Merge branch 'sa11x0-mcp-fixes' into fixes
diff --git a/Makefile b/Makefile
index 156ac69..71e6ed2 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 3
-PATCHLEVEL = 2
+PATCHLEVEL = 3
 SUBLEVEL = 0
-EXTRAVERSION =
+EXTRAVERSION = -rc1
 NAME = Saber-toothed Squirrel
 
 # *DOCUMENTATION*
diff --git a/arch/arm/mach-sa1100/assabet.c b/arch/arm/mach-sa1100/assabet.c
index ebafe8a..0c4b76a 100644
--- a/arch/arm/mach-sa1100/assabet.c
+++ b/arch/arm/mach-sa1100/assabet.c
@@ -202,7 +202,6 @@
 static struct mcp_plat_data assabet_mcp_data = {
 	.mccr0		= MCCR0_ADM,
 	.sclk_rate	= 11981000,
-	.codec		= "ucb1x00",
 };
 
 static void __init assabet_init(void)
@@ -253,17 +252,6 @@
 	sa11x0_register_mtd(&assabet_flash_data, assabet_flash_resources,
 			    ARRAY_SIZE(assabet_flash_resources));
 	sa11x0_register_irda(&assabet_irda_data);
-
-	/*
-	 * Setup the PPC unit correctly.
-	 */
-	PPDR &= ~PPC_RXD4;
-	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
-	PSDR |= PPC_RXD4;
-	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-
-	ASSABET_BCR_set(ASSABET_BCR_CODEC_RST);
 	sa11x0_register_mcp(&assabet_mcp_data);
 }
 
diff --git a/arch/arm/mach-sa1100/cerf.c b/arch/arm/mach-sa1100/cerf.c
index d12d0f4..11bb6d0 100644
--- a/arch/arm/mach-sa1100/cerf.c
+++ b/arch/arm/mach-sa1100/cerf.c
@@ -124,23 +124,12 @@
 static struct mcp_plat_data cerf_mcp_data = {
 	.mccr0		= MCCR0_ADM,
 	.sclk_rate	= 11981000,
-	.codec		= "ucb1x00",
 };
 
 static void __init cerf_init(void)
 {
 	platform_add_devices(cerf_devices, ARRAY_SIZE(cerf_devices));
 	sa11x0_register_mtd(&cerf_flash_data, &cerf_flash_resource, 1);
-
-	/*
-	 * Setup the PPC unit correctly.
-	 */
-	PPDR &= ~PPC_RXD4;
-	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
-	PSDR |= PPC_RXD4;
-	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-
 	sa11x0_register_mcp(&cerf_mcp_data);
 }
 
diff --git a/arch/arm/mach-sa1100/collie.c b/arch/arm/mach-sa1100/collie.c
index cce8763..fd56521 100644
--- a/arch/arm/mach-sa1100/collie.c
+++ b/arch/arm/mach-sa1100/collie.c
@@ -27,7 +27,6 @@
 #include <linux/timer.h>
 #include <linux/gpio.h>
 #include <linux/pda_power.h>
-#include <linux/mfd/ucb1x00.h>
 
 #include <mach/hardware.h>
 #include <asm/mach-types.h>
@@ -86,15 +85,10 @@
 	.num_devs	= 1,
 };
 
-static struct ucb1x00_plat_data collie_ucb1x00_data = {
-	.gpio_base	= COLLIE_TC35143_GPIO_BASE,
-};
-
 static struct mcp_plat_data collie_mcp_data = {
 	.mccr0		= MCCR0_ADM | MCCR0_ExtClk,
 	.sclk_rate	= 9216000,
-	.codec		= "ucb1x00",
-	.codec_pdata	= &collie_ucb1x00_data,
+	.gpio_base	= COLLIE_TC35143_GPIO_BASE,
 };
 
 /*
@@ -356,16 +350,6 @@
 
 	sa11x0_register_mtd(&collie_flash_data, collie_flash_resources,
 			    ARRAY_SIZE(collie_flash_resources));
-
-	/*
-	 * Setup the PPC unit correctly.
-	 */
-	PPDR &= ~PPC_RXD4;
-	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
-	PSDR |= PPC_RXD4;
-	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-
 	sa11x0_register_mcp(&collie_mcp_data);
 
 	sharpsl_save_param();
diff --git a/arch/arm/mach-sa1100/generic.c b/arch/arm/mach-sa1100/generic.c
index a7c0df6..bb10ee2 100644
--- a/arch/arm/mach-sa1100/generic.c
+++ b/arch/arm/mach-sa1100/generic.c
@@ -217,15 +217,10 @@
 static struct resource sa11x0mcp_resources[] = {
 	[0] = {
 		.start	= __PREG(Ser4MCCR0),
-		.end	= __PREG(Ser4MCCR0) + 0x1C - 1,
+		.end	= __PREG(Ser4MCCR0) + 0xffff,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
-		.start	= __PREG(Ser4MCCR1),
-		.end	= __PREG(Ser4MCCR1) + 0x4 - 1,
-		.flags	= IORESOURCE_MEM,
-	},
-	[2] = {
 		.start	= IRQ_Ser4MCP,
 		.end	= IRQ_Ser4MCP,
 		.flags	= IORESOURCE_IRQ,
diff --git a/arch/arm/mach-sa1100/include/mach/mcp.h b/arch/arm/mach-sa1100/include/mach/mcp.h
index 586cec8..ed1a331 100644
--- a/arch/arm/mach-sa1100/include/mach/mcp.h
+++ b/arch/arm/mach-sa1100/include/mach/mcp.h
@@ -17,8 +17,6 @@
 	u32 mccr1;
 	unsigned int sclk_rate;
 	int gpio_base;
-	const char *codec;
-	void *codec_pdata;
 };
 
 #endif
diff --git a/arch/arm/mach-sa1100/lart.c b/arch/arm/mach-sa1100/lart.c
index d117cea..af4e276 100644
--- a/arch/arm/mach-sa1100/lart.c
+++ b/arch/arm/mach-sa1100/lart.c
@@ -24,20 +24,10 @@
 static struct mcp_plat_data lart_mcp_data = {
 	.mccr0		= MCCR0_ADM,
 	.sclk_rate	= 11981000,
-	.codec		= "ucb1x00",
 };
 
 static void __init lart_init(void)
 {
-	/*
-	 * Setup the PPC unit correctly.
-	 */
-	PPDR &= ~PPC_RXD4;
-	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
-	PSDR |= PPC_RXD4;
-	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-
 	sa11x0_register_mcp(&lart_mcp_data);
 }
 
diff --git a/arch/arm/mach-sa1100/shannon.c b/arch/arm/mach-sa1100/shannon.c
index 748d344..318b2b7 100644
--- a/arch/arm/mach-sa1100/shannon.c
+++ b/arch/arm/mach-sa1100/shannon.c
@@ -55,22 +55,11 @@
 static struct mcp_plat_data shannon_mcp_data = {
 	.mccr0		= MCCR0_ADM,
 	.sclk_rate	= 11981000,
-	.codec		= "ucb1x00",
 };
 
 static void __init shannon_init(void)
 {
 	sa11x0_register_mtd(&shannon_flash_data, &shannon_flash_resource, 1);
-
-	/*
-	 * Setup the PPC unit correctly.
-	 */
-	PPDR &= ~PPC_RXD4;
-	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
-	PSDR |= PPC_RXD4;
-	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-
 	sa11x0_register_mcp(&shannon_mcp_data);
 }
 
diff --git a/arch/arm/mach-sa1100/simpad.c b/arch/arm/mach-sa1100/simpad.c
index 458ecec..e17c04d 100644
--- a/arch/arm/mach-sa1100/simpad.c
+++ b/arch/arm/mach-sa1100/simpad.c
@@ -14,7 +14,6 @@
 #include <linux/mtd/partitions.h>
 #include <linux/io.h>
 #include <linux/gpio.h>
-#include <linux/mfd/ucb1x00.h>
 
 #include <asm/irq.h>
 #include <mach/hardware.h>
@@ -188,15 +187,10 @@
 	}
 };
 
-static struct ucb1x00_plat_data simpad_ucb1x00_data = {
-	.gpio_base	= SIMPAD_UCB1X00_GPIO_BASE,
-};
-
 static struct mcp_plat_data simpad_mcp_data = {
 	.mccr0		= MCCR0_ADM,
 	.sclk_rate	= 11981000,
-	.codec		= "ucb1300",
-	.codec_pdata	= &simpad_ucb1x00_data,
+	.gpio_base	= SIMPAD_UCB1X00_GPIO_BASE,
 };
 
 
@@ -384,16 +378,6 @@
 
 	sa11x0_register_mtd(&simpad_flash_data, simpad_flash_resources,
 			      ARRAY_SIZE(simpad_flash_resources));
-
-	/*
-	 * Setup the PPC unit correctly.
-	 */
-	PPDR &= ~PPC_RXD4;
-	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
-	PSDR |= PPC_RXD4;
-	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
-
 	sa11x0_register_mcp(&simpad_mcp_data);
 
 	ret = platform_add_devices(devices, ARRAY_SIZE(devices));
diff --git a/arch/x86/.gitignore b/arch/x86/.gitignore
index 0280790..7cab8c0 100644
--- a/arch/x86/.gitignore
+++ b/arch/x86/.gitignore
@@ -1,3 +1,4 @@
 boot/compressed/vmlinux
 tools/test_get_len
+tools/insn_sanity
 
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6c14ecd..864cc6e 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -125,16 +125,6 @@
 config MMU
 	def_bool y
 
-config ZONE_DMA
-	bool "DMA memory allocation support" if EXPERT
-	default y
-	help
-	  DMA memory allocation support allows devices with less than 32-bit
-	  addressing to allocate within the first 16MB of address space.
-	  Disable if no such devices will be used.
-
-	  If unsure, say Y.
-
 config SBUS
 	bool
 
@@ -255,6 +245,16 @@
 
 menu "Processor type and features"
 
+config ZONE_DMA
+	bool "DMA memory allocation support" if EXPERT
+	default y
+	help
+	  DMA memory allocation support allows devices with less than 32-bit
+	  addressing to allocate within the first 16MB of address space.
+	  Disable if no such devices will be used.
+
+	  If unsure, say Y.
+
 source "kernel/time/Kconfig"
 
 config SMP
diff --git a/arch/x86/include/asm/unistd.h b/arch/x86/include/asm/unistd.h
index b4a3db7..21f77b8 100644
--- a/arch/x86/include/asm/unistd.h
+++ b/arch/x86/include/asm/unistd.h
@@ -7,6 +7,7 @@
 #  include <asm/unistd_32.h>
 #  define __ARCH_WANT_IPC_PARSE_VERSION
 #  define __ARCH_WANT_STAT64
+#  define __ARCH_WANT_SYS_IPC
 #  define __ARCH_WANT_SYS_OLD_MMAP
 #  define __ARCH_WANT_SYS_OLD_SELECT
 
diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h
index 8e862aa..becf47b 100644
--- a/arch/x86/include/asm/uv/uv_bau.h
+++ b/arch/x86/include/asm/uv/uv_bau.h
@@ -65,7 +65,7 @@
  * UV2: Bit 19 selects between
  *  (0): 10 microsecond timebase and
  *  (1): 80 microseconds
- *  we're using 655us, similar to UV1: 65 units of 10us
+ *  we're using 560us, similar to UV1: 65 units of 10us
  */
 #define UV1_INTD_SOFT_ACK_TIMEOUT_PERIOD (9UL)
 #define UV2_INTD_SOFT_ACK_TIMEOUT_PERIOD (15UL)
@@ -167,6 +167,7 @@
 #define FLUSH_RETRY_TIMEOUT		2
 #define FLUSH_GIVEUP			3
 #define FLUSH_COMPLETE			4
+#define FLUSH_RETRY_BUSYBUG		5
 
