 				skl_get_buf_trans_edp(dev_priv, &n_edp_entries);
 		ddi_translations_hdmi =
 				skl_get_buf_trans_hdmi(dev_priv, &n_hdmi_entries);
-		hdmi_default_entry = 8;
 		/* If we're boosting the current, set bit 31 of trans1 */
 		if (dev_priv->vbt.ddi_port_info[port].hdmi_boost_level ||
 		    dev_priv->vbt.ddi_port_info[port].dp_boost_level)
@@ -456,7 +489,6 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 
 		n_dp_entries = ARRAY_SIZE(bdw_ddi_translations_dp);
 		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
-		hdmi_default_entry = 7;
 	} else if (IS_HASWELL(dev_priv)) {
 		ddi_translations_fdi = hsw_ddi_translations_fdi;
 		ddi_translations_dp = hsw_ddi_translations_dp;
@@ -464,7 +496,6 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 		ddi_translations_hdmi = hsw_ddi_translations_hdmi;
 		n_dp_entries = n_edp_entries = ARRAY_SIZE(hsw_ddi_translations_dp);
 		n_hdmi_entries = ARRAY_SIZE(hsw_ddi_translations_hdmi);
-		hdmi_default_entry = 6;
 	} else {
 		WARN(1, "ddi translation table missing\n");
 		ddi_translations_edp = bdw_ddi_translations_dp;
@@ -474,7 +505,6 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 		n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_edp);
 		n_dp_entries = ARRAY_SIZE(bdw_ddi_translations_dp);
 		n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi);
-		hdmi_default_entry = 7;
 	}
 
 	switch (encoder->type) {
@@ -505,11 +535,6 @@ void intel_prepare_ddi_buffer(struct intel_encoder *encoder)
 	if (encoder->type != INTEL_OUTPUT_HDMI)
 		return;
 
-	/* Choose a good default if VBT is badly populated */
-	if (hdmi_level == HDMI_LEVEL_SHIFT_UNKNOWN ||
-	    hdmi_level >= n_hdmi_entries)
-		hdmi_level = hdmi_default_entry;
-
 	/* Entry 9 is for HDMI: */
 	I915_WRITE(DDI_BUF_TRANS_LO(port, i),
 		   ddi_translations_hdmi[hdmi_level].trans1 | iboost_bit);
@@ -1379,14 +1404,30 @@ void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc)
 			   TRANS_CLK_SEL_DISABLED);
 }
 
-static void skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
-			       u32 level, enum port port, int type)
+static void _skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
+				enum port port, uint8_t iboost)
 {
+	u32 tmp;
+
+	tmp = I915_READ(DISPIO_CR_TX_BMU_CR0);
+	tmp &= ~(BALANCE_LEG_MASK(port) | BALANCE_LEG_DISABLE(port));
+	if (iboost)
+		tmp |= iboost << BALANCE_LEG_SHIFT(port);
+	else
+		tmp |= BALANCE_LEG_DISABLE(port);
+	I915_WRITE(DISPIO_CR_TX_BMU_CR0, tmp);
+}
+
+static void skl_ddi_set_iboost(struct intel_encoder *encoder, u32 level)
+{
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(&encoder->base);
+	struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
+	enum port port = intel_dig_port->port;
+	int type = encoder->type;
 	const struct ddi_buf_trans *ddi_translations;
 	uint8_t iboost;
 	uint8_t dp_iboost, hdmi_iboost;
 	int n_entries;
-	u32 reg;
 
 	/* VBT may override standard boost values */
 	dp_iboost = dev_priv->vbt.ddi_port_info[port].dp_boost_level;
@@ -1428,16 +1469,10 @@ static void skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
 		return;
 	}
 
-	reg = I915_READ(DISPIO_CR_TX_BMU_CR0);
-	reg &= ~BALANCE_LEG_MASK(port);
-	reg &= ~(1 << (BALANCE_LEG_DISABLE_SHIFT + port));
-
-	if (iboost)
-		reg |= iboost << BALANCE_LEG_SHIFT(port);
-	else
-		reg |= 1 << (BALANCE_LEG_DISABLE_SHIFT + port);
+	_skl_ddi_set_iboost(dev_priv, port, iboost);
 
-	I915_WRITE(DISPIO_CR_TX_BMU_CR0, reg);
+	if (port == PORT_A && intel_dig_port->max_lanes == 4)
+		_skl_ddi_set_iboost(dev_priv, PORT_E, iboost);
 }
 
 static void bxt_ddi_vswing_sequence(struct drm_i915_private *dev_priv,
@@ -1568,7 +1603,7 @@ uint32_t ddi_signal_levels(struct intel_dp *intel_dp)
 	level = translate_signal_level(signal_levels);
 
 	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
-		skl_ddi_set_iboost(dev_priv, level, port, encoder->type);
+		skl_ddi_set_iboost(encoder, level);
 	else if (IS_BROXTON(dev_priv))
 		bxt_ddi_vswing_sequence(dev_priv, level, port, encoder->type);
 
@@ -1637,6 +1672,10 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 			intel_dp_stop_link_train(intel_dp);
 	} else if (type == INTEL_OUTPUT_HDMI) {
 		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+		int level = intel_ddi_hdmi_level(dev_priv, port);
+
+		if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
+			skl_ddi_set_iboost(intel_encoder, level);
 
 		intel_hdmi->set_infoframes(encoder,
 					   crtc->config->has_hdmi_sink,

drivers/gpu/drm/i915/intel_display.c

@@ -3093,40 +3093,110 @@ static void intel_update_primary_planes(struct drm_device *dev)
 
 	for_each_crtc(dev, crtc) {
 		struct intel_plane *plane = to_intel_plane(crtc->primary);
-		struct intel_plane_state *plane_state;
-
-		drm_modeset_lock_crtc(crtc, &plane->base);
-		plane_state = to_intel_plane_state(plane->base.state);
+		struct intel_plane_state *plane_state =
+			to_intel_plane_state(plane->base.state);
 
 		if (plane_state->visible)
 			plane->update_plane(&plane->base,
 					    to_intel_crtc_state(crtc->state),
 					    plane_state);
+	}
+}
+
+static int
+__intel_display_resume(struct drm_device *dev,
+		       struct drm_atomic_state *state)
+{
+	struct drm_crtc_state *crtc_state;
+	struct drm_crtc *crtc;
+	int i, ret;
+
+	intel_modeset_setup_hw_state(dev);
+	i915_redisable_vga(dev);
 
-		drm_modeset_unlock_crtc(crtc);
+	if (!state)
+		return 0;
+
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		/*
+		 * Force recalculation even if we restore
+		 * current state. With fast modeset this may not result
+		 * in a modeset when the state is compatible.
+		 */
+		crtc_state->mode_changed = true;
 	}
+
+	/* ignore any reset values/BIOS leftovers in the WM registers */
+	to_intel_atomic_state(state)->skip_intermediate_wm = true;
+
+	ret = drm_atomic_commit(state);
+
+	WARN_ON(ret == -EDEADLK);
+	return ret;
 }
 
 void intel_prepare_reset(struct drm_i915_private *dev_priv)
 {
+	struct drm_device *dev = &dev_priv->drm;
+	struct drm_modeset_acquire_ctx *ctx = &dev_priv->reset_ctx;
+	struct drm_atomic_state *state;
+	int ret;
+
 	/* no reset support for gen2 */
 	if (IS_GEN2(dev_priv))
 		return;
 
-	/* reset doesn't touch the display */
+	/*
+	 * Need mode_config.mutex so that we don't
+	 * trample ongoing ->detect() and whatnot.
+	 */
+	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_acquire_init(ctx, 0);
+	while (1) {
+		ret = drm_modeset_lock_all_ctx(dev, ctx);
+		if (ret != -EDEADLK)
+			break;
+
+		drm_modeset_backoff(ctx);
+	}
+
+	/* reset doesn't touch the display, but flips might get nuked anyway, */
 	if (INTEL_GEN(dev_priv) >= 5 || IS_G4X(dev_priv))
 		return;
 
-	drm_modeset_lock_all(&dev_priv->drm);
 	/*
 	 * Disabling the crtcs gracefully seems nicer. Also the
 	 * g33 docs say we should at least disable all the planes.
 	 */
-	intel_display_suspend(&dev_priv->drm);
+	state = drm_atomic_helper_duplicate_state(dev, ctx);
+	if (IS_ERR(state)) {
+		ret = PTR_ERR(state);
+		state = NULL;
+		DRM_ERROR("Duplicating state failed with %i\n", ret);
+		goto err;
+	}
+
+	ret = drm_atomic_helper_disable_all(dev, ctx);
+	if (ret) {
+		DRM_ERROR("Suspending crtc's failed with %i\n", ret);
+		goto err;
+	}
+
+	dev_priv->modeset_restore_state = state;
+	state->acquire_ctx = ctx;
+	return;
+
+err:
+	drm_atomic_state_free(state);
 }
 
 void intel_finish_reset(struct drm_i915_private *dev_priv)
 {
+	struct drm_device *dev = &dev_priv->drm;
+	struct drm_modeset_acquire_ctx *ctx = &dev_priv->reset_ctx;
+	struct drm_atomic_state *state = dev_priv->modeset_restore_state;
+	int ret;
+
 	/*
 	 * Flips in the rings will be nuked by the reset,
 	 * so complete all pending flips so that user space
@@ -3138,6 +3208,8 @@ void intel_finish_reset(struct drm_i915_private *dev_priv)
 	if (IS_GEN2(dev_priv))
 		return;
 
