235 files changed, 6104 insertions, 2355 deletions

diff --git a/Documentation/scsi/cxgb3i.txt b/Documentation/scsi/cxgb3i.txt
index 8141fa01978e..7ac8032ee9b2 100644
--- a/Documentation/scsi/cxgb3i.txt
+++ b/Documentation/scsi/cxgb3i.txt
@@ -4,7 +4,7 @@ Introduction
 ============
 
 The Chelsio T3 ASIC based Adapters (S310, S320, S302, S304, Mezz cards, etc.
-series of products) supports iSCSI acceleration and iSCSI Direct Data Placement
+series of products) support iSCSI acceleration and iSCSI Direct Data Placement
 (DDP) where the hardware handles the expensive byte touching operations, such
 as CRC computation and verification, and direct DMA to the final host memory
 destination:
@@ -31,9 +31,9 @@ destination:
           the TCP segments onto the wire. It handles TCP retransmission if
           needed.
 
-          On receving, S3 h/w recovers the iSCSI PDU by reassembling TCP
+          On receiving, S3 h/w recovers the iSCSI PDU by reassembling TCP
           segments, separating the header and data, calculating and verifying
-          the digests, then forwards the header to the host. The payload data,
+          the digests, then forwarding the header to the host. The payload data,
           if possible, will be directly placed into the pre-posted host DDP
           buffer. Otherwise, the payload data will be sent to the host too.
 
@@ -68,9 +68,8 @@ The following steps need to be taken to accelerates the open-iscsi initiator:
         sure the ip address is unique in the network.
 
 3. edit /etc/iscsi/iscsid.conf
-   The default setting for MaxRecvDataSegmentLength (131072) is too big,
-   replace "node.conn[0].iscsi.MaxRecvDataSegmentLength" to be a value no
-   bigger than 15360 (for example 8192):
+   The default setting for MaxRecvDataSegmentLength (131072) is too big;
+   replace with a value no bigger than 15360 (for example 8192):
 
         node.conn[0].iscsi.MaxRecvDataSegmentLength = 8192
 
diff --git a/Documentation/x86/boot.txt b/Documentation/x86/boot.txt
index 12299697b7cd..e0203662f9e9 100644
--- a/Documentation/x86/boot.txt
+++ b/Documentation/x86/boot.txt
@@ -543,7 +543,10 @@ Protocol:	2.08+
 
   The payload may be compressed. The format of both the compressed and
   uncompressed data should be determined using the standard magic
-  numbers. Currently only gzip compressed ELF is used.
+  numbers.  The currently supported compression formats are gzip
+  (magic numbers 1F 8B or 1F 9E), bzip2 (magic number 42 5A) and LZMA
+  (magic number 5D 00).  The uncompressed payload is currently always ELF
+  (magic number 7F 45 4C 46).
   
 Field name:     payload_length
 Type:           read
diff --git a/MAINTAINERS b/MAINTAINERS
index 59fd2d1d94a7..1c2ca1dc66f2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2464,7 +2464,7 @@ S:	Maintained
 
 ISDN SUBSYSTEM
 P:      Karsten Keil
-M:      kkeil@suse.de
+M:      isdn@linux-pingi.de
 L:      isdn4linux@listserv.isdn4linux.de (subscribers-only)
 W:      http://www.isdn4linux.de
 T:      git kernel.org:/pub/scm/linux/kernel/kkeil/isdn-2.6.git
diff --git a/Makefile b/Makefile
index 27fb890a2bff..c40d83aedebe 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 29
-EXTRAVERSION = -rc6
+EXTRAVERSION = -rc7
 NAME = Erotic Pickled Herring
 
 # *DOCUMENTATION*
diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c
index 5d7a16eab312..af71d38c8e41 100644
--- a/arch/alpha/mm/init.c
+++ b/arch/alpha/mm/init.c
@@ -189,9 +189,21 @@ callback_init(void * kernel_end)
 
         if (alpha_using_srm) {
                 static struct vm_struct console_remap_vm;
-                unsigned long vaddr = VMALLOC_START;
+                unsigned long nr_pages = 0;
+                unsigned long vaddr;
                 unsigned long i, j;
 
+                /* calculate needed size */
+                for (i = 0; i < crb->map_entries; ++i)
+                        nr_pages += crb->map[i].count;
+
+                /* register the vm area */
+                console_remap_vm.flags = VM_ALLOC;
+                console_remap_vm.size = nr_pages << PAGE_SHIFT;
+                vm_area_register_early(&console_remap_vm, PAGE_SIZE);
+
+                vaddr = (unsigned long)console_remap_vm.addr;
+
                 /* Set up the third level PTEs and update the virtual
                    addresses of the CRB entries.  */
                 for (i = 0; i < crb->map_entries; ++i) {
@@ -213,12 +225,6 @@ callback_init(void * kernel_end)
                                 vaddr += PAGE_SIZE;
                         }
                 }
-
-                /* Let vmalloc know that we've allocated some space.  */
-                console_remap_vm.flags = VM_ALLOC;
-                console_remap_vm.addr = (void *) VMALLOC_START;
-                console_remap_vm.size = vaddr - VMALLOC_START;
-                vmlist = &console_remap_vm;
         }
 
         callback_init_done = 1;
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 7049815d66d5..68d6494c0389 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -233,12 +233,13 @@ static void __init cacheid_init(void)
         unsigned int cachetype = read_cpuid_cachetype();
         unsigned int arch = cpu_architecture();
 
-        if (arch >= CPU_ARCH_ARMv7) {
-                cacheid = CACHEID_VIPT_NONALIASING;
-                if ((cachetype & (3 << 14)) == 1 << 14)
-                        cacheid |= CACHEID_ASID_TAGGED;
-        } else if (arch >= CPU_ARCH_ARMv6) {
-                if (cachetype & (1 << 23))
+        if (arch >= CPU_ARCH_ARMv6) {
+                if ((cachetype & (7 << 29)) == 4 << 29) {
+                        /* ARMv7 register format */
+                        cacheid = CACHEID_VIPT_NONALIASING;
+                        if ((cachetype & (3 << 14)) == 1 << 14)
+                                cacheid |= CACHEID_ASID_TAGGED;
+                } else if (cachetype & (1 << 23))
                         cacheid = CACHEID_VIPT_ALIASING;
                 else
                         cacheid = CACHEID_VIPT_NONALIASING;
diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c
index 9bb4f043aa22..7ac812dc055a 100644
--- a/arch/arm/mach-at91/pm.c
+++ b/arch/arm/mach-at91/pm.c
@@ -332,7 +332,6 @@ static int at91_pm_enter(suspend_state_t state)
                         at91_sys_read(AT91_AIC_IPR) & at91_sys_read(AT91_AIC_IMR));
 
 error:
-        sdram_selfrefresh_disable();
         target_state = PM_SUSPEND_ON;
         at91_irq_resume();
         at91_gpio_resume();
diff --git a/arch/arm/mm/abort-ev6.S b/arch/arm/mm/abort-ev6.S
index 8a7f65ba14b7..94077fbd96b7 100644
--- a/arch/arm/mm/abort-ev6.S
+++ b/arch/arm/mm/abort-ev6.S
@@ -23,7 +23,8 @@ ENTRY(v6_early_abort)
 #ifdef CONFIG_CPU_32v6K
         clrex
 #else
-        strex   r0, r1, [sp]                    @ Clear the exclusive monitor
+        sub     r1, sp, #4                      @ Get unused stack location
+        strex   r0, r1, [r1]                    @ Clear the exclusive monitor
 #endif
         mrc     p15, 0, r1, c5, c0, 0           @ get FSR
         mrc     p15, 0, r0, c6, c0, 0           @ get FAR
diff --git a/arch/arm/plat-s3c64xx/irq-eint.c b/arch/arm/plat-s3c64xx/irq-eint.c
index 1f7cc0067f5c..ebb305ce7689 100644
--- a/arch/arm/plat-s3c64xx/irq-eint.c
+++ b/arch/arm/plat-s3c64xx/irq-eint.c
@@ -55,7 +55,7 @@ static void s3c_irq_eint_unmask(unsigned int irq)
         u32 mask;
 
         mask = __raw_readl(S3C64XX_EINT0MASK);
-        mask |= eint_irq_to_bit(irq);
+        mask &= ~eint_irq_to_bit(irq);
         __raw_writel(mask, S3C64XX_EINT0MASK);
 }
 
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
index b189680d18b0..05fe3053dcae 100644
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -181,7 +181,7 @@ source "kernel/Kconfig.preempt"
 config QUICKLIST
         def_bool y
 
-config HAVE_ARCH_BOOTMEM_NODE
+config HAVE_ARCH_BOOTMEM
         def_bool n
 
 config ARCH_HAVE_MEMORY_PRESENT
diff --git a/arch/mips/include/asm/seccomp.h b/arch/mips/include/asm/seccomp.h
index 36ed44070256..a6772e9507f5 100644
--- a/arch/mips/include/asm/seccomp.h
+++ b/arch/mips/include/asm/seccomp.h
@@ -1,6 +1,5 @@
 #ifndef __ASM_SECCOMP_H
 
-#include <linux/thread_info.h>
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/powerpc/include/asm/compat.h b/arch/powerpc/include/asm/compat.h
index d811a8cd7b58..4774c2f92232 100644
--- a/arch/powerpc/include/asm/compat.h
+++ b/arch/powerpc/include/asm/compat.h
@@ -210,5 +210,10 @@ struct compat_shmid64_ds {
         compat_ulong_t __unused6;
 };
 
+static inline int is_compat_task(void)
+{
+        return test_thread_flag(TIF_32BIT);
+}
+
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_COMPAT_H */
diff --git a/arch/powerpc/include/asm/seccomp.h b/arch/powerpc/include/asm/seccomp.h
index 853765eb1f65..00c1d9133cfe 100644
--- a/arch/powerpc/include/asm/seccomp.h
+++ b/arch/powerpc/include/asm/seccomp.h
@@ -1,10 +1,6 @@
 #ifndef _ASM_POWERPC_SECCOMP_H
 #define _ASM_POWERPC_SECCOMP_H
 
-#ifdef __KERNEL__
-#include <linux/thread_info.h>
-#endif
-
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/powerpc/platforms/86xx/gef_sbc610.c b/arch/powerpc/platforms/86xx/gef_sbc610.c
index fb371f5ce132..d6b772ba3b8f 100644
--- a/arch/powerpc/platforms/86xx/gef_sbc610.c
+++ b/arch/powerpc/platforms/86xx/gef_sbc610.c
@@ -142,6 +142,10 @@ static void __init gef_sbc610_nec_fixup(struct pci_dev *pdev)
 {
         unsigned int val;
 
+        /* Do not do the fixup on other platforms! */
+        if (!machine_is(gef_sbc610))
+                return;
+
         printk(KERN_INFO "Running NEC uPD720101 Fixup\n");
 
         /* Ensure ports 1, 2, 3, 4 & 5 are enabled */
diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c
index c42cd898f68b..6118890c946d 100644
--- a/arch/s390/crypto/aes_s390.c
+++ b/arch/s390/crypto/aes_s390.c
@@ -556,7 +556,7 @@ static void __exit aes_s390_fini(void)
 module_init(aes_s390_init);
 module_exit(aes_s390_fini);
 
-MODULE_ALIAS("aes");
+MODULE_ALIAS("aes-all");
 
 MODULE_DESCRIPTION("Rijndael (AES) Cipher Algorithm");
 MODULE_LICENSE("GPL");
diff --git a/arch/sparc/include/asm/compat.h b/arch/sparc/include/asm/compat.h
index f260b58f5ce9..0e706257918f 100644
--- a/arch/sparc/include/asm/compat.h
+++ b/arch/sparc/include/asm/compat.h
@@ -240,4 +240,9 @@ struct compat_shmid64_ds {
         unsigned int    __unused2;
 };
 
+static inline int is_compat_task(void)
+{
+        return test_thread_flag(TIF_32BIT);
+}
+
 #endif /* _ASM_SPARC64_COMPAT_H */
diff --git a/arch/sparc/include/asm/seccomp.h b/arch/sparc/include/asm/seccomp.h
index 7fcd9968192b..adca1bce41d4 100644
--- a/arch/sparc/include/asm/seccomp.h
+++ b/arch/sparc/include/asm/seccomp.h
@@ -1,11 +1,5 @@
 #ifndef _ASM_SECCOMP_H
 
-#include <linux/thread_info.h> /* already defines TIF_32BIT */
-
-#ifndef TIF_32BIT
-#error "unexpected TIF_32BIT on sparc64"
-#endif
-
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 3afc94abe357..31758378bcd2 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -40,6 +40,9 @@ config X86
         select HAVE_GENERIC_DMA_COHERENT if X86_32
         select HAVE_EFFICIENT_UNALIGNED_ACCESS
         select USER_STACKTRACE_SUPPORT
+        select HAVE_KERNEL_GZIP
+        select HAVE_KERNEL_BZIP2
+        select HAVE_KERNEL_LZMA
 
 config ARCH_DEFCONFIG
         string
@@ -135,6 +138,9 @@ config ARCH_HAS_CACHE_LINE_SIZE
 config HAVE_SETUP_PER_CPU_AREA
         def_bool y
 
+config HAVE_DYNAMIC_PER_CPU_AREA
+        def_bool y
+
 config HAVE_CPUMASK_OF_CPU_MAP
         def_bool X86_64_SMP
 
@@ -1127,7 +1133,7 @@ config NODES_SHIFT
           Specify the maximum number of NUMA Nodes available on the target
           system.  Increases memory reserved to accomodate various tables.
 
-config HAVE_ARCH_BOOTMEM_NODE
+config HAVE_ARCH_BOOTMEM
         def_bool y
         depends on X86_32 && NUMA
 
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 1771c804e02f..3ca4c194b8e5 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -4,7 +4,7 @@
 # create a compressed vmlinux image from the original vmlinux
 #
 
-targets := vmlinux vmlinux.bin vmlinux.bin.gz head_$(BITS).o misc.o piggy.o
+targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma head_$(BITS).o misc.o piggy.o
 
 KBUILD_CFLAGS := -m$(BITS) -D__KERNEL__ $(LINUX_INCLUDE) -O2
 KBUILD_CFLAGS += -fno-strict-aliasing -fPIC
@@ -47,18 +47,35 @@ ifeq ($(CONFIG_X86_32),y)
 ifdef CONFIG_RELOCATABLE
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin.all FORCE
         $(call if_changed,gzip)
+$(obj)/vmlinux.bin.bz2: $(obj)/vmlinux.bin.all FORCE
+        $(call if_changed,bzip2)
+$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin.all FORCE
+        $(call if_changed,lzma)
 else
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE
         $(call if_changed,gzip)
+$(obj)/vmlinux.bin.bz2: $(obj)/vmlinux.bin FORCE
+        $(call if_changed,bzip2)
+$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin FORCE
+        $(call if_changed,lzma)
 endif
 LDFLAGS_piggy.o := -r --format binary --oformat elf32-i386 -T
 
 else
+
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE
         $(call if_changed,gzip)
+$(obj)/vmlinux.bin.bz2: $(obj)/vmlinux.bin FORCE
+        $(call if_changed,bzip2)
+$(obj)/vmlinux.bin.lzma: $(obj)/vmlinux.bin FORCE
+        $(call if_changed,lzma)
 
 LDFLAGS_piggy.o := -r --format binary --oformat elf64-x86-64 -T
 endif
 
-$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE
+suffix_$(CONFIG_KERNEL_GZIP)  = gz
+suffix_$(CONFIG_KERNEL_BZIP2) = bz2
+suffix_$(CONFIG_KERNEL_LZMA)  = lzma
+
+$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.$(suffix_y) FORCE
         $(call if_changed,ld)
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
index da062216948a..e45be73684ff 100644
--- a/arch/x86/boot/compressed/misc.c
+++ b/arch/x86/boot/compressed/misc.c
@@ -116,71 +116,13 @@
 /*
  * gzip declarations
  */
-
-#define OF(args)        args
 #define STATIC          static
 
 #undef memset
 #undef memcpy
 #define memzero(s, n)   memset((s), 0, (n))
 
-typedef unsigned char   uch;
-typedef unsigned short  ush;
-typedef unsigned long   ulg;
-
-/*
- * Window size must be at least 32k, and a power of two.
- * We don't actually have a window just a huge output buffer,
- * so we report a 2G window size, as that should always be
- * larger than our output buffer:
- */
-#define WSIZE           0x80000000
-
-/* Input buffer: */
-static unsigned char    *inbuf;
-
-/* Sliding window buffer (and final output buffer): */
-static unsigned char    *window;
-
-/* Valid bytes in inbuf: */
-static unsigned         insize;
-
-/* Index of next byte to be processed in inbuf: */
-static unsigned         inptr;
-
-/* Bytes in output buffer: */
-static unsigned         outcnt;
-
-/* gzip flag byte */
-#define ASCII_FLAG      0x01 /* bit 0 set: file probably ASCII text */
-#define CONTINUATION    0x02 /* bit 1 set: continuation of multi-part gz file */
-#define EXTRA_FIELD     0x04 /* bit 2 set: extra field present */
-#define ORIG_NAM        0x08 /* bit 3 set: original file name present */
-#define COMMENT         0x10 /* bit 4 set: file comment present */
-#define ENCRYPTED       0x20 /* bit 5 set: file is encrypted */
-#define RESERVED        0xC0 /* bit 6, 7:  reserved */
-
-#define get_byte()      (inptr < insize ? inbuf[inptr++] : fill_inbuf())
-
-/* Diagnostic functions */
-#ifdef DEBUG
-#  define Assert(cond, msg) do { if (!(cond)) error(msg); } while (0)
-#  define Trace(x)      do { fprintf x; } while (0)
-#  define Tracev(x)     do { if (verbose) fprintf x ; } while (0)
-#  define Tracevv(x)    do { if (verbose > 1) fprintf x ; } while (0)
-#  define Tracec(c, x)  do { if (verbose && (c)) fprintf x ; } while (0)
-#  define Tracecv(c, x) do { if (verbose > 1 && (c)) fprintf x ; } while (0)
-#else
-#  define Assert(cond, msg)
-#  define Trace(x)
-#  define Tracev(x)
-#  define Tracevv(x)
-#  define Tracec(c, x)
-#  define Tracecv(c, x)
-#endif
 
-static int  fill_inbuf(void);
-static void flush_window(void);
 static void error(char *m);
 
 /*
@@ -189,13 +131,8 @@ static void error(char *m);
 static struct boot_params *real_mode;           /* Pointer to real-mode data */
 static int quiet;
 
-extern unsigned char input_data[];
-extern int input_len;
-
-static long bytes_out;
-
 static void *memset(void *s, int c, unsigned n);
-static void *memcpy(void *dest, const void *src, unsigned n);
+void *memcpy(void *dest, const void *src, unsigned n);
 
 static void __putstr(int, const char *);
 #define putstr(__x)  __putstr(0, __x)
@@ -213,7 +150,17 @@ static char *vidmem;
 static int vidport;
 static int lines, cols;
 
-#include "../../../../lib/inflate.c"
+#ifdef CONFIG_KERNEL_GZIP
+#include "../../../../lib/decompress_inflate.c"
+#endif
+
+#ifdef CONFIG_KERNEL_BZIP2
+#include "../../../../lib/decompress_bunzip2.c"
+#endif
+
+#ifdef CONFIG_KERNEL_LZMA
+#include "../../../../lib/decompress_unlzma.c"
+#endif
 
 static void scroll(void)
 {
@@ -282,7 +229,7 @@ static void *memset(void *s, int c, unsigned n)
         return s;
 }
 
-static void *memcpy(void *dest, const void *src, unsigned n)
+void *memcpy(void *dest, const void *src, unsigned n)
 {
         int i;
         const char *s = src;
@@ -293,38 +240,6 @@ static void *memcpy(void *dest, const void *src, unsigned n)
         return dest;
 }
 
-/* ===========================================================================
- * Fill the input buffer. This is called only when the buffer is empty
- * and at least one byte is really needed.
- */
-static int fill_inbuf(void)
-{
-        error("ran out of input data");
-        return 0;
-}
-
-/* ===========================================================================
- * Write the output window window[0..outcnt-1] and update crc and bytes_out.
- * (Used for the decompressed data only.)
- */
-static void flush_window(void)
-{
-        /* With my window equal to my output buffer
-         * I only need to compute the crc here.
-         */
-        unsigned long c = crc;         /* temporary variable */
-        unsigned n;
-        unsigned char *in, ch;
-
-        in = window;
-        for (n = 0; n < outcnt; n++) {
-                ch = *in++;
-                c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8);
-        }
-        crc = c;
-        bytes_out += (unsigned long)outcnt;
-        outcnt = 0;
-}
 
 static void error(char *x)
 {
@@ -407,12 +322,8 @@ asmlinkage void decompress_kernel(void *rmode, memptr heap,
         lines = real_mode->screen_info.orig_video_lines;
         cols = real_mode->screen_info.orig_video_cols;
 
-        window = output;                /* Output buffer (Normally at 1M) */
         free_mem_ptr     = heap;        /* Heap */
         free_mem_end_ptr = heap + BOOT_HEAP_SIZE;
-        inbuf  = input_data;            /* Input buffer */
-        insize = input_len;
-        inptr  = 0;
 
 #ifdef CONFIG_X86_64
         if ((unsigned long)output & (__KERNEL_ALIGN - 1))
@@ -430,10 +341,9 @@ asmlinkage void decompress_kernel(void *rmode, memptr heap,
 #endif
 #endif
 
-        makecrc();
         if (!quiet)
                 putstr("\nDecompressing Linux... ");
-        gunzip();
+        decompress(input_data, input_len, NULL, NULL, output, NULL, error);
         parse_elf(output);
         if (!quiet)
                 putstr("done.\nBooting the kernel.\n");
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index a6208dc74633..4ef949c1972e 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -75,7 +75,14 @@ static inline void default_inquire_remote_apic(int apicid)
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
 
+#ifdef CONFIG_X86_VSMP
 extern int is_vsmp_box(void);
+#else
+static inline int is_vsmp_box(void)
+{
+        return 0;
+}
+#endif
 extern void xapic_wait_icr_idle(void);
 extern u32 safe_xapic_wait_icr_idle(void);
 extern void xapic_icr_write(u32, u32);
@@ -306,7 +313,7 @@ struct apic {
         void (*send_IPI_self)(int vector);
 
         /* wakeup_secondary_cpu */
-        int (*wakeup_cpu)(int apicid, unsigned long start_eip);
+        int (*wakeup_secondary_cpu)(int apicid, unsigned long start_eip);
 
         int trampoline_phys_low;
         int trampoline_phys_high;
@@ -324,8 +331,21 @@ struct apic {
         u32 (*safe_wait_icr_idle)(void);
 };
 
+/*
+ * Pointer to the local APIC driver in use on this system (there's
+ * always just one such driver in use - the kernel decides via an
+ * early probing process which one it picks - and then sticks to it):
+ */
 extern struct apic *apic;
 
+/*
+ * APIC functionality to boot other CPUs - only used on SMP:
+ */
+#ifdef CONFIG_SMP
+extern atomic_t init_deasserted;
+extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip);
+#endif
+
 static inline u32 apic_read(u32 reg)
 {
         return apic->read(reg);
@@ -384,9 +404,7 @@ static inline unsigned default_get_apic_id(unsigned long x)
 #define DEFAULT_TRAMPOLINE_PHYS_LOW             0x467
 #define DEFAULT_TRAMPOLINE_PHYS_HIGH            0x469
 
-#ifdef CONFIG_X86_32
-extern void es7000_update_apic_to_cluster(void);
-#else
+#ifdef CONFIG_X86_64
 extern struct apic apic_flat;
 extern struct apic apic_physflat;
 extern struct apic apic_x2apic_cluster;
diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index dd61616cb73d..6526cf08b0e4 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -10,17 +10,31 @@
 #define EXTENDED_VGA    0xfffe          /* 80x50 mode */
 #define ASK_VGA         0xfffd          /* ask for it at bootup */
 
+#ifdef __KERNEL__
+
 /* Physical address where kernel should be loaded. */
 #define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \
                                 + (CONFIG_PHYSICAL_ALIGN - 1)) \
                                 & ~(CONFIG_PHYSICAL_ALIGN - 1))
 
+#ifdef CONFIG_KERNEL_BZIP2
+#define BOOT_HEAP_SIZE             0x400000
+#else /* !CONFIG_KERNEL_BZIP2 */
+
 #ifdef CONFIG_X86_64
 #define BOOT_HEAP_SIZE  0x7000
-#define BOOT_STACK_SIZE 0x4000
 #else
 #define BOOT_HEAP_SIZE  0x4000
+#endif
+
+#endif /* !CONFIG_KERNEL_BZIP2 */
+
+#ifdef CONFIG_X86_64
+#define BOOT_STACK_SIZE 0x4000
+#else
 #define BOOT_STACK_SIZE 0x1000
 #endif
 
+#endif /* __KERNEL__ */
+
 #endif /* _ASM_X86_BOOT_H */
diff --git a/arch/x86/include/asm/cacheflush.h b/arch/x86/include/asm/cacheflush.h
index 2f8466540fb5..5b301b7ff5f4 100644
--- a/arch/x86/include/asm/cacheflush.h
+++ b/arch/x86/include/asm/cacheflush.h
@@ -5,24 +5,43 @@
 #include <linux/mm.h>
 
 /* Caches aren't brain-dead on the intel. */
-#define flush_cache_all()                       do { } while (0)
-#define flush_cache_mm(mm)                      do { } while (0)
-#define flush_cache_dup_mm(mm)                  do { } while (0)
-#define flush_cache_range(vma, start, end)      do { } while (0)
-#define flush_cache_page(vma, vmaddr, pfn)      do { } while (0)
-#define flush_dcache_page(page)                 do { } while (0)
-#define flush_dcache_mmap_lock(mapping)         do { } while (0)
-#define flush_dcache_mmap_unlock(mapping)       do { } while (0)
-#define flush_icache_range(start, end)          do { } while (0)
-#define flush_icache_page(vma, pg)              do { } while (0)
-#define flush_icache_user_range(vma, pg, adr, len)      do { } while (0)
-#define flush_cache_vmap(start, end)            do { } while (0)
-#define flush_cache_vunmap(start, end)          do { } while (0)
+static inline void flush_cache_all(void) { }
+static inline void flush_cache_mm(struct mm_struct *mm) { }
+static inline void flush_cache_dup_mm(struct mm_struct *mm) { }
+static inline void flush_cache_range(struct vm_area_struct *vma,
+                                     unsigned long start, unsigned long end) { }
+static inline void flush_cache_page(struct vm_area_struct *vma,
+                                    unsigned long vmaddr, unsigned long pfn) { }
+static inline void flush_dcache_page(struct page *page) { }
+static inline void flush_dcache_mmap_lock(struct address_space *mapping) { }
+static inline void flush_dcache_mmap_unlock(struct address_space *mapping) { }
+static inline void flush_icache_range(unsigned long start,
+                                      unsigned long end) { }
+static inline void flush_icache_page(struct vm_area_struct *vma,
+                                     struct page *page) { }
+static inline void flush_icache_user_range(struct vm_area_struct *vma,
+                                           struct page *page,
+                                           unsigned long addr,
+                                           unsigned long len) { }
+static inline void flush_cache_vmap(unsigned long start, unsigned long end) { }
+static inline void flush_cache_vunmap(unsigned long start,
+                                      unsigned long end) { }
 
-#define copy_to_user_page(vma, page, vaddr, dst, src, len)      \
-        memcpy((dst), (src), (len))
-#define copy_from_user_page(vma, page, vaddr, dst, src, len)    \
-        memcpy((dst), (src), (len))
+static inline void copy_to_user_page(struct vm_area_struct *vma,
+                                     struct page *page, unsigned long vaddr,
+                                     void *dst, const void *src,
+                                     unsigned long len)
+{
+        memcpy(dst, src, len);
+}
+
+static inline void copy_from_user_page(struct vm_area_struct *vma,
+                                       struct page *page, unsigned long vaddr,
+                                       void *dst, const void *src,
+                                       unsigned long len)
+{
+        memcpy(dst, src, len);
+}
 
 #define PG_non_WB                               PG_arch_1
 PAGEFLAG(NonWB, non_WB)
diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h
index ca5ffb2856b6..edc90f23e708 100644
--- a/arch/x86/include/asm/efi.h
+++ b/arch/x86/include/asm/efi.h
@@ -37,8 +37,6 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...);
 
 #else /* !CONFIG_X86_32 */
 
-#define MAX_EFI_IO_PAGES        100
-
 extern u64 efi_call0(void *fp);
 extern u64 efi_call1(void *fp, u64 arg1);
 extern u64 efi_call2(void *fp, u64 arg1, u64 arg2);
diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h
index 23696d44a0af..63a79c77d220 100644
--- a/arch/x86/include/asm/fixmap.h
+++ b/arch/x86/include/asm/fixmap.h
@@ -1,11 +1,145 @@
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1998 Ingo Molnar
+ *
+ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ * x86_32 and x86_64 integration by Gustavo F. Padovan, February 2009
+ */
+
 #ifndef _ASM_X86_FIXMAP_H
 #define _ASM_X86_FIXMAP_H
 
+#ifndef __ASSEMBLY__
+#include <linux/kernel.h>
+#include <asm/acpi.h>
+#include <asm/apicdef.h>
+#include <asm/page.h>
+#ifdef CONFIG_X86_32
+#include <linux/threads.h>
+#include <asm/kmap_types.h>
+#else
+#include <asm/vsyscall.h>
+#endif
+
+/*
+ * We can't declare FIXADDR_TOP as variable for x86_64 because vsyscall
+ * uses fixmaps that relies on FIXADDR_TOP for proper address calculation.
+ * Because of this, FIXADDR_TOP x86 integration was left as later work.
+ */
+#ifdef CONFIG_X86_32
+/* used by vmalloc.c, vsyscall.lds.S.
+ *
+ * Leave one empty page between vmalloc'ed areas and
+ * the start of the fixmap.
+ */
+extern unsigned long __FIXADDR_TOP;
+#define FIXADDR_TOP     ((unsigned long)__FIXADDR_TOP)
+
+#define FIXADDR_USER_START     __fix_to_virt(FIX_VDSO)
+#define FIXADDR_USER_END       __fix_to_virt(FIX_VDSO - 1)
+#else
+#define FIXADDR_TOP     (VSYSCALL_END-PAGE_SIZE)
+
+/* Only covers 32bit vsyscalls currently. Need another set for 64bit. */
+#define FIXADDR_USER_START      ((unsigned long)VSYSCALL32_VSYSCALL)
+#define FIXADDR_USER_END        (FIXADDR_USER_START + PAGE_SIZE)
+#endif
+
+
+/*
+ * Here we define all the compile-time 'special' virtual
+ * addresses. The point is to have a constant address at
+ * compile time, but to set the physical address only
+ * in the boot process.
+ * for x86_32: We allocate these special addresses
+ * from the end of virtual memory (0xfffff000) backwards.
+ * Also this lets us do fail-safe vmalloc(), we
+ * can guarantee that these special addresses and
+ * vmalloc()-ed addresses never overlap.
+ *
+ * These 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages (or larger if used with an increment
+ * higher than 1). Use set_fixmap(idx,phys) to associate
+ * physical memory with fixmap indices.
+ *
+ * TLB entries of such buffers will not be flushed across
+ * task switches.
+ */
+enum fixed_addresses {
 #ifdef CONFIG_X86_32
-# include "fixmap_32.h"
+        FIX_HOLE,
+        FIX_VDSO,
 #else
-# include "fixmap_64.h"
+        VSYSCALL_LAST_PAGE,
+        VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE
+                            + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
+        VSYSCALL_HPET,
 #endif
+        FIX_DBGP_BASE,
+        FIX_EARLYCON_MEM_BASE,
+#ifdef CONFIG_X86_LOCAL_APIC
+        FIX_APIC_BASE,  /* local (CPU) APIC) -- required for SMP or not */
+#endif
+#ifdef CONFIG_X86_IO_APIC
+        FIX_IO_APIC_BASE_0,
+        FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1,
+#endif
+#ifdef CONFIG_X86_VISWS_APIC
+        FIX_CO_CPU,     /* Cobalt timer */
+        FIX_CO_APIC,    /* Cobalt APIC Redirection Table */
+        FIX_LI_PCIA,    /* Lithium PCI Bridge A */
+        FIX_LI_PCIB,    /* Lithium PCI Bridge B */
+#endif
+#ifdef CONFIG_X86_F00F_BUG
+        FIX_F00F_IDT,   /* Virtual mapping for IDT */
+#endif
+#ifdef CONFIG_X86_CYCLONE_TIMER
+        FIX_CYCLONE_TIMER, /*cyclone timer register*/
+#endif
+#ifdef CONFIG_X86_32
+        FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */
+        FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
+#ifdef CONFIG_PCI_MMCONFIG
+        FIX_PCIE_MCFG,
+#endif
+#endif
+#ifdef CONFIG_PARAVIRT
+        FIX_PARAVIRT_BOOTMAP,
+#endif
+        __end_of_permanent_fixed_addresses,
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+        FIX_OHCI1394_BASE,
+#endif
+        /*
+         * 256 temporary boot-time mappings, used by early_ioremap(),
+         * before ioremap() is functional.
+         *
+         * We round it up to the next 256 pages boundary so that we
+         * can have a single pgd entry and a single pte table:
+         */
+#define NR_FIX_BTMAPS           64
+#define FIX_BTMAPS_SLOTS        4
+        FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
+                        (__end_of_permanent_fixed_addresses & 255),
+        FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
+#ifdef CONFIG_X86_32
+        FIX_WP_TEST,
+#endif
+        __end_of_fixed_addresses
+};
+
+
+extern void reserve_top_address(unsigned long reserve);
+
+#define FIXADDR_SIZE    (__end_of_permanent_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_BOOT_SIZE       (__end_of_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_START           (FIXADDR_TOP - FIXADDR_SIZE)
+#define FIXADDR_BOOT_START      (FIXADDR_TOP - FIXADDR_BOOT_SIZE)
 
 extern int fixmaps_set;
 
@@ -69,4 +203,5 @@ static inline unsigned long virt_to_fix(const unsigned long vaddr)
         BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
         return __virt_to_fix(vaddr);
 }
+#endif /* !__ASSEMBLY__ */
 #endif /* _ASM_X86_FIXMAP_H */
diff --git a/arch/x86/include/asm/fixmap_32.h b/arch/x86/include/asm/fixmap_32.h
deleted file mode 100644
index 047d9bab2b31..000000000000
--- a/arch/x86/include/asm/fixmap_32.h
+++ /dev/null
@@ -1,115 +0,0 @@
-/*
- * fixmap.h: compile-time virtual memory allocation
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1998 Ingo Molnar
- *
- * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
- */
-
-#ifndef _ASM_X86_FIXMAP_32_H
-#define _ASM_X86_FIXMAP_32_H
-
-
-/* used by vmalloc.c, vsyscall.lds.S.
- *
- * Leave one empty page between vmalloc'ed areas and
- * the start of the fixmap.
- */
-extern unsigned long __FIXADDR_TOP;
-#define FIXADDR_USER_START     __fix_to_virt(FIX_VDSO)
-#define FIXADDR_USER_END       __fix_to_virt(FIX_VDSO - 1)
-
-#ifndef __ASSEMBLY__
-#include <linux/kernel.h>
-#include <asm/acpi.h>
-#include <asm/apicdef.h>
-#include <asm/page.h>
-#include <linux/threads.h>
-#include <asm/kmap_types.h>
-
-/*
- * Here we define all the compile-time 'special' virtual
- * addresses. The point is to have a constant address at
- * compile time, but to set the physical address only
- * in the boot process. We allocate these special addresses
- * from the end of virtual memory (0xfffff000) backwards.
- * Also this lets us do fail-safe vmalloc(), we
- * can guarantee that these special addresses and
- * vmalloc()-ed addresses never overlap.
- *
- * these 'compile-time allocated' memory buffers are
- * fixed-size 4k pages. (or larger if used with an increment
- * highger than 1) use fixmap_set(idx,phys) to associate
- * physical memory with fixmap indices.
- *
- * TLB entries of such buffers will not be flushed across
- * task switches.
- */
-enum fixed_addresses {
-        FIX_HOLE,
-        FIX_VDSO,
-        FIX_DBGP_BASE,
-        FIX_EARLYCON_MEM_BASE,
-#ifdef CONFIG_X86_LOCAL_APIC
-        FIX_APIC_BASE,  /* local (CPU) APIC) -- required for SMP or not */
-#endif
-#ifdef CONFIG_X86_IO_APIC
-        FIX_IO_APIC_BASE_0,
-        FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
-#endif
-#ifdef CONFIG_X86_VISWS_APIC
-        FIX_CO_CPU,     /* Cobalt timer */
-        FIX_CO_APIC,    /* Cobalt APIC Redirection Table */
-        FIX_LI_PCIA,    /* Lithium PCI Bridge A */
-        FIX_LI_PCIB,    /* Lithium PCI Bridge B */
-#endif
-#ifdef CONFIG_X86_F00F_BUG
-        FIX_F00F_IDT,   /* Virtual mapping for IDT */
-#endif
-#ifdef CONFIG_X86_CYCLONE_TIMER
-        FIX_CYCLONE_TIMER, /*cyclone timer register*/
-#endif
-        FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */
-        FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
-#ifdef CONFIG_PCI_MMCONFIG
-        FIX_PCIE_MCFG,
-#endif
-#ifdef CONFIG_PARAVIRT
-        FIX_PARAVIRT_BOOTMAP,
-#endif
-        __end_of_permanent_fixed_addresses,
-        /*
-         * 256 temporary boot-time mappings, used by early_ioremap(),
-         * before ioremap() is functional.
-         *
-         * We round it up to the next 256 pages boundary so that we
-         * can have a single pgd entry and a single pte table:
-         */
-#define NR_FIX_BTMAPS           64
-#define FIX_BTMAPS_SLOTS        4
-        FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
-                        (__end_of_permanent_fixed_addresses & 255),
-        FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
-        FIX_WP_TEST,
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-        FIX_OHCI1394_BASE,
-#endif
-        __end_of_fixed_addresses
-};
-
-extern void reserve_top_address(unsigned long reserve);
-
-
-#define FIXADDR_TOP     ((unsigned long)__FIXADDR_TOP)
-
-#define __FIXADDR_SIZE  (__end_of_permanent_fixed_addresses << PAGE_SHIFT)
-#define __FIXADDR_BOOT_SIZE     (__end_of_fixed_addresses << PAGE_SHIFT)
-#define FIXADDR_START           (FIXADDR_TOP - __FIXADDR_SIZE)
-#define FIXADDR_BOOT_START      (FIXADDR_TOP - __FIXADDR_BOOT_SIZE)
-
-#endif /* !__ASSEMBLY__ */
-#endif /* _ASM_X86_FIXMAP_32_H */
diff --git a/arch/x86/include/asm/fixmap_64.h b/arch/x86/include/asm/fixmap_64.h
deleted file mode 100644
index 298d9ba3faeb..000000000000
--- a/arch/x86/include/asm/fixmap_64.h
+++ /dev/null
@@ -1,79 +0,0 @@
-/*
- * fixmap.h: compile-time virtual memory allocation
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1998 Ingo Molnar
- */
-
-#ifndef _ASM_X86_FIXMAP_64_H
-#define _ASM_X86_FIXMAP_64_H
-
-#include <linux/kernel.h>
-#include <asm/acpi.h>
-#include <asm/apicdef.h>
-#include <asm/page.h>
-#include <asm/vsyscall.h>
-#include <asm/efi.h>
-
-/*
- * Here we define all the compile-time 'special' virtual
- * addresses. The point is to have a constant address at
- * compile time, but to set the physical address only
- * in the boot process.
- *
- * These 'compile-time allocated' memory buffers are
- * fixed-size 4k pages (or larger if used with an increment
- * higher than 1). Use set_fixmap(idx,phys) to associate
- * physical memory with fixmap indices.
- *
- * TLB entries of such buffers will not be flushed across
- * task switches.
- */
-
-enum fixed_addresses {
-        VSYSCALL_LAST_PAGE,
-        VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE
-                            + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
-        VSYSCALL_HPET,
-        FIX_DBGP_BASE,
-        FIX_EARLYCON_MEM_BASE,
-        FIX_APIC_BASE,  /* local (CPU) APIC) -- required for SMP or not */
-        FIX_IO_APIC_BASE_0,
-        FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1,
-        FIX_EFI_IO_MAP_LAST_PAGE,
-        FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE
-                                  + MAX_EFI_IO_PAGES - 1,
-#ifdef CONFIG_PARAVIRT
-        FIX_PARAVIRT_BOOTMAP,
-#endif
-        __end_of_permanent_fixed_addresses,
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-        FIX_OHCI1394_BASE,
-#endif
-        /*
-         * 256 temporary boot-time mappings, used by early_ioremap(),
-         * before ioremap() is functional.
-         *
-         * We round it up to the next 256 pages boundary so that we
-         * can have a single pgd entry and a single pte table:
-         */
-#define NR_FIX_BTMAPS           64
-#define FIX_BTMAPS_SLOTS        4
-        FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
-                        (__end_of_permanent_fixed_addresses & 255),
-        FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
-        __end_of_fixed_addresses
-};
-
-#define FIXADDR_TOP     (VSYSCALL_END-PAGE_SIZE)
-#define FIXADDR_SIZE    (__end_of_fixed_addresses << PAGE_SHIFT)
-#define FIXADDR_START   (FIXADDR_TOP - FIXADDR_SIZE)
-
-/* Only covers 32bit vsyscalls currently. Need another set for 64bit. */
-#define FIXADDR_USER_START      ((unsigned long)VSYSCALL32_VSYSCALL)
-#define FIXADDR_USER_END        (FIXADDR_USER_START + PAGE_SIZE)
-
-#endif /* _ASM_X86_FIXMAP_64_H */
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index 48f0004db8c9..71c9e5183982 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -172,7 +172,13 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 
 #else  /* CONFIG_X86_32 */
 
-extern void finit(void);
+#ifdef CONFIG_MATH_EMULATION
+extern void finit_task(struct task_struct *tsk);
+#else
+static inline void finit_task(struct task_struct *tsk)
+{
+}
+#endif
 
 static inline void tolerant_fwait(void)
 {
diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h
index 683d0b4c00fc..e5383e3d2f8c 100644
--- a/arch/x86/include/asm/io.h
+++ b/arch/x86/include/asm/io.h
@@ -172,8 +172,6 @@ static inline void __iomem *ioremap(resource_size_t offset, unsigned long size)
 
 extern void iounmap(volatile void __iomem *addr);
 
-extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
-
 
 #ifdef CONFIG_X86_32
 # include "io_32.h"
@@ -198,7 +196,6 @@ extern void early_ioremap_reset(void);
 extern void __iomem *early_ioremap(unsigned long offset, unsigned long size);
 extern void __iomem *early_memremap(unsigned long offset, unsigned long size);
 extern void early_iounmap(void __iomem *addr, unsigned long size);
-extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 
 #define IO_SPACE_LIMIT 0xffff
 
diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h
index 105fb90a0635..ede6998bd92c 100644
--- a/arch/x86/include/asm/mmzone_32.h
+++ b/arch/x86/include/asm/mmzone_32.h
@@ -91,46 +91,9 @@ static inline int pfn_valid(int pfn)
 #endif /* CONFIG_DISCONTIGMEM */
 
 #ifdef CONFIG_NEED_MULTIPLE_NODES
-
-/*
- * Following are macros that are specific to this numa platform.
- */
-#define reserve_bootmem(addr, size, flags) \
-        reserve_bootmem_node(NODE_DATA(0), (addr), (size), (flags))
-#define alloc_bootmem(x) \
-        __alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_nopanic(x) \
-        __alloc_bootmem_node_nopanic(NODE_DATA(0), (x), SMP_CACHE_BYTES, \
-                                __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low(x) \
-        __alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES, 0)
-#define alloc_bootmem_pages(x) \
-        __alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_pages_nopanic(x) \
-        __alloc_bootmem_node_nopanic(NODE_DATA(0), (x), PAGE_SIZE, \
-                                __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low_pages(x) \
-        __alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, 0)
-#define alloc_bootmem_node(pgdat, x)                                    \
-({                                                                      \
-        struct pglist_data  __maybe_unused                      \
-                                *__alloc_bootmem_node__pgdat = (pgdat); \
-        __alloc_bootmem_node(NODE_DATA(0), (x), SMP_CACHE_BYTES,        \
-                                                __pa(MAX_DMA_ADDRESS)); \
-})
-#define alloc_bootmem_pages_node(pgdat, x)                              \
-({                                                                      \
-        struct pglist_data  __maybe_unused                      \
-                                *__alloc_bootmem_node__pgdat = (pgdat); \
-        __alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE,              \
-                                                __pa(MAX_DMA_ADDRESS)); \
-})
-#define alloc_bootmem_low_pages_node(pgdat, x)                          \
-({                                                                      \
-        struct pglist_data  __maybe_unused                      \
-                                *__alloc_bootmem_node__pgdat = (pgdat); \
-        __alloc_bootmem_node(NODE_DATA(0), (x), PAGE_SIZE, 0);          \
-})
+/* always use node 0 for bootmem on this numa platform */
+#define bootmem_arch_preferred_node(__bdata, size, align, goal, limit)  \
+        (NODE_DATA(0)->bdata)
 #endif /* CONFIG_NEED_MULTIPLE_NODES */
 
 #endif /* _ASM_X86_MMZONE_32_H */
diff --git a/arch/x86/include/asm/numa_32.h b/arch/x86/include/asm/numa_32.h
index e9f5db796244..a37229011b56 100644
--- a/arch/x86/include/asm/numa_32.h
+++ b/arch/x86/include/asm/numa_32.h
@@ -4,8 +4,12 @@
 extern int pxm_to_nid(int pxm);
 extern void numa_remove_cpu(int cpu);
 
-#ifdef CONFIG_NUMA
+#ifdef CONFIG_HIGHMEM
 extern void set_highmem_pages_init(void);
+#else
+static inline void set_highmem_pages_init(void)
+{
+}
 #endif
 
 #endif /* _ASM_X86_NUMA_32_H */
diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h
index aee103b26d01..8f1d2fbec1d4 100644
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@ -43,6 +43,14 @@
 #else /* ...!ASSEMBLY */
 
 #include <linux/stringify.h>
+#include <asm/sections.h>
+
+#define __addr_to_pcpu_ptr(addr)                                        \
+        (void *)((unsigned long)(addr) - (unsigned long)pcpu_base_addr  \
+                 + (unsigned long)__per_cpu_start)
+#define __pcpu_ptr_to_addr(ptr)                                         \
+        (void *)((unsigned long)(ptr) + (unsigned long)pcpu_base_addr   \
+                 - (unsigned long)__per_cpu_start)
 
 #ifdef CONFIG_SMP
 #define __percpu_arg(x)         "%%"__stringify(__percpu_seg)":%P" #x
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 1c097a3a6669..d0812e155f1d 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -288,6 +288,8 @@ static inline int is_new_memtype_allowed(unsigned long flags,
         return 1;
 }
 
+pmd_t *populate_extra_pmd(unsigned long vaddr);
+pte_t *populate_extra_pte(unsigned long vaddr);
 #endif  /* __ASSEMBLY__ */
 
 #ifdef CONFIG_X86_32
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index c7a98f738210..76139506c3e4 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -248,7 +248,6 @@ struct x86_hw_tss {
 #define IO_BITMAP_LONGS                 (IO_BITMAP_BYTES/sizeof(long))
 #define IO_BITMAP_OFFSET                offsetof(struct tss_struct, io_bitmap)
 #define INVALID_IO_BITMAP_OFFSET        0x8000
-#define INVALID_IO_BITMAP_OFFSET_LAZY   0x9000
 
 struct tss_struct {
         /*
@@ -263,11 +262,6 @@ struct tss_struct {
          * be within the limit.
          */
         unsigned long           io_bitmap[IO_BITMAP_LONGS + 1];
-        /*
-         * Cache the current maximum and the last task that used the bitmap:
-         */
-        unsigned long           io_bitmap_max;
-        struct thread_struct    *io_bitmap_owner;
 
         /*
          * .. and then another 0x100 bytes for the emergency kernel stack:
diff --git a/arch/x86/include/asm/seccomp_32.h b/arch/x86/include/asm/seccomp_32.h
index a6ad87b352c4..b811d6f5780c 100644
--- a/arch/x86/include/asm/seccomp_32.h
+++ b/arch/x86/include/asm/seccomp_32.h
@@ -1,12 +1,6 @@
 #ifndef _ASM_X86_SECCOMP_32_H
 #define _ASM_X86_SECCOMP_32_H
 
-#include <linux/thread_info.h>
-
-#ifdef TIF_32BIT
-#error "unexpected TIF_32BIT on i386"
-#endif
-
 #include <linux/unistd.h>
 
 #define __NR_seccomp_read __NR_read
diff --git a/arch/x86/include/asm/seccomp_64.h b/arch/x86/include/asm/seccomp_64.h
index 4171bb794e9e..84ec1bd161a5 100644
--- a/arch/x86/include/asm/seccomp_64.h
+++ b/arch/x86/include/asm/seccomp_64.h
@@ -1,14 +1,6 @@
 #ifndef _ASM_X86_SECCOMP_64_H
 #define _ASM_X86_SECCOMP_64_H
 
-#include <linux/thread_info.h>
-
-#ifdef TIF_32BIT
-#error "unexpected TIF_32BIT on x86_64"
-#else
-#define TIF_32BIT TIF_IA32
-#endif
-
 #include <linux/unistd.h>
 #include <asm/ia32_unistd.h>
 
diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h
index 66801cb72f69..05c6f6b11fd5 100644
--- a/arch/x86/include/asm/setup.h
+++ b/arch/x86/include/asm/setup.h
@@ -31,7 +31,6 @@ struct x86_quirks {
         void (*smp_read_mpc_oem)(struct mpc_oemtable *oemtable,
                                 unsigned short oemsize);
         int (*setup_ioapic_ids)(void);
-        int (*update_apic)(void);
 };
 
 extern void x86_quirk_pre_intr_init(void);
@@ -65,7 +64,11 @@ extern void x86_quirk_time_init(void);
 #include <asm/bootparam.h>
 
 /* Interrupt control for vSMPowered x86_64 systems */
+#ifdef CONFIG_X86_VSMP
 void vsmp_init(void);
+#else
+static inline void vsmp_init(void) { }
+#endif
 
 void setup_bios_corruption_check(void);
 
@@ -77,8 +80,6 @@ static inline void visws_early_detect(void) { }
 static inline int is_visws_box(void) { return 0; }
 #endif
 
-extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip);
-extern int wakeup_secondary_cpu_via_init(int apicid, unsigned long start_eip);
 extern struct x86_quirks *x86_quirks;
 extern unsigned long saved_video_mode;
 
diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h
index c00bfdbdd456..643c59b4bc6e 100644
--- a/arch/x86/include/asm/system.h
+++ b/arch/x86/include/asm/system.h
@@ -20,6 +20,9 @@
 struct task_struct; /* one of the stranger aspects of C forward declarations */
 struct task_struct *__switch_to(struct task_struct *prev,
                                 struct task_struct *next);
+struct tss_struct;
+void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
+                      struct tss_struct *tss);
 
 #ifdef CONFIG_X86_32
 
diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h
index a0ba61386972..5e06259e90e5 100644
--- a/arch/x86/include/asm/uaccess_32.h
+++ b/arch/x86/include/asm/uaccess_32.h
@@ -157,7 +157,7 @@ __copy_from_user(void *to, const void __user *from, unsigned long n)
 }
 
 static __always_inline unsigned long __copy_from_user_nocache(void *to,
-                const void __user *from, unsigned long n, unsigned long total)
+                                const void __user *from, unsigned long n)
 {
         might_fault();
         if (__builtin_constant_p(n)) {
@@ -180,7 +180,7 @@ static __always_inline unsigned long __copy_from_user_nocache(void *to,
 
 static __always_inline unsigned long
 __copy_from_user_inatomic_nocache(void *to, const void __user *from,
-                                  unsigned long n, unsigned long total)
+                                  unsigned long n)
 {
        return __copy_from_user_ll_nocache_nozero(to, from, n);
 }
diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h
index dcaa0404cf7b..8cc687326eb8 100644
--- a/arch/x86/include/asm/uaccess_64.h
+++ b/arch/x86/include/asm/uaccess_64.h
@@ -188,29 +188,18 @@ __copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
 extern long __copy_user_nocache(void *dst, const void __user *src,
                                 unsigned size, int zerorest);
 
-static inline int __copy_from_user_nocache(void *dst, const void __user *src,
-                                   unsigned size, unsigned long total)
+static inline int
+__copy_from_user_nocache(void *dst, const void __user *src, unsigned size)
 {
         might_sleep();
-        /*
-         * In practice this limit means that large file write()s
-         * which get chunked to 4K copies get handled via
-         * non-temporal stores here. Smaller writes get handled
-         * via regular __copy_from_user():
-         */
-        if (likely(total >= PAGE_SIZE))
-                return __copy_user_nocache(dst, src, size, 1);
-        else
-                return __copy_from_user(dst, src, size);
+        return __copy_user_nocache(dst, src, size, 1);
 }
 
-static inline int __copy_from_user_inatomic_nocache(void *dst,
-            const void __user *src, unsigned size, unsigned total)
+static inline int
+__copy_from_user_inatomic_nocache(void *dst, const void __user *src,
+                                  unsigned size)
 {
-        if (likely(total >= PAGE_SIZE))
-                return __copy_user_nocache(dst, src, size, 0);
-        else
-                return __copy_from_user_inatomic(dst, src, size);
+        return __copy_user_nocache(dst, src, size, 0);
 }
 
 unsigned long
diff --git a/arch/x86/include/asm/uv/uv.h b/arch/x86/include/asm/uv/uv.h
index 8242bf965812..c0a01b5d985b 100644
--- a/arch/x86/include/asm/uv/uv.h
+++ b/arch/x86/include/asm/uv/uv.h
@@ -12,7 +12,6 @@ extern enum uv_system_type get_uv_system_type(void);
 extern int is_uv_system(void);
 extern void uv_cpu_init(void);
 extern void uv_system_init(void);
-extern int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip);
 extern const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask,
                                                  struct mm_struct *mm,
                                                  unsigned long va,
@@ -24,8 +23,6 @@ static inline enum uv_system_type get_uv_system_type(void) { return UV_NONE; }
 static inline int is_uv_system(void)    { return 0; }
 static inline void uv_cpu_init(void)    { }
 static inline void uv_system_init(void) { }
-static inline int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
-{ return 1; }
 static inline const struct cpumask *
 uv_flush_tlb_others(const struct cpumask *cpumask, struct mm_struct *mm,
                     unsigned long va, unsigned int cpu)
diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h
index 4bd990ee43df..1a918dde46b5 100644
--- a/arch/x86/include/asm/xen/page.h
+++ b/arch/x86/include/asm/xen/page.h
@@ -164,6 +164,7 @@ static inline pte_t __pte_ma(pteval_t x)
 
 
 xmaddr_t arbitrary_virt_to_machine(void *address);
+unsigned long arbitrary_virt_to_mfn(void *vaddr);
 void make_lowmem_page_readonly(void *vaddr);
 void make_lowmem_page_readwrite(void *vaddr);
 
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index de5657c039e9..95f216bbfaf1 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -70,7 +70,7 @@ obj-$(CONFIG_FUNCTION_GRAPH_TRACER)	+= ftrace.o
 obj-$(CONFIG_KEXEC)             += machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)             += relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)        += crash_dump_$(BITS).o
-obj-y                           += vsmp_64.o
+obj-$(CONFIG_X86_VSMP)          += vsmp_64.o
 obj-$(CONFIG_KPROBES)           += kprobes.o
 obj-$(CONFIG_MODULES)           += module_$(BITS).o
 obj-$(CONFIG_EFI)               += efi.o efi_$(BITS).o efi_stub_$(BITS).o
diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c
index 3b002995e145..f933822dba18 100644
--- a/arch/x86/kernel/apic/apic_flat_64.c
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -222,7 +222,6 @@ struct apic apic_flat =  {
         .send_IPI_all                   = flat_send_IPI_all,
         .send_IPI_self                  = apic_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
         .wait_for_init_deassert         = NULL,
@@ -373,7 +372,6 @@ struct apic apic_physflat =  {
         .send_IPI_all                   = physflat_send_IPI_all,
         .send_IPI_self                  = apic_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
         .wait_for_init_deassert         = NULL,
diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c
index 0b1093394fdf..d806ecaa948f 100644
--- a/arch/x86/kernel/apic/bigsmp_32.c
+++ b/arch/x86/kernel/apic/bigsmp_32.c
@@ -16,17 +16,17 @@
 #include <asm/apic.h>
 #include <asm/ipi.h>
 
-static inline unsigned bigsmp_get_apic_id(unsigned long x)
+static unsigned bigsmp_get_apic_id(unsigned long x)
 {
         return (x >> 24) & 0xFF;
 }
 
-static inline int bigsmp_apic_id_registered(void)
+static int bigsmp_apic_id_registered(void)
 {
         return 1;
 }
 
-static inline const cpumask_t *bigsmp_target_cpus(void)
+static const cpumask_t *bigsmp_target_cpus(void)
 {
 #ifdef CONFIG_SMP
         return &cpu_online_map;
@@ -35,13 +35,12 @@ static inline const cpumask_t *bigsmp_target_cpus(void)
 #endif
 }
 
-static inline unsigned long
-bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid)
+static unsigned long bigsmp_check_apicid_used(physid_mask_t bitmap, int apicid)
 {
         return 0;
 }
 
-static inline unsigned long bigsmp_check_apicid_present(int bit)
+static unsigned long bigsmp_check_apicid_present(int bit)
 {
         return 1;
 }
@@ -64,7 +63,7 @@ static inline unsigned long calculate_ldr(int cpu)
  * an APIC.  See e.g. "AP-388 82489DX User's Manual" (Intel
  * document number 292116).  So here it goes...
  */
-static inline void bigsmp_init_apic_ldr(void)
+static void bigsmp_init_apic_ldr(void)
 {
         unsigned long val;
         int cpu = smp_processor_id();
@@ -74,19 +73,19 @@ static inline void bigsmp_init_apic_ldr(void)
         apic_write(APIC_LDR, val);
 }
 
-static inline void bigsmp_setup_apic_routing(void)
+static void bigsmp_setup_apic_routing(void)
 {
         printk(KERN_INFO
                 "Enabling APIC mode:  Physflat.  Using %d I/O APICs\n",
                 nr_ioapics);
 }
 
-static inline int bigsmp_apicid_to_node(int logical_apicid)
+static int bigsmp_apicid_to_node(int logical_apicid)
 {
         return apicid_2_node[hard_smp_processor_id()];
 }
 
-static inline int bigsmp_cpu_present_to_apicid(int mps_cpu)
+static int bigsmp_cpu_present_to_apicid(int mps_cpu)
 {
         if (mps_cpu < nr_cpu_ids)
                 return (int) per_cpu(x86_bios_cpu_apicid, mps_cpu);
@@ -94,7 +93,7 @@ static inline int bigsmp_cpu_present_to_apicid(int mps_cpu)
         return BAD_APICID;
 }
 
-static inline physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid)
+static physid_mask_t bigsmp_apicid_to_cpu_present(int phys_apicid)
 {
         return physid_mask_of_physid(phys_apicid);
 }
@@ -107,29 +106,24 @@ static inline int bigsmp_cpu_to_logical_apicid(int cpu)
         return cpu_physical_id(cpu);
 }
 
-static inline physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map)
+static physid_mask_t bigsmp_ioapic_phys_id_map(physid_mask_t phys_map)
 {
         /* For clustered we don't have a good way to do this yet - hack */
         return physids_promote(0xFFL);
 }
 
-static inline void bigsmp_setup_portio_remap(void)
-{
-}
-
-static inline int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid)
+static int bigsmp_check_phys_apicid_present(int boot_cpu_physical_apicid)
 {
         return 1;
 }
 
 /* As we are using single CPU as destination, pick only one CPU here */
-static inline unsigned int bigsmp_cpu_mask_to_apicid(const cpumask_t *cpumask)
+static unsigned int bigsmp_cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
         return bigsmp_cpu_to_logical_apicid(first_cpu(*cpumask));
 }
 
-static inline unsigned int
-bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+static unsigned int bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
                               const struct cpumask *andmask)
 {
         int cpu;
@@ -148,7 +142,7 @@ bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
         return BAD_APICID;
 }
 
-static inline int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
+static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
 {
         return cpuid_apic >> index_msb;
 }
@@ -158,12 +152,12 @@ static inline void bigsmp_send_IPI_mask(const struct cpumask *mask, int vector)
         default_send_IPI_mask_sequence_phys(mask, vector);
 }
 
-static inline void bigsmp_send_IPI_allbutself(int vector)
+static void bigsmp_send_IPI_allbutself(int vector)
 {
         default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector);
 }
 
-static inline void bigsmp_send_IPI_all(int vector)
+static void bigsmp_send_IPI_all(int vector)
 {
         bigsmp_send_IPI_mask(cpu_online_mask, vector);
 }
@@ -256,7 +250,6 @@ struct apic apic_bigsmp = {
         .send_IPI_all                   = bigsmp_send_IPI_all,
         .send_IPI_self                  = default_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c
index 320f2d2e4e54..19588f2770ee 100644
--- a/arch/x86/kernel/apic/es7000_32.c
+++ b/arch/x86/kernel/apic/es7000_32.c
@@ -163,22 +163,17 @@ static int wakeup_secondary_cpu_via_mip(int cpu, unsigned long eip)
         return 0;
 }
 
-static int __init es7000_update_apic(void)
+static int es7000_apic_is_cluster(void)
 {
-        apic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
-
         /* MPENTIUMIII */
         if (boot_cpu_data.x86 == 6 &&
-            (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) {
-                es7000_update_apic_to_cluster();
-                apic->wait_for_init_deassert = NULL;
-                apic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
-        }
+            (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11))
+                return 1;
 
         return 0;
 }
 
-static void __init setup_unisys(void)
+static void setup_unisys(void)
 {
         /*
          * Determine the generation of the ES7000 currently running.
@@ -192,14 +187,12 @@ static void __init setup_unisys(void)
         else
                 es7000_plat = ES7000_CLASSIC;
         ioapic_renumber_irq = es7000_rename_gsi;
-
-        x86_quirks->update_apic = es7000_update_apic;
 }
 
 /*
  * Parse the OEM Table:
  */
-static int __init parse_unisys_oem(char *oemptr)
+static int parse_unisys_oem(char *oemptr)
 {
         int                     i;
         int                     success = 0;
@@ -261,7 +254,7 @@ static int __init parse_unisys_oem(char *oemptr)
 }
 
 #ifdef CONFIG_ACPI
-static int __init find_unisys_acpi_oem_table(unsigned long *oem_addr)
+static int find_unisys_acpi_oem_table(unsigned long *oem_addr)
 {
         struct acpi_table_header *header = NULL;
         struct es7000_oem_table *table;
@@ -292,7 +285,7 @@ static int __init find_unisys_acpi_oem_table(unsigned long *oem_addr)
         return 0;
 }
 
-static void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr)
+static void unmap_unisys_acpi_oem_table(unsigned long oem_addr)
 {
         if (!oem_addr)
                 return;
@@ -310,8 +303,10 @@ static int es7000_check_dsdt(void)
         return 0;
 }
 
