Merge 4.1-rc7 into staging-testing

We want the staging tree fixes in here too to help with testing and
merge issues.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Documentation/ABI/testing/sysfs-devices-system-cpu

@@ -162,7 +162,7 @@ Description:	Discover CPUs in the same CPU frequency coordination domain
 What:		/sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
 Date:		August 2008
 KernelVersion:	2.6.27
-Contact:	discuss@x86-64.org
+Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
 Description:	Disable L3 cache indices
 
 		These files exist in every CPU's cache/index3 directory. Each

Documentation/devicetree/bindings/clock/silabs,si5351.txt

@@ -17,7 +17,8 @@ Required properties:
 - #clock-cells: from common clock binding; shall be set to 1.
 - clocks: from common clock binding; list of parent clock
   handles, shall be xtal reference clock or xtal and clkin for
-  si5351c only.
+  si5351c only. Corresponding clock input names are "xtal" and
+  "clkin" respectively.
 - #address-cells: shall be set to 1.
 - #size-cells: shall be set to 0.
 
@@ -71,6 +72,7 @@ i2c-master-node {
 
 		/* connect xtal input to 25MHz reference */
 		clocks = <&ref25>;
+		clock-names = "xtal";
 
 		/* connect xtal input as source of pll0 and pll1 */
 		silabs,pll-source = <0 0>, <1 0>;

Documentation/devicetree/bindings/net/cdns-emac.txt

@@ -3,7 +3,8 @@
 Required properties:
 - compatible: Should be "cdns,[<chip>-]{emac}"
   Use "cdns,at91rm9200-emac" Atmel at91rm9200 SoC.
-  or the generic form: "cdns,emac".
+  Use "cdns,zynq-gem" Xilinx Zynq-7xxx SoC.
+  Or the generic form: "cdns,emac".
 - reg: Address and length of the register set for the device
 - interrupts: Should contain macb interrupt
 - phy-mode: see ethernet.txt file in the same directory.

Documentation/devicetree/bindings/usb/renesas_usbhs.txt

@@ -15,10 +15,8 @@ Optional properties:
   - phys: phandle + phy specifier pair
   - phy-names: must be "usb"
   - dmas: Must contain a list of references to DMA specifiers.
-  - dma-names : Must contain a list of DMA names:
+  - dma-names : named "ch%d", where %d is the channel number ranging from zero
-   - tx0 ... tx<n>
+                to the number of channels (DnFIFOs) minus one.
-   - rx0 ... rx<n>
-    - This <n> means DnFIFO in USBHS module.
 
 Example:
 	usbhs: usb@e6590000 {

Documentation/hwmon/tmp401

@@ -20,7 +20,7 @@ Supported chips:
     Datasheet: http://focus.ti.com/docs/prod/folders/print/tmp432.html
   * Texas Instruments TMP435
     Prefix: 'tmp435'
-    Addresses scanned: I2C 0x37, 0x48 - 0x4f
+    Addresses scanned: I2C 0x48 - 0x4f
     Datasheet: http://focus.ti.com/docs/prod/folders/print/tmp435.html
 
 Authors:

Documentation/target/tcmu-design.txt

@@ -15,8 +15,7 @@ Contents:
   a) Discovering and configuring TCMU uio devices
   b) Waiting for events on the device(s)
   c) Managing the command ring
-3) Command filtering and pass_level
+3) A final note
-4) A final note
 
 
 TCM Userspace Design
@@ -324,7 +323,7 @@ int handle_device_events(int fd, void *map)
   /* Process events from cmd ring until we catch up with cmd_head */
   while (ent != (void *)mb + mb->cmdr_off + mb->cmd_head) {
 
-    if (tcmu_hdr_get_op(&ent->hdr) == TCMU_OP_CMD) {
+    if (tcmu_hdr_get_op(ent->hdr.len_op) == TCMU_OP_CMD) {
       uint8_t *cdb = (void *)mb + ent->req.cdb_off;
       bool success = true;
 
@@ -339,8 +338,12 @@ int handle_device_events(int fd, void *map)
         ent->rsp.scsi_status = SCSI_CHECK_CONDITION;
       }
     }
+    else if (tcmu_hdr_get_op(ent->hdr.len_op) != TCMU_OP_PAD) {
+      /* Tell the kernel we didn't handle unknown opcodes */
+      ent->hdr.uflags |= TCMU_UFLAG_UNKNOWN_OP;
+    }
     else {
-      /* Do nothing for PAD entries */
+      /* Do nothing for PAD entries except update cmd_tail */
     }
 
     /* update cmd_tail */
@@ -360,28 +363,6 @@ int handle_device_events(int fd, void *map)
 }
 
 
-Command filtering and pass_level
---------------------------------
-
-TCMU supports a "pass_level" option with valid values of 0 or 1.  When
-the value is 0 (the default), nearly all SCSI commands received for
-the device are passed through to the handler. This allows maximum
-flexibility but increases the amount of code required by the handler,
-to support all mandatory SCSI commands. If pass_level is set to 1,
-then only IO-related commands are presented, and the rest are handled
-by LIO's in-kernel command emulation. The commands presented at level
-1 include all versions of:
-
-READ
-WRITE
-WRITE_VERIFY
-XDWRITEREAD
-WRITE_SAME
-COMPARE_AND_WRITE
-SYNCHRONIZE_CACHE
-UNMAP
-
-
 A final note
 ------------
 

Documentation/virtual/kvm/mmu.txt

@@ -169,6 +169,10 @@ Shadow pages contain the following information:
     Contains the value of cr4.smep && !cr0.wp for which the page is valid
     (pages for which this is true are different from other pages; see the
     treatment of cr0.wp=0 below).
+  role.smap_andnot_wp:
+    Contains the value of cr4.smap && !cr0.wp for which the page is valid
+    (pages for which this is true are different from other pages; see the
+    treatment of cr0.wp=0 below).
   gfn:
     Either the guest page table containing the translations shadowed by this
     page, or the base page frame for linear translations.  See role.direct.
@@ -344,10 +348,16 @@ on fault type:
 
 (user write faults generate a #PF)
 
-In the first case there is an additional complication if CR4.SMEP is
+In the first case there are two additional complications:
-enabled: since we've turned the page into a kernel page, the kernel may now
+- if CR4.SMEP is enabled: since we've turned the page into a kernel page,
-execute it.  We handle this by also setting spte.nx.  If we get a user
+  the kernel may now execute it.  We handle this by also setting spte.nx.
-fetch or read fault, we'll change spte.u=1 and spte.nx=gpte.nx back.
+  If we get a user fetch or read fault, we'll change spte.u=1 and
+  spte.nx=gpte.nx back.
+- if CR4.SMAP is disabled: since the page has been changed to a kernel
+  page, it can not be reused when CR4.SMAP is enabled. We set
+  CR4.SMAP && !CR0.WP into shadow page's role to avoid this case. Note,
+  here we do not care the case that CR4.SMAP is enabled since KVM will
+  directly inject #PF to guest due to failed permission check.
 
 To prevent an spte that was converted into a kernel page with cr0.wp=0
 from being written by the kernel after cr0.wp has changed to 1, we make

MAINTAINERS

@@ -51,9 +51,9 @@ trivial patch so apply some common sense.
 	or does something very odd once a month document it.
 
 	PLEASE remember that submissions must be made under the terms
-	of the OSDL certificate of contribution and should include a
+	of the Linux Foundation certificate of contribution and should
-	Signed-off-by: line.  The current version of this "Developer's
+	include a Signed-off-by: line.  The current version of this
-	Certificate of Origin" (DCO) is listed in the file
+	"Developer's Certificate of Origin" (DCO) is listed in the file
 	Documentation/SubmittingPatches.
 
 6.	Make sure you have the right to send any changes you make. If you
@@ -2427,7 +2427,6 @@ L:	linux-security-module@vger.kernel.org
 S:	Supported
 F:	include/linux/capability.h
 F:	include/uapi/linux/capability.h
-F:	security/capability.c
 F:	security/commoncap.c
 F:	kernel/capability.c
 
@@ -3825,10 +3824,11 @@ M:	David Woodhouse <dwmw2@infradead.org>
 L:	linux-embedded@vger.kernel.org
 S:	Maintained
 
-EMULEX LPFC FC SCSI DRIVER
+EMULEX/AVAGO LPFC FC/FCOE SCSI DRIVER
-M:	James Smart <james.smart@emulex.com>
+M:	James Smart <james.smart@avagotech.com>
+M:	Dick Kennedy <dick.kennedy@avagotech.com>
 L:	linux-scsi@vger.kernel.org
-W:	http://sourceforge.net/projects/lpfcxxxx
+W:	http://www.avagotech.com
 S:	Supported
 F:	drivers/scsi/lpfc/
 
@@ -4536,7 +4536,7 @@ M:	Jean Delvare <jdelvare@suse.de>
 M:	Guenter Roeck <linux@roeck-us.net>
 L:	lm-sensors@lm-sensors.org
 W:	http://www.lm-sensors.org/
-T:	quilt kernel.org/pub/linux/kernel/people/jdelvare/linux-2.6/jdelvare-hwmon/
+T:	quilt http://jdelvare.nerim.net/devel/linux/jdelvare-hwmon/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging.git
 S:	Maintained
 F:	Documentation/hwmon/
@@ -7575,6 +7575,7 @@ F:	drivers/pci/host/pci-exynos.c
 
 PCI DRIVER FOR SYNOPSIS DESIGNWARE
 M:	Jingoo Han <jingoohan1@gmail.com>
+M:	Pratyush Anand <pratyush.anand@gmail.com>
 L:	linux-pci@vger.kernel.org
 S:	Maintained
 F:	drivers/pci/host/*designware*
@@ -7588,8 +7589,9 @@ F:	Documentation/devicetree/bindings/pci/host-generic-pci.txt
 F:	drivers/pci/host/pci-host-generic.c
 
 PCIE DRIVER FOR ST SPEAR13XX
+M:	Pratyush Anand <pratyush.anand@gmail.com>
 L:	linux-pci@vger.kernel.org
-S:	Orphan
+S:	Maintained
 F:	drivers/pci/host/*spear*
 
 PCMCIA SUBSYSTEM
@@ -8829,9 +8831,11 @@ F:	drivers/misc/phantom.c
 F:	include/uapi/linux/phantom.h
 
 SERVER ENGINES 10Gbps iSCSI - BladeEngine 2 DRIVER
-M:	Jayamohan Kallickal <jayamohan.kallickal@emulex.com>
+M:	Jayamohan Kallickal <jayamohan.kallickal@avagotech.com>
+M:	Minh Tran <minh.tran@avagotech.com>
+M:	John Soni Jose <sony.john-n@avagotech.com>
 L:	linux-scsi@vger.kernel.org
-W:	http://www.emulex.com
+W:	http://www.avagotech.com
 S:	Supported
 F:	drivers/scsi/be2iscsi/
 
@@ -10594,8 +10598,7 @@ F:	drivers/virtio/virtio_input.c
 F:	include/uapi/linux/virtio_input.h
 
 VIA RHINE NETWORK DRIVER
-M:	Roger Luethi <rl@hellgate.ch>
+S:	Orphan
-S:	Maintained
 F:	drivers/net/ethernet/via/via-rhine.c
 
 VIA SD/MMC CARD CONTROLLER DRIVER

Makefile

@@ -1,7 +1,7 @@
 VERSION = 4
 PATCHLEVEL = 1
 SUBLEVEL = 0
-EXTRAVERSION = -rc4
+EXTRAVERSION = -rc7
 NAME = Hurr durr I'ma sheep
 
 # *DOCUMENTATION*

arch/alpha/boot/Makefile

@@ -14,6 +14,9 @@ targets		:= vmlinux.gz vmlinux \
 		   tools/bootpzh bootloader bootpheader bootpzheader 
 OBJSTRIP	:= $(obj)/tools/objstrip
 
+HOSTCFLAGS	:= -Wall -I$(objtree)/usr/include
+BOOTCFLAGS	+= -I$(obj) -I$(srctree)/$(obj)
+
 # SRM bootable image.  Copy to offset 512 of a partition.
 $(obj)/bootimage: $(addprefix $(obj)/tools/,mkbb lxboot bootlx) $(obj)/vmlinux.nh
 	( cat $(obj)/tools/lxboot $(obj)/tools/bootlx $(obj)/vmlinux.nh ) > $@ 
@@ -96,13 +99,14 @@ $(obj)/tools/bootph: $(obj)/bootpheader $(OBJSTRIP) FORCE
 $(obj)/tools/bootpzh: $(obj)/bootpzheader $(OBJSTRIP) FORCE
 	$(call if_changed,objstrip)
 
-LDFLAGS_bootloader   := -static -uvsprintf -T  #-N -relax
+LDFLAGS_bootloader   := -static -T # -N -relax
-LDFLAGS_bootpheader  := -static -uvsprintf -T  #-N -relax
+LDFLAGS_bootloader   := -static -T # -N -relax
-LDFLAGS_bootpzheader := -static -uvsprintf -T  #-N -relax
+LDFLAGS_bootpheader  := -static -T # -N -relax
+LDFLAGS_bootpzheader := -static -T # -N -relax
 
-OBJ_bootlx   := $(obj)/head.o $(obj)/main.o
+OBJ_bootlx   := $(obj)/head.o $(obj)/stdio.o $(obj)/main.o
-OBJ_bootph   := $(obj)/head.o $(obj)/bootp.o
+OBJ_bootph   := $(obj)/head.o $(obj)/stdio.o $(obj)/bootp.o
-OBJ_bootpzh  := $(obj)/head.o $(obj)/bootpz.o $(obj)/misc.o
+OBJ_bootpzh  := $(obj)/head.o $(obj)/stdio.o $(obj)/bootpz.o $(obj)/misc.o
 
 $(obj)/bootloader: $(obj)/bootloader.lds $(OBJ_bootlx) $(LIBS_Y) FORCE
 	$(call if_changed,ld)

arch/alpha/boot/main.c

@@ -19,7 +19,6 @@
 
 #include "ksize.h"
 