+	dev_priv->modeset_restore_state = NULL;
+
 	/* reset doesn't touch the display */
 	if (INTEL_GEN(dev_priv) >= 5 || IS_G4X(dev_priv)) {
 		/*
@@ -3149,29 +3221,32 @@ void intel_finish_reset(struct drm_i915_private *dev_priv)
 		 * FIXME: Atomic will make this obsolete since we won't schedule
 		 * CS-based flips (which might get lost in gpu resets) any more.
 		 */
-		intel_update_primary_planes(&dev_priv->drm);
-		return;
-	}
-
-	/*
-	 * The display has been reset as well,
-	 * so need a full re-initialization.
-	 */
-	intel_runtime_pm_disable_interrupts(dev_priv);
-	intel_runtime_pm_enable_interrupts(dev_priv);
+		intel_update_primary_planes(dev);
+	} else {
+		/*
+		 * The display has been reset as well,
+		 * so need a full re-initialization.
+		 */
+		intel_runtime_pm_disable_interrupts(dev_priv);
+		intel_runtime_pm_enable_interrupts(dev_priv);
 
-	intel_modeset_init_hw(&dev_priv->drm);
+		intel_modeset_init_hw(dev);
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	if (dev_priv->display.hpd_irq_setup)
-		dev_priv->display.hpd_irq_setup(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+		spin_lock_irq(&dev_priv->irq_lock);
+		if (dev_priv->display.hpd_irq_setup)
+			dev_priv->display.hpd_irq_setup(dev_priv);
+		spin_unlock_irq(&dev_priv->irq_lock);
 
-	intel_display_resume(&dev_priv->drm);
+		ret = __intel_display_resume(dev, state);
+		if (ret)
+			DRM_ERROR("Restoring old state failed with %i\n", ret);
 
-	intel_hpd_init(dev_priv);
+		intel_hpd_init(dev_priv);
+	}
 
-	drm_modeset_unlock_all(&dev_priv->drm);
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+	mutex_unlock(&dev->mode_config.mutex);
 }
 
 static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
@@ -16156,9 +16231,10 @@ void intel_display_resume(struct drm_device *dev)
 	struct drm_atomic_state *state = dev_priv->modeset_restore_state;
 	struct drm_modeset_acquire_ctx ctx;
 	int ret;
-	bool setup = false;
 
 	dev_priv->modeset_restore_state = NULL;
+	if (state)
+		state->acquire_ctx = &ctx;
 
 	/*
 	 * This is a cludge because with real atomic modeset mode_config.mutex
@@ -16169,43 +16245,17 @@ void intel_display_resume(struct drm_device *dev)
 	mutex_lock(&dev->mode_config.mutex);
 	drm_modeset_acquire_init(&ctx, 0);
 
-retry:
-	ret = drm_modeset_lock_all_ctx(dev, &ctx);
-
-	if (ret == 0 && !setup) {
-		setup = true;
-
-		intel_modeset_setup_hw_state(dev);
-		i915_redisable_vga(dev);
-	}
-
-	if (ret == 0 && state) {
-		struct drm_crtc_state *crtc_state;
-		struct drm_crtc *crtc;
-		int i;
-
-		state->acquire_ctx = &ctx;
-
-		/* ignore any reset values/BIOS leftovers in the WM registers */
-		to_intel_atomic_state(state)->skip_intermediate_wm = true;
-
-		for_each_crtc_in_state(state, crtc, crtc_state, i) {
-			/*
-			 * Force recalculation even if we restore
-			 * current state. With fast modeset this may not result
-			 * in a modeset when the state is compatible.
-			 */
-			crtc_state->mode_changed = true;
-		}
-
-		ret = drm_atomic_commit(state);
-	}
+	while (1) {
+		ret = drm_modeset_lock_all_ctx(dev, &ctx);
+		if (ret != -EDEADLK)
+			break;
 
-	if (ret == -EDEADLK) {
 		drm_modeset_backoff(&ctx);
-		goto retry;
 	}
 
+	if (!ret)
+		ret = __intel_display_resume(dev, state);
+
 	drm_modeset_drop_locks(&ctx);
 	drm_modeset_acquire_fini(&ctx);
 	mutex_unlock(&dev->mode_config.mutex);

drivers/gpu/drm/i915/intel_fbc.c

@@ -1230,12 +1230,29 @@ static int intel_sanitize_fbc_option(struct drm_i915_private *dev_priv)
 	if (i915.enable_fbc >= 0)
 		return !!i915.enable_fbc;
 
+	if (!HAS_FBC(dev_priv))
+		return 0;
+
 	if (IS_BROADWELL(dev_priv))
 		return 1;
 
 	return 0;
 }
 
+static bool need_fbc_vtd_wa(struct drm_i915_private *dev_priv)
+{
+#ifdef CONFIG_INTEL_IOMMU
+	/* WaFbcTurnOffFbcWhenHyperVisorIsUsed:skl,bxt */
+	if (intel_iommu_gfx_mapped &&
+	    (IS_SKYLAKE(dev_priv) || IS_BROXTON(dev_priv))) {
+		DRM_INFO("Disabling framebuffer compression (FBC) to prevent screen flicker with VT-d enabled\n");
+		return true;
+	}
+#endif
+
+	return false;
+}
+
 /**
  * intel_fbc_init - Initialize FBC
  * @dev_priv: the i915 device
@@ -1253,6 +1270,9 @@ void intel_fbc_init(struct drm_i915_private *dev_priv)
 	fbc->active = false;
 	fbc->work.scheduled = false;
 
+	if (need_fbc_vtd_wa(dev_priv))
+		mkwrite_device_info(dev_priv)->has_fbc = false;
+
 	i915.enable_fbc = intel_sanitize_fbc_option(dev_priv);
 	DRM_DEBUG_KMS("Sanitized enable_fbc value: %d\n", i915.enable_fbc);
 

drivers/gpu/drm/i915/intel_pm.c

@@ -3344,6 +3344,8 @@ static uint32_t skl_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal,
 		plane_bytes_per_line *= 4;
 		plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 		plane_blocks_per_line /= 4;
+	} else if (tiling == DRM_FORMAT_MOD_NONE) {
+		plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512) + 1;
 	} else {
 		plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 	}
@@ -6574,9 +6576,7 @@ void intel_init_gt_powersave(struct drm_i915_private *dev_priv)
 
 void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv)
 {
-	if (IS_CHERRYVIEW(dev_priv))
-		return;
-	else if (IS_VALLEYVIEW(dev_priv))
+	if (IS_VALLEYVIEW(dev_priv))
 		valleyview_cleanup_gt_powersave(dev_priv);
 
 	if (!i915.enable_rc6)

drivers/gpu/drm/i915/intel_ringbuffer.c

@@ -1178,8 +1178,8 @@ static int bxt_init_workarounds(struct intel_engine_cs *engine)
 		I915_WRITE(GEN8_L3SQCREG1, L3_GENERAL_PRIO_CREDITS(62) |
 					   L3_HIGH_PRIO_CREDITS(2));
 
-	/* WaInsertDummyPushConstPs:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
+	/* WaToEnableHwFixForPushConstHWBug:bxt */
+	if (IS_BXT_REVID(dev_priv, BXT_REVID_C0, REVID_FOREVER))
 		WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
 				  GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION);
 
@@ -1222,8 +1222,8 @@ static int kbl_init_workarounds(struct intel_engine_cs *engine)
 		I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) |
 			   GEN8_LQSC_RO_PERF_DIS);
 
-	/* WaInsertDummyPushConstPs:kbl */
-	if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0))
+	/* WaToEnableHwFixForPushConstHWBug:kbl */
+	if (IS_KBL_REVID(dev_priv, KBL_REVID_C0, REVID_FOREVER))
 		WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2,
 				  GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION);
 

drivers/gpu/drm/mediatek/Kconfig

@@ -2,6 +2,9 @@ config DRM_MEDIATEK
 	tristate "DRM Support for Mediatek SoCs"
 	depends on DRM
 	depends on ARCH_MEDIATEK || (ARM && COMPILE_TEST)
+	depends on COMMON_CLK
+	depends on HAVE_ARM_SMCCC
+	depends on OF
 	select DRM_GEM_CMA_HELPER
 	select DRM_KMS_HELPER
 	select DRM_MIPI_DSI

drivers/gpu/drm/radeon/radeon_atpx_handler.c

@@ -198,16 +198,7 @@ static int radeon_atpx_validate(struct radeon_atpx *atpx)
 	atpx->is_hybrid = false;
 	if (valid_bits & ATPX_MS_HYBRID_GFX_SUPPORTED) {
 		printk("ATPX Hybrid Graphics\n");
-#if 1
-		/* This is a temporary hack until the D3 cold support
-		 * makes it upstream.  The ATPX power_control method seems
-		 * to still work on even if the system should be using
-		 * the new standardized hybrid D3 cold ACPI interface.
-		 */
-		atpx->functions.power_cntl = true;
-#else
 		atpx->functions.power_cntl = false;
-#endif
 		atpx->is_hybrid = true;
 	}
 

drivers/hwmon/it87.c

@@ -491,7 +491,7 @@ struct it87_sio_data {
 struct it87_data {
 	const struct attribute_group *groups[7];
 	enum chips type;
-	u16 features;
+	u32 features;
 	u8 peci_mask;
 	u8 old_peci_mask;
 

drivers/i2c/busses/i2c-at91.c

@@ -38,6 +38,7 @@
 #define AT91_I2C_TIMEOUT	msecs_to_jiffies(100)	/* transfer timeout */
 #define AT91_I2C_DMA_THRESHOLD	8			/* enable DMA if transfer size is bigger than this threshold */
 #define AUTOSUSPEND_TIMEOUT		2000
+#define AT91_I2C_MAX_ALT_CMD_DATA_SIZE	256
 