+static int es7000_acpi_ret;
+
 /* Hook from generic ACPI tables.c */
-static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 {
         unsigned long oem_addr = 0;
         int check_dsdt;
@@ -332,10 +327,26 @@ static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
                  */
                 unmap_unisys_acpi_oem_table(oem_addr);
         }
-        return ret;
+
+        es7000_acpi_ret = ret;
+
+        return ret && !es7000_apic_is_cluster();
 }
+
+static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id)
+{
+        int ret = es7000_acpi_ret;
+
+        return ret && es7000_apic_is_cluster();
+}
+
 #else /* !CONFIG_ACPI: */
-static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+        return 0;
+}
+
+static int es7000_acpi_madt_oem_check_cluster(char *oem_id, char *oem_table_id)
 {
         return 0;
 }
@@ -349,8 +360,7 @@ static void es7000_spin(int n)
                 rep_nop();
 }
 
-static int __init
-es7000_mip_write(struct mip_reg *mip_reg)
+static int es7000_mip_write(struct mip_reg *mip_reg)
 {
         int status = 0;
         int spin;
@@ -383,7 +393,7 @@ es7000_mip_write(struct mip_reg *mip_reg)
         return status;
 }
 
-static void __init es7000_enable_apic_mode(void)
+static void es7000_enable_apic_mode(void)
 {
         struct mip_reg es7000_mip_reg;
         int mip_status;
@@ -416,11 +426,8 @@ static void es7000_vector_allocation_domain(int cpu, cpumask_t *retmask)
 
 static void es7000_wait_for_init_deassert(atomic_t *deassert)
 {
-#ifndef CONFIG_ES7000_CLUSTERED_APIC
         while (!atomic_read(deassert))
                 cpu_relax();
-#endif
-        return;
 }
 
 static unsigned int es7000_get_apic_id(unsigned long x)
@@ -565,72 +572,24 @@ static int es7000_check_phys_apicid_present(int cpu_physical_apicid)
         return 1;
 }
 
-static unsigned int
-es7000_cpu_mask_to_apicid_cluster(const struct cpumask *cpumask)
-{
-        int cpus_found = 0;
-        int num_bits_set;
-        int apicid;
-        int cpu;
-
-        num_bits_set = cpumask_weight(cpumask);
-        /* Return id to all */
-        if (num_bits_set == nr_cpu_ids)
-                return 0xFF;
-        /*
-         * The cpus in the mask must all be on the apic cluster.  If are not
-         * on the same apicid cluster return default value of target_cpus():
-         */
-        cpu = cpumask_first(cpumask);
-        apicid = es7000_cpu_to_logical_apicid(cpu);
-
-        while (cpus_found < num_bits_set) {
-                if (cpumask_test_cpu(cpu, cpumask)) {
-                        int new_apicid = es7000_cpu_to_logical_apicid(cpu);
-
-                        if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-                                WARN(1, "Not a valid mask!");
-
-                                return 0xFF;
-                        }
-                        apicid = new_apicid;
-                        cpus_found++;
-                }
-                cpu++;
-        }
-        return apicid;
-}
-
 static unsigned int es7000_cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
-        int cpus_found = 0;
-        int num_bits_set;
-        int apicid;
-        int cpu;
+        unsigned int round = 0;
+        int cpu, uninitialized_var(apicid);
 
-        num_bits_set = cpus_weight(*cpumask);
-        /* Return id to all */
-        if (num_bits_set == nr_cpu_ids)
-                return es7000_cpu_to_logical_apicid(0);
         /*
-         * The cpus in the mask must all be on the apic cluster.  If are not
-         * on the same apicid cluster return default value of target_cpus():
+         * The cpus in the mask must all be on the apic cluster.
          */
-        cpu = first_cpu(*cpumask);
-        apicid = es7000_cpu_to_logical_apicid(cpu);
-        while (cpus_found < num_bits_set) {
-                if (cpu_isset(cpu, *cpumask)) {
-                        int new_apicid = es7000_cpu_to_logical_apicid(cpu);
+        for_each_cpu(cpu, cpumask) {
+                int new_apicid = es7000_cpu_to_logical_apicid(cpu);
 
-                        if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-                                printk("%s: Not a valid mask!\n", __func__);
+                if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
+                        WARN(1, "Not a valid mask!");
 
-                                return es7000_cpu_to_logical_apicid(0);
-                        }
-                        apicid = new_apicid;
-                        cpus_found++;
+                        return BAD_APICID;
                 }
-                cpu++;
+                apicid = new_apicid;
+                round++;
         }
         return apicid;
 }
@@ -659,37 +618,103 @@ static int es7000_phys_pkg_id(int cpuid_apic, int index_msb)
         return cpuid_apic >> index_msb;
 }
 
-void __init es7000_update_apic_to_cluster(void)
-{
-        apic->target_cpus = target_cpus_cluster;
-        apic->irq_delivery_mode = dest_LowestPrio;
-        /* logical delivery broadcast to all procs: */
-        apic->irq_dest_mode = 1;
-
-        apic->init_apic_ldr = es7000_init_apic_ldr_cluster;
-
-        apic->cpu_mask_to_apicid = es7000_cpu_mask_to_apicid_cluster;
-}
-
 static int probe_es7000(void)
 {
         /* probed later in mptable/ACPI hooks */
         return 0;
 }
 
-static __init int
-es7000_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid)
+static int es7000_mps_ret;
+static int es7000_mps_oem_check(struct mpc_table *mpc, char *oem,
+                char *productid)
 {
+        int ret = 0;
+
         if (mpc->oemptr) {
                 struct mpc_oemtable *oem_table =
                         (struct mpc_oemtable *)mpc->oemptr;
 
                 if (!strncmp(oem, "UNISYS", 6))
-                        return parse_unisys_oem((char *)oem_table);
+                        ret = parse_unisys_oem((char *)oem_table);
         }
-        return 0;
+
+        es7000_mps_ret = ret;
+
+        return ret && !es7000_apic_is_cluster();
 }
 
+static int es7000_mps_oem_check_cluster(struct mpc_table *mpc, char *oem,
+                char *productid)
+{
+        int ret = es7000_mps_ret;
+
+        return ret && es7000_apic_is_cluster();
+}
+
+struct apic apic_es7000_cluster = {
+
+        .name                           = "es7000",
+        .probe                          = probe_es7000,
+        .acpi_madt_oem_check            = es7000_acpi_madt_oem_check_cluster,
+        .apic_id_registered             = es7000_apic_id_registered,
+
+        .irq_delivery_mode              = dest_LowestPrio,
+        /* logical delivery broadcast to all procs: */
+        .irq_dest_mode                  = 1,
+
+        .target_cpus                    = target_cpus_cluster,
+        .disable_esr                    = 1,
+        .dest_logical                   = 0,
+        .check_apicid_used              = es7000_check_apicid_used,
+        .check_apicid_present           = es7000_check_apicid_present,
+
+        .vector_allocation_domain       = es7000_vector_allocation_domain,
+        .init_apic_ldr                  = es7000_init_apic_ldr_cluster,
+
+        .ioapic_phys_id_map             = es7000_ioapic_phys_id_map,
+        .setup_apic_routing             = es7000_setup_apic_routing,
+        .multi_timer_check              = NULL,
+        .apicid_to_node                 = es7000_apicid_to_node,
+        .cpu_to_logical_apicid          = es7000_cpu_to_logical_apicid,
+        .cpu_present_to_apicid          = es7000_cpu_present_to_apicid,
+        .apicid_to_cpu_present          = es7000_apicid_to_cpu_present,
+        .setup_portio_remap             = NULL,
+        .check_phys_apicid_present      = es7000_check_phys_apicid_present,
+        .enable_apic_mode               = es7000_enable_apic_mode,
+        .phys_pkg_id                    = es7000_phys_pkg_id,
+        .mps_oem_check                  = es7000_mps_oem_check_cluster,
+
+        .get_apic_id                    = es7000_get_apic_id,
+        .set_apic_id                    = NULL,
+        .apic_id_mask                   = 0xFF << 24,
+
+        .cpu_mask_to_apicid             = es7000_cpu_mask_to_apicid,
+        .cpu_mask_to_apicid_and         = es7000_cpu_mask_to_apicid_and,
+
+        .send_IPI_mask                  = es7000_send_IPI_mask,
+        .send_IPI_mask_allbutself       = NULL,
+        .send_IPI_allbutself            = es7000_send_IPI_allbutself,
+        .send_IPI_all                   = es7000_send_IPI_all,
+        .send_IPI_self                  = default_send_IPI_self,
+
+        .wakeup_secondary_cpu           = wakeup_secondary_cpu_via_mip,
+
+        .trampoline_phys_low            = 0x467,
+        .trampoline_phys_high           = 0x469,
+
+        .wait_for_init_deassert         = NULL,
+
+        /* Nothing to do for most platforms, since cleared by the INIT cycle: */
+        .smp_callin_clear_local_apic    = NULL,
+        .inquire_remote_apic            = default_inquire_remote_apic,
+
+        .read                           = native_apic_mem_read,
+        .write                          = native_apic_mem_write,
+        .icr_read                       = native_apic_icr_read,
+        .icr_write                      = native_apic_icr_write,
+        .wait_icr_idle                  = native_apic_wait_icr_idle,
+        .safe_wait_icr_idle             = native_safe_apic_wait_icr_idle,
+};
 
 struct apic apic_es7000 = {
 
@@ -737,8 +762,6 @@ struct apic apic_es7000 = {
         .send_IPI_all                   = es7000_send_IPI_all,
         .send_IPI_self                  = default_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
-
         .trampoline_phys_low            = 0x467,
         .trampoline_phys_high           = 0x469,
 
diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c
index d9d6d61eed82..ba2fc6465534 100644
--- a/arch/x86/kernel/apic/numaq_32.c
+++ b/arch/x86/kernel/apic/numaq_32.c
@@ -69,7 +69,7 @@ struct mpc_trans {
 /* x86_quirks member */
 static int                              mpc_record;
 
-static __cpuinitdata struct mpc_trans   *translation_table[MAX_MPC_ENTRY];
+static struct mpc_trans                 *translation_table[MAX_MPC_ENTRY];
 
 int                                     mp_bus_id_to_node[MAX_MP_BUSSES];
 int                                     mp_bus_id_to_local[MAX_MP_BUSSES];
@@ -256,13 +256,6 @@ static int __init numaq_setup_ioapic_ids(void)
         return 1;
 }
 
-static int __init numaq_update_apic(void)
-{
-        apic->wakeup_cpu = wakeup_secondary_cpu_via_nmi;
-
-        return 0;
-}
-
 static struct x86_quirks numaq_x86_quirks __initdata = {
         .arch_pre_time_init             = numaq_pre_time_init,
         .arch_time_init                 = NULL,
@@ -278,7 +271,6 @@ static struct x86_quirks numaq_x86_quirks __initdata = {
         .mpc_oem_pci_bus                = mpc_oem_pci_bus,
         .smp_read_mpc_oem               = smp_read_mpc_oem,
         .setup_ioapic_ids               = numaq_setup_ioapic_ids,
-        .update_apic                    = numaq_update_apic,
 };
 
 static __init void early_check_numaq(void)
@@ -546,7 +538,7 @@ struct apic apic_numaq = {
         .send_IPI_all                   = numaq_send_IPI_all,
         .send_IPI_self                  = default_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
+        .wakeup_secondary_cpu           = wakeup_secondary_cpu_via_nmi,
         .trampoline_phys_low            = NUMAQ_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = NUMAQ_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c
index 3a730fa574bb..141c99a1c264 100644
--- a/arch/x86/kernel/apic/probe_32.c
+++ b/arch/x86/kernel/apic/probe_32.c
@@ -138,7 +138,6 @@ struct apic apic_default = {
         .send_IPI_all                   = default_send_IPI_all,
         .send_IPI_self                  = default_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
@@ -159,6 +158,7 @@ extern struct apic apic_numaq;
 extern struct apic apic_summit;
 extern struct apic apic_bigsmp;
 extern struct apic apic_es7000;
+extern struct apic apic_es7000_cluster;
 extern struct apic apic_default;
 
 struct apic *apic = &apic_default;
@@ -176,6 +176,7 @@ static struct apic *apic_probe[] __initdata = {
 #endif
 #ifdef CONFIG_X86_ES7000
         &apic_es7000,
+        &apic_es7000_cluster,
 #endif
         &apic_default,  /* must be last */
         NULL,
@@ -197,9 +198,6 @@ static int __init parse_apic(char *arg)
                 }
         }
 
-        if (x86_quirks->update_apic)
-                x86_quirks->update_apic();
-
         /* Parsed again by __setup for debug/verbose */
         return 0;
 }
@@ -218,8 +216,6 @@ void __init generic_bigsmp_probe(void)
         if (!cmdline_apic && apic == &apic_default) {
                 if (apic_bigsmp.probe()) {
                         apic = &apic_bigsmp;
-                        if (x86_quirks->update_apic)
-                                x86_quirks->update_apic();
                         printk(KERN_INFO "Overriding APIC driver with %s\n",
                                apic->name);
                 }
@@ -240,9 +236,6 @@ void __init generic_apic_probe(void)
                 /* Not visible without early console */
                 if (!apic_probe[i])
                         panic("Didn't find an APIC driver");
-
-                if (x86_quirks->update_apic)
-                        x86_quirks->update_apic();
         }
         printk(KERN_INFO "Using APIC driver %s\n", apic->name);
 }
@@ -262,8 +255,6 @@ generic_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid)
 
                 if (!cmdline_apic) {
                         apic = apic_probe[i];
-                        if (x86_quirks->update_apic)
-                                x86_quirks->update_apic();
                         printk(KERN_INFO "Switched to APIC driver `%s'.\n",
                                apic->name);
                 }
@@ -284,8 +275,6 @@ int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 
                 if (!cmdline_apic) {
                         apic = apic_probe[i];
-                        if (x86_quirks->update_apic)
-                                x86_quirks->update_apic();
                         printk(KERN_INFO "Switched to APIC driver `%s'.\n",
                                apic->name);
                 }
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index e7c163661c77..8d7748efe6a8 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -68,9 +68,6 @@ void __init default_setup_apic_routing(void)
                         apic = &apic_physflat;
                 printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
         }
-
-        if (x86_quirks->update_apic)
-                x86_quirks->update_apic();
 }
 
 /* Same for both flat and physical. */
diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c
index 32838b57a945..aac52fa873ff 100644
--- a/arch/x86/kernel/apic/summit_32.c
+++ b/arch/x86/kernel/apic/summit_32.c
@@ -77,9 +77,9 @@ static void summit_send_IPI_all(int vector)
 extern int use_cyclone;
 
 #ifdef CONFIG_X86_SUMMIT_NUMA
-extern void setup_summit(void);
+static void setup_summit(void);
 #else
-#define setup_summit()  {}
+static inline void setup_summit(void) {}
 #endif
 
 static int summit_mps_oem_check(struct mpc_table *mpc, char *oem,
@@ -291,33 +291,21 @@ static int summit_check_phys_apicid_present(int boot_cpu_physical_apicid)
 
 static unsigned int summit_cpu_mask_to_apicid(const cpumask_t *cpumask)
 {
-        int cpus_found = 0;
-        int num_bits_set;
-        int apicid;
-        int cpu;
+        unsigned int round = 0;
+        int cpu, apicid = 0;
 
-        num_bits_set = cpus_weight(*cpumask);
-        if (num_bits_set >= nr_cpu_ids)
-                return BAD_APICID;
         /*
          * The cpus in the mask must all be on the apic cluster.
          */
-        cpu = first_cpu(*cpumask);
-        apicid = summit_cpu_to_logical_apicid(cpu);
-
-        while (cpus_found < num_bits_set) {
-                if (cpu_isset(cpu, *cpumask)) {
-                        int new_apicid = summit_cpu_to_logical_apicid(cpu);
+        for_each_cpu(cpu, cpumask) {
+                int new_apicid = summit_cpu_to_logical_apicid(cpu);
 
-                        if (APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
-                                printk("%s: Not a valid mask!\n", __func__);
-
-                                return BAD_APICID;
-                        }
-                        apicid = apicid | new_apicid;
-                        cpus_found++;
+                if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
+                        printk("%s: Not a valid mask!\n", __func__);
+                        return BAD_APICID;
                 }
-                cpu++;
+                apicid |= new_apicid;
+                round++;
         }
         return apicid;
 }
@@ -372,15 +360,15 @@ static void summit_vector_allocation_domain(int cpu, cpumask_t *retmask)
 }
 
 #ifdef CONFIG_X86_SUMMIT_NUMA
-static struct rio_table_hdr *rio_table_hdr __initdata;
-static struct scal_detail   *scal_devs[MAX_NUMNODES] __initdata;
-static struct rio_detail    *rio_devs[MAX_NUMNODES*4] __initdata;
+static struct rio_table_hdr *rio_table_hdr;
+static struct scal_detail   *scal_devs[MAX_NUMNODES];
+static struct rio_detail    *rio_devs[MAX_NUMNODES*4];
 
 #ifndef CONFIG_X86_NUMAQ
-static int mp_bus_id_to_node[MAX_MP_BUSSES] __initdata;
+static int mp_bus_id_to_node[MAX_MP_BUSSES];
 #endif
 
-static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
+static int setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
 {
         int twister = 0, node = 0;
         int i, bus, num_buses;
@@ -442,7 +430,7 @@ static int __init setup_pci_node_map_for_wpeg(int wpeg_num, int last_bus)
         return bus;
 }
 
-static int __init build_detail_arrays(void)
+static int build_detail_arrays(void)
 {
         unsigned long ptr;
         int i, scal_detail_size, rio_detail_size;
@@ -476,7 +464,7 @@ static int __init build_detail_arrays(void)
         return 1;
 }
 
-void __init setup_summit(void)
+void setup_summit(void)
 {
         unsigned long           ptr;
         unsigned short          offset;
@@ -574,7 +562,6 @@ struct apic apic_summit = {
         .send_IPI_all                   = summit_send_IPI_all,
         .send_IPI_self                  = default_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
 
diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c
index 354b9c45601d..8fb87b6dd633 100644
--- a/arch/x86/kernel/apic/x2apic_cluster.c
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -224,7 +224,6 @@ struct apic apic_x2apic_cluster = {
         .send_IPI_all                   = x2apic_send_IPI_all,
         .send_IPI_self                  = x2apic_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
         .wait_for_init_deassert         = NULL,
diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c
index 5bcb174409bc..23625b9f98b2 100644
--- a/arch/x86/kernel/apic/x2apic_phys.c
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -213,7 +213,6 @@ struct apic apic_x2apic_phys = {
         .send_IPI_all                   = x2apic_send_IPI_all,
         .send_IPI_self                  = x2apic_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
         .wait_for_init_deassert         = NULL,
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 20b4ad07c3a1..1bd6da1f8fad 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -7,28 +7,28 @@
  *
  * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved.
  */
-
-#include <linux/kernel.h>
-#include <linux/threads.h>
-#include <linux/cpu.h>
 #include <linux/cpumask.h>
+#include <linux/hardirq.h>
+#include <linux/proc_fs.h>
+#include <linux/threads.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/string.h>
 #include <linux/ctype.h>
-#include <linux/init.h>
 #include <linux/sched.h>
-#include <linux/module.h>
-#include <linux/hardirq.h>
 #include <linux/timer.h>
-#include <linux/proc_fs.h>
-#include <asm/current.h>
-#include <asm/smp.h>
-#include <asm/apic.h>
-#include <asm/ipi.h>
-#include <asm/pgtable.h>
-#include <asm/uv/uv.h>
+#include <linux/cpu.h>
+#include <linux/init.h>
+
 #include <asm/uv/uv_mmrs.h>
 #include <asm/uv/uv_hub.h>
+#include <asm/current.h>
+#include <asm/pgtable.h>
 #include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+#include <asm/apic.h>
+#include <asm/ipi.h>
+#include <asm/smp.h>
 
 DEFINE_PER_CPU(int, x2apic_extra_bits);
 
@@ -91,24 +91,28 @@ static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask)
         cpumask_set_cpu(cpu, retmask);
 }
 
-int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
+static int uv_wakeup_secondary(int phys_apicid, unsigned long start_rip)
 {
+#ifdef CONFIG_SMP
         unsigned long val;
         int pnode;
 
         pnode = uv_apicid_to_pnode(phys_apicid);
         val = (1UL << UVH_IPI_INT_SEND_SHFT) |
             (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-            (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+            ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
             APIC_DM_INIT;
         uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
         mdelay(10);
 
         val = (1UL << UVH_IPI_INT_SEND_SHFT) |
             (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
-            (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+            ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
             APIC_DM_STARTUP;
         uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+
+        atomic_set(&init_deasserted, 1);
+#endif
         return 0;
 }
 
@@ -285,7 +289,7 @@ struct apic apic_x2apic_uv_x = {
         .send_IPI_all                   = uv_send_IPI_all,
         .send_IPI_self                  = uv_send_IPI_self,
 
-        .wakeup_cpu                     = NULL,
+        .wakeup_secondary_cpu           = uv_wakeup_secondary,
         .trampoline_phys_low            = DEFAULT_TRAMPOLINE_PHYS_LOW,
         .trampoline_phys_high           = DEFAULT_TRAMPOLINE_PHYS_HIGH,
         .wait_for_init_deassert         = NULL,
@@ -365,7 +369,7 @@ static __init void map_high(char *id, unsigned long base, int shift,
         paddr = base << shift;
         bytes = (1UL << shift) * (max_pnode + 1);
         printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr,
-                                                paddr + bytes);
+                                                paddr + bytes);
         if (map_type == map_uc)
                 init_extra_mapping_uc(paddr, bytes);
         else
@@ -528,7 +532,7 @@ late_initcall(uv_init_heartbeat);
 
 /*
  * Called on each cpu to initialize the per_cpu UV data area.
- *      ZZZ hotplug not supported yet
+ * FIXME: hotplug not supported yet
  */
 void __cpuinit uv_cpu_init(void)
 {
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 4b1c319d30c3..22590cf688ae 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -601,7 +601,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
         if (!data)
                 return -ENOMEM;
 
-        data->acpi_data = percpu_ptr(acpi_perf_data, cpu);
+        data->acpi_data = per_cpu_ptr(acpi_perf_data, cpu);
         per_cpu(drv_data, cpu) = data;
 
         if (cpu_has(c, X86_FEATURE_CONSTANT_TSC))
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index 01b1244ef1c0..d67e0e48bc2d 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -7,11 +7,10 @@
 /*
  *      Get CPU information for use by the procfs.
  */
-#ifdef CONFIG_X86_32
 static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
                               unsigned int cpu)
 {
-#ifdef CONFIG_X86_HT
+#ifdef CONFIG_SMP
         if (c->x86_max_cores * smp_num_siblings > 1) {
                 seq_printf(m, "physical id\t: %d\n", c->phys_proc_id);
                 seq_printf(m, "siblings\t: %d\n",
@@ -24,6 +23,7 @@ static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
 #endif
 }
 
+#ifdef CONFIG_X86_32
 static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c)
 {
         /*
@@ -50,22 +50,6 @@ static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c)
                    c->wp_works_ok ? "yes" : "no");
 }
 #else
-static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c,
-                              unsigned int cpu)
-{
-#ifdef CONFIG_SMP
-        if (c->x86_max_cores * smp_num_siblings > 1) {
-                seq_printf(m, "physical id\t: %d\n", c->phys_proc_id);
-                seq_printf(m, "siblings\t: %d\n",
-                           cpus_weight(per_cpu(cpu_core_map, cpu)));
-                seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id);
-                seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
-                seq_printf(m, "apicid\t\t: %d\n", c->apicid);
-                seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid);
-        }
-#endif
-}
-
 static void show_cpuinfo_misc(struct seq_file *m, struct cpuinfo_x86 *c)
 {
         seq_printf(m,
diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index b205272ad394..1736acc4d7aa 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -469,7 +469,7 @@ void __init efi_enter_virtual_mode(void)
         efi_memory_desc_t *md;
         efi_status_t status;
         unsigned long size;
-        u64 end, systab, addr, npages;
+        u64 end, systab, addr, npages, end_pfn;
         void *p, *va;
 
         efi.systab = NULL;
@@ -481,7 +481,10 @@ void __init efi_enter_virtual_mode(void)
                 size = md->num_pages << EFI_PAGE_SHIFT;
                 end = md->phys_addr + size;
 
-                if (PFN_UP(end) <= max_low_pfn_mapped)
+                end_pfn = PFN_UP(end);
+                if (end_pfn <= max_low_pfn_mapped
+                    || (end_pfn > (1UL << (32 - PAGE_SHIFT))
+                        && end_pfn <= max_pfn_mapped))
                         va = __va(md->phys_addr);
                 else
                         va = efi_ioremap(md->phys_addr, size);
diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c
index a4ee29127fdf..22c3b7828c50 100644
--- a/arch/x86/kernel/efi_64.c
+++ b/arch/x86/kernel/efi_64.c
@@ -100,24 +100,11 @@ void __init efi_call_phys_epilog(void)
 
 void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size)
 {
-        static unsigned pages_mapped __initdata;
-        unsigned i, pages;
-        unsigned long offset;
+        unsigned long last_map_pfn;
 
-        pages = PFN_UP(phys_addr + size) - PFN_DOWN(phys_addr);
-        offset = phys_addr & ~PAGE_MASK;
-        phys_addr &= PAGE_MASK;
-
-        if (pages_mapped + pages > MAX_EFI_IO_PAGES)
+        last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size);
+        if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size)
                 return NULL;
 
-        for (i = 0; i < pages; i++) {
-                __set_fixmap(FIX_EFI_IO_MAP_FIRST_PAGE - pages_mapped,
-                             phys_addr, PAGE_KERNEL);
-                phys_addr += PAGE_SIZE;
-                pages_mapped++;
-        }
-
-        return (void __iomem *)__fix_to_virt(FIX_EFI_IO_MAP_FIRST_PAGE - \
-                                             (pages_mapped - pages)) + offset;
+        return (void __iomem *)__va(phys_addr);
 }
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index b0f61f0dcd0a..f2f8540a7f3d 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -136,7 +136,7 @@ int init_fpu(struct task_struct *tsk)
 #ifdef CONFIG_X86_32
         if (!HAVE_HWFP) {
                 memset(tsk->thread.xstate, 0, xstate_size);
-                finit();
+                finit_task(tsk);
                 set_stopped_child_used_math(tsk);
                 return 0;
         }
diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c
index e41980a373ab..99c4d308f16b 100644
--- a/arch/x86/kernel/ioport.c
+++ b/arch/x86/kernel/ioport.c
@@ -85,19 +85,8 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
 
         t->io_bitmap_max = bytes;
 
-#ifdef CONFIG_X86_32
-        /*
-         * Sets the lazy trigger so that the next I/O operation will
-         * reload the correct bitmap.
-         * Reset the owner so that a process switch will not set
-         * tss->io_bitmap_base to IO_BITMAP_OFFSET.
-         */
-        tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY;
-        tss->io_bitmap_owner = NULL;
-#else
         /* Update the TSS: */
         memcpy(tss->io_bitmap, t->io_bitmap_ptr, bytes_updated);
-#endif
 
         put_cpu();
 
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 9dc6b2b24275..3b09634a5153 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -16,6 +16,7 @@
 #include <linux/cpu.h>
 #include <linux/delay.h>
 #include <linux/uaccess.h>
+#include <linux/percpu.h>
 
 #include <asm/apic.h>
 
@@ -55,13 +56,13 @@ static inline void print_stack_overflow(void) { }
 union irq_ctx {
         struct thread_info      tinfo;
         u32                     stack[THREAD_SIZE/sizeof(u32)];
-};
+} __attribute__((aligned(PAGE_SIZE)));
 
-static union irq_ctx *hardirq_ctx[NR_CPUS] __read_mostly;
-static union irq_ctx *softirq_ctx[NR_CPUS] __read_mostly;
+static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx);
+static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx);
 
-static char softirq_stack[NR_CPUS * THREAD_SIZE] __page_aligned_bss;
-static char hardirq_stack[NR_CPUS * THREAD_SIZE] __page_aligned_bss;
+static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, hardirq_stack);
+static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, softirq_stack);
 
 static void call_on_stack(void *func, void *stack)
 {
@@ -81,7 +82,7 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
         u32 *isp, arg1, arg2;
 
         curctx = (union irq_ctx *) current_thread_info();
-        irqctx = hardirq_ctx[smp_processor_id()];
+        irqctx = __get_cpu_var(hardirq_ctx);
 
         /*
          * this is where we switch to the IRQ stack. However, if we are
@@ -125,34 +126,34 @@ void __cpuinit irq_ctx_init(int cpu)
 {
         union irq_ctx *irqctx;
 
-        if (hardirq_ctx[cpu])
+        if (per_cpu(hardirq_ctx, cpu))
                 return;
 
-        irqctx = (union irq_ctx*) &hardirq_stack[cpu*THREAD_SIZE];
+        irqctx = &per_cpu(hardirq_stack, cpu);
         irqctx->tinfo.task              = NULL;
         irqctx->tinfo.exec_domain       = NULL;
         irqctx->tinfo.cpu               = cpu;
         irqctx->tinfo.preempt_count     = HARDIRQ_OFFSET;
         irqctx->tinfo.addr_limit        = MAKE_MM_SEG(0);
 
-        hardirq_ctx[cpu] = irqctx;
+        per_cpu(hardirq_ctx, cpu) = irqctx;
 
-        irqctx = (union irq_ctx *) &softirq_stack[cpu*THREAD_SIZE];
+        irqctx = &per_cpu(softirq_stack, cpu);
         irqctx->tinfo.task              = NULL;
         irqctx->tinfo.exec_domain       = NULL;
         irqctx->tinfo.cpu               = cpu;
         irqctx->tinfo.preempt_count     = 0;
         irqctx->tinfo.addr_limit        = MAKE_MM_SEG(0);
 
-        softirq_ctx[cpu] = irqctx;
+        per_cpu(softirq_ctx, cpu) = irqctx;
 
         printk(KERN_DEBUG "CPU %u irqstacks, hard=%p soft=%p\n",
-               cpu, hardirq_ctx[cpu], softirq_ctx[cpu]);
+               cpu, per_cpu(hardirq_ctx, cpu),  per_cpu(softirq_ctx, cpu));
 }
 
 void irq_ctx_exit(int cpu)
 {
-        hardirq_ctx[cpu] = NULL;
+        per_cpu(hardirq_ctx, cpu) = NULL;
 }
 
 asmlinkage void do_softirq(void)
@@ -169,7 +170,7 @@ asmlinkage void do_softirq(void)
 
         if (local_softirq_pending()) {
                 curctx = current_thread_info();
-                irqctx = softirq_ctx[smp_processor_id()];
+                irqctx = __get_cpu_var(softirq_ctx);
                 irqctx->tinfo.task = curctx->task;
                 irqctx->tinfo.previous_esp = current_stack_pointer;
 
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 87b69d4fac16..6afa5232dbb7 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -1,8 +1,8 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
-#include <asm/idle.h>
 #include <linux/smp.h>
+#include <linux/prctl.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
 #include <linux/module.h>
@@ -11,6 +11,9 @@
 #include <linux/ftrace.h>
 #include <asm/system.h>
 #include <asm/apic.h>
+#include <asm/idle.h>
+#include <asm/uaccess.h>
+#include <asm/i387.h>
 
 unsigned long idle_halt;
 EXPORT_SYMBOL(idle_halt);
@@ -56,6 +59,192 @@ void arch_task_cache_init(void)
 }
 
 /*
+ * Free current thread data structures etc..
+ */
+void exit_thread(void)
+{
+        struct task_struct *me = current;
+        struct thread_struct *t = &me->thread;
+
+        if (me->thread.io_bitmap_ptr) {
+                struct tss_struct *tss = &per_cpu(init_tss, get_cpu());
+
+                kfree(t->io_bitmap_ptr);
+                t->io_bitmap_ptr = NULL;
+                clear_thread_flag(TIF_IO_BITMAP);
+                /*
+                 * Careful, clear this in the TSS too:
+                 */
+                memset(tss->io_bitmap, 0xff, t->io_bitmap_max);
+                t->io_bitmap_max = 0;
+                put_cpu();
+        }
+
+        ds_exit_thread(current);
+}
+
+void flush_thread(void)
+{
+        struct task_struct *tsk = current;
+
+#ifdef CONFIG_X86_64
+        if (test_tsk_thread_flag(tsk, TIF_ABI_PENDING)) {
+                clear_tsk_thread_flag(tsk, TIF_ABI_PENDING);
+                if (test_tsk_thread_flag(tsk, TIF_IA32)) {
+                        clear_tsk_thread_flag(tsk, TIF_IA32);
+                } else {
+                        set_tsk_thread_flag(tsk, TIF_IA32);
+                        current_thread_info()->status |= TS_COMPAT;
+                }
+        }
+#endif
+
+        clear_tsk_thread_flag(tsk, TIF_DEBUG);
+
+        tsk->thread.debugreg0 = 0;
+        tsk->thread.debugreg1 = 0;
+        tsk->thread.debugreg2 = 0;
+        tsk->thread.debugreg3 = 0;
+        tsk->thread.debugreg6 = 0;
+        tsk->thread.debugreg7 = 0;
+        memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
+        /*
+         * Forget coprocessor state..
+         */
+        tsk->fpu_counter = 0;
+        clear_fpu(tsk);
+        clear_used_math();
+}
+
+static void hard_disable_TSC(void)
+{
+        write_cr4(read_cr4() | X86_CR4_TSD);
+}
+
+void disable_TSC(void)
+{
+        preempt_disable();
+        if (!test_and_set_thread_flag(TIF_NOTSC))
+                /*
+                 * Must flip the CPU state synchronously with
+                 * TIF_NOTSC in the current running context.
+                 */
+                hard_disable_TSC();
+        preempt_enable();
+}
+
+static void hard_enable_TSC(void)
+{
+        write_cr4(read_cr4() & ~X86_CR4_TSD);
+}
+
+static void enable_TSC(void)
+{
+        preempt_disable();
+        if (test_and_clear_thread_flag(TIF_NOTSC))
+                /*
+                 * Must flip the CPU state synchronously with
+                 * TIF_NOTSC in the current running context.
+                 */
+                hard_enable_TSC();
+        preempt_enable();
+}
+
+int get_tsc_mode(unsigned long adr)
+{
+        unsigned int val;
+
+        if (test_thread_flag(TIF_NOTSC))
+                val = PR_TSC_SIGSEGV;
+        else
+                val = PR_TSC_ENABLE;
+
+        return put_user(val, (unsigned int __user *)adr);
+}
+
+int set_tsc_mode(unsigned int val)
+{
+        if (val == PR_TSC_SIGSEGV)
+                disable_TSC();
+        else if (val == PR_TSC_ENABLE)
+                enable_TSC();
+        else
+                return -EINVAL;
+
+        return 0;
+}
+
+void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
+                      struct tss_struct *tss)
+{
+        struct thread_struct *prev, *next;
+
+        prev = &prev_p->thread;
+        next = &next_p->thread;
+
+        if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) ||
+            test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR))
+                ds_switch_to(prev_p, next_p);
+        else if (next->debugctlmsr != prev->debugctlmsr)
+                update_debugctlmsr(next->debugctlmsr);
+
+        if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
+                set_debugreg(next->debugreg0, 0);
+                set_debugreg(next->debugreg1, 1);
+                set_debugreg(next->debugreg2, 2);
+                set_debugreg(next->debugreg3, 3);
+                /* no 4 and 5 */
+                set_debugreg(next->debugreg6, 6);
+                set_debugreg(next->debugreg7, 7);
+        }
+
+        if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
+            test_tsk_thread_flag(next_p, TIF_NOTSC)) {
+                /* prev and next are different */
+                if (test_tsk_thread_flag(next_p, TIF_NOTSC))
+                        hard_disable_TSC();
+                else
+                        hard_enable_TSC();
+        }
+
+        if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
+                /*
+                 * Copy the relevant range of the IO bitmap.
+                 * Normally this is 128 bytes or less:
+                 */
+                memcpy(tss->io_bitmap, next->io_bitmap_ptr,
+                       max(prev->io_bitmap_max, next->io_bitmap_max));
+        } else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) {
+                /*
+                 * Clear any possible leftover bits:
+                 */
+                memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
+        }
+}
+
+int sys_fork(struct pt_regs *regs)
+{
+        return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
+}
+
+/*
+ * This is trivial, and on the face of it looks like it
+ * could equally well be done in user mode.
+ *
+ * Not so, for quite unobvious reasons - register pressure.
+ * In user mode vfork() cannot have a stack frame, and if
+ * done by calling the "clone()" system call directly, you
+ * do not have enough call-clobbered registers to hold all
+ * the information you need.
+ */
+int sys_vfork(struct pt_regs *regs)
+{
+        return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0,
+                       NULL, NULL);
+}
+
+
+/*
  * Idle related variables and functions
  */
 unsigned long boot_option_idle_override = 0;
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 646da41a620a..14014d766cad 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -230,55 +230,6 @@ int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
 }
 EXPORT_SYMBOL(kernel_thread);
 
-/*
- * Free current thread data structures etc..
- */
-void exit_thread(void)
-{
-        /* The process may have allocated an io port bitmap... nuke it. */
-        if (unlikely(test_thread_flag(TIF_IO_BITMAP))) {
-                struct task_struct *tsk = current;
-                struct thread_struct *t = &tsk->thread;
-                int cpu = get_cpu();
-                struct tss_struct *tss = &per_cpu(init_tss, cpu);
-
-                kfree(t->io_bitmap_ptr);
-                t->io_bitmap_ptr = NULL;
-                clear_thread_flag(TIF_IO_BITMAP);
-                /*
-                 * Careful, clear this in the TSS too:
-                 */
-                memset(tss->io_bitmap, 0xff, tss->io_bitmap_max);
-                t->io_bitmap_max = 0;
-                tss->io_bitmap_owner = NULL;
-                tss->io_bitmap_max = 0;
-                tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET;
-                put_cpu();
-        }
-
-        ds_exit_thread(current);
-}
-
-void flush_thread(void)
-{
-        struct task_struct *tsk = current;
-
-        tsk->thread.debugreg0 = 0;
-        tsk->thread.debugreg1 = 0;
-        tsk->thread.debugreg2 = 0;
-        tsk->thread.debugreg3 = 0;
-        tsk->thread.debugreg6 = 0;
-        tsk->thread.debugreg7 = 0;
-        memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
-        clear_tsk_thread_flag(tsk, TIF_DEBUG);
-        /*
-         * Forget coprocessor state..
-         */
-        tsk->fpu_counter = 0;
-        clear_fpu(tsk);
-        clear_used_math();
-}
-
 void release_thread(struct task_struct *dead_task)
 {
         BUG_ON(dead_task->mm);
@@ -366,127 +317,6 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
 }
 EXPORT_SYMBOL_GPL(start_thread);
 
-static void hard_disable_TSC(void)
-{
-        write_cr4(read_cr4() | X86_CR4_TSD);
-}
-
-void disable_TSC(void)
-{
-        preempt_disable();
-        if (!test_and_set_thread_flag(TIF_NOTSC))
-                /*
-                 * Must flip the CPU state synchronously with
-                 * TIF_NOTSC in the current running context.
-                 */
-                hard_disable_TSC();
-        preempt_enable();
-}
-
-static void hard_enable_TSC(void)
-{
-        write_cr4(read_cr4() & ~X86_CR4_TSD);
-}
-
-static void enable_TSC(void)
-{
-        preempt_disable();
-        if (test_and_clear_thread_flag(TIF_NOTSC))
-                /*
-                 * Must flip the CPU state synchronously with
-                 * TIF_NOTSC in the current running context.
-                 */
-                hard_enable_TSC();
-        preempt_enable();
-}
-
-int get_tsc_mode(unsigned long adr)
-{
-        unsigned int val;
-
-        if (test_thread_flag(TIF_NOTSC))
-                val = PR_TSC_SIGSEGV;
-        else
-                val = PR_TSC_ENABLE;
-
-        return put_user(val, (unsigned int __user *)adr);
-}
-
-int set_tsc_mode(unsigned int val)
-{
-        if (val == PR_TSC_SIGSEGV)
-                disable_TSC();
-        else if (val == PR_TSC_ENABLE)
-                enable_TSC();
-        else
-                return -EINVAL;
-
-        return 0;
-}
-
-static noinline void
-__switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
-                 struct tss_struct *tss)
-{
-        struct thread_struct *prev, *next;
-
-        prev = &prev_p->thread;
-        next = &next_p->thread;
-
-        if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) ||
-            test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR))
-                ds_switch_to(prev_p, next_p);
-        else if (next->debugctlmsr != prev->debugctlmsr)
-                update_debugctlmsr(next->debugctlmsr);
-
-        if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
-                set_debugreg(next->debugreg0, 0);
-                set_debugreg(next->debugreg1, 1);
-                set_debugreg(next->debugreg2, 2);
-                set_debugreg(next->debugreg3, 3);
-                /* no 4 and 5 */
-                set_debugreg(next->debugreg6, 6);
-                set_debugreg(next->debugreg7, 7);
-        }
-
-        if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
-            test_tsk_thread_flag(next_p, TIF_NOTSC)) {
-                /* prev and next are different */
-                if (test_tsk_thread_flag(next_p, TIF_NOTSC))
-                        hard_disable_TSC();
-                else
-                        hard_enable_TSC();
-        }
-
-        if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
-                /*
-                 * Disable the bitmap via an invalid offset. We still cache
-                 * the previous bitmap owner and the IO bitmap contents:
-                 */
-                tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET;
-                return;
-        }
-
-        if (likely(next == tss->io_bitmap_owner)) {
-                /*
-                 * Previous owner of the bitmap (hence the bitmap content)
-                 * matches the next task, we dont have to do anything but
-                 * to set a valid offset in the TSS:
-                 */
-                tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET;
-                return;
-        }
-        /*
-         * Lazy TSS's I/O bitmap copy. We set an invalid offset here
-         * and we let the task to get a GPF in case an I/O instruction
-         * is performed.  The handler of the GPF will verify that the
-         * faulting task has a valid I/O bitmap and, it true, does the
-         * real copy and restart the instruction.  This will save us
-         * redundant copies when the currently switched task does not
-         * perform any I/O during its timeslice.
-         */
-        tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY;
-}
 
 /*
  *      switch_to(x,yn) should switch tasks from x to y.
@@ -600,11 +430,6 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
         return prev_p;
 }
 
-int sys_fork(struct pt_regs *regs)
-{
-        return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
-}
-
 int sys_clone(struct pt_regs *regs)
 {
         unsigned long clone_flags;
@@ -621,21 +446,6 @@ int sys_clone(struct pt_regs *regs)
 }
 
 /*
- * This is trivial, and on the face of it looks like it
- * could equally well be done in user mode.
- *
- * Not so, for quite unobvious reasons - register pressure.
- * In user mode vfork() cannot have a stack frame, and if
- * done by calling the "clone()" system call directly, you
- * do not have enough call-clobbered registers to hold all
- * the information you need.
- */
-int sys_vfork(struct pt_regs *regs)
-{
-        return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0, NULL, NULL);
-}
-
-/*
  * sys_execve() executes a new program.
  */
 int sys_execve(struct pt_regs *regs)
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 836ef6575f01..abb7e6a7f0c6 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -237,61 +237,6 @@ void show_regs(struct pt_regs *regs)
         show_trace(NULL, regs, (void *)(regs + 1), regs->bp);
 }
 
-/*
- * Free current thread data structures etc..
- */
-void exit_thread(void)
-{
-        struct task_struct *me = current;
-        struct thread_struct *t = &me->thread;
-
-        if (me->thread.io_bitmap_ptr) {
-                struct tss_struct *tss = &per_cpu(init_tss, get_cpu());
-
-                kfree(t->io_bitmap_ptr);
-                t->io_bitmap_ptr = NULL;
-                clear_thread_flag(TIF_IO_BITMAP);
-                /*
-                 * Careful, clear this in the TSS too:
-                 */
-                memset(tss->io_bitmap, 0xff, t->io_bitmap_max);
-                t->io_bitmap_max = 0;
-                put_cpu();
-        }
-
-        ds_exit_thread(current);
-}
-
-void flush_thread(void)
-{
-        struct task_struct *tsk = current;
-
-        if (test_tsk_thread_flag(tsk, TIF_ABI_PENDING)) {
-                clear_tsk_thread_flag(tsk, TIF_ABI_PENDING);
-                if (test_tsk_thread_flag(tsk, TIF_IA32)) {
-                        clear_tsk_thread_flag(tsk, TIF_IA32);
-                } else {
-                        set_tsk_thread_flag(tsk, TIF_IA32);
-                        current_thread_info()->status |= TS_COMPAT;
-                }
-        }
-        clear_tsk_thread_flag(tsk, TIF_DEBUG);
-
-        tsk->thread.debugreg0 = 0;
-        tsk->thread.debugreg1 = 0;
-        tsk->thread.debugreg2 = 0;
-        tsk->thread.debugreg3 = 0;
-        tsk->thread.debugreg6 = 0;
-        tsk->thread.debugreg7 = 0;
-        memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
-        /*
-         * Forget coprocessor state..
-         */
-        tsk->fpu_counter = 0;
-        clear_fpu(tsk);
-        clear_used_math();
-}
-
 void release_thread(struct task_struct *dead_task)
 {
         if (dead_task->mm) {
@@ -425,118 +370,6 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
 }
 EXPORT_SYMBOL_GPL(start_thread);
 
-static void hard_disable_TSC(void)
-{
-        write_cr4(read_cr4() | X86_CR4_TSD);
-}
-
-void disable_TSC(void)
-{
-        preempt_disable();
-        if (!test_and_set_thread_flag(TIF_NOTSC))
-                /*
-                 * Must flip the CPU state synchronously with
-                 * TIF_NOTSC in the current running context.
-                 */
-                hard_disable_TSC();
-        preempt_enable();
-}
-
-static void hard_enable_TSC(void)
-{
-        write_cr4(read_cr4() & ~X86_CR4_TSD);
-}
-
-static void enable_TSC(void)
-{
-        preempt_disable();
-        if (test_and_clear_thread_flag(TIF_NOTSC))
-                /*
-                 * Must flip the CPU state synchronously with
-                 * TIF_NOTSC in the current running context.
-                 */
-                hard_enable_TSC();
-        preempt_enable();
-}
-
-int get_tsc_mode(unsigned long adr)
-{
-        unsigned int val;
-
-        if (test_thread_flag(TIF_NOTSC))
-                val = PR_TSC_SIGSEGV;
-        else
-                val = PR_TSC_ENABLE;
-
-        return put_user(val, (unsigned int __user *)adr);
-}
-
-int set_tsc_mode(unsigned int val)
-{
-        if (val == PR_TSC_SIGSEGV)
-                disable_TSC();
-        else if (val == PR_TSC_ENABLE)
-                enable_TSC();
-        else
-                return -EINVAL;
-
-        return 0;
-}
-
-/*
- * This special macro can be used to load a debugging register
- */
-#define loaddebug(thread, r) set_debugreg(thread->debugreg ## r, r)
-
-static inline void __switch_to_xtra(struct task_struct *prev_p,
-                                    struct task_struct *next_p,
-                                    struct tss_struct *tss)
-{
-        struct thread_struct *prev, *next;
-
-        prev = &prev_p->thread,
-        next = &next_p->thread;
-
-        if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) ||
-            test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR))
-                ds_switch_to(prev_p, next_p);
-        else if (next->debugctlmsr != prev->debugctlmsr)
-                update_debugctlmsr(next->debugctlmsr);
-
-        if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
-                loaddebug(next, 0);
-                loaddebug(next, 1);
-                loaddebug(next, 2);
-                loaddebug(next, 3);
-                /* no 4 and 5 */
-                loaddebug(next, 6);
-                loaddebug(next, 7);
-        }
-
-        if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
-            test_tsk_thread_flag(next_p, TIF_NOTSC)) {
-                /* prev and next are different */
-                if (test_tsk_thread_flag(next_p, TIF_NOTSC))
-                        hard_disable_TSC();
-                else
-                        hard_enable_TSC();
-        }
-
-        if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
-                /*
-                 * Copy the relevant range of the IO bitmap.
-                 * Normally this is 128 bytes or less:
-                 */
-                memcpy(tss->io_bitmap, next->io_bitmap_ptr,
-                       max(prev->io_bitmap_max, next->io_bitmap_max));
-        } else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) {
-                /*
-                 * Clear any possible leftover bits:
-                 */
-                memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
-        }
-}
-
 /*
  *      switch_to(x,y) should switch tasks from x to y.
  *
@@ -694,11 +527,6 @@ void set_personality_64bit(void)
         current->personality &= ~READ_IMPLIES_EXEC;
 }
 
-asmlinkage long sys_fork(struct pt_regs *regs)
-{
-        return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
-}
-
 asmlinkage long
 sys_clone(unsigned long clone_flags, unsigned long newsp,
           void __user *parent_tid, void __user *child_tid, struct pt_regs *regs)
@@ -708,22 +536,6 @@ sys_clone(unsigned long clone_flags, unsigned long newsp,
         return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
 }
 
-/*
- * This is trivial, and on the face of it looks like it
- * could equally well be done in user mode.
- *
- * Not so, for quite unobvious reasons - register pressure.
- * In user mode vfork() cannot have a stack frame, and if
- * done by calling the "clone()" system call directly, you
- * do not have enough call-clobbered registers to hold all
- * the information you need.
- */
-asmlinkage long sys_vfork(struct pt_regs *regs)
-{
-        return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0,
-                    NULL, NULL);
-}
-
 unsigned long get_wchan(struct task_struct *p)
 {
         unsigned long stack;
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index fb2159a5c817..3d9672e59c16 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -1383,7 +1383,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs,
 #ifdef CONFIG_X86_32
 # define IS_IA32        1
 #elif defined CONFIG_IA32_EMULATION
-# define IS_IA32        test_thread_flag(TIF_IA32)
+# define IS_IA32        is_compat_task()
 #else
 # define IS_IA32        0
 #endif
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 1cc18d439bbb..2aef36d8aca2 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -216,6 +216,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = {
                         DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"),
                 },
         },
+        {       /* Handle problems with rebooting on Dell XPS710 */
+                .callback = set_bios_reboot,
+                .ident = "Dell XPS710",
+                .matches = {
+                        DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+                        DMI_MATCH(DMI_PRODUCT_NAME, "Dell XPS710"),
+                },
+        },
         { }
 };
 
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 5b85759e7972..b746deb9ebc6 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -600,19 +600,7 @@ static int __init setup_elfcorehdr(char *arg)
 early_param("elfcorehdr", setup_elfcorehdr);
 #endif
 
-static int __init default_update_apic(void)
-{
-#ifdef CONFIG_SMP
-        if (!apic->wakeup_cpu)
-                apic->wakeup_cpu = wakeup_secondary_cpu_via_init;
-#endif
-
-        return 0;
-}
-
-static struct x86_quirks default_x86_quirks __initdata = {
-        .update_apic         = default_update_apic,
-};
+static struct x86_quirks default_x86_quirks __initdata;
 
 struct x86_quirks *x86_quirks __initdata = &default_x86_quirks;
 
@@ -782,6 +770,9 @@ void __init setup_arch(char **cmdline_p)
 
         finish_e820_parsing();
 
+        if (efi_enabled)
+                efi_init();
+
         dmi_scan_machine();
 
         dmi_check_system(bad_bios_dmi_table);
@@ -801,8 +792,6 @@ void __init setup_arch(char **cmdline_p)
         insert_resource(&iomem_resource, &data_resource);
         insert_resource(&iomem_resource, &bss_resource);
 
-        if (efi_enabled)
-                efi_init();
 
 #ifdef CONFIG_X86_32
         if (ppro_with_ram_bug()) {
@@ -875,9 +864,7 @@ void __init setup_arch(char **cmdline_p)
 
         reserve_initrd();
 
-#ifdef CONFIG_X86_64
         vsmp_init();
-#endif
 
         io_delay_init();
 
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index d992e6cff730..c29f301d3885 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -7,6 +7,7 @@
 #include <linux/crash_dump.h>
 #include <linux/smp.h>
 #include <linux/topology.h>
+#include <linux/pfn.h>
 #include <asm/sections.h>
 #include <asm/processor.h>
 #include <asm/setup.h>
@@ -41,6 +42,321 @@ unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = {
 };
 EXPORT_SYMBOL(__per_cpu_offset);
 
+/**
+ * pcpu_need_numa - determine percpu allocation needs to consider NUMA
+ *
+ * If NUMA is not configured or there is only one NUMA node available,
+ * there is no reason to consider NUMA.  This function determines
+ * whether percpu allocation should consider NUMA or not.
+ *
+ * RETURNS:
+ * true if NUMA should be considered; otherwise, false.
+ */
+static bool __init pcpu_need_numa(void)
+{
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+        pg_data_t *last = NULL;
+        unsigned int cpu;
+
+        for_each_possible_cpu(cpu) {
+                int node = early_cpu_to_node(cpu);
+
+                if (node_online(node) && NODE_DATA(node) &&
+                    last && last != NODE_DATA(node))
+                        return true;
+
+                last = NODE_DATA(node);
+        }
+#endif
+        return false;
+}
+
+/**
+ * pcpu_alloc_bootmem - NUMA friendly alloc_bootmem wrapper for percpu
+ * @cpu: cpu to allocate for
+ * @size: size allocation in bytes
+ * @align: alignment
+ *
+ * Allocate @size bytes aligned at @align for cpu @cpu.  This wrapper
+ * does the right thing for NUMA regardless of the current
+ * configuration.
+ *
+ * RETURNS:
+ * Pointer to the allocated area on success, NULL on failure.
+ */
+static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size,
+                                        unsigned long align)
+{
+        const unsigned long goal = __pa(MAX_DMA_ADDRESS);
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+        int node = early_cpu_to_node(cpu);
+        void *ptr;
+
+        if (!node_online(node) || !NODE_DATA(node)) {
+                ptr = __alloc_bootmem_nopanic(size, align, goal);
+                pr_info("cpu %d has no node %d or node-local memory\n",
+                        cpu, node);
+                pr_debug("per cpu data for cpu%d %lu bytes at %016lx\n",
+                         cpu, size, __pa(ptr));
+        } else {
+                ptr = __alloc_bootmem_node_nopanic(NODE_DATA(node),
+                                                   size, align, goal);
+                pr_debug("per cpu data for cpu%d %lu bytes on node%d at "
+                         "%016lx\n", cpu, size, node, __pa(ptr));
+        }
+        return ptr;
+#else
+        return __alloc_bootmem_nopanic(size, align, goal);
+#endif
+}
+
+/*
+ * Remap allocator
+ *
+ * This allocator uses PMD page as unit.  A PMD page is allocated for
+ * each cpu and each is remapped into vmalloc area using PMD mapping.
+ * As PMD page is quite large, only part of it is used for the first
+ * chunk.  Unused part is returned to the bootmem allocator.
+ *
+ * So, the PMD pages are mapped twice - once to the physical mapping
+ * and to the vmalloc area for the first percpu chunk.  The double
+ * mapping does add one more PMD TLB entry pressure but still is much
+ * better than only using 4k mappings while still being NUMA friendly.
+ */
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+static size_t pcpur_size __initdata;
+static void **pcpur_ptrs __initdata;
+
+static struct page * __init pcpur_get_page(unsigned int cpu, int pageno)
+{
+        size_t off = (size_t)pageno << PAGE_SHIFT;
+
+        if (off >= pcpur_size)
+                return NULL;
+
+        return virt_to_page(pcpur_ptrs[cpu] + off);
+}
+
+static ssize_t __init setup_pcpu_remap(size_t static_size)
+{
+        static struct vm_struct vm;
+        pg_data_t *last;
+        size_t ptrs_size;
+        unsigned int cpu;
+        ssize_t ret;
+
+        /*
+         * If large page isn't supported, there's no benefit in doing
+         * this.  Also, on non-NUMA, embedding is better.
+         */
+        if (!cpu_has_pse || pcpu_need_numa())
+                return -EINVAL;
+
+        last = NULL;
+        for_each_possible_cpu(cpu) {
+                int node = early_cpu_to_node(cpu);
+
+                if (node_online(node) && NODE_DATA(node) &&
+                    last && last != NODE_DATA(node))
+                        goto proceed;
+
+                last = NODE_DATA(node);
+        }
+        return -EINVAL;
+
+proceed:
+        /*
+         * Currently supports only single page.  Supporting multiple
+         * pages won't be too difficult if it ever becomes necessary.
+         */
+        pcpur_size = PFN_ALIGN(static_size + PERCPU_DYNAMIC_RESERVE);
+        if (pcpur_size > PMD_SIZE) {
+                pr_warning("PERCPU: static data is larger than large page, "
+                           "can't use large page\n");
+                return -EINVAL;
+        }
+
+        /* allocate pointer array and alloc large pages */
+        ptrs_size = PFN_ALIGN(num_possible_cpus() * sizeof(pcpur_ptrs[0]));
+        pcpur_ptrs = alloc_bootmem(ptrs_size);
+
+        for_each_possible_cpu(cpu) {
+                pcpur_ptrs[cpu] = pcpu_alloc_bootmem(cpu, PMD_SIZE, PMD_SIZE);
+                if (!pcpur_ptrs[cpu])
+                        goto enomem;
+
+                /*
+                 * Only use pcpur_size bytes and give back the rest.
+                 *
+                 * Ingo: The 2MB up-rounding bootmem is needed to make
+                 * sure the partial 2MB page is still fully RAM - it's
+                 * not well-specified to have a PAT-incompatible area
+                 * (unmapped RAM, device memory, etc.) in that hole.
+                 */
+                free_bootmem(__pa(pcpur_ptrs[cpu] + pcpur_size),
+                             PMD_SIZE - pcpur_size);
+
+                memcpy(pcpur_ptrs[cpu], __per_cpu_load, static_size);
+        }
+
+        /* allocate address and map */
+        vm.flags = VM_ALLOC;
+        vm.size = num_possible_cpus() * PMD_SIZE;
+        vm_area_register_early(&vm, PMD_SIZE);
+
+        for_each_possible_cpu(cpu) {
+                pmd_t *pmd;
+
+                pmd = populate_extra_pmd((unsigned long)vm.addr
+                                         + cpu * PMD_SIZE);
+                set_pmd(pmd, pfn_pmd(page_to_pfn(virt_to_page(pcpur_ptrs[cpu])),
+                                     PAGE_KERNEL_LARGE));
+        }
+
+        /* we're ready, commit */
+        pr_info("PERCPU: Remapped at %p with large pages, static data "
+                "%zu bytes\n", vm.addr, static_size);
+
+        ret = pcpu_setup_first_chunk(pcpur_get_page, static_size, PMD_SIZE,
+                                     pcpur_size - static_size, vm.addr, NULL);
+        goto out_free_ar;
+
+enomem:
+        for_each_possible_cpu(cpu)
+                if (pcpur_ptrs[cpu])
+                        free_bootmem(__pa(pcpur_ptrs[cpu]), PMD_SIZE);
+        ret = -ENOMEM;
+out_free_ar:
+        free_bootmem(__pa(pcpur_ptrs), ptrs_size);
+        return ret;
+}
+#else
+static ssize_t __init setup_pcpu_remap(size_t static_size)
+{
+        return -EINVAL;
+}
+#endif
+
+/*
+ * Embedding allocator
+ *
+ * The first chunk is sized to just contain the static area plus
+ * PERCPU_DYNAMIC_RESERVE and allocated as a contiguous area using
+ * bootmem allocator and used as-is without being mapped into vmalloc
+ * area.  This enables the first chunk to piggy back on the linear
+ * physical PMD mapping and doesn't add any additional pressure to
+ * TLB.
+ */
+static void *pcpue_ptr __initdata;
+static size_t pcpue_unit_size __initdata;
+
+static struct page * __init pcpue_get_page(unsigned int cpu, int pageno)
+{
+        return virt_to_page(pcpue_ptr + cpu * pcpue_unit_size
+                            + ((size_t)pageno << PAGE_SHIFT));
+}
+
+static ssize_t __init setup_pcpu_embed(size_t static_size)
+{
+        unsigned int cpu;
+
+        /*
+         * If large page isn't supported, there's no benefit in doing
+         * this.  Also, embedding allocation doesn't play well with
+         * NUMA.
+         */
+        if (!cpu_has_pse || pcpu_need_numa())
+                return -EINVAL;
+
+        /* allocate and copy */
+        pcpue_unit_size = PFN_ALIGN(static_size + PERCPU_DYNAMIC_RESERVE);
+        pcpue_unit_size = max_t(size_t, pcpue_unit_size, PCPU_MIN_UNIT_SIZE);
+        pcpue_ptr = pcpu_alloc_bootmem(0, num_possible_cpus() * pcpue_unit_size,
+                                       PAGE_SIZE);
+        if (!pcpue_ptr)
+                return -ENOMEM;
+
+        for_each_possible_cpu(cpu)
+                memcpy(pcpue_ptr + cpu * pcpue_unit_size, __per_cpu_load,
+                       static_size);
+
+        /* we're ready, commit */
+        pr_info("PERCPU: Embedded %zu pages at %p, static data %zu bytes\n",
+                pcpue_unit_size >> PAGE_SHIFT, pcpue_ptr, static_size);
+
+        return pcpu_setup_first_chunk(pcpue_get_page, static_size,
+                                      pcpue_unit_size,
+                                      pcpue_unit_size - static_size, pcpue_ptr,
+                                      NULL);
+}
+
+/*
+ * 4k page allocator
+ *
+ * This is the basic allocator.  Static percpu area is allocated
+ * page-by-page and most of initialization is done by the generic
+ * setup function.
+ */
+static struct page **pcpu4k_pages __initdata;
+static int pcpu4k_nr_static_pages __initdata;
+
+static struct page * __init pcpu4k_get_page(unsigned int cpu, int pageno)
+{
+        if (pageno < pcpu4k_nr_static_pages)
+                return pcpu4k_pages[cpu * pcpu4k_nr_static_pages + pageno];
+        return NULL;
+}
+
+static void __init pcpu4k_populate_pte(unsigned long addr)
+{
+        populate_extra_pte(addr);
+}
+
+static ssize_t __init setup_pcpu_4k(size_t static_size)
+{
+        size_t pages_size;
+        unsigned int cpu;
+        int i, j;
+        ssize_t ret;
+
+        pcpu4k_nr_static_pages = PFN_UP(static_size);
+
+        /* unaligned allocations can't be freed, round up to page size */
+        pages_size = PFN_ALIGN(pcpu4k_nr_static_pages * num_possible_cpus()
+                               * sizeof(pcpu4k_pages[0]));
+        pcpu4k_pages = alloc_bootmem(pages_size);
+
+        /* allocate and copy */
+        j = 0;
+        for_each_possible_cpu(cpu)
+                for (i = 0; i < pcpu4k_nr_static_pages; i++) {
+                        void *ptr;
+
+                        ptr = pcpu_alloc_bootmem(cpu, PAGE_SIZE, PAGE_SIZE);
+                        if (!ptr)
+                                goto enomem;
+
+                        memcpy(ptr, __per_cpu_load + i * PAGE_SIZE, PAGE_SIZE);
+                        pcpu4k_pages[j++] = virt_to_page(ptr);
+                }
+
+        /* we're ready, commit */
+        pr_info("PERCPU: Allocated %d 4k pages, static data %zu bytes\n",
+                pcpu4k_nr_static_pages, static_size);
+
+        ret = pcpu_setup_first_chunk(pcpu4k_get_page, static_size, 0, 0, NULL,
+                                     pcpu4k_populate_pte);
+        goto out_free_ar;
+
+enomem:
+        while (--j >= 0)
+                free_bootmem(__pa(page_address(pcpu4k_pages[j])), PAGE_SIZE);
+        ret = -ENOMEM;
+out_free_ar:
+        free_bootmem(__pa(pcpu4k_pages), pages_size);
+        return ret;
+}
+
 static inline void setup_percpu_segment(int cpu)
 {
 #ifdef CONFIG_X86_32
@@ -61,38 +377,35 @@ static inline void setup_percpu_segment(int cpu)
  */
 void __init setup_per_cpu_areas(void)
 {
-        ssize_t size;
-        char *ptr;
-        int cpu;
-
-        /* Copy section for each CPU (we discard the original) */
-        size = roundup(PERCPU_ENOUGH_ROOM, PAGE_SIZE);
+        size_t static_size = __per_cpu_end - __per_cpu_start;
+        unsigned int cpu;
+        unsigned long delta;
+        size_t pcpu_unit_size;
+        ssize_t ret;
 
         pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n",
                 NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids);
 