-extern int vsprintf(char *, const char *, va_list);
 extern unsigned long switch_to_osf_pal(unsigned long nr,
 	struct pcb_struct * pcb_va, struct pcb_struct * pcb_pa,
 	unsigned long *vptb);

arch/alpha/boot/stdio.c

@@ -0,0 +1,306 @@
+/*
+ * Copyright (C) Paul Mackerras 1997.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <stdarg.h>
+#include <stddef.h>
+
+size_t strnlen(const char * s, size_t count)
+{
+	const char *sc;
+
+	for (sc = s; count-- && *sc != '\0'; ++sc)
+		/* nothing */;
+	return sc - s;
+}
+
+# define do_div(n, base) ({						\
+	unsigned int __base = (base);					\
+	unsigned int __rem;						\
+	__rem = ((unsigned long long)(n)) % __base;			\
+	(n) = ((unsigned long long)(n)) / __base;			\
+	__rem;								\
+})
+
+
+static int skip_atoi(const char **s)
+{
+	int i, c;
+
+	for (i = 0; '0' <= (c = **s) && c <= '9'; ++*s)
+		i = i*10 + c - '0';
+	return i;
+}
+
+#define ZEROPAD	1		/* pad with zero */
+#define SIGN	2		/* unsigned/signed long */
+#define PLUS	4		/* show plus */
+#define SPACE	8		/* space if plus */
+#define LEFT	16		/* left justified */
+#define SPECIAL	32		/* 0x */
+#define LARGE	64		/* use 'ABCDEF' instead of 'abcdef' */
+
+static char * number(char * str, unsigned long long num, int base, int size, int precision, int type)
+{
+	char c,sign,tmp[66];
+	const char *digits="0123456789abcdefghijklmnopqrstuvwxyz";
+	int i;
+
+	if (type & LARGE)
+		digits = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
+	if (type & LEFT)
+		type &= ~ZEROPAD;
+	if (base < 2 || base > 36)
+		return 0;
+	c = (type & ZEROPAD) ? '0' : ' ';
+	sign = 0;
+	if (type & SIGN) {
+		if ((signed long long)num < 0) {
+			sign = '-';
+			num = - (signed long long)num;
+			size--;
+		} else if (type & PLUS) {
+			sign = '+';
+			size--;
+		} else if (type & SPACE) {
+			sign = ' ';
+			size--;
+		}
+	}
+	if (type & SPECIAL) {
+		if (base == 16)
+			size -= 2;
+		else if (base == 8)
+			size--;
+	}
+	i = 0;
+	if (num == 0)
+		tmp[i++]='0';
+	else while (num != 0) {
+		tmp[i++] = digits[do_div(num, base)];
+	}
+	if (i > precision)
+		precision = i;
+	size -= precision;
+	if (!(type&(ZEROPAD+LEFT)))
+		while(size-->0)
+			*str++ = ' ';
+	if (sign)
+		*str++ = sign;
+	if (type & SPECIAL) {
+		if (base==8)
+			*str++ = '0';
+		else if (base==16) {
+			*str++ = '0';
+			*str++ = digits[33];
+		}
+	}
+	if (!(type & LEFT))
+		while (size-- > 0)
+			*str++ = c;
+	while (i < precision--)
+		*str++ = '0';
+	while (i-- > 0)
+		*str++ = tmp[i];
+	while (size-- > 0)
+		*str++ = ' ';
+	return str;
+}
+
+int vsprintf(char *buf, const char *fmt, va_list args)
+{
+	int len;
+	unsigned long long num;
+	int i, base;
+	char * str;
+	const char *s;
+
+	int flags;		/* flags to number() */
+
+	int field_width;	/* width of output field */
+	int precision;		/* min. # of digits for integers; max
+				   number of chars for from string */
+	int qualifier;		/* 'h', 'l', or 'L' for integer fields */
+	                        /* 'z' support added 23/7/1999 S.H.    */
+				/* 'z' changed to 'Z' --davidm 1/25/99 */
+
+
+	for (str=buf ; *fmt ; ++fmt) {
+		if (*fmt != '%') {
+			*str++ = *fmt;
+			continue;
+		}
+
+		/* process flags */
+		flags = 0;
+		repeat:
+			++fmt;		/* this also skips first '%' */
+			switch (*fmt) {
+				case '-': flags |= LEFT; goto repeat;
+				case '+': flags |= PLUS; goto repeat;
+				case ' ': flags |= SPACE; goto repeat;
+				case '#': flags |= SPECIAL; goto repeat;
+				case '0': flags |= ZEROPAD; goto repeat;
+				}
+
+		/* get field width */
+		field_width = -1;
+		if ('0' <= *fmt && *fmt <= '9')
+			field_width = skip_atoi(&fmt);
+		else if (*fmt == '*') {
+			++fmt;
+			/* it's the next argument */
+			field_width = va_arg(args, int);
+			if (field_width < 0) {
+				field_width = -field_width;
+				flags |= LEFT;
+			}
+		}
+
+		/* get the precision */
+		precision = -1;
+		if (*fmt == '.') {
+			++fmt;
+			if ('0' <= *fmt && *fmt <= '9')
+				precision = skip_atoi(&fmt);
+			else if (*fmt == '*') {
+				++fmt;
+				/* it's the next argument */
+				precision = va_arg(args, int);
+			}
+			if (precision < 0)
+				precision = 0;
+		}
+
+		/* get the conversion qualifier */
+		qualifier = -1;
+		if (*fmt == 'l' && *(fmt + 1) == 'l') {
+			qualifier = 'q';
+			fmt += 2;
+		} else if (*fmt == 'h' || *fmt == 'l' || *fmt == 'L'
+			|| *fmt == 'Z') {
+			qualifier = *fmt;
+			++fmt;
+		}
+
+		/* default base */
+		base = 10;
+
+		switch (*fmt) {
+		case 'c':
+			if (!(flags & LEFT))
+				while (--field_width > 0)
+					*str++ = ' ';
+			*str++ = (unsigned char) va_arg(args, int);
+			while (--field_width > 0)
+				*str++ = ' ';
+			continue;
+
+		case 's':
+			s = va_arg(args, char *);
+			if (!s)
+				s = "<NULL>";
+
+			len = strnlen(s, precision);
+
+			if (!(flags & LEFT))
+				while (len < field_width--)
+					*str++ = ' ';
+			for (i = 0; i < len; ++i)
+				*str++ = *s++;
+			while (len < field_width--)
+				*str++ = ' ';
+			continue;
+
+		case 'p':
+			if (field_width == -1) {
+				field_width = 2*sizeof(void *);
+				flags |= ZEROPAD;
+			}
+			str = number(str,
+				(unsigned long) va_arg(args, void *), 16,
+				field_width, precision, flags);
+			continue;
+
+
+		case 'n':
+			if (qualifier == 'l') {
+				long * ip = va_arg(args, long *);
+				*ip = (str - buf);
+			} else if (qualifier == 'Z') {
+				size_t * ip = va_arg(args, size_t *);
+				*ip = (str - buf);
+			} else {
+				int * ip = va_arg(args, int *);
+				*ip = (str - buf);
+			}
+			continue;
+
+		case '%':
+			*str++ = '%';
+			continue;
+
+		/* integer number formats - set up the flags and "break" */
+		case 'o':
+			base = 8;
+			break;
+
+		case 'X':
+			flags |= LARGE;
+		case 'x':
+			base = 16;
+			break;
+
+		case 'd':
+		case 'i':
+			flags |= SIGN;
+		case 'u':
+			break;
+
+		default:
+			*str++ = '%';
+			if (*fmt)
+				*str++ = *fmt;
+			else
+				--fmt;
+			continue;
+		}
+		if (qualifier == 'l') {
+			num = va_arg(args, unsigned long);
+			if (flags & SIGN)
+				num = (signed long) num;
+		} else if (qualifier == 'q') {
+			num = va_arg(args, unsigned long long);
+			if (flags & SIGN)
+				num = (signed long long) num;
+		} else if (qualifier == 'Z') {
+			num = va_arg(args, size_t);
+		} else if (qualifier == 'h') {
+			num = (unsigned short) va_arg(args, int);
+			if (flags & SIGN)
+				num = (signed short) num;
+		} else {
+			num = va_arg(args, unsigned int);
+			if (flags & SIGN)
+				num = (signed int) num;
+		}
+		str = number(str, num, base, field_width, precision, flags);
+	}
+	*str = '\0';
+	return str-buf;
+}
+
+int sprintf(char * buf, const char *fmt, ...)
+{
+	va_list args;
+	int i;
+
+	va_start(args, fmt);
+	i=vsprintf(buf,fmt,args);
+	va_end(args);
+	return i;
+}

arch/alpha/boot/tools/objstrip.c

@@ -27,6 +27,9 @@
 #include <linux/param.h>
 #ifdef __ELF__
 # include <linux/elf.h>
+# define elfhdr elf64_hdr
+# define elf_phdr elf64_phdr
+# define elf_check_arch(x) ((x)->e_machine == EM_ALPHA)
 #endif
 
 /* bootfile size must be multiple of BLOCK_SIZE: */

arch/alpha/include/asm/types.h

@@ -2,6 +2,5 @@
 #define _ALPHA_TYPES_H
 
 #include <asm-generic/int-ll64.h>
-#include <uapi/asm/types.h>
 
 #endif /* _ALPHA_TYPES_H */

arch/alpha/include/asm/unistd.h

@@ -3,7 +3,7 @@
 
 #include <uapi/asm/unistd.h>
 
-#define NR_SYSCALLS			511
+#define NR_SYSCALLS			514
 
 #define __ARCH_WANT_OLD_READDIR
 #define __ARCH_WANT_STAT64

arch/alpha/include/uapi/asm/unistd.h

@@ -472,5 +472,8 @@
 #define __NR_sched_setattr		508
 #define __NR_sched_getattr		509
 #define __NR_renameat2			510
+#define __NR_getrandom			511
+#define __NR_memfd_create		512
+#define __NR_execveat			513
 
 #endif /* _UAPI_ALPHA_UNISTD_H */

arch/alpha/kernel/err_ev6.c

@@ -6,7 +6,6 @@
  *	Error handling code supporting Alpha systems
  */
 
-#include <linux/init.h>
 #include <linux/sched.h>
 
 #include <asm/io.h>

arch/alpha/kernel/irq.c

@@ -19,7 +19,6 @@
 #include <linux/ptrace.h>
 #include <linux/interrupt.h>
 #include <linux/random.h>
-#include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>

arch/alpha/kernel/osf_sys.c

@@ -1019,14 +1019,13 @@ SYSCALL_DEFINE2(osf_settimeofday, struct timeval32 __user *, tv,
  	if (tv) {
 		if (get_tv32((struct timeval *)&kts, tv))
 			return -EFAULT;
+		kts.tv_nsec *= 1000;
 	}
 	if (tz) {
 		if (copy_from_user(&ktz, tz, sizeof(*tz)))
 			return -EFAULT;
 	}
 
-	kts.tv_nsec *= 1000;
-
 	return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL);
 }
 

arch/alpha/kernel/process.c

@@ -236,12 +236,11 @@ release_thread(struct task_struct *dead_task)
 }
 
 /*
- * Copy an alpha thread..
+ * Copy architecture-specific thread state
  */
-
 int
 copy_thread(unsigned long clone_flags, unsigned long usp,
-	    unsigned long arg,
+	    unsigned long kthread_arg,
 	    struct task_struct *p)
 {
 	extern void ret_from_fork(void);
@@ -262,7 +261,7 @@ copy_thread(unsigned long clone_flags, unsigned long usp,
 			sizeof(struct switch_stack) + sizeof(struct pt_regs));
 		childstack->r26 = (unsigned long) ret_from_kernel_thread;
 		childstack->r9 = usp;	/* function */
-		childstack->r10 = arg;
+		childstack->r10 = kthread_arg;
 		childregs->hae = alpha_mv.hae_cache,
 		childti->pcb.usp = 0;
 		return 0;

arch/alpha/kernel/smp.c

@@ -63,7 +63,6 @@ static struct {
 enum ipi_message_type {
 	IPI_RESCHEDULE,
 	IPI_CALL_FUNC,
-	IPI_CALL_FUNC_SINGLE,
 	IPI_CPU_STOP,
 };
 
@@ -506,7 +505,6 @@ setup_profiling_timer(unsigned int multiplier)
 	return -EINVAL;
 }
 
-
 static void
 send_ipi_message(const struct cpumask *to_whom, enum ipi_message_type operation)
 {
@@ -552,10 +550,6 @@ handle_ipi(struct pt_regs *regs)
 			generic_smp_call_function_interrupt();
 			break;
 
-		case IPI_CALL_FUNC_SINGLE:
-			generic_smp_call_function_single_interrupt();
-			break;
-
 		case IPI_CPU_STOP:
 			halt();
 
@@ -606,7 +600,7 @@ void arch_send_call_function_ipi_mask(const struct cpumask *mask)
 
 void arch_send_call_function_single_ipi(int cpu)
 {
-	send_ipi_message(cpumask_of(cpu), IPI_CALL_FUNC_SINGLE);
+	send_ipi_message(cpumask_of(cpu), IPI_CALL_FUNC);
 }
 
 static void

arch/alpha/kernel/srmcons.c

@@ -237,8 +237,7 @@ srmcons_init(void)
 
 	return -ENODEV;
 }
-
+device_initcall(srmcons_init);
-module_init(srmcons_init);
 
 
 /*

arch/alpha/kernel/sys_marvel.c

@@ -331,7 +331,7 @@ marvel_map_irq(const struct pci_dev *cdev, u8 slot, u8 pin)
 	pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &intline);
 	irq = intline;
 
-	msi_loc = pci_find_capability(dev, PCI_CAP_ID_MSI);
+	msi_loc = dev->msi_cap;
 	msg_ctl = 0;
 	if (msi_loc) 
 		pci_read_config_word(dev, msi_loc + PCI_MSI_FLAGS, &msg_ctl);

arch/alpha/kernel/systbls.S

@@ -529,6 +529,9 @@ sys_call_table:
 	.quad sys_sched_setattr
 	.quad sys_sched_getattr
 	.quad sys_renameat2			/* 510 */
+	.quad sys_getrandom
+	.quad sys_memfd_create
+	.quad sys_execveat
 
 	.size sys_call_table, . - sys_call_table
 	.type sys_call_table, @object

arch/alpha/kernel/traps.c

@@ -14,7 +14,6 @@
 #include <linux/tty.h>
 #include <linux/delay.h>
 #include <linux/module.h>
-#include <linux/init.h>
 #include <linux/kallsyms.h>
 #include <linux/ratelimit.h>
 

arch/alpha/oprofile/op_model_ev4.c

@@ -8,7 +8,6 @@
  */
 
 #include <linux/oprofile.h>
-#include <linux/init.h>
 #include <linux/smp.h>
 #include <asm/ptrace.h>
 

arch/alpha/oprofile/op_model_ev5.c

@@ -8,7 +8,6 @@
  */
 
 #include <linux/oprofile.h>
-#include <linux/init.h>
 #include <linux/smp.h>
 #include <asm/ptrace.h>
 

arch/alpha/oprofile/op_model_ev6.c

@@ -8,7 +8,6 @@
  */
 
 #include <linux/oprofile.h>
-#include <linux/init.h>
 #include <linux/smp.h>
 #include <asm/ptrace.h>
 

arch/alpha/oprofile/op_model_ev67.c

@@ -9,7 +9,6 @@
  */
 
 #include <linux/oprofile.h>
-#include <linux/init.h>
 #include <linux/smp.h>
 #include <asm/ptrace.h>
 

arch/arm/boot/dts/Makefile

@@ -223,7 +223,7 @@ dtb-$(CONFIG_SOC_IMX25) += \
 	imx25-eukrea-mbimxsd25-baseboard-dvi-vga.dtb \
 	imx25-karo-tx25.dtb \
 	imx25-pdk.dtb
-dtb-$(CONFIG_SOC_IMX31) += \
+dtb-$(CONFIG_SOC_IMX27) += \
 	imx27-apf27.dtb \
 	imx27-apf27dev.dtb \
 	imx27-eukrea-mbimxsd27-baseboard.dtb \

arch/arm/boot/dts/am335x-boneblack.dts

@@ -80,7 +80,3 @@
 		status = "okay";
 	};
 };
-
-&rtc {
-	system-power-controller;
-};

arch/arm/boot/dts/am335x-evmsk.dts

@@ -654,7 +654,7 @@
 	wlcore: wlcore@2 {
 		compatible = "ti,wl1271";
 		reg = <2>;
-		interrupt-parent = <&gpio1>;
+		interrupt-parent = <&gpio0>;
 		interrupts = <31 IRQ_TYPE_LEVEL_HIGH>; /* gpio 31 */
 		ref-clock-frequency = <38400000>;
 	};

arch/arm/boot/dts/exynos4412-trats2.dts

@@ -736,7 +736,7 @@
 
 			display-timings {
 				timing-0 {
-					clock-frequency = <0>;
+					clock-frequency = <57153600>;
 					hactive = <720>;
 					vactive = <1280>;
 					hfront-porch = <5>;