 /* AT91 TWI register definitions */
 #define	AT91_TWI_CR		0x0000	/* Control Register */
@@ -141,6 +142,7 @@ struct at91_twi_dev {
 	unsigned twi_cwgr_reg;
 	struct at91_twi_pdata *pdata;
 	bool use_dma;
+	bool use_alt_cmd;
 	bool recv_len_abort;
 	u32 fifo_size;
 	struct at91_twi_dma dma;
@@ -269,7 +271,7 @@ static void at91_twi_write_next_byte(struct at91_twi_dev *dev)
 
 	/* send stop when last byte has been written */
 	if (--dev->buf_len == 0)
-		if (!dev->pdata->has_alt_cmd)
+		if (!dev->use_alt_cmd)
 			at91_twi_write(dev, AT91_TWI_CR, AT91_TWI_STOP);
 
 	dev_dbg(dev->dev, "wrote 0x%x, to go %d\n", *dev->buf, dev->buf_len);
@@ -292,7 +294,7 @@ static void at91_twi_write_data_dma_callback(void *data)
 	 * we just have to enable TXCOMP one.
 	 */
 	at91_twi_write(dev, AT91_TWI_IER, AT91_TWI_TXCOMP);
-	if (!dev->pdata->has_alt_cmd)
+	if (!dev->use_alt_cmd)
 		at91_twi_write(dev, AT91_TWI_CR, AT91_TWI_STOP);
 }
 
@@ -410,7 +412,7 @@ static void at91_twi_read_next_byte(struct at91_twi_dev *dev)
 	}
 
 	/* send stop if second but last byte has been read */
-	if (!dev->pdata->has_alt_cmd && dev->buf_len == 1)
+	if (!dev->use_alt_cmd && dev->buf_len == 1)
 		at91_twi_write(dev, AT91_TWI_CR, AT91_TWI_STOP);
 
 	dev_dbg(dev->dev, "read 0x%x, to go %d\n", *dev->buf, dev->buf_len);
@@ -426,7 +428,7 @@ static void at91_twi_read_data_dma_callback(void *data)
 	dma_unmap_single(dev->dev, sg_dma_address(&dev->dma.sg[0]),
 			 dev->buf_len, DMA_FROM_DEVICE);
 
-	if (!dev->pdata->has_alt_cmd) {
+	if (!dev->use_alt_cmd) {
 		/* The last two bytes have to be read without using dma */
 		dev->buf += dev->buf_len - 2;
 		dev->buf_len = 2;
@@ -443,7 +445,7 @@ static void at91_twi_read_data_dma(struct at91_twi_dev *dev)
 	struct dma_chan *chan_rx = dma->chan_rx;
 	size_t buf_len;
 
-	buf_len = (dev->pdata->has_alt_cmd) ? dev->buf_len : dev->buf_len - 2;
+	buf_len = (dev->use_alt_cmd) ? dev->buf_len : dev->buf_len - 2;
 	dma->direction = DMA_FROM_DEVICE;
 
 	/* Keep in mind that we won't use dma to read the last two bytes */
@@ -651,7 +653,7 @@ static int at91_do_twi_transfer(struct at91_twi_dev *dev)
 		unsigned start_flags = AT91_TWI_START;
 
 		/* if only one byte is to be read, immediately stop transfer */
-		if (!has_alt_cmd && dev->buf_len <= 1 &&
+		if (!dev->use_alt_cmd && dev->buf_len <= 1 &&
 		    !(dev->msg->flags & I2C_M_RECV_LEN))
 			start_flags |= AT91_TWI_STOP;
 		at91_twi_write(dev, AT91_TWI_CR, start_flags);
@@ -745,7 +747,7 @@ static int at91_twi_xfer(struct i2c_adapter *adap, struct i2c_msg *msg, int num)
 	int ret;
 	unsigned int_addr_flag = 0;
 	struct i2c_msg *m_start = msg;
-	bool is_read, use_alt_cmd = false;
+	bool is_read;
 
 	dev_dbg(&adap->dev, "at91_xfer: processing %d messages:\n", num);
 
@@ -768,14 +770,16 @@ static int at91_twi_xfer(struct i2c_adapter *adap, struct i2c_msg *msg, int num)
 		at91_twi_write(dev, AT91_TWI_IADR, internal_address);
 	}
 
+	dev->use_alt_cmd = false;
 	is_read = (m_start->flags & I2C_M_RD);
 	if (dev->pdata->has_alt_cmd) {
-		if (m_start->len > 0) {
+		if (m_start->len > 0 &&
+		    m_start->len < AT91_I2C_MAX_ALT_CMD_DATA_SIZE) {
 			at91_twi_write(dev, AT91_TWI_CR, AT91_TWI_ACMEN);
 			at91_twi_write(dev, AT91_TWI_ACR,
 				       AT91_TWI_ACR_DATAL(m_start->len) |
 				       ((is_read) ? AT91_TWI_ACR_DIR : 0));
-			use_alt_cmd = true;
+			dev->use_alt_cmd = true;
 		} else {
 			at91_twi_write(dev, AT91_TWI_CR, AT91_TWI_ACMDIS);
 		}
@@ -784,7 +788,7 @@ static int at91_twi_xfer(struct i2c_adapter *adap, struct i2c_msg *msg, int num)
 	at91_twi_write(dev, AT91_TWI_MMR,
 		       (m_start->addr << 16) |
 		       int_addr_flag |
-		       ((!use_alt_cmd && is_read) ? AT91_TWI_MREAD : 0));
+		       ((!dev->use_alt_cmd && is_read) ? AT91_TWI_MREAD : 0));
 
 	dev->buf_len = m_start->len;
 	dev->buf = m_start->buf;

drivers/i2c/busses/i2c-bcm-iproc.c

@@ -158,7 +158,7 @@ static irqreturn_t bcm_iproc_i2c_isr(int irq, void *data)
 
 	if (status & BIT(IS_M_START_BUSY_SHIFT)) {
 		iproc_i2c->xfer_is_done = 1;
-		complete_all(&iproc_i2c->done);
+		complete(&iproc_i2c->done);
 	}
 
 	writel(status, iproc_i2c->base + IS_OFFSET);

drivers/i2c/busses/i2c-bcm-kona.c

@@ -229,7 +229,7 @@ static irqreturn_t bcm_kona_i2c_isr(int irq, void *devid)
 		       dev->base + TXFCR_OFFSET);
 
 	writel(status & ~ISR_RESERVED_MASK, dev->base + ISR_OFFSET);
-	complete_all(&dev->done);
+	complete(&dev->done);
 
 	return IRQ_HANDLED;
 }

drivers/i2c/busses/i2c-brcmstb.c

@@ -228,7 +228,7 @@ static irqreturn_t brcmstb_i2c_isr(int irq, void *devid)
 		return IRQ_NONE;
 
 	brcmstb_i2c_enable_disable_irq(dev, INT_DISABLE);
-	complete_all(&dev->done);
+	complete(&dev->done);
 
 	dev_dbg(dev->device, "isr handled");
 	return IRQ_HANDLED;

drivers/i2c/busses/i2c-cros-ec-tunnel.c

@@ -215,7 +215,7 @@ static int ec_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg i2c_msgs[],
 	msg->outsize = request_len;
 	msg->insize = response_len;
 
-	result = cros_ec_cmd_xfer(bus->ec, msg);
+	result = cros_ec_cmd_xfer_status(bus->ec, msg);
 	if (result < 0) {
 		dev_err(dev, "Error transferring EC i2c message %d\n", result);
 		goto exit;

drivers/i2c/busses/i2c-meson.c

@@ -211,7 +211,7 @@ static void meson_i2c_stop(struct meson_i2c *i2c)
 		meson_i2c_add_token(i2c, TOKEN_STOP);
 	} else {
 		i2c->state = STATE_IDLE;
-		complete_all(&i2c->done);
+		complete(&i2c->done);
 	}
 }
 
@@ -238,7 +238,7 @@ static irqreturn_t meson_i2c_irq(int irqno, void *dev_id)
 		dev_dbg(i2c->dev, "error bit set\n");
 		i2c->error = -ENXIO;
 		i2c->state = STATE_IDLE;
-		complete_all(&i2c->done);
+		complete(&i2c->done);
 		goto out;
 	}
 