-        pr_info("PERCPU: Allocating %zd bytes of per cpu data\n", size);
+        /*
+         * Allocate percpu area.  If PSE is supported, try to make use
+         * of large page mappings.  Please read comments on top of
+         * each allocator for details.
+         */
+        ret = setup_pcpu_remap(static_size);
+        if (ret < 0)
+                ret = setup_pcpu_embed(static_size);
+        if (ret < 0)
+                ret = setup_pcpu_4k(static_size);
+        if (ret < 0)
+                panic("cannot allocate static percpu area (%zu bytes, err=%zd)",
+                      static_size, ret);
 
-        for_each_possible_cpu(cpu) {
-#ifndef CONFIG_NEED_MULTIPLE_NODES
-                ptr = alloc_bootmem_pages(size);
-#else
-                int node = early_cpu_to_node(cpu);
-                if (!node_online(node) || !NODE_DATA(node)) {
-                        ptr = alloc_bootmem_pages(size);
-                        pr_info("cpu %d has no node %d or node-local memory\n",
-                                cpu, node);
-                        pr_debug("per cpu data for cpu%d at %016lx\n",
-                                 cpu, __pa(ptr));
-                } else {
-                        ptr = alloc_bootmem_pages_node(NODE_DATA(node), size);
-                        pr_debug("per cpu data for cpu%d on node%d at %016lx\n",
-                                cpu, node, __pa(ptr));
-                }
-#endif
+        pcpu_unit_size = ret;
 
-                memcpy(ptr, __per_cpu_load, __per_cpu_end - __per_cpu_start);
-                per_cpu_offset(cpu) = ptr - __per_cpu_start;
+        /* alrighty, percpu areas up and running */
+        delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
+        for_each_possible_cpu(cpu) {
+                per_cpu_offset(cpu) = delta + cpu * pcpu_unit_size;
                 per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu);
                 per_cpu(cpu_number, cpu) = cpu;
                 setup_percpu_segment(cpu);
@@ -125,8 +438,6 @@ void __init setup_per_cpu_areas(void)
                  */
                 if (cpu == boot_cpu_id)
                         switch_to_new_gdt(cpu);
-
-                DBG("PERCPU: cpu %4d %p\n", cpu, ptr);
         }
 
         /* indicate the early static arrays will soon be gone */
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index 7cdcd16885ed..d2cc6428c587 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -187,40 +187,35 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate,
 /*
  * Set up a signal frame.
  */
-#ifdef CONFIG_X86_32
-static const struct {
-        u16 poplmovl;
-        u32 val;
-        u16 int80;
-} __attribute__((packed)) retcode = {
-        0xb858,         /* popl %eax; movl $..., %eax */
-        __NR_sigreturn,
-        0x80cd,         /* int $0x80 */
-};
-
-static const struct {
-        u8  movl;
-        u32 val;
-        u16 int80;
-        u8  pad;
-} __attribute__((packed)) rt_retcode = {
-        0xb8,           /* movl $..., %eax */
-        __NR_rt_sigreturn,
-        0x80cd,         /* int $0x80 */
-        0
-};
 
 /*
  * Determine which stack to use..
  */
+static unsigned long align_sigframe(unsigned long sp)
+{
+#ifdef CONFIG_X86_32
+        /*
+         * Align the stack pointer according to the i386 ABI,
+         * i.e. so that on function entry ((sp + 4) & 15) == 0.
+         */
+        sp = ((sp + 4) & -16ul) - 4;
+#else /* !CONFIG_X86_32 */
+        sp = round_down(sp, 16) - 8;
+#endif
+        return sp;
+}
+
 static inline void __user *
 get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
-             void **fpstate)
+             void __user **fpstate)
 {
-        unsigned long sp;
-
         /* Default to using normal stack */
-        sp = regs->sp;
+        unsigned long sp = regs->sp;
+
+#ifdef CONFIG_X86_64
+        /* redzone */
+        sp -= 128;
+#endif /* CONFIG_X86_64 */
 
         /*
          * If we are on the alternate signal stack and would overflow it, don't.
@@ -234,30 +229,52 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
                 if (sas_ss_flags(sp) == 0)
                         sp = current->sas_ss_sp + current->sas_ss_size;
         } else {
+#ifdef CONFIG_X86_32
                 /* This is the legacy signal stack switching. */
                 if ((regs->ss & 0xffff) != __USER_DS &&
                         !(ka->sa.sa_flags & SA_RESTORER) &&
                                 ka->sa.sa_restorer)
                         sp = (unsigned long) ka->sa.sa_restorer;
+#endif /* CONFIG_X86_32 */
         }
 
         if (used_math()) {
-                sp = sp - sig_xstate_size;
-                *fpstate = (struct _fpstate *) sp;
+                sp -= sig_xstate_size;
+#ifdef CONFIG_X86_64
+                sp = round_down(sp, 64);
+#endif /* CONFIG_X86_64 */
+                *fpstate = (void __user *)sp;
+
                 if (save_i387_xstate(*fpstate) < 0)
                         return (void __user *)-1L;
         }
 
-        sp -= frame_size;
-        /*
-         * Align the stack pointer according to the i386 ABI,
-         * i.e. so that on function entry ((sp + 4) & 15) == 0.
-         */
-        sp = ((sp + 4) & -16ul) - 4;
-
-        return (void __user *) sp;
+        return (void __user *)align_sigframe(sp - frame_size);
 }
 
+#ifdef CONFIG_X86_32
+static const struct {
+        u16 poplmovl;
+        u32 val;
+        u16 int80;
+} __attribute__((packed)) retcode = {
+        0xb858,         /* popl %eax; movl $..., %eax */
+        __NR_sigreturn,
+        0x80cd,         /* int $0x80 */
+};
+
+static const struct {
+        u8  movl;
+        u32 val;
+        u16 int80;
+        u8  pad;
+} __attribute__((packed)) rt_retcode = {
+        0xb8,           /* movl $..., %eax */
+        __NR_rt_sigreturn,
+        0x80cd,         /* int $0x80 */
+        0
+};
+
 static int
 __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
               struct pt_regs *regs)
@@ -388,24 +405,6 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
         return 0;
 }
 #else /* !CONFIG_X86_32 */
-/*
- * Determine which stack to use..
- */
-static void __user *
-get_stack(struct k_sigaction *ka, unsigned long sp, unsigned long size)
-{
-        /* Default to using normal stack - redzone*/
-        sp -= 128;
-
-        /* This is the X/Open sanctioned signal stack switching.  */
-        if (ka->sa.sa_flags & SA_ONSTACK) {
-                if (sas_ss_flags(sp) == 0)
-                        sp = current->sas_ss_sp + current->sas_ss_size;
-        }
-
-        return (void __user *)round_down(sp - size, 64);
-}
-
 static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
                             sigset_t *set, struct pt_regs *regs)
 {
@@ -414,15 +413,7 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
         int err = 0;
         struct task_struct *me = current;
 
-        if (used_math()) {
-                fp = get_stack(ka, regs->sp, sig_xstate_size);
-                frame = (void __user *)round_down(
-                        (unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8;
-
-                if (save_i387_xstate(fp) < 0)
-                        return -EFAULT;
-        } else
-                frame = get_stack(ka, regs->sp, sizeof(struct rt_sigframe)) - 8;
+        frame = get_sigframe(ka, regs, sizeof(struct rt_sigframe), &fp);
 
         if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
                 return -EFAULT;
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 9ce666387f37..249334f5080a 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -112,7 +112,7 @@ EXPORT_PER_CPU_SYMBOL(cpu_core_map);
 DEFINE_PER_CPU_SHARED_ALIGNED(struct cpuinfo_x86, cpu_info);
 EXPORT_PER_CPU_SYMBOL(cpu_info);
 
-static atomic_t init_deasserted;
+atomic_t init_deasserted;
 
 
 /* Set if we find a B stepping CPU */
@@ -614,12 +614,6 @@ wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip)
         unsigned long send_status, accept_status = 0;
         int maxlvt, num_starts, j;
 
-        if (get_uv_system_type() == UV_NON_UNIQUE_APIC) {
-                send_status = uv_wakeup_secondary(phys_apicid, start_eip);
-                atomic_set(&init_deasserted, 1);
-                return send_status;
-        }
-
         maxlvt = lapic_get_maxlvt();
 
         /*
@@ -748,7 +742,8 @@ static void __cpuinit do_fork_idle(struct work_struct *work)
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
  * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
- * Returns zero if CPU booted OK, else error code from ->wakeup_cpu.
+ * Returns zero if CPU booted OK, else error code from
+ * ->wakeup_secondary_cpu.
  */
 static int __cpuinit do_boot_cpu(int apicid, int cpu)
 {
@@ -835,9 +830,13 @@ do_rest:
         }
 
         /*
-         * Starting actual IPI sequence...
+         * Kick the secondary CPU. Use the method in the APIC driver
+         * if it's defined - or use an INIT boot APIC message otherwise:
          */
-        boot_error = apic->wakeup_cpu(apicid, start_ip);
+        if (apic->wakeup_secondary_cpu)
+                boot_error = apic->wakeup_secondary_cpu(apicid, start_ip);
+        else
+                boot_error = wakeup_secondary_cpu_via_init(apicid, start_ip);
 
         if (!boot_error) {
                 /*
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index c05430ac1b44..a1d288327ff0 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -118,47 +118,6 @@ die_if_kernel(const char *str, struct pt_regs *regs, long err)
         if (!user_mode_vm(regs))
                 die(str, regs, err);
 }
-
-/*
- * Perform the lazy TSS's I/O bitmap copy. If the TSS has an
- * invalid offset set (the LAZY one) and the faulting thread has
- * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS,
- * we set the offset field correctly and return 1.
- */
-static int lazy_iobitmap_copy(void)
-{
-        struct thread_struct *thread;
-        struct tss_struct *tss;
-        int cpu;
-
-        cpu = get_cpu();
-        tss = &per_cpu(init_tss, cpu);
-        thread = &current->thread;
-
-        if (tss->x86_tss.io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY &&
-            thread->io_bitmap_ptr) {
-                memcpy(tss->io_bitmap, thread->io_bitmap_ptr,
-                       thread->io_bitmap_max);
-                /*
-                 * If the previously set map was extending to higher ports
-                 * than the current one, pad extra space with 0xff (no access).
-                 */
-                if (thread->io_bitmap_max < tss->io_bitmap_max) {
-                        memset((char *) tss->io_bitmap +
-                                thread->io_bitmap_max, 0xff,
-                                tss->io_bitmap_max - thread->io_bitmap_max);
-                }
-                tss->io_bitmap_max = thread->io_bitmap_max;
-                tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET;
-                tss->io_bitmap_owner = thread;
-                put_cpu();
-
-                return 1;
-        }
-        put_cpu();
-
-        return 0;
-}
 #endif
 
 static void __kprobes
@@ -309,11 +268,6 @@ do_general_protection(struct pt_regs *regs, long error_code)
         conditional_sti(regs);
 
 #ifdef CONFIG_X86_32
-        if (lazy_iobitmap_copy()) {
-                /* restart the faulting instruction */
-                return;
-        }
-
         if (regs->flags & X86_VM_MASK)
                 goto gp_in_vm86;
 #endif
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index c609205df594..74de562812cc 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -22,7 +22,7 @@
 #include <asm/paravirt.h>
 #include <asm/setup.h>
 
-#if defined CONFIG_PCI && defined CONFIG_PARAVIRT
+#ifdef CONFIG_PARAVIRT
 /*
  * Interrupt control on vSMPowered systems:
  * ~AC is a shadow of IF.  If IF is 'on' AC should be 'off'
@@ -114,7 +114,6 @@ static void __init set_vsmp_pv_ops(void)
 }
 #endif
 
-#ifdef CONFIG_PCI
 static int is_vsmp = -1;
 
 static void __init detect_vsmp_box(void)
@@ -139,15 +138,6 @@ int is_vsmp_box(void)
                 return 0;
         }
 }
-#else
-static void __init detect_vsmp_box(void)
-{
-}
-int is_vsmp_box(void)
-{
-        return 0;
-}
-#endif
 
 void __init vsmp_init(void)
 {
diff --git a/arch/x86/math-emu/fpu_aux.c b/arch/x86/math-emu/fpu_aux.c
index 491e737ce547..aa0987088774 100644
--- a/arch/x86/math-emu/fpu_aux.c
+++ b/arch/x86/math-emu/fpu_aux.c
@@ -30,20 +30,29 @@ static void fclex(void)
 }
 
 /* Needs to be externally visible */
-void finit(void)
+void finit_task(struct task_struct *tsk)
 {
-        control_word = 0x037f;
-        partial_status = 0;
-        top = 0;                /* We don't keep top in the status word internally. */
-        fpu_tag_word = 0xffff;
+        struct i387_soft_struct *soft = &tsk->thread.xstate->soft;
+        struct address *oaddr, *iaddr;
+        soft->cwd = 0x037f;
+        soft->swd = 0;
+        soft->ftop = 0; /* We don't keep top in the status word internally. */
+        soft->twd = 0xffff;
         /* The behaviour is different from that detailed in
            Section 15.1.6 of the Intel manual */
-        operand_address.offset = 0;
-        operand_address.selector = 0;
-        instruction_address.offset = 0;
-        instruction_address.selector = 0;
-        instruction_address.opcode = 0;
-        no_ip_update = 1;
+        oaddr = (struct address *)&soft->foo;
+        oaddr->offset = 0;
+        oaddr->selector = 0;
+        iaddr = (struct address *)&soft->fip;
+        iaddr->offset = 0;
+        iaddr->selector = 0;
+        iaddr->opcode = 0;
+        soft->no_update = 1;
+}
+
+void finit(void)
+{
+        finit_task(current);
 }
 
 /*
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 2b938a384910..08537747cb58 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,4 +1,4 @@
-obj-y   :=  init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
+obj-y   :=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
             pat.o pgtable.o gup.o
 
 obj-$(CONFIG_SMP)               += tlb.o
diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index bcc079c282dd..00f127c80b0e 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -1,5 +1,6 @@
 #include <linux/highmem.h>
 #include <linux/module.h>
+#include <linux/swap.h> /* for totalram_pages */
 
 void *kmap(struct page *page)
 {
@@ -156,3 +157,36 @@ EXPORT_SYMBOL(kmap);
 EXPORT_SYMBOL(kunmap);
 EXPORT_SYMBOL(kmap_atomic);
 EXPORT_SYMBOL(kunmap_atomic);
+
+#ifdef CONFIG_NUMA
+void __init set_highmem_pages_init(void)
+{
+        struct zone *zone;
+        int nid;
+
+        for_each_zone(zone) {
+                unsigned long zone_start_pfn, zone_end_pfn;
+
+                if (!is_highmem(zone))
+                        continue;
+
+                zone_start_pfn = zone->zone_start_pfn;
+                zone_end_pfn = zone_start_pfn + zone->spanned_pages;
+
+                nid = zone_to_nid(zone);
+                printk(KERN_INFO "Initializing %s for node %d (%08lx:%08lx)\n",
+                                zone->name, nid, zone_start_pfn, zone_end_pfn);
+
+                add_highpages_with_active_regions(nid, zone_start_pfn,
+                                 zone_end_pfn);
+        }
+        totalram_pages += totalhigh_pages;
+}
+#else
+void __init set_highmem_pages_init(void)
+{
+        add_highpages_with_active_regions(0, highstart_pfn, highend_pfn);
+
+        totalram_pages += totalhigh_pages;
+}
+#endif /* CONFIG_NUMA */
diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
new file mode 100644
index 000000000000..ce6a722587d8
--- /dev/null
+++ b/arch/x86/mm/init.c
@@ -0,0 +1,49 @@
+#include <linux/swap.h>
+#include <asm/cacheflush.h>
+#include <asm/page.h>
+#include <asm/sections.h>
+#include <asm/system.h>
+
+void free_init_pages(char *what, unsigned long begin, unsigned long end)
+{
+        unsigned long addr = begin;
+
+        if (addr >= end)
+                return;
+
+        /*
+         * If debugging page accesses then do not free this memory but
+         * mark them not present - any buggy init-section access will
+         * create a kernel page fault:
+         */
+#ifdef CONFIG_DEBUG_PAGEALLOC
+        printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
+                begin, PAGE_ALIGN(end));
+        set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
+#else
+        /*
+         * We just marked the kernel text read only above, now that
+         * we are going to free part of that, we need to make that
+         * writeable first.
+         */
+        set_memory_rw(begin, (end - begin) >> PAGE_SHIFT);
+
+        printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
+
+        for (; addr < end; addr += PAGE_SIZE) {
+                ClearPageReserved(virt_to_page(addr));
+                init_page_count(virt_to_page(addr));
+                memset((void *)(addr & ~(PAGE_SIZE-1)),
+                        POISON_FREE_INITMEM, PAGE_SIZE);
+                free_page(addr);
+                totalram_pages++;
+        }
+#endif
+}
+
+void free_initmem(void)
+{
+        free_init_pages("unused kernel memory",
+                        (unsigned long)(&__init_begin),
+                        (unsigned long)(&__init_end));
+}
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 06708ee94aa4..47df0e1bbeb9 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -50,8 +50,6 @@
 #include <asm/setup.h>
 #include <asm/cacheflush.h>
 
-unsigned int __VMALLOC_RESERVE = 128 << 20;
-
 unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
 
@@ -137,6 +135,23 @@ static pte_t * __init one_page_table_init(pmd_t *pmd)
         return pte_offset_kernel(pmd, 0);
 }
 
+pmd_t * __init populate_extra_pmd(unsigned long vaddr)
+{
+        int pgd_idx = pgd_index(vaddr);
+        int pmd_idx = pmd_index(vaddr);
+
+        return one_md_table_init(swapper_pg_dir + pgd_idx) + pmd_idx;
+}
+
+pte_t * __init populate_extra_pte(unsigned long vaddr)
+{
+        int pte_idx = pte_index(vaddr);
+        pmd_t *pmd;
+
+        pmd = populate_extra_pmd(vaddr);
+        return one_page_table_init(pmd) + pte_idx;
+}
+
 static pte_t *__init page_table_kmap_check(pte_t *pte, pmd_t *pmd,
                                            unsigned long vaddr, pte_t *lastpte)
 {
@@ -469,22 +484,10 @@ void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn,
         work_with_active_regions(nid, add_highpages_work_fn, &data);
 }
 
-#ifndef CONFIG_NUMA
-static void __init set_highmem_pages_init(void)
-{
-        add_highpages_with_active_regions(0, highstart_pfn, highend_pfn);
-
-        totalram_pages += totalhigh_pages;
-}
-#endif /* !CONFIG_NUMA */
-
 #else
 static inline void permanent_kmaps_init(pgd_t *pgd_base)
 {
 }
-static inline void set_highmem_pages_init(void)
-{
-}
 #endif /* CONFIG_HIGHMEM */
 
 void __init native_pagetable_setup_start(pgd_t *base)
@@ -847,10 +850,10 @@ static void __init find_early_table_space(unsigned long end, int use_pse)
         unsigned long puds, pmds, ptes, tables, start;
 
         puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
-        tables = PAGE_ALIGN(puds * sizeof(pud_t));
+        tables = roundup(puds * sizeof(pud_t), PAGE_SIZE);
 
         pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
-        tables += PAGE_ALIGN(pmds * sizeof(pmd_t));
+        tables += roundup(pmds * sizeof(pmd_t), PAGE_SIZE);
 
         if (use_pse) {
                 unsigned long extra;
@@ -861,10 +864,10 @@ static void __init find_early_table_space(unsigned long end, int use_pse)
         } else
                 ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
-        tables += PAGE_ALIGN(ptes * sizeof(pte_t));
+        tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE);
 
         /* for fixmap */
-        tables += PAGE_ALIGN(__end_of_fixed_addresses * sizeof(pte_t));
+        tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE);
 
         /*
          * RED-PEN putting page tables only on node 0 could
@@ -1214,45 +1217,6 @@ void mark_rodata_ro(void)
 }
 #endif
 
-void free_init_pages(char *what, unsigned long begin, unsigned long end)
-{
-#ifdef CONFIG_DEBUG_PAGEALLOC
-        /*
-         * If debugging page accesses then do not free this memory but
-         * mark them not present - any buggy init-section access will
-         * create a kernel page fault:
-         */
-        printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
-                begin, PAGE_ALIGN(end));
-        set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
-#else
-        unsigned long addr;
-
-        /*
-         * We just marked the kernel text read only above, now that
-         * we are going to free part of that, we need to make that
-         * writeable first.
-         */
-        set_memory_rw(begin, (end - begin) >> PAGE_SHIFT);
-
-        for (addr = begin; addr < end; addr += PAGE_SIZE) {
-                ClearPageReserved(virt_to_page(addr));
-                init_page_count(virt_to_page(addr));
-                memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
-                free_page(addr);
-                totalram_pages++;
-        }
-        printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
-#endif
-}
-
-void free_initmem(void)
-{
-        free_init_pages("unused kernel memory",
-                        (unsigned long)(&__init_begin),
-                        (unsigned long)(&__init_end));
-}
-
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index e6d36b490250..07f44d491df1 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -168,34 +168,51 @@ static __ref void *spp_getpage(void)
         return ptr;
 }
 
-void
-set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
+static pud_t *fill_pud(pgd_t *pgd, unsigned long vaddr)
 {
-        pud_t *pud;
-        pmd_t *pmd;
-        pte_t *pte;
+        if (pgd_none(*pgd)) {
+                pud_t *pud = (pud_t *)spp_getpage();
+                pgd_populate(&init_mm, pgd, pud);
+                if (pud != pud_offset(pgd, 0))
+                        printk(KERN_ERR "PAGETABLE BUG #00! %p <-> %p\n",
+                               pud, pud_offset(pgd, 0));
+        }
+        return pud_offset(pgd, vaddr);
+}
 
-        pud = pud_page + pud_index(vaddr);
+static pmd_t *fill_pmd(pud_t *pud, unsigned long vaddr)
+{
         if (pud_none(*pud)) {
-                pmd = (pmd_t *) spp_getpage();
+                pmd_t *pmd = (pmd_t *) spp_getpage();
                 pud_populate(&init_mm, pud, pmd);
-                if (pmd != pmd_offset(pud, 0)) {
+                if (pmd != pmd_offset(pud, 0))
                         printk(KERN_ERR "PAGETABLE BUG #01! %p <-> %p\n",
-                                pmd, pmd_offset(pud, 0));
-                        return;
-                }
+                               pmd, pmd_offset(pud, 0));
         }
-        pmd = pmd_offset(pud, vaddr);
+        return pmd_offset(pud, vaddr);
+}
+
+static pte_t *fill_pte(pmd_t *pmd, unsigned long vaddr)
+{
         if (pmd_none(*pmd)) {
-                pte = (pte_t *) spp_getpage();
+                pte_t *pte = (pte_t *) spp_getpage();
                 pmd_populate_kernel(&init_mm, pmd, pte);
-                if (pte != pte_offset_kernel(pmd, 0)) {
+                if (pte != pte_offset_kernel(pmd, 0))
                         printk(KERN_ERR "PAGETABLE BUG #02!\n");
-                        return;
-                }
         }
+        return pte_offset_kernel(pmd, vaddr);
+}
+
+void set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
+{
+        pud_t *pud;
+        pmd_t *pmd;
+        pte_t *pte;
+
+        pud = pud_page + pud_index(vaddr);
+        pmd = fill_pmd(pud, vaddr);
+        pte = fill_pte(pmd, vaddr);
 
-        pte = pte_offset_kernel(pmd, vaddr);
         set_pte(pte, new_pte);
 
         /*
@@ -205,8 +222,7 @@ set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
         __flush_tlb_one(vaddr);
 }
 
-void
-set_pte_vaddr(unsigned long vaddr, pte_t pteval)
+void set_pte_vaddr(unsigned long vaddr, pte_t pteval)
 {
         pgd_t *pgd;
         pud_t *pud_page;
@@ -223,6 +239,24 @@ set_pte_vaddr(unsigned long vaddr, pte_t pteval)
         set_pte_vaddr_pud(pud_page, vaddr, pteval);
 }
 
+pmd_t * __init populate_extra_pmd(unsigned long vaddr)
+{
+        pgd_t *pgd;
+        pud_t *pud;
+
+        pgd = pgd_offset_k(vaddr);
+        pud = fill_pud(pgd, vaddr);
+        return fill_pmd(pud, vaddr);
+}
+
+pte_t * __init populate_extra_pte(unsigned long vaddr)
+{
+        pmd_t *pmd;
+
+        pmd = populate_extra_pmd(vaddr);
+        return fill_pte(pmd, vaddr);
+}
+
 /*
  * Create large page table mappings for a range of physical addresses.
  */
@@ -714,6 +748,8 @@ unsigned long __init_refok init_memory_mapping(unsigned long start,
         pos = start_pfn << PAGE_SHIFT;
         end_pfn = ((pos + (PMD_SIZE - 1)) >> PMD_SHIFT)
                         << (PMD_SHIFT - PAGE_SHIFT);
+        if (end_pfn > (end >> PAGE_SHIFT))
+                end_pfn = end >> PAGE_SHIFT;
         if (start_pfn < end_pfn) {
                 nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
                 pos = end_pfn << PAGE_SHIFT;
@@ -945,43 +981,6 @@ void __init mem_init(void)
                 initsize >> 10);
 }
 
-void free_init_pages(char *what, unsigned long begin, unsigned long end)
-{
-        unsigned long addr = begin;
-
-        if (addr >= end)
-                return;
-
-        /*
-         * If debugging page accesses then do not free this memory but
-         * mark them not present - any buggy init-section access will
-         * create a kernel page fault:
-         */
-#ifdef CONFIG_DEBUG_PAGEALLOC
-        printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
-                begin, PAGE_ALIGN(end));
-        set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
-#else
-        printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
-
-        for (; addr < end; addr += PAGE_SIZE) {
-                ClearPageReserved(virt_to_page(addr));
-                init_page_count(virt_to_page(addr));
-                memset((void *)(addr & ~(PAGE_SIZE-1)),
-                        POISON_FREE_INITMEM, PAGE_SIZE);
-                free_page(addr);
-                totalram_pages++;
-        }
-#endif
-}
-
-void free_initmem(void)
-{
-        free_init_pages("unused kernel memory",
-                        (unsigned long)(&__init_begin),
-                        (unsigned long)(&__init_end));
-}
-
 #ifdef CONFIG_DEBUG_RODATA
 const int rodata_test_data = 0xC3;
 EXPORT_SYMBOL_GPL(rodata_test_data);
diff --git a/arch/x86/mm/kmmio.c b/arch/x86/mm/kmmio.c
index 93d82038af4b..9f205030d9aa 100644
--- a/arch/x86/mm/kmmio.c
+++ b/arch/x86/mm/kmmio.c
@@ -32,11 +32,14 @@ struct kmmio_fault_page {
         struct list_head list;
         struct kmmio_fault_page *release_next;
         unsigned long page; /* location of the fault page */
+        bool old_presence; /* page presence prior to arming */
+        bool armed;
 
         /*
          * Number of times this page has been registered as a part
          * of a probe. If zero, page is disarmed and this may be freed.
-         * Used only by writers (RCU).
+         * Used only by writers (RCU) and post_kmmio_handler().
+         * Protected by kmmio_lock, when linked into kmmio_page_table.
          */
         int count;
 };
@@ -105,57 +108,85 @@ static struct kmmio_fault_page *get_kmmio_fault_page(unsigned long page)
         return NULL;
 }
 
-static void set_page_present(unsigned long addr, bool present,
-                                                        unsigned int *pglevel)
+static void set_pmd_presence(pmd_t *pmd, bool present, bool *old)
+{
+        pmdval_t v = pmd_val(*pmd);
+        *old = !!(v & _PAGE_PRESENT);
+        v &= ~_PAGE_PRESENT;
+        if (present)
+                v |= _PAGE_PRESENT;
+        set_pmd(pmd, __pmd(v));
+}
+
+static void set_pte_presence(pte_t *pte, bool present, bool *old)
+{
+        pteval_t v = pte_val(*pte);
+        *old = !!(v & _PAGE_PRESENT);
+        v &= ~_PAGE_PRESENT;
+        if (present)
+                v |= _PAGE_PRESENT;
+        set_pte_atomic(pte, __pte(v));
+}
+
+static int set_page_presence(unsigned long addr, bool present, bool *old)
 {
-        pteval_t pteval;
-        pmdval_t pmdval;
         unsigned int level;
-        pmd_t *pmd;
         pte_t *pte = lookup_address(addr, &level);
 
         if (!pte) {
                 pr_err("kmmio: no pte for page 0x%08lx\n", addr);
-                return;
+                return -1;
         }
 
-        if (pglevel)
-                *pglevel = level;
-
         switch (level) {
         case PG_LEVEL_2M:
-                pmd = (pmd_t *)pte;
-                pmdval = pmd_val(*pmd) & ~_PAGE_PRESENT;
-                if (present)
-                        pmdval |= _PAGE_PRESENT;
-                set_pmd(pmd, __pmd(pmdval));
+                set_pmd_presence((pmd_t *)pte, present, old);
                 break;
-
         case PG_LEVEL_4K:
-                pteval = pte_val(*pte) & ~_PAGE_PRESENT;
-                if (present)
-                        pteval |= _PAGE_PRESENT;
-                set_pte_atomic(pte, __pte(pteval));
+                set_pte_presence(pte, present, old);
                 break;
-
         default:
                 pr_err("kmmio: unexpected page level 0x%x.\n", level);
-                return;
+                return -1;
         }
 
         __flush_tlb_one(addr);
+        return 0;
 }
 
-/** Mark the given page as not present. Access to it will trigger a fault. */
-static void arm_kmmio_fault_page(unsigned long page, unsigned int *pglevel)
+/*
+ * Mark the given page as not present. Access to it will trigger a fault.
+ *
+ * Struct kmmio_fault_page is protected by RCU and kmmio_lock, but the
+ * protection is ignored here. RCU read lock is assumed held, so the struct
+ * will not disappear unexpectedly. Furthermore, the caller must guarantee,
+ * that double arming the same virtual address (page) cannot occur.
+ *
+ * Double disarming on the other hand is allowed, and may occur when a fault
+ * and mmiotrace shutdown happen simultaneously.
+ */
+static int arm_kmmio_fault_page(struct kmmio_fault_page *f)
 {
-        set_page_present(page & PAGE_MASK, false, pglevel);
+        int ret;
+        WARN_ONCE(f->armed, KERN_ERR "kmmio page already armed.\n");
+        if (f->armed) {
+                pr_warning("kmmio double-arm: page 0x%08lx, ref %d, old %d\n",
+                                        f->page, f->count, f->old_presence);
+        }
+        ret = set_page_presence(f->page, false, &f->old_presence);
+        WARN_ONCE(ret < 0, KERN_ERR "kmmio arming 0x%08lx failed.\n", f->page);
+        f->armed = true;
+        return ret;
 }
 
-/** Mark the given page as present. */
-static void disarm_kmmio_fault_page(unsigned long page, unsigned int *pglevel)
+/** Restore the given page to saved presence state. */
+static void disarm_kmmio_fault_page(struct kmmio_fault_page *f)
 {
-        set_page_present(page & PAGE_MASK, true, pglevel);
+        bool tmp;
+        int ret = set_page_presence(f->page, f->old_presence, &tmp);
+        WARN_ONCE(ret < 0,
+                        KERN_ERR "kmmio disarming 0x%08lx failed.\n", f->page);
+        f->armed = false;
 }
 
 /*
@@ -202,28 +233,32 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 
         ctx = &get_cpu_var(kmmio_ctx);
         if (ctx->active) {
-                disarm_kmmio_fault_page(faultpage->page, NULL);
                 if (addr == ctx->addr) {
                         /*
-                         * On SMP we sometimes get recursive probe hits on the
-                         * same address. Context is already saved, fall out.
+                         * A second fault on the same page means some other
+                         * condition needs handling by do_page_fault(), the
+                         * page really not being present is the most common.
                          */
-                        pr_debug("kmmio: duplicate probe hit on CPU %d, for "
-                                                "address 0x%08lx.\n",
-                                                smp_processor_id(), addr);
-                        ret = 1;
-                        goto no_kmmio_ctx;
-                }
-                /*
-                 * Prevent overwriting already in-flight context.
-                 * This should not happen, let's hope disarming at least
-                 * prevents a panic.
-                 */
-                pr_emerg("kmmio: recursive probe hit on CPU %d, "
+                        pr_debug("kmmio: secondary hit for 0x%08lx CPU %d.\n",
+                                        addr, smp_processor_id());
+
+                        if (!faultpage->old_presence)
+                                pr_info("kmmio: unexpected secondary hit for "
+                                        "address 0x%08lx on CPU %d.\n", addr,
+                                        smp_processor_id());
+                } else {
+                        /*
+                         * Prevent overwriting already in-flight context.
+                         * This should not happen, let's hope disarming at
+                         * least prevents a panic.
+                         */
+                        pr_emerg("kmmio: recursive probe hit on CPU %d, "
                                         "for address 0x%08lx. Ignoring.\n",
                                         smp_processor_id(), addr);
-                pr_emerg("kmmio: previous hit was at 0x%08lx.\n",
-                                        ctx->addr);
+                        pr_emerg("kmmio: previous hit was at 0x%08lx.\n",
+                                                ctx->addr);
+                        disarm_kmmio_fault_page(faultpage);
+                }
                 goto no_kmmio_ctx;
         }
         ctx->active++;
@@ -244,7 +279,7 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
         regs->flags &= ~X86_EFLAGS_IF;
 
         /* Now we set present bit in PTE and single step. */
-        disarm_kmmio_fault_page(ctx->fpage->page, NULL);
+        disarm_kmmio_fault_page(ctx->fpage);
 
         /*
          * If another cpu accesses the same page while we are stepping,
@@ -275,7 +310,7 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
         struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx);
 
         if (!ctx->active) {
-                pr_debug("kmmio: spurious debug trap on CPU %d.\n",
+                pr_warning("kmmio: spurious debug trap on CPU %d.\n",
                                                         smp_processor_id());
                 goto out;
         }
@@ -283,7 +318,11 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
         if (ctx->probe && ctx->probe->post_handler)
                 ctx->probe->post_handler(ctx->probe, condition, regs);
 
-        arm_kmmio_fault_page(ctx->fpage->page, NULL);
+        /* Prevent racing against release_kmmio_fault_page(). */
+        spin_lock(&kmmio_lock);
+        if (ctx->fpage->count)
+                arm_kmmio_fault_page(ctx->fpage);
+        spin_unlock(&kmmio_lock);
 
         regs->flags &= ~X86_EFLAGS_TF;
         regs->flags |= ctx->saved_flags;
@@ -315,20 +354,24 @@ static int add_kmmio_fault_page(unsigned long page)
         f = get_kmmio_fault_page(page);
         if (f) {
                 if (!f->count)
-                        arm_kmmio_fault_page(f->page, NULL);
+                        arm_kmmio_fault_page(f);
                 f->count++;
                 return 0;
         }
 
-        f = kmalloc(sizeof(*f), GFP_ATOMIC);
+        f = kzalloc(sizeof(*f), GFP_ATOMIC);
         if (!f)
                 return -1;
 
         f->count = 1;
         f->page = page;
-        list_add_rcu(&f->list, kmmio_page_list(f->page));
 
-        arm_kmmio_fault_page(f->page, NULL);
+        if (arm_kmmio_fault_page(f)) {
+                kfree(f);
+                return -1;
+        }
+
+        list_add_rcu(&f->list, kmmio_page_list(f->page));
 
         return 0;
 }
@@ -347,7 +390,7 @@ static void release_kmmio_fault_page(unsigned long page,
         f->count--;
         BUG_ON(f->count < 0);
         if (!f->count) {
-                disarm_kmmio_fault_page(f->page, NULL);
+                disarm_kmmio_fault_page(f);
                 f->release_next = *release_list;
                 *release_list = f;
         }
diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c
index 3957cd6d6454..451fe95a0352 100644
--- a/arch/x86/mm/numa_32.c
+++ b/arch/x86/mm/numa_32.c
@@ -423,32 +423,6 @@ void __init initmem_init(unsigned long start_pfn,
         setup_bootmem_allocator();
 }
 
-void __init set_highmem_pages_init(void)
-{
-#ifdef CONFIG_HIGHMEM
-        struct zone *zone;
-        int nid;
-
-        for_each_zone(zone) {
-                unsigned long zone_start_pfn, zone_end_pfn;
-
-                if (!is_highmem(zone))
-                        continue;
-
-                zone_start_pfn = zone->zone_start_pfn;
-                zone_end_pfn = zone_start_pfn + zone->spanned_pages;
-
-                nid = zone_to_nid(zone);
-                printk(KERN_INFO "Initializing %s for node %d (%08lx:%08lx)\n",
-                                zone->name, nid, zone_start_pfn, zone_end_pfn);
-
-                add_highpages_with_active_regions(nid, zone_start_pfn,
-                                 zone_end_pfn);
-        }
-        totalram_pages += totalhigh_pages;
-#endif
-}
-
 #ifdef CONFIG_MEMORY_HOTPLUG
 static int paddr_to_nid(u64 addr)
 {
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index 86f2ffc43c3d..5b7c7c8464fe 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -313,6 +313,24 @@ int ptep_clear_flush_young(struct vm_area_struct *vma,
         return young;
 }
 
+/**
+ * reserve_top_address - reserves a hole in the top of kernel address space
+ * @reserve - size of hole to reserve
+ *
+ * Can be used to relocate the fixmap area and poke a hole in the top
+ * of kernel address space to make room for a hypervisor.
+ */
+void __init reserve_top_address(unsigned long reserve)
+{
+#ifdef CONFIG_X86_32
+        BUG_ON(fixmaps_set > 0);
+        printk(KERN_INFO "Reserving virtual address space above 0x%08x\n",
+               (int)-reserve);
+        __FIXADDR_TOP = -reserve - PAGE_SIZE;
+        __VMALLOC_RESERVE += reserve;
+#endif
+}
+
 int fixmaps_set;
 
 void __native_set_fixmap(enum fixed_addresses idx, pte_t pte)
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index 0951db9ee519..f2e477c91c1b 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -20,6 +20,8 @@
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 
+unsigned int __VMALLOC_RESERVE = 128 << 20;
+
 /*
  * Associate a virtual page frame with a given physical page frame 
  * and protection flags for that frame.
@@ -97,22 +99,6 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
 unsigned long __FIXADDR_TOP = 0xfffff000;
 EXPORT_SYMBOL(__FIXADDR_TOP);
 
-/**
- * reserve_top_address - reserves a hole in the top of kernel address space
- * @reserve - size of hole to reserve
- *
- * Can be used to relocate the fixmap area and poke a hole in the top
- * of kernel address space to make room for a hypervisor.
- */
-void __init reserve_top_address(unsigned long reserve)
-{
-        BUG_ON(fixmaps_set > 0);
-        printk(KERN_INFO "Reserving virtual address space above 0x%08x\n",
-               (int)-reserve);
-        __FIXADDR_TOP = -reserve - PAGE_SIZE;
-        __VMALLOC_RESERVE += reserve;
-}
-
 /*
  * vmalloc=size forces the vmalloc area to be exactly 'size'
  * bytes. This can be used to increase (or decrease) the
diff --git a/arch/x86/mm/testmmiotrace.c b/arch/x86/mm/testmmiotrace.c
index ab50a8d7402c..427fd1b56df5 100644
--- a/arch/x86/mm/testmmiotrace.c
+++ b/arch/x86/mm/testmmiotrace.c
@@ -1,5 +1,5 @@
 /*
- * Written by Pekka Paalanen, 2008 <pq@iki.fi>
+ * Written by Pekka Paalanen, 2008-2009 <pq@iki.fi>
  */
 #include <linux/module.h>
 #include <linux/io.h>
@@ -9,35 +9,74 @@
 
 static unsigned long mmio_address;
 module_param(mmio_address, ulong, 0);
-MODULE_PARM_DESC(mmio_address, "Start address of the mapping of 16 kB.");
+MODULE_PARM_DESC(mmio_address, " Start address of the mapping of 16 kB "
+                                "(or 8 MB if read_far is non-zero).");
+
+static unsigned long read_far = 0x400100;
+module_param(read_far, ulong, 0);
+MODULE_PARM_DESC(read_far, " Offset of a 32-bit read within 8 MB "
+                                "(default: 0x400100).");
+
+static unsigned v16(unsigned i)
+{
+        return i * 12 + 7;
+}
+
+static unsigned v32(unsigned i)
+{
+        return i * 212371 + 13;
+}
 
 static void do_write_test(void __iomem *p)
 {
         unsigned int i;
+        pr_info(MODULE_NAME ": write test.\n");
         mmiotrace_printk("Write test.\n");
+
         for (i = 0; i < 256; i++)
                 iowrite8(i, p + i);
+
         for (i = 1024; i < (5 * 1024); i += 2)
-                iowrite16(i * 12 + 7, p + i);
+                iowrite16(v16(i), p + i);
+
         for (i = (5 * 1024); i < (16 * 1024); i += 4)
-                iowrite32(i * 212371 + 13, p + i);
+                iowrite32(v32(i), p + i);
 }
 
 static void do_read_test(void __iomem *p)
 {
         unsigned int i;
+        unsigned errs[3] = { 0 };
+        pr_info(MODULE_NAME ": read test.\n");
         mmiotrace_printk("Read test.\n");
+
         for (i = 0; i < 256; i++)
-                ioread8(p + i);
+                if (ioread8(p + i) != i)
+                        ++errs[0];
+
         for (i = 1024; i < (5 * 1024); i += 2)
-                ioread16(p + i);
+                if (ioread16(p + i) != v16(i))
+                        ++errs[1];
+
         for (i = (5 * 1024); i < (16 * 1024); i += 4)
-                ioread32(p + i);
+                if (ioread32(p + i) != v32(i))
+                        ++errs[2];
+
+        mmiotrace_printk("Read errors: 8-bit %d, 16-bit %d, 32-bit %d.\n",
+                                                errs[0], errs[1], errs[2]);
 }
 
-static void do_test(void)
+static void do_read_far_test(void __iomem *p)
 {
-        void __iomem *p = ioremap_nocache(mmio_address, 0x4000);
+        pr_info(MODULE_NAME ": read far test.\n");
+        mmiotrace_printk("Read far test.\n");
+
+        ioread32(p + read_far);
+}
+
+static void do_test(unsigned long size)
+{
+        void __iomem *p = ioremap_nocache(mmio_address, size);
         if (!p) {
                 pr_err(MODULE_NAME ": could not ioremap, aborting.\n");
                 return;
@@ -45,11 +84,15 @@ static void do_test(void)
         mmiotrace_printk("ioremap returned %p.\n", p);
         do_write_test(p);
         do_read_test(p);
+        if (read_far && read_far < size - 4)
+                do_read_far_test(p);
         iounmap(p);
 }
 
 static int __init init(void)
 {
+        unsigned long size = (read_far) ? (8 << 20) : (16 << 10);
+
         if (mmio_address == 0) {
                 pr_err(MODULE_NAME ": you have to use the module argument "
                                                         "mmio_address.\n");
@@ -58,10 +101,11 @@ static int __init init(void)
                 return -ENXIO;
         }
 
-        pr_warning(MODULE_NAME ": WARNING: mapping 16 kB @ 0x%08lx "
-                                        "in PCI address space, and writing "
-                                        "rubbish in there.\n", mmio_address);
-        do_test();
+        pr_warning(MODULE_NAME ": WARNING: mapping %lu kB @ 0x%08lx in PCI "
+                "address space, and writing 16 kB of rubbish in there.\n",
+                 size >> 10, mmio_address);
+        do_test(size);
+        pr_info(MODULE_NAME ": All done.\n");
         return 0;
 }
 
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index e9f80c744cf3..10131fbdaada 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -78,8 +78,18 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
         if (cpu_has_arch_perfmon) {
                 union cpuid10_eax eax;
                 eax.full = cpuid_eax(0xa);
-                if (counter_width < eax.split.bit_width)
-                        counter_width = eax.split.bit_width;
+
+                /*
+                 * For Core2 (family 6, model 15), don't reset the
+                 * counter width:
+                 */
+                if (!(eax.split.version_id == 0 &&
+                        current_cpu_data.x86 == 6 &&
+                                current_cpu_data.x86_model == 15)) {
+
+                        if (counter_width < eax.split.bit_width)
+                                counter_width = eax.split.bit_width;
+                }
         }
 
         /* clear all counters */
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index c52f4034c7fd..82cd39a6cbd3 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -103,7 +103,7 @@ static void xen_vcpu_setup(int cpu)
 
         vcpup = &per_cpu(xen_vcpu_info, cpu);
 
-        info.mfn = virt_to_mfn(vcpup);
+        info.mfn = arbitrary_virt_to_mfn(vcpup);
         info.offset = offset_in_page(vcpup);
 
         printk(KERN_DEBUG "trying to map vcpu_info %d at %p, mfn %llx, offset %d\n",
@@ -301,8 +301,10 @@ static void xen_load_gdt(const struct desc_ptr *dtr)
         frames = mcs.args;
 
         for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) {
-                frames[f] = virt_to_mfn(va);
+                frames[f] = arbitrary_virt_to_mfn((void *)va);
+
                 make_lowmem_page_readonly((void *)va);
+                make_lowmem_page_readonly(mfn_to_virt(frames[f]));
         }
 
         MULTI_set_gdt(mcs.mc, frames, size / sizeof(struct desc_struct));
@@ -314,7 +316,7 @@ static void load_TLS_descriptor(struct thread_struct *t,
                                 unsigned int cpu, unsigned int i)
 {
         struct desc_struct *gdt = get_cpu_gdt_table(cpu);
-        xmaddr_t maddr = virt_to_machine(&gdt[GDT_ENTRY_TLS_MIN+i]);
+        xmaddr_t maddr = arbitrary_virt_to_machine(&gdt[GDT_ENTRY_TLS_MIN+i]);
         struct multicall_space mc = __xen_mc_entry(0);
 
         MULTI_update_descriptor(mc.mc, maddr.maddr, t->tls_array[i]);
@@ -488,7 +490,7 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
                 break;
 
         default: {
-                xmaddr_t maddr = virt_to_machine(&dt[entry]);
+                xmaddr_t maddr = arbitrary_virt_to_machine(&dt[entry]);
 
                 xen_mc_flush();
                 if (HYPERVISOR_update_descriptor(maddr.maddr, *(u64 *)desc))
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 319bd40a57c2..cb6afa4ec95c 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -276,6 +276,13 @@ void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
         p2m_top[topidx][idx] = mfn;
 }
 
+unsigned long arbitrary_virt_to_mfn(void *vaddr)
+{
+        xmaddr_t maddr = arbitrary_virt_to_machine(vaddr);
+
+        return PFN_DOWN(maddr.maddr);
+}
+
 xmaddr_t arbitrary_virt_to_machine(void *vaddr)
 {
         unsigned long address = (unsigned long)vaddr;
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 035582ae815d..8d470562ffc9 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -219,6 +219,7 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 {
         struct vcpu_guest_context *ctxt;
         struct desc_struct *gdt;
+        unsigned long gdt_mfn;
 
         if (cpumask_test_and_set_cpu(cpu, xen_cpu_initialized_map))
                 return 0;
@@ -248,9 +249,12 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
         ctxt->ldt_ents = 0;
 
         BUG_ON((unsigned long)gdt & ~PAGE_MASK);
+
+        gdt_mfn = arbitrary_virt_to_mfn(gdt);
         make_lowmem_page_readonly(gdt);
+        make_lowmem_page_readonly(mfn_to_virt(gdt_mfn));
 
-        ctxt->gdt_frames[0] = virt_to_mfn(gdt);
+        ctxt->gdt_frames[0] = gdt_mfn;
         ctxt->gdt_ents      = GDT_ENTRIES;
 
         ctxt->user_regs.cs = __KERNEL_CS;
diff --git a/block/blktrace.c b/block/blktrace.c
index 7cf9d1ff45a0..028120a0965a 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -363,7 +363,7 @@ int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
         if (!bt->sequence)
                 goto err;
 
-        bt->msg_data = __alloc_percpu(BLK_TN_MAX_MSG);
+        bt->msg_data = __alloc_percpu(BLK_TN_MAX_MSG, __alignof__(char));
         if (!bt->msg_data)
                 goto err;
 
diff --git a/crypto/api.c b/crypto/api.c
index efe77df6863f..38a2bc02a98c 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -215,8 +215,19 @@ struct crypto_alg *crypto_larval_lookup(const char *name, u32 type, u32 mask)
         mask &= ~(CRYPTO_ALG_LARVAL | CRYPTO_ALG_DEAD);
         type &= mask;
 
-        alg = try_then_request_module(crypto_alg_lookup(name, type, mask),
-                                      name);
+        alg = crypto_alg_lookup(name, type, mask);
+        if (!alg) {
+                char tmp[CRYPTO_MAX_ALG_NAME];
+
+                request_module(name);
+
+                if (!((type ^ CRYPTO_ALG_NEED_FALLBACK) & mask) &&
+                    snprintf(tmp, sizeof(tmp), "%s-all", name) < sizeof(tmp))
+                        request_module(tmp);
+
+                alg = crypto_alg_lookup(name, type, mask);
+        }
+
         if (alg)
                 return crypto_is_larval(alg) ? crypto_larval_wait(alg) : alg;
 
diff --git a/drivers/acpi/processor_perflib.c b/drivers/acpi/processor_perflib.c
index 9cc769b587ff..68fd3d292799 100644
--- a/drivers/acpi/processor_perflib.c
+++ b/drivers/acpi/processor_perflib.c
@@ -516,12 +516,12 @@ int acpi_processor_preregister_performance(
                         continue;
                 }
 
-                if (!performance || !percpu_ptr(performance, i)) {
+                if (!performance || !per_cpu_ptr(performance, i)) {
                         retval = -EINVAL;
                         continue;
                 }
 
-                pr->performance = percpu_ptr(performance, i);
+                pr->performance = per_cpu_ptr(performance, i);
                 cpumask_set_cpu(i, pr->performance->shared_cpu_map);
                 if (acpi_processor_get_psd(pr)) {
                         retval = -EINVAL;
diff --git a/drivers/crypto/ixp4xx_crypto.c b/drivers/crypto/ixp4xx_crypto.c
index 2d637e0fbc03..d9e751be8c5f 100644
--- a/drivers/crypto/ixp4xx_crypto.c
+++ b/drivers/crypto/ixp4xx_crypto.c
@@ -457,10 +457,12 @@ static int init_ixp_crypto(void)
         if (!ctx_pool) {
                 goto err;
         }
-        ret = qmgr_request_queue(SEND_QID, NPE_QLEN_TOTAL, 0, 0);
+        ret = qmgr_request_queue(SEND_QID, NPE_QLEN_TOTAL, 0, 0,
+                                 "ixp_crypto:out", NULL);
         if (ret)
                 goto err;
-        ret = qmgr_request_queue(RECV_QID, NPE_QLEN, 0, 0);
+        ret = qmgr_request_queue(RECV_QID, NPE_QLEN, 0, 0,
+                                 "ixp_crypto:in", NULL);
         if (ret) {
                 qmgr_release_queue(SEND_QID);
                 goto err;
diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c
index 856b3cc25583..3f0fdd18255d 100644
--- a/drivers/crypto/padlock-aes.c
+++ b/drivers/crypto/padlock-aes.c
@@ -489,4 +489,4 @@ MODULE_DESCRIPTION("VIA PadLock AES algorithm support");
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Michal Ludvig");
 
-MODULE_ALIAS("aes");
+MODULE_ALIAS("aes-all");
diff --git a/drivers/crypto/padlock-sha.c b/drivers/crypto/padlock-sha.c
index a7fbadebf623..a2c8e8514b63 100644
--- a/drivers/crypto/padlock-sha.c
+++ b/drivers/crypto/padlock-sha.c
@@ -304,7 +304,7 @@ MODULE_DESCRIPTION("VIA PadLock SHA1/SHA256 algorithms support.");
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Michal Ludvig");
 
-MODULE_ALIAS("sha1");
-MODULE_ALIAS("sha256");
+MODULE_ALIAS("sha1-all");
+MODULE_ALIAS("sha256-all");
 MODULE_ALIAS("sha1-padlock");
 MODULE_ALIAS("sha256-padlock");
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index ea5440dd10dc..647374acba94 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -1401,7 +1401,7 @@ MODULE_ALIAS("platform:iop-adma");
 
 static struct platform_driver iop_adma_driver = {
         .probe          = iop_adma_probe,
-        .remove         = iop_adma_remove,
+        .remove         = __devexit_p(iop_adma_remove),
         .driver         = {
                 .owner  = THIS_MODULE,
                 .name   = "iop-adma",
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index d35cbd1ff0b3..5d5d5b31867f 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -1287,7 +1287,7 @@ mv_xor_conf_mbus_windows(struct mv_xor_shared_private *msp,
 
 static struct platform_driver mv_xor_driver = {
         .probe          = mv_xor_probe,
-        .remove         = mv_xor_remove,
+        .remove         = __devexit_p(mv_xor_remove),
         .driver         = {
                 .owner  = THIS_MODULE,
                 .name   = MV_XOR_NAME,
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index 72c667f9bee1..12715d3c078d 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -420,7 +420,7 @@ int drm_rmmap_locked(struct drm_device *dev, drm_local_map_t *map)
                                 dev->sigdata.lock = NULL;
                         master->lock.hw_lock = NULL;   /* SHM removed */
                         master->lock.file_priv = NULL;
-                        wake_up_interruptible(&master->lock.lock_queue);
+                        wake_up_interruptible_all(&master->lock.lock_queue);
                 }
                 break;
         case _DRM_AGP:
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 6c020fe5431c..f52663ebe016 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -484,6 +484,7 @@ int drm_release(struct inode *inode, struct file *filp)
         mutex_lock(&dev->struct_mutex);
 
         if (file_priv->is_master) {
+                struct drm_master *master = file_priv->master;
                 struct drm_file *temp;
                 list_for_each_entry(temp, &dev->filelist, lhead) {
                         if ((temp->master == file_priv->master) &&
@@ -491,6 +492,19 @@ int drm_release(struct inode *inode, struct file *filp)
                                 temp->authenticated = 0;
                 }
 
+                /**
+                 * Since the master is disappearing, so is the
+                 * possibility to lock.
+                 */
+
+                if (master->lock.hw_lock) {
+                        if (dev->sigdata.lock == master->lock.hw_lock)
+                                dev->sigdata.lock = NULL;
+                        master->lock.hw_lock = NULL;
+                        master->lock.file_priv = NULL;
+                        wake_up_interruptible_all(&master->lock.lock_queue);
+                }
+
                 if (file_priv->minor->master == file_priv->master) {
                         /* drop the reference held my the minor */
                         drm_master_put(&file_priv->minor->master);
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index 46e7b28f0707..e2f70a516c34 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -80,6 +80,7 @@ int drm_lock(struct drm_device *dev, void *data, struct drm_file *file_priv)
                 __set_current_state(TASK_INTERRUPTIBLE);
                 if (!master->lock.hw_lock) {
                         /* Device has been unregistered */
+                        send_sig(SIGTERM, current, 0);
                         ret = -EINTR;
                         break;
                 }
@@ -93,7 +94,7 @@ int drm_lock(struct drm_device *dev, void *data, struct drm_file *file_priv)
                 /* Contention */
                 schedule();
                 if (signal_pending(current)) {
-                        ret = -ERESTARTSYS;
+                        ret = -EINTR;
                         break;
                 }
         }
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index 46bb923b097c..7c8b15b22bf2 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -146,14 +146,6 @@ static void drm_master_destroy(struct kref *kref)
 
         drm_ht_remove(&master->magiclist);
 
-        if (master->lock.hw_lock) {
-                if (dev->sigdata.lock == master->lock.hw_lock)
-                        dev->sigdata.lock = NULL;
-                master->lock.hw_lock = NULL;
-                master->lock.file_priv = NULL;
-                wake_up_interruptible(&master->lock.lock_queue);
-        }
-
         drm_free(master, sizeof(*master), DRM_MEM_DRIVER);
 }
 
@@ -176,7 +168,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
             file_priv->minor->master != file_priv->master) {
                 mutex_lock(&dev->struct_mutex);
                 file_priv->minor->master = drm_master_get(file_priv->master);
-                mutex_lock(&dev->struct_mutex);
+                mutex_unlock(&dev->struct_mutex);
         }
 
         return 0;
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index c29feb97c36d..85685bfd12da 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -211,7 +211,7 @@ fast_user_write(struct io_mapping *mapping,
 
         vaddr_atomic = io_mapping_map_atomic_wc(mapping, page_base);
         unwritten = __copy_from_user_inatomic_nocache(vaddr_atomic + page_offset,
-                                                      user_data, length, length);
+                                                      user_data, length);
         io_mapping_unmap_atomic(vaddr_atomic);
         if (unwritten)
                 return -EFAULT;
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 548ff2c66431..87b6b603469e 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -383,12 +383,13 @@ int i915_irq_emit(struct drm_device *dev, void *data,
         drm_i915_irq_emit_t *emit = data;
         int result;
 
-        RING_LOCK_TEST_WITH_RETURN(dev, file_priv);
-
         if (!dev_priv) {
                 DRM_ERROR("called with no initialization\n");
                 return -EINVAL;
         }
+
+        RING_LOCK_TEST_WITH_RETURN(dev, file_priv);
+
         mutex_lock(&dev->struct_mutex);
         result = i915_emit_irq(dev);
         mutex_unlock(&dev->struct_mutex);
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index eeda276f8f16..7f186bbcb99d 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -482,7 +482,7 @@ mv64xxx_i2c_map_regs(struct platform_device *pd,
         return 0;
 }
 
-static void __devexit
+static void
 mv64xxx_i2c_unmap_regs(struct mv64xxx_i2c_data *drv_data)
 {
         if (drv_data->reg_base) {
@@ -577,7 +577,7 @@ mv64xxx_i2c_remove(struct platform_device *dev)
 
 static struct platform_driver mv64xxx_i2c_driver = {
         .probe  = mv64xxx_i2c_probe,
-        .remove = mv64xxx_i2c_remove,
+        .remove = __devexit_p(mv64xxx_i2c_remove),
         .driver = {
                 .owner  = THIS_MODULE,
                 .name   = MV64XXX_I2C_CTLR_NAME,
diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c
index c3c8b9bc40ae..45470f18d7e9 100644
--- a/drivers/input/keyboard/atkbd.c
+++ b/drivers/input/keyboard/atkbd.c
@@ -839,7 +839,7 @@ static void atkbd_disconnect(struct serio *serio)
  */
 static void atkbd_dell_laptop_keymap_fixup(struct atkbd *atkbd)
 {
-        const unsigned int forced_release_keys[] = {
+        static const unsigned int forced_release_keys[] = {
                 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x8b, 0x8f, 0x93,
         };
         int i;
@@ -856,7 +856,7 @@ static void atkbd_dell_laptop_keymap_fixup(struct atkbd *atkbd)
  */
 static void atkbd_hp_keymap_fixup(struct atkbd *atkbd)
 {
-        const unsigned int forced_release_keys[] = {
+        static const unsigned int forced_release_keys[] = {
                 0x94,
         };
         int i;
diff --git a/drivers/input/keyboard/bf54x-keys.c b/drivers/input/keyboard/bf54x-keys.c
index 19284016e0f4..ee855c5202e8 100644
--- a/drivers/input/keyboard/bf54x-keys.c
+++ b/drivers/input/keyboard/bf54x-keys.c
@@ -209,8 +209,8 @@ static int __devinit bfin_kpad_probe(struct platform_device *pdev)
                 goto out;
         }
 
-        if (!pdata->debounce_time || !pdata->debounce_time > MAX_MULT ||
-            !pdata->coldrive_time || !pdata->coldrive_time > MAX_MULT) {
+        if (!pdata->debounce_time || pdata->debounce_time > MAX_MULT ||
+            !pdata->coldrive_time || pdata->coldrive_time > MAX_MULT) {
                 printk(KERN_ERR DRV_NAME
                         ": Invalid Debounce/Columdrive Time from pdata\n");
                 bfin_write_KPAD_MSEL(0xFF0);    /* Default MSEL */
diff --git a/drivers/input/keyboard/corgikbd.c b/drivers/input/keyboard/corgikbd.c
index c8ed065ea0cb..abb04c82c622 100644
--- a/drivers/input/keyboard/corgikbd.c
+++ b/drivers/input/keyboard/corgikbd.c
@@ -288,7 +288,7 @@ static int corgikbd_resume(struct platform_device *dev)
 #define corgikbd_resume         NULL
 #endif
 
-static int __init corgikbd_probe(struct platform_device *pdev)
+static int __devinit corgikbd_probe(struct platform_device *pdev)
 {
         struct corgikbd *corgikbd;
         struct input_dev *input_dev;
@@ -368,7 +368,7 @@ static int __init corgikbd_probe(struct platform_device *pdev)
         return err;
 }
 
-static int corgikbd_remove(struct platform_device *pdev)
+static int __devexit corgikbd_remove(struct platform_device *pdev)
 {
         int i;
         struct corgikbd *corgikbd = platform_get_drvdata(pdev);
@@ -388,7 +388,7 @@ static int corgikbd_remove(struct platform_device *pdev)
 
 static struct platform_driver corgikbd_driver = {
         .probe          = corgikbd_probe,
-        .remove         = corgikbd_remove,
+        .remove         = __devexit_p(corgikbd_remove),
         .suspend        = corgikbd_suspend,
         .resume         = corgikbd_resume,
         .driver         = {
@@ -397,7 +397,7 @@ static struct platform_driver corgikbd_driver = {
         },
 };
 
-static int __devinit corgikbd_init(void)
+static int __init corgikbd_init(void)
 {
         return platform_driver_register(&corgikbd_driver);
 }
diff --git a/drivers/input/keyboard/omap-keypad.c b/drivers/input/keyboard/omap-keypad.c
index 3f3d1198cdb1..058fa8b02c21 100644
--- a/drivers/input/keyboard/omap-keypad.c
+++ b/drivers/input/keyboard/omap-keypad.c
@@ -279,7 +279,7 @@ static int omap_kp_resume(struct platform_device *dev)
 #define omap_kp_resume  NULL
 #endif
 