arch/arm/boot/dts/imx27.dtsi

@@ -533,7 +533,7 @@
 
 			fec: ethernet@1002b000 {
 				compatible = "fsl,imx27-fec";
-				reg = <0x1002b000 0x4000>;
+				reg = <0x1002b000 0x1000>;
 				interrupts = <50>;
 				clocks = <&clks IMX27_CLK_FEC_IPG_GATE>,
 					 <&clks IMX27_CLK_FEC_AHB_GATE>;

arch/arm/boot/dts/omap3-devkit8000.dts

@@ -110,6 +110,8 @@
 	nand@0,0 {
 		reg = <0 0 4>; /* CS0, offset 0, IO size 4 */
 		nand-bus-width = <16>;
+		gpmc,device-width = <2>;
+		ti,nand-ecc-opt = "sw";
 
 		gpmc,sync-clk-ps = <0>;
 		gpmc,cs-on-ns = <0>;

arch/arm/boot/dts/zynq-7000.dtsi

@@ -193,7 +193,7 @@
 		};
 
 		gem0: ethernet@e000b000 {
-			compatible = "cdns,gem";
+			compatible = "cdns,zynq-gem";
 			reg = <0xe000b000 0x1000>;
 			status = "disabled";
 			interrupts = <0 22 4>;
@@ -204,7 +204,7 @@
 		};
 
 		gem1: ethernet@e000c000 {
-			compatible = "cdns,gem";
+			compatible = "cdns,zynq-gem";
 			reg = <0xe000c000 0x1000>;
 			status = "disabled";
 			interrupts = <0 45 4>;

arch/arm/configs/multi_v7_defconfig

@@ -429,7 +429,7 @@ CONFIG_USB_EHCI_EXYNOS=y
 CONFIG_USB_EHCI_TEGRA=y
 CONFIG_USB_EHCI_HCD_STI=y
 CONFIG_USB_EHCI_HCD_PLATFORM=y
-CONFIG_USB_ISP1760_HCD=y
+CONFIG_USB_ISP1760=y
 CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_OHCI_HCD_STI=y
 CONFIG_USB_OHCI_HCD_PLATFORM=y

arch/arm/kernel/entry-common.S

@@ -33,7 +33,9 @@ ret_fast_syscall:
  UNWIND(.fnstart	)
  UNWIND(.cantunwind	)
 	disable_irq				@ disable interrupts
-	ldr	r1, [tsk, #TI_FLAGS]
+	ldr	r1, [tsk, #TI_FLAGS]		@ re-check for syscall tracing
+	tst	r1, #_TIF_SYSCALL_WORK
+	bne	__sys_trace_return
 	tst	r1, #_TIF_WORK_MASK
 	bne	fast_work_pending
 	asm_trace_hardirqs_on

arch/arm/kernel/perf_event_cpu.c

@@ -304,16 +304,17 @@ static int probe_current_pmu(struct arm_pmu *pmu)
 static int of_pmu_irq_cfg(struct platform_device *pdev)
 {
 	int i, irq;
-	int *irqs = kcalloc(pdev->num_resources, sizeof(*irqs), GFP_KERNEL);
+	int *irqs;
-
-	if (!irqs)
-		return -ENOMEM;
 
 	/* Don't bother with PPIs; they're already affine */
 	irq = platform_get_irq(pdev, 0);
 	if (irq >= 0 && irq_is_percpu(irq))
 		return 0;
 
+	irqs = kcalloc(pdev->num_resources, sizeof(*irqs), GFP_KERNEL);
+	if (!irqs)
+		return -ENOMEM;
+
 	for (i = 0; i < pdev->num_resources; ++i) {
 		struct device_node *dn;
 		int cpu;

arch/arm/mach-imx/gpc.c

@@ -280,9 +280,15 @@ void __init imx_gpc_check_dt(void)
 	struct device_node *np;
 
 	np = of_find_compatible_node(NULL, NULL, "fsl,imx6q-gpc");
-	if (WARN_ON(!np ||
+	if (WARN_ON(!np))
-		    !of_find_property(np, "interrupt-controller", NULL)))
+		return;
-		pr_warn("Outdated DT detected, system is about to crash!!!\n");
+
+	if (WARN_ON(!of_find_property(np, "interrupt-controller", NULL))) {
+		pr_warn("Outdated DT detected, suspend/resume will NOT work\n");
+
+		/* map GPC, so that at least CPUidle and WARs keep working */
+		gpc_base = of_iomap(np, 0);
+	}
 }
 
 #ifdef CONFIG_PM_GENERIC_DOMAINS
@@ -443,6 +449,10 @@ static int imx_gpc_probe(struct platform_device *pdev)
 	struct regulator *pu_reg;
 	int ret;
 
+	/* bail out if DT too old and doesn't provide the necessary info */
+	if (!of_property_read_bool(pdev->dev.of_node, "#power-domain-cells"))
+		return 0;
+
 	pu_reg = devm_regulator_get_optional(&pdev->dev, "pu");
 	if (PTR_ERR(pu_reg) == -ENODEV)
 		pu_reg = NULL;

arch/arm/mach-pxa/pxa_cplds_irqs.c

@@ -107,7 +107,7 @@ static int cplds_probe(struct platform_device *pdev)
 	struct resource *res;
 	struct cplds *fpga;
 	int ret;
-	unsigned int base_irq = 0;
+	int base_irq;
 	unsigned long irqflags = 0;
 
 	fpga = devm_kzalloc(&pdev->dev, sizeof(*fpga), GFP_KERNEL);

arch/arm/mm/mmu.c

@@ -1112,22 +1112,22 @@ void __init sanity_check_meminfo(void)
 			}
 
 			/*
-			 * Find the first non-section-aligned page, and point
+			 * Find the first non-pmd-aligned page, and point
 			 * memblock_limit at it. This relies on rounding the
-			 * limit down to be section-aligned, which happens at
+			 * limit down to be pmd-aligned, which happens at the
-			 * the end of this function.
+			 * end of this function.
 			 *
 			 * With this algorithm, the start or end of almost any
-			 * bank can be non-section-aligned. The only exception
+			 * bank can be non-pmd-aligned. The only exception is
-			 * is that the start of the bank 0 must be section-
+			 * that the start of the bank 0 must be section-
 			 * aligned, since otherwise memory would need to be
 			 * allocated when mapping the start of bank 0, which
 			 * occurs before any free memory is mapped.
 			 */
 			if (!memblock_limit) {
-				if (!IS_ALIGNED(block_start, SECTION_SIZE))
+				if (!IS_ALIGNED(block_start, PMD_SIZE))
 					memblock_limit = block_start;
-				else if (!IS_ALIGNED(block_end, SECTION_SIZE))
+				else if (!IS_ALIGNED(block_end, PMD_SIZE))
 					memblock_limit = arm_lowmem_limit;
 			}
 
@@ -1137,12 +1137,12 @@ void __init sanity_check_meminfo(void)
 	high_memory = __va(arm_lowmem_limit - 1) + 1;
 
 	/*
-	 * Round the memblock limit down to a section size.  This
+	 * Round the memblock limit down to a pmd size.  This
 	 * helps to ensure that we will allocate memory from the
-	 * last full section, which should be mapped.
+	 * last full pmd, which should be mapped.
 	 */
 	if (memblock_limit)
-		memblock_limit = round_down(memblock_limit, SECTION_SIZE);
+		memblock_limit = round_down(memblock_limit, PMD_SIZE);
 	if (!memblock_limit)
 		memblock_limit = arm_lowmem_limit;
 

arch/arm/xen/enlighten.c

@@ -272,6 +272,7 @@ void xen_arch_pre_suspend(void) { }
 void xen_arch_post_suspend(int suspend_cancelled) { }
 void xen_timer_resume(void) { }
 void xen_arch_resume(void) { }
+void xen_arch_suspend(void) { }
 
 
 /* In the hypervisor.S file. */

arch/ia64/kernel/smpboot.c

@@ -127,7 +127,7 @@ int smp_num_siblings = 1;
 volatile int ia64_cpu_to_sapicid[NR_CPUS];
 EXPORT_SYMBOL(ia64_cpu_to_sapicid);
 
-static volatile cpumask_t cpu_callin_map;
+static cpumask_t cpu_callin_map;
 
 struct smp_boot_data smp_boot_data __initdata;
 
@@ -477,6 +477,7 @@ do_boot_cpu (int sapicid, int cpu, struct task_struct *idle)
 	for (timeout = 0; timeout < 100000; timeout++) {
 		if (cpumask_test_cpu(cpu, &cpu_callin_map))
 			break;  /* It has booted */
+		barrier(); /* Make sure we re-read cpu_callin_map */
 		udelay(100);
 	}
 	Dprintk("\n");

arch/ia64/pci/pci.c

@@ -478,9 +478,16 @@ struct pci_bus *pci_acpi_scan_root(struct acpi_pci_root *root)
 
 int pcibios_root_bridge_prepare(struct pci_host_bridge *bridge)
 {
-	struct pci_controller *controller = bridge->bus->sysdata;
+	/*
-
+	 * We pass NULL as parent to pci_create_root_bus(), so if it is not NULL
-	ACPI_COMPANION_SET(&bridge->dev, controller->companion);
+	 * here, pci_create_root_bus() has been called by someone else and
+	 * sysdata is likely to be different from what we expect.  Let it go in
+	 * that case.
+	 */
+	if (!bridge->dev.parent) {
+		struct pci_controller *controller = bridge->bus->sysdata;
+		ACPI_COMPANION_SET(&bridge->dev, controller->companion);
+	}
 	return 0;
 }
 

arch/mips/ath79/prom.c

@@ -1,6 +1,7 @@
 /*
  *  Atheros AR71XX/AR724X/AR913X specific prom routines
  *
+ *  Copyright (C) 2015 Laurent Fasnacht <l@libres.ch>
  *  Copyright (C) 2008-2010 Gabor Juhos <juhosg@openwrt.org>
  *  Copyright (C) 2008 Imre Kaloz <kaloz@openwrt.org>
  *
@@ -25,12 +26,14 @@ void __init prom_init(void)
 {
 	fw_init_cmdline();
 
+#ifdef CONFIG_BLK_DEV_INITRD
 	/* Read the initrd address from the firmware environment */
 	initrd_start = fw_getenvl("initrd_start");
 	if (initrd_start) {
 		initrd_start = KSEG0ADDR(initrd_start);
 		initrd_end = initrd_start + fw_getenvl("initrd_size");
 	}
+#endif
 }
 
 void __init prom_free_prom_memory(void)

arch/mips/ath79/setup.c

@@ -225,7 +225,7 @@ void __init plat_time_init(void)
 	ddr_clk_rate = ath79_get_sys_clk_rate("ddr");
 	ref_clk_rate = ath79_get_sys_clk_rate("ref");
 
-	pr_info("Clocks: CPU:%lu.%03luMHz, DDR:%lu.%03luMHz, AHB:%lu.%03luMHz, Ref:%lu.%03luMHz",
+	pr_info("Clocks: CPU:%lu.%03luMHz, DDR:%lu.%03luMHz, AHB:%lu.%03luMHz, Ref:%lu.%03luMHz\n",
 		cpu_clk_rate / 1000000, (cpu_clk_rate / 1000) % 1000,
 		ddr_clk_rate / 1000000, (ddr_clk_rate / 1000) % 1000,
 		ahb_clk_rate / 1000000, (ahb_clk_rate / 1000) % 1000,

arch/mips/configs/fuloong2e_defconfig

@@ -194,7 +194,7 @@ CONFIG_USB_WUSB_CBAF=m
 CONFIG_USB_C67X00_HCD=m
 CONFIG_USB_EHCI_HCD=y
 CONFIG_USB_EHCI_ROOT_HUB_TT=y
-CONFIG_USB_ISP1760_HCD=m
+CONFIG_USB_ISP1760=m
 CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_UHCI_HCD=m
 CONFIG_USB_R8A66597_HCD=m

arch/mips/kernel/cpu-probe.c

@@ -74,13 +74,12 @@ static inline void cpu_set_fpu_fcsr_mask(struct cpuinfo_mips *c)
 {
 	unsigned long sr, mask, fcsr, fcsr0, fcsr1;
 
+	fcsr = c->fpu_csr31;
 	mask = FPU_CSR_ALL_X | FPU_CSR_ALL_E | FPU_CSR_ALL_S | FPU_CSR_RM;
 
 	sr = read_c0_status();
 	__enable_fpu(FPU_AS_IS);
 
-	fcsr = read_32bit_cp1_register(CP1_STATUS);
-
 	fcsr0 = fcsr & mask;
 	write_32bit_cp1_register(CP1_STATUS, fcsr0);
 	fcsr0 = read_32bit_cp1_register(CP1_STATUS);

arch/mips/kernel/irq.c

@@ -29,7 +29,7 @@
 int kgdb_early_setup;
 #endif
 
-static unsigned long irq_map[NR_IRQS / BITS_PER_LONG];
+static DECLARE_BITMAP(irq_map, NR_IRQS);
 
 int allocate_irqno(void)
 {
@@ -109,7 +109,7 @@ void __init init_IRQ(void)
 #endif
 }
 
-#ifdef DEBUG_STACKOVERFLOW
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
 static inline void check_stack_overflow(void)
 {
 	unsigned long sp;

arch/mips/kernel/smp-bmips.c

@@ -444,7 +444,7 @@ struct plat_smp_ops bmips5000_smp_ops = {
 static void bmips_wr_vec(unsigned long dst, char *start, char *end)
 {
 	memcpy((void *)dst, start, end - start);
-	dma_cache_wback((unsigned long)start, end - start);
+	dma_cache_wback(dst, end - start);
 	local_flush_icache_range(dst, dst + (end - start));
 	instruction_hazard();
 }

arch/mips/kvm/emulate.c

@@ -2409,7 +2409,7 @@ enum emulation_result kvm_mips_complete_mmio_load(struct kvm_vcpu *vcpu,
 		if (vcpu->mmio_needed == 2)
 			*gpr = *(int16_t *) run->mmio.data;
 		else
-			*gpr = *(int16_t *) run->mmio.data;
+			*gpr = *(uint16_t *)run->mmio.data;
 
 		break;
 	case 1:

arch/mips/lib/strnlen_user.S

@@ -34,7 +34,12 @@ LEAF(__strnlen_\func\()_asm)
 FEXPORT(__strnlen_\func\()_nocheck_asm)
 	move		v0, a0
 	PTR_ADDU	a1, a0			# stop pointer
-1:	beq		v0, a1, 1f		# limit reached?
+1:
+#ifdef CONFIG_CPU_DADDI_WORKAROUNDS
+	.set		noat
+	li		AT, 1
+#endif
+	beq		v0, a1, 1f		# limit reached?
 .ifeqs "\func", "kernel"
 	EX(lb, t0, (v0), .Lfault\@)
 .else
@@ -42,7 +47,13 @@ FEXPORT(__strnlen_\func\()_nocheck_asm)
 .endif
 	.set		noreorder
 	bnez		t0, 1b
-1:	 PTR_ADDIU	v0, 1
+1:
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
+	 PTR_ADDIU	v0, 1
+#else
+	 PTR_ADDU	v0, AT
+	.set		at
+#endif
 	.set		reorder
 	PTR_SUBU	v0, a0
 	jr		ra

arch/mips/loongson/loongson-3/smp.c

@@ -272,7 +272,7 @@ void loongson3_ipi_interrupt(struct pt_regs *regs)
 	if (action & SMP_ASK_C0COUNT) {
 		BUG_ON(cpu != 0);
 		c0count = read_c0_count();
-		for (i = 1; i < loongson_sysconf.nr_cpus; i++)
+		for (i = 1; i < num_possible_cpus(); i++)
 			per_cpu(core0_c0count, i) = c0count;
 	}
 }

arch/mips/mm/c-r4k.c

@@ -1372,7 +1372,7 @@ static int probe_scache(void)
 	scache_size = addr;
 	c->scache.linesz = 16 << ((config & R4K_CONF_SB) >> 22);
 	c->scache.ways = 1;
-	c->dcache.waybit = 0;		/* does not matter */
+	c->scache.waybit = 0;		/* does not matter */
 
 	return 1;
 }

arch/mips/net/bpf_jit.c

@@ -681,11 +681,7 @@ static unsigned int get_stack_depth(struct jit_ctx *ctx)
 		sp_off += config_enabled(CONFIG_64BIT) ?
 			(ARGS_USED_BY_JIT + 1) * RSIZE : RSIZE;
 