@@ -269,7 +269,7 @@ static irqreturn_t meson_i2c_irq(int irqno, void *dev_id)
 		break;
 	case STATE_STOP:
 		i2c->state = STATE_IDLE;
-		complete_all(&i2c->done);
+		complete(&i2c->done);
 		break;
 	case STATE_IDLE:
 		break;

drivers/i2c/busses/i2c-ocores.c

@@ -379,6 +379,7 @@ static int ocores_i2c_of_probe(struct platform_device *pdev,
 			if (!clock_frequency_present) {
 				dev_err(&pdev->dev,
 					"Missing required parameter 'opencores,ip-clock-frequency'\n");
+				clk_disable_unprepare(i2c->clk);
 				return -ENODEV;
 			}
 			i2c->ip_clock_khz = clock_frequency / 1000;
@@ -467,20 +468,21 @@ static int ocores_i2c_probe(struct platform_device *pdev)
 		default:
 			dev_err(&pdev->dev, "Unsupported I/O width (%d)\n",
 				i2c->reg_io_width);
-			return -EINVAL;
+			ret = -EINVAL;
+			goto err_clk;
 		}
 	}
 
 	ret = ocores_init(&pdev->dev, i2c);
 	if (ret)
-		return ret;
+		goto err_clk;
 
 	init_waitqueue_head(&i2c->wait);
 	ret = devm_request_irq(&pdev->dev, irq, ocores_isr, 0,
 			       pdev->name, i2c);
 	if (ret) {
 		dev_err(&pdev->dev, "Cannot claim IRQ\n");
-		return ret;
+		goto err_clk;
 	}
 
 	/* hook up driver to tree */
@@ -494,7 +496,7 @@ static int ocores_i2c_probe(struct platform_device *pdev)
 	ret = i2c_add_adapter(&i2c->adap);
 	if (ret) {
 		dev_err(&pdev->dev, "Failed to add adapter\n");
-		return ret;
+		goto err_clk;
 	}
 
 	/* add in known devices to the bus */
@@ -504,6 +506,10 @@ static int ocores_i2c_probe(struct platform_device *pdev)
 	}
 
 	return 0;
+
+err_clk:
+	clk_disable_unprepare(i2c->clk);
+	return ret;
 }
 
 static int ocores_i2c_remove(struct platform_device *pdev)

drivers/i2c/muxes/i2c-demux-pinctrl.c

@@ -68,7 +68,7 @@ static int i2c_demux_activate_master(struct i2c_demux_pinctrl_priv *priv, u32 ne
 	adap = of_find_i2c_adapter_by_node(priv->chan[new_chan].parent_np);
 	if (!adap) {
 		ret = -ENODEV;
-		goto err;
+		goto err_with_revert;
 	}
 
 	p = devm_pinctrl_get_select(adap->dev.parent, priv->bus_name);
@@ -103,6 +103,8 @@ static int i2c_demux_activate_master(struct i2c_demux_pinctrl_priv *priv, u32 ne
 
  err_with_put:
 	i2c_put_adapter(adap);
+ err_with_revert:
+	of_changeset_revert(&priv->chan[new_chan].chgset);
  err:
 	dev_err(priv->dev, "failed to setup demux-adapter %d (%d)\n", new_chan, ret);
 	return ret;

drivers/md/dm-crypt.c

@@ -181,7 +181,7 @@ struct crypt_config {
 	u8 key[0];
 };
 
-#define MIN_IOS        16
+#define MIN_IOS        64
 
 static void clone_init(struct dm_crypt_io *, struct bio *);
 static void kcryptd_queue_crypt(struct dm_crypt_io *io);

drivers/md/dm-raid.c

@@ -191,7 +191,6 @@ struct raid_dev {
 #define RT_FLAG_RS_BITMAP_LOADED	2
 #define RT_FLAG_UPDATE_SBS		3
 #define RT_FLAG_RESHAPE_RS		4
-#define RT_FLAG_KEEP_RS_FROZEN		5
 
 /* Array elements of 64 bit needed for rebuild/failed disk bits */
 #define DISKS_ARRAY_ELEMS ((MAX_RAID_DEVICES + (sizeof(uint64_t) * 8 - 1)) / sizeof(uint64_t) / 8)
@@ -861,6 +860,9 @@ static int validate_region_size(struct raid_set *rs, unsigned long region_size)
 {
 	unsigned long min_region_size = rs->ti->len / (1 << 21);
 
+	if (rs_is_raid0(rs))
+		return 0;
+
 	if (!region_size) {
 		/*
 		 * Choose a reasonable default.	 All figures in sectors.
@@ -930,6 +932,8 @@ static int validate_raid_redundancy(struct raid_set *rs)
 			rebuild_cnt++;
 
 	switch (rs->raid_type->level) {
+	case 0:
+		break;
 	case 1:
 		if (rebuild_cnt >= rs->md.raid_disks)
 			goto too_many;
@@ -2335,6 +2339,13 @@ static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
 		case 0:
 			break;
 		default:
+			/*
+			 * We have to keep any raid0 data/metadata device pairs or
+			 * the MD raid0 personality will fail to start the array.
+			 */
+			if (rs_is_raid0(rs))
+				continue;
+
 			dev = container_of(rdev, struct raid_dev, rdev);
 			if (dev->meta_dev)
 				dm_put_device(ti, dev->meta_dev);
@@ -2579,7 +2590,6 @@ static int rs_prepare_reshape(struct raid_set *rs)
 		} else {
 			/* Process raid1 without delta_disks */
 			mddev->raid_disks = rs->raid_disks;
-			set_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags);
 			reshape = false;
 		}
 	} else {
@@ -2590,7 +2600,6 @@ static int rs_prepare_reshape(struct raid_set *rs)
 	if (reshape) {
 		set_bit(RT_FLAG_RESHAPE_RS, &rs->runtime_flags);
 		set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
-		set_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags);
 	} else if (mddev->raid_disks < rs->raid_disks)
 		/* Create new superblocks and bitmaps, if any new disks */
 		set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
@@ -2902,7 +2911,6 @@ static int raid_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 			goto bad;
 
 		set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
-		set_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags);
 		/* Takeover ain't recovery, so disable recovery */
 		rs_setup_recovery(rs, MaxSector);
 		rs_set_new(rs);
@@ -3386,21 +3394,28 @@ static void raid_postsuspend(struct dm_target *ti)
 {
 	struct raid_set *rs = ti->private;
 
-	if (test_and_clear_bit(RT_FLAG_RS_RESUMED, &rs->runtime_flags)) {
-		if (!rs->md.suspended)
-			mddev_suspend(&rs->md);
-		rs->md.ro = 1;
-	}
+	if (!rs->md.suspended)
+		mddev_suspend(&rs->md);
+
+	rs->md.ro = 1;
 }
 
 static void attempt_restore_of_faulty_devices(struct raid_set *rs)
 {
 	int i;
-	uint64_t failed_devices, cleared_failed_devices = 0;
+	uint64_t cleared_failed_devices[DISKS_ARRAY_ELEMS];
 	unsigned long flags;
+	bool cleared = false;
 	struct dm_raid_superblock *sb;
+	struct mddev *mddev = &rs->md;
 	struct md_rdev *r;
 
+	/* RAID personalities have to provide hot add/remove methods or we need to bail out. */
+	if (!mddev->pers || !mddev->pers->hot_add_disk || !mddev->pers->hot_remove_disk)
+		return;
+
+	memset(cleared_failed_devices, 0, sizeof(cleared_failed_devices));
+
 	for (i = 0; i < rs->md.raid_disks; i++) {
 		r = &rs->dev[i].rdev;
 		if (test_bit(Faulty, &r->flags) && r->sb_page &&
@@ -3420,7 +3435,7 @@ static void attempt_restore_of_faulty_devices(struct raid_set *rs)
 			 * ourselves.
 			 */
 			if ((r->raid_disk >= 0) &&
-			    (r->mddev->pers->hot_remove_disk(r->mddev, r) != 0))
+			    (mddev->pers->hot_remove_disk(mddev, r) != 0))
 				/* Failed to revive this device, try next */
 				continue;
 
@@ -3430,22 +3445,30 @@ static void attempt_restore_of_faulty_devices(struct raid_set *rs)
 			clear_bit(Faulty, &r->flags);
 			clear_bit(WriteErrorSeen, &r->flags);
 			clear_bit(In_sync, &r->flags);
-			if (r->mddev->pers->hot_add_disk(r->mddev, r)) {
+			if (mddev->pers->hot_add_disk(mddev, r)) {
 				r->raid_disk = -1;
 				r->saved_raid_disk = -1;
 				r->flags = flags;
 			} else {
 				r->recovery_offset = 0;
-				cleared_failed_devices |= 1 << i;
+				set_bit(i, (void *) cleared_failed_devices);
+				cleared = true;
 			}
 		}
 	}
-	if (cleared_failed_devices) {
+
+	/* If any failed devices could be cleared, update all sbs failed_devices bits */
+	if (cleared) {
+		uint64_t failed_devices[DISKS_ARRAY_ELEMS];
+
 		rdev_for_each(r, &rs->md) {
 			sb = page_address(r->sb_page);
-			failed_devices = le64_to_cpu(sb->failed_devices);
-			failed_devices &= ~cleared_failed_devices;
-			sb->failed_devices = cpu_to_le64(failed_devices);
+			sb_retrieve_failed_devices(sb, failed_devices);
+
+			for (i = 0; i < DISKS_ARRAY_ELEMS; i++)
+				failed_devices[i] &= ~cleared_failed_devices[i];
+
+			sb_update_failed_devices(sb, failed_devices);
 		}
 	}
 }
@@ -3610,26 +3633,15 @@ static void raid_resume(struct dm_target *ti)
 		 * devices are reachable again.
 		 */
 		attempt_restore_of_faulty_devices(rs);
-	} else {
-		mddev->ro = 0;
-		mddev->in_sync = 0;
+	}
 