-static int __init omap_kp_probe(struct platform_device *pdev)
+static int __devinit omap_kp_probe(struct platform_device *pdev)
 {
         struct omap_kp *omap_kp;
         struct input_dev *input_dev;
@@ -422,7 +422,7 @@ err1:
         return -EINVAL;
 }
 
-static int omap_kp_remove(struct platform_device *pdev)
+static int __devexit omap_kp_remove(struct platform_device *pdev)
 {
         struct omap_kp *omap_kp = platform_get_drvdata(pdev);
 
@@ -454,7 +454,7 @@ static int omap_kp_remove(struct platform_device *pdev)
 
 static struct platform_driver omap_kp_driver = {
         .probe          = omap_kp_probe,
-        .remove         = omap_kp_remove,
+        .remove         = __devexit_p(omap_kp_remove),
         .suspend        = omap_kp_suspend,
         .resume         = omap_kp_resume,
         .driver         = {
@@ -463,7 +463,7 @@ static struct platform_driver omap_kp_driver = {
         },
 };
 
-static int __devinit omap_kp_init(void)
+static int __init omap_kp_init(void)
 {
         printk(KERN_INFO "OMAP Keypad Driver\n");
         return platform_driver_register(&omap_kp_driver);
diff --git a/drivers/input/keyboard/spitzkbd.c b/drivers/input/keyboard/spitzkbd.c
index c48b76a46a58..9d1781a618e9 100644
--- a/drivers/input/keyboard/spitzkbd.c
+++ b/drivers/input/keyboard/spitzkbd.c
@@ -343,7 +343,7 @@ static int spitzkbd_resume(struct platform_device *dev)
 #define spitzkbd_resume         NULL
 #endif
 
-static int __init spitzkbd_probe(struct platform_device *dev)
+static int __devinit spitzkbd_probe(struct platform_device *dev)
 {
         struct spitzkbd *spitzkbd;
         struct input_dev *input_dev;
@@ -444,7 +444,7 @@ static int __init spitzkbd_probe(struct platform_device *dev)
         return err;
 }
 
-static int spitzkbd_remove(struct platform_device *dev)
+static int __devexit spitzkbd_remove(struct platform_device *dev)
 {
         int i;
         struct spitzkbd *spitzkbd = platform_get_drvdata(dev);
@@ -470,7 +470,7 @@ static int spitzkbd_remove(struct platform_device *dev)
 
 static struct platform_driver spitzkbd_driver = {
         .probe          = spitzkbd_probe,
-        .remove         = spitzkbd_remove,
+        .remove         = __devexit_p(spitzkbd_remove),
         .suspend        = spitzkbd_suspend,
         .resume         = spitzkbd_resume,
         .driver         = {
@@ -479,7 +479,7 @@ static struct platform_driver spitzkbd_driver = {
         },
 };
 
-static int __devinit spitzkbd_init(void)
+static int __init spitzkbd_init(void)
 {
         return platform_driver_register(&spitzkbd_driver);
 }
diff --git a/drivers/input/mouse/Kconfig b/drivers/input/mouse/Kconfig
index 9bef935ef19f..4f38e6f7dfdd 100644
--- a/drivers/input/mouse/Kconfig
+++ b/drivers/input/mouse/Kconfig
@@ -70,7 +70,7 @@ config MOUSE_PS2_SYNAPTICS
 config MOUSE_PS2_LIFEBOOK
         bool "Fujitsu Lifebook PS/2 mouse protocol extension" if EMBEDDED
         default y
-        depends on MOUSE_PS2
+        depends on MOUSE_PS2 && X86
         help
           Say Y here if you have a Fujitsu B-series Lifebook PS/2
           TouchScreen connected to your system.
diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c
index b9a25d57bc5e..6ab0eb1ada1c 100644
--- a/drivers/input/mouse/elantech.c
+++ b/drivers/input/mouse/elantech.c
@@ -542,7 +542,7 @@ int elantech_detect(struct psmouse *psmouse, int set_properties)
             ps2_command(ps2dev,  NULL, PSMOUSE_CMD_SETSCALE11) ||
             ps2_command(ps2dev,  NULL, PSMOUSE_CMD_SETSCALE11) ||
             ps2_command(ps2dev, param, PSMOUSE_CMD_GETINFO)) {
-                pr_err("elantech.c: sending Elantech magic knock failed.\n");
+                pr_debug("elantech.c: sending Elantech magic knock failed.\n");
                 return -1;
         }
 
@@ -551,8 +551,27 @@ int elantech_detect(struct psmouse *psmouse, int set_properties)
          * set of magic numbers
          */
         if (param[0] != 0x3c || param[1] != 0x03 || param[2] != 0xc8) {
-                pr_info("elantech.c: unexpected magic knock result 0x%02x, 0x%02x, 0x%02x.\n",
-                        param[0], param[1], param[2]);
+                pr_debug("elantech.c: "
+                         "unexpected magic knock result 0x%02x, 0x%02x, 0x%02x.\n",
+                         param[0], param[1], param[2]);
+                return -1;
+        }
+
+        /*
+         * Query touchpad's firmware version and see if it reports known
+         * value to avoid mis-detection. Logitech mice are known to respond
+         * to Elantech magic knock and there might be more.
+         */
+        if (synaptics_send_cmd(psmouse, ETP_FW_VERSION_QUERY, param)) {
+                pr_debug("elantech.c: failed to query firmware version.\n");
+                return -1;
+        }
+
+        pr_debug("elantech.c: Elantech version query result 0x%02x, 0x%02x, 0x%02x.\n",
+                 param[0], param[1], param[2]);
+
+        if (param[0] == 0 || param[1] != 0) {
+                pr_debug("elantech.c: Probably not a real Elantech touchpad. Aborting.\n");
                 return -1;
         }
 
@@ -600,8 +619,7 @@ int elantech_init(struct psmouse *psmouse)
         int i, error;
         unsigned char param[3];
 
-        etd = kzalloc(sizeof(struct elantech_data), GFP_KERNEL);
-        psmouse->private = etd;
+        psmouse->private = etd = kzalloc(sizeof(struct elantech_data), GFP_KERNEL);
         if (!etd)
                 return -1;
 
@@ -610,14 +628,12 @@ int elantech_init(struct psmouse *psmouse)
                 etd->parity[i] = etd->parity[i & (i - 1)] ^ 1;
 
         /*
-         * Find out what version hardware this is
+         * Do the version query again so we can store the result
          */
         if (synaptics_send_cmd(psmouse, ETP_FW_VERSION_QUERY, param)) {
                 pr_err("elantech.c: failed to query firmware version.\n");
                 goto init_fail;
         }
-        pr_info("elantech.c: Elantech version query result 0x%02x, 0x%02x, 0x%02x.\n",
-                param[0], param[1], param[2]);
         etd->fw_version_maj = param[0];
         etd->fw_version_min = param[2];
 
diff --git a/drivers/input/mouse/pxa930_trkball.c b/drivers/input/mouse/pxa930_trkball.c
index d297accf9a7f..1e827ad0afbe 100644
--- a/drivers/input/mouse/pxa930_trkball.c
+++ b/drivers/input/mouse/pxa930_trkball.c
@@ -83,7 +83,7 @@ static int write_tbcr(struct pxa930_trkball *trkball, int v)
 
         __raw_writel(v, trkball->mmio_base + TBCR);
 
-        while (i--) {
+        while (--i) {
                 if (__raw_readl(trkball->mmio_base + TBCR) == v)
                         break;
                 msleep(1);
diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c
index 865fc69e9bc3..f3e4f7b0240d 100644
--- a/drivers/input/mouse/synaptics.c
+++ b/drivers/input/mouse/synaptics.c
@@ -182,11 +182,6 @@ static int synaptics_identify(struct psmouse *psmouse)
 
 static int synaptics_query_hardware(struct psmouse *psmouse)
 {
-        int retries = 0;
-
-        while ((retries++ < 3) && psmouse_reset(psmouse))
-                /* empty */;
-
         if (synaptics_identify(psmouse))
                 return -1;
         if (synaptics_model_id(psmouse))
@@ -582,6 +577,8 @@ static int synaptics_reconnect(struct psmouse *psmouse)
         struct synaptics_data *priv = psmouse->private;
         struct synaptics_data old_priv = *priv;
 
+        psmouse_reset(psmouse);
+
         if (synaptics_detect(psmouse, 0))
                 return -1;
 
@@ -640,6 +637,8 @@ int synaptics_init(struct psmouse *psmouse)
         if (!priv)
                 return -1;
 
+        psmouse_reset(psmouse);
+
         if (synaptics_query_hardware(psmouse)) {
                 printk(KERN_ERR "Unable to query Synaptics hardware.\n");
                 goto init_fail;
diff --git a/drivers/input/serio/ambakmi.c b/drivers/input/serio/ambakmi.c
index b10ffae7c39b..e29cdc13a199 100644
--- a/drivers/input/serio/ambakmi.c
+++ b/drivers/input/serio/ambakmi.c
@@ -57,7 +57,7 @@ static int amba_kmi_write(struct serio *io, unsigned char val)
         struct amba_kmi_port *kmi = io->port_data;
         unsigned int timeleft = 10000; /* timeout in 100ms */
 
-        while ((readb(KMISTAT) & KMISTAT_TXEMPTY) == 0 && timeleft--)
+        while ((readb(KMISTAT) & KMISTAT_TXEMPTY) == 0 && --timeleft)
                 udelay(10);
 
         if (timeleft)
@@ -129,8 +129,8 @@ static int amba_kmi_probe(struct amba_device *dev, void *id)
         io->write       = amba_kmi_write;
         io->open        = amba_kmi_open;
         io->close       = amba_kmi_close;
-        strlcpy(io->name, dev->dev.bus_id, sizeof(io->name));
-        strlcpy(io->phys, dev->dev.bus_id, sizeof(io->phys));
+        strlcpy(io->name, dev_name(&dev->dev), sizeof(io->name));
+        strlcpy(io->phys, dev_name(&dev->dev), sizeof(io->phys));
         io->port_data   = kmi;
         io->dev.parent  = &dev->dev;
 
diff --git a/drivers/input/serio/gscps2.c b/drivers/input/serio/gscps2.c
index adc3bd6e7f7b..bd0f92d9f40f 100644
--- a/drivers/input/serio/gscps2.c
+++ b/drivers/input/serio/gscps2.c
@@ -359,7 +359,7 @@ static int __init gscps2_probe(struct parisc_device *dev)
 
         snprintf(serio->name, sizeof(serio->name), "GSC PS/2 %s",
                  (ps2port->id == GSC_ID_KEYBOARD) ? "keyboard" : "mouse");
-        strlcpy(serio->phys, dev->dev.bus_id, sizeof(serio->phys));
+        strlcpy(serio->phys, dev_name(&dev->dev), sizeof(serio->phys));
         serio->id.type          = SERIO_8042;
         serio->write            = gscps2_write;
         serio->open             = gscps2_open;
diff --git a/drivers/input/serio/sa1111ps2.c b/drivers/input/serio/sa1111ps2.c
index 2ad88780a170..57953c0eb82f 100644
--- a/drivers/input/serio/sa1111ps2.c
+++ b/drivers/input/serio/sa1111ps2.c
@@ -246,8 +246,8 @@ static int __devinit ps2_probe(struct sa1111_dev *dev)
         serio->write            = ps2_write;
         serio->open             = ps2_open;
         serio->close            = ps2_close;
-        strlcpy(serio->name, dev->dev.bus_id, sizeof(serio->name));
-        strlcpy(serio->phys, dev->dev.bus_id, sizeof(serio->phys));
+        strlcpy(serio->name, dev_name(&dev->dev), sizeof(serio->name));
+        strlcpy(serio->phys, dev_name(&dev->dev), sizeof(serio->phys));
         serio->port_data        = ps2if;
         serio->dev.parent       = &dev->dev;
         ps2if->io               = serio;
diff --git a/drivers/input/touchscreen/atmel_tsadcc.c b/drivers/input/touchscreen/atmel_tsadcc.c
index a89a6a8f05e6..055969e8be13 100644
--- a/drivers/input/touchscreen/atmel_tsadcc.c
+++ b/drivers/input/touchscreen/atmel_tsadcc.c
@@ -236,7 +236,7 @@ static int __devinit atmel_tsadcc_probe(struct platform_device *pdev)
         ts_dev->bufferedmeasure = 0;
 
         snprintf(ts_dev->phys, sizeof(ts_dev->phys),
-                 "%s/input0", pdev->dev.bus_id);
+                 "%s/input0", dev_name(&pdev->dev));
 
         input_dev->name = "atmel touch screen controller";
         input_dev->phys = ts_dev->phys;
diff --git a/drivers/input/touchscreen/corgi_ts.c b/drivers/input/touchscreen/corgi_ts.c
index 65202c9f63ff..3fb51b54fe61 100644
--- a/drivers/input/touchscreen/corgi_ts.c
+++ b/drivers/input/touchscreen/corgi_ts.c
@@ -268,7 +268,7 @@ static int corgits_resume(struct platform_device *dev)
 #define corgits_resume          NULL
 #endif
 
-static int __init corgits_probe(struct platform_device *pdev)
+static int __devinit corgits_probe(struct platform_device *pdev)
 {
         struct corgi_ts *corgi_ts;
         struct input_dev *input_dev;
@@ -343,7 +343,7 @@ static int __init corgits_probe(struct platform_device *pdev)
         return err;
 }
 
-static int corgits_remove(struct platform_device *pdev)
+static int __devexit corgits_remove(struct platform_device *pdev)
 {
         struct corgi_ts *corgi_ts = platform_get_drvdata(pdev);
 
@@ -352,12 +352,13 @@ static int corgits_remove(struct platform_device *pdev)
         corgi_ts->machinfo->put_hsync();
         input_unregister_device(corgi_ts->input);
         kfree(corgi_ts);
+
         return 0;
 }
 
 static struct platform_driver corgits_driver = {
         .probe          = corgits_probe,
-        .remove         = corgits_remove,
+        .remove         = __devexit_p(corgits_remove),
         .suspend        = corgits_suspend,
         .resume         = corgits_resume,
         .driver         = {
@@ -366,7 +367,7 @@ static struct platform_driver corgits_driver = {
         },
 };
 
-static int __devinit corgits_init(void)
+static int __init corgits_init(void)
 {
         return platform_driver_register(&corgits_driver);
 }
diff --git a/drivers/input/touchscreen/tsc2007.c b/drivers/input/touchscreen/tsc2007.c
index b75dc2990574..4ab070246892 100644
--- a/drivers/input/touchscreen/tsc2007.c
+++ b/drivers/input/touchscreen/tsc2007.c
@@ -289,7 +289,8 @@ static int tsc2007_probe(struct i2c_client *client,
 
         pdata->init_platform_hw();
 
-        snprintf(ts->phys, sizeof(ts->phys), "%s/input0", client->dev.bus_id);
+        snprintf(ts->phys, sizeof(ts->phys),
+                 "%s/input0", dev_name(&client->dev));
 
         input_dev->name = "TSC2007 Touchscreen";
         input_dev->phys = ts->phys;
diff --git a/drivers/input/touchscreen/usbtouchscreen.c b/drivers/input/touchscreen/usbtouchscreen.c
index 5080b26ba160..fb7cb9bdfbd5 100644
--- a/drivers/input/touchscreen/usbtouchscreen.c
+++ b/drivers/input/touchscreen/usbtouchscreen.c
@@ -60,6 +60,10 @@ static int swap_xy;
 module_param(swap_xy, bool, 0644);
 MODULE_PARM_DESC(swap_xy, "If set X and Y axes are swapped.");
 
+static int hwcalib_xy;
+module_param(hwcalib_xy, bool, 0644);
+MODULE_PARM_DESC(hwcalib_xy, "If set hw-calibrated X/Y are used if available");
+
 /* device specifc data/functions */
 struct usbtouch_usb;
 struct usbtouch_device_info {
@@ -118,6 +122,7 @@ enum {
 
 #define USB_DEVICE_HID_CLASS(vend, prod) \
         .match_flags = USB_DEVICE_ID_MATCH_INT_CLASS \
+                | USB_DEVICE_ID_MATCH_INT_PROTOCOL \
                 | USB_DEVICE_ID_MATCH_DEVICE, \
         .idVendor = (vend), \
         .idProduct = (prod), \
@@ -260,8 +265,13 @@ static int panjit_read_data(struct usbtouch_usb *dev, unsigned char *pkt)
 
 static int mtouch_read_data(struct usbtouch_usb *dev, unsigned char *pkt)
 {
-        dev->x = (pkt[8] << 8) | pkt[7];
-        dev->y = (pkt[10] << 8) | pkt[9];
+        if (hwcalib_xy) {
+                dev->x = (pkt[4] << 8) | pkt[3];
+                dev->y = 0xffff - ((pkt[6] << 8) | pkt[5]);
+        } else {
+                dev->x = (pkt[8] << 8) | pkt[7];
+                dev->y = (pkt[10] << 8) | pkt[9];
+        }
         dev->touch = (pkt[2] & 0x40) ? 1 : 0;
 
         return 1;
@@ -294,6 +304,12 @@ static int mtouch_init(struct usbtouch_usb *usbtouch)
                         return ret;
         }
 
+        /* Default min/max xy are the raw values, override if using hw-calib */
+        if (hwcalib_xy) {
+                input_set_abs_params(usbtouch->input, ABS_X, 0, 0xffff, 0, 0);
+                input_set_abs_params(usbtouch->input, ABS_Y, 0, 0xffff, 0, 0);
+        }
+
         return 0;
 }
 #endif
diff --git a/drivers/media/video/uvc/uvc_status.c b/drivers/media/video/uvc/uvc_status.c
index c1e4ae27c613..c705f248da88 100644
--- a/drivers/media/video/uvc/uvc_status.c
+++ b/drivers/media/video/uvc/uvc_status.c
@@ -46,8 +46,8 @@ static int uvc_input_init(struct uvc_device *dev)
         usb_to_input_id(udev, &input->id);
         input->dev.parent = &dev->intf->dev;
 
-        set_bit(EV_KEY, input->evbit);
-        set_bit(BTN_0, input->keybit);
+        __set_bit(EV_KEY, input->evbit);
+        __set_bit(KEY_CAMERA, input->keybit);
 
         if ((ret = input_register_device(input)) < 0)
                 goto error;
@@ -70,8 +70,10 @@ static void uvc_input_cleanup(struct uvc_device *dev)
 static void uvc_input_report_key(struct uvc_device *dev, unsigned int code,
         int value)
 {
-        if (dev->input)
+        if (dev->input) {
                 input_report_key(dev->input, code, value);
+                input_sync(dev->input);
+        }
 }
 
 #else
@@ -96,7 +98,7 @@ static void uvc_event_streaming(struct uvc_device *dev, __u8 *data, int len)
                         return;
                 uvc_trace(UVC_TRACE_STATUS, "Button (intf %u) %s len %d\n",
                         data[1], data[3] ? "pressed" : "released", len);
-                uvc_input_report_key(dev, BTN_0, data[3]);
+                uvc_input_report_key(dev, KEY_CAMERA, data[3]);
         } else {
                 uvc_trace(UVC_TRACE_STATUS, "Stream %u error event %02x %02x "
                         "len %d.\n", data[1], data[2], data[3], len);
diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index 96ac88317b8e..ea3aafbbda44 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -91,9 +91,9 @@ MODULE_PARM_DESC(mpt_msi_enable_fc, " Enable MSI Support for FC \
                 controllers (default=0)");
 
 static int mpt_msi_enable_sas;
-module_param(mpt_msi_enable_sas, int, 1);
+module_param(mpt_msi_enable_sas, int, 0);
 MODULE_PARM_DESC(mpt_msi_enable_sas, " Enable MSI Support for SAS \
-                controllers (default=1)");
+                controllers (default=0)");
 
 
 static int mpt_channel_mapping;
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 8cff5f5e7f86..406da9a8d453 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -107,6 +107,7 @@ static const struct sdhci_pci_fixes sdhci_ene_714 = {
 
 static const struct sdhci_pci_fixes sdhci_cafe = {
         .quirks         = SDHCI_QUIRK_NO_SIMULT_VDD_AND_POWER |
+                          SDHCI_QUIRK_NO_BUSY_IRQ |
                           SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
 };
 
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index f52f3053ed92..accb592764ed 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1291,8 +1291,11 @@ static void sdhci_cmd_irq(struct sdhci_host *host, u32 intmask)
                 if (host->cmd->data)
                         DBG("Cannot wait for busy signal when also "
                                 "doing a data transfer");
-                else
+                else if (!(host->quirks & SDHCI_QUIRK_NO_BUSY_IRQ))
                         return;
+
+                /* The controller does not support the end-of-busy IRQ,
+                 * fall through and take the SDHCI_INT_RESPONSE */
         }
 
         if (intmask & SDHCI_INT_RESPONSE)
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index ebb83657e27a..43c37c68d07a 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -208,6 +208,8 @@ struct sdhci_host {
 #define SDHCI_QUIRK_BROKEN_TIMEOUT_VAL                  (1<<12)
 /* Controller has an issue with buffer bits for small transfers */
 #define SDHCI_QUIRK_BROKEN_SMALL_PIO                    (1<<13)
+/* Controller does not provide transfer-complete interrupt when not busy */
+#define SDHCI_QUIRK_NO_BUSY_IRQ                         (1<<14)
 
         int                     irq;            /* Device IRQ */
         void __iomem *          ioaddr;         /* Mapped address */
diff --git a/drivers/mtd/nand/orion_nand.c b/drivers/mtd/nand/orion_nand.c
index 917cf8d3ae95..c2dfd3ea353d 100644
--- a/drivers/mtd/nand/orion_nand.c
+++ b/drivers/mtd/nand/orion_nand.c
@@ -149,7 +149,7 @@ static int __devexit orion_nand_remove(struct platform_device *pdev)
 
 static struct platform_driver orion_nand_driver = {
         .probe          = orion_nand_probe,
-        .remove         = orion_nand_remove,
+        .remove         = __devexit_p(orion_nand_remove),
         .driver         = {
                 .name   = "orion_nand",
                 .owner  = THIS_MODULE,
diff --git a/drivers/net/arm/Makefile b/drivers/net/arm/Makefile
index c69c0cdba4a2..811a3ccd14c1 100644
--- a/drivers/net/arm/Makefile
+++ b/drivers/net/arm/Makefile
@@ -4,7 +4,7 @@
 #
 
 obj-$(CONFIG_ARM_AM79C961A)     += am79c961a.o
-obj-$(CONFIG_ARM_ETHERH)        += etherh.o ../8390.o
+obj-$(CONFIG_ARM_ETHERH)        += etherh.o
 obj-$(CONFIG_ARM_ETHER3)        += ether3.o
 obj-$(CONFIG_ARM_ETHER1)        += ether1.o
 obj-$(CONFIG_ARM_AT91_ETHER)    += at91_ether.o
diff --git a/drivers/net/arm/etherh.c b/drivers/net/arm/etherh.c
index 54b52e5b1821..f52f668c49bf 100644
--- a/drivers/net/arm/etherh.c
+++ b/drivers/net/arm/etherh.c
@@ -641,15 +641,15 @@ static const struct net_device_ops etherh_netdev_ops = {
         .ndo_open               = etherh_open,
         .ndo_stop               = etherh_close,
         .ndo_set_config         = etherh_set_config,
-        .ndo_start_xmit         = ei_start_xmit,
-        .ndo_tx_timeout         = ei_tx_timeout,
-        .ndo_get_stats          = ei_get_stats,
-        .ndo_set_multicast_list = ei_set_multicast_list,
+        .ndo_start_xmit         = __ei_start_xmit,
+        .ndo_tx_timeout         = __ei_tx_timeout,
+        .ndo_get_stats          = __ei_get_stats,
+        .ndo_set_multicast_list = __ei_set_multicast_list,
         .ndo_validate_addr      = eth_validate_addr,
         .ndo_set_mac_address    = eth_mac_addr,
         .ndo_change_mtu         = eth_change_mtu,
 #ifdef CONFIG_NET_POLL_CONTROLLER
-        .ndo_poll_controller    = ei_poll,
+        .ndo_poll_controller    = __ei_poll,
 #endif
 };
 
diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index c38512ebcea6..dc5f051005fa 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -1264,8 +1264,14 @@ static void b44_clear_stats(struct b44 *bp)
 static void b44_chip_reset(struct b44 *bp, int reset_kind)
 {
         struct ssb_device *sdev = bp->sdev;
+        bool was_enabled;
 
-        if (ssb_device_is_enabled(bp->sdev)) {
+        was_enabled = ssb_device_is_enabled(bp->sdev);
+
+        ssb_device_enable(bp->sdev, 0);
+        ssb_pcicore_dev_irqvecs_enable(&sdev->bus->pcicore, sdev);
+
+        if (was_enabled) {
                 bw32(bp, B44_RCV_LAZY, 0);
                 bw32(bp, B44_ENET_CTRL, ENET_CTRL_DISABLE);
                 b44_wait_bit(bp, B44_ENET_CTRL, ENET_CTRL_DISABLE, 200, 1);
@@ -1277,10 +1283,8 @@ static void b44_chip_reset(struct b44 *bp, int reset_kind)
                 }
                 bw32(bp, B44_DMARX_CTRL, 0);
                 bp->rx_prod = bp->rx_cons = 0;
-        } else
-                ssb_pcicore_dev_irqvecs_enable(&sdev->bus->pcicore, sdev);
+        }
 
-        ssb_device_enable(bp->sdev, 0);
         b44_clear_stats(bp);
 
         /*
@@ -2236,6 +2240,7 @@ static void __devexit b44_remove_one(struct ssb_device *sdev)
         struct net_device *dev = ssb_get_drvdata(sdev);
 
         unregister_netdev(dev);
+        ssb_device_disable(sdev, 0);
         ssb_bus_may_powerdown(sdev->bus);
         free_netdev(dev);
         ssb_pcihost_set_power_state(sdev, PCI_D3hot);
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index 9b12a13a640f..9831b3f408aa 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -1284,7 +1284,7 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
         spin_lock_irqsave(&priv->txlock, flags);
 
         /* check if there is space to queue this packet */
-        if (nr_frags > priv->num_txbdfree) {
+        if ((nr_frags+1) > priv->num_txbdfree) {
                 /* no space, stop the queue */
                 netif_stop_queue(dev);
                 dev->stats.tx_fifo_errors++;
diff --git a/drivers/net/hp-plus.c b/drivers/net/hp-plus.c
index 5e070f446635..0486cbe01adb 100644
--- a/drivers/net/hp-plus.c
+++ b/drivers/net/hp-plus.c
@@ -467,7 +467,7 @@ init_module(void)
                         if (this_dev != 0) break; /* only autoprobe 1st one */
                         printk(KERN_NOTICE "hp-plus.c: Presently autoprobing (not recommended) for a single card.\n");
                 }
-                dev = alloc_ei_netdev();
+                dev = alloc_eip_netdev();
                 if (!dev)
                         break;
                 dev->irq = irq[this_dev];
diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c
index 9f33e442f403..13087782ac40 100644
--- a/drivers/net/netxen/netxen_nic_main.c
+++ b/drivers/net/netxen/netxen_nic_main.c
@@ -588,7 +588,12 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
                 adapter->pci_mem_read = netxen_nic_pci_mem_read_2M;
                 adapter->pci_mem_write = netxen_nic_pci_mem_write_2M;
 
-                mem_ptr0 = ioremap(mem_base, mem_len);
+                mem_ptr0 = pci_ioremap_bar(pdev, 0);
+                if (mem_ptr0 == NULL) {
+                        dev_err(&pdev->dev, "failed to map PCI bar 0\n");
+                        return -EIO;
+                }
+
                 pci_len0 = mem_len;
                 first_page_group_start = 0;
                 first_page_group_end   = 0;
@@ -795,9 +800,12 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
          * See if the firmware gave us a virtual-physical port mapping.
          */
         adapter->physical_port = adapter->portnum;
-        i = adapter->pci_read_normalize(adapter, CRB_V2P(adapter->portnum));
-        if (i != 0x55555555)
-                adapter->physical_port = i;
+        if (adapter->fw_major < 4) {
+                i = adapter->pci_read_normalize(adapter,
+                                CRB_V2P(adapter->portnum));
+                if (i != 0x55555555)
+                        adapter->physical_port = i;
+        }
 
         adapter->flags &= ~(NETXEN_NIC_MSI_ENABLED | NETXEN_NIC_MSIX_ENABLED);
 
diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 0771eb6fc6eb..b3473401c83a 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -81,9 +81,9 @@ static const int multicast_filter_limit = 32;
 #define RTL8169_TX_TIMEOUT      (6*HZ)
 #define RTL8169_PHY_TIMEOUT     (10*HZ)
 
-#define RTL_EEPROM_SIG          cpu_to_le32(0x8129)
-#define RTL_EEPROM_SIG_MASK     cpu_to_le32(0xffff)
+#define RTL_EEPROM_SIG          0x8129
 #define RTL_EEPROM_SIG_ADDR     0x0000
+#define RTL_EEPROM_MAC_ADDR     0x0007
 
 /* write/read MMIO register */
 #define RTL_W8(reg, val8)       writeb ((val8), ioaddr + (reg))
@@ -293,6 +293,11 @@ enum rtl_register_content {
         /* Cfg9346Bits */
         Cfg9346_Lock    = 0x00,
         Cfg9346_Unlock  = 0xc0,
+        Cfg9346_Program = 0x80,         /* Programming mode */
+        Cfg9346_EECS    = 0x08,         /* Chip select */
+        Cfg9346_EESK    = 0x04,         /* Serial data clock */
+        Cfg9346_EEDI    = 0x02,         /* Data input */
+        Cfg9346_EEDO    = 0x01,         /* Data output */
 
         /* rx_mode_bits */
         AcceptErr       = 0x20,
@@ -305,6 +310,7 @@ enum rtl_register_content {
         /* RxConfigBits */
         RxCfgFIFOShift  = 13,
         RxCfgDMAShift   =  8,
+        RxCfg9356SEL    =  6,           /* EEPROM type: 0 = 9346, 1 = 9356 */
 
         /* TxConfigBits */
         TxInterFrameGapShift = 24,
@@ -1963,6 +1969,108 @@ static const struct net_device_ops rtl8169_netdev_ops = {
 
 };
 
+/* Delay between EEPROM clock transitions. Force out buffered PCI writes. */
+#define RTL_EEPROM_DELAY()      RTL_R8(Cfg9346)
+#define RTL_EEPROM_READ_CMD     6
+
+/* read 16bit word stored in EEPROM. EEPROM is addressed by words. */
+static u16 rtl_eeprom_read(void __iomem *ioaddr, int addr)
+{
+        u16 result = 0;
+        int cmd, cmd_len, i;
+
+        /* check for EEPROM address size (in bits) */
+        if (RTL_R32(RxConfig) & (1 << RxCfg9356SEL)) {
+                /* EEPROM is 93C56 */
+                cmd_len = 3 + 8; /* 3 bits for command id and 8 for address */
+                cmd = (RTL_EEPROM_READ_CMD << 8) | (addr & 0xff);
+        } else {
+                /* EEPROM is 93C46 */
+                cmd_len = 3 + 6; /* 3 bits for command id and 6 for address */
+                cmd = (RTL_EEPROM_READ_CMD << 6) | (addr & 0x3f);
+        }
+
+        /* enter programming mode */
+        RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS);
+        RTL_EEPROM_DELAY();
+
+        /* write command and requested address */
+        while (cmd_len--) {
+                u8 x = Cfg9346_Program | Cfg9346_EECS;
+
+                x |= (cmd & (1 << cmd_len)) ? Cfg9346_EEDI : 0;
+
+                /* write a bit */
+                RTL_W8(Cfg9346, x);
+                RTL_EEPROM_DELAY();
+
+                /* raise clock */
+                RTL_W8(Cfg9346, x | Cfg9346_EESK);
+                RTL_EEPROM_DELAY();
+        }
+
+        /* lower clock */
+        RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS);
+        RTL_EEPROM_DELAY();
+
+        /* read back 16bit value */
+        for (i = 16; i > 0; i--) {
+                /* raise clock */
+                RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS | Cfg9346_EESK);
+                RTL_EEPROM_DELAY();
+
+                result <<= 1;
+                result |= (RTL_R8(Cfg9346) & Cfg9346_EEDO) ? 1 : 0;
+
+                /* lower clock */
+                RTL_W8(Cfg9346, Cfg9346_Program | Cfg9346_EECS);
+                RTL_EEPROM_DELAY();
+        }
+
+        RTL_W8(Cfg9346, Cfg9346_Program);
+        /* leave programming mode */
+        RTL_W8(Cfg9346, Cfg9346_Lock);
+
+        return result;
+}
+
+static void rtl_init_mac_address(struct rtl8169_private *tp,
+                                 void __iomem *ioaddr)
+{
+        struct pci_dev *pdev = tp->pci_dev;
+        u16 x;
+        u8 mac[8];
+
+        /* read EEPROM signature */
+        x = rtl_eeprom_read(ioaddr, RTL_EEPROM_SIG_ADDR);
+
+        if (x != RTL_EEPROM_SIG) {
+                dev_info(&pdev->dev, "Missing EEPROM signature: %04x\n", x);
+                return;
+        }
+
+        /* read MAC address */
+        x = rtl_eeprom_read(ioaddr, RTL_EEPROM_MAC_ADDR);
+        mac[0] = x & 0xff;
+        mac[1] = x >> 8;
+        x = rtl_eeprom_read(ioaddr, RTL_EEPROM_MAC_ADDR + 1);
+        mac[2] = x & 0xff;
+        mac[3] = x >> 8;
+        x = rtl_eeprom_read(ioaddr, RTL_EEPROM_MAC_ADDR + 2);
+        mac[4] = x & 0xff;
+        mac[5] = x >> 8;
+
+        if (netif_msg_probe(tp)) {
+                DECLARE_MAC_BUF(buf);
+
+                dev_info(&pdev->dev, "MAC address found in EEPROM: %s\n",
+                         print_mac(buf, mac));
+        }
+
+        if (is_valid_ether_addr(mac))
+                rtl_rar_set(tp, mac);
+}
+
 static int __devinit
 rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
@@ -2141,6 +2249,8 @@ rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
         tp->mmio_addr = ioaddr;
 
+        rtl_init_mac_address(tp, ioaddr);
+
         /* Get MAC address */
         for (i = 0; i < MAC_ADDR_LEN; i++)
                 dev->dev_addr[i] = RTL_R8(MAC0 + i);
diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c
index e009481c606c..396f821b5ff0 100644
--- a/drivers/net/usb/asix.c
+++ b/drivers/net/usb/asix.c
@@ -1451,6 +1451,14 @@ static const struct usb_device_id	products [] = {
         // Cables-to-Go USB Ethernet Adapter
         USB_DEVICE(0x0b95, 0x772a),
         .driver_info = (unsigned long) &ax88772_info,
+}, {
+        // ABOCOM for pci
+        USB_DEVICE(0x14ea, 0xab11),
+        .driver_info = (unsigned long) &ax88178_info,
+}, {
+        // ASIX 88772a
+        USB_DEVICE(0x0db0, 0xa877),
+        .driver_info = (unsigned long) &ax88772_info,
 },
         { },            // END
 };
diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c
index 0e061dfea78d..55e8ecc3a9e5 100644
--- a/drivers/net/usb/cdc_ether.c
+++ b/drivers/net/usb/cdc_ether.c
@@ -559,6 +559,11 @@ static const struct usb_device_id	products [] = {
         USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ETHERNET,
                         USB_CDC_PROTO_NONE),
         .driver_info = (unsigned long) &cdc_info,
+}, {
+        /* Ericsson F3507g */
+        USB_DEVICE_AND_INTERFACE_INFO(0x0bdb, 0x1900, USB_CLASS_COMM,
+                        USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+        .driver_info = (unsigned long) &cdc_info,
 },
         { },            // END
 };
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index aa3149078888..c32284ff3f54 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -723,8 +723,8 @@ u32 usbnet_get_link (struct net_device *net)
         if (dev->mii.mdio_read)
                 return mii_link_ok(&dev->mii);
 
-        /* Otherwise, say we're up (to avoid breaking scripts) */
-        return 1;
+        /* Otherwise, dtrt for drivers calling netif_carrier_{on,off} */
+        return ethtool_op_get_link(net);
 }
 EXPORT_SYMBOL_GPL(usbnet_get_link);
 
diff --git a/drivers/net/usb/zaurus.c b/drivers/net/usb/zaurus.c
index e24f7b3ace4b..04882c8f9bf1 100644
--- a/drivers/net/usb/zaurus.c
+++ b/drivers/net/usb/zaurus.c
@@ -341,6 +341,11 @@ static const struct usb_device_id	products [] = {
         USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_MDLM,
                         USB_CDC_PROTO_NONE),
         .driver_info = (unsigned long) &bogus_mdlm_info,
+}, {
+        /* Motorola MOTOMAGX phones */
+        USB_DEVICE_AND_INTERFACE_INFO(0x22b8, 0x6425, USB_CLASS_COMM,
+                        USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+        .driver_info = (unsigned long) &bogus_mdlm_info,
 },
 
 /* Olympus has some models with a Zaurus-compatible option.
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 108bbbeacfb6..124fe75b8a8a 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -239,6 +239,16 @@ static int veth_open(struct net_device *dev)
         return 0;
 }
 
+static int veth_close(struct net_device *dev)
+{
+        struct veth_priv *priv = netdev_priv(dev);
+
+        netif_carrier_off(dev);
+        netif_carrier_off(priv->peer);
+
+        return 0;
+}
+
 static int veth_dev_init(struct net_device *dev)
 {
         struct veth_net_stats *stats;
@@ -265,6 +275,7 @@ static void veth_dev_free(struct net_device *dev)
 static const struct net_device_ops veth_netdev_ops = {
         .ndo_init            = veth_dev_init,
         .ndo_open            = veth_open,
+        .ndo_stop            = veth_close,
         .ndo_start_xmit      = veth_xmit,
         .ndo_get_stats       = veth_get_stats,
         .ndo_set_mac_address = eth_mac_addr,
@@ -280,44 +291,6 @@ static void veth_setup(struct net_device *dev)
         dev->destructor = veth_dev_free;
 }
 
-static void veth_change_state(struct net_device *dev)
-{
-        struct net_device *peer;
-        struct veth_priv *priv;
-
-        priv = netdev_priv(dev);
-        peer = priv->peer;
-
-        if (netif_carrier_ok(peer)) {
-                if (!netif_carrier_ok(dev))
-                        netif_carrier_on(dev);
-        } else {
-                if (netif_carrier_ok(dev))
-                        netif_carrier_off(dev);
-        }
-}
-
-static int veth_device_event(struct notifier_block *unused,
-                             unsigned long event, void *ptr)
-{
-        struct net_device *dev = ptr;
-
-        if (dev->netdev_ops->ndo_open != veth_open)
-                goto out;
-
-        switch (event) {
-        case NETDEV_CHANGE:
-                veth_change_state(dev);
-                break;
-        }
-out:
-        return NOTIFY_DONE;
-}
-
-static struct notifier_block veth_notifier_block __read_mostly = {
-        .notifier_call  = veth_device_event,
-};
-
 /*
  * netlink interface
  */
@@ -468,14 +441,12 @@ static struct rtnl_link_ops veth_link_ops = {
 
 static __init int veth_init(void)
 {
-        register_netdevice_notifier(&veth_notifier_block);
         return rtnl_link_register(&veth_link_ops);
 }
 
 static __exit void veth_exit(void)
 {
         rtnl_link_unregister(&veth_link_ops);
-        unregister_netdevice_notifier(&veth_notifier_block);
 }
 
 module_init(veth_init);
diff --git a/drivers/net/wireless/ath9k/main.c b/drivers/net/wireless/ath9k/main.c
index 727f067aca4f..0e80990d8e84 100644
--- a/drivers/net/wireless/ath9k/main.c
+++ b/drivers/net/wireless/ath9k/main.c
@@ -1538,6 +1538,7 @@ bad2:
 bad:
         if (ah)
                 ath9k_hw_detach(ah);
+        ath9k_exit_debug(sc);
 
         return error;
 }
@@ -1545,7 +1546,7 @@ bad:
 static int ath_attach(u16 devid, struct ath_softc *sc)
 {
         struct ieee80211_hw *hw = sc->hw;
-        int error = 0;
+        int error = 0, i;
 
         DPRINTF(sc, ATH_DBG_CONFIG, "Attach ATH hw\n");
 
@@ -1589,11 +1590,11 @@ static int ath_attach(u16 devid, struct ath_softc *sc)
         /* initialize tx/rx engine */
         error = ath_tx_init(sc, ATH_TXBUF);
         if (error != 0)
-                goto detach;
+                goto error_attach;
 
         error = ath_rx_init(sc, ATH_RXBUF);
         if (error != 0)
-                goto detach;
+                goto error_attach;
 
 #if defined(CONFIG_RFKILL) || defined(CONFIG_RFKILL_MODULE)
         /* Initialze h/w Rfkill */
@@ -1601,8 +1602,9 @@ static int ath_attach(u16 devid, struct ath_softc *sc)
                 INIT_DELAYED_WORK(&sc->rf_kill.rfkill_poll, ath_rfkill_poll);
 
         /* Initialize s/w rfkill */
-        if (ath_init_sw_rfkill(sc))
-                goto detach;
+        error = ath_init_sw_rfkill(sc);
+        if (error)
+                goto error_attach;
 #endif
 
         error = ieee80211_register_hw(hw);
@@ -1611,8 +1613,16 @@ static int ath_attach(u16 devid, struct ath_softc *sc)
         ath_init_leds(sc);
 
         return 0;
-detach:
-        ath_detach(sc);
+
+error_attach:
+        /* cleanup tx queues */
+        for (i = 0; i < ATH9K_NUM_TX_QUEUES; i++)
+                if (ATH_TXQ_SETUP(sc, i))
+                        ath_tx_cleanupq(sc, &sc->tx.txq[i]);
+
+        ath9k_hw_detach(sc->sc_ah);
+        ath9k_exit_debug(sc);
+
         return error;
 }
 
diff --git a/drivers/net/wireless/libertas/ethtool.c b/drivers/net/wireless/libertas/ethtool.c
index 61d2f50470c8..b118a35ec605 100644
--- a/drivers/net/wireless/libertas/ethtool.c
+++ b/drivers/net/wireless/libertas/ethtool.c
@@ -23,7 +23,7 @@ static const char * mesh_stat_strings[]= {
 static void lbs_ethtool_get_drvinfo(struct net_device *dev,
                                          struct ethtool_drvinfo *info)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         snprintf(info->fw_version, 32, "%u.%u.%u.p%u",
                 priv->fwrelease >> 24 & 0xff,
@@ -47,7 +47,7 @@ static int lbs_ethtool_get_eeprom_len(struct net_device *dev)
 static int lbs_ethtool_get_eeprom(struct net_device *dev,
                                   struct ethtool_eeprom *eeprom, u8 * bytes)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct cmd_ds_802_11_eeprom_access cmd;
         int ret;
 
@@ -76,7 +76,7 @@ out:
 static void lbs_ethtool_get_stats(struct net_device *dev,
                                   struct ethtool_stats *stats, uint64_t *data)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct cmd_ds_mesh_access mesh_access;
         int ret;
 
@@ -113,7 +113,7 @@ static void lbs_ethtool_get_stats(struct net_device *dev,
 
 static int lbs_ethtool_get_sset_count(struct net_device *dev, int sset)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         if (sset == ETH_SS_STATS && dev == priv->mesh_dev)
                 return MESH_STATS_NUM;
@@ -143,7 +143,7 @@ static void lbs_ethtool_get_strings(struct net_device *dev,
 static void lbs_ethtool_get_wol(struct net_device *dev,
                                 struct ethtool_wolinfo *wol)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         if (priv->wol_criteria == 0xffffffff) {
                 /* Interface driver didn't configure wake */
@@ -166,7 +166,7 @@ static void lbs_ethtool_get_wol(struct net_device *dev,
 static int lbs_ethtool_set_wol(struct net_device *dev,
                                struct ethtool_wolinfo *wol)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         uint32_t criteria = 0;
 
         if (priv->wol_criteria == 0xffffffff && wol->wolopts)
diff --git a/drivers/net/wireless/libertas/if_usb.c b/drivers/net/wireless/libertas/if_usb.c
index 2fc637ad85c7..ea3dc038be76 100644
--- a/drivers/net/wireless/libertas/if_usb.c
+++ b/drivers/net/wireless/libertas/if_usb.c
@@ -59,7 +59,7 @@ static int if_usb_reset_device(struct if_usb_card *cardp);
 static ssize_t if_usb_firmware_set(struct device *dev,
                 struct device_attribute *attr, const char *buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct if_usb_card *cardp = priv->card;
         char fwname[FIRMWARE_NAME_MAX];
         int ret;
@@ -86,7 +86,7 @@ static DEVICE_ATTR(lbs_flash_fw, 0200, NULL, if_usb_firmware_set);
 static ssize_t if_usb_boot2_set(struct device *dev,
                 struct device_attribute *attr, const char *buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct if_usb_card *cardp = priv->card;
         char fwname[FIRMWARE_NAME_MAX];
         int ret;
diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c
index 4e0007d20030..f76623e0ff9a 100644
--- a/drivers/net/wireless/libertas/main.c
+++ b/drivers/net/wireless/libertas/main.c
@@ -222,7 +222,7 @@ u8 lbs_data_rate_to_fw_index(u32 rate)
 static ssize_t lbs_anycast_get(struct device *dev,
                 struct device_attribute *attr, char * buf)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_access mesh_access;
         int ret;
 
@@ -241,7 +241,7 @@ static ssize_t lbs_anycast_get(struct device *dev,
 static ssize_t lbs_anycast_set(struct device *dev,
                 struct device_attribute *attr, const char * buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_access mesh_access;
         uint32_t datum;
         int ret;
@@ -263,7 +263,7 @@ static ssize_t lbs_anycast_set(struct device *dev,
 static ssize_t lbs_prb_rsp_limit_get(struct device *dev,
                 struct device_attribute *attr, char *buf)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_access mesh_access;
         int ret;
         u32 retry_limit;
@@ -286,7 +286,7 @@ static ssize_t lbs_prb_rsp_limit_get(struct device *dev,
 static ssize_t lbs_prb_rsp_limit_set(struct device *dev,
                 struct device_attribute *attr, const char *buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_access mesh_access;
         int ret;
         unsigned long retry_limit;
@@ -321,7 +321,7 @@ static void lbs_remove_mesh(struct lbs_private *priv);
 static ssize_t lbs_rtap_get(struct device *dev,
                 struct device_attribute *attr, char * buf)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         return snprintf(buf, 5, "0x%X\n", priv->monitormode);
 }
 
@@ -332,7 +332,7 @@ static ssize_t lbs_rtap_set(struct device *dev,
                 struct device_attribute *attr, const char * buf, size_t count)
 {
         int monitor_mode;
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
 
         sscanf(buf, "%x", &monitor_mode);
         if (monitor_mode) {
@@ -383,7 +383,7 @@ static DEVICE_ATTR(lbs_rtap, 0644, lbs_rtap_get, lbs_rtap_set );
 static ssize_t lbs_mesh_get(struct device *dev,
                 struct device_attribute *attr, char * buf)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         return snprintf(buf, 5, "0x%X\n", !!priv->mesh_dev);
 }
 
@@ -393,7 +393,7 @@ static ssize_t lbs_mesh_get(struct device *dev,
 static ssize_t lbs_mesh_set(struct device *dev,
                 struct device_attribute *attr, const char * buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         int enable;
         int ret, action = CMD_ACT_MESH_CONFIG_STOP;
 
@@ -452,7 +452,7 @@ static struct attribute_group lbs_mesh_attr_group = {
  */
 static int lbs_dev_open(struct net_device *dev)
 {
-        struct lbs_private *priv = netdev_priv(dev) ;
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
 
         lbs_deb_enter(LBS_DEB_NET);
@@ -521,7 +521,7 @@ static int lbs_mesh_stop(struct net_device *dev)
  */
 static int lbs_eth_stop(struct net_device *dev)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_NET);
 
@@ -538,7 +538,7 @@ static int lbs_eth_stop(struct net_device *dev)
 
 static void lbs_tx_timeout(struct net_device *dev)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_TX);
 
@@ -590,7 +590,7 @@ EXPORT_SYMBOL_GPL(lbs_host_to_card_done);
  */
 static struct net_device_stats *lbs_get_stats(struct net_device *dev)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_NET);
         return &priv->stats;
@@ -599,7 +599,7 @@ static struct net_device_stats *lbs_get_stats(struct net_device *dev)
 static int lbs_set_mac_address(struct net_device *dev, void *addr)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct sockaddr *phwaddr = addr;
         struct cmd_ds_802_11_mac_address cmd;
 
@@ -732,7 +732,7 @@ static void lbs_set_mcast_worker(struct work_struct *work)
 
 static void lbs_set_multicast_list(struct net_device *dev)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         schedule_work(&priv->mcast_work);
 }
@@ -748,7 +748,7 @@ static void lbs_set_multicast_list(struct net_device *dev)
 static int lbs_thread(void *data)
 {
         struct net_device *dev = data;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         wait_queue_t wait;
 
         lbs_deb_enter(LBS_DEB_THREAD);
@@ -1184,6 +1184,7 @@ struct lbs_private *lbs_add_card(void *card, struct device *dmdev)
                 goto done;
         }
         priv = netdev_priv(dev);
+        dev->ml_priv = priv;
 
         if (lbs_init_adapter(priv)) {
                 lbs_pr_err("failed to initialize adapter structure.\n");
diff --git a/drivers/net/wireless/libertas/persistcfg.c b/drivers/net/wireless/libertas/persistcfg.c
index d42b7a5a1b3f..18fe29faf99b 100644
--- a/drivers/net/wireless/libertas/persistcfg.c
+++ b/drivers/net/wireless/libertas/persistcfg.c
@@ -18,7 +18,7 @@
 static int mesh_get_default_parameters(struct device *dev,
                                        struct mrvl_mesh_defaults *defs)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_config cmd;
         int ret;
 
@@ -57,7 +57,7 @@ static ssize_t bootflag_get(struct device *dev,
 static ssize_t bootflag_set(struct device *dev, struct device_attribute *attr,
                             const char *buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_config cmd;
         uint32_t datum;
         int ret;
@@ -100,7 +100,7 @@ static ssize_t boottime_get(struct device *dev,
 static ssize_t boottime_set(struct device *dev,
                 struct device_attribute *attr, const char *buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_config cmd;
         uint32_t datum;
         int ret;
@@ -152,7 +152,7 @@ static ssize_t channel_get(struct device *dev,
 static ssize_t channel_set(struct device *dev, struct device_attribute *attr,
                            const char *buf, size_t count)
 {
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         struct cmd_ds_mesh_config cmd;
         uint32_t datum;
         int ret;
@@ -210,7 +210,7 @@ static ssize_t mesh_id_set(struct device *dev, struct device_attribute *attr,
         struct cmd_ds_mesh_config cmd;
         struct mrvl_mesh_defaults defs;
         struct mrvl_meshie *ie;
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         int len;
         int ret;
 
@@ -269,7 +269,7 @@ static ssize_t protocol_id_set(struct device *dev,
         struct cmd_ds_mesh_config cmd;
         struct mrvl_mesh_defaults defs;
         struct mrvl_meshie *ie;
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         uint32_t datum;
         int ret;
 
@@ -323,7 +323,7 @@ static ssize_t metric_id_set(struct device *dev, struct device_attribute *attr,
         struct cmd_ds_mesh_config cmd;
         struct mrvl_mesh_defaults defs;
         struct mrvl_meshie *ie;
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         uint32_t datum;
         int ret;
 
@@ -377,7 +377,7 @@ static ssize_t capability_set(struct device *dev, struct device_attribute *attr,
         struct cmd_ds_mesh_config cmd;
         struct mrvl_mesh_defaults defs;
         struct mrvl_meshie *ie;
-        struct lbs_private *priv = netdev_priv(to_net_dev(dev));
+        struct lbs_private *priv = to_net_dev(dev)->ml_priv;
         uint32_t datum;
         int ret;
 
diff --git a/drivers/net/wireless/libertas/scan.c b/drivers/net/wireless/libertas/scan.c
index 57f6c12cda20..9014950f4328 100644
--- a/drivers/net/wireless/libertas/scan.c
+++ b/drivers/net/wireless/libertas/scan.c
@@ -945,7 +945,7 @@ int lbs_set_scan(struct net_device *dev, struct iw_request_info *info,
                  union iwreq_data *wrqu, char *extra)
 {
         DECLARE_SSID_BUF(ssid);
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -1008,7 +1008,7 @@ int lbs_get_scan(struct net_device *dev, struct iw_request_info *info,
                  struct iw_point *dwrq, char *extra)
 {
 #define SCAN_ITEM_SIZE 128
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int err = 0;
         char *ev = extra;
         char *stop = ev + dwrq->length;
diff --git a/drivers/net/wireless/libertas/tx.c b/drivers/net/wireless/libertas/tx.c
index dac462641170..68bec31ae03b 100644
--- a/drivers/net/wireless/libertas/tx.c
+++ b/drivers/net/wireless/libertas/tx.c
@@ -60,7 +60,7 @@ static u32 convert_radiotap_rate_to_mv(u8 rate)
 int lbs_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
         unsigned long flags;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct txpd *txpd;
         char *p802x_hdr;
         uint16_t pkt_len;
diff --git a/drivers/net/wireless/libertas/wext.c b/drivers/net/wireless/libertas/wext.c
index c6102e08179e..f16d136ab4bb 100644
--- a/drivers/net/wireless/libertas/wext.c
+++ b/drivers/net/wireless/libertas/wext.c
@@ -163,7 +163,7 @@ static int lbs_get_name(struct net_device *dev, struct iw_request_info *info,
 static int lbs_get_freq(struct net_device *dev, struct iw_request_info *info,
                          struct iw_freq *fwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct chan_freq_power *cfp;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -189,7 +189,7 @@ static int lbs_get_freq(struct net_device *dev, struct iw_request_info *info,
 static int lbs_get_wap(struct net_device *dev, struct iw_request_info *info,
                         struct sockaddr *awrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -207,7 +207,7 @@ static int lbs_get_wap(struct net_device *dev, struct iw_request_info *info,
 static int lbs_set_nick(struct net_device *dev, struct iw_request_info *info,
                          struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -231,7 +231,7 @@ static int lbs_set_nick(struct net_device *dev, struct iw_request_info *info,
 static int lbs_get_nick(struct net_device *dev, struct iw_request_info *info,
                          struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -248,7 +248,7 @@ static int lbs_get_nick(struct net_device *dev, struct iw_request_info *info,
 static int mesh_get_nick(struct net_device *dev, struct iw_request_info *info,
                          struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -273,7 +273,7 @@ static int lbs_set_rts(struct net_device *dev, struct iw_request_info *info,
                         struct iw_param *vwrq, char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         u32 val = vwrq->value;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -293,7 +293,7 @@ static int lbs_set_rts(struct net_device *dev, struct iw_request_info *info,
 static int lbs_get_rts(struct net_device *dev, struct iw_request_info *info,
                         struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         u16 val = 0;
 
@@ -315,7 +315,7 @@ out:
 static int lbs_set_frag(struct net_device *dev, struct iw_request_info *info,
                          struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         u32 val = vwrq->value;
 
@@ -336,7 +336,7 @@ static int lbs_set_frag(struct net_device *dev, struct iw_request_info *info,
 static int lbs_get_frag(struct net_device *dev, struct iw_request_info *info,
                          struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         u16 val = 0;
 
@@ -359,7 +359,7 @@ out:
 static int lbs_get_mode(struct net_device *dev,
                          struct iw_request_info *info, u32 * uwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -385,7 +385,7 @@ static int lbs_get_txpow(struct net_device *dev,
                           struct iw_request_info *info,
                           struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         s16 curlevel = 0;
         int ret = 0;
 
@@ -418,7 +418,7 @@ out:
 static int lbs_set_retry(struct net_device *dev, struct iw_request_info *info,
                           struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         u16 slimit = 0, llimit = 0;
 
@@ -466,7 +466,7 @@ out:
 static int lbs_get_retry(struct net_device *dev, struct iw_request_info *info,
                           struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         u16 val = 0;
 
@@ -542,7 +542,7 @@ static int lbs_get_range(struct net_device *dev, struct iw_request_info *info,
                           struct iw_point *dwrq, char *extra)
 {
         int i, j;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct iw_range *range = (struct iw_range *)extra;
         struct chan_freq_power *cfp;
         u8 rates[MAX_RATES + 1];
@@ -708,7 +708,7 @@ out:
 static int lbs_set_power(struct net_device *dev, struct iw_request_info *info,
                           struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -758,7 +758,7 @@ static int lbs_set_power(struct net_device *dev, struct iw_request_info *info,
 static int lbs_get_power(struct net_device *dev, struct iw_request_info *info,
                           struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -781,7 +781,7 @@ static struct iw_statistics *lbs_get_wireless_stats(struct net_device *dev)
                 EXCELLENT = 95,
                 PERFECT = 100
         };
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         u32 rssi_qual;
         u32 tx_qual;
         u32 quality = 0;
@@ -886,7 +886,7 @@ static int lbs_set_freq(struct net_device *dev, struct iw_request_info *info,
                   struct iw_freq *fwrq, char *extra)
 {
         int ret = -EINVAL;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct chan_freq_power *cfp;
         struct assoc_request * assoc_req;
 
@@ -943,7 +943,7 @@ static int lbs_mesh_set_freq(struct net_device *dev,
                              struct iw_request_info *info,
                              struct iw_freq *fwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct chan_freq_power *cfp;
         int ret = -EINVAL;
 
@@ -994,7 +994,7 @@ out:
 static int lbs_set_rate(struct net_device *dev, struct iw_request_info *info,
                   struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         u8 new_rate = 0;
         int ret = -EINVAL;
         u8 rates[MAX_RATES + 1];
@@ -1054,7 +1054,7 @@ out:
 static int lbs_get_rate(struct net_device *dev, struct iw_request_info *info,
                   struct iw_param *vwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1079,7 +1079,7 @@ static int lbs_set_mode(struct net_device *dev,
                   struct iw_request_info *info, u32 * uwrq, char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct assoc_request * assoc_req;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -1124,7 +1124,7 @@ static int lbs_get_encode(struct net_device *dev,
                            struct iw_request_info *info,
                            struct iw_point *dwrq, u8 * extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int index = (dwrq->flags & IW_ENCODE_INDEX) - 1;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -1319,7 +1319,7 @@ static int lbs_set_encode(struct net_device *dev,
                     struct iw_point *dwrq, char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct assoc_request * assoc_req;
         u16 is_default = 0, index = 0, set_tx_key = 0;
 
@@ -1395,7 +1395,7 @@ static int lbs_get_encodeext(struct net_device *dev,
                               char *extra)
 {
         int ret = -EINVAL;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
         int index, max_key_len;
 
@@ -1501,7 +1501,7 @@ static int lbs_set_encodeext(struct net_device *dev,
                               char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
         int alg = ext->alg;
         struct assoc_request * assoc_req;
@@ -1639,7 +1639,7 @@ static int lbs_set_genie(struct net_device *dev,
                           struct iw_point *dwrq,
                           char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         struct assoc_request * assoc_req;
 
@@ -1685,7 +1685,7 @@ static int lbs_get_genie(struct net_device *dev,
                           char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1713,7 +1713,7 @@ static int lbs_set_auth(struct net_device *dev,
                          struct iw_param *dwrq,
                          char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct assoc_request * assoc_req;
         int ret = 0;
         int updated = 0;
@@ -1816,7 +1816,7 @@ static int lbs_get_auth(struct net_device *dev,
                          char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1857,7 +1857,7 @@ static int lbs_set_txpow(struct net_device *dev, struct iw_request_info *info,
                    struct iw_param *vwrq, char *extra)
 {
         int ret = 0;
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         s16 dbm = (s16) vwrq->value;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -1936,7 +1936,7 @@ out:
 static int lbs_get_essid(struct net_device *dev, struct iw_request_info *info,
                    struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -1971,7 +1971,7 @@ static int lbs_get_essid(struct net_device *dev, struct iw_request_info *info,
 static int lbs_set_essid(struct net_device *dev, struct iw_request_info *info,
                    struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
         u8 ssid[IW_ESSID_MAX_SIZE];
         u8 ssid_len = 0;
@@ -2040,7 +2040,7 @@ static int lbs_mesh_get_essid(struct net_device *dev,
                               struct iw_request_info *info,
                               struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
 
         lbs_deb_enter(LBS_DEB_WEXT);
 
@@ -2058,7 +2058,7 @@ static int lbs_mesh_set_essid(struct net_device *dev,
                               struct iw_request_info *info,
                               struct iw_point *dwrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         int ret = 0;
 
         lbs_deb_enter(LBS_DEB_WEXT);
@@ -2102,7 +2102,7 @@ static int lbs_mesh_set_essid(struct net_device *dev,
 static int lbs_set_wap(struct net_device *dev, struct iw_request_info *info,
                  struct sockaddr *awrq, char *extra)
 {
-        struct lbs_private *priv = netdev_priv(dev);
+        struct lbs_private *priv = dev->ml_priv;
         struct assoc_request * assoc_req;
         int ret = 0;
 
diff --git a/drivers/net/wireless/orinoco/orinoco.c b/drivers/net/wireless/orinoco/orinoco.c
index 45a04faa7818..067d1a9c728b 100644
--- a/drivers/net/wireless/orinoco/orinoco.c
+++ b/drivers/net/wireless/orinoco/orinoco.c
@@ -3157,8 +3157,20 @@ static int orinoco_pm_notifier(struct notifier_block *notifier,
 
         return NOTIFY_DONE;
 }
+
+static void orinoco_register_pm_notifier(struct orinoco_private *priv)
+{
+        priv->pm_notifier.notifier_call = orinoco_pm_notifier;
+        register_pm_notifier(&priv->pm_notifier);
+}
+
+static void orinoco_unregister_pm_notifier(struct orinoco_private *priv)
+{
+        unregister_pm_notifier(&priv->pm_notifier);
+}
 #else /* !PM_SLEEP || HERMES_CACHE_FW_ON_INIT */
-#define orinoco_pm_notifier NULL
+#define orinoco_register_pm_notifier(priv) do { } while(0)
+#define orinoco_unregister_pm_notifier(priv) do { } while(0)
 #endif
 
 /********************************************************************/
@@ -3648,8 +3660,7 @@ struct net_device
         priv->cached_fw = NULL;
 
         /* Register PM notifiers */
-        priv->pm_notifier.notifier_call = orinoco_pm_notifier;
-        register_pm_notifier(&priv->pm_notifier);
+        orinoco_register_pm_notifier(priv);
 
         return dev;
 }
@@ -3673,7 +3684,7 @@ void free_orinocodev(struct net_device *dev)
                 kfree(rx_data);
         }
 