-	/*
+	return sp_off;
-	 * Subtract the bytes for the last registers since we only care about
-	 * the location on the stack pointer.
-	 */
-	return sp_off - RSIZE;
 }
 
 static void build_prologue(struct jit_ctx *ctx)

arch/mips/ralink/ill_acc.c

@@ -41,7 +41,7 @@ static irqreturn_t ill_acc_irq_handler(int irq, void *_priv)
 		addr, (type >> ILL_ACC_OFF_S) & ILL_ACC_OFF_M,
 		type & ILL_ACC_LEN_M);
 
-	rt_memc_w32(REG_ILL_ACC_TYPE, REG_ILL_ACC_TYPE);
+	rt_memc_w32(ILL_INT_STATUS, REG_ILL_ACC_TYPE);
 
 	return IRQ_HANDLED;
 }

arch/powerpc/kernel/mce.c

@@ -73,7 +73,7 @@ void save_mce_event(struct pt_regs *regs, long handled,
 		    uint64_t nip, uint64_t addr)
 {
 	uint64_t srr1;
-	int index = __this_cpu_inc_return(mce_nest_count);
+	int index = __this_cpu_inc_return(mce_nest_count) - 1;
 	struct machine_check_event *mce = this_cpu_ptr(&mce_event[index]);
 
 	/*
@@ -184,7 +184,7 @@ void machine_check_queue_event(void)
 	if (!get_mce_event(&evt, MCE_EVENT_RELEASE))
 		return;
 
-	index = __this_cpu_inc_return(mce_queue_count);
+	index = __this_cpu_inc_return(mce_queue_count) - 1;
 	/* If queue is full, just return for now. */
 	if (index >= MAX_MC_EVT) {
 		__this_cpu_dec(mce_queue_count);

arch/powerpc/kernel/vmlinux.lds.S

@@ -213,6 +213,7 @@ SECTIONS
 		*(.opd)
 	}
 
+	. = ALIGN(256);
 	.got : AT(ADDR(.got) - LOAD_OFFSET) {
 		__toc_start = .;
 #ifndef CONFIG_RELOCATABLE

arch/powerpc/kvm/book3s_hv.c

@@ -1952,7 +1952,7 @@ static void post_guest_process(struct kvmppc_vcore *vc)
  */
 static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
 {
-	struct kvm_vcpu *vcpu;
+	struct kvm_vcpu *vcpu, *vnext;
 	int i;
 	int srcu_idx;
 
@@ -1982,7 +1982,8 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
 	 */
 	if ((threads_per_core > 1) &&
 	    ((vc->num_threads > threads_per_subcore) || !on_primary_thread())) {
-		list_for_each_entry(vcpu, &vc->runnable_threads, arch.run_list) {
+		list_for_each_entry_safe(vcpu, vnext, &vc->runnable_threads,
+					 arch.run_list) {
 			vcpu->arch.ret = -EBUSY;
 			kvmppc_remove_runnable(vc, vcpu);
 			wake_up(&vcpu->arch.cpu_run);

arch/powerpc/mm/hugetlbpage.c

@@ -689,27 +689,34 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 struct page *
 follow_huge_addr(struct mm_struct *mm, unsigned long address, int write)
 {
-	pte_t *ptep;
+	pte_t *ptep, pte;
-	struct page *page;
 	unsigned shift;
 	unsigned long mask, flags;
+	struct page *page = ERR_PTR(-EINVAL);
+
+	local_irq_save(flags);
+	ptep = find_linux_pte_or_hugepte(mm->pgd, address, &shift);
+	if (!ptep)
+		goto no_page;
+	pte = READ_ONCE(*ptep);
 	/*
+	 * Verify it is a huge page else bail.
 	 * Transparent hugepages are handled by generic code. We can skip them
 	 * here.
 	 */
-	local_irq_save(flags);
+	if (!shift || pmd_trans_huge(__pmd(pte_val(pte))))
-	ptep = find_linux_pte_or_hugepte(mm->pgd, address, &shift);
+		goto no_page;
 
-	/* Verify it is a huge page else bail. */
+	if (!pte_present(pte)) {
-	if (!ptep || !shift || pmd_trans_huge(*(pmd_t *)ptep)) {
+		page = NULL;
-		local_irq_restore(flags);
+		goto no_page;
-		return ERR_PTR(-EINVAL);
 	}
 	mask = (1UL << shift) - 1;
-	page = pte_page(*ptep);
+	page = pte_page(pte);
 	if (page)
 		page += (address & mask) / PAGE_SIZE;
 
+no_page:
 	local_irq_restore(flags);
 	return page;
 }

arch/powerpc/mm/pgtable_64.c

@@ -839,6 +839,17 @@ pmd_t pmdp_get_and_clear(struct mm_struct *mm,
 	 * hash fault look at them.
 	 */
 	memset(pgtable, 0, PTE_FRAG_SIZE);
+	/*
+	 * Serialize against find_linux_pte_or_hugepte which does lock-less
+	 * lookup in page tables with local interrupts disabled. For huge pages
+	 * it casts pmd_t to pte_t. Since format of pte_t is different from
+	 * pmd_t we want to prevent transit from pmd pointing to page table
+	 * to pmd pointing to huge page (and back) while interrupts are disabled.
+	 * We clear pmd to possibly replace it with page table pointer in
+	 * different code paths. So make sure we wait for the parallel
+	 * find_linux_pte_or_hugepage to finish.
+	 */
+	kick_all_cpus_sync();
 	return old_pmd;
 }
 

arch/s390/crypto/ghash_s390.c

@@ -16,11 +16,12 @@
 #define GHASH_DIGEST_SIZE	16
 
 struct ghash_ctx {
-	u8 icv[16];
+	u8 key[GHASH_BLOCK_SIZE];
-	u8 key[16];
 };
 
 struct ghash_desc_ctx {
+	u8 icv[GHASH_BLOCK_SIZE];
+	u8 key[GHASH_BLOCK_SIZE];
 	u8 buffer[GHASH_BLOCK_SIZE];
 	u32 bytes;
 };
@@ -28,8 +29,10 @@ struct ghash_desc_ctx {
 static int ghash_init(struct shash_desc *desc)
 {
 	struct ghash_desc_ctx *dctx = shash_desc_ctx(desc);
+	struct ghash_ctx *ctx = crypto_shash_ctx(desc->tfm);
 
 	memset(dctx, 0, sizeof(*dctx));
+	memcpy(dctx->key, ctx->key, GHASH_BLOCK_SIZE);
 
 	return 0;
 }
@@ -45,7 +48,6 @@ static int ghash_setkey(struct crypto_shash *tfm,
 	}
 
 	memcpy(ctx->key, key, GHASH_BLOCK_SIZE);
-	memset(ctx->icv, 0, GHASH_BLOCK_SIZE);
 
 	return 0;
 }
@@ -54,7 +56,6 @@ static int ghash_update(struct shash_desc *desc,
 			 const u8 *src, unsigned int srclen)
 {
 	struct ghash_desc_ctx *dctx = shash_desc_ctx(desc);
-	struct ghash_ctx *ctx = crypto_shash_ctx(desc->tfm);
 	unsigned int n;
 	u8 *buf = dctx->buffer;
 	int ret;
@@ -70,7 +71,7 @@ static int ghash_update(struct shash_desc *desc,
 		src += n;
 
 		if (!dctx->bytes) {
-			ret = crypt_s390_kimd(KIMD_GHASH, ctx, buf,
+			ret = crypt_s390_kimd(KIMD_GHASH, dctx, buf,
 					      GHASH_BLOCK_SIZE);
 			if (ret != GHASH_BLOCK_SIZE)
 				return -EIO;
@@ -79,7 +80,7 @@ static int ghash_update(struct shash_desc *desc,
 
 	n = srclen & ~(GHASH_BLOCK_SIZE - 1);
 	if (n) {
-		ret = crypt_s390_kimd(KIMD_GHASH, ctx, src, n);
+		ret = crypt_s390_kimd(KIMD_GHASH, dctx, src, n);
 		if (ret != n)
 			return -EIO;
 		src += n;
@@ -94,7 +95,7 @@ static int ghash_update(struct shash_desc *desc,
 	return 0;
 }
 
-static int ghash_flush(struct ghash_ctx *ctx, struct ghash_desc_ctx *dctx)
+static int ghash_flush(struct ghash_desc_ctx *dctx)
 {
 	u8 *buf = dctx->buffer;
 	int ret;
@@ -104,24 +105,24 @@ static int ghash_flush(struct ghash_ctx *ctx, struct ghash_desc_ctx *dctx)
 
 		memset(pos, 0, dctx->bytes);
 
-		ret = crypt_s390_kimd(KIMD_GHASH, ctx, buf, GHASH_BLOCK_SIZE);
+		ret = crypt_s390_kimd(KIMD_GHASH, dctx, buf, GHASH_BLOCK_SIZE);
 		if (ret != GHASH_BLOCK_SIZE)
 			return -EIO;
+
+		dctx->bytes = 0;
 	}
 
-	dctx->bytes = 0;
 	return 0;
 }
 
 static int ghash_final(struct shash_desc *desc, u8 *dst)
 {
 	struct ghash_desc_ctx *dctx = shash_desc_ctx(desc);
-	struct ghash_ctx *ctx = crypto_shash_ctx(desc->tfm);
 	int ret;
 
-	ret = ghash_flush(ctx, dctx);
+	ret = ghash_flush(dctx);
 	if (!ret)
-		memcpy(dst, ctx->icv, GHASH_BLOCK_SIZE);
+		memcpy(dst, dctx->icv, GHASH_BLOCK_SIZE);
 	return ret;
 }
 

arch/s390/crypto/prng.c

@@ -125,7 +125,7 @@ static int generate_entropy(u8 *ebuf, size_t nbytes)
 		/* fill page with urandom bytes */
 		get_random_bytes(pg, PAGE_SIZE);
 		/* exor page with stckf values */
-		for (n = 0; n < sizeof(PAGE_SIZE/sizeof(u64)); n++) {
+		for (n = 0; n < PAGE_SIZE / sizeof(u64); n++) {
 			u64 *p = ((u64 *)pg) + n;
 			*p ^= get_tod_clock_fast();
 		}

arch/s390/include/asm/pgtable.h

@@ -494,7 +494,7 @@ static inline int pmd_large(pmd_t pmd)
 	return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0;
 }
 
-static inline int pmd_pfn(pmd_t pmd)
+static inline unsigned long pmd_pfn(pmd_t pmd)
 {
 	unsigned long origin_mask;
 

arch/s390/net/bpf_jit_comp.c

@@ -443,8 +443,11 @@ static void bpf_jit_epilogue(struct bpf_jit *jit)
 
 /*
  * Compile one eBPF instruction into s390x code
+ *
+ * NOTE: Use noinline because for gcov (-fprofile-arcs) gcc allocates a lot of
+ * stack space for the large switch statement.
  */
-static int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i)
+static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i)
 {
 	struct bpf_insn *insn = &fp->insnsi[i];
 	int jmp_off, last, insn_count = 1;
@@ -588,8 +591,8 @@ static int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i)
 		EMIT4(0xb9160000, dst_reg, rc_reg);
 		break;
 	}
-	case BPF_ALU64 | BPF_DIV | BPF_X: /* dst = dst / (u32) src */
+	case BPF_ALU64 | BPF_DIV | BPF_X: /* dst = dst / src */
-	case BPF_ALU64 | BPF_MOD | BPF_X: /* dst = dst % (u32) src */
+	case BPF_ALU64 | BPF_MOD | BPF_X: /* dst = dst % src */
 	{
 		int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0;
 
@@ -602,10 +605,8 @@ static int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i)
 		EMIT4_IMM(0xa7090000, REG_W0, 0);
 		/* lgr %w1,%dst */
 		EMIT4(0xb9040000, REG_W1, dst_reg);
-		/* llgfr %dst,%src (u32 cast) */
-		EMIT4(0xb9160000, dst_reg, src_reg);
 		/* dlgr %w0,%dst */
-		EMIT4(0xb9870000, REG_W0, dst_reg);
+		EMIT4(0xb9870000, REG_W0, src_reg);
 		/* lgr %dst,%rc */
 		EMIT4(0xb9040000, dst_reg, rc_reg);
 		break;
@@ -632,8 +633,8 @@ static int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i)
 		EMIT4(0xb9160000, dst_reg, rc_reg);
 		break;
 	}
-	case BPF_ALU64 | BPF_DIV | BPF_K: /* dst = dst / (u32) imm */
+	case BPF_ALU64 | BPF_DIV | BPF_K: /* dst = dst / imm */
-	case BPF_ALU64 | BPF_MOD | BPF_K: /* dst = dst % (u32) imm */
+	case BPF_ALU64 | BPF_MOD | BPF_K: /* dst = dst % imm */
 	{
 		int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0;
 