 /*
  * tuning the action when the numalink network is extremely delayed
@@ -235,10 +236,10 @@
 
 
 /*
- * Message header:  16 bytes (128 bits) (bytes 0x30-0x3f of descriptor)
+ * UV1 Message header:  16 bytes (128 bits) (bytes 0x30-0x3f of descriptor)
  * see table 4.2.3.0.1 in broacast_assist spec.
  */
-struct bau_msg_header {
+struct uv1_bau_msg_header {
 	unsigned int	dest_subnodeid:6;	/* must be 0x10, for the LB */
 	/* bits 5:0 */
 	unsigned int	base_dest_nasid:15;	/* nasid of the first bit */
@@ -318,19 +319,87 @@
 };
 
 /*
+ * UV2 Message header:  16 bytes (128 bits) (bytes 0x30-0x3f of descriptor)
+ * see figure 9-2 of harp_sys.pdf
+ */
+struct uv2_bau_msg_header {
+	unsigned int	base_dest_nasid:15;	/* nasid of the first bit */
+	/* bits 14:0 */				/* in uvhub map */
+	unsigned int	dest_subnodeid:5;	/* must be 0x10, for the LB */
+	/* bits 19:15 */
+	unsigned int	rsvd_1:1;		/* must be zero */
+	/* bit 20 */
+	/* Address bits 59:21 */
+	/* bits 25:2 of address (44:21) are payload */
+	/* these next 24 bits become bytes 12-14 of msg */
+	/* bits 28:21 land in byte 12 */
+	unsigned int	replied_to:1;		/* sent as 0 by the source to
+						   byte 12 */
+	/* bit 21 */
+	unsigned int	msg_type:3;		/* software type of the
+						   message */
+	/* bits 24:22 */
+	unsigned int	canceled:1;		/* message canceled, resource
+						   is to be freed*/
+	/* bit 25 */
+	unsigned int	payload_1:3;		/* not currently used */
+	/* bits 28:26 */
+
+	/* bits 36:29 land in byte 13 */
+	unsigned int	payload_2a:3;		/* not currently used */
+	unsigned int	payload_2b:5;		/* not currently used */
+	/* bits 36:29 */
+
+	/* bits 44:37 land in byte 14 */
+	unsigned int	payload_3:8;		/* not currently used */
+	/* bits 44:37 */
+
+	unsigned int	rsvd_2:7;		/* reserved */
+	/* bits 51:45 */
+	unsigned int	swack_flag:1;		/* software acknowledge flag */
+	/* bit 52 */
+	unsigned int	rsvd_3a:3;		/* must be zero */
+	unsigned int	rsvd_3b:8;		/* must be zero */
+	unsigned int	rsvd_3c:8;		/* must be zero */
+	unsigned int	rsvd_3d:3;		/* must be zero */
+	/* bits 74:53 */
+	unsigned int	fairness:3;		/* usually zero */
+	/* bits 77:75 */
+
+	unsigned int	sequence:16;		/* message sequence number */
+	/* bits 93:78  Suppl_A  */
+	unsigned int	chaining:1;		/* next descriptor is part of
+						   this activation*/
+	/* bit 94 */
+	unsigned int	multilevel:1;		/* multi-level multicast
+						   format */
+	/* bit 95 */
+	unsigned int	rsvd_4:24;		/* ordered / source node /
+						   source subnode / aging
+						   must be zero */
+	/* bits 119:96 */
+	unsigned int	command:8;		/* message type */
+	/* bits 127:120 */
+};
+
+/*
  * The activation descriptor:
  * The format of the message to send, plus all accompanying control
  * Should be 64 bytes
  */
 struct bau_desc {
-	struct pnmask			distribution;
+	struct pnmask				distribution;
 	/*
 	 * message template, consisting of header and payload:
 	 */
-	struct bau_msg_header		header;
-	struct bau_msg_payload		payload;
+	union bau_msg_header {
+		struct uv1_bau_msg_header	uv1_hdr;
+		struct uv2_bau_msg_header	uv2_hdr;
+	} header;
+
+	struct bau_msg_payload			payload;
 };
-/*
+/* UV1:
  *   -payload--    ---------header------
  *   bytes 0-11    bits 41-56  bits 58-81
  *       A           B  (2)      C (3)
@@ -340,6 +409,16 @@
  *   bytes 0-11  bytes 12-14  bytes 16-17  (byte 15 filled in by hw as vector)
  *   ------------payload queue-----------
  */
+/* UV2:
+ *   -payload--    ---------header------
+ *   bytes 0-11    bits 70-78  bits 21-44
+ *       A           B  (2)      C (3)
+ *
+ *            A/B/C are moved to:
+ *       A            C          B
+ *   bytes 0-11  bytes 12-14  bytes 16-17  (byte 15 filled in by hw as vector)
+ *   ------------payload queue-----------
+ */
 
 /*
  * The payload queue on the destination side is an array of these.
@@ -385,7 +464,6 @@
 struct msg_desc {
 	struct bau_pq_entry	*msg;
 	int			msg_slot;
-	int			swack_slot;
 	struct bau_pq_entry	*queue_first;
 	struct bau_pq_entry	*queue_last;
 };
@@ -405,6 +483,7 @@
 						   requests */
 	unsigned long	s_stimeout;		/* source side timeouts */
 	unsigned long	s_dtimeout;		/* destination side timeouts */
+	unsigned long	s_strongnacks;		/* number of strong nack's */
 	unsigned long	s_time;			/* time spent in sending side */
 	unsigned long	s_retriesok;		/* successful retries */
 	unsigned long	s_ntargcpu;		/* total number of cpu's
@@ -439,6 +518,9 @@
 	unsigned long	s_retry_messages;	/* retry broadcasts */
 	unsigned long	s_bau_reenabled;	/* for bau enable/disable */
 	unsigned long	s_bau_disabled;		/* for bau enable/disable */
+	unsigned long	s_uv2_wars;		/* uv2 workaround, perm. busy */
+	unsigned long	s_uv2_wars_hw;		/* uv2 workaround, hiwater */
+	unsigned long	s_uv2_war_waits;	/* uv2 workaround, long waits */
 	/* destination statistics */
 	unsigned long	d_alltlb;		/* times all tlb's on this
 						   cpu were flushed */
@@ -511,9 +593,12 @@
 	short			osnode;
 	short			uvhub_cpu;
 	short			uvhub;
+	short			uvhub_version;
 	short			cpus_in_socket;
 	short			cpus_in_uvhub;
 	short			partition_base_pnode;
+	short			using_desc; /* an index, like uvhub_cpu */
+	unsigned int		inuse_map;
 	unsigned short		message_number;
 	unsigned short		uvhub_quiesce;
 	short			socket_acknowledge_count[DEST_Q_SIZE];
@@ -531,6 +616,7 @@
 	int			cong_response_us;
 	int			cong_reps;
 	int			cong_period;
+	unsigned long		clocks_per_100_usec;
 	cycles_t		period_time;
 	long			period_requests;
 	struct hub_and_pnode	*thp;
@@ -591,6 +677,11 @@
 	uv_write_local_mmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, mr);
 }
 
+static inline void write_gmmr_sw_ack(int pnode, unsigned long mr)
+{
+	write_gmmr(pnode, UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, mr);
+}
+
 static inline unsigned long read_mmr_sw_ack(void)
 {
 	return read_lmmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE);
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index c0dd5b6..a62c201 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -290,14 +290,15 @@
 static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap)
 {
 	int count;
-	u64 tsc = 0;
+	u64 tsc = 0, prev_tsc = 0;
 
 	for (count = 0; count < 50000; count++) {
 		if (!pit_verify_msb(val))
 			break;
+		prev_tsc = tsc;
 		tsc = get_cycles();
 	}
-	*deltap = get_cycles() - tsc;
+	*deltap = get_cycles() - prev_tsc;
 	*tscp = tsc;
 
 	/*
@@ -311,9 +312,9 @@
  * How many MSB values do we want to see? We aim for
  * a maximum error rate of 500ppm (in practice the
  * real error is much smaller), but refuse to spend
- * more than 25ms on it.
+ * more than 50ms on it.
  */
-#define MAX_QUICK_PIT_MS 25
+#define MAX_QUICK_PIT_MS 50
 #define MAX_QUICK_PIT_ITERATIONS (MAX_QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256)
 
 static unsigned long quick_pit_calibrate(void)
@@ -383,15 +384,12 @@
 	 *
 	 * As a result, we can depend on there not being
 	 * any odd delays anywhere, and the TSC reads are
-	 * reliable (within the error). We also adjust the
-	 * delta to the middle of the error bars, just
-	 * because it looks nicer.
+	 * reliable (within the error).
 	 *
 	 * kHz = ticks / time-in-seconds / 1000;
 	 * kHz = (t2 - t1) / (I * 256 / PIT_TICK_RATE) / 1000
 	 * kHz = ((t2 - t1) * PIT_TICK_RATE) / (I * 256 * 1000)
 	 */
-	delta += (long)(d2 - d1)/2;
 	delta *= PIT_TICK_RATE;
 	do_div(delta, i*256*1000);
 	printk("Fast TSC calibration using PIT\n");
diff --git a/arch/x86/lib/x86-opcode-map.txt b/arch/x86/lib/x86-opcode-map.txt
index 5b83c51..8191379 100644
--- a/arch/x86/lib/x86-opcode-map.txt
+++ b/arch/x86/lib/x86-opcode-map.txt
@@ -219,7 +219,9 @@
 ac: LODS/B AL,Xb
 ad: LODS/W/D/Q rAX,Xv
 ae: SCAS/B AL,Yb
-af: SCAS/W/D/Q rAX,Xv
+# Note: The May 2011 Intel manual shows Xv for the second parameter of the
+# next instruction but Yv is correct
+af: SCAS/W/D/Q rAX,Yv
 # 0xb0 - 0xbf
 b0: MOV AL/R8L,Ib
 b1: MOV CL/R9L,Ib
@@ -729,8 +731,8 @@
 df: VAESDECLAST Vdq,Hdq,Wdq (66),(v1)
 f0: MOVBE Gy,My | MOVBE Gw,Mw (66) | CRC32 Gd,Eb (F2)
 f1: MOVBE My,Gy | MOVBE Mw,Gw (66) | CRC32 Gd,Ey (F2)
-f3: ANDN Gy,By,Ey (v)
-f4: Grp17 (1A)
+f2: ANDN Gy,By,Ey (v)
+f3: Grp17 (1A)
 f5: BZHI Gy,Ey,By (v) | PEXT Gy,By,Ey (F3),(v) | PDEP Gy,By,Ey (F2),(v)
 f6: MULX By,Gy,rDX,Ey (F2),(v)
 f7: BEXTR Gy,Ey,By (v) | SHLX Gy,Ey,By (66),(v) | SARX Gy,Ey,By (F3),(v) | SHRX Gy,Ey,By (F2),(v)
diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c
index 5b55219..9be4cff 100644
--- a/arch/x86/platform/uv/tlb_uv.c
+++ b/arch/x86/platform/uv/tlb_uv.c
@@ -157,13 +157,14 @@
  * clear of the Timeout bit (as well) will free the resource. No reply will
  * be sent (the hardware will only do one reply per message).
  */
-static void reply_to_message(struct msg_desc *mdp, struct bau_control *bcp)
+static void reply_to_message(struct msg_desc *mdp, struct bau_control *bcp,
+						int do_acknowledge)
 {
 	unsigned long dw;
 	struct bau_pq_entry *msg;
 
 	msg = mdp->msg;
-	if (!msg->canceled) {
+	if (!msg->canceled && do_acknowledge) {
 		dw = (msg->swack_vec << UV_SW_ACK_NPENDING) | msg->swack_vec;
 		write_mmr_sw_ack(dw);
 	}
@@ -212,8 +213,8 @@
 			if (mmr & (msg_res << UV_SW_ACK_NPENDING)) {
 				unsigned long mr;
 				/*
-				 * is the resource timed out?
-				 * make everyone ignore the cancelled message.
+				 * Is the resource timed out?
+				 * Make everyone ignore the cancelled message.
 				 */
 				msg2->canceled = 1;
 				stat->d_canceled++;
@@ -231,8 +232,8 @@
  * Do all the things a cpu should do for a TLB shootdown message.
  * Other cpu's may come here at the same time for this message.
  */
-static void bau_process_message(struct msg_desc *mdp,
-					struct bau_control *bcp)
+static void bau_process_message(struct msg_desc *mdp, struct bau_control *bcp,
+						int do_acknowledge)
 {
 	short socket_ack_count = 0;
 	short *sp;
@@ -284,8 +285,9 @@
 		if (msg_ack_count == bcp->cpus_in_uvhub) {
 			/*
 			 * All cpus in uvhub saw it; reply
+			 * (unless we are in the UV2 workaround)
 			 */
-			reply_to_message(mdp, bcp);
+			reply_to_message(mdp, bcp, do_acknowledge);
 		}
 	}
 