-        unregister_pm_notifier(&priv->pm_notifier);
+        orinoco_unregister_pm_notifier(priv);
         orinoco_uncache_fw(priv);
 
         priv->wpa_ie_len = 0;
diff --git a/drivers/net/wireless/rtl818x/rtl8187_dev.c b/drivers/net/wireless/rtl818x/rtl8187_dev.c
index 22bc07ef2f37..f4747a1134ba 100644
--- a/drivers/net/wireless/rtl818x/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl818x/rtl8187_dev.c
@@ -48,6 +48,10 @@ static struct usb_device_id rtl8187_table[] __devinitdata = {
         {USB_DEVICE(0x0bda, 0x8189), .driver_info = DEVICE_RTL8187B},
         {USB_DEVICE(0x0bda, 0x8197), .driver_info = DEVICE_RTL8187B},
         {USB_DEVICE(0x0bda, 0x8198), .driver_info = DEVICE_RTL8187B},
+        /* Surecom */
+        {USB_DEVICE(0x0769, 0x11F2), .driver_info = DEVICE_RTL8187},
+        /* Logitech */
+        {USB_DEVICE(0x0789, 0x010C), .driver_info = DEVICE_RTL8187},
         /* Netgear */
         {USB_DEVICE(0x0846, 0x6100), .driver_info = DEVICE_RTL8187},
         {USB_DEVICE(0x0846, 0x6a00), .driver_info = DEVICE_RTL8187},
@@ -57,8 +61,16 @@ static struct usb_device_id rtl8187_table[] __devinitdata = {
         /* Sitecom */
         {USB_DEVICE(0x0df6, 0x000d), .driver_info = DEVICE_RTL8187},
         {USB_DEVICE(0x0df6, 0x0028), .driver_info = DEVICE_RTL8187B},
+        /* Sphairon Access Systems GmbH */
+        {USB_DEVICE(0x114B, 0x0150), .driver_info = DEVICE_RTL8187},
+        /* Dick Smith Electronics */
+        {USB_DEVICE(0x1371, 0x9401), .driver_info = DEVICE_RTL8187},
         /* Abocom */
         {USB_DEVICE(0x13d1, 0xabe6), .driver_info = DEVICE_RTL8187},
+        /* Qcom */
+        {USB_DEVICE(0x18E8, 0x6232), .driver_info = DEVICE_RTL8187},
+        /* AirLive */
+        {USB_DEVICE(0x1b75, 0x8187), .driver_info = DEVICE_RTL8187},
         {}
 };
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i.h b/drivers/scsi/cxgb3i/cxgb3i.h
index fde6e4c634e7..a7cf550b9cca 100644
--- a/drivers/scsi/cxgb3i/cxgb3i.h
+++ b/drivers/scsi/cxgb3i/cxgb3i.h
@@ -20,6 +20,7 @@
 #include <linux/list.h>
 #include <linux/netdevice.h>
 #include <linux/scatterlist.h>
+#include <linux/skbuff.h>
 #include <scsi/libiscsi_tcp.h>
 
 /* from cxgb3 LLD */
@@ -113,6 +114,26 @@ struct cxgb3i_endpoint {
         struct cxgb3i_conn *cconn;
 };
 
+/**
+ * struct cxgb3i_task_data - private iscsi task data
+ *
+ * @nr_frags:   # of coalesced page frags (from scsi sgl)
+ * @frags:      coalesced page frags (from scsi sgl)
+ * @skb:        tx pdu skb
+ * @offset:     data offset for the next pdu
+ * @count:      max. possible pdu payload
+ * @sgoffset:   offset to the first sg entry for a given offset
+ */
+#define MAX_PDU_FRAGS   ((ULP2_MAX_PDU_PAYLOAD + 512 - 1) / 512)
+struct cxgb3i_task_data {
+        unsigned short nr_frags;
+        skb_frag_t frags[MAX_PDU_FRAGS];
+        struct sk_buff *skb;
+        unsigned int offset;
+        unsigned int count;
+        unsigned int sgoffset;
+};
+
 int cxgb3i_iscsi_init(void);
 void cxgb3i_iscsi_cleanup(void);
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.c b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
index 08f3a09d9233..a83d36e4926f 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
@@ -639,10 +639,11 @@ static int ddp_init(struct t3cdev *tdev)
         write_unlock(&cxgb3i_ddp_rwlock);
 
         ddp_log_info("nppods %u (0x%x ~ 0x%x), bits %u, mask 0x%x,0x%x "
-                        "pkt %u,%u.\n",
+                        "pkt %u/%u, %u/%u.\n",
                         ppmax, ddp->llimit, ddp->ulimit, ddp->idx_bits,
                         ddp->idx_mask, ddp->rsvd_tag_mask,
-                        ddp->max_txsz, ddp->max_rxsz);
+                        ddp->max_txsz, uinfo.max_txsz,
+                        ddp->max_rxsz, uinfo.max_rxsz);
         return 0;
 
 free_ddp_map:
@@ -654,8 +655,8 @@ free_ddp_map:
  * cxgb3i_adapter_ddp_init - initialize the adapter's ddp resource
  * @tdev: t3cdev adapter
  * @tformat: tag format
- * @txsz: max tx pkt size, filled in by this func.
- * @rxsz: max rx pkt size, filled in by this func.
+ * @txsz: max tx pdu payload size, filled in by this func.
+ * @rxsz: max rx pdu payload size, filled in by this func.
  * initialize the ddp pagepod manager for a given adapter if needed and
  * setup the tag format for a given iscsi entity
  */
@@ -685,10 +686,12 @@ int cxgb3i_adapter_ddp_init(struct t3cdev *tdev,
                       tformat->sw_bits, tformat->rsvd_bits,
                       tformat->rsvd_shift, tformat->rsvd_mask);
 
-        *txsz = ddp->max_txsz;
-        *rxsz = ddp->max_rxsz;
-        ddp_log_info("ddp max pkt size: %u, %u.\n",
-                     ddp->max_txsz, ddp->max_rxsz);
+        *txsz = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
+                        ddp->max_txsz - ISCSI_PDU_NONPAYLOAD_LEN);
+        *rxsz = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
+                        ddp->max_rxsz - ISCSI_PDU_NONPAYLOAD_LEN);
+        ddp_log_info("max payload size: %u/%u, %u/%u.\n",
+                     *txsz, ddp->max_txsz, *rxsz, ddp->max_rxsz);
         return 0;
 }
 EXPORT_SYMBOL_GPL(cxgb3i_adapter_ddp_init);
diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.h b/drivers/scsi/cxgb3i/cxgb3i_ddp.h
index 5c7c4d95c493..3faae7831c83 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.h
@@ -13,6 +13,8 @@
 #ifndef __CXGB3I_ULP2_DDP_H__
 #define __CXGB3I_ULP2_DDP_H__
 
+#include <linux/vmalloc.h>
+
 /**
  * struct cxgb3i_tag_format - cxgb3i ulp tag format for an iscsi entity
  *
@@ -85,8 +87,9 @@ struct cxgb3i_ddp_info {
         struct sk_buff **gl_skb;
 };
 
+#define ISCSI_PDU_NONPAYLOAD_LEN        312 /* bhs(48) + ahs(256) + digest(8) */
 #define ULP2_MAX_PKT_SIZE       16224
-#define ULP2_MAX_PDU_PAYLOAD    (ULP2_MAX_PKT_SIZE - ISCSI_PDU_NONPAYLOAD_MAX)
+#define ULP2_MAX_PDU_PAYLOAD    (ULP2_MAX_PKT_SIZE - ISCSI_PDU_NONPAYLOAD_LEN)
 #define PPOD_PAGES_MAX          4
 #define PPOD_PAGES_SHIFT        2       /* 4 pages per pod */
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i_init.c b/drivers/scsi/cxgb3i/cxgb3i_init.c
index 091ecb4d9f3d..1ce9f244e46c 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_init.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_init.c
@@ -12,8 +12,8 @@
 #include "cxgb3i.h"
 
 #define DRV_MODULE_NAME         "cxgb3i"
-#define DRV_MODULE_VERSION      "1.0.0"
-#define DRV_MODULE_RELDATE      "Jun. 1, 2008"
+#define DRV_MODULE_VERSION      "1.0.1"
+#define DRV_MODULE_RELDATE      "Jan. 2009"
 
 static char version[] =
         "Chelsio S3xx iSCSI Driver " DRV_MODULE_NAME
diff --git a/drivers/scsi/cxgb3i/cxgb3i_iscsi.c b/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
index d83464b9b3f9..fa2a44f37b36 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
@@ -364,7 +364,8 @@ cxgb3i_session_create(struct iscsi_endpoint *ep, u16 cmds_max, u16 qdepth,
 
         cls_session = iscsi_session_setup(&cxgb3i_iscsi_transport, shost,
                                           cmds_max,
-                                          sizeof(struct iscsi_tcp_task),
+                                          sizeof(struct iscsi_tcp_task) +
+                                          sizeof(struct cxgb3i_task_data),
                                           initial_cmdsn, ISCSI_MAX_TARGET);
         if (!cls_session)
                 return NULL;
@@ -402,17 +403,15 @@ static inline int cxgb3i_conn_max_xmit_dlength(struct iscsi_conn *conn)
 {
         struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
         struct cxgb3i_conn *cconn = tcp_conn->dd_data;
-        unsigned int max = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
-                                 cconn->hba->snic->tx_max_size -
-                                 ISCSI_PDU_NONPAYLOAD_MAX);
+        unsigned int max = max(512 * MAX_SKB_FRAGS, SKB_TX_HEADROOM);
 
+        max = min(cconn->hba->snic->tx_max_size, max);
         if (conn->max_xmit_dlength)
-                conn->max_xmit_dlength = min_t(unsigned int,
-                                                conn->max_xmit_dlength, max);
+                conn->max_xmit_dlength = min(conn->max_xmit_dlength, max);
         else
                 conn->max_xmit_dlength = max;
         align_pdu_size(conn->max_xmit_dlength);
-        cxgb3i_log_info("conn 0x%p, max xmit %u.\n",
+        cxgb3i_api_debug("conn 0x%p, max xmit %u.\n",
                          conn, conn->max_xmit_dlength);
         return 0;
 }
@@ -427,9 +426,7 @@ static inline int cxgb3i_conn_max_recv_dlength(struct iscsi_conn *conn)
 {
         struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
         struct cxgb3i_conn *cconn = tcp_conn->dd_data;
-        unsigned int max = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD,
-                                 cconn->hba->snic->rx_max_size -
-                                 ISCSI_PDU_NONPAYLOAD_MAX);
+        unsigned int max = cconn->hba->snic->rx_max_size;
 
         align_pdu_size(max);
         if (conn->max_recv_dlength) {
@@ -439,8 +436,7 @@ static inline int cxgb3i_conn_max_recv_dlength(struct iscsi_conn *conn)
                                          conn->max_recv_dlength, max);
                         return -EINVAL;
                 }
-                conn->max_recv_dlength = min_t(unsigned int,
-                                                conn->max_recv_dlength, max);
+                conn->max_recv_dlength = min(conn->max_recv_dlength, max);
                 align_pdu_size(conn->max_recv_dlength);
         } else
                 conn->max_recv_dlength = max;
@@ -844,7 +840,7 @@ static struct scsi_host_template cxgb3i_host_template = {
         .proc_name              = "cxgb3i",
         .queuecommand           = iscsi_queuecommand,
         .change_queue_depth     = iscsi_change_queue_depth,
-        .can_queue              = 128 * (ISCSI_DEF_XMIT_CMDS_MAX - 1),
+        .can_queue              = CXGB3I_SCSI_QDEPTH_DFLT - 1,
         .sg_tablesize           = SG_ALL,
         .max_sectors            = 0xFFFF,
         .cmd_per_lun            = ISCSI_DEF_CMD_PER_LUN,
diff --git a/drivers/scsi/cxgb3i/cxgb3i_offload.c b/drivers/scsi/cxgb3i/cxgb3i_offload.c
index a865f1fefe8b..de3b3b614cca 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_offload.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_offload.c
@@ -23,19 +23,19 @@
 #include "cxgb3i_ddp.h"
 
 #ifdef __DEBUG_C3CN_CONN__
-#define c3cn_conn_debug         cxgb3i_log_info
+#define c3cn_conn_debug         cxgb3i_log_debug
 #else
 #define c3cn_conn_debug(fmt...)
 #endif
 
 #ifdef __DEBUG_C3CN_TX__
-#define c3cn_tx_debug         cxgb3i_log_debug
+#define c3cn_tx_debug           cxgb3i_log_debug
 #else
 #define c3cn_tx_debug(fmt...)
 #endif
 
 #ifdef __DEBUG_C3CN_RX__
-#define c3cn_rx_debug         cxgb3i_log_debug
+#define c3cn_rx_debug           cxgb3i_log_debug
 #else
 #define c3cn_rx_debug(fmt...)
 #endif
@@ -47,9 +47,9 @@ static int cxgb3_rcv_win = 256 * 1024;
 module_param(cxgb3_rcv_win, int, 0644);
 MODULE_PARM_DESC(cxgb3_rcv_win, "TCP receive window in bytes (default=256KB)");
 
-static int cxgb3_snd_win = 64 * 1024;
+static int cxgb3_snd_win = 128 * 1024;
 module_param(cxgb3_snd_win, int, 0644);
-MODULE_PARM_DESC(cxgb3_snd_win, "TCP send window in bytes (default=64KB)");
+MODULE_PARM_DESC(cxgb3_snd_win, "TCP send window in bytes (default=128KB)");
 
 static int cxgb3_rx_credit_thres = 10 * 1024;
 module_param(cxgb3_rx_credit_thres, int, 0644);
@@ -301,8 +301,8 @@ static void act_open_req_arp_failure(struct t3cdev *dev, struct sk_buff *skb)
 static void skb_entail(struct s3_conn *c3cn, struct sk_buff *skb,
                        int flags)
 {
-        CXGB3_SKB_CB(skb)->seq = c3cn->write_seq;
-        CXGB3_SKB_CB(skb)->flags = flags;
+        skb_tcp_seq(skb) = c3cn->write_seq;
+        skb_flags(skb) = flags;
         __skb_queue_tail(&c3cn->write_queue, skb);
 }
 
@@ -457,12 +457,9 @@ static unsigned int wrlen __read_mostly;
  * The number of WRs needed for an skb depends on the number of fragments
  * in the skb and whether it has any payload in its main body.  This maps the
  * length of the gather list represented by an skb into the # of necessary WRs.
- *
- * The max. length of an skb is controlled by the max pdu size which is ~16K.
- * Also, assume the min. fragment length is the sector size (512), then add
- * extra fragment counts for iscsi bhs and payload padding.
+ * The extra two fragments are for iscsi bhs and payload padding.
  */
-#define SKB_WR_LIST_SIZE        (16384/512 + 3)
+#define SKB_WR_LIST_SIZE        (MAX_SKB_FRAGS + 2)
 static unsigned int skb_wrs[SKB_WR_LIST_SIZE] __read_mostly;
 
 static void s3_init_wr_tab(unsigned int wr_len)
@@ -485,7 +482,7 @@ static void s3_init_wr_tab(unsigned int wr_len)
 
 static inline void reset_wr_list(struct s3_conn *c3cn)
 {
-        c3cn->wr_pending_head = NULL;
+        c3cn->wr_pending_head = c3cn->wr_pending_tail = NULL;
 }
 
 /*
@@ -496,7 +493,7 @@ static inline void reset_wr_list(struct s3_conn *c3cn)
 static inline void enqueue_wr(struct s3_conn *c3cn,
                               struct sk_buff *skb)
 {
-        skb_wr_data(skb) = NULL;
+        skb_tx_wr_next(skb) = NULL;
 
         /*
          * We want to take an extra reference since both us and the driver
@@ -509,10 +506,22 @@ static inline void enqueue_wr(struct s3_conn *c3cn,
         if (!c3cn->wr_pending_head)
                 c3cn->wr_pending_head = skb;
         else
-                skb_wr_data(skb) = skb;
+                skb_tx_wr_next(c3cn->wr_pending_tail) = skb;
         c3cn->wr_pending_tail = skb;
 }
 
+static int count_pending_wrs(struct s3_conn *c3cn)
+{
+        int n = 0;
+        const struct sk_buff *skb = c3cn->wr_pending_head;
+
+        while (skb) {
+                n += skb->csum;
+                skb = skb_tx_wr_next(skb);
+        }
+        return n;
+}
+
 static inline struct sk_buff *peek_wr(const struct s3_conn *c3cn)
 {
         return c3cn->wr_pending_head;
@@ -529,8 +538,8 @@ static inline struct sk_buff *dequeue_wr(struct s3_conn *c3cn)
 
         if (likely(skb)) {
                 /* Don't bother clearing the tail */
-                c3cn->wr_pending_head = skb_wr_data(skb);
-                skb_wr_data(skb) = NULL;
+                c3cn->wr_pending_head = skb_tx_wr_next(skb);
+                skb_tx_wr_next(skb) = NULL;
         }
         return skb;
 }
@@ -543,13 +552,14 @@ static void purge_wr_queue(struct s3_conn *c3cn)
 }
 
 static inline void make_tx_data_wr(struct s3_conn *c3cn, struct sk_buff *skb,
-                                   int len)
+                                   int len, int req_completion)
 {
         struct tx_data_wr *req;
 
         skb_reset_transport_header(skb);
         req = (struct tx_data_wr *)__skb_push(skb, sizeof(*req));
-        req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA));
+        req->wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_TX_DATA) |
+                        (req_completion ? F_WR_COMPL : 0));
         req->wr_lo = htonl(V_WR_TID(c3cn->tid));
         req->sndseq = htonl(c3cn->snd_nxt);
         /* len includes the length of any HW ULP additions */
@@ -592,7 +602,7 @@ static int c3cn_push_tx_frames(struct s3_conn *c3cn, int req_completion)
 
         if (unlikely(c3cn->state == C3CN_STATE_CONNECTING ||
                      c3cn->state == C3CN_STATE_CLOSE_WAIT_1 ||
-                     c3cn->state == C3CN_STATE_ABORTING)) {
+                     c3cn->state >= C3CN_STATE_ABORTING)) {
                 c3cn_tx_debug("c3cn 0x%p, in closing state %u.\n",
                               c3cn, c3cn->state);
                 return 0;
@@ -615,7 +625,7 @@ static int c3cn_push_tx_frames(struct s3_conn *c3cn, int req_completion)
                 if (c3cn->wr_avail < wrs_needed) {
                         c3cn_tx_debug("c3cn 0x%p, skb len %u/%u, frag %u, "
                                       "wr %d < %u.\n",
-                                      c3cn, skb->len, skb->datalen, frags,
+                                      c3cn, skb->len, skb->data_len, frags,
                                       wrs_needed, c3cn->wr_avail);
                         break;
                 }
@@ -627,20 +637,24 @@ static int c3cn_push_tx_frames(struct s3_conn *c3cn, int req_completion)
                 c3cn->wr_unacked += wrs_needed;
                 enqueue_wr(c3cn, skb);
 
-                if (likely(CXGB3_SKB_CB(skb)->flags & C3CB_FLAG_NEED_HDR)) {
-                        len += ulp_extra_len(skb);
-                        make_tx_data_wr(c3cn, skb, len);
-                        c3cn->snd_nxt += len;
-                        if ((req_completion
-                             && c3cn->wr_unacked == wrs_needed)
-                            || (CXGB3_SKB_CB(skb)->flags & C3CB_FLAG_COMPL)
-                            || c3cn->wr_unacked >= c3cn->wr_max / 2) {
-                                struct work_request_hdr *wr = cplhdr(skb);
+                c3cn_tx_debug("c3cn 0x%p, enqueue, skb len %u/%u, frag %u, "
+                                "wr %d, left %u, unack %u.\n",
+                                c3cn, skb->len, skb->data_len, frags,
+                                wrs_needed, c3cn->wr_avail, c3cn->wr_unacked);
+
 
-                                wr->wr_hi |= htonl(F_WR_COMPL);
+                if (likely(skb_flags(skb) & C3CB_FLAG_NEED_HDR)) {
+                        if ((req_completion &&
+                                c3cn->wr_unacked == wrs_needed) ||
+                            (skb_flags(skb) & C3CB_FLAG_COMPL) ||
+                            c3cn->wr_unacked >= c3cn->wr_max / 2) {
+                                req_completion = 1;
                                 c3cn->wr_unacked = 0;
                         }
-                        CXGB3_SKB_CB(skb)->flags &= ~C3CB_FLAG_NEED_HDR;
+                        len += ulp_extra_len(skb);
+                        make_tx_data_wr(c3cn, skb, len, req_completion);
+                        c3cn->snd_nxt += len;
+                        skb_flags(skb) &= ~C3CB_FLAG_NEED_HDR;
                 }
 
                 total_size += skb->truesize;
@@ -735,8 +749,11 @@ static void process_act_establish(struct s3_conn *c3cn, struct sk_buff *skb)
         if (unlikely(c3cn_flag(c3cn, C3CN_ACTIVE_CLOSE_NEEDED)))
                 /* upper layer has requested closing */
                 send_abort_req(c3cn);
-        else if (c3cn_push_tx_frames(c3cn, 1))
+        else {
+                if (skb_queue_len(&c3cn->write_queue))
+                        c3cn_push_tx_frames(c3cn, 1);
                 cxgb3i_conn_tx_open(c3cn);
+        }
 }
 
 static int do_act_establish(struct t3cdev *cdev, struct sk_buff *skb,
@@ -1082,8 +1099,8 @@ static void process_rx_iscsi_hdr(struct s3_conn *c3cn, struct sk_buff *skb)
                 return;
         }
 
-        CXGB3_SKB_CB(skb)->seq = ntohl(hdr_cpl->seq);
-        CXGB3_SKB_CB(skb)->flags = 0;
+        skb_tcp_seq(skb) = ntohl(hdr_cpl->seq);
+        skb_flags(skb) = 0;
 
         skb_reset_transport_header(skb);
         __skb_pull(skb, sizeof(struct cpl_iscsi_hdr));
@@ -1103,12 +1120,12 @@ static void process_rx_iscsi_hdr(struct s3_conn *c3cn, struct sk_buff *skb)
                 goto abort_conn;
 
         skb_ulp_mode(skb) = ULP2_FLAG_DATA_READY;
-        skb_ulp_pdulen(skb) = ntohs(ddp_cpl.len);
-        skb_ulp_ddigest(skb) = ntohl(ddp_cpl.ulp_crc);
+        skb_rx_pdulen(skb) = ntohs(ddp_cpl.len);
+        skb_rx_ddigest(skb) = ntohl(ddp_cpl.ulp_crc);
         status = ntohl(ddp_cpl.ddp_status);
 
         c3cn_rx_debug("rx skb 0x%p, len %u, pdulen %u, ddp status 0x%x.\n",
-                      skb, skb->len, skb_ulp_pdulen(skb), status);
+                      skb, skb->len, skb_rx_pdulen(skb), status);
 
         if (status & (1 << RX_DDP_STATUS_HCRC_SHIFT))
                 skb_ulp_mode(skb) |= ULP2_FLAG_HCRC_ERROR;
@@ -1126,7 +1143,7 @@ static void process_rx_iscsi_hdr(struct s3_conn *c3cn, struct sk_buff *skb)
         } else if (status & (1 << RX_DDP_STATUS_DDP_SHIFT))
                 skb_ulp_mode(skb) |= ULP2_FLAG_DATA_DDPED;
 
-        c3cn->rcv_nxt = ntohl(ddp_cpl.seq) + skb_ulp_pdulen(skb);
+        c3cn->rcv_nxt = ntohl(ddp_cpl.seq) + skb_rx_pdulen(skb);
         __pskb_trim(skb, len);
         __skb_queue_tail(&c3cn->receive_queue, skb);
         cxgb3i_conn_pdu_ready(c3cn);
@@ -1151,12 +1168,27 @@ static int do_iscsi_hdr(struct t3cdev *t3dev, struct sk_buff *skb, void *ctx)
  * Process an acknowledgment of WR completion.  Advance snd_una and send the
  * next batch of work requests from the write queue.
  */
+static void check_wr_invariants(struct s3_conn *c3cn)
+{
+        int pending = count_pending_wrs(c3cn);
+
+        if (unlikely(c3cn->wr_avail + pending != c3cn->wr_max))
+                cxgb3i_log_error("TID %u: credit imbalance: avail %u, "
+                                "pending %u, total should be %u\n",
+                                c3cn->tid, c3cn->wr_avail, pending,
+                                c3cn->wr_max);
+}
+
 static void process_wr_ack(struct s3_conn *c3cn, struct sk_buff *skb)
 {
         struct cpl_wr_ack *hdr = cplhdr(skb);
         unsigned int credits = ntohs(hdr->credits);
         u32 snd_una = ntohl(hdr->snd_una);
 
+        c3cn_tx_debug("%u WR credits, avail %u, unack %u, TID %u, state %u.\n",
+                        credits, c3cn->wr_avail, c3cn->wr_unacked,
+                        c3cn->tid, c3cn->state);
+
         c3cn->wr_avail += credits;
         if (c3cn->wr_unacked > c3cn->wr_max - c3cn->wr_avail)
                 c3cn->wr_unacked = c3cn->wr_max - c3cn->wr_avail;
@@ -1171,6 +1203,17 @@ static void process_wr_ack(struct s3_conn *c3cn, struct sk_buff *skb)
                         break;
                 }
                 if (unlikely(credits < p->csum)) {
+                        struct tx_data_wr *w = cplhdr(p);
+                        cxgb3i_log_error("TID %u got %u WR credits need %u, "
+                                         "len %u, main body %u, frags %u, "
+                                         "seq # %u, ACK una %u, ACK nxt %u, "
+                                         "WR_AVAIL %u, WRs pending %u\n",
+                                         c3cn->tid, credits, p->csum, p->len,
+                                         p->len - p->data_len,
+                                         skb_shinfo(p)->nr_frags,
+                                         ntohl(w->sndseq), snd_una,
+                                         ntohl(hdr->snd_nxt), c3cn->wr_avail,
+                                         count_pending_wrs(c3cn) - credits);
                         p->csum -= credits;
                         break;
                 } else {
@@ -1180,15 +1223,24 @@ static void process_wr_ack(struct s3_conn *c3cn, struct sk_buff *skb)
                 }
         }
 
-        if (unlikely(before(snd_una, c3cn->snd_una)))
+        check_wr_invariants(c3cn);
+
+        if (unlikely(before(snd_una, c3cn->snd_una))) {
+                cxgb3i_log_error("TID %u, unexpected sequence # %u in WR_ACK "
+                                 "snd_una %u\n",
+                                 c3cn->tid, snd_una, c3cn->snd_una);
                 goto out_free;
+        }
 
         if (c3cn->snd_una != snd_una) {
                 c3cn->snd_una = snd_una;
                 dst_confirm(c3cn->dst_cache);
         }
 
-        if (skb_queue_len(&c3cn->write_queue) && c3cn_push_tx_frames(c3cn, 0))
+        if (skb_queue_len(&c3cn->write_queue)) {
+                if (c3cn_push_tx_frames(c3cn, 0))
+                        cxgb3i_conn_tx_open(c3cn);
+        } else
                 cxgb3i_conn_tx_open(c3cn);
 out_free:
         __kfree_skb(skb);
@@ -1452,7 +1504,7 @@ static void init_offload_conn(struct s3_conn *c3cn,
                               struct dst_entry *dst)
 {
         BUG_ON(c3cn->cdev != cdev);
-        c3cn->wr_max = c3cn->wr_avail = T3C_DATA(cdev)->max_wrs;
+        c3cn->wr_max = c3cn->wr_avail = T3C_DATA(cdev)->max_wrs - 1;
         c3cn->wr_unacked = 0;
         c3cn->mss_idx = select_mss(c3cn, dst_mtu(dst));
 
@@ -1671,9 +1723,17 @@ int cxgb3i_c3cn_send_pdus(struct s3_conn *c3cn, struct sk_buff *skb)
                 goto out_err;
         }
 
-        err = -EPIPE;
         if (c3cn->err) {
                 c3cn_tx_debug("c3cn 0x%p, err %d.\n", c3cn, c3cn->err);
+                err = -EPIPE;
+                goto out_err;
+        }
+
+        if (c3cn->write_seq - c3cn->snd_una >= cxgb3_snd_win) {
+                c3cn_tx_debug("c3cn 0x%p, snd %u - %u > %u.\n",
+                                c3cn, c3cn->write_seq, c3cn->snd_una,
+                                cxgb3_snd_win);
+                err = -EAGAIN;
                 goto out_err;
         }
 
diff --git a/drivers/scsi/cxgb3i/cxgb3i_offload.h b/drivers/scsi/cxgb3i/cxgb3i_offload.h
index d23156907ffd..6344b9eb2589 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_offload.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_offload.h
@@ -178,25 +178,33 @@ void cxgb3i_c3cn_release(struct s3_conn *);
  * @flag:       see C3CB_FLAG_* below
  * @ulp_mode:   ULP mode/submode of sk_buff
  * @seq:        tcp sequence number
- * @ddigest:    pdu data digest
- * @pdulen:     recovered pdu length
- * @wr_data:    scratch area for tx wr
  */
+struct cxgb3_skb_rx_cb {
+        __u32 ddigest;                  /* data digest */
+        __u32 pdulen;                   /* recovered pdu length */
+};
+
+struct cxgb3_skb_tx_cb {
+        struct sk_buff *wr_next;        /* next wr */
+};
+
 struct cxgb3_skb_cb {
         __u8 flags;
         __u8 ulp_mode;
         __u32 seq;
-        __u32 ddigest;
-        __u32 pdulen;
-        struct sk_buff *wr_data;
+        union {
+                struct cxgb3_skb_rx_cb rx;
+                struct cxgb3_skb_tx_cb tx;
+        };
 };
 
 #define CXGB3_SKB_CB(skb)       ((struct cxgb3_skb_cb *)&((skb)->cb[0]))
-
+#define skb_flags(skb)          (CXGB3_SKB_CB(skb)->flags)
 #define skb_ulp_mode(skb)       (CXGB3_SKB_CB(skb)->ulp_mode)
-#define skb_ulp_ddigest(skb)    (CXGB3_SKB_CB(skb)->ddigest)
-#define skb_ulp_pdulen(skb)     (CXGB3_SKB_CB(skb)->pdulen)
-#define skb_wr_data(skb)        (CXGB3_SKB_CB(skb)->wr_data)
+#define skb_tcp_seq(skb)        (CXGB3_SKB_CB(skb)->seq)
+#define skb_rx_ddigest(skb)     (CXGB3_SKB_CB(skb)->rx.ddigest)
+#define skb_rx_pdulen(skb)      (CXGB3_SKB_CB(skb)->rx.pdulen)
+#define skb_tx_wr_next(skb)     (CXGB3_SKB_CB(skb)->tx.wr_next)
 
 enum c3cb_flags {
         C3CB_FLAG_NEED_HDR = 1 << 0,    /* packet needs a TX_DATA_WR header */
@@ -217,6 +225,7 @@ struct sge_opaque_hdr {
 /* for TX: a skb must have a headroom of at least TX_HEADER_LEN bytes */
 #define TX_HEADER_LEN \
                 (sizeof(struct tx_data_wr) + sizeof(struct sge_opaque_hdr))
+#define SKB_TX_HEADROOM         SKB_MAX_HEAD(TX_HEADER_LEN)
 
 /*
  * get and set private ip for iscsi traffic
diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.c b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
index ce7ce8c6094c..17115c230d65 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
@@ -32,6 +32,10 @@
 #define cxgb3i_tx_debug(fmt...)
 #endif
 
+/* always allocate rooms for AHS */
+#define SKB_TX_PDU_HEADER_LEN   \
+        (sizeof(struct iscsi_hdr) + ISCSI_MAX_AHS_SIZE)
+static unsigned int skb_extra_headroom;
 static struct page *pad_page;
 
 /*
@@ -146,12 +150,13 @@ static inline void tx_skb_setmode(struct sk_buff *skb, int hcrc, int dcrc)
 
 void cxgb3i_conn_cleanup_task(struct iscsi_task *task)
 {
-        struct iscsi_tcp_task *tcp_task = task->dd_data;
+        struct cxgb3i_task_data *tdata = task->dd_data +
+                                        sizeof(struct iscsi_tcp_task);
 
         /* never reached the xmit task callout */
-        if (tcp_task->dd_data)
-                kfree_skb(tcp_task->dd_data);
-        tcp_task->dd_data = NULL;
+        if (tdata->skb)
+                __kfree_skb(tdata->skb);
+        memset(tdata, 0, sizeof(struct cxgb3i_task_data));
 
         /* MNC - Do we need a check in case this is called but
          * cxgb3i_conn_alloc_pdu has never been called on the task */
@@ -159,28 +164,102 @@ void cxgb3i_conn_cleanup_task(struct iscsi_task *task)
         iscsi_tcp_cleanup_task(task);
 }
 
-/*
- * We do not support ahs yet
- */
+static int sgl_seek_offset(struct scatterlist *sgl, unsigned int sgcnt,
+                                unsigned int offset, unsigned int *off,
+                                struct scatterlist **sgp)
+{
+        int i;
+        struct scatterlist *sg;
+
+        for_each_sg(sgl, sg, sgcnt, i) {
+                if (offset < sg->length) {
+                        *off = offset;
+                        *sgp = sg;
+                        return 0;
+                }
+                offset -= sg->length;
+        }
+        return -EFAULT;
+}
+
+static int sgl_read_to_frags(struct scatterlist *sg, unsigned int sgoffset,
+                                unsigned int dlen, skb_frag_t *frags,
+                                int frag_max)
+{
+        unsigned int datalen = dlen;
+        unsigned int sglen = sg->length - sgoffset;
+        struct page *page = sg_page(sg);
+        int i;
+
+        i = 0;
+        do {
+                unsigned int copy;
+
+                if (!sglen) {
+                        sg = sg_next(sg);
+                        if (!sg) {
+                                cxgb3i_log_error("%s, sg NULL, len %u/%u.\n",
+                                                 __func__, datalen, dlen);
+                                return -EINVAL;
+                        }
+                        sgoffset = 0;
+                        sglen = sg->length;
+                        page = sg_page(sg);
+
+                }
+                copy = min(datalen, sglen);
+                if (i && page == frags[i - 1].page &&
+                    sgoffset + sg->offset ==
+                        frags[i - 1].page_offset + frags[i - 1].size) {
+                        frags[i - 1].size += copy;
+                } else {
+                        if (i >= frag_max) {
+                                cxgb3i_log_error("%s, too many pages %u, "
+                                                 "dlen %u.\n", __func__,
+                                                 frag_max, dlen);
+                                return -EINVAL;
+                        }
+
+                        frags[i].page = page;
+                        frags[i].page_offset = sg->offset + sgoffset;
+                        frags[i].size = copy;
+                        i++;
+                }
+                datalen -= copy;
+                sgoffset += copy;
+                sglen -= copy;
+        } while (datalen);
+
+        return i;
+}
+
 int cxgb3i_conn_alloc_pdu(struct iscsi_task *task, u8 opcode)
 {
+        struct iscsi_conn *conn = task->conn;
         struct iscsi_tcp_task *tcp_task = task->dd_data;
-        struct sk_buff *skb;
+        struct cxgb3i_task_data *tdata = task->dd_data + sizeof(*tcp_task);
+        struct scsi_cmnd *sc = task->sc;
+        int headroom = SKB_TX_PDU_HEADER_LEN;
 
+        tcp_task->dd_data = tdata;
         task->hdr = NULL;
-        /* always allocate rooms for AHS */
-        skb = alloc_skb(sizeof(struct iscsi_hdr) + ISCSI_MAX_AHS_SIZE +
-                        TX_HEADER_LEN,  GFP_ATOMIC);
-        if (!skb)
+
+        /* write command, need to send data pdus */
+        if (skb_extra_headroom && (opcode == ISCSI_OP_SCSI_DATA_OUT ||
+            (opcode == ISCSI_OP_SCSI_CMD &&
+            (scsi_bidi_cmnd(sc) || sc->sc_data_direction == DMA_TO_DEVICE))))
+                headroom += min(skb_extra_headroom, conn->max_xmit_dlength);
+
+        tdata->skb = alloc_skb(TX_HEADER_LEN + headroom, GFP_ATOMIC);
+        if (!tdata->skb)
                 return -ENOMEM;
+        skb_reserve(tdata->skb, TX_HEADER_LEN);
 
         cxgb3i_tx_debug("task 0x%p, opcode 0x%x, skb 0x%p.\n",
-                        task, opcode, skb);
+                        task, opcode, tdata->skb);
 
-        tcp_task->dd_data = skb;
-        skb_reserve(skb, TX_HEADER_LEN);
-        task->hdr = (struct iscsi_hdr *)skb->data;
-        task->hdr_max = sizeof(struct iscsi_hdr);
+        task->hdr = (struct iscsi_hdr *)tdata->skb->data;
+        task->hdr_max = SKB_TX_PDU_HEADER_LEN;
 
         /* data_out uses scsi_cmd's itt */
         if (opcode != ISCSI_OP_SCSI_DATA_OUT)
@@ -192,13 +271,13 @@ int cxgb3i_conn_alloc_pdu(struct iscsi_task *task, u8 opcode)
 int cxgb3i_conn_init_pdu(struct iscsi_task *task, unsigned int offset,
                               unsigned int count)
 {
-        struct iscsi_tcp_task *tcp_task = task->dd_data;
-        struct sk_buff *skb = tcp_task->dd_data;
         struct iscsi_conn *conn = task->conn;
-        struct page *pg;
+        struct iscsi_tcp_task *tcp_task = task->dd_data;
+        struct cxgb3i_task_data *tdata = tcp_task->dd_data;
+        struct sk_buff *skb = tdata->skb;
         unsigned int datalen = count;
         int i, padlen = iscsi_padding(count);
-        skb_frag_t *frag;
+        struct page *pg;
 
         cxgb3i_tx_debug("task 0x%p,0x%p, offset %u, count %u, skb 0x%p.\n",
                         task, task->sc, offset, count, skb);
@@ -209,90 +288,94 @@ int cxgb3i_conn_init_pdu(struct iscsi_task *task, unsigned int offset,
                 return 0;
 
         if (task->sc) {
-                struct scatterlist *sg;
-                struct scsi_data_buffer *sdb;
-                unsigned int sgoffset = offset;
-                struct page *sgpg;
-                unsigned int sglen;
-
-                sdb = scsi_out(task->sc);
-                sg = sdb->table.sgl;
-
-                for_each_sg(sdb->table.sgl, sg, sdb->table.nents, i) {
-                        cxgb3i_tx_debug("sg %d, page 0x%p, len %u offset %u\n",
-                                        i, sg_page(sg), sg->length, sg->offset);
-
-                        if (sgoffset < sg->length)
-                                break;
-                        sgoffset -= sg->length;
+                struct scsi_data_buffer *sdb = scsi_out(task->sc);
+                struct scatterlist *sg = NULL;
+                int err;
+
+                tdata->offset = offset;
+                tdata->count = count;
+                err = sgl_seek_offset(sdb->table.sgl, sdb->table.nents,
+                                        tdata->offset, &tdata->sgoffset, &sg);
+                if (err < 0) {
+                        cxgb3i_log_warn("tpdu, sgl %u, bad offset %u/%u.\n",
+                                        sdb->table.nents, tdata->offset,
+                                        sdb->length);
+                        return err;
                 }
-                sgpg = sg_page(sg);
-                sglen = sg->length - sgoffset;
-
-                do {
-                        int j = skb_shinfo(skb)->nr_frags;
-                        unsigned int copy;
-
-                        if (!sglen) {
-                                sg = sg_next(sg);
-                                sgpg = sg_page(sg);
-                                sgoffset = 0;
-                                sglen = sg->length;
-                                ++i;
+                err = sgl_read_to_frags(sg, tdata->sgoffset, tdata->count,
+                                        tdata->frags, MAX_PDU_FRAGS);
+                if (err < 0) {
+                        cxgb3i_log_warn("tpdu, sgl %u, bad offset %u + %u.\n",
+                                        sdb->table.nents, tdata->offset,
+                                        tdata->count);
+                        return err;
+                }
+                tdata->nr_frags = err;
+
+                if (tdata->nr_frags > MAX_SKB_FRAGS ||
+                    (padlen && tdata->nr_frags == MAX_SKB_FRAGS)) {
+                        char *dst = skb->data + task->hdr_len;
+                        skb_frag_t *frag = tdata->frags;
+
+                        /* data fits in the skb's headroom */
+                        for (i = 0; i < tdata->nr_frags; i++, frag++) {
+                                char *src = kmap_atomic(frag->page,
+                                                        KM_SOFTIRQ0);
+
+                                memcpy(dst, src+frag->page_offset, frag->size);
+                                dst += frag->size;
+                                kunmap_atomic(src, KM_SOFTIRQ0);
                         }
-                        copy = min(sglen, datalen);
-                        if (j && skb_can_coalesce(skb, j, sgpg,
-                                                  sg->offset + sgoffset)) {
-                                skb_shinfo(skb)->frags[j - 1].size += copy;
-                        } else {
-                                get_page(sgpg);
-                                skb_fill_page_desc(skb, j, sgpg,
-                                                   sg->offset + sgoffset, copy);
+                        if (padlen) {
+                                memset(dst, 0, padlen);
+                                padlen = 0;
                         }
-                        sgoffset += copy;
-                        sglen -= copy;
-                        datalen -= copy;
-                } while (datalen);
+                        skb_put(skb, count + padlen);
+                } else {
+                        /* data fit into frag_list */
+                        for (i = 0; i < tdata->nr_frags; i++)
+                                get_page(tdata->frags[i].page);
+
+                        memcpy(skb_shinfo(skb)->frags, tdata->frags,
+                                sizeof(skb_frag_t) * tdata->nr_frags);
+                        skb_shinfo(skb)->nr_frags = tdata->nr_frags;
+                        skb->len += count;
+                        skb->data_len += count;
+                        skb->truesize += count;
+                }
+
         } else {
                 pg = virt_to_page(task->data);
 
-                while (datalen) {
-                        i = skb_shinfo(skb)->nr_frags;
-                        frag = &skb_shinfo(skb)->frags[i];
-
-                        get_page(pg);
-                        frag->page = pg;
-                        frag->page_offset = 0;
-                        frag->size = min((unsigned int)PAGE_SIZE, datalen);
-
-                        skb_shinfo(skb)->nr_frags++;
-                        datalen -= frag->size;
-                        pg++;
-                }
+                get_page(pg);
+                skb_fill_page_desc(skb, 0, pg, offset_in_page(task->data),
+                                        count);
+                skb->len += count;
+                skb->data_len += count;
+                skb->truesize += count;
         }
 
         if (padlen) {
                 i = skb_shinfo(skb)->nr_frags;
-                frag = &skb_shinfo(skb)->frags[i];
-                frag->page = pad_page;
-                frag->page_offset = 0;
-                frag->size = padlen;
-                skb_shinfo(skb)->nr_frags++;
+                get_page(pad_page);
+                skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, pad_page, 0,
+                                 padlen);
+
+                skb->data_len += padlen;
+                skb->truesize += padlen;
+                skb->len += padlen;
         }
 
-        datalen = count + padlen;
-        skb->data_len += datalen;
-        skb->truesize += datalen;
-        skb->len += datalen;
         return 0;
 }
 
 int cxgb3i_conn_xmit_pdu(struct iscsi_task *task)
 {
-        struct iscsi_tcp_task *tcp_task = task->dd_data;
-        struct sk_buff *skb = tcp_task->dd_data;
         struct iscsi_tcp_conn *tcp_conn = task->conn->dd_data;
         struct cxgb3i_conn *cconn = tcp_conn->dd_data;
+        struct iscsi_tcp_task *tcp_task = task->dd_data;
+        struct cxgb3i_task_data *tdata = tcp_task->dd_data;
+        struct sk_buff *skb = tdata->skb;
         unsigned int datalen;
         int err;
 
@@ -300,13 +383,14 @@ int cxgb3i_conn_xmit_pdu(struct iscsi_task *task)
                 return 0;
 
         datalen = skb->data_len;
-        tcp_task->dd_data = NULL;
+        tdata->skb = NULL;
         err = cxgb3i_c3cn_send_pdus(cconn->cep->c3cn, skb);
-        cxgb3i_tx_debug("task 0x%p, skb 0x%p, len %u/%u, rv %d.\n",
-                        task, skb, skb->len, skb->data_len, err);
         if (err > 0) {
                 int pdulen = err;
 
+        cxgb3i_tx_debug("task 0x%p, skb 0x%p, len %u/%u, rv %d.\n",
+                        task, skb, skb->len, skb->data_len, err);
+
                 if (task->conn->hdrdgst_en)
                         pdulen += ISCSI_DIGEST_SIZE;
                 if (datalen && task->conn->datadgst_en)
@@ -325,12 +409,14 @@ int cxgb3i_conn_xmit_pdu(struct iscsi_task *task)
                 return err;
         }
         /* reset skb to send when we are called again */
-        tcp_task->dd_data = skb;
+        tdata->skb = skb;
         return -EAGAIN;
 }
 
 int cxgb3i_pdu_init(void)
 {
+        if (SKB_TX_HEADROOM > (512 * MAX_SKB_FRAGS))
+                skb_extra_headroom = SKB_TX_HEADROOM;
         pad_page = alloc_page(GFP_KERNEL);
         if (!pad_page)
                 return -ENOMEM;
@@ -366,7 +452,9 @@ void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn)
         skb = skb_peek(&c3cn->receive_queue);
         while (!err && skb) {
                 __skb_unlink(skb, &c3cn->receive_queue);
-                read += skb_ulp_pdulen(skb);
+                read += skb_rx_pdulen(skb);
+                cxgb3i_rx_debug("conn 0x%p, cn 0x%p, rx skb 0x%p, pdulen %u.\n",
+                                conn, c3cn, skb, skb_rx_pdulen(skb));
                 err = cxgb3i_conn_read_pdu_skb(conn, skb);
                 __kfree_skb(skb);
                 skb = skb_peek(&c3cn->receive_queue);
@@ -377,6 +465,11 @@ void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn)
                 cxgb3i_c3cn_rx_credits(c3cn, read);
         }
         conn->rxdata_octets += read;
+
+        if (err) {
+                cxgb3i_log_info("conn 0x%p rx failed err %d.\n", conn, err);
+                iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+        }
 }
 
 void cxgb3i_conn_tx_open(struct s3_conn *c3cn)
diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.h b/drivers/scsi/cxgb3i/cxgb3i_pdu.h
index a3f685cc2362..0770b23d90da 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.h
@@ -53,7 +53,7 @@ struct cpl_rx_data_ddp_norss {
 #define ULP2_FLAG_DCRC_ERROR            0x20
 #define ULP2_FLAG_PAD_ERROR             0x40
 
-void cxgb3i_conn_closing(struct s3_conn *);
+void cxgb3i_conn_closing(struct s3_conn *c3cn);
 void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn);
 void cxgb3i_conn_tx_open(struct s3_conn *c3cn);
 #endif
diff --git a/drivers/scsi/hptiop.c b/drivers/scsi/hptiop.c
index a48e4990fe12..34be88d7afa5 100644
--- a/drivers/scsi/hptiop.c
+++ b/drivers/scsi/hptiop.c
@@ -1251,6 +1251,7 @@ static struct pci_device_id hptiop_id_table[] = {
         { PCI_VDEVICE(TTI, 0x3530), (kernel_ulong_t)&hptiop_itl_ops },
         { PCI_VDEVICE(TTI, 0x3560), (kernel_ulong_t)&hptiop_itl_ops },
         { PCI_VDEVICE(TTI, 0x4322), (kernel_ulong_t)&hptiop_itl_ops },
+        { PCI_VDEVICE(TTI, 0x4321), (kernel_ulong_t)&hptiop_itl_ops },
         { PCI_VDEVICE(TTI, 0x4210), (kernel_ulong_t)&hptiop_itl_ops },
         { PCI_VDEVICE(TTI, 0x4211), (kernel_ulong_t)&hptiop_itl_ops },
         { PCI_VDEVICE(TTI, 0x4310), (kernel_ulong_t)&hptiop_itl_ops },
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 940dc32ff0dc..b82ffd90632e 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1040,12 +1040,11 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
                                 action = ACTION_FAIL;
                         break;
                 case ABORTED_COMMAND:
+                        action = ACTION_FAIL;
                         if (sshdr.asc == 0x10) { /* DIF */
                                 description = "Target Data Integrity Failure";
-                                action = ACTION_FAIL;
                                 error = -EILSEQ;
-                        } else
-                                action = ACTION_RETRY;
+                        }
                         break;
                 case NOT_READY:
                         /* If the device is in the process of becoming
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index d57566b8be0a..55310dbc10a6 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -107,6 +107,7 @@ static void scsi_disk_release(struct device *cdev);
 static void sd_print_sense_hdr(struct scsi_disk *, struct scsi_sense_hdr *);
 static void sd_print_result(struct scsi_disk *, int);
 
+static DEFINE_SPINLOCK(sd_index_lock);
 static DEFINE_IDA(sd_index_ida);
 
 /* This semaphore is used to mediate the 0->1 reference get in the
@@ -1914,7 +1915,9 @@ static int sd_probe(struct device *dev)
                 if (!ida_pre_get(&sd_index_ida, GFP_KERNEL))
                         goto out_put;
 
+                spin_lock(&sd_index_lock);
                 error = ida_get_new(&sd_index_ida, &index);
+                spin_unlock(&sd_index_lock);
         } while (error == -EAGAIN);
 
         if (error)
@@ -1936,7 +1939,9 @@ static int sd_probe(struct device *dev)
         return 0;
 
  out_free_index:
+        spin_lock(&sd_index_lock);
         ida_remove(&sd_index_ida, index);
+        spin_unlock(&sd_index_lock);
  out_put:
         put_disk(gd);
  out_free:
@@ -1986,7 +1991,9 @@ static void scsi_disk_release(struct device *dev)
         struct scsi_disk *sdkp = to_scsi_disk(dev);
         struct gendisk *disk = sdkp->disk;
         
+        spin_lock(&sd_index_lock);
         ida_remove(&sd_index_ida, sdkp->index);
+        spin_unlock(&sd_index_lock);
 
         disk->private_data = NULL;
         put_disk(disk);
diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c
index 48ff701d3a72..2552b9f325ee 100644
--- a/drivers/video/pxafb.c
+++ b/drivers/video/pxafb.c
@@ -2230,7 +2230,7 @@ static int __devexit pxafb_remove(struct platform_device *dev)
 
 static struct platform_driver pxafb_driver = {
         .probe          = pxafb_probe,
-        .remove         = pxafb_remove,
+        .remove         = __devexit_p(pxafb_remove),
         .suspend        = pxafb_suspend,
         .resume         = pxafb_resume,
         .driver         = {
diff --git a/fs/Makefile b/fs/Makefile
index 38bc735c67ad..dc20db348679 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -69,10 +69,12 @@ obj-$(CONFIG_DLM)		+= dlm/
 # Do not add any filesystems before this line
 obj-$(CONFIG_REISERFS_FS)       += reiserfs/
 obj-$(CONFIG_EXT3_FS)           += ext3/ # Before ext2 so root fs can be ext3
-obj-$(CONFIG_EXT4_FS)           += ext4/ # Before ext2 so root fs can be ext4
+obj-$(CONFIG_EXT2_FS)           += ext2/
+# We place ext4 after ext2 so plain ext2 root fs's are mounted using ext2
+# unless explicitly requested by rootfstype
+obj-$(CONFIG_EXT4_FS)           += ext4/
 obj-$(CONFIG_JBD)               += jbd/
 obj-$(CONFIG_JBD2)              += jbd2/
-obj-$(CONFIG_EXT2_FS)           += ext2/
 obj-$(CONFIG_CRAMFS)            += cramfs/
 obj-$(CONFIG_SQUASHFS)          += squashfs/
 obj-y                           += ramfs/
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 9a50b8052dcf..de9459b4cb94 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -609,7 +609,9 @@ int ext4_claim_free_blocks(struct ext4_sb_info *sbi,
  */
 int ext4_should_retry_alloc(struct super_block *sb, int *retries)
 {
-        if (!ext4_has_free_blocks(EXT4_SB(sb), 1) || (*retries)++ > 3)
+        if (!ext4_has_free_blocks(EXT4_SB(sb), 1) ||
+            (*retries)++ > 3 ||
+            !EXT4_SB(sb)->s_journal)
                 return 0;
 
         jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 51cdd13e1c31..c7fed5b18745 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2544,7 +2544,7 @@ retry:
 
                 ext4_journal_stop(handle);
 
-                if (mpd.retval == -ENOSPC) {
+                if ((mpd.retval == -ENOSPC) && sbi->s_journal) {
                         /* commit the transaction which would
                          * free blocks released in the transaction
                          * and try again
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index a5732c58f676..39d1993cfa13 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -3091,7 +3091,6 @@ static int ext4_freeze(struct super_block *sb)
 
                 /* Journal blocked and flushed, clear needs_recovery flag. */
                 EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
-                ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
                 error = ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
                 if (error)
                         goto out;
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index b97cdc516a8f..106c3ba50844 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -52,6 +52,7 @@ header-y += const.h
 header-y += cgroupstats.h
 header-y += cramfs_fs.h
 header-y += cycx_cfm.h
+header-y += dcbnl.h
 header-y += dlmconstants.h
 header-y += dlm_device.h
 header-y += dlm_netlink.h
diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 95837bfb5256..455d83219fae 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -65,23 +65,20 @@ extern void free_bootmem(unsigned long addr, unsigned long size);
 #define BOOTMEM_DEFAULT         0
 #define BOOTMEM_EXCLUSIVE       (1<<0)
 
+extern int reserve_bootmem(unsigned long addr,
+                           unsigned long size,
+                           int flags);
 extern int reserve_bootmem_node(pg_data_t *pgdat,
-                                 unsigned long physaddr,
-                                 unsigned long size,
-                                 int flags);
-#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags);
-#endif
+                                unsigned long physaddr,
+                                unsigned long size,
+                                int flags);
 
-extern void *__alloc_bootmem_nopanic(unsigned long size,
+extern void *__alloc_bootmem(unsigned long size,
                              unsigned long align,
                              unsigned long goal);
-extern void *__alloc_bootmem(unsigned long size,
+extern void *__alloc_bootmem_nopanic(unsigned long size,
                                      unsigned long align,
                                      unsigned long goal);
-extern void *__alloc_bootmem_low(unsigned long size,
-                                 unsigned long align,
-                                 unsigned long goal);
 extern void *__alloc_bootmem_node(pg_data_t *pgdat,
                                   unsigned long size,
                                   unsigned long align,
@@ -90,30 +87,35 @@ extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat,
                                   unsigned long size,
                                   unsigned long align,
                                   unsigned long goal);
+extern void *__alloc_bootmem_low(unsigned long size,
+                                 unsigned long align,
+                                 unsigned long goal);
 extern void *__alloc_bootmem_low_node(pg_data_t *pgdat,
                                       unsigned long size,
                                       unsigned long align,
                                       unsigned long goal);
-#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
+
 #define alloc_bootmem(x) \
         __alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_nopanic(x) \
         __alloc_bootmem_nopanic(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low(x) \
-        __alloc_bootmem_low(x, SMP_CACHE_BYTES, 0)
 #define alloc_bootmem_pages(x) \
         __alloc_bootmem(x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_pages_nopanic(x) \
         __alloc_bootmem_nopanic(x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
-#define alloc_bootmem_low_pages(x) \
-        __alloc_bootmem_low(x, PAGE_SIZE, 0)
 #define alloc_bootmem_node(pgdat, x) \
         __alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_pages_node(pgdat, x) \
         __alloc_bootmem_node(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+#define alloc_bootmem_pages_node_nopanic(pgdat, x) \
+        __alloc_bootmem_node_nopanic(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+
+#define alloc_bootmem_low(x) \
+        __alloc_bootmem_low(x, SMP_CACHE_BYTES, 0)
+#define alloc_bootmem_low_pages(x) \
+        __alloc_bootmem_low(x, PAGE_SIZE, 0)
 #define alloc_bootmem_low_pages_node(pgdat, x) \
         __alloc_bootmem_low_node(pgdat, x, PAGE_SIZE, 0)
-#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
 extern int reserve_bootmem_generic(unsigned long addr, unsigned long size,
                                    int flags);
diff --git a/include/linux/dcbnl.h b/include/linux/dcbnl.h
index b0ef274e0031..7d2e10006188 100644
--- a/include/linux/dcbnl.h
+++ b/include/linux/dcbnl.h
@@ -20,10 +20,12 @@
 #ifndef __LINUX_DCBNL_H__
 #define __LINUX_DCBNL_H__
 
+#include <linux/types.h>
+
 #define DCB_PROTO_VERSION 1
 
 struct dcbmsg {
-        unsigned char      dcb_family;
+        __u8               dcb_family;
         __u8               cmd;
         __u16              dcb_pad;
 };
diff --git a/include/linux/decompress/bunzip2.h b/include/linux/decompress/bunzip2.h
new file mode 100644
index 000000000000..115272137a9c
--- /dev/null
+++ b/include/linux/decompress/bunzip2.h
@@ -0,0 +1,10 @@
+#ifndef DECOMPRESS_BUNZIP2_H
+#define DECOMPRESS_BUNZIP2_H
+
+int bunzip2(unsigned char *inbuf, int len,
+            int(*fill)(void*, unsigned int),
+            int(*flush)(void*, unsigned int),
+            unsigned char *output,
+            int *pos,
+            void(*error)(char *x));
+#endif
diff --git a/include/linux/decompress/generic.h b/include/linux/decompress/generic.h
new file mode 100644
index 000000000000..6dfb856327bb
--- /dev/null
+++ b/include/linux/decompress/generic.h
@@ -0,0 +1,33 @@
+#ifndef DECOMPRESS_GENERIC_H
+#define DECOMPRESS_GENERIC_H
+
+/* Minimal chunksize to be read.
+ *Bzip2 prefers at least 4096
+ *Lzma prefers 0x10000 */
+#define COMPR_IOBUF_SIZE        4096
+
+typedef int (*decompress_fn) (unsigned char *inbuf, int len,
+                              int(*fill)(void*, unsigned int),
+                              int(*writebb)(void*, unsigned int),
+                              unsigned char *output,
+                              int *posp,
+                              void(*error)(char *x));
+
+/* inbuf   - input buffer
+ *len     - len of pre-read data in inbuf
+ *fill    - function to fill inbuf if empty
+ *writebb - function to write out outbug
+ *posp    - if non-null, input position (number of bytes read) will be
+ *        returned here
+ *
+ *If len != 0, the inbuf is initialized (with as much data), and fill
+ *should not be called
+ *If len = 0, the inbuf is allocated, but empty. Its size is IOBUF_SIZE
+ *fill should be called (repeatedly...) to read data, at most IOBUF_SIZE
+ */
+
+/* Utility routine to detect the decompression method */
+decompress_fn decompress_method(const unsigned char *inbuf, int len,
+                                const char **name);
+
+#endif
diff --git a/include/linux/decompress/inflate.h b/include/linux/decompress/inflate.h
new file mode 100644
index 000000000000..f9b06ccc3e5c
--- /dev/null
+++ b/include/linux/decompress/inflate.h
@@ -0,0 +1,13 @@
+#ifndef INFLATE_H
+#define INFLATE_H
+
+/* Other housekeeping constants */
+#define INBUFSIZ 4096
+
+int gunzip(unsigned char *inbuf, int len,
+           int(*fill)(void*, unsigned int),
+           int(*flush)(void*, unsigned int),
+           unsigned char *output,
+           int *pos,
+           void(*error_fn)(char *x));
+#endif
diff --git a/include/linux/decompress/mm.h b/include/linux/decompress/mm.h
new file mode 100644
index 000000000000..12ff8c3f1d05
--- /dev/null
+++ b/include/linux/decompress/mm.h
@@ -0,0 +1,87 @@
+/*
+ * linux/compr_mm.h
+ *
+ * Memory management for pre-boot and ramdisk uncompressors
+ *
+ * Authors: Alain Knaff <alain@knaff.lu>
+ *
+ */
+
+#ifndef DECOMPR_MM_H
+#define DECOMPR_MM_H
+
+#ifdef STATIC
+
+/* Code active when included from pre-boot environment: */
+
+/* A trivial malloc implementation, adapted from
+ *  malloc by Hannu Savolainen 1993 and Matthias Urlichs 1994
+ */
+static unsigned long malloc_ptr;
+static int malloc_count;
+
+static void *malloc(int size)
+{
+        void *p;
+
+        if (size < 0)
+                error("Malloc error");
+        if (!malloc_ptr)
+                malloc_ptr = free_mem_ptr;
+
+        malloc_ptr = (malloc_ptr + 3) & ~3;     /* Align */
+
+        p = (void *)malloc_ptr;
+        malloc_ptr += size;
+
+        if (free_mem_end_ptr && malloc_ptr >= free_mem_end_ptr)
+                error("Out of memory");
+
+        malloc_count++;
+        return p;
+}
+
+static void free(void *where)
+{
+        malloc_count--;
+        if (!malloc_count)
+                malloc_ptr = free_mem_ptr;
+}
+
+#define large_malloc(a) malloc(a)
+#define large_free(a) free(a)
+
+#define set_error_fn(x)
+
+#define INIT
+
+#else /* STATIC */
+
+/* Code active when compiled standalone for use when loading ramdisk: */
+
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/vmalloc.h>
+
+/* Use defines rather than static inline in order to avoid spurious
+ * warnings when not needed (indeed large_malloc / large_free are not
+ * needed by inflate */
+
+#define malloc(a) kmalloc(a, GFP_KERNEL)
+#define free(a) kfree(a)
+
+#define large_malloc(a) vmalloc(a)
+#define large_free(a) vfree(a)
+
+static void(*error)(char *m);
+#define set_error_fn(x) error = x;
+
+#define INIT __init
+#define STATIC
+
+#include <linux/init.h>
+
+#endif /* STATIC */
+
+#endif /* DECOMPR_MM_H */
diff --git a/include/linux/decompress/unlzma.h b/include/linux/decompress/unlzma.h
new file mode 100644
index 000000000000..7796538f1bf4
--- /dev/null
+++ b/include/linux/decompress/unlzma.h
@@ -0,0 +1,12 @@
+#ifndef DECOMPRESS_UNLZMA_H
+#define DECOMPRESS_UNLZMA_H
+
+int unlzma(unsigned char *, int,
+           int(*fill)(void*, unsigned int),
+           int(*flush)(void*, unsigned int),
+           unsigned char *output,
+           int *posp,
+           void(*error)(char *x)
+        );
+
+#endif
diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h
index cbc2f0cd631b..0adb0f91568c 100644
--- a/include/linux/io-mapping.h
+++ b/include/linux/io-mapping.h
@@ -91,8 +91,11 @@ io_mapping_unmap_atomic(void *vaddr)
 static inline void *
 io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset)
 {
+        resource_size_t phys_addr;
+
         BUG_ON(offset >= mapping->size);
-        resource_size_t phys_addr = mapping->base + offset;
+        phys_addr = mapping->base + offset;
+
         return ioremap_wc(phys_addr, PAGE_SIZE);
 }
 
diff --git a/include/linux/netfilter/xt_NFLOG.h b/include/linux/netfilter/xt_NFLOG.h
index cdcd0ed58f7a..4b36aeb46a10 100644
--- a/include/linux/netfilter/xt_NFLOG.h
+++ b/include/linux/netfilter/xt_NFLOG.h
@@ -2,7 +2,7 @@
 #define _XT_NFLOG_TARGET
 
 #define XT_NFLOG_DEFAULT_GROUP          0x1
-#define XT_NFLOG_DEFAULT_THRESHOLD      1
+#define XT_NFLOG_DEFAULT_THRESHOLD      0
 
 #define XT_NFLOG_MASK                   0x0
 
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 3577ffd90d45..545b068bcb70 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -76,52 +76,98 @@
 