@@ -649,7 +650,7 @@ static int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i)
 		EMIT4(0xb9040000, REG_W1, dst_reg);
 		/* dlg %w0,<d(imm)>(%l) */
 		EMIT6_DISP_LH(0xe3000000, 0x0087, REG_W0, REG_0, REG_L,
-			      EMIT_CONST_U64((u32) imm));
+			      EMIT_CONST_U64(imm));
 		/* lgr %dst,%rc */
 		EMIT4(0xb9040000, dst_reg, rc_reg);
 		break;

arch/sparc/include/asm/cpudata_64.h

@@ -24,7 +24,8 @@ typedef struct {
 	unsigned int	icache_line_size;
 	unsigned int	ecache_size;
 	unsigned int	ecache_line_size;
-	int		core_id;
+	unsigned short	sock_id;
+	unsigned short	core_id;
 	int		proc_id;
 } cpuinfo_sparc;
 

arch/sparc/include/asm/pgtable_64.h

@@ -308,12 +308,26 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot)
 	"	sllx		%1, 32, %1\n"
 	"	or		%0, %1, %0\n"
 	"	.previous\n"
+	"	.section	.sun_m7_2insn_patch, \"ax\"\n"
+	"	.word		661b\n"
+	"	sethi		%%uhi(%4), %1\n"
+	"	sethi		%%hi(%4), %0\n"
+	"	.word		662b\n"
+	"	or		%1, %%ulo(%4), %1\n"
+	"	or		%0, %%lo(%4), %0\n"
+	"	.word		663b\n"
+	"	sllx		%1, 32, %1\n"
+	"	or		%0, %1, %0\n"
+	"	.previous\n"
 	: "=r" (mask), "=r" (tmp)
 	: "i" (_PAGE_PADDR_4U | _PAGE_MODIFIED_4U | _PAGE_ACCESSED_4U |
 	       _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U |
 	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4U),
 	  "i" (_PAGE_PADDR_4V | _PAGE_MODIFIED_4V | _PAGE_ACCESSED_4V |
 	       _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V |
+	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4V),
+	  "i" (_PAGE_PADDR_4V | _PAGE_MODIFIED_4V | _PAGE_ACCESSED_4V |
+	       _PAGE_CP_4V | _PAGE_E_4V |
 	       _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4V));
 
 	return __pte((pte_val(pte) & mask) | (pgprot_val(prot) & ~mask));
@@ -342,9 +356,15 @@ static inline pgprot_t pgprot_noncached(pgprot_t prot)
 	"	andn		%0, %4, %0\n"
 	"	or		%0, %5, %0\n"
 	"	.previous\n"
+	"	.section	.sun_m7_2insn_patch, \"ax\"\n"
+	"	.word		661b\n"
+	"	andn		%0, %6, %0\n"
+	"	or		%0, %5, %0\n"
+	"	.previous\n"
 	: "=r" (val)
 	: "0" (val), "i" (_PAGE_CP_4U | _PAGE_CV_4U), "i" (_PAGE_E_4U),
-	             "i" (_PAGE_CP_4V | _PAGE_CV_4V), "i" (_PAGE_E_4V));
+	             "i" (_PAGE_CP_4V | _PAGE_CV_4V), "i" (_PAGE_E_4V),
+	             "i" (_PAGE_CP_4V));
 
 	return __pgprot(val);
 }

arch/sparc/include/asm/topology_64.h

@@ -40,11 +40,12 @@ static inline int pcibus_to_node(struct pci_bus *pbus)
 #ifdef CONFIG_SMP
 #define topology_physical_package_id(cpu)	(cpu_data(cpu).proc_id)
 #define topology_core_id(cpu)			(cpu_data(cpu).core_id)
-#define topology_core_cpumask(cpu)		(&cpu_core_map[cpu])
+#define topology_core_cpumask(cpu)		(&cpu_core_sib_map[cpu])
 #define topology_thread_cpumask(cpu)		(&per_cpu(cpu_sibling_map, cpu))
 #endif /* CONFIG_SMP */
 
 extern cpumask_t cpu_core_map[NR_CPUS];
+extern cpumask_t cpu_core_sib_map[NR_CPUS];
 static inline const struct cpumask *cpu_coregroup_mask(int cpu)
 {
         return &cpu_core_map[cpu];

arch/sparc/include/asm/trap_block.h

@@ -79,6 +79,8 @@ struct sun4v_2insn_patch_entry {
 };
 extern struct sun4v_2insn_patch_entry __sun4v_2insn_patch,
 	__sun4v_2insn_patch_end;
+extern struct sun4v_2insn_patch_entry __sun_m7_2insn_patch,
+	__sun_m7_2insn_patch_end;
 
 
 #endif /* !(__ASSEMBLY__) */

arch/sparc/kernel/entry.h

@@ -69,6 +69,8 @@ void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *,
 			     struct sun4v_1insn_patch_entry *);
 void sun4v_patch_2insn_range(struct sun4v_2insn_patch_entry *,
 			     struct sun4v_2insn_patch_entry *);
+void sun_m7_patch_2insn_range(struct sun4v_2insn_patch_entry *,
+			     struct sun4v_2insn_patch_entry *);
 extern unsigned int dcache_parity_tl1_occurred;
 extern unsigned int icache_parity_tl1_occurred;
 

arch/sparc/kernel/leon_pci_grpci2.c

@@ -723,7 +723,6 @@ static int grpci2_of_probe(struct platform_device *ofdev)
 		err = -ENOMEM;
 		goto err1;
 	}
-	memset(grpci2priv, 0, sizeof(*grpci2priv));
 	priv->regs = regs;
 	priv->irq = ofdev->archdata.irqs[0]; /* BASE IRQ */
 	priv->irq_mode = (capability & STS_IRQMODE) >> STS_IRQMODE_BIT;

arch/sparc/kernel/mdesc.c

@@ -614,45 +614,68 @@ static void fill_in_one_cache(cpuinfo_sparc *c, struct mdesc_handle *hp, u64 mp)
 	}
 }
 
-static void mark_core_ids(struct mdesc_handle *hp, u64 mp, int core_id)
+static void find_back_node_value(struct mdesc_handle *hp, u64 node,
+				 char *srch_val,
+				 void (*func)(struct mdesc_handle *, u64, int),
+				 u64 val, int depth)
 {
-	u64 a;
+	u64 arc;
 
-	mdesc_for_each_arc(a, hp, mp, MDESC_ARC_TYPE_BACK) {
+	/* Since we have an estimate of recursion depth, do a sanity check. */
-		u64 t = mdesc_arc_target(hp, a);
+	if (depth == 0)
-		const char *name;
+		return;
-		const u64 *id;
 
-		name = mdesc_node_name(hp, t);
+	mdesc_for_each_arc(arc, hp, node, MDESC_ARC_TYPE_BACK) {
-		if (!strcmp(name, "cpu")) {
+		u64 n = mdesc_arc_target(hp, arc);
-			id = mdesc_get_property(hp, t, "id", NULL);
+		const char *name = mdesc_node_name(hp, n);
-			if (*id < NR_CPUS)
-				cpu_data(*id).core_id = core_id;
-		} else {
-			u64 j;
 
-			mdesc_for_each_arc(j, hp, t, MDESC_ARC_TYPE_BACK) {
+		if (!strcmp(srch_val, name))
-				u64 n = mdesc_arc_target(hp, j);
+			(*func)(hp, n, val);
-				const char *n_name;
 
-				n_name = mdesc_node_name(hp, n);
+		find_back_node_value(hp, n, srch_val, func, val, depth-1);
-				if (strcmp(n_name, "cpu"))
-					continue;
-
-				id = mdesc_get_property(hp, n, "id", NULL);
-				if (*id < NR_CPUS)
-					cpu_data(*id).core_id = core_id;
-			}
-		}
 	}
 }
 
+static void __mark_core_id(struct mdesc_handle *hp, u64 node,
+			   int core_id)
+{
+	const u64 *id = mdesc_get_property(hp, node, "id", NULL);
+
+	if (*id < num_possible_cpus())
+		cpu_data(*id).core_id = core_id;
+}
+
+static void __mark_sock_id(struct mdesc_handle *hp, u64 node,
+			   int sock_id)
+{
+	const u64 *id = mdesc_get_property(hp, node, "id", NULL);
+
+	if (*id < num_possible_cpus())
+		cpu_data(*id).sock_id = sock_id;
+}
+
+static void mark_core_ids(struct mdesc_handle *hp, u64 mp,
+			  int core_id)
+{
+	find_back_node_value(hp, mp, "cpu", __mark_core_id, core_id, 10);
+}
+
+static void mark_sock_ids(struct mdesc_handle *hp, u64 mp,
+			  int sock_id)
+{
+	find_back_node_value(hp, mp, "cpu", __mark_sock_id, sock_id, 10);
+}
+
 static void set_core_ids(struct mdesc_handle *hp)
 {
 	int idx;
 	u64 mp;
 
 	idx = 1;
+
+	/* Identify unique cores by looking for cpus backpointed to by
+	 * level 1 instruction caches.
+	 */
 	mdesc_for_each_node_by_name(hp, mp, "cache") {
 		const u64 *level;
 		const char *type;
@@ -667,11 +690,72 @@ static void set_core_ids(struct mdesc_handle *hp)
 			continue;
 
 		mark_core_ids(hp, mp, idx);
-
 		idx++;
 	}
 }
 
+static int set_sock_ids_by_cache(struct mdesc_handle *hp, int level)
+{
+	u64 mp;
+	int idx = 1;
+	int fnd = 0;
+
+	/* Identify unique sockets by looking for cpus backpointed to by
+	 * shared level n caches.
+	 */
+	mdesc_for_each_node_by_name(hp, mp, "cache") {
+		const u64 *cur_lvl;
+
+		cur_lvl = mdesc_get_property(hp, mp, "level", NULL);
+		if (*cur_lvl != level)
+			continue;
+
+		mark_sock_ids(hp, mp, idx);
+		idx++;
+		fnd = 1;
+	}
+	return fnd;
+}
+
+static void set_sock_ids_by_socket(struct mdesc_handle *hp, u64 mp)
+{
+	int idx = 1;
+
+	mdesc_for_each_node_by_name(hp, mp, "socket") {
+		u64 a;
+
+		mdesc_for_each_arc(a, hp, mp, MDESC_ARC_TYPE_FWD) {
+			u64 t = mdesc_arc_target(hp, a);
+			const char *name;
+			const u64 *id;
+
+			name = mdesc_node_name(hp, t);
+			if (strcmp(name, "cpu"))
+				continue;
+
+			id = mdesc_get_property(hp, t, "id", NULL);
+			if (*id < num_possible_cpus())
+				cpu_data(*id).sock_id = idx;
+		}
+		idx++;
+	}
+}
+
+static void set_sock_ids(struct mdesc_handle *hp)
+{
+	u64 mp;
+
+	/* If machine description exposes sockets data use it.
+	 * Otherwise fallback to use shared L3 or L2 caches.
+	 */
+	mp = mdesc_node_by_name(hp, MDESC_NODE_NULL, "sockets");
+	if (mp != MDESC_NODE_NULL)
+		return set_sock_ids_by_socket(hp, mp);
+
+	if (!set_sock_ids_by_cache(hp, 3))
+		set_sock_ids_by_cache(hp, 2);
+}
+
 static void mark_proc_ids(struct mdesc_handle *hp, u64 mp, int proc_id)
 {
 	u64 a;
@@ -707,7 +791,6 @@ static void __set_proc_ids(struct mdesc_handle *hp, const char *exec_unit_name)
 			continue;
 
 		mark_proc_ids(hp, mp, idx);
-
 		idx++;
 	}
 }
@@ -900,6 +983,7 @@ void mdesc_fill_in_cpu_data(cpumask_t *mask)
 
 	set_core_ids(hp);
 	set_proc_ids(hp);
+	set_sock_ids(hp);
 
 	mdesc_release(hp);
 

arch/sparc/kernel/pci.c

@@ -1002,6 +1002,38 @@ static int __init pcibios_init(void)
 subsys_initcall(pcibios_init);
 
 #ifdef CONFIG_SYSFS
+
+#define SLOT_NAME_SIZE  11  /* Max decimal digits + null in u32 */
+
+static void pcie_bus_slot_names(struct pci_bus *pbus)
+{
+	struct pci_dev *pdev;
+	struct pci_bus *bus;
+
+	list_for_each_entry(pdev, &pbus->devices, bus_list) {
+		char name[SLOT_NAME_SIZE];
+		struct pci_slot *pci_slot;
+		const u32 *slot_num;
+		int len;
+
+		slot_num = of_get_property(pdev->dev.of_node,
+					   "physical-slot#", &len);
+
+		if (slot_num == NULL || len != 4)
+			continue;
+
+		snprintf(name, sizeof(name), "%u", slot_num[0]);
+		pci_slot = pci_create_slot(pbus, slot_num[0], name, NULL);
+
+		if (IS_ERR(pci_slot))
+			pr_err("PCI: pci_create_slot returned %ld.\n",
+			       PTR_ERR(pci_slot));
+	}
+
+	list_for_each_entry(bus, &pbus->children, node)
+		pcie_bus_slot_names(bus);
+}
+
 static void pci_bus_slot_names(struct device_node *node, struct pci_bus *bus)
 {
 	const struct pci_slot_names {
@@ -1053,18 +1085,29 @@ static int __init of_pci_slot_init(void)
 
 	while ((pbus = pci_find_next_bus(pbus)) != NULL) {
 		struct device_node *node;
+		struct pci_dev *pdev;
 
-		if (pbus->self) {
+		pdev = list_first_entry(&pbus->devices, struct pci_dev,
-			/* PCI->PCI bridge */
+					bus_list);
-			node = pbus->self->dev.of_node;
+
+		if (pdev && pci_is_pcie(pdev)) {
+			pcie_bus_slot_names(pbus);
 		} else {
-			struct pci_pbm_info *pbm = pbus->sysdata;
 
-			/* Host PCI controller */
+			if (pbus->self) {
-			node = pbm->op->dev.of_node;
+
+				/* PCI->PCI bridge */
+				node = pbus->self->dev.of_node;
+
+			} else {
+				struct pci_pbm_info *pbm = pbus->sysdata;
+
+				/* Host PCI controller */
+				node = pbm->op->dev.of_node;
+			}
+
+			pci_bus_slot_names(node, pbus);
 		}
-
-		pci_bus_slot_names(node, pbus);
 	}
 
 	return 0;

arch/sparc/kernel/setup_64.c

@@ -255,6 +255,24 @@ void sun4v_patch_2insn_range(struct sun4v_2insn_patch_entry *start,
 	}
 }
 
+void sun_m7_patch_2insn_range(struct sun4v_2insn_patch_entry *start,
+			     struct sun4v_2insn_patch_entry *end)
+{
+	while (start < end) {
+		unsigned long addr = start->addr;
+
+		*(unsigned int *) (addr +  0) = start->insns[0];
+		wmb();
+		__asm__ __volatile__("flush	%0" : : "r" (addr +  0));
+
+		*(unsigned int *) (addr +  4) = start->insns[1];
+		wmb();
+		__asm__ __volatile__("flush	%0" : : "r" (addr +  4));
+
+		start++;
+	}
+}
+
 static void __init sun4v_patch(void)
 {
 	extern void sun4v_hvapi_init(void);
@@ -267,6 +285,9 @@ static void __init sun4v_patch(void)
 
 	sun4v_patch_2insn_range(&__sun4v_2insn_patch,
 				&__sun4v_2insn_patch_end);
+	if (sun4v_chip_type == SUN4V_CHIP_SPARC_M7)
+		sun_m7_patch_2insn_range(&__sun_m7_2insn_patch,
+					 &__sun_m7_2insn_patch_end);
 
 	sun4v_hvapi_init();
 }

arch/sparc/kernel/smp_64.c

@@ -60,8 +60,12 @@ DEFINE_PER_CPU(cpumask_t, cpu_sibling_map) = CPU_MASK_NONE;
 cpumask_t cpu_core_map[NR_CPUS] __read_mostly =
 	{ [0 ... NR_CPUS-1] = CPU_MASK_NONE };
 
+cpumask_t cpu_core_sib_map[NR_CPUS] __read_mostly = {
+	[0 ... NR_CPUS-1] = CPU_MASK_NONE };
+
 EXPORT_PER_CPU_SYMBOL(cpu_sibling_map);
 EXPORT_SYMBOL(cpu_core_map);
+EXPORT_SYMBOL(cpu_core_sib_map);
 
 static cpumask_t smp_commenced_mask;
 
@@ -1243,6 +1247,15 @@ void smp_fill_in_sib_core_maps(void)
 		}
 	}
 
+	for_each_present_cpu(i)  {
+		unsigned int j;
+
+		for_each_present_cpu(j)  {
+			if (cpu_data(i).sock_id == cpu_data(j).sock_id)
+				cpumask_set_cpu(j, &cpu_core_sib_map[i]);
+		}
+	}
+
 	for_each_present_cpu(i) {
 		unsigned int j;
 

arch/sparc/kernel/vmlinux.lds.S

@@ -138,6 +138,11 @@ SECTIONS
 		*(.pause_3insn_patch)
 		__pause_3insn_patch_end = .;
 	}
+	.sun_m7_2insn_patch : {
+		__sun_m7_2insn_patch = .;
+		*(.sun_m7_2insn_patch)
+		__sun_m7_2insn_patch_end = .;
+	}
 	PERCPU_SECTION(SMP_CACHE_BYTES)
 