-		/*
-		 * When passing in flags to the ctr, we expect userspace
-		 * to reset them because they made it to the superblocks
-		 * and reload the mapping anyway.
-		 *
-		 * -> only unfreeze recovery in case of a table reload or
-		 *    we'll have a bogus recovery/reshape position
-		 *    retrieved from the superblock by the ctr because
-		 *    the ongoing recovery/reshape will change it after read.
-		 */
-		if (!test_bit(RT_FLAG_KEEP_RS_FROZEN, &rs->runtime_flags))
-			clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+	mddev->ro = 0;
+	mddev->in_sync = 0;
 
-		if (mddev->suspended)
-			mddev_resume(mddev);
-	}
+	clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+
+	if (mddev->suspended)
+		mddev_resume(mddev);
 }
 
 static struct target_type raid_target = {

drivers/md/dm-round-robin.c

@@ -210,14 +210,17 @@ static struct dm_path *rr_select_path(struct path_selector *ps, size_t nr_bytes)
 	struct path_info *pi = NULL;
 	struct dm_path *current_path = NULL;
 
+	local_irq_save(flags);
 	current_path = *this_cpu_ptr(s->current_path);
 	if (current_path) {
 		percpu_counter_dec(&s->repeat_count);
-		if (percpu_counter_read_positive(&s->repeat_count) > 0)
+		if (percpu_counter_read_positive(&s->repeat_count) > 0) {
+			local_irq_restore(flags);
 			return current_path;
+		}
 	}
 
-	spin_lock_irqsave(&s->lock, flags);
+	spin_lock(&s->lock);
 	if (!list_empty(&s->valid_paths)) {
 		pi = list_entry(s->valid_paths.next, struct path_info, list);
 		list_move_tail(&pi->list, &s->valid_paths);

drivers/of/base.c

@@ -1631,8 +1631,7 @@ static int __of_parse_phandle_with_args(const struct device_node *np,
 	 */
 
  err:
-	if (it.node)
-		of_node_put(it.node);
+	of_node_put(it.node);
 	return rc;
 }
 
@@ -2343,20 +2342,13 @@ struct device_node *of_graph_get_endpoint_by_regs(
 	const struct device_node *parent, int port_reg, int reg)
 {
 	struct of_endpoint endpoint;
-	struct device_node *node, *prev_node = NULL;
-
-	while (1) {
-		node = of_graph_get_next_endpoint(parent, prev_node);
-		of_node_put(prev_node);
-		if (!node)
-			break;
+	struct device_node *node = NULL;
 
+	for_each_endpoint_of_node(parent, node) {
 		of_graph_parse_endpoint(node, &endpoint);
 		if (((port_reg == -1) || (endpoint.port == port_reg)) &&
 			((reg == -1) || (endpoint.id == reg)))
 			return node;
-
-		prev_node = node;
 	}
 
 	return NULL;

drivers/of/fdt.c

@@ -517,7 +517,7 @@ static void *__unflatten_device_tree(const void *blob,
 		pr_warning("End of tree marker overwritten: %08x\n",
 			   be32_to_cpup(mem + size));
 
-	if (detached) {
+	if (detached && mynodes) {
 		of_node_set_flag(*mynodes, OF_DETACHED);
 		pr_debug("unflattened tree is detached\n");
 	}

drivers/of/irq.c

@@ -544,12 +544,15 @@ void __init of_irq_init(const struct of_device_id *matches)
 
 			list_del(&desc->list);
 
+			of_node_set_flag(desc->dev, OF_POPULATED);
+
 			pr_debug("of_irq_init: init %s (%p), parent %p\n",
 				 desc->dev->full_name,
 				 desc->dev, desc->interrupt_parent);
 			ret = desc->irq_init_cb(desc->dev,
 						desc->interrupt_parent);
 			if (ret) {
+				of_node_clear_flag(desc->dev, OF_POPULATED);
 				kfree(desc);
 				continue;
 			}
@@ -559,8 +562,6 @@ void __init of_irq_init(const struct of_device_id *matches)
 			 * its children can get processed in a subsequent pass.
 			 */
 			list_add_tail(&desc->list, &intc_parent_list);
-
-			of_node_set_flag(desc->dev, OF_POPULATED);
 		}
 
 		/* Get the next pending parent that might have children */

drivers/of/platform.c

@@ -497,6 +497,7 @@ int of_platform_default_populate(struct device_node *root,
 }
 EXPORT_SYMBOL_GPL(of_platform_default_populate);
 
+#ifndef CONFIG_PPC
 static int __init of_platform_default_populate_init(void)
 {
 	struct device_node *node;
@@ -521,6 +522,7 @@ static int __init of_platform_default_populate_init(void)
 	return 0;
 }
 arch_initcall_sync(of_platform_default_populate_init);
+#endif
 
 static int of_platform_device_destroy(struct device *dev, void *data)
 {

drivers/scsi/aacraid/commctrl.c

@@ -63,7 +63,7 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 	struct fib *fibptr;
 	struct hw_fib * hw_fib = (struct hw_fib *)0;
 	dma_addr_t hw_fib_pa = (dma_addr_t)0LL;
-	unsigned size;
+	unsigned int size, osize;
 	int retval;
 
 	if (dev->in_reset) {
@@ -87,7 +87,8 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 	 *	will not overrun the buffer when we copy the memory. Return
 	 *	an error if we would.
 	 */
-	size = le16_to_cpu(kfib->header.Size) + sizeof(struct aac_fibhdr);
+	osize = size = le16_to_cpu(kfib->header.Size) +
+		sizeof(struct aac_fibhdr);
 	if (size < le16_to_cpu(kfib->header.SenderSize))
 		size = le16_to_cpu(kfib->header.SenderSize);
 	if (size > dev->max_fib_size) {
@@ -118,6 +119,14 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 		goto cleanup;
 	}
 
+	/* Sanity check the second copy */
+	if ((osize != le16_to_cpu(kfib->header.Size) +
+		sizeof(struct aac_fibhdr))
+		|| (size < le16_to_cpu(kfib->header.SenderSize))) {
+		retval = -EINVAL;
+		goto cleanup;
+	}
+
 	if (kfib->header.Command == cpu_to_le16(TakeABreakPt)) {
 		aac_adapter_interrupt(dev);
 		/*

drivers/scsi/fcoe/fcoe_ctlr.c

@@ -2923,7 +2923,7 @@ static int fcoe_ctlr_vlan_recv(struct fcoe_ctlr *fip, struct sk_buff *skb)
 	mutex_unlock(&fip->ctlr_mutex);
 
 drop:
-	kfree(skb);
+	kfree_skb(skb);
 	return rc;
 }
 

drivers/scsi/megaraid/megaraid_sas_base.c

@@ -5037,7 +5037,7 @@ static int megasas_init_fw(struct megasas_instance *instance)
 	/* Find first memory bar */
 	bar_list = pci_select_bars(instance->pdev, IORESOURCE_MEM);
 	instance->bar = find_first_bit(&bar_list, sizeof(unsigned long));
-	if (pci_request_selected_regions(instance->pdev, instance->bar,
+	if (pci_request_selected_regions(instance->pdev, 1<<instance->bar,
 					 "megasas: LSI")) {
 		dev_printk(KERN_DEBUG, &instance->pdev->dev, "IO memory region busy!\n");
 		return -EBUSY;
@@ -5339,7 +5339,7 @@ fail_ready_state:
 	iounmap(instance->reg_set);
 
       fail_ioremap:
-	pci_release_selected_regions(instance->pdev, instance->bar);
+	pci_release_selected_regions(instance->pdev, 1<<instance->bar);
 
 	return -EINVAL;
 }
@@ -5360,7 +5360,7 @@ static void megasas_release_mfi(struct megasas_instance *instance)
 
 	iounmap(instance->reg_set);
 
-	pci_release_selected_regions(instance->pdev, instance->bar);
+	pci_release_selected_regions(instance->pdev, 1<<instance->bar);
 }
 
 /**

drivers/scsi/megaraid/megaraid_sas_fusion.c

@@ -2603,7 +2603,7 @@ megasas_release_fusion(struct megasas_instance *instance)
 
 	iounmap(instance->reg_set);
 
-	pci_release_selected_regions(instance->pdev, instance->bar);
+	pci_release_selected_regions(instance->pdev, 1<<instance->bar);
 }
 
 /**

drivers/scsi/mpt3sas/mpt3sas_base.c

@@ -2188,6 +2188,17 @@ mpt3sas_base_map_resources(struct MPT3SAS_ADAPTER *ioc)
 	} else
 		ioc->msix96_vector = 0;
 
+	if (ioc->is_warpdrive) {
+		ioc->reply_post_host_index[0] = (resource_size_t __iomem *)
+		    &ioc->chip->ReplyPostHostIndex;
+
+		for (i = 1; i < ioc->cpu_msix_table_sz; i++)
+			ioc->reply_post_host_index[i] =
+			(resource_size_t __iomem *)
+			((u8 __iomem *)&ioc->chip->Doorbell + (0x4000 + ((i - 1)
+			* 4)));
+	}
+
 	list_for_each_entry(reply_q, &ioc->reply_queue_list, list)
 		pr_info(MPT3SAS_FMT "%s: IRQ %d\n",
 		    reply_q->name,  ((ioc->msix_enable) ? "PCI-MSI-X enabled" :
@@ -5280,17 +5291,6 @@ mpt3sas_base_attach(struct MPT3SAS_ADAPTER *ioc)
 	if (r)
 		goto out_free_resources;
 