@@ -491,27 +493,138 @@
 /*
  * UV2 has an extra bit of status in the ACTIVATION_STATUS_2 register.
  */
-static unsigned long uv2_read_status(unsigned long offset, int rshft, int cpu)
+static unsigned long uv2_read_status(unsigned long offset, int rshft, int desc)
 {
 	unsigned long descriptor_status;
 	unsigned long descriptor_status2;
 
 	descriptor_status = ((read_lmmr(offset) >> rshft) & UV_ACT_STATUS_MASK);
-	descriptor_status2 = (read_mmr_uv2_status() >> cpu) & 0x1UL;
+	descriptor_status2 = (read_mmr_uv2_status() >> desc) & 0x1UL;
 	descriptor_status = (descriptor_status << 1) | descriptor_status2;
 	return descriptor_status;
 }
 
+/*
+ * Return whether the status of the descriptor that is normally used for this
+ * cpu (the one indexed by its hub-relative cpu number) is busy.
+ * The status of the original 32 descriptors is always reflected in the 64
+ * bits of UVH_LB_BAU_SB_ACTIVATION_STATUS_0.
+ * The bit provided by the activation_status_2 register is irrelevant to
+ * the status if it is only being tested for busy or not busy.
+ */
+int normal_busy(struct bau_control *bcp)
+{
+	int cpu = bcp->uvhub_cpu;
+	int mmr_offset;
+	int right_shift;
+
+	mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_0;
+	right_shift = cpu * UV_ACT_STATUS_SIZE;
+	return (((((read_lmmr(mmr_offset) >> right_shift) &
+				UV_ACT_STATUS_MASK)) << 1) == UV2H_DESC_BUSY);
+}
+
+/*
+ * Entered when a bau descriptor has gone into a permanent busy wait because
+ * of a hardware bug.
+ * Workaround the bug.
+ */
+int handle_uv2_busy(struct bau_control *bcp)
+{
+	int busy_one = bcp->using_desc;
+	int normal = bcp->uvhub_cpu;
+	int selected = -1;
+	int i;
+	unsigned long descriptor_status;
+	unsigned long status;
+	int mmr_offset;
+	struct bau_desc *bau_desc_old;
+	struct bau_desc *bau_desc_new;
+	struct bau_control *hmaster = bcp->uvhub_master;
+	struct ptc_stats *stat = bcp->statp;
+	cycles_t ttm;
+
+	stat->s_uv2_wars++;
+	spin_lock(&hmaster->uvhub_lock);
+	/* try for the original first */
+	if (busy_one != normal) {
+		if (!normal_busy(bcp))
+			selected = normal;
+	}
+	if (selected < 0) {
+		/* can't use the normal, select an alternate */
+		mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_1;
+		descriptor_status = read_lmmr(mmr_offset);
+
+		/* scan available descriptors 32-63 */
+		for (i = 0; i < UV_CPUS_PER_AS; i++) {
+			if ((hmaster->inuse_map & (1 << i)) == 0) {
+				status = ((descriptor_status >>
+						(i * UV_ACT_STATUS_SIZE)) &
+						UV_ACT_STATUS_MASK) << 1;
+				if (status != UV2H_DESC_BUSY) {
+					selected = i + UV_CPUS_PER_AS;
+					break;
+				}
+			}
+		}
+	}
+
+	if (busy_one != normal)
+		/* mark the busy alternate as not in-use */
+		hmaster->inuse_map &= ~(1 << (busy_one - UV_CPUS_PER_AS));
+
+	if (selected >= 0) {
+		/* switch to the selected descriptor */
+		if (selected != normal) {
+			/* set the selected alternate as in-use */
+			hmaster->inuse_map |=
+					(1 << (selected - UV_CPUS_PER_AS));
+			if (selected > stat->s_uv2_wars_hw)
+				stat->s_uv2_wars_hw = selected;
+		}
+		bau_desc_old = bcp->descriptor_base;
+		bau_desc_old += (ITEMS_PER_DESC * busy_one);
+		bcp->using_desc = selected;
+		bau_desc_new = bcp->descriptor_base;
+		bau_desc_new += (ITEMS_PER_DESC * selected);
+		*bau_desc_new = *bau_desc_old;
+	} else {
+		/*
+		 * All are busy. Wait for the normal one for this cpu to
+		 * free up.
+		 */
+		stat->s_uv2_war_waits++;
+		spin_unlock(&hmaster->uvhub_lock);
+		ttm = get_cycles();
+		do {
+			cpu_relax();
+		} while (normal_busy(bcp));
+		spin_lock(&hmaster->uvhub_lock);
+		/* switch to the original descriptor */
+		bcp->using_desc = normal;
+		bau_desc_old = bcp->descriptor_base;
+		bau_desc_old += (ITEMS_PER_DESC * bcp->using_desc);
+		bcp->using_desc = (ITEMS_PER_DESC * normal);
+		bau_desc_new = bcp->descriptor_base;
+		bau_desc_new += (ITEMS_PER_DESC * normal);
+		*bau_desc_new = *bau_desc_old; /* copy the entire descriptor */
+	}
+	spin_unlock(&hmaster->uvhub_lock);
+	return FLUSH_RETRY_BUSYBUG;
+}
+
 static int uv2_wait_completion(struct bau_desc *bau_desc,
 				unsigned long mmr_offset, int right_shift,
 				struct bau_control *bcp, long try)
 {
 	unsigned long descriptor_stat;
 	cycles_t ttm;
-	int cpu = bcp->uvhub_cpu;
+	int desc = bcp->using_desc;
+	long busy_reps = 0;
 	struct ptc_stats *stat = bcp->statp;
 
-	descriptor_stat = uv2_read_status(mmr_offset, right_shift, cpu);
+	descriptor_stat = uv2_read_status(mmr_offset, right_shift, desc);
 
 	/* spin on the status MMR, waiting for it to go idle */
 	while (descriptor_stat != UV2H_DESC_IDLE) {
@@ -522,32 +635,35 @@
 		 * our message and its state will stay IDLE.
 		 */
 		if ((descriptor_stat == UV2H_DESC_SOURCE_TIMEOUT) ||
-		    (descriptor_stat == UV2H_DESC_DEST_STRONG_NACK) ||
 		    (descriptor_stat == UV2H_DESC_DEST_PUT_ERR)) {
 			stat->s_stimeout++;
 			return FLUSH_GIVEUP;
+		} else if (descriptor_stat == UV2H_DESC_DEST_STRONG_NACK) {
+			stat->s_strongnacks++;
+			bcp->conseccompletes = 0;
+			return FLUSH_GIVEUP;
 		} else if (descriptor_stat == UV2H_DESC_DEST_TIMEOUT) {
 			stat->s_dtimeout++;
-			ttm = get_cycles();
-			/*
-			 * Our retries may be blocked by all destination
-			 * swack resources being consumed, and a timeout
-			 * pending.  In that case hardware returns the
-			 * ERROR that looks like a destination timeout.
-			 */
-			if (cycles_2_us(ttm - bcp->send_message) < timeout_us) {
-				bcp->conseccompletes = 0;
-				return FLUSH_RETRY_PLUGGED;
-			}
 			bcp->conseccompletes = 0;
 			return FLUSH_RETRY_TIMEOUT;
 		} else {
+			busy_reps++;
+			if (busy_reps > 1000000) {
+				/* not to hammer on the clock */
+				busy_reps = 0;
+				ttm = get_cycles();
+				if ((ttm - bcp->send_message) >
+					(bcp->clocks_per_100_usec)) {
+					return handle_uv2_busy(bcp);
+				}
+			}
 			/*
 			 * descriptor_stat is still BUSY
 			 */
 			cpu_relax();
 		}
-		descriptor_stat = uv2_read_status(mmr_offset, right_shift, cpu);
+		descriptor_stat = uv2_read_status(mmr_offset, right_shift,
+									desc);
 	}
 	bcp->conseccompletes++;
 	return FLUSH_COMPLETE;
@@ -563,17 +679,17 @@
 {
 	int right_shift;
 	unsigned long mmr_offset;
-	int cpu = bcp->uvhub_cpu;
+	int desc = bcp->using_desc;
 
-	if (cpu < UV_CPUS_PER_AS) {
+	if (desc < UV_CPUS_PER_AS) {
 		mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_0;
-		right_shift = cpu * UV_ACT_STATUS_SIZE;
+		right_shift = desc * UV_ACT_STATUS_SIZE;
 	} else {
 		mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_1;
-		right_shift = ((cpu - UV_CPUS_PER_AS) * UV_ACT_STATUS_SIZE);
+		right_shift = ((desc - UV_CPUS_PER_AS) * UV_ACT_STATUS_SIZE);
 	}
 
-	if (is_uv1_hub())
+	if (bcp->uvhub_version == 1)
 		return uv1_wait_completion(bau_desc, mmr_offset, right_shift,
 								bcp, try);
 	else
@@ -752,19 +868,22 @@
  * Returns 1 if it gives up entirely and the original cpu mask is to be
  * returned to the kernel.
  */
-int uv_flush_send_and_wait(struct bau_desc *bau_desc,
-			struct cpumask *flush_mask, struct bau_control *bcp)
+int uv_flush_send_and_wait(struct cpumask *flush_mask, struct bau_control *bcp)
 {
 	int seq_number = 0;
 	int completion_stat = 0;
+	int uv1 = 0;
 	long try = 0;
 	unsigned long index;
 	cycles_t time1;
 	cycles_t time2;
 	struct ptc_stats *stat = bcp->statp;
 	struct bau_control *hmaster = bcp->uvhub_master;
+	struct uv1_bau_msg_header *uv1_hdr = NULL;
+	struct uv2_bau_msg_header *uv2_hdr = NULL;
+	struct bau_desc *bau_desc;
 