 #ifdef CONFIG_SMP
 
+#ifdef CONFIG_HAVE_DYNAMIC_PER_CPU_AREA
+
+/* minimum unit size, also is the maximum supported allocation size */
+#define PCPU_MIN_UNIT_SIZE              (16UL << PAGE_SHIFT)
+
+/*
+ * PERCPU_DYNAMIC_RESERVE indicates the amount of free area to piggy
+ * back on the first chunk if arch is manually allocating and mapping
+ * it for faster access (as a part of large page mapping for example).
+ * Note that dynamic percpu allocator covers both static and dynamic
+ * areas, so these values are bigger than PERCPU_MODULE_RESERVE.
+ *
+ * On typical configuration with modules, the following values leave
+ * about 8k of free space on the first chunk after boot on both x86_32
+ * and 64 when module support is enabled.  When module support is
+ * disabled, it's much tighter.
+ */
+#ifndef PERCPU_DYNAMIC_RESERVE
+#  if BITS_PER_LONG > 32
+#    ifdef CONFIG_MODULES
+#      define PERCPU_DYNAMIC_RESERVE    (6 << PAGE_SHIFT)
+#    else
+#      define PERCPU_DYNAMIC_RESERVE    (4 << PAGE_SHIFT)
+#    endif
+#  else
+#    ifdef CONFIG_MODULES
+#      define PERCPU_DYNAMIC_RESERVE    (4 << PAGE_SHIFT)
+#    else
+#      define PERCPU_DYNAMIC_RESERVE    (2 << PAGE_SHIFT)
+#    endif
+#  endif
+#endif  /* PERCPU_DYNAMIC_RESERVE */
+
+extern void *pcpu_base_addr;
+
+typedef struct page * (*pcpu_get_page_fn_t)(unsigned int cpu, int pageno);
+typedef void (*pcpu_populate_pte_fn_t)(unsigned long addr);
+
+extern size_t __init pcpu_setup_first_chunk(pcpu_get_page_fn_t get_page_fn,
+                                        size_t static_size, size_t unit_size,
+                                        size_t free_size, void *base_addr,
+                                        pcpu_populate_pte_fn_t populate_pte_fn);
+
+/*
+ * Use this to get to a cpu's version of the per-cpu object
+ * dynamically allocated. Non-atomic access to the current CPU's
+ * version should probably be combined with get_cpu()/put_cpu().
+ */
+#define per_cpu_ptr(ptr, cpu)   SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu)))
+
+#else /* CONFIG_HAVE_DYNAMIC_PER_CPU_AREA */
+
 struct percpu_data {
         void *ptrs[1];
 };
 
 #define __percpu_disguise(pdata) (struct percpu_data *)~(unsigned long)(pdata)
-/* 
- * Use this to get to a cpu's version of the per-cpu object dynamically
- * allocated. Non-atomic access to the current CPU's version should
- * probably be combined with get_cpu()/put_cpu().
- */ 
-#define percpu_ptr(ptr, cpu)                              \
-({                                                        \
-        struct percpu_data *__p = __percpu_disguise(ptr); \
-        (__typeof__(ptr))__p->ptrs[(cpu)];                \
+
+#define per_cpu_ptr(ptr, cpu)                                           \
+({                                                                      \
+        struct percpu_data *__p = __percpu_disguise(ptr);               \
+        (__typeof__(ptr))__p->ptrs[(cpu)];                              \
 })
 
-extern void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask);
-extern void percpu_free(void *__pdata);
+#endif /* CONFIG_HAVE_DYNAMIC_PER_CPU_AREA */
+
+extern void *__alloc_percpu(size_t size, size_t align);
+extern void free_percpu(void *__pdata);
 
 #else /* CONFIG_SMP */
 
-#define percpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
+#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
 
-static __always_inline void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+static inline void *__alloc_percpu(size_t size, size_t align)
 {
-        return kzalloc(size, gfp);
+        /*
+         * Can't easily make larger alignment work with kmalloc.  WARN
+         * on it.  Larger alignment should only be used for module
+         * percpu sections on SMP for which this path isn't used.
+         */
+        WARN_ON_ONCE(align > SMP_CACHE_BYTES);
+        return kzalloc(size, GFP_KERNEL);
 }
 
-static inline void percpu_free(void *__pdata)
+static inline void free_percpu(void *p)
 {
-        kfree(__pdata);
+        kfree(p);
 }
 
 #endif /* CONFIG_SMP */
 
-#define percpu_alloc_mask(size, gfp, mask) \
-        __percpu_alloc_mask((size), (gfp), &(mask))
-
-#define percpu_alloc(size, gfp) percpu_alloc_mask((size), (gfp), cpu_online_map)
-
-/* (legacy) interface for use without CPU hotplug handling */
-
-#define __alloc_percpu(size)    percpu_alloc_mask((size), GFP_KERNEL, \
-                                                  cpu_possible_map)
-#define alloc_percpu(type)      (type *)__alloc_percpu(sizeof(type))
-#define free_percpu(ptr)        percpu_free((ptr))
-#define per_cpu_ptr(ptr, cpu)   percpu_ptr((ptr), (cpu))
+#define alloc_percpu(type)      (type *)__alloc_percpu(sizeof(type), \
+                                                       __alignof__(type))
 
 #endif /* __LINUX_PERCPU_H */
diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
index f3f697df1d71..80044a4f3ab9 100644
--- a/include/linux/rcuclassic.h
+++ b/include/linux/rcuclassic.h
@@ -181,4 +181,10 @@ extern long rcu_batches_completed_bh(void);
 #define rcu_enter_nohz()        do { } while (0)
 #define rcu_exit_nohz()         do { } while (0)
 
+/* A context switch is a grace period for rcuclassic. */
+static inline int rcu_blocking_is_gp(void)
+{
+        return num_online_cpus() == 1;
+}
+
 #endif /* __LINUX_RCUCLASSIC_H */
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 921340a7b71c..528343e6da51 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -52,6 +52,9 @@ struct rcu_head {
         void (*func)(struct rcu_head *head);
 };
 
+/* Internal to kernel, but needed by rcupreempt.h. */
+extern int rcu_scheduler_active;
+
 #if defined(CONFIG_CLASSIC_RCU)
 #include <linux/rcuclassic.h>
 #elif defined(CONFIG_TREE_RCU)
@@ -265,6 +268,7 @@ extern void rcu_barrier_sched(void);
 
 /* Internal to kernel */
 extern void rcu_init(void);
+extern void rcu_scheduler_starting(void);
 extern int rcu_needs_cpu(int cpu);
 
 #endif /* __LINUX_RCUPDATE_H */
diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h
index 3e05c09b54a2..74304b4538d8 100644
--- a/include/linux/rcupreempt.h
+++ b/include/linux/rcupreempt.h
@@ -142,4 +142,19 @@ static inline void rcu_exit_nohz(void)
 #define rcu_exit_nohz()         do { } while (0)
 #endif /* CONFIG_NO_HZ */
 
+/*
+ * A context switch is a grace period for rcupreempt synchronize_rcu()
+ * only during early boot, before the scheduler has been initialized.
+ * So, how the heck do we get a context switch?  Well, if the caller
+ * invokes synchronize_rcu(), they are willing to accept a context
+ * switch, so we simply pretend that one happened.
+ *
+ * After boot, there might be a blocked or preempted task in an RCU
+ * read-side critical section, so we cannot then take the fastpath.
+ */
+static inline int rcu_blocking_is_gp(void)
+{
+        return num_online_cpus() == 1 && !rcu_scheduler_active;
+}
+
 #endif /* __LINUX_RCUPREEMPT_H */
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index d4368b7975c3..a722fb67bb2d 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -326,4 +326,10 @@ static inline void rcu_exit_nohz(void)
 }
 #endif /* CONFIG_NO_HZ */
 
+/* A context switch is a grace period for rcutree. */
+static inline int rcu_blocking_is_gp(void)
+{
+        return num_online_cpus() == 1;
+}
+
 #endif /* __LINUX_RCUTREE_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index f0a50b20e8a0..a7c7698583bb 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2303,9 +2303,13 @@ extern long sched_group_rt_runtime(struct task_group *tg);
 extern int sched_group_set_rt_period(struct task_group *tg,
                                       long rt_period_us);
 extern long sched_group_rt_period(struct task_group *tg);
+extern int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk);
 #endif
 #endif
 
+extern int task_can_switch_user(struct user_struct *up,
+                                        struct task_struct *tsk);
+
 #ifdef CONFIG_TASK_XACCT
 static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
 {
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 6f3c603b0d67..6b58367d145e 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -41,13 +41,13 @@ static inline void pagefault_enable(void)
 #ifndef ARCH_HAS_NOCACHE_UACCESS
 
 static inline unsigned long __copy_from_user_inatomic_nocache(void *to,
-                const void __user *from, unsigned long n, unsigned long total)
+                                const void __user *from, unsigned long n)
 {
         return __copy_from_user_inatomic(to, from, n);
 }
 
 static inline unsigned long __copy_from_user_nocache(void *to,
-                const void __user *from, unsigned long n, unsigned long total)
+                                const void __user *from, unsigned long n)
 {
         return __copy_from_user(to, from, n);
 }
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 9c0890c7a06a..a43ebec3a7b9 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -95,6 +95,9 @@ extern struct vm_struct *remove_vm_area(const void *addr);
 
 extern int map_vm_area(struct vm_struct *area, pgprot_t prot,
                         struct page ***pages);
+extern int map_kernel_range_noflush(unsigned long start, unsigned long size,
+                                    pgprot_t prot, struct page **pages);
+extern void unmap_kernel_range_noflush(unsigned long addr, unsigned long size);
 extern void unmap_kernel_range(unsigned long addr, unsigned long size);
 
 /* Allocate/destroy a 'vmalloc' VM area. */
@@ -110,5 +113,6 @@ extern long vwrite(char *buf, char *addr, unsigned long count);
  */
 extern rwlock_t vmlist_lock;
 extern struct vm_struct *vmlist;
+extern __init void vm_area_register_early(struct vm_struct *vm, size_t align);
 
 #endif /* _LINUX_VMALLOC_H */
diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h
index e78afe7f28e3..c25068e38516 100644
--- a/include/net/netfilter/nf_conntrack_core.h
+++ b/include/net/netfilter/nf_conntrack_core.h
@@ -59,7 +59,7 @@ static inline int nf_conntrack_confirm(struct sk_buff *skb)
         struct nf_conn *ct = (struct nf_conn *)skb->nfct;
         int ret = NF_ACCEPT;
 
-        if (ct) {
+        if (ct && ct != &nf_conntrack_untracked) {
                 if (!nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct))
                         ret = __nf_conntrack_confirm(skb);
                 nf_ct_deliver_cached_events(ct);
diff --git a/init/Kconfig b/init/Kconfig
index f068071fcc5d..95a66131403a 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -101,6 +101,66 @@ config LOCALVERSION_AUTO
 
           which is done within the script "scripts/setlocalversion".)
 
+config HAVE_KERNEL_GZIP
+        bool
+
+config HAVE_KERNEL_BZIP2
+        bool
+
+config HAVE_KERNEL_LZMA
+        bool
+
+choice
+        prompt "Kernel compression mode"
+        default KERNEL_GZIP
+        depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA
+        help
+          The linux kernel is a kind of self-extracting executable.
+          Several compression algorithms are available, which differ
+          in efficiency, compression and decompression speed.
+          Compression speed is only relevant when building a kernel.
+          Decompression speed is relevant at each boot.
+
+          If you have any problems with bzip2 or lzma compressed
+          kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
+          version of this functionality (bzip2 only), for 2.4, was
+          supplied by Christian Ludwig)
+
+          High compression options are mostly useful for users, who
+          are low on disk space (embedded systems), but for whom ram
+          size matters less.
+
+          If in doubt, select 'gzip'
+
+config KERNEL_GZIP
+        bool "Gzip"
+        depends on HAVE_KERNEL_GZIP
+        help
+          The old and tried gzip compression. Its compression ratio is
+          the poorest among the 3 choices; however its speed (both
+          compression and decompression) is the fastest.
+
+config KERNEL_BZIP2
+        bool "Bzip2"
+        depends on HAVE_KERNEL_BZIP2
+        help
+          Its compression ratio and speed is intermediate.
+          Decompression speed is slowest among the three.  The kernel
+          size is about 10% smaller with bzip2, in comparison to gzip.
+          Bzip2 uses a large amount of memory. For modern kernels you
+          will need at least 8MB RAM or more for booting.
+
+config KERNEL_LZMA
+        bool "LZMA"
+        depends on HAVE_KERNEL_LZMA
+        help
+          The most recent compression algorithm.
+          Its ratio is best, decompression speed is between the other
+          two. Compression is slowest.  The kernel size is about 33%
+          smaller with LZMA in comparison to gzip.
+
+endchoice
+
 config SWAP
         bool "Support for paging of anonymous memory (swap)"
         depends on MMU && BLOCK
diff --git a/init/do_mounts_rd.c b/init/do_mounts_rd.c
index 0f0f0cf3ba9a..027a402708de 100644
--- a/init/do_mounts_rd.c
+++ b/init/do_mounts_rd.c
@@ -11,6 +11,9 @@
 #include "do_mounts.h"
 #include "../fs/squashfs/squashfs_fs.h"
 
+#include <linux/decompress/generic.h>
+
+
 int __initdata rd_prompt = 1;/* 1 = prompt for RAM disk, 0 = don't prompt */
 
 static int __init prompt_ramdisk(char *str)
@@ -29,7 +32,7 @@ static int __init ramdisk_start_setup(char *str)
 }
 __setup("ramdisk_start=", ramdisk_start_setup);
 
-static int __init crd_load(int in_fd, int out_fd);
+static int __init crd_load(int in_fd, int out_fd, decompress_fn deco);
 
 /*
  * This routine tries to find a RAM disk image to load, and returns the
@@ -38,15 +41,15 @@ static int __init crd_load(int in_fd, int out_fd);
  * numbers could not be found.
  *
  * We currently check for the following magic numbers:
- *      minix
- *      ext2
+ *      minix
+ *      ext2
  *      romfs
  *      cramfs
  *      squashfs
- *      gzip
+ *      gzip
  */
-static int __init 
-identify_ramdisk_image(int fd, int start_block)
+static int __init
+identify_ramdisk_image(int fd, int start_block, decompress_fn *decompressor)
 {
         const int size = 512;
         struct minix_super_block *minixsb;
@@ -56,6 +59,7 @@ identify_ramdisk_image(int fd, int start_block)
         struct squashfs_super_block *squashfsb;
         int nblocks = -1;
         unsigned char *buf;
+        const char *compress_name;
 
         buf = kmalloc(size, GFP_KERNEL);
         if (!buf)
@@ -69,18 +73,19 @@ identify_ramdisk_image(int fd, int start_block)
         memset(buf, 0xe5, size);
 
         /*
-         * Read block 0 to test for gzipped kernel
+         * Read block 0 to test for compressed kernel
          */
         sys_lseek(fd, start_block * BLOCK_SIZE, 0);
         sys_read(fd, buf, size);
 
-        /*
-         * If it matches the gzip magic numbers, return 0
-         */
-        if (buf[0] == 037 && ((buf[1] == 0213) || (buf[1] == 0236))) {
-                printk(KERN_NOTICE
-                       "RAMDISK: Compressed image found at block %d\n",
-                       start_block);
+        *decompressor = decompress_method(buf, size, &compress_name);
+        if (compress_name) {
+                printk(KERN_NOTICE "RAMDISK: %s image found at block %d\n",
+                       compress_name, start_block);
+                if (!*decompressor)
+                        printk(KERN_EMERG
+                               "RAMDISK: %s decompressor not configured!\n",
+                               compress_name);
                 nblocks = 0;
                 goto done;
         }
@@ -142,7 +147,7 @@ identify_ramdisk_image(int fd, int start_block)
         printk(KERN_NOTICE
                "RAMDISK: Couldn't find valid RAM disk image starting at %d.\n",
                start_block);
-        
+
 done:
         sys_lseek(fd, start_block * BLOCK_SIZE, 0);
         kfree(buf);
@@ -157,6 +162,7 @@ int __init rd_load_image(char *from)
         int nblocks, i, disk;
         char *buf = NULL;
         unsigned short rotate = 0;
+        decompress_fn decompressor = NULL;
 #if !defined(CONFIG_S390) && !defined(CONFIG_PPC_ISERIES)
         char rotator[4] = { '|' , '/' , '-' , '\\' };
 #endif
@@ -169,12 +175,12 @@ int __init rd_load_image(char *from)
         if (in_fd < 0)
                 goto noclose_input;
 
-        nblocks = identify_ramdisk_image(in_fd, rd_image_start);
+        nblocks = identify_ramdisk_image(in_fd, rd_image_start, &decompressor);
         if (nblocks < 0)
                 goto done;
 
         if (nblocks == 0) {
-                if (crd_load(in_fd, out_fd) == 0)
+                if (crd_load(in_fd, out_fd, decompressor) == 0)
                         goto successful_load;
                 goto done;
         }
@@ -200,7 +206,7 @@ int __init rd_load_image(char *from)
                        nblocks, rd_blocks);
                 goto done;
         }
-                
+
         /*
          * OK, time to copy in the data
          */
@@ -273,138 +279,48 @@ int __init rd_load_disk(int n)
         return rd_load_image("/dev/root");
 }
 
-/*
- * gzip declarations
- */
-
-#define OF(args)  args
-
-#ifndef memzero
-#define memzero(s, n)     memset ((s), 0, (n))
-#endif
-
-typedef unsigned char  uch;
-typedef unsigned short ush;
-typedef unsigned long  ulg;
-
-#define INBUFSIZ 4096
-#define WSIZE 0x8000    /* window size--must be a power of two, and */
-                        /*  at least 32K for zip's deflate method */
-
-static uch *inbuf;
-static uch *window;
-
-static unsigned insize;  /* valid bytes in inbuf */
-static unsigned inptr;   /* index of next byte to be processed in inbuf */
-static unsigned outcnt;  /* bytes in output buffer */
 static int exit_code;
-static int unzip_error;
-static long bytes_out;
+static int decompress_error;
 static int crd_infd, crd_outfd;
 
-#define get_byte()  (inptr < insize ? inbuf[inptr++] : fill_inbuf())
-                
-/* Diagnostic functions (stubbed out) */
-#define Assert(cond,msg)
-#define Trace(x)
-#define Tracev(x)
-#define Tracevv(x)
-#define Tracec(c,x)
-#define Tracecv(c,x)
-
-#define STATIC static
-#define INIT __init
-
-static int  __init fill_inbuf(void);
-static void __init flush_window(void);
-static void __init error(char *m);
-
-#define NO_INFLATE_MALLOC
-
-#include "../lib/inflate.c"
-
-/* ===========================================================================
- * Fill the input buffer. This is called only when the buffer is empty
- * and at least one byte is really needed.
- * Returning -1 does not guarantee that gunzip() will ever return.
- */
-static int __init fill_inbuf(void)
+static int __init compr_fill(void *buf, unsigned int len)
 {
-        if (exit_code) return -1;
-        
-        insize = sys_read(crd_infd, inbuf, INBUFSIZ);
-        if (insize == 0) {
-                error("RAMDISK: ran out of compressed data");
-                return -1;
-        }
-
-        inptr = 1;
-
-        return inbuf[0];
+        int r = sys_read(crd_infd, buf, len);
+        if (r < 0)
+                printk(KERN_ERR "RAMDISK: error while reading compressed data");
+        else if (r == 0)
+                printk(KERN_ERR "RAMDISK: EOF while reading compressed data");
+        return r;
 }
 
-/* ===========================================================================
- * Write the output window window[0..outcnt-1] and update crc and bytes_out.
- * (Used for the decompressed data only.)
- */
-static void __init flush_window(void)
+static int __init compr_flush(void *window, unsigned int outcnt)
 {
-    ulg c = crc;         /* temporary variable */
-    unsigned n, written;
-    uch *in, ch;
-    
-    written = sys_write(crd_outfd, window, outcnt);
-    if (written != outcnt && unzip_error == 0) {
-        printk(KERN_ERR "RAMDISK: incomplete write (%d != %d) %ld\n",
-               written, outcnt, bytes_out);
-        unzip_error = 1;
-    }
-    in = window;
-    for (n = 0; n < outcnt; n++) {
-            ch = *in++;
-            c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8);
-    }
-    crc = c;
-    bytes_out += (ulg)outcnt;
-    outcnt = 0;
+        int written = sys_write(crd_outfd, window, outcnt);
+        if (written != outcnt) {
+                if (decompress_error == 0)
+                        printk(KERN_ERR
+                               "RAMDISK: incomplete write (%d != %d)\n",
+                               written, outcnt);
+                decompress_error = 1;
+                return -1;
+        }
+        return outcnt;
 }
 
 static void __init error(char *x)
 {
         printk(KERN_ERR "%s\n", x);
         exit_code = 1;
-        unzip_error = 1;
+        decompress_error = 1;
 }
 
-static int __init crd_load(int in_fd, int out_fd)
+static int __init crd_load(int in_fd, int out_fd, decompress_fn deco)
 {
         int result;
-
-        insize = 0;             /* valid bytes in inbuf */
-        inptr = 0;              /* index of next byte to be processed in inbuf */
-        outcnt = 0;             /* bytes in output buffer */
-        exit_code = 0;
-        bytes_out = 0;
-        crc = (ulg)0xffffffffL; /* shift register contents */
-
         crd_infd = in_fd;
         crd_outfd = out_fd;
-        inbuf = kmalloc(INBUFSIZ, GFP_KERNEL);
-        if (!inbuf) {
-                printk(KERN_ERR "RAMDISK: Couldn't allocate gzip buffer\n");
-                return -1;
-        }
-        window = kmalloc(WSIZE, GFP_KERNEL);
-        if (!window) {
-                printk(KERN_ERR "RAMDISK: Couldn't allocate gzip window\n");
-                kfree(inbuf);
-                return -1;
-        }
-        makecrc();
-        result = gunzip();
-        if (unzip_error)
+        result = deco(NULL, 0, compr_fill, compr_flush, NULL, NULL, error);
+        if (decompress_error)
                 result = 1;
-        kfree(inbuf);
-        kfree(window);
         return result;
 }
diff --git a/init/initramfs.c b/init/initramfs.c
index d9c941c0c3ca..7dcde7ea6603 100644
--- a/init/initramfs.c
+++ b/init/initramfs.c
@@ -390,11 +390,13 @@ static int __init write_buffer(char *buf, unsigned len)
         return len - count;
 }
 
-static void __init flush_buffer(char *buf, unsigned len)
+static int __init flush_buffer(void *bufv, unsigned len)
 {
+        char *buf = (char *) bufv;
         int written;
+        int origLen = len;
         if (message)
-                return;
+                return -1;
         while ((written = write_buffer(buf, len)) < len && !message) {
                 char c = buf[written];
                 if (c == '0') {
@@ -408,84 +410,28 @@ static void __init flush_buffer(char *buf, unsigned len)
                 } else
                         error("junk in compressed archive");
         }
+        return origLen;
 }
 
-/*
- * gzip declarations
- */
+static unsigned my_inptr;   /* index of next byte to be processed in inbuf */
 
-#define OF(args)  args
-
-#ifndef memzero
-#define memzero(s, n)     memset ((s), 0, (n))
-#endif
-
-typedef unsigned char  uch;
-typedef unsigned short ush;
-typedef unsigned long  ulg;
-
-#define WSIZE 0x8000    /* window size--must be a power of two, and */
-                        /*  at least 32K for zip's deflate method */
-
-static uch *inbuf;
-static uch *window;
-
-static unsigned insize;  /* valid bytes in inbuf */
-static unsigned inptr;   /* index of next byte to be processed in inbuf */
-static unsigned outcnt;  /* bytes in output buffer */
-static long bytes_out;
-
-#define get_byte()  (inptr < insize ? inbuf[inptr++] : -1)
-                
-/* Diagnostic functions (stubbed out) */
-#define Assert(cond,msg)
-#define Trace(x)
-#define Tracev(x)
-#define Tracevv(x)
-#define Tracec(c,x)
-#define Tracecv(c,x)
-
-#define STATIC static
-#define INIT __init
-
-static void __init flush_window(void);
-static void __init error(char *m);
-
-#define NO_INFLATE_MALLOC
-
-#include "../lib/inflate.c"
-
-/* ===========================================================================
- * Write the output window window[0..outcnt-1] and update crc and bytes_out.
- * (Used for the decompressed data only.)
- */
-static void __init flush_window(void)
-{
-        ulg c = crc;         /* temporary variable */
-        unsigned n;
-        uch *in, ch;
-
-        flush_buffer(window, outcnt);
-        in = window;
-        for (n = 0; n < outcnt; n++) {
-                ch = *in++;
-                c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8);
-        }
-        crc = c;
-        bytes_out += (ulg)outcnt;
-        outcnt = 0;
-}
+#include <linux/decompress/generic.h>
 
 static char * __init unpack_to_rootfs(char *buf, unsigned len, int check_only)
 {
         int written;
+        decompress_fn decompress;
+        const char *compress_name;
+        static __initdata char msg_buf[64];
+
         dry_run = check_only;
         header_buf = kmalloc(110, GFP_KERNEL);
         symlink_buf = kmalloc(PATH_MAX + N_ALIGN(PATH_MAX) + 1, GFP_KERNEL);
         name_buf = kmalloc(N_ALIGN(PATH_MAX), GFP_KERNEL);
-        window = kmalloc(WSIZE, GFP_KERNEL);
-        if (!window || !header_buf || !symlink_buf || !name_buf)
+
+        if (!header_buf || !symlink_buf || !name_buf)
                 panic("can't allocate buffers");
+
         state = Start;
         this_header = 0;
         message = NULL;
@@ -505,22 +451,25 @@ static char * __init unpack_to_rootfs(char *buf, unsigned len, int check_only)
                         continue;
                 }
                 this_header = 0;
-                insize = len;
-                inbuf = buf;
-                inptr = 0;
-                outcnt = 0;             /* bytes in output buffer */
-                bytes_out = 0;
-                crc = (ulg)0xffffffffL; /* shift register contents */
-                makecrc();
-                gunzip();
+                decompress = decompress_method(buf, len, &compress_name);
+                if (decompress)
+                        decompress(buf, len, NULL, flush_buffer, NULL,
+                                   &my_inptr, error);
+                else if (compress_name) {
+                        if (!message) {
+                                snprintf(msg_buf, sizeof msg_buf,
+                                         "compression method %s not configured",
+                                         compress_name);
+                                message = msg_buf;
+                        }
+                }
                 if (state != Reset)
-                        error("junk in gzipped archive");
-                this_header = saved_offset + inptr;
-                buf += inptr;
-                len -= inptr;
+                        error("junk in compressed archive");
+                this_header = saved_offset + my_inptr;
+                buf += my_inptr;
+                len -= my_inptr;
         }
         dir_utime();
-        kfree(window);
         kfree(name_buf);
         kfree(symlink_buf);
         kfree(header_buf);
@@ -579,7 +528,7 @@ static int __init populate_rootfs(void)
         char *err = unpack_to_rootfs(__initramfs_start,
                          __initramfs_end - __initramfs_start, 0);
         if (err)
-                panic(err);
+                panic(err);     /* Failed to decompress INTERNAL initramfs */
         if (initrd_start) {
 #ifdef CONFIG_BLK_DEV_RAM
                 int fd;
@@ -605,9 +554,12 @@ static int __init populate_rootfs(void)
                 printk(KERN_INFO "Unpacking initramfs...");
                 err = unpack_to_rootfs((char *)initrd_start,
                         initrd_end - initrd_start, 0);
-                if (err)
-                        panic(err);
-                printk(" done\n");
+                if (err) {
+                        printk(" failed!\n");
+                        printk(KERN_EMERG "%s\n", err);
+                } else {
+                        printk(" done\n");
+                }
                 free_initrd();
 #endif
         }
diff --git a/init/main.c b/init/main.c
index 6441083f8273..6bf83afd654d 100644
--- a/init/main.c
+++ b/init/main.c
@@ -98,7 +98,7 @@ static inline void mark_rodata_ro(void) { }
 extern void tc_init(void);
 #endif
 
-enum system_states system_state;
+enum system_states system_state __read_mostly;
 EXPORT_SYMBOL(system_state);
 
 /*
@@ -464,6 +464,7 @@ static noinline void __init_refok rest_init(void)
          * at least once to get things moving:
          */
         init_idle_bootup_task(current);
+        rcu_scheduler_starting();
         preempt_enable_no_resched();
         schedule();
         preempt_disable();
diff --git a/kernel/module.c b/kernel/module.c
index ba22484a987e..1f0657ae555b 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -51,6 +51,7 @@
 #include <linux/tracepoint.h>
 #include <linux/ftrace.h>
 #include <linux/async.h>
+#include <linux/percpu.h>
 
 #if 0
 #define DEBUGP printk
@@ -366,6 +367,34 @@ static struct module *find_module(const char *name)
 }
 
 #ifdef CONFIG_SMP
+
+#ifdef CONFIG_HAVE_DYNAMIC_PER_CPU_AREA
+
+static void *percpu_modalloc(unsigned long size, unsigned long align,
+                             const char *name)
+{
+        void *ptr;
+
+        if (align > PAGE_SIZE) {
+                printk(KERN_WARNING "%s: per-cpu alignment %li > %li\n",
+                       name, align, PAGE_SIZE);
+                align = PAGE_SIZE;
+        }
+
+        ptr = __alloc_percpu(size, align);
+        if (!ptr)
+                printk(KERN_WARNING
+                       "Could not allocate %lu bytes percpu data\n", size);
+        return ptr;
+}
+
+static void percpu_modfree(void *freeme)
+{
+        free_percpu(freeme);
+}
+
+#else /* ... !CONFIG_HAVE_DYNAMIC_PER_CPU_AREA */
+
 /* Number of blocks used and allocated. */
 static unsigned int pcpu_num_used, pcpu_num_allocated;
 /* Size of each block.  -ve means used. */
@@ -480,21 +509,6 @@ static void percpu_modfree(void *freeme)
         }
 }
 
-static unsigned int find_pcpusec(Elf_Ehdr *hdr,
-                                 Elf_Shdr *sechdrs,
-                                 const char *secstrings)
-{
-        return find_sec(hdr, sechdrs, secstrings, ".data.percpu");
-}
-
-static void percpu_modcopy(void *pcpudest, const void *from, unsigned long size)
-{
-        int cpu;
-
-        for_each_possible_cpu(cpu)
-                memcpy(pcpudest + per_cpu_offset(cpu), from, size);
-}
-
 static int percpu_modinit(void)
 {
         pcpu_num_used = 2;
@@ -513,7 +527,26 @@ static int percpu_modinit(void)
         return 0;
 }
 __initcall(percpu_modinit);
+
+#endif /* CONFIG_HAVE_DYNAMIC_PER_CPU_AREA */
+
+static unsigned int find_pcpusec(Elf_Ehdr *hdr,
+                                 Elf_Shdr *sechdrs,
+                                 const char *secstrings)
+{
+        return find_sec(hdr, sechdrs, secstrings, ".data.percpu");
+}
+
+static void percpu_modcopy(void *pcpudest, const void *from, unsigned long size)
+{
+        int cpu;
+
+        for_each_possible_cpu(cpu)
+                memcpy(pcpudest + per_cpu_offset(cpu), from, size);
+}
+
 #else /* ... !CONFIG_SMP */
+
 static inline void *percpu_modalloc(unsigned long size, unsigned long align,
                                     const char *name)
 {
@@ -535,6 +568,7 @@ static inline void percpu_modcopy(void *pcpudst, const void *src,
         /* pcpusec should be 0, and size of that section should be 0. */
         BUG_ON(size != 0);
 }
+
 #endif /* CONFIG_SMP */
 
 #define MODINFO_ATTR(field)     \
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index bd5a9003497c..654c640a6b9c 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -679,8 +679,8 @@ int rcu_needs_cpu(int cpu)
 void rcu_check_callbacks(int cpu, int user)
 {
         if (user ||
-            (idle_cpu(cpu) && !in_softirq() &&
-                                hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
+            (idle_cpu(cpu) && rcu_scheduler_active &&
+             !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
 
                 /*
                  * Get here if this CPU took its interrupt from user
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index d92a76a881aa..cae8a059cf47 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -44,6 +44,7 @@
 #include <linux/cpu.h>
 #include <linux/mutex.h>
 #include <linux/module.h>
+#include <linux/kernel_stat.h>
 
 enum rcu_barrier {
         RCU_BARRIER_STD,
@@ -55,6 +56,7 @@ static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL};
 static atomic_t rcu_barrier_cpu_count;
 static DEFINE_MUTEX(rcu_barrier_mutex);
 static struct completion rcu_barrier_completion;
+int rcu_scheduler_active __read_mostly;
 
 /*
  * Awaken the corresponding synchronize_rcu() instance now that a
@@ -80,6 +82,10 @@ void wakeme_after_rcu(struct rcu_head  *head)
 void synchronize_rcu(void)
 {
         struct rcu_synchronize rcu;
+
+        if (rcu_blocking_is_gp())
+                return;
+
         init_completion(&rcu.completion);
         /* Will wake me after RCU finished. */
         call_rcu(&rcu.head, wakeme_after_rcu);
@@ -175,3 +181,9 @@ void __init rcu_init(void)
         __rcu_init();
 }
 
+void rcu_scheduler_starting(void)
+{
+        WARN_ON(num_online_cpus() != 1);
+        WARN_ON(nr_context_switches() > 0);
+        rcu_scheduler_active = 1;
+}
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index 33cfc50781f9..5d59e850fb71 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -1181,6 +1181,9 @@ void __synchronize_sched(void)
 {
         struct rcu_synchronize rcu;
 
+        if (num_online_cpus() == 1)
+                return;  /* blocking is gp if only one CPU! */
+
         init_completion(&rcu.completion);
         /* Will wake me after RCU finished. */
         call_rcu_sched(&rcu.head, wakeme_after_rcu);
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index b2fd602a6f6f..97ce31579ec0 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -948,8 +948,8 @@ static void rcu_do_batch(struct rcu_data *rdp)
 void rcu_check_callbacks(int cpu, int user)
 {
         if (user ||
-            (idle_cpu(cpu) && !in_softirq() &&
-                                hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
+            (idle_cpu(cpu) && rcu_scheduler_active &&
+             !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
 
                 /*
                  * Get here if this CPU took its interrupt from user
diff --git a/kernel/sched.c b/kernel/sched.c
index 7d97ff7c4478..0a76d0b6f215 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -223,7 +223,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
 {
         ktime_t now;
 
-        if (rt_bandwidth_enabled() && rt_b->rt_runtime == RUNTIME_INF)
+        if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF)
                 return;
 
         if (hrtimer_active(&rt_b->rt_period_timer))
@@ -9219,6 +9219,16 @@ static int sched_rt_global_constraints(void)
 
         return ret;
 }
+
+int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk)
+{
+        /* Don't accept realtime tasks when there is no way for them to run */
+        if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0)
+                return 0;
+
+        return 1;
+}
+
 #else /* !CONFIG_RT_GROUP_SCHED */
 static int sched_rt_global_constraints(void)
 {
@@ -9312,8 +9322,7 @@ cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp,
                       struct task_struct *tsk)
 {
 #ifdef CONFIG_RT_GROUP_SCHED
-        /* Don't accept realtime tasks when there is no way for them to run */
-        if (rt_task(tsk) && cgroup_tg(cgrp)->rt_bandwidth.rt_runtime == 0)
+        if (!sched_rt_can_attach(cgroup_tg(cgrp), tsk))
                 return -EINVAL;
 #else
         /* We don't support RT-tasks being in separate groups */
@@ -9476,7 +9485,7 @@ cpuacct_destroy(struct cgroup_subsys *ss, struct cgroup *cgrp)
 
 static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu)
 {
-        u64 *cpuusage = percpu_ptr(ca->cpuusage, cpu);
+        u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
         u64 data;
 
 #ifndef CONFIG_64BIT
@@ -9495,7 +9504,7 @@ static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu)
 
 static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val)
 {
-        u64 *cpuusage = percpu_ptr(ca->cpuusage, cpu);
+        u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
 
 #ifndef CONFIG_64BIT
         /*
@@ -9591,7 +9600,7 @@ static void cpuacct_charge(struct task_struct *tsk, u64 cputime)
         ca = task_ca(tsk);
 
         for (; ca; ca = ca->parent) {
-                u64 *cpuusage = percpu_ptr(ca->cpuusage, cpu);
+                u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
                 *cpuusage += cputime;
         }
 }
diff --git a/kernel/seccomp.c b/kernel/seccomp.c
index ad64fcb731f2..57d4b13b631d 100644
--- a/kernel/seccomp.c
+++ b/kernel/seccomp.c
@@ -8,6 +8,7 @@
 
 #include <linux/seccomp.h>
 #include <linux/sched.h>
+#include <linux/compat.h>
 
 /* #define SECCOMP_DEBUG 1 */
 #define NR_SECCOMP_MODES 1
@@ -22,7 +23,7 @@ static int mode1_syscalls[] = {
         0, /* null terminated */
 };
 
-#ifdef TIF_32BIT
+#ifdef CONFIG_COMPAT
 static int mode1_syscalls_32[] = {
         __NR_seccomp_read_32, __NR_seccomp_write_32, __NR_seccomp_exit_32, __NR_seccomp_sigreturn_32,
         0, /* null terminated */
@@ -37,8 +38,8 @@ void __secure_computing(int this_syscall)
         switch (mode) {
         case 1:
                 syscall = mode1_syscalls;
-#ifdef TIF_32BIT
-                if (test_thread_flag(TIF_32BIT))
+#ifdef CONFIG_COMPAT
+                if (is_compat_task())
                         syscall = mode1_syscalls_32;
 #endif
                 do {
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 0cd415ee62a2..74541ca49536 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -170,7 +170,7 @@ int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
          * doesn't hit this CPU until we're ready. */
         get_cpu();
         for_each_online_cpu(i) {
-                sm_work = percpu_ptr(stop_machine_work, i);
+                sm_work = per_cpu_ptr(stop_machine_work, i);
                 INIT_WORK(sm_work, stop_cpu);
                 queue_work_on(i, stop_machine_wq, sm_work);
         }
diff --git a/kernel/sys.c b/kernel/sys.c
index f145c415bc16..37f458e6882a 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -559,7 +559,7 @@ error:
         abort_creds(new);
         return retval;
 }
-  
+
 /*
  * change the user struct in a credentials set to match the new UID
  */
@@ -571,6 +571,11 @@ static int set_user(struct cred *new)
         if (!new_user)
                 return -EAGAIN;
 
+        if (!task_can_switch_user(new_user, current)) {
+                free_uid(new_user);
+                return -EINVAL;
+        }
+
         if (atomic_read(&new_user->processes) >=
                                 current->signal->rlim[RLIMIT_NPROC].rlim_cur &&
                         new_user != INIT_USER) {
@@ -631,10 +636,11 @@ SYSCALL_DEFINE2(setreuid, uid_t, ruid, uid_t, euid)
                         goto error;
         }
 
-        retval = -EAGAIN;
-        if (new->uid != old->uid && set_user(new) < 0)
-                goto error;
-
+        if (new->uid != old->uid) {
+                retval = set_user(new);
+                if (retval < 0)
+                        goto error;
+        }
         if (ruid != (uid_t) -1 ||
             (euid != (uid_t) -1 && euid != old->uid))
                 new->suid = new->euid;
@@ -680,9 +686,10 @@ SYSCALL_DEFINE1(setuid, uid_t, uid)
         retval = -EPERM;
         if (capable(CAP_SETUID)) {
                 new->suid = new->uid = uid;
-                if (uid != old->uid && set_user(new) < 0) {
-                        retval = -EAGAIN;
-                        goto error;
+                if (uid != old->uid) {
+                        retval = set_user(new);
+                        if (retval < 0)
+                                goto error;
                 }
         } else if (uid != old->uid && uid != new->suid) {
                 goto error;
@@ -734,11 +741,13 @@ SYSCALL_DEFINE3(setresuid, uid_t, ruid, uid_t, euid, uid_t, suid)
                         goto error;
         }
 
-        retval = -EAGAIN;
         if (ruid != (uid_t) -1) {
                 new->uid = ruid;
-                if (ruid != old->uid && set_user(new) < 0)
-                        goto error;
+                if (ruid != old->uid) {
+                        retval = set_user(new);
+                        if (retval < 0)
+                                goto error;
+                }
         }
         if (euid != (uid_t) -1)
                 new->euid = euid;
diff --git a/kernel/user.c b/kernel/user.c
index 3551ac742395..6a9b696128c8 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -362,6 +362,24 @@ static void free_user(struct user_struct *up, unsigned long flags)
 
 #endif
 
+#if defined(CONFIG_RT_GROUP_SCHED) && defined(CONFIG_USER_SCHED)
+/*
+ * We need to check if a setuid can take place. This function should be called
+ * before successfully completing the setuid.
+ */
+int task_can_switch_user(struct user_struct *up, struct task_struct *tsk)
+{
+
+        return sched_rt_can_attach(up->tg, tsk);
+
+}
+#else
+int task_can_switch_user(struct user_struct *up, struct task_struct *tsk)
+{
+        return 1;
+}
+#endif
+
 /*
  * Locate the user_struct for the passed UID.  If found, take a ref on it.  The
  * caller must undo that ref with free_uid().
diff --git a/lib/Kconfig b/lib/Kconfig
index 03c2c24b9083..daa481824d9c 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -98,6 +98,20 @@ config LZO_DECOMPRESS
         tristate
 
 #
+# These all provide a common interface (hence the apparent duplication with
+# ZLIB_INFLATE; DECOMPRESS_GZIP is just a wrapper.)
+#
+config DECOMPRESS_GZIP
+        select ZLIB_INFLATE
+        tristate
+
+config DECOMPRESS_BZIP2
+        tristate
+
+config DECOMPRESS_LZMA
+        tristate
+
+#
 # Generic allocator support is selected if needed
 #
 config GENERIC_ALLOCATOR
diff --git a/lib/Makefile b/lib/Makefile
index 32b0e64ded27..790de7c25d0d 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -11,7 +11,8 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \
          rbtree.o radix-tree.o dump_stack.o \
          idr.o int_sqrt.o extable.o prio_tree.o \
          sha1.o irq_regs.o reciprocal_div.o argv_split.o \
-         proportions.o prio_heap.o ratelimit.o show_mem.o is_single_threaded.o
+         proportions.o prio_heap.o ratelimit.o show_mem.o \
+         is_single_threaded.o decompress.o
 
 lib-$(CONFIG_MMU) += ioremap.o
 lib-$(CONFIG_SMP) += cpumask.o
@@ -65,6 +66,10 @@ obj-$(CONFIG_REED_SOLOMON) += reed_solomon/
 obj-$(CONFIG_LZO_COMPRESS) += lzo/
 obj-$(CONFIG_LZO_DECOMPRESS) += lzo/
 