-	if (ioc->is_warpdrive) {
-		ioc->reply_post_host_index[0] = (resource_size_t __iomem *)
-		    &ioc->chip->ReplyPostHostIndex;
-
-		for (i = 1; i < ioc->cpu_msix_table_sz; i++)
-			ioc->reply_post_host_index[i] =
-			(resource_size_t __iomem *)
-			((u8 __iomem *)&ioc->chip->Doorbell + (0x4000 + ((i - 1)
-			* 4)));
-	}
-
 	pci_set_drvdata(ioc->pdev, ioc->shost);
 	r = _base_get_ioc_facts(ioc, CAN_SLEEP);
 	if (r)

drivers/scsi/ses.c

@@ -778,6 +778,8 @@ static void ses_intf_remove_enclosure(struct scsi_device *sdev)
 	if (!edev)
 		return;
 
+	enclosure_unregister(edev);
+
 	ses_dev = edev->scratch;
 	edev->scratch = NULL;
 
@@ -789,7 +791,6 @@ static void ses_intf_remove_enclosure(struct scsi_device *sdev)
 	kfree(edev->component[0].scratch);
 
 	put_device(&edev->edev);
-	enclosure_unregister(edev);
 }
 
 static void ses_intf_remove(struct device *cdev,

drivers/usb/class/cdc-acm.c

@@ -1354,7 +1354,6 @@ made_compressed_probe:
 	spin_lock_init(&acm->write_lock);
 	spin_lock_init(&acm->read_lock);
 	mutex_init(&acm->mutex);
-	acm->rx_endpoint = usb_rcvbulkpipe(usb_dev, epread->bEndpointAddress);
 	acm->is_int_ep = usb_endpoint_xfer_int(epread);
 	if (acm->is_int_ep)
 		acm->bInterval = epread->bInterval;
@@ -1394,14 +1393,14 @@ made_compressed_probe:
 		urb->transfer_dma = rb->dma;
 		if (acm->is_int_ep) {
 			usb_fill_int_urb(urb, acm->dev,
-					 acm->rx_endpoint,
+					 usb_rcvintpipe(usb_dev, epread->bEndpointAddress),
 					 rb->base,
 					 acm->readsize,
 					 acm_read_bulk_callback, rb,
 					 acm->bInterval);
 		} else {
 			usb_fill_bulk_urb(urb, acm->dev,
-					  acm->rx_endpoint,
+					  usb_rcvbulkpipe(usb_dev, epread->bEndpointAddress),
 					  rb->base,
 					  acm->readsize,
 					  acm_read_bulk_callback, rb);

drivers/usb/class/cdc-acm.h

@@ -96,7 +96,6 @@ struct acm {
 	struct acm_rb read_buffers[ACM_NR];
 	struct acm_wb *putbuffer;			/* for acm_tty_put_char() */
 	int rx_buflimit;
-	int rx_endpoint;
 	spinlock_t read_lock;
 	int write_used;					/* number of non-empty write buffers */
 	int transmitting;

drivers/usb/core/config.c

@@ -171,6 +171,31 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno,
 							ep, buffer, size);
 }
 
+static const unsigned short low_speed_maxpacket_maxes[4] = {
+	[USB_ENDPOINT_XFER_CONTROL] = 8,
+	[USB_ENDPOINT_XFER_ISOC] = 0,
+	[USB_ENDPOINT_XFER_BULK] = 0,
+	[USB_ENDPOINT_XFER_INT] = 8,
+};
+static const unsigned short full_speed_maxpacket_maxes[4] = {
+	[USB_ENDPOINT_XFER_CONTROL] = 64,
+	[USB_ENDPOINT_XFER_ISOC] = 1023,
+	[USB_ENDPOINT_XFER_BULK] = 64,
+	[USB_ENDPOINT_XFER_INT] = 64,
+};
+static const unsigned short high_speed_maxpacket_maxes[4] = {
+	[USB_ENDPOINT_XFER_CONTROL] = 64,
+	[USB_ENDPOINT_XFER_ISOC] = 1024,
+	[USB_ENDPOINT_XFER_BULK] = 512,
+	[USB_ENDPOINT_XFER_INT] = 1023,
+};
+static const unsigned short super_speed_maxpacket_maxes[4] = {
+	[USB_ENDPOINT_XFER_CONTROL] = 512,
+	[USB_ENDPOINT_XFER_ISOC] = 1024,
+	[USB_ENDPOINT_XFER_BULK] = 1024,
+	[USB_ENDPOINT_XFER_INT] = 1024,
+};
+
 static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum,
     int asnum, struct usb_host_interface *ifp, int num_ep,
     unsigned char *buffer, int size)
@@ -179,6 +204,8 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum,
 	struct usb_endpoint_descriptor *d;
 	struct usb_host_endpoint *endpoint;
 	int n, i, j, retval;
+	unsigned int maxp;
+	const unsigned short *maxpacket_maxes;
 
 	d = (struct usb_endpoint_descriptor *) buffer;
 	buffer += d->bLength;
@@ -286,6 +313,42 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum,
 			endpoint->desc.wMaxPacketSize = cpu_to_le16(8);
 	}
 
+	/* Validate the wMaxPacketSize field */
+	maxp = usb_endpoint_maxp(&endpoint->desc);
+
+	/* Find the highest legal maxpacket size for this endpoint */
+	i = 0;		/* additional transactions per microframe */
+	switch (to_usb_device(ddev)->speed) {
+	case USB_SPEED_LOW:
+		maxpacket_maxes = low_speed_maxpacket_maxes;
+		break;
+	case USB_SPEED_FULL:
+		maxpacket_maxes = full_speed_maxpacket_maxes;
+		break;
+	case USB_SPEED_HIGH:
+		/* Bits 12..11 are allowed only for HS periodic endpoints */
+		if (usb_endpoint_xfer_int(d) || usb_endpoint_xfer_isoc(d)) {
+			i = maxp & (BIT(12) | BIT(11));
+			maxp &= ~i;
+		}
+		/* fallthrough */
+	default:
+		maxpacket_maxes = high_speed_maxpacket_maxes;
+		break;
+	case USB_SPEED_SUPER:
+	case USB_SPEED_SUPER_PLUS:
+		maxpacket_maxes = super_speed_maxpacket_maxes;
+		break;
+	}
+	j = maxpacket_maxes[usb_endpoint_type(&endpoint->desc)];
+
+	if (maxp > j) {
+		dev_warn(ddev, "config %d interface %d altsetting %d endpoint 0x%X has invalid maxpacket %d, setting to %d\n",
+		    cfgno, inum, asnum, d->bEndpointAddress, maxp, j);
+		maxp = j;
+		endpoint->desc.wMaxPacketSize = cpu_to_le16(i | maxp);
+	}
+
 	/*
 	 * Some buggy high speed devices have bulk endpoints using
 	 * maxpacket sizes other than 512.  High speed HCDs may not
@@ -293,9 +356,6 @@ static int usb_parse_endpoint(struct device *ddev, int cfgno, int inum,
 	 */
 	if (to_usb_device(ddev)->speed == USB_SPEED_HIGH
 			&& usb_endpoint_xfer_bulk(d)) {
-		unsigned maxp;
-
-		maxp = usb_endpoint_maxp(&endpoint->desc) & 0x07ff;
 		if (maxp != 512)
 			dev_warn(ddev, "config %d interface %d altsetting %d "
 				"bulk endpoint 0x%X has invalid maxpacket %d\n",

drivers/usb/core/devio.c

@@ -241,7 +241,8 @@ static int usbdev_mmap(struct file *file, struct vm_area_struct *vma)
 		goto error_decrease_mem;
 	}
 
-	mem = usb_alloc_coherent(ps->dev, size, GFP_USER, &dma_handle);
+	mem = usb_alloc_coherent(ps->dev, size, GFP_USER | __GFP_NOWARN,
+			&dma_handle);
 	if (!mem) {
 		ret = -ENOMEM;
 		goto error_free_usbm;
@@ -2582,7 +2583,9 @@ static unsigned int usbdev_poll(struct file *file,
 	if (file->f_mode & FMODE_WRITE && !list_empty(&ps->async_completed))
 		mask |= POLLOUT | POLLWRNORM;
 	if (!connected(ps))
-		mask |= POLLERR | POLLHUP;
+		mask |= POLLHUP;
+	if (list_empty(&ps->list))
+		mask |= POLLERR;
 	return mask;
 }
 

drivers/usb/core/hub.c

@@ -1052,14 +1052,11 @@ static void hub_activate(struct usb_hub *hub, enum hub_activation_type type)
 
 	/* Continue a partial initialization */
 	if (type == HUB_INIT2 || type == HUB_INIT3) {
-		device_lock(hub->intfdev);
+		device_lock(&hdev->dev);
 