 	. = ALIGN(PAGE_SIZE);

arch/sparc/mm/init_64.c

@@ -54,6 +54,7 @@
 #include "init_64.h"
 
 unsigned long kern_linear_pte_xor[4] __read_mostly;
+static unsigned long page_cache4v_flag;
 
 /* A bitmap, two bits for every 256MB of physical memory.  These two
  * bits determine what page size we use for kernel linear
@@ -1909,11 +1910,24 @@ static void __init sun4u_linear_pte_xor_finalize(void)
 
 static void __init sun4v_linear_pte_xor_finalize(void)
 {
+	unsigned long pagecv_flag;
+
+	/* Bit 9 of TTE is no longer CV bit on M7 processor and it instead
+	 * enables MCD error. Do not set bit 9 on M7 processor.
+	 */
+	switch (sun4v_chip_type) {
+	case SUN4V_CHIP_SPARC_M7:
+		pagecv_flag = 0x00;
+		break;
+	default:
+		pagecv_flag = _PAGE_CV_4V;
+		break;
+	}
 #ifndef CONFIG_DEBUG_PAGEALLOC
 	if (cpu_pgsz_mask & HV_PGSZ_MASK_256MB) {
 		kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^
 			PAGE_OFFSET;
-		kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V |
+		kern_linear_pte_xor[1] |= (_PAGE_CP_4V | pagecv_flag |
 					   _PAGE_P_4V | _PAGE_W_4V);
 	} else {
 		kern_linear_pte_xor[1] = kern_linear_pte_xor[0];
@@ -1922,7 +1936,7 @@ static void __init sun4v_linear_pte_xor_finalize(void)
 	if (cpu_pgsz_mask & HV_PGSZ_MASK_2GB) {
 		kern_linear_pte_xor[2] = (_PAGE_VALID | _PAGE_SZ2GB_4V) ^
 			PAGE_OFFSET;
-		kern_linear_pte_xor[2] |= (_PAGE_CP_4V | _PAGE_CV_4V |
+		kern_linear_pte_xor[2] |= (_PAGE_CP_4V | pagecv_flag |
 					   _PAGE_P_4V | _PAGE_W_4V);
 	} else {
 		kern_linear_pte_xor[2] = kern_linear_pte_xor[1];
@@ -1931,7 +1945,7 @@ static void __init sun4v_linear_pte_xor_finalize(void)
 	if (cpu_pgsz_mask & HV_PGSZ_MASK_16GB) {
 		kern_linear_pte_xor[3] = (_PAGE_VALID | _PAGE_SZ16GB_4V) ^
 			PAGE_OFFSET;
-		kern_linear_pte_xor[3] |= (_PAGE_CP_4V | _PAGE_CV_4V |
+		kern_linear_pte_xor[3] |= (_PAGE_CP_4V | pagecv_flag |
 					   _PAGE_P_4V | _PAGE_W_4V);
 	} else {
 		kern_linear_pte_xor[3] = kern_linear_pte_xor[2];
@@ -1958,6 +1972,13 @@ static phys_addr_t __init available_memory(void)
 	return available;
 }
 
+#define _PAGE_CACHE_4U	(_PAGE_CP_4U | _PAGE_CV_4U)
+#define _PAGE_CACHE_4V	(_PAGE_CP_4V | _PAGE_CV_4V)
+#define __DIRTY_BITS_4U	 (_PAGE_MODIFIED_4U | _PAGE_WRITE_4U | _PAGE_W_4U)
+#define __DIRTY_BITS_4V	 (_PAGE_MODIFIED_4V | _PAGE_WRITE_4V | _PAGE_W_4V)
+#define __ACCESS_BITS_4U (_PAGE_ACCESSED_4U | _PAGE_READ_4U | _PAGE_R)
+#define __ACCESS_BITS_4V (_PAGE_ACCESSED_4V | _PAGE_READ_4V | _PAGE_R)
+
 /* We need to exclude reserved regions. This exclusion will include
  * vmlinux and initrd. To be more precise the initrd size could be used to
  * compute a new lower limit because it is freed later during initialization.
@@ -2034,6 +2055,25 @@ void __init paging_init(void)
 	memset(swapper_4m_tsb, 0x40, sizeof(swapper_4m_tsb));
 #endif
 
+	/* TTE.cv bit on sparc v9 occupies the same position as TTE.mcde
+	 * bit on M7 processor. This is a conflicting usage of the same
+	 * bit. Enabling TTE.cv on M7 would turn on Memory Corruption
+	 * Detection error on all pages and this will lead to problems
+	 * later. Kernel does not run with MCD enabled and hence rest
+	 * of the required steps to fully configure memory corruption
+	 * detection are not taken. We need to ensure TTE.mcde is not
+	 * set on M7 processor. Compute the value of cacheability
+	 * flag for use later taking this into consideration.
+	 */
+	switch (sun4v_chip_type) {
+	case SUN4V_CHIP_SPARC_M7:
+		page_cache4v_flag = _PAGE_CP_4V;
+		break;
+	default:
+		page_cache4v_flag = _PAGE_CACHE_4V;
+		break;
+	}
+
 	if (tlb_type == hypervisor)
 		sun4v_pgprot_init();
 	else
@@ -2274,13 +2314,6 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 }
 #endif
 
-#define _PAGE_CACHE_4U	(_PAGE_CP_4U | _PAGE_CV_4U)
-#define _PAGE_CACHE_4V	(_PAGE_CP_4V | _PAGE_CV_4V)
-#define __DIRTY_BITS_4U	 (_PAGE_MODIFIED_4U | _PAGE_WRITE_4U | _PAGE_W_4U)
-#define __DIRTY_BITS_4V	 (_PAGE_MODIFIED_4V | _PAGE_WRITE_4V | _PAGE_W_4V)
-#define __ACCESS_BITS_4U (_PAGE_ACCESSED_4U | _PAGE_READ_4U | _PAGE_R)
-#define __ACCESS_BITS_4V (_PAGE_ACCESSED_4V | _PAGE_READ_4V | _PAGE_R)
-
 pgprot_t PAGE_KERNEL __read_mostly;
 EXPORT_SYMBOL(PAGE_KERNEL);
 
@@ -2312,8 +2345,7 @@ int __meminit vmemmap_populate(unsigned long vstart, unsigned long vend,
 		    _PAGE_P_4U | _PAGE_W_4U);
 	if (tlb_type == hypervisor)
 		pte_base = (_PAGE_VALID | _PAGE_SZ4MB_4V |
-			    _PAGE_CP_4V | _PAGE_CV_4V |
+			    page_cache4v_flag | _PAGE_P_4V | _PAGE_W_4V);
-			    _PAGE_P_4V | _PAGE_W_4V);
 
 	pte_base |= _PAGE_PMD_HUGE;
 
@@ -2450,14 +2482,14 @@ static void __init sun4v_pgprot_init(void)
 	int i;
 
 	PAGE_KERNEL = __pgprot (_PAGE_PRESENT_4V | _PAGE_VALID |
-				_PAGE_CACHE_4V | _PAGE_P_4V |
+				page_cache4v_flag | _PAGE_P_4V |
 				__ACCESS_BITS_4V | __DIRTY_BITS_4V |
 				_PAGE_EXEC_4V);
 	PAGE_KERNEL_LOCKED = PAGE_KERNEL;
 
 	_PAGE_IE = _PAGE_IE_4V;
 	_PAGE_E = _PAGE_E_4V;
-	_PAGE_CACHE = _PAGE_CACHE_4V;
+	_PAGE_CACHE = page_cache4v_flag;
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[0] = _PAGE_VALID ^ PAGE_OFFSET;
@@ -2465,8 +2497,8 @@ static void __init sun4v_pgprot_init(void)
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4V) ^
 		PAGE_OFFSET;
 #endif
-	kern_linear_pte_xor[0] |= (_PAGE_CP_4V | _PAGE_CV_4V |
+	kern_linear_pte_xor[0] |= (page_cache4v_flag | _PAGE_P_4V |
-				   _PAGE_P_4V | _PAGE_W_4V);
+				   _PAGE_W_4V);
 
 	for (i = 1; i < 4; i++)
 		kern_linear_pte_xor[i] = kern_linear_pte_xor[0];
@@ -2479,12 +2511,12 @@ static void __init sun4v_pgprot_init(void)
 			     _PAGE_SZ4MB_4V | _PAGE_SZ512K_4V |
 			     _PAGE_SZ64K_4V | _PAGE_SZ8K_4V);
 
-	page_none = _PAGE_PRESENT_4V | _PAGE_ACCESSED_4V | _PAGE_CACHE_4V;
+	page_none = _PAGE_PRESENT_4V | _PAGE_ACCESSED_4V | page_cache4v_flag;
-	page_shared = (_PAGE_VALID | _PAGE_PRESENT_4V | _PAGE_CACHE_4V |
+	page_shared = (_PAGE_VALID | _PAGE_PRESENT_4V | page_cache4v_flag |
 		       __ACCESS_BITS_4V | _PAGE_WRITE_4V | _PAGE_EXEC_4V);
-	page_copy   = (_PAGE_VALID | _PAGE_PRESENT_4V | _PAGE_CACHE_4V |
+	page_copy   = (_PAGE_VALID | _PAGE_PRESENT_4V | page_cache4v_flag |
 		       __ACCESS_BITS_4V | _PAGE_EXEC_4V);
-	page_readonly = (_PAGE_VALID | _PAGE_PRESENT_4V | _PAGE_CACHE_4V |
+	page_readonly = (_PAGE_VALID | _PAGE_PRESENT_4V | page_cache4v_flag |
 			 __ACCESS_BITS_4V | _PAGE_EXEC_4V);
 
 	page_exec_bit = _PAGE_EXEC_4V;
@@ -2542,7 +2574,7 @@ static unsigned long kern_large_tte(unsigned long paddr)
 	       _PAGE_EXEC_4U | _PAGE_L_4U | _PAGE_W_4U);
 	if (tlb_type == hypervisor)
 		val = (_PAGE_VALID | _PAGE_SZ4MB_4V |
-		       _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_P_4V |
+		       page_cache4v_flag | _PAGE_P_4V |
 		       _PAGE_EXEC_4V | _PAGE_W_4V);
 
 	return val | paddr;

arch/x86/boot/compressed/misc.h

@@ -2,15 +2,14 @@
 #define BOOT_COMPRESSED_MISC_H
 
 /*
- * we have to be careful, because no indirections are allowed here, and
+ * Special hack: we have to be careful, because no indirections are allowed here,
- * paravirt_ops is a kind of one. As it will only run in baremetal anyway,
+ * and paravirt_ops is a kind of one. As it will only run in baremetal anyway,
- * we just keep it from happening
+ * we just keep it from happening. (This list needs to be extended when new
+ * paravirt and debugging variants are added.)
  */
 #undef CONFIG_PARAVIRT
+#undef CONFIG_PARAVIRT_SPINLOCKS
 #undef CONFIG_KASAN
-#ifdef CONFIG_X86_32
-#define _ASM_X86_DESC_H 1
-#endif
 
 #include <linux/linkage.h>
 #include <linux/screen_info.h>

arch/x86/include/asm/kvm_host.h

@@ -207,6 +207,7 @@ union kvm_mmu_page_role {
 		unsigned nxe:1;
 		unsigned cr0_wp:1;
 		unsigned smep_andnot_wp:1;
+		unsigned smap_andnot_wp:1;
 	};
 };
 
@@ -400,6 +401,7 @@ struct kvm_vcpu_arch {
 	struct kvm_mmu_memory_cache mmu_page_header_cache;
 
 	struct fpu guest_fpu;
+	bool eager_fpu;
 	u64 xcr0;
 	u64 guest_supported_xcr0;
 	u32 guest_xstate_size;
@@ -743,6 +745,7 @@ struct kvm_x86_ops {
 	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
 	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
 	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
+	void (*fpu_activate)(struct kvm_vcpu *vcpu);
 	void (*fpu_deactivate)(struct kvm_vcpu *vcpu);
 
 	void (*tlb_flush)(struct kvm_vcpu *vcpu);

arch/x86/include/asm/ptrace.h

@@ -107,7 +107,7 @@ static inline unsigned long regs_return_value(struct pt_regs *regs)
 static inline int user_mode(struct pt_regs *regs)
 {
 #ifdef CONFIG_X86_32
-	return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL;
+	return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >= USER_RPL;
 #else
 	return !!(regs->cs & 3);
 #endif

arch/x86/include/asm/segment.h

@@ -231,11 +231,21 @@
 #define TLS_SIZE			(GDT_ENTRY_TLS_ENTRIES* 8)
 
 #ifdef __KERNEL__
+
+/*
+ * early_idt_handler_array is an array of entry points referenced in the
+ * early IDT.  For simplicity, it's a real array with one entry point
+ * every nine bytes.  That leaves room for an optional 'push $0' if the
+ * vector has no error code (two bytes), a 'push $vector_number' (two
+ * bytes), and a jump to the common entry code (up to five bytes).
+ */
+#define EARLY_IDT_HANDLER_SIZE 9
+
 #ifndef __ASSEMBLY__
 
-extern const char early_idt_handlers[NUM_EXCEPTION_VECTORS][2+2+5];
+extern const char early_idt_handler_array[NUM_EXCEPTION_VECTORS][EARLY_IDT_HANDLER_SIZE];
 #ifdef CONFIG_TRACING
-# define trace_early_idt_handlers early_idt_handlers
+# define trace_early_idt_handler_array early_idt_handler_array
 #endif
 
 /*

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

@@ -140,6 +140,7 @@
 #define MSR_CORE_C3_RESIDENCY		0x000003fc
 #define MSR_CORE_C6_RESIDENCY		0x000003fd
 #define MSR_CORE_C7_RESIDENCY		0x000003fe
+#define MSR_KNL_CORE_C6_RESIDENCY	0x000003ff
 #define MSR_PKG_C2_RESIDENCY		0x0000060d
 #define MSR_PKG_C8_RESIDENCY		0x00000630
 #define MSR_PKG_C9_RESIDENCY		0x00000631

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

@@ -708,6 +708,7 @@ static int mce_no_way_out(struct mce *m, char **msg, unsigned long *validp,
 			  struct pt_regs *regs)
 {
 	int i, ret = 0;
+	char *tmp;
 
 	for (i = 0; i < mca_cfg.banks; i++) {
 		m->status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i));
@@ -716,9 +717,11 @@ static int mce_no_way_out(struct mce *m, char **msg, unsigned long *validp,
 			if (quirk_no_way_out)
 				quirk_no_way_out(i, m, regs);
 		}
-		if (mce_severity(m, mca_cfg.tolerant, msg, true) >=
+
-		    MCE_PANIC_SEVERITY)
+		if (mce_severity(m, mca_cfg.tolerant, &tmp, true) >= MCE_PANIC_SEVERITY) {
+			*msg = tmp;
 			ret = 1;
+		}
 	}
 	return ret;
 }

arch/x86/kernel/cpu/perf_event.c

@@ -190,6 +190,7 @@ static bool check_hw_exists(void)
 	u64 val, val_fail, val_new= ~0;
 	int i, reg, reg_fail, ret = 0;
 	int bios_fail = 0;
+	int reg_safe = -1;
 
 	/*
 	 * Check to see if the BIOS enabled any of the counters, if so
@@ -204,6 +205,8 @@ static bool check_hw_exists(void)
 			bios_fail = 1;
 			val_fail = val;
 			reg_fail = reg;
+		} else {
+			reg_safe = i;
 		}
 	}
 