-	if (is_uv1_hub())
+	if (bcp->uvhub_version == 1)
 		uv1_throttle(hmaster, stat);
 
 	while (hmaster->uvhub_quiesce)
@@ -772,22 +891,39 @@
 
 	time1 = get_cycles();
 	do {
-		if (try == 0) {
-			bau_desc->header.msg_type = MSG_REGULAR;
+		bau_desc = bcp->descriptor_base;
+		bau_desc += (ITEMS_PER_DESC * bcp->using_desc);
+		if (bcp->uvhub_version == 1) {
+			uv1 = 1;
+			uv1_hdr = &bau_desc->header.uv1_hdr;
+		} else
+			uv2_hdr = &bau_desc->header.uv2_hdr;
+		if ((try == 0) || (completion_stat == FLUSH_RETRY_BUSYBUG)) {
+			if (uv1)
+				uv1_hdr->msg_type = MSG_REGULAR;
+			else
+				uv2_hdr->msg_type = MSG_REGULAR;
 			seq_number = bcp->message_number++;
 		} else {
-			bau_desc->header.msg_type = MSG_RETRY;
+			if (uv1)
+				uv1_hdr->msg_type = MSG_RETRY;
+			else
+				uv2_hdr->msg_type = MSG_RETRY;
 			stat->s_retry_messages++;
 		}
 
-		bau_desc->header.sequence = seq_number;
-		index = (1UL << AS_PUSH_SHIFT) | bcp->uvhub_cpu;
+		if (uv1)
+			uv1_hdr->sequence = seq_number;
+		else
+			uv2_hdr->sequence = seq_number;
+		index = (1UL << AS_PUSH_SHIFT) | bcp->using_desc;
 		bcp->send_message = get_cycles();
 
 		write_mmr_activation(index);
 
 		try++;
 		completion_stat = wait_completion(bau_desc, bcp, try);
+		/* UV2: wait_completion() may change the bcp->using_desc */
 
 		handle_cmplt(completion_stat, bau_desc, bcp, hmaster, stat);
 
@@ -798,6 +934,7 @@
 		}
 		cpu_relax();
 	} while ((completion_stat == FLUSH_RETRY_PLUGGED) ||
+		 (completion_stat == FLUSH_RETRY_BUSYBUG) ||
 		 (completion_stat == FLUSH_RETRY_TIMEOUT));
 
 	time2 = get_cycles();
@@ -812,6 +949,7 @@
 	record_send_stats(time1, time2, bcp, stat, completion_stat, try);
 
 	if (completion_stat == FLUSH_GIVEUP)
+		/* FLUSH_GIVEUP will fall back to using IPI's for tlb flush */
 		return 1;
 	return 0;
 }
@@ -967,7 +1105,7 @@
 		stat->s_ntargself++;
 
 	bau_desc = bcp->descriptor_base;
-	bau_desc += ITEMS_PER_DESC * bcp->uvhub_cpu;
+	bau_desc += (ITEMS_PER_DESC * bcp->using_desc);
 	bau_uvhubs_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE);
 	if (set_distrib_bits(flush_mask, bcp, bau_desc, &locals, &remotes))
 		return NULL;
@@ -980,13 +1118,86 @@
 	 * uv_flush_send_and_wait returns 0 if all cpu's were messaged,
 	 * or 1 if it gave up and the original cpumask should be returned.
 	 */
-	if (!uv_flush_send_and_wait(bau_desc, flush_mask, bcp))
+	if (!uv_flush_send_and_wait(flush_mask, bcp))
 		return NULL;
 	else
 		return cpumask;
 }
 
 /*
+ * Search the message queue for any 'other' message with the same software
+ * acknowledge resource bit vector.
+ */
+struct bau_pq_entry *find_another_by_swack(struct bau_pq_entry *msg,
+			struct bau_control *bcp, unsigned char swack_vec)
+{
+	struct bau_pq_entry *msg_next = msg + 1;
+
+	if (msg_next > bcp->queue_last)
+		msg_next = bcp->queue_first;
+	while ((msg_next->swack_vec != 0) && (msg_next != msg)) {
+		if (msg_next->swack_vec == swack_vec)
+			return msg_next;
+		msg_next++;
+		if (msg_next > bcp->queue_last)
+			msg_next = bcp->queue_first;
+	}
+	return NULL;
+}
+
+/*
+ * UV2 needs to work around a bug in which an arriving message has not
+ * set a bit in the UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE register.
+ * Such a message must be ignored.
+ */
+void process_uv2_message(struct msg_desc *mdp, struct bau_control *bcp)
+{
+	unsigned long mmr_image;
+	unsigned char swack_vec;
+	struct bau_pq_entry *msg = mdp->msg;
+	struct bau_pq_entry *other_msg;
+
+	mmr_image = read_mmr_sw_ack();
+	swack_vec = msg->swack_vec;
+
+	if ((swack_vec & mmr_image) == 0) {
+		/*
+		 * This message was assigned a swack resource, but no
+		 * reserved acknowlegment is pending.
+		 * The bug has prevented this message from setting the MMR.
+		 * And no other message has used the same sw_ack resource.
+		 * Do the requested shootdown but do not reply to the msg.
+		 * (the 0 means make no acknowledge)
+		 */
+		bau_process_message(mdp, bcp, 0);
+		return;
+	}
+
+	/*
+	 * Some message has set the MMR 'pending' bit; it might have been
+	 * another message.  Look for that message.
+	 */
+	other_msg = find_another_by_swack(msg, bcp, msg->swack_vec);
+	if (other_msg) {
+		/* There is another.  Do not ack the current one. */
+		bau_process_message(mdp, bcp, 0);
+		/*
+		 * Let the natural processing of that message acknowledge
+		 * it. Don't get the processing of sw_ack's out of order.
+		 */
+		return;
+	}
+
+	/*
+	 * There is no other message using this sw_ack, so it is safe to
+	 * acknowledge it.
+	 */
+	bau_process_message(mdp, bcp, 1);
+
+	return;
+}
+
+/*
  * The BAU message interrupt comes here. (registered by set_intr_gate)
  * See entry_64.S
  *
@@ -1009,6 +1220,7 @@
 	struct ptc_stats *stat;
 	struct msg_desc msgdesc;
 
+	ack_APIC_irq();
 	time_start = get_cycles();
 
 	bcp = &per_cpu(bau_control, smp_processor_id());
@@ -1022,9 +1234,11 @@
 		count++;
 
 		msgdesc.msg_slot = msg - msgdesc.queue_first;
-		msgdesc.swack_slot = ffs(msg->swack_vec) - 1;
 		msgdesc.msg = msg;
-		bau_process_message(&msgdesc, bcp);
+		if (bcp->uvhub_version == 2)
+			process_uv2_message(&msgdesc, bcp);
+		else
+			bau_process_message(&msgdesc, bcp, 1);
 
 		msg++;
 		if (msg > msgdesc.queue_last)
@@ -1036,8 +1250,6 @@
 		stat->d_nomsg++;
 	else if (count > 1)
 		stat->d_multmsg++;
-
-	ack_APIC_irq();
 }
 
 /*
@@ -1083,7 +1295,7 @@
 		 */
 		mmr_image |= (1L << SOFTACK_MSHIFT);
 		if (is_uv2_hub()) {
-			mmr_image |= (1L << UV2_LEG_SHFT);
+			mmr_image &= ~(1L << UV2_LEG_SHFT);
 			mmr_image |= (1L << UV2_EXT_SHFT);
 		}
 		write_mmr_misc_control(pnode, mmr_image);
@@ -1136,13 +1348,13 @@
 		seq_printf(file,
 			"remotehub numuvhubs numuvhubs16 numuvhubs8 ");
 		seq_printf(file,
-			"numuvhubs4 numuvhubs2 numuvhubs1 dto retries rok ");
+		    "numuvhubs4 numuvhubs2 numuvhubs1 dto snacks retries rok ");
 		seq_printf(file,
 			"resetp resett giveup sto bz throt swack recv rtime ");
 		seq_printf(file,
 			"all one mult none retry canc nocan reset rcan ");
 		seq_printf(file,
-			"disable enable\n");
+			"disable enable wars warshw warwaits\n");
 	}
 	if (cpu < num_possible_cpus() && cpu_online(cpu)) {
 		stat = &per_cpu(ptcstats, cpu);
@@ -1154,10 +1366,10 @@
 			   stat->s_ntargremotes, stat->s_ntargcpu,
 			   stat->s_ntarglocaluvhub, stat->s_ntargremoteuvhub,
 			   stat->s_ntarguvhub, stat->s_ntarguvhub16);
-		seq_printf(file, "%ld %ld %ld %ld %ld ",
+		seq_printf(file, "%ld %ld %ld %ld %ld %ld ",
 			   stat->s_ntarguvhub8, stat->s_ntarguvhub4,
 			   stat->s_ntarguvhub2, stat->s_ntarguvhub1,
-			   stat->s_dtimeout);
+			   stat->s_dtimeout, stat->s_strongnacks);
 		seq_printf(file, "%ld %ld %ld %ld %ld %ld %ld %ld ",
 			   stat->s_retry_messages, stat->s_retriesok,
 			   stat->s_resets_plug, stat->s_resets_timeout,
@@ -1173,8 +1385,10 @@
 			   stat->d_nomsg, stat->d_retries, stat->d_canceled,
 			   stat->d_nocanceled, stat->d_resets,
 			   stat->d_rcanceled);
-		seq_printf(file, "%ld %ld\n",
-			stat->s_bau_disabled, stat->s_bau_reenabled);
+		seq_printf(file, "%ld %ld %ld %ld %ld\n",
+			stat->s_bau_disabled, stat->s_bau_reenabled,
+			stat->s_uv2_wars, stat->s_uv2_wars_hw,
+			stat->s_uv2_war_waits);
 	}
 	return 0;
 }
@@ -1432,12 +1646,15 @@
 {
 	int i;
 	int cpu;
+	int uv1 = 0;
 	unsigned long gpa;
 	unsigned long m;
 	unsigned long n;
 	size_t dsize;
 	struct bau_desc *bau_desc;
 	struct bau_desc *bd2;
+	struct uv1_bau_msg_header *uv1_hdr;
+	struct uv2_bau_msg_header *uv2_hdr;
 	struct bau_control *bcp;
 
 	/*
@@ -1451,6 +1668,8 @@
 	gpa = uv_gpa(bau_desc);
 	n = uv_gpa_to_gnode(gpa);
 	m = uv_gpa_to_offset(gpa);
+	if (is_uv1_hub())
+		uv1 = 1;
 
 	/* the 14-bit pnode */
 	write_mmr_descriptor_base(pnode, (n << UV_DESC_PSHIFT | m));
@@ -1461,21 +1680,33 @@
 	 */
 	for (i = 0, bd2 = bau_desc; i < (ADP_SZ * ITEMS_PER_DESC); i++, bd2++) {
 		memset(bd2, 0, sizeof(struct bau_desc));
-		bd2->header.swack_flag =	1;
-		/*
-		 * The base_dest_nasid set in the message header is the nasid
-		 * of the first uvhub in the partition. The bit map will
-		 * indicate destination pnode numbers relative to that base.
-		 * They may not be consecutive if nasid striding is being used.
-		 */
-		bd2->header.base_dest_nasid =	UV_PNODE_TO_NASID(base_pnode);
-		bd2->header.dest_subnodeid =	UV_LB_SUBNODEID;
-		bd2->header.command =		UV_NET_ENDPOINT_INTD;
-		bd2->header.int_both =		1;
-		/*
-		 * all others need to be set to zero:
-		 *   fairness chaining multilevel count replied_to
-		 */
+		if (uv1) {
+			uv1_hdr = &bd2->header.uv1_hdr;
+			uv1_hdr->swack_flag =	1;
+			/*
+			 * The base_dest_nasid set in the message header
+			 * is the nasid of the first uvhub in the partition.
+			 * The bit map will indicate destination pnode numbers
+			 * relative to that base. They may not be consecutive
+			 * if nasid striding is being used.
+			 */
+			uv1_hdr->base_dest_nasid =
+						UV_PNODE_TO_NASID(base_pnode);
+			uv1_hdr->dest_subnodeid =	UV_LB_SUBNODEID;
+			uv1_hdr->command =		UV_NET_ENDPOINT_INTD;
+			uv1_hdr->int_both =		1;
+			/*
+			 * all others need to be set to zero:
+			 *   fairness chaining multilevel count replied_to
+			 */
+		} else {
+			uv2_hdr = &bd2->header.uv2_hdr;
+			uv2_hdr->swack_flag =	1;
+			uv2_hdr->base_dest_nasid =
+						UV_PNODE_TO_NASID(base_pnode);
+			uv2_hdr->dest_subnodeid =	UV_LB_SUBNODEID;
+			uv2_hdr->command =		UV_NET_ENDPOINT_INTD;
+		}
 	}
 	for_each_present_cpu(cpu) {
 		if (pnode != uv_blade_to_pnode(uv_cpu_to_blade_id(cpu)))
@@ -1531,6 +1762,7 @@
 	write_mmr_payload_first(pnode, pn_first);
 	write_mmr_payload_tail(pnode, first);
 	write_mmr_payload_last(pnode, last);
+	write_gmmr_sw_ack(pnode, 0xffffUL);
 