 		/* Was the hub disconnected while we were waiting? */
-		if (hub->disconnected) {
-			device_unlock(hub->intfdev);
-			kref_put(&hub->kref, hub_release);
-			return;
-		}
+		if (hub->disconnected)
+			goto disconnected;
 		if (type == HUB_INIT2)
 			goto init2;
 		goto init3;
@@ -1262,7 +1259,7 @@ static void hub_activate(struct usb_hub *hub, enum hub_activation_type type)
 			queue_delayed_work(system_power_efficient_wq,
 					&hub->init_work,
 					msecs_to_jiffies(delay));
-			device_unlock(hub->intfdev);
+			device_unlock(&hdev->dev);
 			return;		/* Continues at init3: below */
 		} else {
 			msleep(delay);
@@ -1281,12 +1278,12 @@ static void hub_activate(struct usb_hub *hub, enum hub_activation_type type)
 	/* Scan all ports that need attention */
 	kick_hub_wq(hub);
 
-	/* Allow autosuspend if it was suppressed */
-	if (type <= HUB_INIT3)
+	if (type == HUB_INIT2 || type == HUB_INIT3) {
+		/* Allow autosuspend if it was suppressed */
+ disconnected:
 		usb_autopm_put_interface_async(to_usb_interface(hub->intfdev));
-
-	if (type == HUB_INIT2 || type == HUB_INIT3)
-		device_unlock(hub->intfdev);
+		device_unlock(&hdev->dev);
+	}
 
 	kref_put(&hub->kref, hub_release);
 }
@@ -1315,8 +1312,6 @@ static void hub_quiesce(struct usb_hub *hub, enum hub_quiescing_type type)
 	struct usb_device *hdev = hub->hdev;
 	int i;
 
-	cancel_delayed_work_sync(&hub->init_work);
-
 	/* hub_wq and related activity won't re-trigger */
 	hub->quiescing = 1;
 

drivers/usb/dwc3/dwc3-of-simple.c

@@ -61,6 +61,7 @@ static int dwc3_of_simple_probe(struct platform_device *pdev)
 	if (!simple->clks)
 		return -ENOMEM;
 
+	platform_set_drvdata(pdev, simple);
 	simple->dev = dev;
 
 	for (i = 0; i < simple->num_clocks; i++) {

drivers/usb/dwc3/dwc3-pci.c

@@ -37,6 +37,7 @@
 #define PCI_DEVICE_ID_INTEL_BXT			0x0aaa
 #define PCI_DEVICE_ID_INTEL_BXT_M		0x1aaa
 #define PCI_DEVICE_ID_INTEL_APL			0x5aaa
+#define PCI_DEVICE_ID_INTEL_KBP			0xa2b0
 
 static const struct acpi_gpio_params reset_gpios = { 0, 0, false };
 static const struct acpi_gpio_params cs_gpios = { 1, 0, false };
@@ -227,6 +228,7 @@ static const struct pci_device_id dwc3_pci_id_table[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BXT), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BXT_M), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_APL), },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBP), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_NL_USB), },
 	{  }	/* Terminating Entry */
 };

drivers/usb/dwc3/gadget.c

@@ -829,7 +829,7 @@ static void dwc3_prepare_one_trb(struct dwc3_ep *dep,
 	if (!req->request.no_interrupt && !chain)
 		trb->ctrl |= DWC3_TRB_CTRL_IOC | DWC3_TRB_CTRL_ISP_IMI;
 
-	if (last)
+	if (last && !usb_endpoint_xfer_isoc(dep->endpoint.desc))
 		trb->ctrl |= DWC3_TRB_CTRL_LST;
 
 	if (chain)
@@ -1955,7 +1955,8 @@ static void dwc3_gadget_free_endpoints(struct dwc3 *dwc)
 
 static int __dwc3_cleanup_done_trbs(struct dwc3 *dwc, struct dwc3_ep *dep,
 		struct dwc3_request *req, struct dwc3_trb *trb,
-		const struct dwc3_event_depevt *event, int status)
+		const struct dwc3_event_depevt *event, int status,
+		int chain)
 {
 	unsigned int		count;
 	unsigned int		s_pkt = 0;
@@ -1964,17 +1965,22 @@ static int __dwc3_cleanup_done_trbs(struct dwc3 *dwc, struct dwc3_ep *dep,
 	dep->queued_requests--;
 	trace_dwc3_complete_trb(dep, trb);
 
+	/*
+	 * If we're in the middle of series of chained TRBs and we
+	 * receive a short transfer along the way, DWC3 will skip
+	 * through all TRBs including the last TRB in the chain (the
+	 * where CHN bit is zero. DWC3 will also avoid clearing HWO
+	 * bit and SW has to do it manually.
+	 *
+	 * We're going to do that here to avoid problems of HW trying
+	 * to use bogus TRBs for transfers.
+	 */
+	if (chain && (trb->ctrl & DWC3_TRB_CTRL_HWO))
+		trb->ctrl &= ~DWC3_TRB_CTRL_HWO;
+
 	if ((trb->ctrl & DWC3_TRB_CTRL_HWO) && status != -ESHUTDOWN)
-		/*
-		 * We continue despite the error. There is not much we
-		 * can do. If we don't clean it up we loop forever. If
-		 * we skip the TRB then it gets overwritten after a
-		 * while since we use them in a ring buffer. A BUG()
-		 * would help. Lets hope that if this occurs, someone
-		 * fixes the root cause instead of looking away :)
-		 */
-		dev_err(dwc->dev, "%s's TRB (%p) still owned by HW\n",
-				dep->name, trb);
+		return 1;
+
 	count = trb->size & DWC3_TRB_SIZE_MASK;
 
 	if (dep->direction) {
@@ -2013,15 +2019,7 @@ static int __dwc3_cleanup_done_trbs(struct dwc3 *dwc, struct dwc3_ep *dep,
 			s_pkt = 1;
 	}
 
-	/*
-	 * We assume here we will always receive the entire data block
-	 * which we should receive. Meaning, if we program RX to
-	 * receive 4K but we receive only 2K, we assume that's all we
-	 * should receive and we simply bounce the request back to the
-	 * gadget driver for further processing.
-	 */
-	req->request.actual += req->request.length - count;
-	if (s_pkt)
+	if (s_pkt && !chain)
 		return 1;
 	if ((event->status & DEPEVT_STATUS_LST) &&
 			(trb->ctrl & (DWC3_TRB_CTRL_LST |
@@ -2040,13 +2038,17 @@ static int dwc3_cleanup_done_reqs(struct dwc3 *dwc, struct dwc3_ep *dep,
 	struct dwc3_trb		*trb;
 	unsigned int		slot;
 	unsigned int		i;
+	int			count = 0;
 	int			ret;
 
 	do {
+		int chain;
+
 		req = next_request(&dep->started_list);
 		if (WARN_ON_ONCE(!req))
 			return 1;
 
+		chain = req->request.num_mapped_sgs > 0;
 		i = 0;
 		do {
 			slot = req->first_trb_index + i;
@@ -2054,13 +2056,22 @@ static int dwc3_cleanup_done_reqs(struct dwc3 *dwc, struct dwc3_ep *dep,
 				slot++;
 			slot %= DWC3_TRB_NUM;
 			trb = &dep->trb_pool[slot];
+			count += trb->size & DWC3_TRB_SIZE_MASK;
 
 			ret = __dwc3_cleanup_done_trbs(dwc, dep, req, trb,
-					event, status);
+					event, status, chain);
 			if (ret)
 				break;
 		} while (++i < req->request.num_mapped_sgs);
 
+		/*
+		 * We assume here we will always receive the entire data block
+		 * which we should receive. Meaning, if we program RX to
+		 * receive 4K but we receive only 2K, we assume that's all we
+		 * should receive and we simply bounce the request back to the
+		 * gadget driver for further processing.
+		 */
+		req->request.actual += req->request.length - count;
 		dwc3_gadget_giveback(dep, req, status);
 
 		if (ret)

drivers/usb/gadget/composite.c

@@ -1913,6 +1913,8 @@ unknown:
 			break;
 
 		case USB_RECIP_ENDPOINT:
+			if (!cdev->config)
+				break;
 			endp = ((w_index & 0x80) >> 3) | (w_index & 0x0f);
 			list_for_each_entry(f, &cdev->config->functions, list) {
 				if (test_bit(endp, f->endpoints))
@@ -2124,14 +2126,14 @@ int composite_os_desc_req_prepare(struct usb_composite_dev *cdev,
 
 	cdev->os_desc_req = usb_ep_alloc_request(ep0, GFP_KERNEL);
 	if (!cdev->os_desc_req) {
-		ret = PTR_ERR(cdev->os_desc_req);
+		ret = -ENOMEM;
 		goto end;
 	}
 
 	/* OS feature descriptor length <= 4kB */
 	cdev->os_desc_req->buf = kmalloc(4096, GFP_KERNEL);
 	if (!cdev->os_desc_req->buf) {
-		ret = PTR_ERR(cdev->os_desc_req->buf);
+		ret = -ENOMEM;
 		kfree(cdev->os_desc_req);
 		goto end;
 	}

drivers/usb/gadget/configfs.c

@@ -1490,7 +1490,9 @@ void unregister_gadget_item(struct config_item *item)
 {
 	struct gadget_info *gi = to_gadget_info(item);
 