@@ -221,12 +224,23 @@ static bool check_hw_exists(void)
 		}
 	}
 
+	/*
+	 * If all the counters are enabled, the below test will always
+	 * fail.  The tools will also become useless in this scenario.
+	 * Just fail and disable the hardware counters.
+	 */
+
+	if (reg_safe == -1) {
+		reg = reg_safe;
+		goto msr_fail;
+	}
+
 	/*
 	 * Read the current value, change it and read it back to see if it
 	 * matches, this is needed to detect certain hardware emulators
 	 * (qemu/kvm) that don't trap on the MSR access and always return 0s.
 	 */
-	reg = x86_pmu_event_addr(0);
+	reg = x86_pmu_event_addr(reg_safe);
 	if (rdmsrl_safe(reg, &val))
 		goto msr_fail;
 	val ^= 0xffffUL;
@@ -611,6 +625,7 @@ struct sched_state {
 	int	event;		/* event index */
 	int	counter;	/* counter index */
 	int	unassigned;	/* number of events to be assigned left */
+	int	nr_gp;		/* number of GP counters used */
 	unsigned long used[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
 };
 
@@ -620,27 +635,29 @@ struct sched_state {
 struct perf_sched {
 	int			max_weight;
 	int			max_events;
-	struct perf_event	**events;
+	int			max_gp;
-	struct sched_state	state;
 	int			saved_states;
+	struct event_constraint	**constraints;
+	struct sched_state	state;
 	struct sched_state	saved[SCHED_STATES_MAX];
 };
 
 /*
  * Initialize interator that runs through all events and counters.
  */
-static void perf_sched_init(struct perf_sched *sched, struct perf_event **events,
+static void perf_sched_init(struct perf_sched *sched, struct event_constraint **constraints,
-			    int num, int wmin, int wmax)
+			    int num, int wmin, int wmax, int gpmax)
 {
 	int idx;
 
 	memset(sched, 0, sizeof(*sched));
 	sched->max_events	= num;
 	sched->max_weight	= wmax;
-	sched->events		= events;
+	sched->max_gp		= gpmax;
+	sched->constraints	= constraints;
 
 	for (idx = 0; idx < num; idx++) {
-		if (events[idx]->hw.constraint->weight == wmin)
+		if (constraints[idx]->weight == wmin)
 			break;
 	}
 
@@ -687,7 +704,7 @@ static bool __perf_sched_find_counter(struct perf_sched *sched)
 	if (sched->state.event >= sched->max_events)
 		return false;
 
-	c = sched->events[sched->state.event]->hw.constraint;
+	c = sched->constraints[sched->state.event];
 	/* Prefer fixed purpose counters */
 	if (c->idxmsk64 & (~0ULL << INTEL_PMC_IDX_FIXED)) {
 		idx = INTEL_PMC_IDX_FIXED;
@@ -696,11 +713,16 @@ static bool __perf_sched_find_counter(struct perf_sched *sched)
 				goto done;
 		}
 	}
+
 	/* Grab the first unused counter starting with idx */
 	idx = sched->state.counter;
 	for_each_set_bit_from(idx, c->idxmsk, INTEL_PMC_IDX_FIXED) {
-		if (!__test_and_set_bit(idx, sched->state.used))
+		if (!__test_and_set_bit(idx, sched->state.used)) {
+			if (sched->state.nr_gp++ >= sched->max_gp)
+				return false;
+
 			goto done;
+		}
 	}
 
 	return false;
@@ -745,7 +767,7 @@ static bool perf_sched_next_event(struct perf_sched *sched)
 			if (sched->state.weight > sched->max_weight)
 				return false;
 		}
-		c = sched->events[sched->state.event]->hw.constraint;
+		c = sched->constraints[sched->state.event];
 	} while (c->weight != sched->state.weight);
 
 	sched->state.counter = 0;	/* start with first counter */
@@ -756,12 +778,12 @@ static bool perf_sched_next_event(struct perf_sched *sched)
 /*
  * Assign a counter for each event.
  */
-int perf_assign_events(struct perf_event **events, int n,
+int perf_assign_events(struct event_constraint **constraints, int n,
-			int wmin, int wmax, int *assign)
+			int wmin, int wmax, int gpmax, int *assign)
 {
 	struct perf_sched sched;
 
-	perf_sched_init(&sched, events, n, wmin, wmax);
+	perf_sched_init(&sched, constraints, n, wmin, wmax, gpmax);
 
 	do {
 		if (!perf_sched_find_counter(&sched))
@@ -788,9 +810,9 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 		x86_pmu.start_scheduling(cpuc);
 
 	for (i = 0, wmin = X86_PMC_IDX_MAX, wmax = 0; i < n; i++) {
-		hwc = &cpuc->event_list[i]->hw;
+		cpuc->event_constraint[i] = NULL;
 		c = x86_pmu.get_event_constraints(cpuc, i, cpuc->event_list[i]);
-		hwc->constraint = c;
+		cpuc->event_constraint[i] = c;
 
 		wmin = min(wmin, c->weight);
 		wmax = max(wmax, c->weight);
@@ -801,7 +823,7 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 	 */
 	for (i = 0; i < n; i++) {
 		hwc = &cpuc->event_list[i]->hw;
-		c = hwc->constraint;
+		c = cpuc->event_constraint[i];
 
 		/* never assigned */
 		if (hwc->idx == -1)
@@ -821,9 +843,26 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 	}
 
 	/* slow path */
-	if (i != n)
+	if (i != n) {
-		unsched = perf_assign_events(cpuc->event_list, n, wmin,
+		int gpmax = x86_pmu.num_counters;
-					     wmax, assign);
+
+		/*
+		 * Do not allow scheduling of more than half the available
+		 * generic counters.
+		 *
+		 * This helps avoid counter starvation of sibling thread by
+		 * ensuring at most half the counters cannot be in exclusive
+		 * mode. There is no designated counters for the limits. Any
+		 * N/2 counters can be used. This helps with events with
+		 * specific counter constraints.
+		 */
+		if (is_ht_workaround_enabled() && !cpuc->is_fake &&
+		    READ_ONCE(cpuc->excl_cntrs->exclusive_present))
+			gpmax /= 2;
+
+		unsched = perf_assign_events(cpuc->event_constraint, n, wmin,
+					     wmax, gpmax, assign);
+	}
 
 	/*
 	 * In case of success (unsched = 0), mark events as committed,
@@ -840,7 +879,7 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 			e = cpuc->event_list[i];
 			e->hw.flags |= PERF_X86_EVENT_COMMITTED;
 			if (x86_pmu.commit_scheduling)
-				x86_pmu.commit_scheduling(cpuc, e, assign[i]);
+				x86_pmu.commit_scheduling(cpuc, i, assign[i]);
 		}
 	}
 
@@ -1292,8 +1331,10 @@ static void x86_pmu_del(struct perf_event *event, int flags)
 		x86_pmu.put_event_constraints(cpuc, event);
 
 	/* Delete the array entry. */
-	while (++i < cpuc->n_events)
+	while (++i < cpuc->n_events) {
 		cpuc->event_list[i-1] = cpuc->event_list[i];
+		cpuc->event_constraint[i-1] = cpuc->event_constraint[i];
+	}
 	--cpuc->n_events;
 
 	perf_event_update_userpage(event);

arch/x86/kernel/cpu/perf_event.h

@@ -74,6 +74,7 @@ struct event_constraint {
 #define PERF_X86_EVENT_EXCL		0x0040 /* HT exclusivity on counter */
 #define PERF_X86_EVENT_DYNAMIC		0x0080 /* dynamic alloc'd constraint */
 #define PERF_X86_EVENT_RDPMC_ALLOWED	0x0100 /* grant rdpmc permission */
+#define PERF_X86_EVENT_EXCL_ACCT	0x0200 /* accounted EXCL event */
 
 
 struct amd_nb {
@@ -134,8 +135,6 @@ enum intel_excl_state_type {
 struct intel_excl_states {
 	enum intel_excl_state_type init_state[X86_PMC_IDX_MAX];
 	enum intel_excl_state_type state[X86_PMC_IDX_MAX];
-	int  num_alloc_cntrs;/* #counters allocated */
-	int  max_alloc_cntrs;/* max #counters allowed */
 	bool sched_started; /* true if scheduling has started */
 };
 
@@ -144,6 +143,11 @@ struct intel_excl_cntrs {
 
 	struct intel_excl_states states[2];
 
+	union {
+		u16	has_exclusive[2];
+		u32	exclusive_present;
+	};
+
 	int		refcnt;		/* per-core: #HT threads */
 	unsigned	core_id;	/* per-core: core id */
 };
@@ -172,7 +176,11 @@ struct cpu_hw_events {
 					     added in the current transaction */
 	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
 	u64			tags[X86_PMC_IDX_MAX];
+
 	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
+	struct event_constraint	*event_constraint[X86_PMC_IDX_MAX];
+
+	int			n_excl; /* the number of exclusive events */
 
 	unsigned int		group_flag;
 	int			is_fake;
@@ -519,9 +527,7 @@ struct x86_pmu {
 	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
 						 struct perf_event *event);
 
-	void		(*commit_scheduling)(struct cpu_hw_events *cpuc,
+	void		(*commit_scheduling)(struct cpu_hw_events *cpuc, int idx, int cntr);
-					     struct perf_event *event,
-					     int cntr);
 
 	void		(*start_scheduling)(struct cpu_hw_events *cpuc);
 
@@ -717,8 +723,8 @@ static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
 
 void x86_pmu_enable_all(int added);
 
-int perf_assign_events(struct perf_event **events, int n,
+int perf_assign_events(struct event_constraint **constraints, int n,
-			int wmin, int wmax, int *assign);
+			int wmin, int wmax, int gpmax, int *assign);
 int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign);
 
 void x86_pmu_stop(struct perf_event *event, int flags);
@@ -929,4 +935,8 @@ static inline struct intel_shared_regs *allocate_shared_regs(int cpu)
 	return NULL;
 }
 
+static inline int is_ht_workaround_enabled(void)
+{
+	return 0;
+}
 #endif /* CONFIG_CPU_SUP_INTEL */

arch/x86/kernel/cpu/perf_event_intel.c

@@ -1923,7 +1923,6 @@ intel_start_scheduling(struct cpu_hw_events *cpuc)
 	xl = &excl_cntrs->states[tid];
 
 	xl->sched_started = true;
-	xl->num_alloc_cntrs = 0;
 	/*
 	 * lock shared state until we are done scheduling
 	 * in stop_event_scheduling()
@@ -2000,6 +1999,11 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
 	 * across HT threads
 	 */
 	is_excl = c->flags & PERF_X86_EVENT_EXCL;
+	if (is_excl && !(event->hw.flags & PERF_X86_EVENT_EXCL_ACCT)) {
+		event->hw.flags |= PERF_X86_EVENT_EXCL_ACCT;
+		if (!cpuc->n_excl++)
+			WRITE_ONCE(excl_cntrs->has_exclusive[tid], 1);
+	}
 
 	/*
 	 * xl = state of current HT
@@ -2008,18 +2012,6 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
 	xl = &excl_cntrs->states[tid];
 	xlo = &excl_cntrs->states[o_tid];
 
-	/*
-	 * do not allow scheduling of more than max_alloc_cntrs
-	 * which is set to half the available generic counters.
-	 * this helps avoid counter starvation of sibling thread
-	 * by ensuring at most half the counters cannot be in
-	 * exclusive mode. There is not designated counters for the
-	 * limits. Any N/2 counters can be used. This helps with
-	 * events with specifix counter constraints
-	 */
-	if (xl->num_alloc_cntrs++ == xl->max_alloc_cntrs)
-		return &emptyconstraint;
-
 	cx = c;
 
 	/*
@@ -2106,7 +2098,7 @@ static struct event_constraint *
 intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
 			    struct perf_event *event)
 {
-	struct event_constraint *c1 = event->hw.constraint;
+	struct event_constraint *c1 = cpuc->event_constraint[idx];
 	struct event_constraint *c2;
 
 	/*
@@ -2150,6 +2142,11 @@ static void intel_put_excl_constraints(struct cpu_hw_events *cpuc,
 
 	xl = &excl_cntrs->states[tid];
 	xlo = &excl_cntrs->states[o_tid];
+	if (hwc->flags & PERF_X86_EVENT_EXCL_ACCT) {
+		hwc->flags &= ~PERF_X86_EVENT_EXCL_ACCT;
+		if (!--cpuc->n_excl)
+			WRITE_ONCE(excl_cntrs->has_exclusive[tid], 0);
+	}
 
 	/*
 	 * put_constraint may be called from x86_schedule_events()
@@ -2188,8 +2185,6 @@ intel_put_shared_regs_event_constraints(struct cpu_hw_events *cpuc,
 static void intel_put_event_constraints(struct cpu_hw_events *cpuc,
 					struct perf_event *event)
 {
-	struct event_constraint *c = event->hw.constraint;
-
 	intel_put_shared_regs_event_constraints(cpuc, event);
 
 	/*
@@ -2197,19 +2192,14 @@ static void intel_put_event_constraints(struct cpu_hw_events *cpuc,
 	 * all events are subject to and must call the
 	 * put_excl_constraints() routine
 	 */
-	if (c && cpuc->excl_cntrs)
+	if (cpuc->excl_cntrs)
 		intel_put_excl_constraints(cpuc, event);
-
-	/* cleanup dynamic constraint */
-	if (c && (c->flags & PERF_X86_EVENT_DYNAMIC))
-		event->hw.constraint = NULL;
 }
 
-static void intel_commit_scheduling(struct cpu_hw_events *cpuc,
+static void intel_commit_scheduling(struct cpu_hw_events *cpuc, int idx, int cntr)
-				    struct perf_event *event, int cntr)
 {
 	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
-	struct event_constraint *c = event->hw.constraint;
+	struct event_constraint *c = cpuc->event_constraint[idx];
 	struct intel_excl_states *xlo, *xl;
 	int tid = cpuc->excl_thread_id;
 	int o_tid = 1 - tid;
@@ -2639,8 +2629,6 @@ static void intel_pmu_cpu_starting(int cpu)
 		cpuc->lbr_sel = &cpuc->shared_regs->regs[EXTRA_REG_LBR];
 
 	if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) {
-		int h = x86_pmu.num_counters >> 1;
-
 		for_each_cpu(i, topology_thread_cpumask(cpu)) {
 			struct intel_excl_cntrs *c;
 
@@ -2654,11 +2642,6 @@ static void intel_pmu_cpu_starting(int cpu)
 		}
 		cpuc->excl_cntrs->core_id = core_id;
 		cpuc->excl_cntrs->refcnt++;
-		/*
-		 * set hard limit to half the number of generic counters
-		 */
-		cpuc->excl_cntrs->states[0].max_alloc_cntrs = h;
-		cpuc->excl_cntrs->states[1].max_alloc_cntrs = h;
 	}
 }
 

arch/x86/kernel/cpu/perf_event_intel_ds.c

@@ -706,9 +706,9 @@ void intel_pmu_pebs_disable(struct perf_event *event)
 
 	cpuc->pebs_enabled &= ~(1ULL << hwc->idx);
 
-	if (event->hw.constraint->flags & PERF_X86_EVENT_PEBS_LDLAT)
+	if (event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT)
 		cpuc->pebs_enabled &= ~(1ULL << (hwc->idx + 32));
-	else if (event->hw.constraint->flags & PERF_X86_EVENT_PEBS_ST)
+	else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST)
 		cpuc->pebs_enabled &= ~(1ULL << 63);
 
 	if (cpuc->enabled)

arch/x86/kernel/cpu/perf_event_intel_pt.c

@@ -151,7 +151,7 @@ static int __init pt_pmu_hw_init(void)
 
 		de_attr->attr.attr.name = pt_caps[i].name;
 