+lib-$(CONFIG_DECOMPRESS_GZIP) += decompress_inflate.o
+lib-$(CONFIG_DECOMPRESS_BZIP2) += decompress_bunzip2.o
+lib-$(CONFIG_DECOMPRESS_LZMA) += decompress_unlzma.o
+
 obj-$(CONFIG_TEXTSEARCH) += textsearch.o
 obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
 obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o
diff --git a/lib/decompress.c b/lib/decompress.c
new file mode 100644
index 000000000000..d2842f571674
--- /dev/null
+++ b/lib/decompress.c
@@ -0,0 +1,54 @@
+/*
+ * decompress.c
+ *
+ * Detect the decompression method based on magic number
+ */
+
+#include <linux/decompress/generic.h>
+
+#include <linux/decompress/bunzip2.h>
+#include <linux/decompress/unlzma.h>
+#include <linux/decompress/inflate.h>
+
+#include <linux/types.h>
+#include <linux/string.h>
+
+#ifndef CONFIG_DECOMPRESS_GZIP
+# define gunzip NULL
+#endif
+#ifndef CONFIG_DECOMPRESS_BZIP2
+# define bunzip2 NULL
+#endif
+#ifndef CONFIG_DECOMPRESS_LZMA
+# define unlzma NULL
+#endif
+
+static const struct compress_format {
+        unsigned char magic[2];
+        const char *name;
+        decompress_fn decompressor;
+} compressed_formats[] = {
+        { {037, 0213}, "gzip", gunzip },
+        { {037, 0236}, "gzip", gunzip },
+        { {0x42, 0x5a}, "bzip2", bunzip2 },
+        { {0x5d, 0x00}, "lzma", unlzma },
+        { {0, 0}, NULL, NULL }
+};
+
+decompress_fn decompress_method(const unsigned char *inbuf, int len,
+                                const char **name)
+{
+        const struct compress_format *cf;
+
+        if (len < 2)
+                return NULL;    /* Need at least this much... */
+
+        for (cf = compressed_formats; cf->name; cf++) {
+                if (!memcmp(inbuf, cf->magic, 2))
+                        break;
+
+        }
+        if (name)
+                *name = cf->name;
+        return cf->decompressor;
+}
diff --git a/lib/decompress_bunzip2.c b/lib/decompress_bunzip2.c
new file mode 100644
index 000000000000..5d3ddb5fcfd9
--- /dev/null
+++ b/lib/decompress_bunzip2.c
@@ -0,0 +1,735 @@
+/* vi: set sw = 4 ts = 4: */
+/*      Small bzip2 deflate implementation, by Rob Landley (rob@landley.net).
+
+        Based on bzip2 decompression code by Julian R Seward (jseward@acm.org),
+        which also acknowledges contributions by Mike Burrows, David Wheeler,
+        Peter Fenwick, Alistair Moffat, Radford Neal, Ian H. Witten,
+        Robert Sedgewick, and Jon L. Bentley.
+
+        This code is licensed under the LGPLv2:
+                LGPL (http://www.gnu.org/copyleft/lgpl.html
+*/
+
+/*
+        Size and speed optimizations by Manuel Novoa III  (mjn3@codepoet.org).
+
+        More efficient reading of Huffman codes, a streamlined read_bunzip()
+        function, and various other tweaks.  In (limited) tests, approximately
+        20% faster than bzcat on x86 and about 10% faster on arm.
+
+        Note that about 2/3 of the time is spent in read_unzip() reversing
+        the Burrows-Wheeler transformation.  Much of that time is delay
+        resulting from cache misses.
+
+        I would ask that anyone benefiting from this work, especially those
+        using it in commercial products, consider making a donation to my local
+        non-profit hospice organization in the name of the woman I loved, who
+        passed away Feb. 12, 2003.
+
+                In memory of Toni W. Hagan
+
+                Hospice of Acadiana, Inc.
+                2600 Johnston St., Suite 200
+                Lafayette, LA 70503-3240
+
+                Phone (337) 232-1234 or 1-800-738-2226
+                Fax   (337) 232-1297
+
+                http://www.hospiceacadiana.com/
+
+        Manuel
+ */
+
+/*
+        Made it fit for running in Linux Kernel by Alain Knaff (alain@knaff.lu)
+*/
+
+
+#ifndef STATIC
+#include <linux/decompress/bunzip2.h>
+#endif /* !STATIC */
+
+#include <linux/decompress/mm.h>
+
+#ifndef INT_MAX
+#define INT_MAX 0x7fffffff
+#endif
+
+/* Constants for Huffman coding */
+#define MAX_GROUPS              6
+#define GROUP_SIZE              50      /* 64 would have been more efficient */
+#define MAX_HUFCODE_BITS        20      /* Longest Huffman code allowed */
+#define MAX_SYMBOLS             258     /* 256 literals + RUNA + RUNB */
+#define SYMBOL_RUNA             0
+#define SYMBOL_RUNB             1
+
+/* Status return values */
+#define RETVAL_OK                       0
+#define RETVAL_LAST_BLOCK               (-1)
+#define RETVAL_NOT_BZIP_DATA            (-2)
+#define RETVAL_UNEXPECTED_INPUT_EOF     (-3)
+#define RETVAL_UNEXPECTED_OUTPUT_EOF    (-4)
+#define RETVAL_DATA_ERROR               (-5)
+#define RETVAL_OUT_OF_MEMORY            (-6)
+#define RETVAL_OBSOLETE_INPUT           (-7)
+
+/* Other housekeeping constants */
+#define BZIP2_IOBUF_SIZE                4096
+
+/* This is what we know about each Huffman coding group */
+struct group_data {
+        /* We have an extra slot at the end of limit[] for a sentinal value. */
+        int limit[MAX_HUFCODE_BITS+1];
+        int base[MAX_HUFCODE_BITS];
+        int permute[MAX_SYMBOLS];
+        int minLen, maxLen;
+};
+
+/* Structure holding all the housekeeping data, including IO buffers and
+   memory that persists between calls to bunzip */
+struct bunzip_data {
+        /* State for interrupting output loop */
+        int writeCopies, writePos, writeRunCountdown, writeCount, writeCurrent;
+        /* I/O tracking data (file handles, buffers, positions, etc.) */
+        int (*fill)(void*, unsigned int);
+        int inbufCount, inbufPos /*, outbufPos*/;
+        unsigned char *inbuf /*,*outbuf*/;
+        unsigned int inbufBitCount, inbufBits;
+        /* The CRC values stored in the block header and calculated from the
+        data */
+        unsigned int crc32Table[256], headerCRC, totalCRC, writeCRC;
+        /* Intermediate buffer and its size (in bytes) */
+        unsigned int *dbuf, dbufSize;
+        /* These things are a bit too big to go on the stack */
+        unsigned char selectors[32768];         /* nSelectors = 15 bits */
+        struct group_data groups[MAX_GROUPS];   /* Huffman coding tables */
+        int io_error;                   /* non-zero if we have IO error */
+};
+
+
+/* Return the next nnn bits of input.  All reads from the compressed input
+   are done through this function.  All reads are big endian */
+static unsigned int INIT get_bits(struct bunzip_data *bd, char bits_wanted)
+{
+        unsigned int bits = 0;
+
+        /* If we need to get more data from the byte buffer, do so.
+           (Loop getting one byte at a time to enforce endianness and avoid
+           unaligned access.) */
+        while (bd->inbufBitCount < bits_wanted) {
+                /* If we need to read more data from file into byte buffer, do
+                   so */
+                if (bd->inbufPos == bd->inbufCount) {
+                        if (bd->io_error)
+                                return 0;
+                        bd->inbufCount = bd->fill(bd->inbuf, BZIP2_IOBUF_SIZE);
+                        if (bd->inbufCount <= 0) {
+                                bd->io_error = RETVAL_UNEXPECTED_INPUT_EOF;
+                                return 0;
+                        }
+                        bd->inbufPos = 0;
+                }
+                /* Avoid 32-bit overflow (dump bit buffer to top of output) */
+                if (bd->inbufBitCount >= 24) {
+                        bits = bd->inbufBits&((1 << bd->inbufBitCount)-1);
+                        bits_wanted -= bd->inbufBitCount;
+                        bits <<= bits_wanted;
+                        bd->inbufBitCount = 0;
+                }
+                /* Grab next 8 bits of input from buffer. */
+                bd->inbufBits = (bd->inbufBits << 8)|bd->inbuf[bd->inbufPos++];
+                bd->inbufBitCount += 8;
+        }
+        /* Calculate result */
+        bd->inbufBitCount -= bits_wanted;
+        bits |= (bd->inbufBits >> bd->inbufBitCount)&((1 << bits_wanted)-1);
+
+        return bits;
+}
+
+/* Unpacks the next block and sets up for the inverse burrows-wheeler step. */
+
+static int INIT get_next_block(struct bunzip_data *bd)
+{
+        struct group_data *hufGroup = NULL;
+        int *base = NULL;
+        int *limit = NULL;
+        int dbufCount, nextSym, dbufSize, groupCount, selector,
+                i, j, k, t, runPos, symCount, symTotal, nSelectors,
+                byteCount[256];
+        unsigned char uc, symToByte[256], mtfSymbol[256], *selectors;
+        unsigned int *dbuf, origPtr;
+
+        dbuf = bd->dbuf;
+        dbufSize = bd->dbufSize;
+        selectors = bd->selectors;
+
+        /* Read in header signature and CRC, then validate signature.
+           (last block signature means CRC is for whole file, return now) */
+        i = get_bits(bd, 24);
+        j = get_bits(bd, 24);
+        bd->headerCRC = get_bits(bd, 32);
+        if ((i == 0x177245) && (j == 0x385090))
+                return RETVAL_LAST_BLOCK;
+        if ((i != 0x314159) || (j != 0x265359))
+                return RETVAL_NOT_BZIP_DATA;
+        /* We can add support for blockRandomised if anybody complains.
+           There was some code for this in busybox 1.0.0-pre3, but nobody ever
+           noticed that it didn't actually work. */
+        if (get_bits(bd, 1))
+                return RETVAL_OBSOLETE_INPUT;
+        origPtr = get_bits(bd, 24);
+        if (origPtr > dbufSize)
+                return RETVAL_DATA_ERROR;
+        /* mapping table: if some byte values are never used (encoding things
+           like ascii text), the compression code removes the gaps to have fewer
+           symbols to deal with, and writes a sparse bitfield indicating which
+           values were present.  We make a translation table to convert the
+           symbols back to the corresponding bytes. */
+        t = get_bits(bd, 16);
+        symTotal = 0;
+        for (i = 0; i < 16; i++) {
+                if (t&(1 << (15-i))) {
+                        k = get_bits(bd, 16);
+                        for (j = 0; j < 16; j++)
+                                if (k&(1 << (15-j)))
+                                        symToByte[symTotal++] = (16*i)+j;
+                }
+        }
+        /* How many different Huffman coding groups does this block use? */
+        groupCount = get_bits(bd, 3);
+        if (groupCount < 2 || groupCount > MAX_GROUPS)
+                return RETVAL_DATA_ERROR;
+        /* nSelectors: Every GROUP_SIZE many symbols we select a new
+           Huffman coding group.  Read in the group selector list,
+           which is stored as MTF encoded bit runs.  (MTF = Move To
+           Front, as each value is used it's moved to the start of the
+           list.) */
+        nSelectors = get_bits(bd, 15);
+        if (!nSelectors)
+                return RETVAL_DATA_ERROR;
+        for (i = 0; i < groupCount; i++)
+                mtfSymbol[i] = i;
+        for (i = 0; i < nSelectors; i++) {
+                /* Get next value */
+                for (j = 0; get_bits(bd, 1); j++)
+                        if (j >= groupCount)
+                                return RETVAL_DATA_ERROR;
+                /* Decode MTF to get the next selector */
+                uc = mtfSymbol[j];
+                for (; j; j--)
+                        mtfSymbol[j] = mtfSymbol[j-1];
+                mtfSymbol[0] = selectors[i] = uc;
+        }
+        /* Read the Huffman coding tables for each group, which code
+           for symTotal literal symbols, plus two run symbols (RUNA,
+           RUNB) */
+        symCount = symTotal+2;
+        for (j = 0; j < groupCount; j++) {
+                unsigned char length[MAX_SYMBOLS], temp[MAX_HUFCODE_BITS+1];
+                int     minLen, maxLen, pp;
+                /* Read Huffman code lengths for each symbol.  They're
+                   stored in a way similar to mtf; record a starting
+                   value for the first symbol, and an offset from the
+                   previous value for everys symbol after that.
+                   (Subtracting 1 before the loop and then adding it
+                   back at the end is an optimization that makes the
+                   test inside the loop simpler: symbol length 0
+                   becomes negative, so an unsigned inequality catches
+                   it.) */
+                t = get_bits(bd, 5)-1;
+                for (i = 0; i < symCount; i++) {
+                        for (;;) {
+                                if (((unsigned)t) > (MAX_HUFCODE_BITS-1))
+                                        return RETVAL_DATA_ERROR;
+
+                                /* If first bit is 0, stop.  Else
+                                   second bit indicates whether to
+                                   increment or decrement the value.
+                                   Optimization: grab 2 bits and unget
+                                   the second if the first was 0. */
+
+                                k = get_bits(bd, 2);
+                                if (k < 2) {
+                                        bd->inbufBitCount++;
+                                        break;
+                                }
+                                /* Add one if second bit 1, else
+                                 * subtract 1.  Avoids if/else */
+                                t += (((k+1)&2)-1);
+                        }
+                        /* Correct for the initial -1, to get the
+                         * final symbol length */
+                        length[i] = t+1;
+                }
+                /* Find largest and smallest lengths in this group */
+                minLen = maxLen = length[0];
+
+                for (i = 1; i < symCount; i++) {
+                        if (length[i] > maxLen)
+                                maxLen = length[i];
+                        else if (length[i] < minLen)
+                                minLen = length[i];
+                }
+
+                /* Calculate permute[], base[], and limit[] tables from
+                 * length[].
+                 *
+                 * permute[] is the lookup table for converting
+                 * Huffman coded symbols into decoded symbols.  base[]
+                 * is the amount to subtract from the value of a
+                 * Huffman symbol of a given length when using
+                 * permute[].
+                 *
+                 * limit[] indicates the largest numerical value a
+                 * symbol with a given number of bits can have.  This
+                 * is how the Huffman codes can vary in length: each
+                 * code with a value > limit[length] needs another
+                 * bit.
+                 */
+                hufGroup = bd->groups+j;
+                hufGroup->minLen = minLen;
+                hufGroup->maxLen = maxLen;
+                /* Note that minLen can't be smaller than 1, so we
+                   adjust the base and limit array pointers so we're
+                   not always wasting the first entry.  We do this
+                   again when using them (during symbol decoding).*/
+                base = hufGroup->base-1;
+                limit = hufGroup->limit-1;
+                /* Calculate permute[].  Concurently, initialize
+                 * temp[] and limit[]. */
+                pp = 0;
+                for (i = minLen; i <= maxLen; i++) {
+                        temp[i] = limit[i] = 0;
+                        for (t = 0; t < symCount; t++)
+                                if (length[t] == i)
+                                        hufGroup->permute[pp++] = t;
+                }
+                /* Count symbols coded for at each bit length */
+                for (i = 0; i < symCount; i++)
+                        temp[length[i]]++;
+                /* Calculate limit[] (the largest symbol-coding value
+                 *at each bit length, which is (previous limit <<
+                 *1)+symbols at this level), and base[] (number of
+                 *symbols to ignore at each bit length, which is limit
+                 *minus the cumulative count of symbols coded for
+                 *already). */
+                pp = t = 0;
+                for (i = minLen; i < maxLen; i++) {
+                        pp += temp[i];
+                        /* We read the largest possible symbol size
+                           and then unget bits after determining how
+                           many we need, and those extra bits could be
+                           set to anything.  (They're noise from
+                           future symbols.)  At each level we're
+                           really only interested in the first few
+                           bits, so here we set all the trailing
+                           to-be-ignored bits to 1 so they don't
+                           affect the value > limit[length]
+                           comparison. */
+                        limit[i] = (pp << (maxLen - i)) - 1;
+                        pp <<= 1;
+                        base[i+1] = pp-(t += temp[i]);
+                }
+                limit[maxLen+1] = INT_MAX; /* Sentinal value for
+                                            * reading next sym. */
+                limit[maxLen] = pp+temp[maxLen]-1;
+                base[minLen] = 0;
+        }
+        /* We've finished reading and digesting the block header.  Now
+           read this block's Huffman coded symbols from the file and
+           undo the Huffman coding and run length encoding, saving the
+           result into dbuf[dbufCount++] = uc */
+
+        /* Initialize symbol occurrence counters and symbol Move To
+         * Front table */
+        for (i = 0; i < 256; i++) {
+                byteCount[i] = 0;
+                mtfSymbol[i] = (unsigned char)i;
+        }
+        /* Loop through compressed symbols. */
+        runPos = dbufCount = symCount = selector = 0;
+        for (;;) {
+                /* Determine which Huffman coding group to use. */
+                if (!(symCount--)) {
+                        symCount = GROUP_SIZE-1;
+                        if (selector >= nSelectors)
+                                return RETVAL_DATA_ERROR;
+                        hufGroup = bd->groups+selectors[selector++];
+                        base = hufGroup->base-1;
+                        limit = hufGroup->limit-1;
+                }
+                /* Read next Huffman-coded symbol. */
+                /* Note: It is far cheaper to read maxLen bits and
+                   back up than it is to read minLen bits and then an
+                   additional bit at a time, testing as we go.
+                   Because there is a trailing last block (with file
+                   CRC), there is no danger of the overread causing an
+                   unexpected EOF for a valid compressed file.  As a
+                   further optimization, we do the read inline
+                   (falling back to a call to get_bits if the buffer
+                   runs dry).  The following (up to got_huff_bits:) is
+                   equivalent to j = get_bits(bd, hufGroup->maxLen);
+                 */
+                while (bd->inbufBitCount < hufGroup->maxLen) {
+                        if (bd->inbufPos == bd->inbufCount) {
+                                j = get_bits(bd, hufGroup->maxLen);
+                                goto got_huff_bits;
+                        }
+                        bd->inbufBits =
+                                (bd->inbufBits << 8)|bd->inbuf[bd->inbufPos++];
+                        bd->inbufBitCount += 8;
+                };
+                bd->inbufBitCount -= hufGroup->maxLen;
+                j = (bd->inbufBits >> bd->inbufBitCount)&
+                        ((1 << hufGroup->maxLen)-1);
+got_huff_bits:
+                /* Figure how how many bits are in next symbol and
+                 * unget extras */
+                i = hufGroup->minLen;
+                while (j > limit[i])
+                        ++i;
+                bd->inbufBitCount += (hufGroup->maxLen - i);
+                /* Huffman decode value to get nextSym (with bounds checking) */
+                if ((i > hufGroup->maxLen)
+                        || (((unsigned)(j = (j>>(hufGroup->maxLen-i))-base[i]))
+                                >= MAX_SYMBOLS))
+                        return RETVAL_DATA_ERROR;
+                nextSym = hufGroup->permute[j];
+                /* We have now decoded the symbol, which indicates
+                   either a new literal byte, or a repeated run of the
+                   most recent literal byte.  First, check if nextSym
+                   indicates a repeated run, and if so loop collecting
+                   how many times to repeat the last literal. */
+                if (((unsigned)nextSym) <= SYMBOL_RUNB) { /* RUNA or RUNB */
+                        /* If this is the start of a new run, zero out
+                         * counter */
+                        if (!runPos) {
+                                runPos = 1;
+                                t = 0;
+                        }
+                        /* Neat trick that saves 1 symbol: instead of
+                           or-ing 0 or 1 at each bit position, add 1
+                           or 2 instead.  For example, 1011 is 1 << 0
+                           + 1 << 1 + 2 << 2.  1010 is 2 << 0 + 2 << 1
+                           + 1 << 2.  You can make any bit pattern
+                           that way using 1 less symbol than the basic
+                           or 0/1 method (except all bits 0, which
+                           would use no symbols, but a run of length 0
+                           doesn't mean anything in this context).
+                           Thus space is saved. */
+                        t += (runPos << nextSym);
+                        /* +runPos if RUNA; +2*runPos if RUNB */
+
+                        runPos <<= 1;
+                        continue;
+                }
+                /* When we hit the first non-run symbol after a run,
+                   we now know how many times to repeat the last
+                   literal, so append that many copies to our buffer
+                   of decoded symbols (dbuf) now.  (The last literal
+                   used is the one at the head of the mtfSymbol
+                   array.) */
+                if (runPos) {
+                        runPos = 0;
+                        if (dbufCount+t >= dbufSize)
+                                return RETVAL_DATA_ERROR;
+
+                        uc = symToByte[mtfSymbol[0]];
+                        byteCount[uc] += t;
+                        while (t--)
+                                dbuf[dbufCount++] = uc;
+                }
+                /* Is this the terminating symbol? */
+                if (nextSym > symTotal)
+                        break;
+                /* At this point, nextSym indicates a new literal
+                   character.  Subtract one to get the position in the
+                   MTF array at which this literal is currently to be
+                   found.  (Note that the result can't be -1 or 0,
+                   because 0 and 1 are RUNA and RUNB.  But another
+                   instance of the first symbol in the mtf array,
+                   position 0, would have been handled as part of a
+                   run above.  Therefore 1 unused mtf position minus 2
+                   non-literal nextSym values equals -1.) */
+                if (dbufCount >= dbufSize)
+                        return RETVAL_DATA_ERROR;
+                i = nextSym - 1;
+                uc = mtfSymbol[i];
+                /* Adjust the MTF array.  Since we typically expect to
+                 *move only a small number of symbols, and are bound
+                 *by 256 in any case, using memmove here would
+                 *typically be bigger and slower due to function call
+                 *overhead and other assorted setup costs. */
+                do {
+                        mtfSymbol[i] = mtfSymbol[i-1];
+                } while (--i);
+                mtfSymbol[0] = uc;
+                uc = symToByte[uc];
+                /* We have our literal byte.  Save it into dbuf. */
+                byteCount[uc]++;
+                dbuf[dbufCount++] = (unsigned int)uc;
+        }
+        /* At this point, we've read all the Huffman-coded symbols
+           (and repeated runs) for this block from the input stream,
+           and decoded them into the intermediate buffer.  There are
+           dbufCount many decoded bytes in dbuf[].  Now undo the
+           Burrows-Wheeler transform on dbuf.  See
+           http://dogma.net/markn/articles/bwt/bwt.htm
+         */
+        /* Turn byteCount into cumulative occurrence counts of 0 to n-1. */
+        j = 0;
+        for (i = 0; i < 256; i++) {
+                k = j+byteCount[i];
+                byteCount[i] = j;
+                j = k;
+        }
+        /* Figure out what order dbuf would be in if we sorted it. */
+        for (i = 0; i < dbufCount; i++) {
+                uc = (unsigned char)(dbuf[i] & 0xff);
+                dbuf[byteCount[uc]] |= (i << 8);
+                byteCount[uc]++;
+        }
+        /* Decode first byte by hand to initialize "previous" byte.
+           Note that it doesn't get output, and if the first three
+           characters are identical it doesn't qualify as a run (hence
+           writeRunCountdown = 5). */
+        if (dbufCount) {
+                if (origPtr >= dbufCount)
+                        return RETVAL_DATA_ERROR;
+                bd->writePos = dbuf[origPtr];
+                bd->writeCurrent = (unsigned char)(bd->writePos&0xff);
+                bd->writePos >>= 8;
+                bd->writeRunCountdown = 5;
+        }
+        bd->writeCount = dbufCount;
+
+        return RETVAL_OK;
+}
+
+/* Undo burrows-wheeler transform on intermediate buffer to produce output.
+   If start_bunzip was initialized with out_fd =-1, then up to len bytes of
+   data are written to outbuf.  Return value is number of bytes written or
+   error (all errors are negative numbers).  If out_fd!=-1, outbuf and len
+   are ignored, data is written to out_fd and return is RETVAL_OK or error.
+*/
+
+static int INIT read_bunzip(struct bunzip_data *bd, char *outbuf, int len)
+{
+        const unsigned int *dbuf;
+        int pos, xcurrent, previous, gotcount;
+
+        /* If last read was short due to end of file, return last block now */
+        if (bd->writeCount < 0)
+                return bd->writeCount;
+
+        gotcount = 0;
+        dbuf = bd->dbuf;
+        pos = bd->writePos;
+        xcurrent = bd->writeCurrent;
+
+        /* We will always have pending decoded data to write into the output
+           buffer unless this is the very first call (in which case we haven't
+           Huffman-decoded a block into the intermediate buffer yet). */
+
+        if (bd->writeCopies) {
+                /* Inside the loop, writeCopies means extra copies (beyond 1) */
+                --bd->writeCopies;
+                /* Loop outputting bytes */
+                for (;;) {
+                        /* If the output buffer is full, snapshot
+                         * state and return */
+                        if (gotcount >= len) {
+                                bd->writePos = pos;
+                                bd->writeCurrent = xcurrent;
+                                bd->writeCopies++;
+                                return len;
+                        }
+                        /* Write next byte into output buffer, updating CRC */
+                        outbuf[gotcount++] = xcurrent;
+                        bd->writeCRC = (((bd->writeCRC) << 8)
+                                ^bd->crc32Table[((bd->writeCRC) >> 24)
+                                ^xcurrent]);
+                        /* Loop now if we're outputting multiple
+                         * copies of this byte */
+                        if (bd->writeCopies) {
+                                --bd->writeCopies;
+                                continue;
+                        }
+decode_next_byte:
+                        if (!bd->writeCount--)
+                                break;
+                        /* Follow sequence vector to undo
+                         * Burrows-Wheeler transform */
+                        previous = xcurrent;
+                        pos = dbuf[pos];
+                        xcurrent = pos&0xff;
+                        pos >>= 8;
+                        /* After 3 consecutive copies of the same
+                           byte, the 4th is a repeat count.  We count
+                           down from 4 instead *of counting up because
+                           testing for non-zero is faster */
+                        if (--bd->writeRunCountdown) {
+                                if (xcurrent != previous)
+                                        bd->writeRunCountdown = 4;
+                        } else {
+                                /* We have a repeated run, this byte
+                                 * indicates the count */
+                                bd->writeCopies = xcurrent;
+                                xcurrent = previous;
+                                bd->writeRunCountdown = 5;
+                                /* Sometimes there are just 3 bytes
+                                 * (run length 0) */
+                                if (!bd->writeCopies)
+                                        goto decode_next_byte;
+                                /* Subtract the 1 copy we'd output
+                                 * anyway to get extras */
+                                --bd->writeCopies;
+                        }
+                }
+                /* Decompression of this block completed successfully */
+                bd->writeCRC = ~bd->writeCRC;
+                bd->totalCRC = ((bd->totalCRC << 1) |
+                                (bd->totalCRC >> 31)) ^ bd->writeCRC;
+                /* If this block had a CRC error, force file level CRC error. */
+                if (bd->writeCRC != bd->headerCRC) {
+                        bd->totalCRC = bd->headerCRC+1;
+                        return RETVAL_LAST_BLOCK;
+                }
+        }
+
+        /* Refill the intermediate buffer by Huffman-decoding next
+         * block of input */
+        /* (previous is just a convenient unused temp variable here) */
+        previous = get_next_block(bd);
+        if (previous) {
+                bd->writeCount = previous;
+                return (previous != RETVAL_LAST_BLOCK) ? previous : gotcount;
+        }
+        bd->writeCRC = 0xffffffffUL;
+        pos = bd->writePos;
+        xcurrent = bd->writeCurrent;
+        goto decode_next_byte;
+}
+
+static int INIT nofill(void *buf, unsigned int len)
+{
+        return -1;
+}
+
+/* Allocate the structure, read file header.  If in_fd ==-1, inbuf must contain
+   a complete bunzip file (len bytes long).  If in_fd!=-1, inbuf and len are
+   ignored, and data is read from file handle into temporary buffer. */
+static int INIT start_bunzip(struct bunzip_data **bdp, void *inbuf, int len,
+                             int (*fill)(void*, unsigned int))
+{
+        struct bunzip_data *bd;
+        unsigned int i, j, c;
+        const unsigned int BZh0 =
+                (((unsigned int)'B') << 24)+(((unsigned int)'Z') << 16)
+                +(((unsigned int)'h') << 8)+(unsigned int)'0';
+
+        /* Figure out how much data to allocate */
+        i = sizeof(struct bunzip_data);
+
+        /* Allocate bunzip_data.  Most fields initialize to zero. */
+        bd = *bdp = malloc(i);
+        memset(bd, 0, sizeof(struct bunzip_data));
+        /* Setup input buffer */
+        bd->inbuf = inbuf;
+        bd->inbufCount = len;
+        if (fill != NULL)
+                bd->fill = fill;
+        else
+                bd->fill = nofill;
+
+        /* Init the CRC32 table (big endian) */
+        for (i = 0; i < 256; i++) {
+                c = i << 24;
+                for (j = 8; j; j--)
+                        c = c&0x80000000 ? (c << 1)^0x04c11db7 : (c << 1);
+                bd->crc32Table[i] = c;
+        }
+
+        /* Ensure that file starts with "BZh['1'-'9']." */
+        i = get_bits(bd, 32);
+        if (((unsigned int)(i-BZh0-1)) >= 9)
+                return RETVAL_NOT_BZIP_DATA;
+
+        /* Fourth byte (ascii '1'-'9'), indicates block size in units of 100k of
+           uncompressed data.  Allocate intermediate buffer for block. */
+        bd->dbufSize = 100000*(i-BZh0);
+
+        bd->dbuf = large_malloc(bd->dbufSize * sizeof(int));
+        return RETVAL_OK;
+}
+
+/* Example usage: decompress src_fd to dst_fd.  (Stops at end of bzip2 data,
+   not end of file.) */
+STATIC int INIT bunzip2(unsigned char *buf, int len,
+                        int(*fill)(void*, unsigned int),
+                        int(*flush)(void*, unsigned int),
+                        unsigned char *outbuf,
+                        int *pos,
+                        void(*error_fn)(char *x))
+{
+        struct bunzip_data *bd;
+        int i = -1;
+        unsigned char *inbuf;
+
+        set_error_fn(error_fn);
+        if (flush)
+                outbuf = malloc(BZIP2_IOBUF_SIZE);
+        else
+                len -= 4; /* Uncompressed size hack active in pre-boot
+                             environment */
+        if (!outbuf) {
+                error("Could not allocate output bufer");
+                return -1;
+        }
+        if (buf)
+                inbuf = buf;
+        else
+                inbuf = malloc(BZIP2_IOBUF_SIZE);
+        if (!inbuf) {
+                error("Could not allocate input bufer");
+                goto exit_0;
+        }
+        i = start_bunzip(&bd, inbuf, len, fill);
+        if (!i) {
+                for (;;) {
+                        i = read_bunzip(bd, outbuf, BZIP2_IOBUF_SIZE);
+                        if (i <= 0)
+                                break;
+                        if (!flush)
+                                outbuf += i;
+                        else
+                                if (i != flush(outbuf, i)) {
+                                        i = RETVAL_UNEXPECTED_OUTPUT_EOF;
+                                        break;
+                                }
+                }
+        }
+        /* Check CRC and release memory */
+        if (i == RETVAL_LAST_BLOCK) {
+                if (bd->headerCRC != bd->totalCRC)
+                        error("Data integrity error when decompressing.");
+                else
+                        i = RETVAL_OK;
+        } else if (i == RETVAL_UNEXPECTED_OUTPUT_EOF) {
+                error("Compressed file ends unexpectedly");
+        }
+        if (bd->dbuf)
+                large_free(bd->dbuf);
+        if (pos)
+                *pos = bd->inbufPos;
+        free(bd);
+        if (!buf)
+                free(inbuf);
+exit_0:
+        if (flush)
+                free(outbuf);
+        return i;
+}
+
+#define decompress bunzip2
diff --git a/lib/decompress_inflate.c b/lib/decompress_inflate.c
new file mode 100644
index 000000000000..839a329b4fc4
--- /dev/null
+++ b/lib/decompress_inflate.c
@@ -0,0 +1,167 @@
+#ifdef STATIC
+/* Pre-boot environment: included */
+
+/* prevent inclusion of _LINUX_KERNEL_H in pre-boot environment: lots
+ * errors about console_printk etc... on ARM */
+#define _LINUX_KERNEL_H
+
+#include "zlib_inflate/inftrees.c"
+#include "zlib_inflate/inffast.c"
+#include "zlib_inflate/inflate.c"
+
+#else /* STATIC */
+/* initramfs et al: linked */
+
+#include <linux/zutil.h>
+
+#include "zlib_inflate/inftrees.h"
+#include "zlib_inflate/inffast.h"
+#include "zlib_inflate/inflate.h"
+
+#include "zlib_inflate/infutil.h"
+
+#endif /* STATIC */
+
+#include <linux/decompress/mm.h>
+
+#define INBUF_LEN (16*1024)
+
+/* Included from initramfs et al code */
+STATIC int INIT gunzip(unsigned char *buf, int len,
+                       int(*fill)(void*, unsigned int),
+                       int(*flush)(void*, unsigned int),
+                       unsigned char *out_buf,
+                       int *pos,
+                       void(*error_fn)(char *x)) {
+        u8 *zbuf;
+        struct z_stream_s *strm;
+        int rc;
+        size_t out_len;
+
+        set_error_fn(error_fn);
+        rc = -1;
+        if (flush) {
+                out_len = 0x8000; /* 32 K */
+                out_buf = malloc(out_len);
+        } else {
+                out_len = 0x7fffffff; /* no limit */
+        }
+        if (!out_buf) {
+                error("Out of memory while allocating output buffer");
+                goto gunzip_nomem1;
+        }
+
+        if (buf)
+                zbuf = buf;
+        else {
+                zbuf = malloc(INBUF_LEN);
+                len = 0;
+        }
+        if (!zbuf) {
+                error("Out of memory while allocating input buffer");
+                goto gunzip_nomem2;
+        }
+
+        strm = malloc(sizeof(*strm));
+        if (strm == NULL) {
+                error("Out of memory while allocating z_stream");
+                goto gunzip_nomem3;
+        }
+
+        strm->workspace = malloc(flush ? zlib_inflate_workspacesize() :
+                                 sizeof(struct inflate_state));
+        if (strm->workspace == NULL) {
+                error("Out of memory while allocating workspace");
+                goto gunzip_nomem4;
+        }
+
+        if (len == 0)
+                len = fill(zbuf, INBUF_LEN);
+
+        /* verify the gzip header */
+        if (len < 10 ||
+           zbuf[0] != 0x1f || zbuf[1] != 0x8b || zbuf[2] != 0x08) {
+                if (pos)
+                        *pos = 0;
+                error("Not a gzip file");
+                goto gunzip_5;
+        }
+
+        /* skip over gzip header (1f,8b,08... 10 bytes total +
+         * possible asciz filename)
+         */
+        strm->next_in = zbuf + 10;
+        /* skip over asciz filename */
+        if (zbuf[3] & 0x8) {
+                while (strm->next_in[0])
+                        strm->next_in++;
+                strm->next_in++;
+        }
+        strm->avail_in = len - (strm->next_in - zbuf);
+
+        strm->next_out = out_buf;
+        strm->avail_out = out_len;
+
+        rc = zlib_inflateInit2(strm, -MAX_WBITS);
+
+        if (!flush) {
+                WS(strm)->inflate_state.wsize = 0;
+                WS(strm)->inflate_state.window = NULL;
+        }
+
+        while (rc == Z_OK) {
+                if (strm->avail_in == 0) {
+                        /* TODO: handle case where both pos and fill are set */
+                        len = fill(zbuf, INBUF_LEN);
+                        if (len < 0) {
+                                rc = -1;
+                                error("read error");
+                                break;
+                        }
+                        strm->next_in = zbuf;
+                        strm->avail_in = len;
+                }
+                rc = zlib_inflate(strm, 0);
+
+                /* Write any data generated */
+                if (flush && strm->next_out > out_buf) {
+                        int l = strm->next_out - out_buf;
+                        if (l != flush(out_buf, l)) {
+                                rc = -1;
+                                error("write error");
+                                break;
+                        }
+                        strm->next_out = out_buf;
+                        strm->avail_out = out_len;
+                }
+
+                /* after Z_FINISH, only Z_STREAM_END is "we unpacked it all" */
+                if (rc == Z_STREAM_END) {
+                        rc = 0;
+                        break;
+                } else if (rc != Z_OK) {
+                        error("uncompression error");
+                        rc = -1;
+                }
+        }
+
+        zlib_inflateEnd(strm);
+        if (pos)
+                /* add + 8 to skip over trailer */
+                *pos = strm->next_in - zbuf+8;
+
+gunzip_5:
+        free(strm->workspace);
+gunzip_nomem4:
+        free(strm);
+gunzip_nomem3:
+        if (!buf)
+                free(zbuf);
+gunzip_nomem2:
+        if (flush)
+                free(out_buf);
+gunzip_nomem1:
+        return rc; /* returns Z_OK (0) if successful */
+}
+
+#define decompress gunzip
diff --git a/lib/decompress_unlzma.c b/lib/decompress_unlzma.c
new file mode 100644
index 000000000000..546f2f4c157e
--- /dev/null
+++ b/lib/decompress_unlzma.c
@@ -0,0 +1,647 @@
+/* Lzma decompressor for Linux kernel. Shamelessly snarfed
+ *from busybox 1.1.1
+ *
+ *Linux kernel adaptation
+ *Copyright (C) 2006  Alain < alain@knaff.lu >
+ *
+ *Based on small lzma deflate implementation/Small range coder
+ *implementation for lzma.
+ *Copyright (C) 2006  Aurelien Jacobs < aurel@gnuage.org >
+ *
+ *Based on LzmaDecode.c from the LZMA SDK 4.22 (http://www.7-zip.org/)
+ *Copyright (C) 1999-2005  Igor Pavlov
+ *
+ *Copyrights of the parts, see headers below.
+ *
+ *
+ *This program is free software; you can redistribute it and/or
+ *modify it under the terms of the GNU Lesser General Public
+ *License as published by the Free Software Foundation; either
+ *version 2.1 of the License, or (at your option) any later version.
+ *
+ *This program is distributed in the hope that it will be useful,
+ *but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ *Lesser General Public License for more details.
+ *
+ *You should have received a copy of the GNU Lesser General Public
+ *License along with this library; if not, write to the Free Software
+ *Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#ifndef STATIC
+#include <linux/decompress/unlzma.h>
+#endif /* STATIC */
+
+#include <linux/decompress/mm.h>
+
+#define MIN(a, b) (((a) < (b)) ? (a) : (b))
+
+static long long INIT read_int(unsigned char *ptr, int size)
+{
+        int i;
+        long long ret = 0;
+
+        for (i = 0; i < size; i++)
+                ret = (ret << 8) | ptr[size-i-1];
+        return ret;
+}
+
+#define ENDIAN_CONVERT(x) \
+  x = (typeof(x))read_int((unsigned char *)&x, sizeof(x))
+
+
+/* Small range coder implementation for lzma.
+ *Copyright (C) 2006  Aurelien Jacobs < aurel@gnuage.org >
+ *
+ *Based on LzmaDecode.c from the LZMA SDK 4.22 (http://www.7-zip.org/)
+ *Copyright (c) 1999-2005  Igor Pavlov
+ */
+
+#include <linux/compiler.h>
+
+#define LZMA_IOBUF_SIZE 0x10000
+
+struct rc {
+        int (*fill)(void*, unsigned int);
+        uint8_t *ptr;
+        uint8_t *buffer;
+        uint8_t *buffer_end;
+        int buffer_size;
+        uint32_t code;
+        uint32_t range;
+        uint32_t bound;
+};
+
+
+#define RC_TOP_BITS 24
+#define RC_MOVE_BITS 5
+#define RC_MODEL_TOTAL_BITS 11
+
+
+/* Called twice: once at startup and once in rc_normalize() */
+static void INIT rc_read(struct rc *rc)
+{
+        rc->buffer_size = rc->fill((char *)rc->buffer, LZMA_IOBUF_SIZE);
+        if (rc->buffer_size <= 0)
+                error("unexpected EOF");
+        rc->ptr = rc->buffer;
+        rc->buffer_end = rc->buffer + rc->buffer_size;
+}
+
+/* Called once */
+static inline void INIT rc_init(struct rc *rc,
+                                       int (*fill)(void*, unsigned int),
+                                       char *buffer, int buffer_size)
+{
+        rc->fill = fill;
+        rc->buffer = (uint8_t *)buffer;
+        rc->buffer_size = buffer_size;
+        rc->buffer_end = rc->buffer + rc->buffer_size;
+        rc->ptr = rc->buffer;
+
+        rc->code = 0;
+        rc->range = 0xFFFFFFFF;
+}
+
+static inline void INIT rc_init_code(struct rc *rc)
+{
+        int i;
+
+        for (i = 0; i < 5; i++) {
+                if (rc->ptr >= rc->buffer_end)
+                        rc_read(rc);
+                rc->code = (rc->code << 8) | *rc->ptr++;
+        }
+}
+
+
+/* Called once. TODO: bb_maybe_free() */
+static inline void INIT rc_free(struct rc *rc)
+{
+        free(rc->buffer);
+}
+
+/* Called twice, but one callsite is in inline'd rc_is_bit_0_helper() */
+static void INIT rc_do_normalize(struct rc *rc)
+{
+        if (rc->ptr >= rc->buffer_end)
+                rc_read(rc);
+        rc->range <<= 8;
+        rc->code = (rc->code << 8) | *rc->ptr++;
+}
+static inline void INIT rc_normalize(struct rc *rc)
+{
+        if (rc->range < (1 << RC_TOP_BITS))
+                rc_do_normalize(rc);
+}
+
+/* Called 9 times */
+/* Why rc_is_bit_0_helper exists?
+ *Because we want to always expose (rc->code < rc->bound) to optimizer
+ */
+static inline uint32_t INIT rc_is_bit_0_helper(struct rc *rc, uint16_t *p)
+{
+        rc_normalize(rc);
+        rc->bound = *p * (rc->range >> RC_MODEL_TOTAL_BITS);
+        return rc->bound;
+}
+static inline int INIT rc_is_bit_0(struct rc *rc, uint16_t *p)
+{
+        uint32_t t = rc_is_bit_0_helper(rc, p);
+        return rc->code < t;
+}
+
+/* Called ~10 times, but very small, thus inlined */
+static inline void INIT rc_update_bit_0(struct rc *rc, uint16_t *p)
+{
+        rc->range = rc->bound;
+        *p += ((1 << RC_MODEL_TOTAL_BITS) - *p) >> RC_MOVE_BITS;
+}
+static inline void rc_update_bit_1(struct rc *rc, uint16_t *p)
+{
+        rc->range -= rc->bound;
+        rc->code -= rc->bound;
+        *p -= *p >> RC_MOVE_BITS;
+}
+
+/* Called 4 times in unlzma loop */
+static int INIT rc_get_bit(struct rc *rc, uint16_t *p, int *symbol)
+{
+        if (rc_is_bit_0(rc, p)) {
+                rc_update_bit_0(rc, p);
+                *symbol *= 2;
+                return 0;
+        } else {
+                rc_update_bit_1(rc, p);
+                *symbol = *symbol * 2 + 1;
+                return 1;
+        }
+}
+
+/* Called once */
+static inline int INIT rc_direct_bit(struct rc *rc)
+{
+        rc_normalize(rc);
+        rc->range >>= 1;
+        if (rc->code >= rc->range) {
+                rc->code -= rc->range;
+                return 1;
+        }
+        return 0;
+}
+
+/* Called twice */
+static inline void INIT
+rc_bit_tree_decode(struct rc *rc, uint16_t *p, int num_levels, int *symbol)
+{
+        int i = num_levels;
+
+        *symbol = 1;
+        while (i--)
+                rc_get_bit(rc, p + *symbol, symbol);
+        *symbol -= 1 << num_levels;
+}
+
+
+/*
+ * Small lzma deflate implementation.
+ * Copyright (C) 2006  Aurelien Jacobs < aurel@gnuage.org >
+ *
+ * Based on LzmaDecode.c from the LZMA SDK 4.22 (http://www.7-zip.org/)
+ * Copyright (C) 1999-2005  Igor Pavlov
+ */
+
+
+struct lzma_header {
+        uint8_t pos;
+        uint32_t dict_size;
+        uint64_t dst_size;
+} __attribute__ ((packed)) ;
+
+
+#define LZMA_BASE_SIZE 1846
+#define LZMA_LIT_SIZE 768
+
+#define LZMA_NUM_POS_BITS_MAX 4
+
+#define LZMA_LEN_NUM_LOW_BITS 3
+#define LZMA_LEN_NUM_MID_BITS 3
+#define LZMA_LEN_NUM_HIGH_BITS 8
+
+#define LZMA_LEN_CHOICE 0
+#define LZMA_LEN_CHOICE_2 (LZMA_LEN_CHOICE + 1)
+#define LZMA_LEN_LOW (LZMA_LEN_CHOICE_2 + 1)
+#define LZMA_LEN_MID (LZMA_LEN_LOW \
+                      + (1 << (LZMA_NUM_POS_BITS_MAX + LZMA_LEN_NUM_LOW_BITS)))
+#define LZMA_LEN_HIGH (LZMA_LEN_MID \
+                       +(1 << (LZMA_NUM_POS_BITS_MAX + LZMA_LEN_NUM_MID_BITS)))
+#define LZMA_NUM_LEN_PROBS (LZMA_LEN_HIGH + (1 << LZMA_LEN_NUM_HIGH_BITS))
+
+#define LZMA_NUM_STATES 12
+#define LZMA_NUM_LIT_STATES 7
+
+#define LZMA_START_POS_MODEL_INDEX 4
+#define LZMA_END_POS_MODEL_INDEX 14
+#define LZMA_NUM_FULL_DISTANCES (1 << (LZMA_END_POS_MODEL_INDEX >> 1))
+
+#define LZMA_NUM_POS_SLOT_BITS 6
+#define LZMA_NUM_LEN_TO_POS_STATES 4
+
+#define LZMA_NUM_ALIGN_BITS 4
+
+#define LZMA_MATCH_MIN_LEN 2
+
+#define LZMA_IS_MATCH 0
+#define LZMA_IS_REP (LZMA_IS_MATCH + (LZMA_NUM_STATES << LZMA_NUM_POS_BITS_MAX))
+#define LZMA_IS_REP_G0 (LZMA_IS_REP + LZMA_NUM_STATES)
+#define LZMA_IS_REP_G1 (LZMA_IS_REP_G0 + LZMA_NUM_STATES)
+#define LZMA_IS_REP_G2 (LZMA_IS_REP_G1 + LZMA_NUM_STATES)
+#define LZMA_IS_REP_0_LONG (LZMA_IS_REP_G2 + LZMA_NUM_STATES)
+#define LZMA_POS_SLOT (LZMA_IS_REP_0_LONG \
+                       + (LZMA_NUM_STATES << LZMA_NUM_POS_BITS_MAX))
+#define LZMA_SPEC_POS (LZMA_POS_SLOT \
+                       +(LZMA_NUM_LEN_TO_POS_STATES << LZMA_NUM_POS_SLOT_BITS))
+#define LZMA_ALIGN (LZMA_SPEC_POS \
+                    + LZMA_NUM_FULL_DISTANCES - LZMA_END_POS_MODEL_INDEX)
+#define LZMA_LEN_CODER (LZMA_ALIGN + (1 << LZMA_NUM_ALIGN_BITS))
+#define LZMA_REP_LEN_CODER (LZMA_LEN_CODER + LZMA_NUM_LEN_PROBS)
+#define LZMA_LITERAL (LZMA_REP_LEN_CODER + LZMA_NUM_LEN_PROBS)
+
+
+struct writer {
+        uint8_t *buffer;
+        uint8_t previous_byte;
+        size_t buffer_pos;
+        int bufsize;
+        size_t global_pos;
+        int(*flush)(void*, unsigned int);
+        struct lzma_header *header;
+};
+
+struct cstate {
+        int state;
+        uint32_t rep0, rep1, rep2, rep3;
+};
+
+static inline size_t INIT get_pos(struct writer *wr)
+{
+        return
+                wr->global_pos + wr->buffer_pos;
+}
+
+static inline uint8_t INIT peek_old_byte(struct writer *wr,
+                                                uint32_t offs)
+{
+        if (!wr->flush) {
+                int32_t pos;
+                while (offs > wr->header->dict_size)
+                        offs -= wr->header->dict_size;
+                pos = wr->buffer_pos - offs;
+                return wr->buffer[pos];
+        } else {
+                uint32_t pos = wr->buffer_pos - offs;
+                while (pos >= wr->header->dict_size)
+                        pos += wr->header->dict_size;
+                return wr->buffer[pos];
+        }
+
+}
+
+static inline void INIT write_byte(struct writer *wr, uint8_t byte)
+{
+        wr->buffer[wr->buffer_pos++] = wr->previous_byte = byte;
+        if (wr->flush && wr->buffer_pos == wr->header->dict_size) {
+                wr->buffer_pos = 0;
+                wr->global_pos += wr->header->dict_size;
+                wr->flush((char *)wr->buffer, wr->header->dict_size);
+        }
+}
+
+
+static inline void INIT copy_byte(struct writer *wr, uint32_t offs)
+{
+        write_byte(wr, peek_old_byte(wr, offs));
+}
+
+static inline void INIT copy_bytes(struct writer *wr,
+                                         uint32_t rep0, int len)
+{
+        do {
+                copy_byte(wr, rep0);
+                len--;
+        } while (len != 0 && wr->buffer_pos < wr->header->dst_size);
+}
+
+static inline void INIT process_bit0(struct writer *wr, struct rc *rc,
+                                     struct cstate *cst, uint16_t *p,
+                                     int pos_state, uint16_t *prob,
+                                     int lc, uint32_t literal_pos_mask) {
+        int mi = 1;
+        rc_update_bit_0(rc, prob);
+        prob = (p + LZMA_LITERAL +
+                (LZMA_LIT_SIZE
+                 * (((get_pos(wr) & literal_pos_mask) << lc)
+                    + (wr->previous_byte >> (8 - lc))))
+                );
+
+        if (cst->state >= LZMA_NUM_LIT_STATES) {
+                int match_byte = peek_old_byte(wr, cst->rep0);
+                do {
+                        int bit;
+                        uint16_t *prob_lit;
+
+                        match_byte <<= 1;
+                        bit = match_byte & 0x100;
+                        prob_lit = prob + 0x100 + bit + mi;
+                        if (rc_get_bit(rc, prob_lit, &mi)) {
+                                if (!bit)
+                                        break;
+                        } else {
+                                if (bit)
+                                        break;
+                        }
+                } while (mi < 0x100);
+        }
+        while (mi < 0x100) {
+                uint16_t *prob_lit = prob + mi;
+                rc_get_bit(rc, prob_lit, &mi);
+        }
+        write_byte(wr, mi);
+        if (cst->state < 4)
+                cst->state = 0;
+        else if (cst->state < 10)
+                cst->state -= 3;
+        else
+                cst->state -= 6;
+}
+
+static inline void INIT process_bit1(struct writer *wr, struct rc *rc,
+                                            struct cstate *cst, uint16_t *p,
+                                            int pos_state, uint16_t *prob) {
+  int offset;
+        uint16_t *prob_len;
+        int num_bits;
+        int len;
+
+        rc_update_bit_1(rc, prob);
+        prob = p + LZMA_IS_REP + cst->state;
+        if (rc_is_bit_0(rc, prob)) {
+                rc_update_bit_0(rc, prob);
+                cst->rep3 = cst->rep2;
+                cst->rep2 = cst->rep1;
+                cst->rep1 = cst->rep0;
+                cst->state = cst->state < LZMA_NUM_LIT_STATES ? 0 : 3;
+                prob = p + LZMA_LEN_CODER;
+        } else {
+                rc_update_bit_1(rc, prob);
+                prob = p + LZMA_IS_REP_G0 + cst->state;
+                if (rc_is_bit_0(rc, prob)) {
+                        rc_update_bit_0(rc, prob);
+                        prob = (p + LZMA_IS_REP_0_LONG
+                                + (cst->state <<
+                                   LZMA_NUM_POS_BITS_MAX) +
+                                pos_state);
+                        if (rc_is_bit_0(rc, prob)) {
+                                rc_update_bit_0(rc, prob);
+
+                                cst->state = cst->state < LZMA_NUM_LIT_STATES ?
+                                        9 : 11;
+                                copy_byte(wr, cst->rep0);
+                                return;
+                        } else {
+                                rc_update_bit_1(rc, prob);
+                        }
+                } else {
+                        uint32_t distance;
+
+                        rc_update_bit_1(rc, prob);
+                        prob = p + LZMA_IS_REP_G1 + cst->state;
+                        if (rc_is_bit_0(rc, prob)) {
+                                rc_update_bit_0(rc, prob);
+                                distance = cst->rep1;
+                        } else {
+                                rc_update_bit_1(rc, prob);
+                                prob = p + LZMA_IS_REP_G2 + cst->state;
+                                if (rc_is_bit_0(rc, prob)) {
+                                        rc_update_bit_0(rc, prob);
+                                        distance = cst->rep2;
+                                } else {
+                                        rc_update_bit_1(rc, prob);
+                                        distance = cst->rep3;
+                                        cst->rep3 = cst->rep2;
+                                }
+                                cst->rep2 = cst->rep1;
+                        }
+                        cst->rep1 = cst->rep0;
+                        cst->rep0 = distance;
+                }
+                cst->state = cst->state < LZMA_NUM_LIT_STATES ? 8 : 11;
+                prob = p + LZMA_REP_LEN_CODER;
+        }
+
+        prob_len = prob + LZMA_LEN_CHOICE;
+        if (rc_is_bit_0(rc, prob_len)) {
+                rc_update_bit_0(rc, prob_len);
+                prob_len = (prob + LZMA_LEN_LOW
+                            + (pos_state <<
+                               LZMA_LEN_NUM_LOW_BITS));
+                offset = 0;
+                num_bits = LZMA_LEN_NUM_LOW_BITS;
+        } else {
+                rc_update_bit_1(rc, prob_len);
+                prob_len = prob + LZMA_LEN_CHOICE_2;
+                if (rc_is_bit_0(rc, prob_len)) {
+                        rc_update_bit_0(rc, prob_len);
+                        prob_len = (prob + LZMA_LEN_MID
+                                    + (pos_state <<
+                                       LZMA_LEN_NUM_MID_BITS));
+                        offset = 1 << LZMA_LEN_NUM_LOW_BITS;
+                        num_bits = LZMA_LEN_NUM_MID_BITS;
+                } else {
+                        rc_update_bit_1(rc, prob_len);
+                        prob_len = prob + LZMA_LEN_HIGH;
+                        offset = ((1 << LZMA_LEN_NUM_LOW_BITS)
+                                  + (1 << LZMA_LEN_NUM_MID_BITS));
+                        num_bits = LZMA_LEN_NUM_HIGH_BITS;
+                }
+        }
+
+        rc_bit_tree_decode(rc, prob_len, num_bits, &len);
+        len += offset;
+
+        if (cst->state < 4) {
+                int pos_slot;
+
+                cst->state += LZMA_NUM_LIT_STATES;
+                prob =
+                        p + LZMA_POS_SLOT +
+                        ((len <
+                          LZMA_NUM_LEN_TO_POS_STATES ? len :
+                          LZMA_NUM_LEN_TO_POS_STATES - 1)
+                         << LZMA_NUM_POS_SLOT_BITS);
+                rc_bit_tree_decode(rc, prob,
+                                   LZMA_NUM_POS_SLOT_BITS,
+                                   &pos_slot);
+                if (pos_slot >= LZMA_START_POS_MODEL_INDEX) {
+                        int i, mi;
+                        num_bits = (pos_slot >> 1) - 1;
+                        cst->rep0 = 2 | (pos_slot & 1);
+                        if (pos_slot < LZMA_END_POS_MODEL_INDEX) {
+                                cst->rep0 <<= num_bits;
+                                prob = p + LZMA_SPEC_POS +
+                                        cst->rep0 - pos_slot - 1;
+                        } else {
+                                num_bits -= LZMA_NUM_ALIGN_BITS;
+                                while (num_bits--)
+                                        cst->rep0 = (cst->rep0 << 1) |
+                                                rc_direct_bit(rc);
+                                prob = p + LZMA_ALIGN;
+                                cst->rep0 <<= LZMA_NUM_ALIGN_BITS;
+                                num_bits = LZMA_NUM_ALIGN_BITS;
+                        }
+                        i = 1;
+                        mi = 1;
+                        while (num_bits--) {
+                                if (rc_get_bit(rc, prob + mi, &mi))
+                                        cst->rep0 |= i;
+                                i <<= 1;
+                        }
+                } else
+                        cst->rep0 = pos_slot;
+                if (++(cst->rep0) == 0)
+                        return;
+        }
+
+        len += LZMA_MATCH_MIN_LEN;
+
+        copy_bytes(wr, cst->rep0, len);
+}
+
+
+
+STATIC inline int INIT unlzma(unsigned char *buf, int in_len,
+                              int(*fill)(void*, unsigned int),
+                              int(*flush)(void*, unsigned int),
+                              unsigned char *output,
+                              int *posp,
+                              void(*error_fn)(char *x)
+        )
+{
+        struct lzma_header header;
+        int lc, pb, lp;
+        uint32_t pos_state_mask;
+        uint32_t literal_pos_mask;
+        uint16_t *p;
+        int num_probs;
+        struct rc rc;
+        int i, mi;
+        struct writer wr;
+        struct cstate cst;
+        unsigned char *inbuf;
+        int ret = -1;
+
+        set_error_fn(error_fn);
+        if (!flush)
+                in_len -= 4; /* Uncompressed size hack active in pre-boot
+                                environment */
+        if (buf)
+                inbuf = buf;
+        else
+                inbuf = malloc(LZMA_IOBUF_SIZE);
+        if (!inbuf) {
+                error("Could not allocate input bufer");
+                goto exit_0;
+        }
+
+        cst.state = 0;
+        cst.rep0 = cst.rep1 = cst.rep2 = cst.rep3 = 1;
+
+        wr.header = &header;
+        wr.flush = flush;
+        wr.global_pos = 0;
+        wr.previous_byte = 0;
+        wr.buffer_pos = 0;
+
+        rc_init(&rc, fill, inbuf, in_len);
+
+        for (i = 0; i < sizeof(header); i++) {
+                if (rc.ptr >= rc.buffer_end)
+                        rc_read(&rc);
+                ((unsigned char *)&header)[i] = *rc.ptr++;
+        }
+
+        if (header.pos >= (9 * 5 * 5))
+                error("bad header");
+
+        mi = 0;
+        lc = header.pos;
+        while (lc >= 9) {
+                mi++;
+                lc -= 9;
+        }
+        pb = 0;
+        lp = mi;
+        while (lp >= 5) {
+                pb++;
+                lp -= 5;
+        }
+        pos_state_mask = (1 << pb) - 1;
+        literal_pos_mask = (1 << lp) - 1;
+
+        ENDIAN_CONVERT(header.dict_size);
+        ENDIAN_CONVERT(header.dst_size);
+
+        if (header.dict_size == 0)
+                header.dict_size = 1;
+
+        if (output)
+                wr.buffer = output;
+        else {
+                wr.bufsize = MIN(header.dst_size, header.dict_size);
+                wr.buffer = large_malloc(wr.bufsize);
+        }
+        if (wr.buffer == NULL)
+                goto exit_1;
+
+        num_probs = LZMA_BASE_SIZE + (LZMA_LIT_SIZE << (lc + lp));
+        p = (uint16_t *) large_malloc(num_probs * sizeof(*p));
+        if (p == 0)
+                goto exit_2;
+        num_probs = LZMA_LITERAL + (LZMA_LIT_SIZE << (lc + lp));
+        for (i = 0; i < num_probs; i++)
+                p[i] = (1 << RC_MODEL_TOTAL_BITS) >> 1;
+
+        rc_init_code(&rc);
+
+        while (get_pos(&wr) < header.dst_size) {
+                int pos_state = get_pos(&wr) & pos_state_mask;
+                uint16_t *prob = p + LZMA_IS_MATCH +
+                        (cst.state << LZMA_NUM_POS_BITS_MAX) + pos_state;
+                if (rc_is_bit_0(&rc, prob))
+                        process_bit0(&wr, &rc, &cst, p, pos_state, prob,
+                                     lc, literal_pos_mask);
+                else {
+                        process_bit1(&wr, &rc, &cst, p, pos_state, prob);
+                        if (cst.rep0 == 0)
+                                break;
+                }
+        }
+
+        if (posp)
+                *posp = rc.ptr-rc.buffer;
+        if (wr.flush)
+                wr.flush(wr.buffer, wr.buffer_pos);
+        ret = 0;
+        large_free(p);
+exit_2:
+        if (!output)
+                large_free(wr.buffer);
+exit_1:
+        if (!buf)
+                free(inbuf);
+exit_0:
+        return ret;
+}
+
+#define decompress unlzma
diff --git a/lib/zlib_inflate/inflate.h b/lib/zlib_inflate/inflate.h
index df8a6c92052d..3d17b3d1b21f 100644
--- a/lib/zlib_inflate/inflate.h
+++ b/lib/zlib_inflate/inflate.h
@@ -1,3 +1,6 @@
+#ifndef INFLATE_H
+#define INFLATE_H
+
 /* inflate.h -- internal inflate state definition
  * Copyright (C) 1995-2004 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
@@ -105,3 +108,4 @@ struct inflate_state {
     unsigned short work[288];   /* work area for code table building */
     code codes[ENOUGH];         /* space for code tables */
 };
+#endif
diff --git a/lib/zlib_inflate/inftrees.h b/lib/zlib_inflate/inftrees.h
index 5f5219b1240e..b70b4731ac7a 100644
--- a/lib/zlib_inflate/inftrees.h
+++ b/lib/zlib_inflate/inftrees.h
@@ -1,3 +1,6 @@
+#ifndef INFTREES_H
+#define INFTREES_H
+
 /* inftrees.h -- header to use inftrees.c
  * Copyright (C) 1995-2005 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
@@ -53,3 +56,4 @@ typedef enum {
 extern int zlib_inflate_table (codetype type, unsigned short *lens,
                              unsigned codes, code **table,
                              unsigned *bits, unsigned short *work);
+#endif
diff --git a/mm/Makefile b/mm/Makefile
index 72255be57f89..818569b68f46 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -30,6 +30,10 @@ obj-$(CONFIG_FAILSLAB) += failslab.o
 obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
 obj-$(CONFIG_MIGRATION) += migrate.o
+ifdef CONFIG_HAVE_DYNAMIC_PER_CPU_AREA
+obj-$(CONFIG_SMP) += percpu.o
+else
 obj-$(CONFIG_SMP) += allocpercpu.o
+endif
 obj-$(CONFIG_QUICKLIST) += quicklist.o
 obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o page_cgroup.o
diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c
index 4297bc41bfd2..3653c570232b 100644
--- a/mm/allocpercpu.c
+++ b/mm/allocpercpu.c
@@ -99,45 +99,51 @@ static int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
         __percpu_populate_mask((__pdata), (size), (gfp), &(mask))
 
 /**
- * percpu_alloc_mask - initial setup of per-cpu data
+ * alloc_percpu - initial setup of per-cpu data
  * @size: size of per-cpu object
- * @gfp: may sleep or not etc.
- * @mask: populate per-data for cpu's selected through mask bits
+ * @align: alignment
  *
- * Populating per-cpu data for all online cpu's would be a typical use case,
- * which is simplified by the percpu_alloc() wrapper.
- * Per-cpu objects are populated with zeroed buffers.
+ * Allocate dynamic percpu area.  Percpu objects are populated with
+ * zeroed buffers.
  */
-void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+void *__alloc_percpu(size_t size, size_t align)
 {
         /*
          * We allocate whole cache lines to avoid false sharing
          */
         size_t sz = roundup(nr_cpu_ids * sizeof(void *), cache_line_size());
-        void *pdata = kzalloc(sz, gfp);
+        void *pdata = kzalloc(sz, GFP_KERNEL);
         void *__pdata = __percpu_disguise(pdata);
 
+        /*
+         * Can't easily make larger alignment work with kmalloc.  WARN
+         * on it.  Larger alignment should only be used for module
+         * percpu sections on SMP for which this path isn't used.
+         */
+        WARN_ON_ONCE(align > __alignof__(unsigned long long));
+
         if (unlikely(!pdata))
                 return NULL;
-        if (likely(!__percpu_populate_mask(__pdata, size, gfp, mask)))
+        if (likely(!__percpu_populate_mask(__pdata, size, GFP_KERNEL,
+                                           &cpu_possible_map)))
                 return __pdata;
         kfree(pdata);
         return NULL;
 }
-EXPORT_SYMBOL_GPL(__percpu_alloc_mask);
+EXPORT_SYMBOL_GPL(__alloc_percpu);
 
 /**
- * percpu_free - final cleanup of per-cpu data
+ * free_percpu - final cleanup of per-cpu data
  * @__pdata: object to clean up
  *
  * We simply clean up any per-cpu object left. No need for the client to
  * track and specify through a bis mask which per-cpu objects are to free.
  */
-void percpu_free(void *__pdata)
+void free_percpu(void *__pdata)
 {
         if (unlikely(!__pdata))
                 return;
         __percpu_depopulate_mask(__pdata, &cpu_possible_map);
         kfree(__percpu_disguise(__pdata));
 }
-EXPORT_SYMBOL_GPL(percpu_free);
+EXPORT_SYMBOL_GPL(free_percpu);
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 51a0ccf61e0e..daf92713f7de 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -382,7 +382,6 @@ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
         return mark_bootmem_node(pgdat->bdata, start, end, 1, flags);
 }
 
-#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 /**
  * reserve_bootmem - mark a page range as usable
  * @addr: starting address of the range
@@ -403,7 +402,6 @@ int __init reserve_bootmem(unsigned long addr, unsigned long size,
 
         return mark_bootmem(start, end, 1, flags);
 }
-#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
 static unsigned long align_idx(struct bootmem_data *bdata, unsigned long idx,
                         unsigned long step)
@@ -429,8 +427,8 @@ static unsigned long align_off(struct bootmem_data *bdata, unsigned long off,
 }
 
 static void * __init alloc_bootmem_core(struct bootmem_data *bdata,
-                                unsigned long size, unsigned long align,
-                                unsigned long goal, unsigned long limit)
+                                        unsigned long size, unsigned long align,
+                                        unsigned long goal, unsigned long limit)
 {
         unsigned long fallback = 0;
         unsigned long min, max, start, sidx, midx, step;
@@ -530,17 +528,34 @@ find_block:
         return NULL;
 }
 
+static void * __init alloc_arch_preferred_bootmem(bootmem_data_t *bdata,
+                                        unsigned long size, unsigned long align,
+                                        unsigned long goal, unsigned long limit)
+{
+#ifdef CONFIG_HAVE_ARCH_BOOTMEM
+        bootmem_data_t *p_bdata;
+
+        p_bdata = bootmem_arch_preferred_node(bdata, size, align, goal, limit);
+        if (p_bdata)
+                return alloc_bootmem_core(p_bdata, size, align, goal, limit);
+#endif
+        return NULL;
+}
+
 static void * __init ___alloc_bootmem_nopanic(unsigned long size,
                                         unsigned long align,
                                         unsigned long goal,
                                         unsigned long limit)
 {
         bootmem_data_t *bdata;
+        void *region;
 
 restart:
-        list_for_each_entry(bdata, &bdata_list, list) {
-                void *region;
+        region = alloc_arch_preferred_bootmem(NULL, size, align, goal, limit);
+        if (region)
+                return region;
 
+        list_for_each_entry(bdata, &bdata_list, list) {
                 if (goal && bdata->node_low_pfn <= PFN_DOWN(goal))
                         continue;
                 if (limit && bdata->node_min_pfn >= PFN_DOWN(limit))
@@ -618,6 +633,10 @@ static void * __init ___alloc_bootmem_node(bootmem_data_t *bdata,
 {
         void *ptr;
 
+        ptr = alloc_arch_preferred_bootmem(bdata, size, align, goal, limit);
+        if (ptr)
+                return ptr;
+
         ptr = alloc_bootmem_core(bdata, size, align, goal, limit);
         if (ptr)
                 return ptr;
@@ -674,6 +693,10 @@ void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size,
 {
         void *ptr;
 
+        ptr = alloc_arch_preferred_bootmem(pgdat->bdata, size, align, goal, 0);
+        if (ptr)
+                return ptr;
+
         ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
         if (ptr)
                 return ptr;
diff --git a/mm/filemap.c b/mm/filemap.c
index 60fd56772cc6..126d3973b3d1 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1816,14 +1816,14 @@ EXPORT_SYMBOL(file_remove_suid);
 static size_t __iovec_copy_from_user_inatomic(char *vaddr,
                         const struct iovec *iov, size_t base, size_t bytes)
 {
-        size_t copied = 0, left = 0, total = bytes;
+        size_t copied = 0, left = 0;
 
         while (bytes) {
                 char __user *buf = iov->iov_base + base;
                 int copy = min(bytes, iov->iov_len - base);
 
                 base = 0;
-                left = __copy_from_user_inatomic_nocache(vaddr, buf, copy, total);
+                left = __copy_from_user_inatomic(vaddr, buf, copy);
                 copied += copy;
                 bytes -= copy;
                 vaddr += copy;
@@ -1851,9 +1851,7 @@ size_t iov_iter_copy_from_user_atomic(struct page *page,
         if (likely(i->nr_segs == 1)) {
                 int left;
                 char __user *buf = i->iov->iov_base + i->iov_offset;
-
-                left = __copy_from_user_inatomic_nocache(kaddr + offset,
-                                                        buf, bytes, bytes);
+                left = __copy_from_user_inatomic(kaddr + offset, buf, bytes);
                 copied = bytes - left;
         } else {
                 copied = __iovec_copy_from_user_inatomic(kaddr + offset,
@@ -1881,8 +1879,7 @@ size_t iov_iter_copy_from_user(struct page *page,
         if (likely(i->nr_segs == 1)) {
                 int left;
                 char __user *buf = i->iov->iov_base + i->iov_offset;
-
-                left = __copy_from_user_nocache(kaddr + offset, buf, bytes, bytes);
+                left = __copy_from_user(kaddr + offset, buf, bytes);
                 copied = bytes - left;
         } else {
                 copied = __iovec_copy_from_user_inatomic(kaddr + offset,
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
index bf54f8a2cf1d..0c04615651b7 100644
--- a/mm/filemap_xip.c
+++ b/mm/filemap_xip.c
@@ -354,7 +354,7 @@ __xip_file_write(struct file *filp, const char __user *buf,
                         break;
 
                 copied = bytes -
-                        __copy_from_user_nocache(xip_mem + offset, buf, bytes, bytes);
+                        __copy_from_user_nocache(xip_mem + offset, buf, bytes);
 