 	/* in effect, all msg_type's are set to MSG_NOOP */
 	memset(pqp, 0, sizeof(struct bau_pq_entry) * DEST_Q_SIZE);
@@ -1584,14 +1816,14 @@
 		ts_ns = base * mult1 * mult2;
 		ret = ts_ns / 1000;
 	} else {
-		/* 4 bits  0/1 for 10/80us, 3 bits of multiplier */
-		mmr_image = uv_read_local_mmr(UVH_AGING_PRESCALE_SEL);
+		/* 4 bits  0/1 for 10/80us base, 3 bits of multiplier */
+		mmr_image = uv_read_local_mmr(UVH_LB_BAU_MISC_CONTROL);
 		mmr_image = (mmr_image & UV_SA_MASK) >> UV_SA_SHFT;
 		if (mmr_image & (1L << UV2_ACK_UNITS_SHFT))
-			mult1 = 80;
+			base = 80;
 		else
-			mult1 = 10;
-		base = mmr_image & UV2_ACK_MASK;
+			base = 10;
+		mult1 = mmr_image & UV2_ACK_MASK;
 		ret = mult1 * base;
 	}
 	return ret;
@@ -1618,6 +1850,7 @@
 		bcp->cong_response_us		= congested_respns_us;
 		bcp->cong_reps			= congested_reps;
 		bcp->cong_period		= congested_period;
+		bcp->clocks_per_100_usec =	usec_2_cycles(100);
 	}
 }
 
@@ -1728,8 +1961,17 @@
 		bcp->cpus_in_socket = sdp->num_cpus;
 		bcp->socket_master = *smasterp;
 		bcp->uvhub = bdp->uvhub;
+		if (is_uv1_hub())
+			bcp->uvhub_version = 1;
+		else if (is_uv2_hub())
+			bcp->uvhub_version = 2;
+		else {
+			printk(KERN_EMERG "uvhub version not 1 or 2\n");
+			return 1;
+		}
 		bcp->uvhub_master = *hmasterp;
 		bcp->uvhub_cpu = uv_cpu_hub_info(cpu)->blade_processor_id;
+		bcp->using_desc = bcp->uvhub_cpu;
 		if (bcp->uvhub_cpu >= MAX_CPUS_PER_UVHUB) {
 			printk(KERN_EMERG "%d cpus per uvhub invalid\n",
 				bcp->uvhub_cpu);
@@ -1845,6 +2087,8 @@
 			uv_base_pnode = uv_blade_to_pnode(uvhub);
 	}
 
+	enable_timeouts();
+
 	if (init_per_cpu(nuvhubs, uv_base_pnode)) {
 		nobau = 1;
 		return 0;
@@ -1855,7 +2099,6 @@
 		if (uv_blade_nr_possible_cpus(uvhub))
 			init_uvhub(uvhub, vector, uv_base_pnode);
 
-	enable_timeouts();
 	alloc_intr_gate(vector, uv_bau_message_intr1);
 
 	for_each_possible_blade(uvhub) {
@@ -1867,7 +2110,8 @@
 			val = 1L << 63;
 			write_gmmr_activation(pnode, val);
 			mmr = 1; /* should be 1 to broadcast to both sockets */
-			write_mmr_data_broadcast(pnode, mmr);
+			if (!is_uv1_hub())
+				write_mmr_data_broadcast(pnode, mmr);
 		}
 	}
 
diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 37c4bd1..d0c4118 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -87,6 +87,7 @@
 
 config GPIO_IT8761E
 	tristate "IT8761E GPIO support"
+	depends on X86  # unconditional access to IO space.
 	help
 	  Say yes here to support GPIO functionality of IT8761E super I/O chip.
 
diff --git a/drivers/gpio/gpio-ml-ioh.c b/drivers/gpio/gpio-ml-ioh.c
index 461958f..03d6dd5 100644
--- a/drivers/gpio/gpio-ml-ioh.c
+++ b/drivers/gpio/gpio-ml-ioh.c
@@ -248,7 +248,7 @@
 static int ioh_irq_type(struct irq_data *d, unsigned int type)
 {
 	u32 im;
-	u32 *im_reg;
+	void __iomem *im_reg;
 	u32 ien;
 	u32 im_pos;
 	int ch;
@@ -412,7 +412,7 @@
 	int i, j;
 	struct ioh_gpio *chip;
 	void __iomem *base;
-	void __iomem *chip_save;
+	void *chip_save;
 	int irq_base;
 
 	ret = pci_enable_device(pdev);
@@ -428,7 +428,7 @@
 	}
 
 	base = pci_iomap(pdev, 1, 0);
-	if (base == 0) {
+	if (!base) {
 		dev_err(&pdev->dev, "%s : pci_iomap failed", __func__);
 		ret = -ENOMEM;
 		goto err_iomap;
@@ -521,7 +521,7 @@
 	int err;
 	int i;
 	struct ioh_gpio *chip = pci_get_drvdata(pdev);
-	void __iomem *chip_save;
+	void *chip_save;
 
 	chip_save = chip;
 
diff --git a/drivers/gpio/gpio-pch.c b/drivers/gpio/gpio-pch.c
index f060329..68fa55e 100644
--- a/drivers/gpio/gpio-pch.c
+++ b/drivers/gpio/gpio-pch.c
@@ -231,7 +231,7 @@
 static int pch_irq_type(struct irq_data *d, unsigned int type)
 {
 	u32 im;
-	u32 *im_reg;
+	u32 __iomem *im_reg;
 	u32 ien;
 	u32 im_pos;
 	int ch;
@@ -376,7 +376,7 @@
 	}
 
 	chip->base = pci_iomap(pdev, 1, 0);
-	if (chip->base == 0) {
+	if (!chip->base) {
 		dev_err(&pdev->dev, "%s : pci_iomap FAILED", __func__);
 		ret = -ENOMEM;
 		goto err_iomap;
diff --git a/drivers/gpio/gpio-tps65910.c b/drivers/gpio/gpio-tps65910.c
index b9c1c29..91f45b9 100644
--- a/drivers/gpio/gpio-tps65910.c
+++ b/drivers/gpio/gpio-tps65910.c
@@ -52,7 +52,7 @@
 	struct tps65910 *tps65910 = container_of(gc, struct tps65910, gpio);
 
 	/* Set the initial value */
-	tps65910_gpio_set(gc, 0, value);
+	tps65910_gpio_set(gc, offset, value);
 
 	return tps65910_set_bits(tps65910, TPS65910_GPIO0 + offset,
 						GPIO_CFG_MASK);
diff --git a/drivers/mfd/mcp-core.c b/drivers/mfd/mcp-core.c
index 63be60b..86cc3f7 100644
--- a/drivers/mfd/mcp-core.c
+++ b/drivers/mfd/mcp-core.c
@@ -26,35 +26,9 @@
 #define to_mcp(d)		container_of(d, struct mcp, attached_device)
 #define to_mcp_driver(d)	container_of(d, struct mcp_driver, drv)
 
-static const struct mcp_device_id *mcp_match_id(const struct mcp_device_id *id,
-						const char *codec)
-{
-	while (id->name[0]) {
-		if (strcmp(codec, id->name) == 0)
-			return id;
-		id++;
-	}
-	return NULL;
-}
-
-const struct mcp_device_id *mcp_get_device_id(const struct mcp *mcp)
-{
-	const struct mcp_driver *driver =
-		to_mcp_driver(mcp->attached_device.driver);
-
-	return mcp_match_id(driver->id_table, mcp->codec);
-}
-EXPORT_SYMBOL(mcp_get_device_id);
-
 static int mcp_bus_match(struct device *dev, struct device_driver *drv)
 {
-	const struct mcp *mcp = to_mcp(dev);
-	const struct mcp_driver *driver = to_mcp_driver(drv);
-
-	if (driver->id_table)
-		return !!mcp_match_id(driver->id_table, mcp->codec);
-
-	return 0;
+	return 1;
 }
 
 static int mcp_bus_probe(struct device *dev)
@@ -100,18 +74,9 @@
 	return ret;
 }
 
-static int mcp_bus_uevent(struct device *dev, struct kobj_uevent_env *env)
-{
-	struct mcp *mcp = to_mcp(dev);
-
-	add_uevent_var(env, "MODALIAS=%s%s", MCP_MODULE_PREFIX, mcp->codec);
-	return 0;
-}
-
 static struct bus_type mcp_bus_type = {
 	.name		= "mcp",
 	.match		= mcp_bus_match,
-	.uevent		= mcp_bus_uevent,
 	.probe		= mcp_bus_probe,
 	.remove		= mcp_bus_remove,
 	.suspend	= mcp_bus_suspend,
@@ -128,9 +93,11 @@
  */
 void mcp_set_telecom_divisor(struct mcp *mcp, unsigned int div)
 {
-	spin_lock_irq(&mcp->lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&mcp->lock, flags);
 	mcp->ops->set_telecom_divisor(mcp, div);
-	spin_unlock_irq(&mcp->lock);
+	spin_unlock_irqrestore(&mcp->lock, flags);
 }
 EXPORT_SYMBOL(mcp_set_telecom_divisor);
 
@@ -143,9 +110,11 @@
  */
 void mcp_set_audio_divisor(struct mcp *mcp, unsigned int div)
 {
-	spin_lock_irq(&mcp->lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&mcp->lock, flags);
 	mcp->ops->set_audio_divisor(mcp, div);
-	spin_unlock_irq(&mcp->lock);
+	spin_unlock_irqrestore(&mcp->lock, flags);
 }
 EXPORT_SYMBOL(mcp_set_audio_divisor);
 
@@ -198,10 +167,11 @@
  */
 void mcp_enable(struct mcp *mcp)
 {
-	spin_lock_irq(&mcp->lock);
+	unsigned long flags;
+	spin_lock_irqsave(&mcp->lock, flags);
 	if (mcp->use_count++ == 0)
 		mcp->ops->enable(mcp);
-	spin_unlock_irq(&mcp->lock);
+	spin_unlock_irqrestore(&mcp->lock, flags);
 }
 EXPORT_SYMBOL(mcp_enable);
 
@@ -247,14 +217,9 @@
 }
 EXPORT_SYMBOL(mcp_host_alloc);
 