-		sysfs_attr_init(&de_attrs->attr.attr);
+		sysfs_attr_init(&de_attr->attr.attr);
 
 		de_attr->attr.attr.mode		= S_IRUGO;
 		de_attr->attr.show		= pt_cap_show;
@@ -615,7 +615,8 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf,
 				   struct perf_output_handle *handle)
 
 {
-	unsigned long idx, npages, end;
+	unsigned long head = local64_read(&buf->head);
+	unsigned long idx, npages, wakeup;
 
 	if (buf->snapshot)
 		return 0;
@@ -634,17 +635,26 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf,
 	buf->topa_index[buf->stop_pos]->stop = 0;
 	buf->topa_index[buf->intr_pos]->intr = 0;
 
-	if (pt_cap_get(PT_CAP_topa_multiple_entries)) {
+	/* how many pages till the STOP marker */
-		npages = (handle->size + 1) >> PAGE_SHIFT;
+	npages = handle->size >> PAGE_SHIFT;
-		end = (local64_read(&buf->head) >> PAGE_SHIFT) + npages;
+
-		/*if (end > handle->wakeup >> PAGE_SHIFT)
+	/* if it's on a page boundary, fill up one more page */
-		  end = handle->wakeup >> PAGE_SHIFT;*/
+	if (!offset_in_page(head + handle->size + 1))
-		idx = end & (buf->nr_pages - 1);
+		npages++;
-		buf->stop_pos = idx;
+
-		idx = (local64_read(&buf->head) >> PAGE_SHIFT) + npages - 1;
+	idx = (head >> PAGE_SHIFT) + npages;
-		idx &= buf->nr_pages - 1;
+	idx &= buf->nr_pages - 1;
-		buf->intr_pos = idx;
+	buf->stop_pos = idx;
-	}
+
+	wakeup = handle->wakeup >> PAGE_SHIFT;
+
+	/* in the worst case, wake up the consumer one page before hard stop */
+	idx = (head >> PAGE_SHIFT) + npages - 1;
+	if (idx > wakeup)
+		idx = wakeup;
+
+	idx &= buf->nr_pages - 1;
+	buf->intr_pos = idx;
 
 	buf->topa_index[buf->stop_pos]->stop = 1;
 	buf->topa_index[buf->intr_pos]->intr = 1;

arch/x86/kernel/cpu/perf_event_intel_uncore.c

@@ -365,9 +365,8 @@ static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int
 	bitmap_zero(used_mask, UNCORE_PMC_IDX_MAX);
 
 	for (i = 0, wmin = UNCORE_PMC_IDX_MAX, wmax = 0; i < n; i++) {
-		hwc = &box->event_list[i]->hw;
 		c = uncore_get_event_constraint(box, box->event_list[i]);
-		hwc->constraint = c;
+		box->event_constraint[i] = c;
 		wmin = min(wmin, c->weight);
 		wmax = max(wmax, c->weight);
 	}
@@ -375,7 +374,7 @@ static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int
 	/* fastpath, try to reuse previous register */
 	for (i = 0; i < n; i++) {
 		hwc = &box->event_list[i]->hw;
-		c = hwc->constraint;
+		c = box->event_constraint[i];
 
 		/* never assigned */
 		if (hwc->idx == -1)
@@ -395,8 +394,8 @@ static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int
 	}
 	/* slow path */
 	if (i != n)
-		ret = perf_assign_events(box->event_list, n,
+		ret = perf_assign_events(box->event_constraint, n,
-					 wmin, wmax, assign);
+					 wmin, wmax, n, assign);
 
 	if (!assign || ret) {
 		for (i = 0; i < n; i++)

arch/x86/kernel/cpu/perf_event_intel_uncore.h

@@ -97,6 +97,7 @@ struct intel_uncore_box {
 	atomic_t refcnt;
 	struct perf_event *events[UNCORE_PMC_IDX_MAX];
 	struct perf_event *event_list[UNCORE_PMC_IDX_MAX];
+	struct event_constraint *event_constraint[UNCORE_PMC_IDX_MAX];
 	unsigned long active_mask[BITS_TO_LONGS(UNCORE_PMC_IDX_MAX)];
 	u64 tags[UNCORE_PMC_IDX_MAX];
 	struct pci_dev *pci_dev;

arch/x86/kernel/head64.c

@@ -167,7 +167,7 @@ asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data)
 	clear_bss();
 
 	for (i = 0; i < NUM_EXCEPTION_VECTORS; i++)
-		set_intr_gate(i, early_idt_handlers[i]);
+		set_intr_gate(i, early_idt_handler_array[i]);
 	load_idt((const struct desc_ptr *)&idt_descr);
 
 	copy_bootdata(__va(real_mode_data));

arch/x86/kernel/head_32.S

@@ -478,21 +478,22 @@ is486:
 __INIT
 setup_once:
 	/*
-	 * Set up a idt with 256 entries pointing to ignore_int,
+	 * Set up a idt with 256 interrupt gates that push zero if there
-	 * interrupt gates. It doesn't actually load idt - that needs
+	 * is no error code and then jump to early_idt_handler_common.
-	 * to be done on each CPU. Interrupts are enabled elsewhere,
+	 * It doesn't actually load the idt - that needs to be done on
-	 * when we can be relatively sure everything is ok.
+	 * each CPU. Interrupts are enabled elsewhere, when we can be
+	 * relatively sure everything is ok.
 	 */
 
 	movl $idt_table,%edi
-	movl $early_idt_handlers,%eax
+	movl $early_idt_handler_array,%eax
 	movl $NUM_EXCEPTION_VECTORS,%ecx
 1:
 	movl %eax,(%edi)
 	movl %eax,4(%edi)
 	/* interrupt gate, dpl=0, present */
 	movl $(0x8E000000 + __KERNEL_CS),2(%edi)
-	addl $9,%eax
+	addl $EARLY_IDT_HANDLER_SIZE,%eax
 	addl $8,%edi
 	loop 1b
 
@@ -524,26 +525,28 @@ setup_once:
 	andl $0,setup_once_ref	/* Once is enough, thanks */
 	ret
 
-ENTRY(early_idt_handlers)
+ENTRY(early_idt_handler_array)
 	# 36(%esp) %eflags
 	# 32(%esp) %cs
 	# 28(%esp) %eip
 	# 24(%rsp) error code
 	i = 0
 	.rept NUM_EXCEPTION_VECTORS
-	.if (EXCEPTION_ERRCODE_MASK >> i) & 1
+	.ifeq (EXCEPTION_ERRCODE_MASK >> i) & 1
-	ASM_NOP2
-	.else
 	pushl $0		# Dummy error code, to make stack frame uniform
 	.endif
 	pushl $i		# 20(%esp) Vector number
-	jmp early_idt_handler
+	jmp early_idt_handler_common
 	i = i + 1
+	.fill early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
 	.endr
-ENDPROC(early_idt_handlers)
+ENDPROC(early_idt_handler_array)
 	
-	/* This is global to keep gas from relaxing the jumps */
+early_idt_handler_common:
-ENTRY(early_idt_handler)
+	/*
+	 * The stack is the hardware frame, an error code or zero, and the
+	 * vector number.
+	 */
 	cld
 
 	cmpl $2,(%esp)		# X86_TRAP_NMI
@@ -603,7 +606,7 @@ ex_entry:
 is_nmi:
 	addl $8,%esp		/* drop vector number and error code */
 	iret
-ENDPROC(early_idt_handler)
+ENDPROC(early_idt_handler_common)
 
 /* This is the default interrupt "handler" :-) */
 	ALIGN

arch/x86/kernel/head_64.S

@@ -321,26 +321,28 @@ bad_address:
 	jmp bad_address
 
 	__INIT
-	.globl early_idt_handlers
+ENTRY(early_idt_handler_array)
-early_idt_handlers:
 	# 104(%rsp) %rflags
 	#  96(%rsp) %cs
 	#  88(%rsp) %rip
 	#  80(%rsp) error code
 	i = 0
 	.rept NUM_EXCEPTION_VECTORS
-	.if (EXCEPTION_ERRCODE_MASK >> i) & 1
+	.ifeq (EXCEPTION_ERRCODE_MASK >> i) & 1
-	ASM_NOP2
-	.else
 	pushq $0		# Dummy error code, to make stack frame uniform
 	.endif
 	pushq $i		# 72(%rsp) Vector number
-	jmp early_idt_handler
+	jmp early_idt_handler_common
 	i = i + 1
+	.fill early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
 	.endr
+ENDPROC(early_idt_handler_array)
 
-/* This is global to keep gas from relaxing the jumps */
+early_idt_handler_common:
-ENTRY(early_idt_handler)
+	/*
+	 * The stack is the hardware frame, an error code or zero, and the
+	 * vector number.
+	 */
 	cld
 
 	cmpl $2,(%rsp)		# X86_TRAP_NMI
@@ -412,7 +414,7 @@ ENTRY(early_idt_handler)
 is_nmi:
 	addq $16,%rsp		# drop vector number and error code
 	INTERRUPT_RETURN
-ENDPROC(early_idt_handler)
+ENDPROC(early_idt_handler_common)
 
 	__INITDATA
 

arch/x86/kernel/i387.c

@@ -173,6 +173,21 @@ static void init_thread_xstate(void)
 		xstate_size = sizeof(struct i387_fxsave_struct);
 	else
 		xstate_size = sizeof(struct i387_fsave_struct);
+
+	/*
+	 * Quirk: we don't yet handle the XSAVES* instructions
+	 * correctly, as we don't correctly convert between
+	 * standard and compacted format when interfacing
+	 * with user-space - so disable it for now.
+	 *
+	 * The difference is small: with recent CPUs the
+	 * compacted format is only marginally smaller than
+	 * the standard FPU state format.
+	 *
+	 * ( This is easy to backport while we are fixing
+	 *   XSAVES* support. )
+	 */
+	setup_clear_cpu_cap(X86_FEATURE_XSAVES);
 }
 
 /*

arch/x86/kvm/cpuid.c

@@ -16,6 +16,8 @@
 #include <linux/module.h>
 #include <linux/vmalloc.h>
 #include <linux/uaccess.h>
+#include <asm/i387.h> /* For use_eager_fpu.  Ugh! */
+#include <asm/fpu-internal.h> /* For use_eager_fpu.  Ugh! */
 #include <asm/user.h>
 #include <asm/xsave.h>
 #include "cpuid.h"
@@ -95,6 +97,8 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu)
 	if (best && (best->eax & (F(XSAVES) | F(XSAVEC))))
 		best->ebx = xstate_required_size(vcpu->arch.xcr0, true);
 
+	vcpu->arch.eager_fpu = guest_cpuid_has_mpx(vcpu);
+
 	/*
 	 * The existing code assumes virtual address is 48-bit in the canonical
 	 * address checks; exit if it is ever changed.

arch/x86/kvm/cpuid.h

@@ -117,4 +117,12 @@ static inline bool guest_cpuid_has_rtm(struct kvm_vcpu *vcpu)
 	best = kvm_find_cpuid_entry(vcpu, 7, 0);
 	return best && (best->ebx & bit(X86_FEATURE_RTM));
 }
+
+static inline bool guest_cpuid_has_mpx(struct kvm_vcpu *vcpu)
+{
+	struct kvm_cpuid_entry2 *best;
+
+	best = kvm_find_cpuid_entry(vcpu, 7, 0);
+	return best && (best->ebx & bit(X86_FEATURE_MPX));
+}
 #endif

arch/x86/kvm/mmu.c

@@ -3736,8 +3736,8 @@ static void reset_rsvds_bits_mask_ept(struct kvm_vcpu *vcpu,
 	}
 }
 
-void update_permission_bitmask(struct kvm_vcpu *vcpu,
+static void update_permission_bitmask(struct kvm_vcpu *vcpu,
-		struct kvm_mmu *mmu, bool ept)
+				      struct kvm_mmu *mmu, bool ept)
 {
 	unsigned bit, byte, pfec;
 	u8 map;
@@ -3918,6 +3918,7 @@ static void init_kvm_tdp_mmu(struct kvm_vcpu *vcpu)
 void kvm_init_shadow_mmu(struct kvm_vcpu *vcpu)
 {
 	bool smep = kvm_read_cr4_bits(vcpu, X86_CR4_SMEP);
+	bool smap = kvm_read_cr4_bits(vcpu, X86_CR4_SMAP);
 	struct kvm_mmu *context = &vcpu->arch.mmu;
 
 	MMU_WARN_ON(VALID_PAGE(context->root_hpa));
@@ -3936,6 +3937,8 @@ void kvm_init_shadow_mmu(struct kvm_vcpu *vcpu)
 	context->base_role.cr0_wp  = is_write_protection(vcpu);
 	context->base_role.smep_andnot_wp
 		= smep && !is_write_protection(vcpu);
+	context->base_role.smap_andnot_wp
+		= smap && !is_write_protection(vcpu);
 }
 EXPORT_SYMBOL_GPL(kvm_init_shadow_mmu);
 
@@ -4207,12 +4210,18 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 		       const u8 *new, int bytes)
 {
 	gfn_t gfn = gpa >> PAGE_SHIFT;
-	union kvm_mmu_page_role mask = { .word = 0 };
 	struct kvm_mmu_page *sp;
 	LIST_HEAD(invalid_list);
 	u64 entry, gentry, *spte;
 	int npte;
 	bool remote_flush, local_flush, zap_page;
+	union kvm_mmu_page_role mask = (union kvm_mmu_page_role) {
+		.cr0_wp = 1,
+		.cr4_pae = 1,
+		.nxe = 1,
+		.smep_andnot_wp = 1,
+		.smap_andnot_wp = 1,
+	};
 
 	/*
 	 * If we don't have indirect shadow pages, it means no page is
@@ -4238,7 +4247,6 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
 	++vcpu->kvm->stat.mmu_pte_write;
 	kvm_mmu_audit(vcpu, AUDIT_PRE_PTE_WRITE);
 
-	mask.cr0_wp = mask.cr4_pae = mask.nxe = 1;
 	for_each_gfn_indirect_valid_sp(vcpu->kvm, sp, gfn) {
 		if (detect_write_misaligned(sp, gpa, bytes) ||
 		      detect_write_flooding(sp)) {

arch/x86/kvm/mmu.h

@@ -71,8 +71,6 @@ enum {
 int handle_mmio_page_fault_common(struct kvm_vcpu *vcpu, u64 addr, bool direct);
 void kvm_init_shadow_mmu(struct kvm_vcpu *vcpu);
 void kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, bool execonly);
-void update_permission_bitmask(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
-		bool ept);
 
 static inline unsigned int kvm_mmu_available_pages(struct kvm *kvm)
 {
@@ -166,6 +164,8 @@ static inline bool permission_fault(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
 	int index = (pfec >> 1) +
 		    (smap >> (X86_EFLAGS_AC_BIT - PFERR_RSVD_BIT + 1));
 
+	WARN_ON(pfec & PFERR_RSVD_MASK);
+
 	return (mmu->permissions[index] >> pte_access) & 1;
 }
 

arch/x86/kvm/paging_tmpl.h

@@ -718,6 +718,13 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
 					      mmu_is_nested(vcpu));
 		if (likely(r != RET_MMIO_PF_INVALID))
 			return r;
+
+		/*
+		 * page fault with PFEC.RSVD  = 1 is caused by shadow
+		 * page fault, should not be used to walk guest page
+		 * table.
+		 */
+		error_code &= ~PFERR_RSVD_MASK;
 	};
 
 	r = mmu_topup_memory_caches(vcpu);