+	mutex_lock(&gi->lock);
 	unregister_gadget(gi);
+	mutex_unlock(&gi->lock);
 }
 EXPORT_SYMBOL_GPL(unregister_gadget_item);
 

drivers/usb/gadget/function/rndis.c

@@ -680,6 +680,12 @@ static int rndis_reset_response(struct rndis_params *params,
 {
 	rndis_reset_cmplt_type *resp;
 	rndis_resp_t *r;
+	u8 *xbuf;
+	u32 length;
+
+	/* drain the response queue */
+	while ((xbuf = rndis_get_next_response(params, &length)))
+		rndis_free_response(params, xbuf);
 
 	r = rndis_add_response(params, sizeof(rndis_reset_cmplt_type));
 	if (!r)

drivers/usb/gadget/function/u_ether.c

@@ -556,7 +556,8 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb,
 			/* Multi frame CDC protocols may store the frame for
 			 * later which is not a dropped frame.
 			 */
-			if (dev->port_usb->supports_multi_frame)
+			if (dev->port_usb &&
+					dev->port_usb->supports_multi_frame)
 				goto multiframe;
 			goto drop;
 		}

drivers/usb/gadget/function/uvc_configfs.c

@@ -2023,7 +2023,7 @@ static int uvcg_streaming_class_allow_link(struct config_item *src,
 	if (!data) {
 		kfree(*class_array);
 		*class_array = NULL;
-		ret = PTR_ERR(data);
+		ret = -ENOMEM;
 		goto unlock;
 	}
 	cl_arr = *class_array;

drivers/usb/gadget/legacy/inode.c

@@ -542,7 +542,7 @@ static ssize_t ep_aio(struct kiocb *iocb,
 	 */
 	spin_lock_irq(&epdata->dev->lock);
 	value = -ENODEV;
-	if (unlikely(epdata->ep))
+	if (unlikely(epdata->ep == NULL))
 		goto fail;
 
 	req = usb_ep_alloc_request(epdata->ep, GFP_ATOMIC);
@@ -606,7 +606,7 @@ ep_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	}
 	if (is_sync_kiocb(iocb)) {
 		value = ep_io(epdata, buf, len);
-		if (value >= 0 && copy_to_iter(buf, value, to))
+		if (value >= 0 && (copy_to_iter(buf, value, to) != value))
 			value = -EFAULT;
 	} else {
 		struct kiocb_priv *priv = kzalloc(sizeof *priv, GFP_KERNEL);

drivers/usb/gadget/udc/core.c

@@ -1145,7 +1145,7 @@ int usb_add_gadget_udc_release(struct device *parent, struct usb_gadget *gadget,
 			if (ret != -EPROBE_DEFER)
 				list_del(&driver->pending);
 			if (ret)
-				goto err4;
+				goto err5;
 			break;
 		}
 	}
@@ -1154,6 +1154,9 @@ int usb_add_gadget_udc_release(struct device *parent, struct usb_gadget *gadget,
 
 	return 0;
 
+err5:
+	device_del(&udc->dev);
+
 err4:
 	list_del(&udc->list);
 	mutex_unlock(&udc_lock);

drivers/usb/gadget/udc/fsl_qe_udc.c

@@ -2053,7 +2053,7 @@ static void setup_received_handle(struct qe_udc *udc,
 			struct qe_ep *ep;
 
 			if (wValue != 0 || wLength != 0
-				|| pipe > USB_MAX_ENDPOINTS)
+				|| pipe >= USB_MAX_ENDPOINTS)
 				break;
 			ep = &udc->eps[pipe];
 

drivers/usb/host/ehci-hcd.c

@@ -332,11 +332,11 @@ static void ehci_turn_off_all_ports(struct ehci_hcd *ehci)
 	int	port = HCS_N_PORTS(ehci->hcs_params);
 
 	while (port--) {
-		ehci_writel(ehci, PORT_RWC_BITS,
-				&ehci->regs->port_status[port]);
 		spin_unlock_irq(&ehci->lock);
 		ehci_port_power(ehci, port, false);
 		spin_lock_irq(&ehci->lock);
+		ehci_writel(ehci, PORT_RWC_BITS,
+				&ehci->regs->port_status[port]);
 	}
 }
 

drivers/usb/host/max3421-hcd.c

@@ -1675,7 +1675,7 @@ max3421_gpout_set_value(struct usb_hcd *hcd, u8 pin_number, u8 value)
 	if (pin_number > 7)
 		return;
 
-	mask = 1u << pin_number;
+	mask = 1u << (pin_number % 4);
 	idx = pin_number / 4;
 
 	if (value)

drivers/usb/host/xhci-hub.c

@@ -386,6 +386,9 @@ static int xhci_stop_device(struct xhci_hcd *xhci, int slot_id, int suspend)
 
 	ret = 0;
 	virt_dev = xhci->devs[slot_id];
+	if (!virt_dev)
+		return -ENODEV;
+
 	cmd = xhci_alloc_command(xhci, false, true, GFP_NOIO);
 	if (!cmd) {
 		xhci_dbg(xhci, "Couldn't allocate command structure.\n");

drivers/usb/host/xhci-pci.c

@@ -314,11 +314,12 @@ static void xhci_pci_remove(struct pci_dev *dev)
 		usb_remove_hcd(xhci->shared_hcd);
 		usb_put_hcd(xhci->shared_hcd);
 	}
-	usb_hcd_pci_remove(dev);
 
 	/* Workaround for spurious wakeups at shutdown with HSW */
 	if (xhci->quirks & XHCI_SPURIOUS_WAKEUP)
 		pci_set_power_state(dev, PCI_D3hot);
+
+	usb_hcd_pci_remove(dev);
 }
 
 #ifdef CONFIG_PM

drivers/usb/host/xhci-ring.c

@@ -1334,12 +1334,6 @@ static void handle_cmd_completion(struct xhci_hcd *xhci,
 
 	cmd = list_entry(xhci->cmd_list.next, struct xhci_command, cmd_list);
 
-	if (cmd->command_trb != xhci->cmd_ring->dequeue) {
-		xhci_err(xhci,
-			 "Command completion event does not match command\n");
-		return;
-	}
-
 	del_timer(&xhci->cmd_timer);
 
 	trace_xhci_cmd_completion(cmd_trb, (struct xhci_generic_trb *) event);
@@ -1351,6 +1345,13 @@ static void handle_cmd_completion(struct xhci_hcd *xhci,
 		xhci_handle_stopped_cmd_ring(xhci, cmd);
 		return;
 	}
+
+	if (cmd->command_trb != xhci->cmd_ring->dequeue) {
+		xhci_err(xhci,
+			 "Command completion event does not match command\n");
+		return;
+	}
+
 	/*
 	 * Host aborted the command ring, check if the current command was
 	 * supposed to be aborted, otherwise continue normally.
@@ -3243,7 +3244,8 @@ int xhci_queue_bulk_tx(struct xhci_hcd *xhci, gfp_t mem_flags,
 	send_addr = addr;
 
 	/* Queue the TRBs, even if they are zero-length */
-	for (enqd_len = 0; enqd_len < full_len; enqd_len += trb_buff_len) {
+	for (enqd_len = 0; first_trb || enqd_len < full_len;
+			enqd_len += trb_buff_len) {
 		field = TRB_TYPE(TRB_NORMAL);
 
 		/* TRB buffer should not cross 64KB boundaries */

drivers/usb/misc/ftdi-elan.c

@@ -665,7 +665,7 @@ static ssize_t ftdi_elan_read(struct file *file, char __user *buffer,
 {
 	char data[30 *3 + 4];
 	char *d = data;
-	int m = (sizeof(data) - 1) / 3;
+	int m = (sizeof(data) - 1) / 3 - 1;
 	int bytes_read = 0;
 	int retry_on_empty = 10;
 	int retry_on_timeout = 5;
@@ -1684,7 +1684,7 @@ wait:if (ftdi->disconnected > 0) {
 			int i = 0;
 			char data[30 *3 + 4];
 			char *d = data;
-			int m = (sizeof(data) - 1) / 3;
+			int m = (sizeof(data) - 1) / 3 - 1;
 			int l = 0;
 			struct u132_target *target = &ftdi->target[ed];
 			struct u132_command *command = &ftdi->command[
@@ -1876,7 +1876,7 @@ more:{
 		if (packet_bytes > 2) {
 			char diag[30 *3 + 4];
 			char *d = diag;
-			int m = (sizeof(diag) - 1) / 3;
+			int m = (sizeof(diag) - 1) / 3 - 1;
 			char *b = ftdi->bulk_in_buffer;
 			int bytes_read = 0;
 			diag[0] = 0;
@@ -2053,7 +2053,7 @@ static int ftdi_elan_synchronize(struct usb_ftdi *ftdi)
 			if (packet_bytes > 2) {
 				char diag[30 *3 + 4];
 				char *d = diag;
-				int m = (sizeof(diag) - 1) / 3;
+				int m = (sizeof(diag) - 1) / 3 - 1;
 				char *b = ftdi->bulk_in_buffer;
 				int bytes_read = 0;
 				unsigned char c = 0;
@@ -2155,7 +2155,7 @@ more:{
 		if (packet_bytes > 2) {
 			char diag[30 *3 + 4];
 			char *d = diag;
-			int m = (sizeof(diag) - 1) / 3;
+			int m = (sizeof(diag) - 1) / 3 - 1;
 			char *b = ftdi->bulk_in_buffer;
 			int bytes_read = 0;
 			diag[0] = 0;