-int mcp_host_register(struct mcp *mcp, void *pdata)
+int mcp_host_register(struct mcp *mcp)
 {
-	if (!mcp->codec)
-		return -EINVAL;
-
-	mcp->attached_device.platform_data = pdata;
 	dev_set_name(&mcp->attached_device, "mcp0");
-	request_module("%s%s", MCP_MODULE_PREFIX, mcp->codec);
 	return device_register(&mcp->attached_device);
 }
 EXPORT_SYMBOL(mcp_host_register);
diff --git a/drivers/mfd/mcp-sa11x0.c b/drivers/mfd/mcp-sa11x0.c
index 9adc2eb..02c53a0 100644
--- a/drivers/mfd/mcp-sa11x0.c
+++ b/drivers/mfd/mcp-sa11x0.c
@@ -19,7 +19,6 @@
 #include <linux/spinlock.h>
 #include <linux/platform_device.h>
 #include <linux/mfd/mcp.h>
-#include <linux/io.h>
 
 #include <mach/dma.h>
 #include <mach/hardware.h>
@@ -27,19 +26,12 @@
 #include <asm/system.h>
 #include <mach/mcp.h>
 
-/* Register offsets */
-#define MCCR0	0x00
-#define MCDR0	0x08
-#define MCDR1	0x0C
-#define MCDR2	0x10
-#define MCSR	0x18
-#define MCCR1	0x00
+#include <mach/assabet.h>
+
 
 struct mcp_sa11x0 {
-	u32		mccr0;
-	u32		mccr1;
-	unsigned char	*mccr0_base;
-	unsigned char	*mccr1_base;
+	u32	mccr0;
+	u32	mccr1;
 };
 
 #define priv(mcp)	((struct mcp_sa11x0 *)mcp_priv(mcp))
@@ -47,25 +39,25 @@
 static void
 mcp_sa11x0_set_telecom_divisor(struct mcp *mcp, unsigned int divisor)
 {
-	struct mcp_sa11x0 *priv = priv(mcp);
+	unsigned int mccr0;
 
 	divisor /= 32;
 
-	priv->mccr0 &= ~0x00007f00;
-	priv->mccr0 |= divisor << 8;
-	__raw_writel(priv->mccr0, priv->mccr0_base + MCCR0);
+	mccr0 = Ser4MCCR0 & ~0x00007f00;
+	mccr0 |= divisor << 8;
+	Ser4MCCR0 = mccr0;
 }
 
 static void
 mcp_sa11x0_set_audio_divisor(struct mcp *mcp, unsigned int divisor)
 {
-	struct mcp_sa11x0 *priv = priv(mcp);
+	unsigned int mccr0;
 
 	divisor /= 32;
 
-	priv->mccr0 &= ~0x0000007f;
-	priv->mccr0 |= divisor;
-	__raw_writel(priv->mccr0, priv->mccr0_base + MCCR0);
+	mccr0 = Ser4MCCR0 & ~0x0000007f;
+	mccr0 |= divisor;
+	Ser4MCCR0 = mccr0;
 }
 
 /*
@@ -79,16 +71,12 @@
 {
 	int ret = -ETIME;
 	int i;
-	u32 mcpreg;
-	struct mcp_sa11x0 *priv = priv(mcp);
 
-	mcpreg = reg << 17 | MCDR2_Wr | (val & 0xffff);
-	__raw_writel(mcpreg, priv->mccr0_base + MCDR2);
+	Ser4MCDR2 = reg << 17 | MCDR2_Wr | (val & 0xffff);
 
 	for (i = 0; i < 2; i++) {
 		udelay(mcp->rw_timeout);
-		mcpreg = __raw_readl(priv->mccr0_base + MCSR);
-		if (mcpreg & MCSR_CWC) {
+		if (Ser4MCSR & MCSR_CWC) {
 			ret = 0;
 			break;
 		}
@@ -109,18 +97,13 @@
 {
 	int ret = -ETIME;
 	int i;
-	u32 mcpreg;
-	struct mcp_sa11x0 *priv = priv(mcp);
 
-	mcpreg = reg << 17 | MCDR2_Rd;
-	__raw_writel(mcpreg, priv->mccr0_base + MCDR2);
+	Ser4MCDR2 = reg << 17 | MCDR2_Rd;
 
 	for (i = 0; i < 2; i++) {
 		udelay(mcp->rw_timeout);
-		mcpreg = __raw_readl(priv->mccr0_base + MCSR);
-		if (mcpreg & MCSR_CRC) {
-			ret = __raw_readl(priv->mccr0_base + MCDR2)
-				& 0xffff;
+		if (Ser4MCSR & MCSR_CRC) {
+			ret = Ser4MCDR2 & 0xffff;
 			break;
 		}
 	}
@@ -133,19 +116,13 @@
 
 static void mcp_sa11x0_enable(struct mcp *mcp)
 {
-	struct mcp_sa11x0 *priv = priv(mcp);
-
-	__raw_writel(-1, priv->mccr0_base + MCSR);
-	priv->mccr0 |= MCCR0_MCE;
-	__raw_writel(priv->mccr0, priv->mccr0_base + MCCR0);
+	Ser4MCSR = -1;
+	Ser4MCCR0 |= MCCR0_MCE;
 }
 
 static void mcp_sa11x0_disable(struct mcp *mcp)
 {
-	struct mcp_sa11x0 *priv = priv(mcp);
-
-	priv->mccr0 &= ~MCCR0_MCE;
-	__raw_writel(priv->mccr0, priv->mccr0_base + MCCR0);
+	Ser4MCCR0 &= ~MCCR0_MCE;
 }
 
 /*
@@ -165,69 +142,50 @@
 	struct mcp_plat_data *data = pdev->dev.platform_data;
 	struct mcp *mcp;
 	int ret;
-	struct mcp_sa11x0 *priv;
-	struct resource *res_mem0, *res_mem1;
-	u32 size0, size1;
 
 	if (!data)
 		return -ENODEV;
 
-	if (!data->codec)
-		return -ENODEV;
-
-	res_mem0 = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res_mem0)
-		return -ENODEV;
-	size0 = res_mem0->end - res_mem0->start + 1;
-
-	res_mem1 = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	if (!res_mem1)
-		return -ENODEV;
-	size1 = res_mem1->end - res_mem1->start + 1;
-
-	if (!request_mem_region(res_mem0->start, size0, "sa11x0-mcp"))
+	if (!request_mem_region(0x80060000, 0x60, "sa11x0-mcp"))
 		return -EBUSY;
 
-	if (!request_mem_region(res_mem1->start, size1, "sa11x0-mcp")) {
-		ret = -EBUSY;
-		goto release;
-	}
-
 	mcp = mcp_host_alloc(&pdev->dev, sizeof(struct mcp_sa11x0));
 	if (!mcp) {
 		ret = -ENOMEM;
-		goto release2;
+		goto release;
 	}
 
-	priv = priv(mcp);
-
 	mcp->owner		= THIS_MODULE;
 	mcp->ops		= &mcp_sa11x0;
 	mcp->sclk_rate		= data->sclk_rate;
-	mcp->dma_audio_rd	= DDAR_DevAdd(res_mem0->start + MCDR0)
-				+ DDAR_DevRd + DDAR_Brst4 + DDAR_8BitDev;
-	mcp->dma_audio_wr	= DDAR_DevAdd(res_mem0->start + MCDR0)
-				+ DDAR_DevWr + DDAR_Brst4 + DDAR_8BitDev;
-	mcp->dma_telco_rd	= DDAR_DevAdd(res_mem0->start + MCDR1)
-				+ DDAR_DevRd + DDAR_Brst4 + DDAR_8BitDev;
-	mcp->dma_telco_wr	= DDAR_DevAdd(res_mem0->start + MCDR1)
-				+ DDAR_DevWr + DDAR_Brst4 + DDAR_8BitDev;
-	mcp->codec		= data->codec;
+	mcp->dma_audio_rd	= DMA_Ser4MCP0Rd;
+	mcp->dma_audio_wr	= DMA_Ser4MCP0Wr;
+	mcp->dma_telco_rd	= DMA_Ser4MCP1Rd;
+	mcp->dma_telco_wr	= DMA_Ser4MCP1Wr;
+	mcp->gpio_base		= data->gpio_base;
 
 	platform_set_drvdata(pdev, mcp);
 
+	if (machine_is_assabet()) {
+		ASSABET_BCR_set(ASSABET_BCR_CODEC_RST);
+	}
+
+	/*
+	 * Setup the PPC unit correctly.
+	 */
+	PPDR &= ~PPC_RXD4;
+	PPDR |= PPC_TXD4 | PPC_SCLK | PPC_SFRM;
+	PSDR |= PPC_RXD4;
+	PSDR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
+	PPSR &= ~(PPC_TXD4 | PPC_SCLK | PPC_SFRM);
+
 	/*
 	 * Initialise device.  Note that we initially
 	 * set the sampling rate to minimum.
 	 */
-	priv->mccr0_base = ioremap(res_mem0->start, size0);
-	priv->mccr1_base = ioremap(res_mem1->start, size1);
-
-	__raw_writel(-1, priv->mccr0_base + MCSR);
-	priv->mccr1 = data->mccr1;
-	priv->mccr0 = data->mccr0 | 0x7f7f;
-	__raw_writel(priv->mccr0, priv->mccr0_base + MCCR0);
-	__raw_writel(priv->mccr1, priv->mccr1_base + MCCR1);
+	Ser4MCSR = -1;
+	Ser4MCCR1 = data->mccr1;
+	Ser4MCCR0 = data->mccr0 | 0x7f7f;
 
 	/*
 	 * Calculate the read/write timeout (us) from the bit clock
@@ -237,53 +195,36 @@
 	mcp->rw_timeout = (64 * 3 * 1000000 + mcp->sclk_rate - 1) /
 			  mcp->sclk_rate;
 
-	ret = mcp_host_register(mcp, data->codec_pdata);
+	ret = mcp_host_register(mcp);
 	if (ret == 0)
 		goto out;
 
- release2:
-	release_mem_region(res_mem1->start, size1);
  release:
-	release_mem_region(res_mem0->start, size0);
+	release_mem_region(0x80060000, 0x60);
 	platform_set_drvdata(pdev, NULL);
 
  out:
 	return ret;
 }
 
-static int mcp_sa11x0_remove(struct platform_device *pdev)
+static int mcp_sa11x0_remove(struct platform_device *dev)
 {
-	struct mcp *mcp = platform_get_drvdata(pdev);
-	struct mcp_sa11x0 *priv = priv(mcp);
-	struct resource *res_mem;
-	u32 size;
+	struct mcp *mcp = platform_get_drvdata(dev);
 
-	platform_set_drvdata(pdev, NULL);
+	platform_set_drvdata(dev, NULL);
 	mcp_host_unregister(mcp);
+	release_mem_region(0x80060000, 0x60);
 
-	res_mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (res_mem) {
-		size = res_mem->end - res_mem->start + 1;
-		release_mem_region(res_mem->start, size);
-	}
-	res_mem = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	if (res_mem) {
-		size = res_mem->end - res_mem->start + 1;
-		release_mem_region(res_mem->start, size);
-	}
-	iounmap(priv->mccr0_base);
-	iounmap(priv->mccr1_base);
 	return 0;
 }
 
 static int mcp_sa11x0_suspend(struct platform_device *dev, pm_message_t state)
 {
 	struct mcp *mcp = platform_get_drvdata(dev);
-	struct mcp_sa11x0 *priv = priv(mcp);
-	u32 mccr0;
 
-	mccr0 = priv->mccr0 & ~MCCR0_MCE;
-	__raw_writel(mccr0, priv->mccr0_base + MCCR0);
+	priv(mcp)->mccr0 = Ser4MCCR0;
+	priv(mcp)->mccr1 = Ser4MCCR1;
+	Ser4MCCR0 &= ~MCCR0_MCE;
 
 	return 0;
 }
@@ -291,10 +232,9 @@
 static int mcp_sa11x0_resume(struct platform_device *dev)
 {
 	struct mcp *mcp = platform_get_drvdata(dev);
-	struct mcp_sa11x0 *priv = priv(mcp);
 
-	__raw_writel(priv->mccr0, priv->mccr0_base + MCCR0);
-	__raw_writel(priv->mccr1, priv->mccr1_base + MCCR1);
+	Ser4MCCR1 = priv(mcp)->mccr1;
+	Ser4MCCR0 = priv(mcp)->mccr0;
 
 	return 0;
 }
@@ -311,7 +251,6 @@
 	.resume		= mcp_sa11x0_resume,
 	.driver		= {
 		.name	= "sa11x0-mcp",
-		.owner  = THIS_MODULE,
 	},
 };
 
diff --git a/drivers/mfd/ucb1x00-core.c b/drivers/mfd/ucb1x00-core.c
index 91c4f25..febc90c 100644
--- a/drivers/mfd/ucb1x00-core.c
+++ b/drivers/mfd/ucb1x00-core.c
@@ -36,15 +36,6 @@
 static LIST_HEAD(ucb1x00_drivers);
 static LIST_HEAD(ucb1x00_devices);
 
-static struct mcp_device_id ucb1x00_id[] = {
-	{ "ucb1x00", 0 },  /* auto-detection */
-	{ "ucb1200", UCB_ID_1200 },
-	{ "ucb1300", UCB_ID_1300 },
-	{ "tc35143", UCB_ID_TC35143 },
-	{ }
-};
-MODULE_DEVICE_TABLE(mcp, ucb1x00_id);
-
 /**
  *	ucb1x00_io_set_dir - set IO direction
  *	@ucb: UCB1x00 structure describing chip
@@ -157,16 +148,22 @@
 {
 	struct ucb1x00 *ucb = container_of(chip, struct ucb1x00, gpio);
 	unsigned long flags;
+	unsigned old, mask = 1 << offset;
 
 	spin_lock_irqsave(&ucb->io_lock, flags);
-	ucb->io_dir |= (1 << offset);
-	ucb1x00_reg_write(ucb, UCB_IO_DIR, ucb->io_dir);
-
+	old = ucb->io_out;
 	if (value)
-		ucb->io_out |= 1 << offset;
+		ucb->io_out |= mask;
 	else
-		ucb->io_out &= ~(1 << offset);
-	ucb1x00_reg_write(ucb, UCB_IO_DATA, ucb->io_out);
+		ucb->io_out &= ~mask;
+
+	if (old != ucb->io_out)
+		ucb1x00_reg_write(ucb, UCB_IO_DATA, ucb->io_out);
+
+	if (!(ucb->io_dir & mask)) {
+		ucb->io_dir |= mask;
+		ucb1x00_reg_write(ucb, UCB_IO_DIR, ucb->io_dir);
+	}
 	spin_unlock_irqrestore(&ucb->io_lock, flags);
 
 	return 0;
@@ -536,33 +533,17 @@
 
 static int ucb1x00_probe(struct mcp *mcp)
 {
-	const struct mcp_device_id *mid;
 	struct ucb1x00 *ucb;
 	struct ucb1x00_driver *drv;
-	struct ucb1x00_plat_data *pdata;
 	unsigned int id;
 	int ret = -ENODEV;
 	int temp;
 
 	mcp_enable(mcp);
 	id = mcp_reg_read(mcp, UCB_ID);
-	mid = mcp_get_device_id(mcp);
 
-	if (mid && mid->driver_data) {
-		if (id != mid->driver_data) {
-			printk(KERN_WARNING "%s wrong ID %04x found: %04x\n",
-				mid->name, (unsigned int) mid->driver_data, id);
-			goto err_disable;
-		}
-	} else {
-		mid = &ucb1x00_id[1];
-		while (mid->driver_data) {
-			if (id == mid->driver_data)
-				break;
-			mid++;
-		}
-		printk(KERN_WARNING "%s ID not found: %04x\n",
-			ucb1x00_id[0].name, id);
+	if (id != UCB_ID_1200 && id != UCB_ID_1300 && id != UCB_ID_TC35143) {
+		printk(KERN_WARNING "UCB1x00 ID not found: %04x\n", id);
 		goto err_disable;
 	}
 
@@ -571,28 +552,28 @@
 	if (!ucb)
 		goto err_disable;
 
-	pdata = mcp->attached_device.platform_data;
+
 	ucb->dev.class = &ucb1x00_class;
 	ucb->dev.parent = &mcp->attached_device;
-	dev_set_name(&ucb->dev, mid->name);
+	dev_set_name(&ucb->dev, "ucb1x00");
 
 	spin_lock_init(&ucb->lock);
 	spin_lock_init(&ucb->io_lock);
 	sema_init(&ucb->adc_sem, 1);
 
-	ucb->id  = mid;
+	ucb->id  = id;
 	ucb->mcp = mcp;
 	ucb->irq = ucb1x00_detect_irq(ucb);
 	if (ucb->irq == NO_IRQ) {
-		printk(KERN_ERR "%s: IRQ probe failed\n", mid->name);
+		printk(KERN_ERR "UCB1x00: IRQ probe failed\n");
 		ret = -ENODEV;
 		goto err_free;
 	}
 
 	ucb->gpio.base = -1;
-	if (pdata && (pdata->gpio_base >= 0)) {
+	if (mcp->gpio_base != 0) {
 		ucb->gpio.label = dev_name(&ucb->dev);
-		ucb->gpio.base = pdata->gpio_base;
+		ucb->gpio.base = mcp->gpio_base;
 		ucb->gpio.ngpio = 10;
 		ucb->gpio.set = ucb1x00_gpio_set;
 		ucb->gpio.get = ucb1x00_gpio_get;
@@ -605,10 +586,10 @@
 		dev_info(&ucb->dev, "gpio_base not set so no gpiolib support");
 
 	ret = request_irq(ucb->irq, ucb1x00_irq, IRQF_TRIGGER_RISING,
-			  mid->name, ucb);
+			  "UCB1x00", ucb);
 	if (ret) {
-		printk(KERN_ERR "%s: unable to grab irq%d: %d\n",
-			mid->name, ucb->irq, ret);
+		printk(KERN_ERR "ucb1x00: unable to grab irq%d: %d\n",
+			ucb->irq, ret);
 		goto err_gpio;
 	}
 
@@ -712,6 +693,7 @@
 	struct ucb1x00 *ucb = mcp_get_drvdata(mcp);
 	struct ucb1x00_dev *dev;
 
+	ucb1x00_reg_write(ucb, UCB_IO_DATA, ucb->io_out);
 	ucb1x00_reg_write(ucb, UCB_IO_DIR, ucb->io_dir);
 	mutex_lock(&ucb1x00_mutex);
 	list_for_each_entry(dev, &ucb->devs, dev_node) {
@@ -730,7 +712,6 @@
 	.remove		= ucb1x00_remove,
 	.suspend	= ucb1x00_suspend,
 	.resume		= ucb1x00_resume,
-	.id_table	= ucb1x00_id,
 };
 
 static int __init ucb1x00_init(void)
diff --git a/drivers/mfd/ucb1x00-ts.c b/drivers/mfd/ucb1x00-ts.c
index 40ec3c1..63a3cbd 100644
--- a/drivers/mfd/ucb1x00-ts.c
+++ b/drivers/mfd/ucb1x00-ts.c
@@ -47,7 +47,6 @@
 	u16			x_res;
 	u16			y_res;
 
-	unsigned int		restart:1;
 	unsigned int		adcsync:1;
 };
 
@@ -207,15 +206,17 @@
 {
 	struct ucb1x00_ts *ts = _ts;
 	DECLARE_WAITQUEUE(wait, current);
+	bool frozen, ignore = false;
 	int valid = 0;
 
 	set_freezable();
 	add_wait_queue(&ts->irq_wait, &wait);
-	while (!kthread_should_stop()) {
+	while (!kthread_freezable_should_stop(&frozen)) {
 		unsigned int x, y, p;
 		signed long timeout;
 
-		ts->restart = 0;
+		if (frozen)
+			ignore = true;
 
 		ucb1x00_adc_enable(ts->ucb);
 
@@ -258,7 +259,7 @@
 			 * space.  We therefore leave it to user space
 			 * to do any filtering they please.
 			 */
-			if (!ts->restart) {
+			if (!ignore) {
 				ucb1x00_ts_evt_add(ts, p, x, y);
 				valid = 1;
 			}
@@ -267,8 +268,6 @@
 			timeout = HZ / 100;
 		}
 
-		try_to_freeze();
-
 		schedule_timeout(timeout);
 	}
 
@@ -340,26 +339,6 @@
 	ucb1x00_disable(ts->ucb);
 }
 
-#ifdef CONFIG_PM
-static int ucb1x00_ts_resume(struct ucb1x00_dev *dev)
-{
-	struct ucb1x00_ts *ts = dev->priv;
-
-	if (ts->rtask != NULL) {
-		/*
-		 * Restart the TS thread to ensure the
-		 * TS interrupt mode is set up again
-		 * after sleep.
-		 */
-		ts->restart = 1;
-		wake_up(&ts->irq_wait);
-	}
-	return 0;
-}
-#else
-#define ucb1x00_ts_resume NULL
-#endif
-
 
 /*
  * Initialisation.
@@ -382,7 +361,7 @@
 	ts->adcsync = adcsync ? UCB_SYNC : UCB_NOSYNC;
 
 	idev->name       = "Touchscreen panel";
-	idev->id.product = ts->ucb->id->driver_data;
+	idev->id.product = ts->ucb->id;
 	idev->open       = ucb1x00_ts_open;
 	idev->close      = ucb1x00_ts_close;
 
@@ -425,7 +404,6 @@
 static struct ucb1x00_driver ucb1x00_ts_driver = {
 	.add		= ucb1x00_ts_add,
 	.remove		= ucb1x00_ts_remove,
-	.resume		= ucb1x00_ts_resume,
 };
 
 static int __init ucb1x00_ts_init(void)
diff --git a/fs/inode.c b/fs/inode.c
index 4fa4f09..fb10d86 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -322,9 +322,6 @@
 void set_nlink(struct inode *inode, unsigned int nlink)
 {
 	if (!nlink) {
-		printk_ratelimited(KERN_INFO
-			"set_nlink() clearing i_nlink on %s inode %li\n",
-			inode->i_sb->s_type->name, inode->i_ino);
 		clear_nlink(inode);
 	} else {
 		/* Yes, some filesystems do change nlink from zero to one */
diff --git a/fs/proc/stat.c b/fs/proc/stat.c
index d76ca6a..121f77c 100644
--- a/fs/proc/stat.c
+++ b/fs/proc/stat.c
@@ -77,6 +77,8 @@
 		steal += kcpustat_cpu(i).cpustat[CPUTIME_STEAL];
 		guest += kcpustat_cpu(i).cpustat[CPUTIME_GUEST];
 		guest_nice += kcpustat_cpu(i).cpustat[CPUTIME_GUEST_NICE];
+		sum += kstat_cpu_irqs_sum(i);
+		sum += arch_irq_stat_cpu(i);
 
 		for (j = 0; j < NR_SOFTIRQS; j++) {
 			unsigned int softirq_stat = kstat_softirqs_cpu(j, i);
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 2bfd987..6b00954 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -179,47 +179,33 @@
 	struct qnx4_inode_entry *rootdir;
 	int rd, rl;
 	int i, j;
-	int found = 0;
 
-	if (*(qnx4_sb(sb)->sb->RootDir.di_fname) != '/') {
+	if (*(qnx4_sb(sb)->sb->RootDir.di_fname) != '/')
 		return "no qnx4 filesystem (no root dir).";
-	} else {
-		QNX4DEBUG((KERN_NOTICE "QNX4 filesystem found on dev %s.\n", sb->s_id));
-		rd = le32_to_cpu(qnx4_sb(sb)->sb->RootDir.di_first_xtnt.xtnt_blk) - 1;
-		rl = le32_to_cpu(qnx4_sb(sb)->sb->RootDir.di_first_xtnt.xtnt_size);
-		for (j = 0; j < rl; j++) {
-			bh = sb_bread(sb, rd + j);	/* root dir, first block */
-			if (bh == NULL) {
-				return "unable to read root entry.";
-			}
-			for (i = 0; i < QNX4_INODES_PER_BLOCK; i++) {
-				rootdir = (struct qnx4_inode_entry *) (bh->b_data + i * QNX4_DIR_ENTRY_SIZE);
-				if (rootdir->di_fname != NULL) {
-					QNX4DEBUG((KERN_INFO "rootdir entry found : [%s]\n", rootdir->di_fname));
-					if (!strcmp(rootdir->di_fname,
-						    QNX4_BMNAME)) {
-						found = 1;
-						qnx4_sb(sb)->BitMap = kmemdup(rootdir,
-									      sizeof(struct qnx4_inode_entry),
-									      GFP_KERNEL);
-						if (!qnx4_sb(sb)->BitMap) {
-							brelse (bh);
-							return "not enough memory for bitmap inode";
-						}/* keep bitmap inode known */
-						break;
-					}
-				}
-			}
+	QNX4DEBUG((KERN_NOTICE "QNX4 filesystem found on dev %s.\n", sb->s_id));
+	rd = le32_to_cpu(qnx4_sb(sb)->sb->RootDir.di_first_xtnt.xtnt_blk) - 1;
+	rl = le32_to_cpu(qnx4_sb(sb)->sb->RootDir.di_first_xtnt.xtnt_size);
+	for (j = 0; j < rl; j++) {
+		bh = sb_bread(sb, rd + j);	/* root dir, first block */
+		if (bh == NULL)
+			return "unable to read root entry.";
+		rootdir = (struct qnx4_inode_entry *) bh->b_data;
+		for (i = 0; i < QNX4_INODES_PER_BLOCK; i++, rootdir++) {
+			QNX4DEBUG((KERN_INFO "rootdir entry found : [%s]\n", rootdir->di_fname));
+			if (strcmp(rootdir->di_fname, QNX4_BMNAME) != 0)
+				continue;
+			qnx4_sb(sb)->BitMap = kmemdup(rootdir,
+						      sizeof(struct qnx4_inode_entry),
+						      GFP_KERNEL);
 			brelse(bh);
-			if (found != 0) {
-				break;
-			}
+			if (!qnx4_sb(sb)->BitMap)
+				return "not enough memory for bitmap inode";
+			/* keep bitmap inode known */
+			return NULL;
 		}
-		if (found == 0) {
-			return "bitmap file not found.";
-		}
+		brelse(bh);
 	}
-	return NULL;
+	return "bitmap file not found.";
 }
 
 static int qnx4_fill_super(struct super_block *s, void *data, int silent)
@@ -270,7 +256,7 @@
 	if (IS_ERR(root)) {
 		printk(KERN_ERR "qnx4: get inode failed\n");
 		ret = PTR_ERR(root);
- 		goto out;
+ 		goto outb;
  	}
 
 	ret = -ENOMEM;
@@ -283,6 +269,8 @@
 
       outi:
 	iput(root);
+      outb:
+	kfree(qs->BitMap);
       out:
 	brelse(bh);
       outnobh:
diff --git a/fs/super.c b/fs/super.c
index de41e1e..6015c02 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -1186,6 +1186,8 @@
 			printk(KERN_ERR
 				"VFS:Filesystem freeze failed\n");
 			sb->s_frozen = SB_UNFROZEN;
+			smp_wmb();
+			wake_up(&sb->s_wait_unfrozen);
 			deactivate_locked_super(sb);
 			return ret;
 		}
diff --git a/include/linux/key.h b/include/linux/key.h
index bfc014c..5253471 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -271,7 +271,7 @@
 
 extern struct key *key_lookup(key_serial_t id);
 
-static inline key_serial_t key_serial(struct key *key)
+static inline key_serial_t key_serial(const struct key *key)
 {
 	return key ? key->serial : 0;
 }
diff --git a/include/linux/mfd/mcp.h b/include/linux/mfd/mcp.h
index 1515e64..f88c1cc 100644
--- a/include/linux/mfd/mcp.h
+++ b/include/linux/mfd/mcp.h
@@ -10,7 +10,6 @@
 #ifndef MCP_H
 #define MCP_H
 
-#include <linux/mod_devicetable.h>
 #include <mach/dma.h>
 
 struct mcp_ops;
@@ -27,7 +26,7 @@
 	dma_device_t	dma_telco_rd;
 	dma_device_t	dma_telco_wr;
 	struct device	attached_device;
-	const char	*codec;
+	int		gpio_base;
 };
 
 struct mcp_ops {
@@ -45,11 +44,10 @@
 unsigned int mcp_reg_read(struct mcp *, unsigned int);
 void mcp_enable(struct mcp *);
 void mcp_disable(struct mcp *);
-const struct mcp_device_id *mcp_get_device_id(const struct mcp *mcp);
 #define mcp_get_sclk_rate(mcp)	((mcp)->sclk_rate)
 
 struct mcp *mcp_host_alloc(struct device *, size_t);
-int mcp_host_register(struct mcp *, void *);
+int mcp_host_register(struct mcp *);
 void mcp_host_unregister(struct mcp *);
 
 struct mcp_driver {
@@ -58,7 +56,6 @@
 	void (*remove)(struct mcp *);
 	int (*suspend)(struct mcp *, pm_message_t);
 	int (*resume)(struct mcp *);
-	const struct mcp_device_id *id_table;
 };
 
 int mcp_driver_register(struct mcp_driver *);
@@ -67,6 +64,9 @@
 #define mcp_get_drvdata(mcp)	dev_get_drvdata(&(mcp)->attached_device)
 #define mcp_set_drvdata(mcp,d)	dev_set_drvdata(&(mcp)->attached_device, d)
 
-#define mcp_priv(mcp)		((void *)((mcp)+1))
+static inline void *mcp_priv(struct mcp *mcp)
+{
+	return mcp + 1;
+}
 
 #endif
diff --git a/include/linux/mfd/ucb1x00.h b/include/linux/mfd/ucb1x00.h
index bc19e5f..4321f04 100644
--- a/include/linux/mfd/ucb1x00.h
+++ b/include/linux/mfd/ucb1x00.h
@@ -104,9 +104,6 @@
 #define UCB_MODE_DYN_VFLAG_ENA	(1 << 12)
 #define UCB_MODE_AUD_OFF_CAN	(1 << 13)
 
-struct ucb1x00_plat_data {
-	int		gpio_base;
-};
 
 struct ucb1x00_irq {
 	void *devid;
@@ -119,7 +116,7 @@
 	unsigned int		irq;
 	struct semaphore	adc_sem;
 	spinlock_t		io_lock;
-	const struct mcp_device_id *id;
+	u16			id;
 	u16			io_dir;
 	u16			io_out;
 	u16			adc_cr;
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index b29e7f6..83ac071 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -436,17 +436,6 @@
 			__attribute__((aligned(sizeof(kernel_ulong_t))));
 };
 
-/* mcp */
-
-#define MCP_NAME_SIZE	20
-#define MCP_MODULE_PREFIX "mcp:"
-
-struct mcp_device_id {
-	char name[MCP_NAME_SIZE];
-	kernel_ulong_t driver_data	/* Data private to the driver */
-			__attribute__((aligned(sizeof(kernel_ulong_t))));
-};
-
 /* dmi */
 enum dmi_field {
 	DMI_NONE,
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index db110b8..f1539de 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -634,10 +634,11 @@
 	int ret = 0;
 
 	/*
-	 * We skip modules that tain the kernel, especially those with different
-	 * module header (for forced load), to make sure we don't cause a crash.
+	 * We skip modules that taint the kernel, especially those with different
+	 * module headers (for forced load), to make sure we don't cause a crash.
+	 * Staging and out-of-tree GPL modules are fine.
 	 */
-	if (mod->taints)
+	if (mod->taints & ~((1 << TAINT_OOT_MODULE) | (1 << TAINT_CRAP)))
 		return 0;
 	mutex_lock(&tracepoints_mutex);
 	tp_mod = kmalloc(sizeof(struct tp_module), GFP_KERNEL);
diff --git a/lib/mpi/mpicoder.c b/lib/mpi/mpicoder.c
index fe84bb9..716802b 100644
--- a/lib/mpi/mpicoder.c
+++ b/lib/mpi/mpicoder.c
@@ -255,6 +255,8 @@
 	if (!n)
 		n++;		/* avoid zero length allocation */
 	p = buffer = kmalloc(n, GFP_KERNEL);
+	if (!p)
+		return NULL;
 
 	for (i = a->nlimbs - 1; i >= 0; i--) {
 		alimb = a->d[i];
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index c0e14b3..e8c9695 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -823,16 +823,6 @@
 }
 ADD_TO_DEVTABLE("spi", struct spi_device_id, do_spi_entry);
 
-/* Looks like: mcp:S */
-static int do_mcp_entry(const char *filename, struct mcp_device_id *id,
-			char *alias)
-{
-	sprintf(alias, MCP_MODULE_PREFIX "%s", id->name);
-
-	return 1;
-}
-ADD_TO_DEVTABLE("mcp", struct mcp_device_id, do_mcp_entry); 
-
 static const struct dmifield {
 	const char *prefix;
 	int field;
diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c
index d661afb..d45061d 100644
--- a/security/integrity/ima/ima_policy.c
+++ b/security/integrity/ima/ima_policy.c
@@ -99,6 +99,7 @@
 			    struct inode *inode, enum ima_hooks func, int mask)
 {
 	struct task_struct *tsk = current;
+	const struct cred *cred = current_cred();
 	int i;
 
 	if ((rule->flags & IMA_FUNC) && rule->func != func)
@@ -108,7 +109,7 @@
 	if ((rule->flags & IMA_FSMAGIC)
 	    && rule->fsmagic != inode->i_sb->s_magic)
 		return false;
-	if ((rule->flags & IMA_UID) && rule->uid != tsk->cred->uid)
+	if ((rule->flags & IMA_UID) && rule->uid != cred->uid)
 		return false;
 	for (i = 0; i < MAX_LSM_RULES; i++) {
 		int rc = 0;
diff --git a/security/keys/user_defined.c b/security/keys/user_defined.c
index 69ff52c..2aee3c5 100644
--- a/security/keys/user_defined.c
+++ b/security/keys/user_defined.c
@@ -59,7 +59,7 @@
 	/* attach the data */
 	upayload->datalen = datalen;
 	memcpy(upayload->data, data, datalen);
-	rcu_assign_pointer(key->payload.data, upayload);
+	rcu_assign_keypointer(key, upayload);
 	ret = 0;
 
 error:
@@ -98,7 +98,7 @@
 	if (ret == 0) {
 		/* attach the new data, displacing the old */
 		zap = key->payload.data;
-		rcu_assign_pointer(key->payload.data, upayload);
+		rcu_assign_keypointer(key, upayload);
 		key->expiry = 0;
 	}
 
@@ -133,7 +133,7 @@
 	key_payload_reserve(key, 0);
 
 	if (upayload) {
-		rcu_assign_pointer(key->payload.data, NULL);
+		rcu_assign_keypointer(key, NULL);
 		kfree_rcu(upayload, rcu);
 	}
 }