                 if (likely(copied > 0)) {
                         status = copied;
diff --git a/mm/percpu.c b/mm/percpu.c
new file mode 100644
index 000000000000..3d0f5456827c
--- /dev/null
+++ b/mm/percpu.c
@@ -0,0 +1,979 @@
+/*
+ * linux/mm/percpu.c - percpu memory allocator
+ *
+ * Copyright (C) 2009           SUSE Linux Products GmbH
+ * Copyright (C) 2009           Tejun Heo <tj@kernel.org>
+ *
+ * This file is released under the GPLv2.
+ *
+ * This is percpu allocator which can handle both static and dynamic
+ * areas.  Percpu areas are allocated in chunks in vmalloc area.  Each
+ * chunk is consisted of num_possible_cpus() units and the first chunk
+ * is used for static percpu variables in the kernel image (special
+ * boot time alloc/init handling necessary as these areas need to be
+ * brought up before allocation services are running).  Unit grows as
+ * necessary and all units grow or shrink in unison.  When a chunk is
+ * filled up, another chunk is allocated.  ie. in vmalloc area
+ *
+ *  c0                           c1                         c2
+ *  -------------------          -------------------        ------------
+ * | u0 | u1 | u2 | u3 |        | u0 | u1 | u2 | u3 |      | u0 | u1 | u
+ *  -------------------  ......  -------------------  ....  ------------
+ *
+ * Allocation is done in offset-size areas of single unit space.  Ie,
+ * an area of 512 bytes at 6k in c1 occupies 512 bytes at 6k of c1:u0,
+ * c1:u1, c1:u2 and c1:u3.  Percpu access can be done by configuring
+ * percpu base registers UNIT_SIZE apart.
+ *
+ * There are usually many small percpu allocations many of them as
+ * small as 4 bytes.  The allocator organizes chunks into lists
+ * according to free size and tries to allocate from the fullest one.
+ * Each chunk keeps the maximum contiguous area size hint which is
+ * guaranteed to be eqaul to or larger than the maximum contiguous
+ * area in the chunk.  This helps the allocator not to iterate the
+ * chunk maps unnecessarily.
+ *
+ * Allocation state in each chunk is kept using an array of integers
+ * on chunk->map.  A positive value in the map represents a free
+ * region and negative allocated.  Allocation inside a chunk is done
+ * by scanning this map sequentially and serving the first matching
+ * entry.  This is mostly copied from the percpu_modalloc() allocator.
+ * Chunks are also linked into a rb tree to ease address to chunk
+ * mapping during free.
+ *
+ * To use this allocator, arch code should do the followings.
+ *
+ * - define CONFIG_HAVE_DYNAMIC_PER_CPU_AREA
+ *
+ * - define __addr_to_pcpu_ptr() and __pcpu_ptr_to_addr() to translate
+ *   regular address to percpu pointer and back
+ *
+ * - use pcpu_setup_first_chunk() during percpu area initialization to
+ *   setup the first chunk containing the kernel static percpu area
+ */
+
+#include <linux/bitmap.h>
+#include <linux/bootmem.h>
+#include <linux/list.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/percpu.h>
+#include <linux/pfn.h>
+#include <linux/rbtree.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+
+#define PCPU_SLOT_BASE_SHIFT            5       /* 1-31 shares the same slot */
+#define PCPU_DFL_MAP_ALLOC              16      /* start a map with 16 ents */
+
+struct pcpu_chunk {
+        struct list_head        list;           /* linked to pcpu_slot lists */
+        struct rb_node          rb_node;        /* key is chunk->vm->addr */
+        int                     free_size;      /* free bytes in the chunk */
+        int                     contig_hint;    /* max contiguous size hint */
+        struct vm_struct        *vm;            /* mapped vmalloc region */
+        int                     map_used;       /* # of map entries used */
+        int                     map_alloc;      /* # of map entries allocated */
+        int                     *map;           /* allocation map */
+        bool                    immutable;      /* no [de]population allowed */
+        struct page             *page[];        /* #cpus * UNIT_PAGES */
+};
+
+static int pcpu_unit_pages __read_mostly;
+static int pcpu_unit_size __read_mostly;
+static int pcpu_chunk_size __read_mostly;
+static int pcpu_nr_slots __read_mostly;
+static size_t pcpu_chunk_struct_size __read_mostly;
+
+/* the address of the first chunk which starts with the kernel static area */
+void *pcpu_base_addr __read_mostly;
+EXPORT_SYMBOL_GPL(pcpu_base_addr);
+
+/* the size of kernel static area */
+static int pcpu_static_size __read_mostly;
+
+/*
+ * One mutex to rule them all.
+ *
+ * The following mutex is grabbed in the outermost public alloc/free
+ * interface functions and released only when the operation is
+ * complete.  As such, every function in this file other than the
+ * outermost functions are called under pcpu_mutex.
+ *
+ * It can easily be switched to use spinlock such that only the area
+ * allocation and page population commit are protected with it doing
+ * actual [de]allocation without holding any lock.  However, given
+ * what this allocator does, I think it's better to let them run
+ * sequentially.
+ */
+static DEFINE_MUTEX(pcpu_mutex);
+
+static struct list_head *pcpu_slot __read_mostly; /* chunk list slots */
+static struct rb_root pcpu_addr_root = RB_ROOT; /* chunks by address */
+
+static int __pcpu_size_to_slot(int size)
+{
+        int highbit = fls(size);        /* size is in bytes */
+        return max(highbit - PCPU_SLOT_BASE_SHIFT + 2, 1);
+}
+
+static int pcpu_size_to_slot(int size)
+{
+        if (size == pcpu_unit_size)
+                return pcpu_nr_slots - 1;
+        return __pcpu_size_to_slot(size);
+}
+
+static int pcpu_chunk_slot(const struct pcpu_chunk *chunk)
+{
+        if (chunk->free_size < sizeof(int) || chunk->contig_hint < sizeof(int))
+                return 0;
+
+        return pcpu_size_to_slot(chunk->free_size);
+}
+
+static int pcpu_page_idx(unsigned int cpu, int page_idx)
+{
+        return cpu * pcpu_unit_pages + page_idx;
+}
+
+static struct page **pcpu_chunk_pagep(struct pcpu_chunk *chunk,
+                                      unsigned int cpu, int page_idx)
+{
+        return &chunk->page[pcpu_page_idx(cpu, page_idx)];
+}
+
+static unsigned long pcpu_chunk_addr(struct pcpu_chunk *chunk,
+                                     unsigned int cpu, int page_idx)
+{
+        return (unsigned long)chunk->vm->addr +
+                (pcpu_page_idx(cpu, page_idx) << PAGE_SHIFT);
+}
+
+static bool pcpu_chunk_page_occupied(struct pcpu_chunk *chunk,
+                                     int page_idx)
+{
+        return *pcpu_chunk_pagep(chunk, 0, page_idx) != NULL;
+}
+
+/**
+ * pcpu_realloc - versatile realloc
+ * @p: the current pointer (can be NULL for new allocations)
+ * @size: the current size in bytes (can be 0 for new allocations)
+ * @new_size: the wanted new size in bytes (can be 0 for free)
+ *
+ * More robust realloc which can be used to allocate, resize or free a
+ * memory area of arbitrary size.  If the needed size goes over
+ * PAGE_SIZE, kernel VM is used.
+ *
+ * RETURNS:
+ * The new pointer on success, NULL on failure.
+ */
+static void *pcpu_realloc(void *p, size_t size, size_t new_size)
+{
+        void *new;
+
+        if (new_size <= PAGE_SIZE)
+                new = kmalloc(new_size, GFP_KERNEL);
+        else
+                new = vmalloc(new_size);
+        if (new_size && !new)
+                return NULL;
+
+        memcpy(new, p, min(size, new_size));
+        if (new_size > size)
+                memset(new + size, 0, new_size - size);
+
+        if (size <= PAGE_SIZE)
+                kfree(p);
+        else
+                vfree(p);
+
+        return new;
+}
+
+/**
+ * pcpu_chunk_relocate - put chunk in the appropriate chunk slot
+ * @chunk: chunk of interest
+ * @oslot: the previous slot it was on
+ *
+ * This function is called after an allocation or free changed @chunk.
+ * New slot according to the changed state is determined and @chunk is
+ * moved to the slot.
+ */
+static void pcpu_chunk_relocate(struct pcpu_chunk *chunk, int oslot)
+{
+        int nslot = pcpu_chunk_slot(chunk);
+
+        if (oslot != nslot) {
+                if (oslot < nslot)
+                        list_move(&chunk->list, &pcpu_slot[nslot]);
+                else
+                        list_move_tail(&chunk->list, &pcpu_slot[nslot]);
+        }
+}
+
+static struct rb_node **pcpu_chunk_rb_search(void *addr,
+                                             struct rb_node **parentp)
+{
+        struct rb_node **p = &pcpu_addr_root.rb_node;
+        struct rb_node *parent = NULL;
+        struct pcpu_chunk *chunk;
+
+        while (*p) {
+                parent = *p;
+                chunk = rb_entry(parent, struct pcpu_chunk, rb_node);
+
+                if (addr < chunk->vm->addr)
+                        p = &(*p)->rb_left;
+                else if (addr > chunk->vm->addr)
+                        p = &(*p)->rb_right;
+                else
+                        break;
+        }
+
+        if (parentp)
+                *parentp = parent;
+        return p;
+}
+
+/**
+ * pcpu_chunk_addr_search - search for chunk containing specified address
+ * @addr: address to search for
+ *
+ * Look for chunk which might contain @addr.  More specifically, it
+ * searchs for the chunk with the highest start address which isn't
+ * beyond @addr.
+ *
+ * RETURNS:
+ * The address of the found chunk.
+ */
+static struct pcpu_chunk *pcpu_chunk_addr_search(void *addr)
+{
+        struct rb_node *n, *parent;
+        struct pcpu_chunk *chunk;
+
+        n = *pcpu_chunk_rb_search(addr, &parent);
+        if (!n) {
+                /* no exactly matching chunk, the parent is the closest */
+                n = parent;
+                BUG_ON(!n);
+        }
+        chunk = rb_entry(n, struct pcpu_chunk, rb_node);
+
+        if (addr < chunk->vm->addr) {
+                /* the parent was the next one, look for the previous one */
+                n = rb_prev(n);
+                BUG_ON(!n);
+                chunk = rb_entry(n, struct pcpu_chunk, rb_node);
+        }
+
+        return chunk;
+}
+
+/**
+ * pcpu_chunk_addr_insert - insert chunk into address rb tree
+ * @new: chunk to insert
+ *
+ * Insert @new into address rb tree.
+ */
+static void pcpu_chunk_addr_insert(struct pcpu_chunk *new)
+{
+        struct rb_node **p, *parent;
+
+        p = pcpu_chunk_rb_search(new->vm->addr, &parent);
+        BUG_ON(*p);
+        rb_link_node(&new->rb_node, parent, p);
+        rb_insert_color(&new->rb_node, &pcpu_addr_root);
+}
+
+/**
+ * pcpu_split_block - split a map block
+ * @chunk: chunk of interest
+ * @i: index of map block to split
+ * @head: head size in bytes (can be 0)
+ * @tail: tail size in bytes (can be 0)
+ *
+ * Split the @i'th map block into two or three blocks.  If @head is
+ * non-zero, @head bytes block is inserted before block @i moving it
+ * to @i+1 and reducing its size by @head bytes.
+ *
+ * If @tail is non-zero, the target block, which can be @i or @i+1
+ * depending on @head, is reduced by @tail bytes and @tail byte block
+ * is inserted after the target block.
+ *
+ * RETURNS:
+ * 0 on success, -errno on failure.
+ */
+static int pcpu_split_block(struct pcpu_chunk *chunk, int i, int head, int tail)
+{
+        int nr_extra = !!head + !!tail;
+        int target = chunk->map_used + nr_extra;
+
+        /* reallocation required? */
+        if (chunk->map_alloc < target) {
+                int new_alloc = chunk->map_alloc;
+                int *new;
+
+                while (new_alloc < target)
+                        new_alloc *= 2;
+
+                new = pcpu_realloc(chunk->map,
+                                   chunk->map_alloc * sizeof(new[0]),
+                                   new_alloc * sizeof(new[0]));
+                if (!new)
+                        return -ENOMEM;
+
+                chunk->map_alloc = new_alloc;
+                chunk->map = new;
+        }
+
+        /* insert a new subblock */
+        memmove(&chunk->map[i + nr_extra], &chunk->map[i],
+                sizeof(chunk->map[0]) * (chunk->map_used - i));
+        chunk->map_used += nr_extra;
+
+        if (head) {
+                chunk->map[i + 1] = chunk->map[i] - head;
+                chunk->map[i++] = head;
+        }
+        if (tail) {
+                chunk->map[i++] -= tail;
+                chunk->map[i] = tail;
+        }
+        return 0;
+}
+
+/**
+ * pcpu_alloc_area - allocate area from a pcpu_chunk
+ * @chunk: chunk of interest
+ * @size: wanted size in bytes
+ * @align: wanted align
+ *
+ * Try to allocate @size bytes area aligned at @align from @chunk.
+ * Note that this function only allocates the offset.  It doesn't
+ * populate or map the area.
+ *
+ * RETURNS:
+ * Allocated offset in @chunk on success, -errno on failure.
+ */
+static int pcpu_alloc_area(struct pcpu_chunk *chunk, int size, int align)
+{
+        int oslot = pcpu_chunk_slot(chunk);
+        int max_contig = 0;
+        int i, off;
+
+        /*
+         * The static chunk initially doesn't have map attached
+         * because kmalloc wasn't available during init.  Give it one.
+         */
+        if (unlikely(!chunk->map)) {
+                chunk->map = pcpu_realloc(NULL, 0,
+                                PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0]));
+                if (!chunk->map)
+                        return -ENOMEM;
+
+                chunk->map_alloc = PCPU_DFL_MAP_ALLOC;
+                chunk->map[chunk->map_used++] = -pcpu_static_size;
+                if (chunk->free_size)
+                        chunk->map[chunk->map_used++] = chunk->free_size;
+        }
+
+        for (i = 0, off = 0; i < chunk->map_used; off += abs(chunk->map[i++])) {
+                bool is_last = i + 1 == chunk->map_used;
+                int head, tail;
+
+                /* extra for alignment requirement */
+                head = ALIGN(off, align) - off;
+                BUG_ON(i == 0 && head != 0);
+
+                if (chunk->map[i] < 0)
+                        continue;
+                if (chunk->map[i] < head + size) {
+                        max_contig = max(chunk->map[i], max_contig);
+                        continue;
+                }
+
+                /*
+                 * If head is small or the previous block is free,
+                 * merge'em.  Note that 'small' is defined as smaller
+                 * than sizeof(int), which is very small but isn't too
+                 * uncommon for percpu allocations.
+                 */
+                if (head && (head < sizeof(int) || chunk->map[i - 1] > 0)) {
+                        if (chunk->map[i - 1] > 0)
+                                chunk->map[i - 1] += head;
+                        else {
+                                chunk->map[i - 1] -= head;
+                                chunk->free_size -= head;
+                        }
+                        chunk->map[i] -= head;
+                        off += head;
+                        head = 0;
+                }
+
+                /* if tail is small, just keep it around */
+                tail = chunk->map[i] - head - size;
+                if (tail < sizeof(int))
+                        tail = 0;
+
+                /* split if warranted */
+                if (head || tail) {
+                        if (pcpu_split_block(chunk, i, head, tail))
+                                return -ENOMEM;
+                        if (head) {
+                                i++;
+                                off += head;
+                                max_contig = max(chunk->map[i - 1], max_contig);
+                        }
+                        if (tail)
+                                max_contig = max(chunk->map[i + 1], max_contig);
+                }
+
+                /* update hint and mark allocated */
+                if (is_last)
+                        chunk->contig_hint = max_contig; /* fully scanned */
+                else
+                        chunk->contig_hint = max(chunk->contig_hint,
+                                                 max_contig);
+
+                chunk->free_size -= chunk->map[i];
+                chunk->map[i] = -chunk->map[i];
+
+                pcpu_chunk_relocate(chunk, oslot);
+                return off;
+        }
+
+        chunk->contig_hint = max_contig;        /* fully scanned */
+        pcpu_chunk_relocate(chunk, oslot);
+
+        /*
+         * Tell the upper layer that this chunk has no area left.
+         * Note that this is not an error condition but a notification
+         * to upper layer that it needs to look at other chunks.
+         * -ENOSPC is chosen as it isn't used in memory subsystem and
+         * matches the meaning in a way.
+         */
+        return -ENOSPC;
+}
+
+/**
+ * pcpu_free_area - free area to a pcpu_chunk
+ * @chunk: chunk of interest
+ * @freeme: offset of area to free
+ *
+ * Free area starting from @freeme to @chunk.  Note that this function
+ * only modifies the allocation map.  It doesn't depopulate or unmap
+ * the area.
+ */
+static void pcpu_free_area(struct pcpu_chunk *chunk, int freeme)
+{
+        int oslot = pcpu_chunk_slot(chunk);
+        int i, off;
+
+        for (i = 0, off = 0; i < chunk->map_used; off += abs(chunk->map[i++]))
+                if (off == freeme)
+                        break;
+        BUG_ON(off != freeme);
+        BUG_ON(chunk->map[i] > 0);
+
+        chunk->map[i] = -chunk->map[i];
+        chunk->free_size += chunk->map[i];
+
+        /* merge with previous? */
+        if (i > 0 && chunk->map[i - 1] >= 0) {
+                chunk->map[i - 1] += chunk->map[i];
+                chunk->map_used--;
+                memmove(&chunk->map[i], &chunk->map[i + 1],
+                        (chunk->map_used - i) * sizeof(chunk->map[0]));
+                i--;
+        }
+        /* merge with next? */
+        if (i + 1 < chunk->map_used && chunk->map[i + 1] >= 0) {
+                chunk->map[i] += chunk->map[i + 1];
+                chunk->map_used--;
+                memmove(&chunk->map[i + 1], &chunk->map[i + 2],
+                        (chunk->map_used - (i + 1)) * sizeof(chunk->map[0]));
+        }
+
+        chunk->contig_hint = max(chunk->map[i], chunk->contig_hint);
+        pcpu_chunk_relocate(chunk, oslot);
+}
+
+/**
+ * pcpu_unmap - unmap pages out of a pcpu_chunk
+ * @chunk: chunk of interest
+ * @page_start: page index of the first page to unmap
+ * @page_end: page index of the last page to unmap + 1
+ * @flush: whether to flush cache and tlb or not
+ *
+ * For each cpu, unmap pages [@page_start,@page_end) out of @chunk.
+ * If @flush is true, vcache is flushed before unmapping and tlb
+ * after.
+ */
+static void pcpu_unmap(struct pcpu_chunk *chunk, int page_start, int page_end,
+                       bool flush)
+{
+        unsigned int last = num_possible_cpus() - 1;
+        unsigned int cpu;
+
+        /* unmap must not be done on immutable chunk */
+        WARN_ON(chunk->immutable);
+
+        /*
+         * Each flushing trial can be very expensive, issue flush on
+         * the whole region at once rather than doing it for each cpu.
+         * This could be an overkill but is more scalable.
+         */
+        if (flush)
+                flush_cache_vunmap(pcpu_chunk_addr(chunk, 0, page_start),
+                                   pcpu_chunk_addr(chunk, last, page_end));
+
+        for_each_possible_cpu(cpu)
+                unmap_kernel_range_noflush(
+                                pcpu_chunk_addr(chunk, cpu, page_start),
+                                (page_end - page_start) << PAGE_SHIFT);
+
+        /* ditto as flush_cache_vunmap() */
+        if (flush)
+                flush_tlb_kernel_range(pcpu_chunk_addr(chunk, 0, page_start),
+                                       pcpu_chunk_addr(chunk, last, page_end));
+}
+
+/**
+ * pcpu_depopulate_chunk - depopulate and unmap an area of a pcpu_chunk
+ * @chunk: chunk to depopulate
+ * @off: offset to the area to depopulate
+ * @size: size of the area to depopulate in bytes
+ * @flush: whether to flush cache and tlb or not
+ *
+ * For each cpu, depopulate and unmap pages [@page_start,@page_end)
+ * from @chunk.  If @flush is true, vcache is flushed before unmapping
+ * and tlb after.
+ */
+static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk, int off, int size,
+                                  bool flush)
+{
+        int page_start = PFN_DOWN(off);
+        int page_end = PFN_UP(off + size);
+        int unmap_start = -1;
+        int uninitialized_var(unmap_end);
+        unsigned int cpu;
+        int i;
+
+        for (i = page_start; i < page_end; i++) {
+                for_each_possible_cpu(cpu) {
+                        struct page **pagep = pcpu_chunk_pagep(chunk, cpu, i);
+
+                        if (!*pagep)
+                                continue;
+
+                        __free_page(*pagep);
+
+                        /*
+                         * If it's partial depopulation, it might get
+                         * populated or depopulated again.  Mark the
+                         * page gone.
+                         */
+                        *pagep = NULL;
+
+                        unmap_start = unmap_start < 0 ? i : unmap_start;
+                        unmap_end = i + 1;
+                }
+        }
+
+        if (unmap_start >= 0)
+                pcpu_unmap(chunk, unmap_start, unmap_end, flush);
+}
+
+/**
+ * pcpu_map - map pages into a pcpu_chunk
+ * @chunk: chunk of interest
+ * @page_start: page index of the first page to map
+ * @page_end: page index of the last page to map + 1
+ *
+ * For each cpu, map pages [@page_start,@page_end) into @chunk.
+ * vcache is flushed afterwards.
+ */
+static int pcpu_map(struct pcpu_chunk *chunk, int page_start, int page_end)
+{
+        unsigned int last = num_possible_cpus() - 1;
+        unsigned int cpu;
+        int err;
+
+        /* map must not be done on immutable chunk */
+        WARN_ON(chunk->immutable);
+
+        for_each_possible_cpu(cpu) {
+                err = map_kernel_range_noflush(
+                                pcpu_chunk_addr(chunk, cpu, page_start),
+                                (page_end - page_start) << PAGE_SHIFT,
+                                PAGE_KERNEL,
+                                pcpu_chunk_pagep(chunk, cpu, page_start));
+                if (err < 0)
+                        return err;
+        }
+
+        /* flush at once, please read comments in pcpu_unmap() */
+        flush_cache_vmap(pcpu_chunk_addr(chunk, 0, page_start),
+                         pcpu_chunk_addr(chunk, last, page_end));
+        return 0;
+}
+
+/**
+ * pcpu_populate_chunk - populate and map an area of a pcpu_chunk
+ * @chunk: chunk of interest
+ * @off: offset to the area to populate
+ * @size: size of the area to populate in bytes
+ *
+ * For each cpu, populate and map pages [@page_start,@page_end) into
+ * @chunk.  The area is cleared on return.
+ */
+static int pcpu_populate_chunk(struct pcpu_chunk *chunk, int off, int size)
+{
+        const gfp_t alloc_mask = GFP_KERNEL | __GFP_HIGHMEM | __GFP_COLD;
+        int page_start = PFN_DOWN(off);
+        int page_end = PFN_UP(off + size);
+        int map_start = -1;
+        int uninitialized_var(map_end);
+        unsigned int cpu;
+        int i;
+
+        for (i = page_start; i < page_end; i++) {
+                if (pcpu_chunk_page_occupied(chunk, i)) {
+                        if (map_start >= 0) {
+                                if (pcpu_map(chunk, map_start, map_end))
+                                        goto err;
+                                map_start = -1;
+                        }
+                        continue;
+                }
+
+                map_start = map_start < 0 ? i : map_start;
+                map_end = i + 1;
+
+                for_each_possible_cpu(cpu) {
+                        struct page **pagep = pcpu_chunk_pagep(chunk, cpu, i);
+
+                        *pagep = alloc_pages_node(cpu_to_node(cpu),
+                                                  alloc_mask, 0);
+                        if (!*pagep)
+                                goto err;
+                }
+        }
+
+        if (map_start >= 0 && pcpu_map(chunk, map_start, map_end))
+                goto err;
+
+        for_each_possible_cpu(cpu)
+                memset(chunk->vm->addr + cpu * pcpu_unit_size + off, 0,
+                       size);
+
+        return 0;
+err:
+        /* likely under heavy memory pressure, give memory back */
+        pcpu_depopulate_chunk(chunk, off, size, true);
+        return -ENOMEM;
+}
+
+static void free_pcpu_chunk(struct pcpu_chunk *chunk)
+{
+        if (!chunk)
+                return;
+        if (chunk->vm)
+                free_vm_area(chunk->vm);
+        pcpu_realloc(chunk->map, chunk->map_alloc * sizeof(chunk->map[0]), 0);
+        kfree(chunk);
+}
+
+static struct pcpu_chunk *alloc_pcpu_chunk(void)
+{
+        struct pcpu_chunk *chunk;
+
+        chunk = kzalloc(pcpu_chunk_struct_size, GFP_KERNEL);
+        if (!chunk)
+                return NULL;
+
+        chunk->map = pcpu_realloc(NULL, 0,
+                                  PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0]));
+        chunk->map_alloc = PCPU_DFL_MAP_ALLOC;
+        chunk->map[chunk->map_used++] = pcpu_unit_size;
+
+        chunk->vm = get_vm_area(pcpu_chunk_size, GFP_KERNEL);
+        if (!chunk->vm) {
+                free_pcpu_chunk(chunk);
+                return NULL;
+        }
+
+        INIT_LIST_HEAD(&chunk->list);
+        chunk->free_size = pcpu_unit_size;
+        chunk->contig_hint = pcpu_unit_size;
+
+        return chunk;
+}
+
+/**
+ * __alloc_percpu - allocate percpu area
+ * @size: size of area to allocate in bytes
+ * @align: alignment of area (max PAGE_SIZE)
+ *
+ * Allocate percpu area of @size bytes aligned at @align.  Might
+ * sleep.  Might trigger writeouts.
+ *
+ * RETURNS:
+ * Percpu pointer to the allocated area on success, NULL on failure.
+ */
+void *__alloc_percpu(size_t size, size_t align)
+{
+        void *ptr = NULL;
+        struct pcpu_chunk *chunk;
+        int slot, off;
+
+        if (unlikely(!size || size > PCPU_MIN_UNIT_SIZE || align > PAGE_SIZE)) {
+                WARN(true, "illegal size (%zu) or align (%zu) for "
+                     "percpu allocation\n", size, align);
+                return NULL;
+        }
+
+        mutex_lock(&pcpu_mutex);
+
+        /* allocate area */
+        for (slot = pcpu_size_to_slot(size); slot < pcpu_nr_slots; slot++) {
+                list_for_each_entry(chunk, &pcpu_slot[slot], list) {
+                        if (size > chunk->contig_hint)
+                                continue;
+                        off = pcpu_alloc_area(chunk, size, align);
+                        if (off >= 0)
+                                goto area_found;
+                        if (off != -ENOSPC)
+                                goto out_unlock;
+                }
+        }
+
+        /* hmmm... no space left, create a new chunk */
+        chunk = alloc_pcpu_chunk();
+        if (!chunk)
+                goto out_unlock;
+        pcpu_chunk_relocate(chunk, -1);
+        pcpu_chunk_addr_insert(chunk);
+
+        off = pcpu_alloc_area(chunk, size, align);
+        if (off < 0)
+                goto out_unlock;
+
+area_found:
+        /* populate, map and clear the area */
+        if (pcpu_populate_chunk(chunk, off, size)) {
+                pcpu_free_area(chunk, off);
+                goto out_unlock;
+        }
+
+        ptr = __addr_to_pcpu_ptr(chunk->vm->addr + off);
+out_unlock:
+        mutex_unlock(&pcpu_mutex);
+        return ptr;
+}
+EXPORT_SYMBOL_GPL(__alloc_percpu);
+
+static void pcpu_kill_chunk(struct pcpu_chunk *chunk)
+{
+        WARN_ON(chunk->immutable);
+        pcpu_depopulate_chunk(chunk, 0, pcpu_unit_size, false);
+        list_del(&chunk->list);
+        rb_erase(&chunk->rb_node, &pcpu_addr_root);
+        free_pcpu_chunk(chunk);
+}
+
+/**
+ * free_percpu - free percpu area
+ * @ptr: pointer to area to free
+ *
+ * Free percpu area @ptr.  Might sleep.
+ */
+void free_percpu(void *ptr)
+{
+        void *addr = __pcpu_ptr_to_addr(ptr);
+        struct pcpu_chunk *chunk;
+        int off;
+
+        if (!ptr)
+                return;
+
+        mutex_lock(&pcpu_mutex);
+
+        chunk = pcpu_chunk_addr_search(addr);
+        off = addr - chunk->vm->addr;
+
+        pcpu_free_area(chunk, off);
+
+        /* the chunk became fully free, kill one if there are other free ones */
+        if (chunk->free_size == pcpu_unit_size) {
+                struct pcpu_chunk *pos;
+
+                list_for_each_entry(pos,
+                                    &pcpu_slot[pcpu_chunk_slot(chunk)], list)
+                        if (pos != chunk) {
+                                pcpu_kill_chunk(pos);
+                                break;
+                        }
+        }
+
+        mutex_unlock(&pcpu_mutex);
+}
+EXPORT_SYMBOL_GPL(free_percpu);
+
+/**
+ * pcpu_setup_first_chunk - initialize the first percpu chunk
+ * @get_page_fn: callback to fetch page pointer
+ * @static_size: the size of static percpu area in bytes
+ * @unit_size: unit size in bytes, must be multiple of PAGE_SIZE, 0 for auto
+ * @free_size: free size in bytes, 0 for auto
+ * @base_addr: mapped address, NULL for auto
+ * @populate_pte_fn: callback to allocate pagetable, NULL if unnecessary
+ *
+ * Initialize the first percpu chunk which contains the kernel static
+ * perpcu area.  This function is to be called from arch percpu area
+ * setup path.  The first two parameters are mandatory.  The rest are
+ * optional.
+ *
+ * @get_page_fn() should return pointer to percpu page given cpu
+ * number and page number.  It should at least return enough pages to
+ * cover the static area.  The returned pages for static area should
+ * have been initialized with valid data.  If @unit_size is specified,
+ * it can also return pages after the static area.  NULL return
+ * indicates end of pages for the cpu.  Note that @get_page_fn() must
+ * return the same number of pages for all cpus.
+ *
+ * @unit_size, if non-zero, determines unit size and must be aligned
+ * to PAGE_SIZE and equal to or larger than @static_size + @free_size.
+ *
+ * @free_size determines the number of free bytes after the static
+ * area in the first chunk.  If zero, whatever left is available.
+ * Specifying non-zero value make percpu leave the area after
+ * @static_size + @free_size alone.
+ *
+ * Non-null @base_addr means that the caller already allocated virtual
+ * region for the first chunk and mapped it.  percpu must not mess
+ * with the chunk.  Note that @base_addr with 0 @unit_size or non-NULL
+ * @populate_pte_fn doesn't make any sense.
+ *
+ * @populate_pte_fn is used to populate the pagetable.  NULL means the
+ * caller already populated the pagetable.
+ *
+ * RETURNS:
+ * The determined pcpu_unit_size which can be used to initialize
+ * percpu access.
+ */
+size_t __init pcpu_setup_first_chunk(pcpu_get_page_fn_t get_page_fn,
+                                     size_t static_size, size_t unit_size,
+                                     size_t free_size, void *base_addr,
+                                     pcpu_populate_pte_fn_t populate_pte_fn)
+{
+        static struct vm_struct static_vm;
+        struct pcpu_chunk *static_chunk;
+        unsigned int cpu;
+        int nr_pages;
+        int err, i;
+
+        /* santiy checks */
+        BUG_ON(!static_size);
+        BUG_ON(!unit_size && free_size);
+        BUG_ON(unit_size && unit_size < static_size + free_size);
+        BUG_ON(unit_size & ~PAGE_MASK);
+        BUG_ON(base_addr && !unit_size);
+        BUG_ON(base_addr && populate_pte_fn);
+
+        if (unit_size)
+                pcpu_unit_pages = unit_size >> PAGE_SHIFT;
+        else
+                pcpu_unit_pages = max_t(int, PCPU_MIN_UNIT_SIZE >> PAGE_SHIFT,
+                                        PFN_UP(static_size));
+
+        pcpu_static_size = static_size;
+        pcpu_unit_size = pcpu_unit_pages << PAGE_SHIFT;
+        pcpu_chunk_size = num_possible_cpus() * pcpu_unit_size;
+        pcpu_chunk_struct_size = sizeof(struct pcpu_chunk)
+                + num_possible_cpus() * pcpu_unit_pages * sizeof(struct page *);
+
+        /*
+         * Allocate chunk slots.  The additional last slot is for
+         * empty chunks.
+         */
+        pcpu_nr_slots = __pcpu_size_to_slot(pcpu_unit_size) + 2;
+        pcpu_slot = alloc_bootmem(pcpu_nr_slots * sizeof(pcpu_slot[0]));
+        for (i = 0; i < pcpu_nr_slots; i++)
+                INIT_LIST_HEAD(&pcpu_slot[i]);
+
+        /* init static_chunk */
+        static_chunk = alloc_bootmem(pcpu_chunk_struct_size);
+        INIT_LIST_HEAD(&static_chunk->list);
+        static_chunk->vm = &static_vm;
+
+        if (free_size)
+                static_chunk->free_size = free_size;
+        else
+                static_chunk->free_size = pcpu_unit_size - pcpu_static_size;
+
+        static_chunk->contig_hint = static_chunk->free_size;
+
+        /* allocate vm address */
+        static_vm.flags = VM_ALLOC;
+        static_vm.size = pcpu_chunk_size;
+
+        if (!base_addr)
+                vm_area_register_early(&static_vm, PAGE_SIZE);
+        else {
+                /*
+                 * Pages already mapped.  No need to remap into
+                 * vmalloc area.  In this case the static chunk can't
+                 * be mapped or unmapped by percpu and is marked
+                 * immutable.
+                 */
+                static_vm.addr = base_addr;
+                static_chunk->immutable = true;
+        }
+
+        /* assign pages */
+        nr_pages = -1;
+        for_each_possible_cpu(cpu) {
+                for (i = 0; i < pcpu_unit_pages; i++) {
+                        struct page *page = get_page_fn(cpu, i);
+
+                        if (!page)
+                                break;
+                        *pcpu_chunk_pagep(static_chunk, cpu, i) = page;
+                }
+
+                BUG_ON(i < PFN_UP(pcpu_static_size));
+
+                if (nr_pages < 0)
+                        nr_pages = i;
+                else
+                        BUG_ON(nr_pages != i);
+        }
+
+        /* map them */
+        if (populate_pte_fn) {
+                for_each_possible_cpu(cpu)
+                        for (i = 0; i < nr_pages; i++)
+                                populate_pte_fn(pcpu_chunk_addr(static_chunk,
+                                                                cpu, i));
+
+                err = pcpu_map(static_chunk, 0, nr_pages);
+                if (err)
+                        panic("failed to setup static percpu area, err=%d\n",
+                              err);
+        }
+
+        /* link static_chunk in */
+        pcpu_chunk_relocate(static_chunk, -1);
+        pcpu_chunk_addr_insert(static_chunk);
+
+        /* we're done */
+        pcpu_base_addr = (void *)pcpu_chunk_addr(static_chunk, 0, 0);
+        return pcpu_unit_size;
+}
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 11a929872ebd..af58324c361a 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -24,6 +24,7 @@
 #include <linux/radix-tree.h>
 #include <linux/rcupdate.h>
 #include <linux/bootmem.h>
+#include <linux/pfn.h>
 
 #include <asm/atomic.h>
 #include <asm/uaccess.h>
@@ -152,8 +153,8 @@ static int vmap_pud_range(pgd_t *pgd, unsigned long addr,
  *
  * Ie. pte at addr+N*PAGE_SIZE shall point to pfn corresponding to pages[N]
  */
-static int vmap_page_range(unsigned long start, unsigned long end,
-                                pgprot_t prot, struct page **pages)
+static int vmap_page_range_noflush(unsigned long start, unsigned long end,
+                                   pgprot_t prot, struct page **pages)
 {
         pgd_t *pgd;
         unsigned long next;
@@ -169,13 +170,22 @@ static int vmap_page_range(unsigned long start, unsigned long end,
                 if (err)
                         break;
         } while (pgd++, addr = next, addr != end);
-        flush_cache_vmap(start, end);
 
         if (unlikely(err))
                 return err;
         return nr;
 }
 
+static int vmap_page_range(unsigned long start, unsigned long end,
+                           pgprot_t prot, struct page **pages)
+{
+        int ret;
+
+        ret = vmap_page_range_noflush(start, end, prot, pages);
+        flush_cache_vmap(start, end);
+        return ret;
+}
+
 static inline int is_vmalloc_or_module_addr(const void *x)
 {
         /*
@@ -990,6 +1000,32 @@ void *vm_map_ram(struct page **pages, unsigned int count, int node, pgprot_t pro
 }
 EXPORT_SYMBOL(vm_map_ram);
 
+/**
+ * vm_area_register_early - register vmap area early during boot
+ * @vm: vm_struct to register
+ * @align: requested alignment
+ *
+ * This function is used to register kernel vm area before
+ * vmalloc_init() is called.  @vm->size and @vm->flags should contain
+ * proper values on entry and other fields should be zero.  On return,
+ * vm->addr contains the allocated address.
+ *
+ * DO NOT USE THIS FUNCTION UNLESS YOU KNOW WHAT YOU'RE DOING.
+ */
+void __init vm_area_register_early(struct vm_struct *vm, size_t align)
+{
+        static size_t vm_init_off __initdata;
+        unsigned long addr;
+
+        addr = ALIGN(VMALLOC_START + vm_init_off, align);
+        vm_init_off = PFN_ALIGN(addr + vm->size) - VMALLOC_START;
+
+        vm->addr = (void *)addr;
+
+        vm->next = vmlist;
+        vmlist = vm;
+}
+
 void __init vmalloc_init(void)
 {
         struct vmap_area *va;
@@ -1017,6 +1053,58 @@ void __init vmalloc_init(void)
         vmap_initialized = true;
 }
 
+/**
+ * map_kernel_range_noflush - map kernel VM area with the specified pages
+ * @addr: start of the VM area to map
+ * @size: size of the VM area to map
+ * @prot: page protection flags to use
+ * @pages: pages to map
+ *
+ * Map PFN_UP(@size) pages at @addr.  The VM area @addr and @size
+ * specify should have been allocated using get_vm_area() and its
+ * friends.
+ *
+ * NOTE:
+ * This function does NOT do any cache flushing.  The caller is
+ * responsible for calling flush_cache_vmap() on to-be-mapped areas
+ * before calling this function.
+ *
+ * RETURNS:
+ * The number of pages mapped on success, -errno on failure.
+ */
+int map_kernel_range_noflush(unsigned long addr, unsigned long size,
+                             pgprot_t prot, struct page **pages)
+{
+        return vmap_page_range_noflush(addr, addr + size, prot, pages);
+}
+
+/**
+ * unmap_kernel_range_noflush - unmap kernel VM area
+ * @addr: start of the VM area to unmap
+ * @size: size of the VM area to unmap
+ *
+ * Unmap PFN_UP(@size) pages at @addr.  The VM area @addr and @size
+ * specify should have been allocated using get_vm_area() and its
+ * friends.
+ *
+ * NOTE:
+ * This function does NOT do any cache flushing.  The caller is
+ * responsible for calling flush_cache_vunmap() on to-be-mapped areas
+ * before calling this function and flush_tlb_kernel_range() after.
+ */
+void unmap_kernel_range_noflush(unsigned long addr, unsigned long size)
+{
+        vunmap_page_range(addr, addr + size);
+}
+
+/**
+ * unmap_kernel_range - unmap kernel VM area and flush cache and TLB
+ * @addr: start of the VM area to unmap
+ * @size: size of the VM area to unmap
+ *
+ * Similar to unmap_kernel_range_noflush() but flushes vcache before
+ * the unmapping and tlb after.
+ */
 void unmap_kernel_range(unsigned long addr, unsigned long size)
 {
         unsigned long end = addr + size;
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index e9db889d6222..2886d2fb9ab5 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -1,12 +1,16 @@
 #include <linux/skbuff.h>
 #include <linux/netdevice.h>
 #include <linux/if_vlan.h>
+#include <linux/netpoll.h>
 #include "vlan.h"
 
 /* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
 int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
                       u16 vlan_tci, int polling)
 {
+        if (netpoll_rx(skb))
+                return NET_RX_DROP;
+
         if (skb_bond_should_drop(skb))
                 goto drop;
 
@@ -100,6 +104,9 @@ int vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp,
 {
         int err = NET_RX_SUCCESS;
 
+        if (netpoll_receive_skb(skb))
+                return NET_RX_DROP;
+
         switch (vlan_gro_common(napi, grp, vlan_tci, skb)) {
         case -1:
                 return netif_receive_skb(skb);
@@ -126,6 +133,9 @@ int vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp,
         if (!skb)
                 goto out;
 
+        if (netpoll_receive_skb(skb))
+                goto out;
+
         err = NET_RX_SUCCESS;
 
         switch (vlan_gro_common(napi, grp, vlan_tci, skb)) {
diff --git a/net/core/dev.c b/net/core/dev.c
index a17e00662363..72b0d26fd46d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2488,6 +2488,9 @@ static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 
 int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
+        if (netpoll_receive_skb(skb))
+                return NET_RX_DROP;
+
         switch (__napi_gro_receive(napi, skb)) {
         case -1:
                 return netif_receive_skb(skb);
@@ -2558,6 +2561,9 @@ int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info)
         if (!skb)
                 goto out;
 
+        if (netpoll_receive_skb(skb))
+                goto out;
+
         err = NET_RX_SUCCESS;
 
         switch (__napi_gro_receive(napi, skb)) {
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 743f5542d65a..3a3dad801354 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1375,10 +1375,10 @@ EXPORT_SYMBOL_GPL(snmp_fold_field);
 int snmp_mib_init(void *ptr[2], size_t mibsize)
 {
         BUG_ON(ptr == NULL);
-        ptr[0] = __alloc_percpu(mibsize);
+        ptr[0] = __alloc_percpu(mibsize, __alignof__(unsigned long long));
         if (!ptr[0])
                 goto err0;
-        ptr[1] = __alloc_percpu(mibsize);
+        ptr[1] = __alloc_percpu(mibsize, __alignof__(unsigned long long));
         if (!ptr[1])
                 goto err1;
         return 0;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 97f71153584f..bf895401218f 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -3376,7 +3376,7 @@ int __init ip_rt_init(void)
         int rc = 0;
 
 #ifdef CONFIG_NET_CLS_ROUTE
-        ip_rt_acct = __alloc_percpu(256 * sizeof(struct ip_rt_acct));
+        ip_rt_acct = __alloc_percpu(256 * sizeof(struct ip_rt_acct), __alignof__(struct ip_rt_acct));
         if (!ip_rt_acct)
                 panic("IP: failed to allocate ip_rt_acct\n");
 #endif
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a6961d75c7ea..c28976a7e596 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1374,7 +1374,8 @@ static u8 tcp_sacktag_one(struct sk_buff *skb, struct sock *sk,
 
 static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
                            struct tcp_sacktag_state *state,
-                           unsigned int pcount, int shifted, int mss)
+                           unsigned int pcount, int shifted, int mss,
+                           int dup_sack)
 {
         struct tcp_sock *tp = tcp_sk(sk);
         struct sk_buff *prev = tcp_write_queue_prev(sk, skb);
@@ -1410,7 +1411,7 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb,
         }
 
         /* We discard results */
-        tcp_sacktag_one(skb, sk, state, 0, pcount);
+        tcp_sacktag_one(skb, sk, state, dup_sack, pcount);
 
         /* Difference in this won't matter, both ACKed by the same cumul. ACK */
         TCP_SKB_CB(prev)->sacked |= (TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS);
@@ -1561,7 +1562,7 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb,
 
         if (!skb_shift(prev, skb, len))
                 goto fallback;
-        if (!tcp_shifted_skb(sk, skb, state, pcount, len, mss))
+        if (!tcp_shifted_skb(sk, skb, state, pcount, len, mss, dup_sack))
                 goto out;
 
         /* Hole filled allows collapsing with the next as well, this is very
@@ -1580,7 +1581,7 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb,
         len = skb->len;
         if (skb_shift(prev, skb, len)) {
                 pcount += tcp_skb_pcount(skb);
-                tcp_shifted_skb(sk, skb, state, tcp_skb_pcount(skb), len, mss);
+                tcp_shifted_skb(sk, skb, state, tcp_skb_pcount(skb), len, mss, 0);
         }
 
 out:
diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c
index 2747ec7bfb63..4660b088a8ce 100644
--- a/net/ipv4/tcp_scalable.c
+++ b/net/ipv4/tcp_scalable.c
@@ -1,6 +1,6 @@
 /* Tom Kelly's Scalable TCP
  *
- * See htt://www-lce.eng.cam.ac.uk/~ctk21/scalable/
+ * See http://www.deneholme.net/tom/scalable/
  *
  * John Heffner <jheffner@sc.edu>
  */
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 8fe267feb81e..1bcc3431859e 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -258,11 +258,11 @@ unique:
 
         if (twp != NULL) {
                 *twp = tw;
-                NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITRECYCLED);
+                NET_INC_STATS_BH(net, LINUX_MIB_TIMEWAITRECYCLED);
         } else if (tw != NULL) {
                 /* Silly. Should hash-dance instead... */
                 inet_twsk_deschedule(tw, death_row);
-                NET_INC_STATS_BH(twsk_net(tw), LINUX_MIB_TIMEWAITRECYCLED);
+                NET_INC_STATS_BH(net, LINUX_MIB_TIMEWAITRECYCLED);
 
                 inet_twsk_put(tw);
         }
diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
index c323643ffcf9..72dbb6d1a6b3 100644
--- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c
@@ -201,8 +201,9 @@ icmpv6_error(struct net *net, struct sk_buff *skb, unsigned int dataoff,
 
         if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING &&
             nf_ip6_checksum(skb, hooknum, dataoff, IPPROTO_ICMPV6)) {
-                nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL,
-                              "nf_ct_icmpv6: ICMPv6 checksum failed\n");
+                if (LOG_INVALID(net, IPPROTO_ICMPV6))
+                        nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL,
+                                      "nf_ct_icmpv6: ICMPv6 checksum failed ");
                 return -NF_ACCEPT;
         }
 
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index fa49dc7fe100..c712e9fc6bba 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -39,7 +39,7 @@
 #endif
 
 #define NFULNL_NLBUFSIZ_DEFAULT NLMSG_GOODSIZE
-#define NFULNL_TIMEOUT_DEFAULT  HZ      /* every second */
+#define NFULNL_TIMEOUT_DEFAULT  100     /* every second */
 #define NFULNL_QTHRESH_DEFAULT  100     /* 100 packets */
 #define NFULNL_COPY_RANGE_MAX   0xFFFF  /* max packet size is limited by 16-bit struct nfattr nfa_len field */
 
@@ -590,8 +590,10 @@ nfulnl_log_packet(u_int8_t pf,
 
         qthreshold = inst->qthreshold;
         /* per-rule qthreshold overrides per-instance */
-        if (qthreshold > li->u.ulog.qthreshold)
-                qthreshold = li->u.ulog.qthreshold;
+        if (li->u.ulog.qthreshold)
+                if (qthreshold > li->u.ulog.qthreshold)
+                        qthreshold = li->u.ulog.qthreshold;
+
 
         switch (inst->copy_mode) {
         case NFULNL_COPY_META:
diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c
index bfbf521f6ea5..5baccfa5a0de 100644
--- a/net/netfilter/x_tables.c
+++ b/net/netfilter/x_tables.c
@@ -827,59 +827,143 @@ static const struct file_operations xt_table_ops = {
         .release = seq_release_net,
 };
 
-static void *xt_match_seq_start(struct seq_file *seq, loff_t *pos)
+/*
+ * Traverse state for ip{,6}_{tables,matches} for helping crossing
+ * the multi-AF mutexes.
+ */
+struct nf_mttg_trav {
+        struct list_head *head, *curr;
+        uint8_t class, nfproto;
+};
+
+enum {
+        MTTG_TRAV_INIT,
+        MTTG_TRAV_NFP_UNSPEC,
+        MTTG_TRAV_NFP_SPEC,
+        MTTG_TRAV_DONE,
+};
+
+static void *xt_mttg_seq_next(struct seq_file *seq, void *v, loff_t *ppos,
+    bool is_target)
 {
-        struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-        u_int16_t af = (unsigned long)pde->data;
+        static const uint8_t next_class[] = {
+                [MTTG_TRAV_NFP_UNSPEC] = MTTG_TRAV_NFP_SPEC,
+                [MTTG_TRAV_NFP_SPEC]   = MTTG_TRAV_DONE,
+        };
+        struct nf_mttg_trav *trav = seq->private;
+
+        switch (trav->class) {
+        case MTTG_TRAV_INIT:
+                trav->class = MTTG_TRAV_NFP_UNSPEC;
+                mutex_lock(&xt[NFPROTO_UNSPEC].mutex);
+                trav->head = trav->curr = is_target ?
+                        &xt[NFPROTO_UNSPEC].target : &xt[NFPROTO_UNSPEC].match;
+                break;
+        case MTTG_TRAV_NFP_UNSPEC:
+                trav->curr = trav->curr->next;
+                if (trav->curr != trav->head)
+                        break;
+                mutex_unlock(&xt[NFPROTO_UNSPEC].mutex);
+                mutex_lock(&xt[trav->nfproto].mutex);
+                trav->head = trav->curr = is_target ?
+                        &xt[trav->nfproto].target : &xt[trav->nfproto].match;
+                trav->class = next_class[trav->class];
+                break;
+        case MTTG_TRAV_NFP_SPEC:
+                trav->curr = trav->curr->next;
+                if (trav->curr != trav->head)
+                        break;
+                /* fallthru, _stop will unlock */
+        default:
+                return NULL;
+        }
 
-        mutex_lock(&xt[af].mutex);
-        return seq_list_start(&xt[af].match, *pos);
+        if (ppos != NULL)
+                ++*ppos;
+        return trav;
 }
 
-static void *xt_match_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+static void *xt_mttg_seq_start(struct seq_file *seq, loff_t *pos,
+    bool is_target)
 {
-        struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-        u_int16_t af = (unsigned long)pde->data;
+        struct nf_mttg_trav *trav = seq->private;
+        unsigned int j;
 
-        return seq_list_next(v, &xt[af].match, pos);
+        trav->class = MTTG_TRAV_INIT;
+        for (j = 0; j < *pos; ++j)
+                if (xt_mttg_seq_next(seq, NULL, NULL, is_target) == NULL)
+                        return NULL;
+        return trav;
 }
 
-static void xt_match_seq_stop(struct seq_file *seq, void *v)
+static void xt_mttg_seq_stop(struct seq_file *seq, void *v)
 {
-        struct proc_dir_entry *pde = seq->private;
-        u_int16_t af = (unsigned long)pde->data;
+        struct nf_mttg_trav *trav = seq->private;
+
+        switch (trav->class) {
+        case MTTG_TRAV_NFP_UNSPEC:
+                mutex_unlock(&xt[NFPROTO_UNSPEC].mutex);
+                break;
+        case MTTG_TRAV_NFP_SPEC:
+                mutex_unlock(&xt[trav->nfproto].mutex);
+                break;
+        }
+}
 
-        mutex_unlock(&xt[af].mutex);
+static void *xt_match_seq_start(struct seq_file *seq, loff_t *pos)
+{
+        return xt_mttg_seq_start(seq, pos, false);
 }
 
-static int xt_match_seq_show(struct seq_file *seq, void *v)
+static void *xt_match_seq_next(struct seq_file *seq, void *v, loff_t *ppos)
 {
-        struct xt_match *match = list_entry(v, struct xt_match, list);
+        return xt_mttg_seq_next(seq, v, ppos, false);
+}
 
-        if (strlen(match->name))
-                return seq_printf(seq, "%s\n", match->name);
-        else
-                return 0;
+static int xt_match_seq_show(struct seq_file *seq, void *v)
+{
+        const struct nf_mttg_trav *trav = seq->private;
+        const struct xt_match *match;
+
+        switch (trav->class) {
+        case MTTG_TRAV_NFP_UNSPEC:
+        case MTTG_TRAV_NFP_SPEC:
+                if (trav->curr == trav->head)
+                        return 0;
+                match = list_entry(trav->curr, struct xt_match, list);
+                return (*match->name == '\0') ? 0 :
+                       seq_printf(seq, "%s\n", match->name);
+        }
+        return 0;
 }
 
 static const struct seq_operations xt_match_seq_ops = {
         .start  = xt_match_seq_start,
         .next   = xt_match_seq_next,
-        .stop   = xt_match_seq_stop,
+        .stop   = xt_mttg_seq_stop,
         .show   = xt_match_seq_show,
 };
 
 static int xt_match_open(struct inode *inode, struct file *file)
 {
+        struct seq_file *seq;
+        struct nf_mttg_trav *trav;
         int ret;
 
-        ret = seq_open(file, &xt_match_seq_ops);
-        if (!ret) {
-                struct seq_file *seq = file->private_data;
+        trav = kmalloc(sizeof(*trav), GFP_KERNEL);
+        if (trav == NULL)
+                return -ENOMEM;
 
-                seq->private = PDE(inode);
+        ret = seq_open(file, &xt_match_seq_ops);
+        if (ret < 0) {
+                kfree(trav);
+                return ret;
         }
-        return ret;
+
+        seq = file->private_data;
+        seq->private = trav;
+        trav->nfproto = (unsigned long)PDE(inode)->data;
+        return 0;
 }
 
 static const struct file_operations xt_match_ops = {
@@ -887,62 +971,63 @@ static const struct file_operations xt_match_ops = {
         .open    = xt_match_open,
         .read    = seq_read,
         .llseek  = seq_lseek,
-        .release = seq_release,
+        .release = seq_release_private,
 };
 
 static void *xt_target_seq_start(struct seq_file *seq, loff_t *pos)
 {
-        struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-        u_int16_t af = (unsigned long)pde->data;
-
-        mutex_lock(&xt[af].mutex);
-        return seq_list_start(&xt[af].target, *pos);
+        return xt_mttg_seq_start(seq, pos, true);
 }
 
-static void *xt_target_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+static void *xt_target_seq_next(struct seq_file *seq, void *v, loff_t *ppos)
 {
-        struct proc_dir_entry *pde = (struct proc_dir_entry *)seq->private;
-        u_int16_t af = (unsigned long)pde->data;
-
-        return seq_list_next(v, &xt[af].target, pos);
-}
-
-static void xt_target_seq_stop(struct seq_file *seq, void *v)
-{
-        struct proc_dir_entry *pde = seq->private;
-        u_int16_t af = (unsigned long)pde->data;
-
-        mutex_unlock(&xt[af].mutex);
+        return xt_mttg_seq_next(seq, v, ppos, true);
 }
 
 static int xt_target_seq_show(struct seq_file *seq, void *v)
 {
-        struct xt_target *target = list_entry(v, struct xt_target, list);
-
-        if (strlen(target->name))
-                return seq_printf(seq, "%s\n", target->name);
-        else
-                return 0;
+        const struct nf_mttg_trav *trav = seq->private;
+        const struct xt_target *target;
+
+        switch (trav->class) {
+        case MTTG_TRAV_NFP_UNSPEC:
+        case MTTG_TRAV_NFP_SPEC:
+                if (trav->curr == trav->head)
+                        return 0;
+                target = list_entry(trav->curr, struct xt_target, list);
+                return (*target->name == '\0') ? 0 :
+                       seq_printf(seq, "%s\n", target->name);
+        }
+        return 0;
 }
 
 static const struct seq_operations xt_target_seq_ops = {
         .start  = xt_target_seq_start,
         .next   = xt_target_seq_next,
-        .stop   = xt_target_seq_stop,
+        .stop   = xt_mttg_seq_stop,
         .show   = xt_target_seq_show,
 };
 
 static int xt_target_open(struct inode *inode, struct file *file)
 {
+        struct seq_file *seq;
+        struct nf_mttg_trav *trav;
         int ret;
 
-        ret = seq_open(file, &xt_target_seq_ops);
-        if (!ret) {
-                struct seq_file *seq = file->private_data;
+        trav = kmalloc(sizeof(*trav), GFP_KERNEL);
+        if (trav == NULL)
+                return -ENOMEM;
 
-                seq->private = PDE(inode);
+        ret = seq_open(file, &xt_target_seq_ops);
+        if (ret < 0) {
+                kfree(trav);
+                return ret;
         }
-        return ret;
+
+        seq = file->private_data;
+        seq->private = trav;
+        trav->nfproto = (unsigned long)PDE(inode)->data;
+        return 0;
 }
 
 static const struct file_operations xt_target_ops = {
@@ -950,7 +1035,7 @@ static const struct file_operations xt_target_ops = {
         .open    = xt_target_open,
         .read    = seq_read,
         .llseek  = seq_lseek,
-        .release = seq_release,
+        .release = seq_release_private,
 };
 
 #define FORMAT_TABLES   "_tables_names"
diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c
index fe80b614a400..791e030ea903 100644
--- a/net/netfilter/xt_recent.c
+++ b/net/netfilter/xt_recent.c
@@ -542,7 +542,7 @@ recent_mt_proc_write(struct file *file, const char __user *input,
         struct recent_entry *e;
         char buf[sizeof("+b335:1d35:1e55:dead:c0de:1715:5afe:c0de")];
         const char *c = buf;
-        union nf_inet_addr addr;
+        union nf_inet_addr addr = {};
         u_int16_t family;
         bool add, succ;
 
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index f6b4fa97df70..e36e94ab4e10 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -66,11 +66,15 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 {
         struct drr_sched *q = qdisc_priv(sch);
         struct drr_class *cl = (struct drr_class *)*arg;
+        struct nlattr *opt = tca[TCA_OPTIONS];
         struct nlattr *tb[TCA_DRR_MAX + 1];
         u32 quantum;
         int err;
 
-        err = nla_parse_nested(tb, TCA_DRR_MAX, tca[TCA_OPTIONS], drr_policy);
+        if (!opt)
+                return -EINVAL;
+
+        err = nla_parse_nested(tb, TCA_DRR_MAX, opt, drr_policy);
         if (err < 0)
                 return err;
 
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index e06365775bdf..3b949a354470 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -186,3 +186,17 @@ quiet_cmd_gzip = GZIP    $@
 cmd_gzip = gzip -f -9 < $< > $@
 
 
+# Bzip2
+# ---------------------------------------------------------------------------
+
+# Bzip2 does not include size in file... so we have to fake that
+size_append=$(CONFIG_SHELL) $(srctree)/scripts/bin_size
+
+quiet_cmd_bzip2 = BZIP2    $@
+cmd_bzip2 = (bzip2 -9 < $< && $(size_append) $<) > $@ || (rm -f $@ ; false)
+
+# Lzma
+# ---------------------------------------------------------------------------
+
+quiet_cmd_lzma = LZMA    $@
+cmd_lzma = (lzma -9 -c $< && $(size_append) $<) >$@ || (rm -f $@ ; false)
diff --git a/scripts/bin_size b/scripts/bin_size
new file mode 100644
index 000000000000..43e1b360cee6
--- /dev/null
+++ b/scripts/bin_size
@@ -0,0 +1,10 @@
+#!/bin/sh
+
+if [ $# = 0 ] ; then
+   echo Usage: $0 file
+fi
+
+size_dec=`stat -c "%s" $1`
+size_hex_echo_string=`printf "%08x" $size_dec |
+     sed 's/\(..\)\(..\)\(..\)\(..\)/\\\\x\4\\\\x\3\\\\x\2\\\\x\1/g'`
+/bin/echo -ne $size_hex_echo_string
diff --git a/scripts/gen_initramfs_list.sh b/scripts/gen_initramfs_list.sh
index 5f3415f28736..3eea8f15131b 100644
--- a/scripts/gen_initramfs_list.sh
+++ b/scripts/gen_initramfs_list.sh
@@ -5,7 +5,7 @@
 # Released under the terms of the GNU GPL
 #
 # Generate a cpio packed initramfs. It uses gen_init_cpio to generate
-# the cpio archive, and gzip to pack it.
+# the cpio archive, and then compresses it.
 # The script may also be used to generate the inputfile used for gen_init_cpio
 # This script assumes that gen_init_cpio is located in usr/ directory
 
@@ -16,8 +16,8 @@ usage() {
 cat << EOF
 Usage:
 $0 [-o <file>] [-u <uid>] [-g <gid>] {-d | <cpio_source>} ...
-        -o <file>      Create gzipped initramfs file named <file> using
-                       gen_init_cpio and gzip
+        -o <file>      Create compressed initramfs file named <file> using
+                       gen_init_cpio and compressor depending on the extension
         -u <uid>       User ID to map to user ID 0 (root).
                        <uid> is only meaningful if <cpio_source> is a
                        directory.  "squash" forces all files to uid 0.
@@ -225,6 +225,7 @@ cpio_list=
 output="/dev/stdout"
 output_file=""
 is_cpio_compressed=
+compr="gzip -9 -f"
 
 arg="$1"
 case "$arg" in
@@ -233,11 +234,15 @@ case "$arg" in
                 echo "deps_initramfs := \\"
                 shift
                 ;;
-        "-o")   # generate gzipped cpio image named $1
+        "-o")   # generate compressed cpio image named $1
                 shift
                 output_file="$1"
                 cpio_list="$(mktemp ${TMPDIR:-/tmp}/cpiolist.XXXXXX)"
                 output=${cpio_list}
+                echo "$output_file" | grep -q "\.gz$" && compr="gzip -9 -f"
+                echo "$output_file" | grep -q "\.bz2$" && compr="bzip2 -9 -f"
+                echo "$output_file" | grep -q "\.lzma$" && compr="lzma -9 -f"
+                echo "$output_file" | grep -q "\.cpio$" && compr="cat"
                 shift
                 ;;
 esac
@@ -274,7 +279,7 @@ while [ $# -gt 0 ]; do
         esac
 done
 
-# If output_file is set we will generate cpio archive and gzip it
+# If output_file is set we will generate cpio archive and compress it
 # we are carefull to delete tmp files
 if [ ! -z ${output_file} ]; then
         if [ -z ${cpio_file} ]; then
@@ -287,7 +292,8 @@ if [ ! -z ${output_file} ]; then
         if [ "${is_cpio_compressed}" = "compressed" ]; then
                 cat ${cpio_tfile} > ${output_file}
         else
-                cat ${cpio_tfile} | gzip -f -9 - > ${output_file}
+                (cat ${cpio_tfile} | ${compr}  - > ${output_file}) \
+                || (rm -f ${output_file} ; false)
         fi
         [ -z ${cpio_file} ] && rm ${cpio_tfile}
 fi
diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c
index 3f4b26647386..350794ab9b42 100644
--- a/security/selinux/netlabel.c
+++ b/security/selinux/netlabel.c
@@ -386,11 +386,12 @@ int selinux_netlbl_inode_permission(struct inode *inode, int mask)
         if (!S_ISSOCK(inode->i_mode) ||
             ((mask & (MAY_WRITE | MAY_APPEND)) == 0))
                 return 0;
-
         sock = SOCKET_I(inode);
         sk = sock->sk;
+        if (sk == NULL)
+                return 0;
         sksec = sk->sk_security;
-        if (sksec->nlbl_state != NLBL_REQUIRE)
+        if (sksec == NULL || sksec->nlbl_state != NLBL_REQUIRE)
                 return 0;
 
         local_bh_disable();
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index c8d9178f47e5..5e909e0da04b 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -2095,6 +2095,8 @@ static struct snd_pci_quirk probe_mask_list[] __devinitdata = {
         SND_PCI_QUIRK(0x1028, 0x20ac, "Dell Studio Desktop", 0x01),
         /* including bogus ALC268 in slot#2 that conflicts with ALC888 */
         SND_PCI_QUIRK(0x17c0, 0x4085, "Medion MD96630", 0x01),
+        /* conflict of ALC268 in slot#3 (digital I/O); a temporary fix */
+        SND_PCI_QUIRK(0x1179, 0xff00, "Toshiba laptop", 0x03),
         {}
 };
 
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index a680be0d4534..6c26afcb8262 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -10557,6 +10557,7 @@ static struct snd_pci_quirk alc262_cfg_tbl[] = {
         SND_PCI_QUIRK(0x103c, 0x1309, "HP xw4*00", ALC262_HP_BPC),
         SND_PCI_QUIRK(0x103c, 0x130a, "HP xw6*00", ALC262_HP_BPC),
         SND_PCI_QUIRK(0x103c, 0x130b, "HP xw8*00", ALC262_HP_BPC),
+        SND_PCI_QUIRK(0x103c, 0x170b, "HP xw*", ALC262_HP_BPC),
         SND_PCI_QUIRK(0x103c, 0x2800, "HP D7000", ALC262_HP_BPC_D7000_WL),
         SND_PCI_QUIRK(0x103c, 0x2801, "HP D7000", ALC262_HP_BPC_D7000_WF),
         SND_PCI_QUIRK(0x103c, 0x2802, "HP D7000", ALC262_HP_BPC_D7000_WL),
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 8027edf3c8f2..3bc427645da8 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -4989,7 +4989,7 @@ again:
         case STAC_DELL_M4_3:
                 spec->num_dmics = 1;
                 spec->num_smuxes = 0;
-                spec->num_dmuxes = 0;
+                spec->num_dmuxes = 1;
                 break;
         default:
                 spec->num_dmics = STAC92HD71BXX_NUM_DMICS;
diff --git a/usr/Kconfig b/usr/Kconfig
index 86cecb59dd07..43a3a0fe8f29 100644
--- a/usr/Kconfig
+++ b/usr/Kconfig
@@ -44,3 +44,92 @@ config INITRAMFS_ROOT_GID
           owned by group root in the initial ramdisk image.
 
           If you are not sure, leave it set to "0".
+
+config RD_GZIP
+        bool "Initial ramdisk compressed using gzip"
+        default y
+        depends on BLK_DEV_INITRD=y
+        select DECOMPRESS_GZIP
+        help
+          Support loading of a gzip encoded initial ramdisk or cpio buffer.
+          If unsure, say Y.
+
+config RD_BZIP2
+        bool "Initial ramdisk compressed using bzip2"
+        default n
+        depends on BLK_DEV_INITRD=y
+        select DECOMPRESS_BZIP2
+        help
+          Support loading of a bzip2 encoded initial ramdisk or cpio buffer
+          If unsure, say N.
+
+config RD_LZMA
+        bool "Initial ramdisk compressed using lzma"
+        default n
+        depends on BLK_DEV_INITRD=y
+        select DECOMPRESS_LZMA
+        help
+          Support loading of a lzma encoded initial ramdisk or cpio buffer
+          If unsure, say N.
+
+choice
+        prompt "Built-in initramfs compression mode"
+        help
+          This setting is only meaningful if the INITRAMFS_SOURCE is
+          set. It decides by which algorithm the INITRAMFS_SOURCE will
+          be compressed.
+          Several compression algorithms are available, which differ
+          in efficiency, compression and decompression speed.
+          Compression speed is only relevant when building a kernel.
+          Decompression speed is relevant at each boot.
+
+          If you have any problems with bzip2 or lzma compressed
+          initramfs, mail me (Alain Knaff) <alain@knaff.lu>.
+
+          High compression options are mostly useful for users who
+          are low on disk space (embedded systems), but for whom ram
+          size matters less.
+
+          If in doubt, select 'gzip'
+
+config INITRAMFS_COMPRESSION_NONE
+        bool "None"
+        help
+          Do not compress the built-in initramfs at all. This may
+          sound wasteful in space, but, you should be aware that the
+          built-in initramfs will be compressed at a later stage
+          anyways along with the rest of the kernel, on those
+          architectures that support this.
+          However, not compressing the initramfs may lead to slightly
+          higher memory consumption during a short time at boot, while
+          both the cpio image and the unpacked filesystem image will
+          be present in memory simultaneously
+
+config INITRAMFS_COMPRESSION_GZIP
+        bool "Gzip"
+        depends on RD_GZIP
+        help
+          The old and tried gzip compression. Its compression ratio is
+          the poorest among the 3 choices; however its speed (both
+          compression and decompression) is the fastest.
+
+config INITRAMFS_COMPRESSION_BZIP2
+        bool "Bzip2"
+        depends on RD_BZIP2
+        help
+          Its compression ratio and speed is intermediate.
+          Decompression speed is slowest among the three.  The initramfs
+          size is about 10% smaller with bzip2, in comparison to gzip.
+          Bzip2 uses a large amount of memory. For modern kernels you
+          will need at least 8MB RAM or more for booting.
+
+config INITRAMFS_COMPRESSION_LZMA
+        bool "LZMA"
+        depends on RD_LZMA
+        help
+          The most recent compression algorithm.
+          Its ratio is best, decompression speed is between the other
+          two. Compression is slowest.  The initramfs size is about 33%
+          smaller with LZMA in comparison to gzip.
+
+endchoice
diff --git a/usr/Makefile b/usr/Makefile
index 201f27f8cbaf..b84894b3929d 100644
--- a/usr/Makefile
+++ b/usr/Makefile
@@ -6,13 +6,25 @@ klibcdirs:;
 PHONY += klibcdirs
 
 
+# No compression
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_NONE)   =
+
+# Gzip, but no bzip2
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_GZIP)   = .gz
+
+# Bzip2
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_BZIP2)  = .bz2
+
+# Lzma
+suffix_$(CONFIG_INITRAMFS_COMPRESSION_LZMA)   = .lzma
+
 # Generate builtin.o based on initramfs_data.o
-obj-$(CONFIG_BLK_DEV_INITRD) := initramfs_data.o
+obj-$(CONFIG_BLK_DEV_INITRD) := initramfs_data$(suffix_y).o
 
-# initramfs_data.o contains the initramfs_data.cpio.gz image.
+# initramfs_data.o contains the compressed initramfs_data.cpio image.
 # The image is included using .incbin, a dependency which is not
 # tracked automatically.
-$(obj)/initramfs_data.o: $(obj)/initramfs_data.cpio.gz FORCE
+$(obj)/initramfs_data$(suffix_y).o: $(obj)/initramfs_data.cpio$(suffix_y) FORCE
 
 #####
 # Generate the initramfs cpio archive
@@ -25,28 +37,28 @@ ramfs-args  := \
         $(if $(CONFIG_INITRAMFS_ROOT_UID), -u $(CONFIG_INITRAMFS_ROOT_UID)) \
         $(if $(CONFIG_INITRAMFS_ROOT_GID), -g $(CONFIG_INITRAMFS_ROOT_GID))
 
-# .initramfs_data.cpio.gz.d is used to identify all files included
+# .initramfs_data.cpio.d is used to identify all files included
 # in initramfs and to detect if any files are added/removed.
 # Removed files are identified by directory timestamp being updated
 # The dependency list is generated by gen_initramfs.sh -l
-ifneq ($(wildcard $(obj)/.initramfs_data.cpio.gz.d),)
-        include $(obj)/.initramfs_data.cpio.gz.d
+ifneq ($(wildcard $(obj)/.initramfs_data.cpio.d),)
+        include $(obj)/.initramfs_data.cpio.d
 endif
 
 quiet_cmd_initfs = GEN     $@
       cmd_initfs = $(initramfs) -o $@ $(ramfs-args) $(ramfs-input)
 
-targets := initramfs_data.cpio.gz
+targets := initramfs_data.cpio.gz initramfs_data.cpio.bz2 initramfs_data.cpio.lzma initramfs_data.cpio
 # do not try to update files included in initramfs
 $(deps_initramfs): ;
 
 $(deps_initramfs): klibcdirs
-# We rebuild initramfs_data.cpio.gz if:
-# 1) Any included file is newer then initramfs_data.cpio.gz
+# We rebuild initramfs_data.cpio if:
+# 1) Any included file is newer then initramfs_data.cpio
 # 2) There are changes in which files are included (added or deleted)
-# 3) If gen_init_cpio are newer than initramfs_data.cpio.gz
+# 3) If gen_init_cpio are newer than initramfs_data.cpio
 # 4) arguments to gen_initramfs.sh changes
-$(obj)/initramfs_data.cpio.gz: $(obj)/gen_init_cpio $(deps_initramfs) klibcdirs
-        $(Q)$(initramfs) -l $(ramfs-input) > $(obj)/.initramfs_data.cpio.gz.d
+$(obj)/initramfs_data.cpio$(suffix_y): $(obj)/gen_init_cpio $(deps_initramfs) klibcdirs
+        $(Q)$(initramfs) -l $(ramfs-input) > $(obj)/.initramfs_data.cpio.d
         $(call if_changed,initfs)
 
diff --git a/usr/initramfs_data.S b/usr/initramfs_data.S
index c2e1ad424f4a..7c6973d8d829 100644
--- a/usr/initramfs_data.S
+++ b/usr/initramfs_data.S
@@ -26,5 +26,5 @@ SECTIONS
 */
 
 .section .init.ramfs,"a"
-.incbin "usr/initramfs_data.cpio.gz"
+.incbin "usr/initramfs_data.cpio"
 
diff --git a/usr/initramfs_data.bz2.S b/usr/initramfs_data.bz2.S
new file mode 100644
index 000000000000..bc54d090365c
--- /dev/null
+++ b/usr/initramfs_data.bz2.S
@@ -0,0 +1,29 @@
+/*
+  initramfs_data includes the compressed binary that is the
+  filesystem used for early user space.
+  Note: Older versions of "as" (prior to binutils 2.11.90.0.23
+  released on 2001-07-14) dit not support .incbin.
+  If you are forced to use older binutils than that then the
+  following trick can be applied to create the resulting binary:
+
+
+  ld -m elf_i386  --format binary --oformat elf32-i386 -r \
+  -T initramfs_data.scr initramfs_data.cpio.gz -o initramfs_data.o
+   ld -m elf_i386  -r -o built-in.o initramfs_data.o
+
+  initramfs_data.scr looks like this:
+SECTIONS
+{
+       .init.ramfs : { *(.data) }
+}
+
+  The above example is for i386 - the parameters vary from architectures.
+  Eventually look up LDFLAGS_BLOB in an older version of the
+  arch/$(ARCH)/Makefile to see the flags used before .incbin was introduced.
+
+  Using .incbin has the advantage over ld that the correct flags are set
+  in the ELF header, as required by certain architectures.
+*/
+
+.section .init.ramfs,"a"
+.incbin "usr/initramfs_data.cpio.bz2"
diff --git a/usr/initramfs_data.gz.S b/usr/initramfs_data.gz.S
new file mode 100644
index 000000000000..890c8dd1d6bd
--- /dev/null
+++ b/usr/initramfs_data.gz.S
@@ -0,0 +1,29 @@
+/*
+  initramfs_data includes the compressed binary that is the
+  filesystem used for early user space.
+  Note: Older versions of "as" (prior to binutils 2.11.90.0.23
+  released on 2001-07-14) dit not support .incbin.
+  If you are forced to use older binutils than that then the
+  following trick can be applied to create the resulting binary:
+
+
+  ld -m elf_i386  --format binary --oformat elf32-i386 -r \
+  -T initramfs_data.scr initramfs_data.cpio.gz -o initramfs_data.o
+   ld -m elf_i386  -r -o built-in.o initramfs_data.o
+
+  initramfs_data.scr looks like this:
+SECTIONS
+{
+       .init.ramfs : { *(.data) }
+}
+
+  The above example is for i386 - the parameters vary from architectures.
+  Eventually look up LDFLAGS_BLOB in an older version of the
+  arch/$(ARCH)/Makefile to see the flags used before .incbin was introduced.
+
+  Using .incbin has the advantage over ld that the correct flags are set
+  in the ELF header, as required by certain architectures.
+*/
+
+.section .init.ramfs,"a"
+.incbin "usr/initramfs_data.cpio.gz"
diff --git a/usr/initramfs_data.lzma.S b/usr/initramfs_data.lzma.S
new file mode 100644
index 000000000000..e11469e48562
--- /dev/null
+++ b/usr/initramfs_data.lzma.S
@@ -0,0 +1,29 @@
+/*
+  initramfs_data includes the compressed binary that is the
+  filesystem used for early user space.
+  Note: Older versions of "as" (prior to binutils 2.11.90.0.23
+  released on 2001-07-14) dit not support .incbin.
+  If you are forced to use older binutils than that then the
+  following trick can be applied to create the resulting binary:
+
+
+  ld -m elf_i386  --format binary --oformat elf32-i386 -r \
+  -T initramfs_data.scr initramfs_data.cpio.gz -o initramfs_data.o
+   ld -m elf_i386  -r -o built-in.o initramfs_data.o
+
+  initramfs_data.scr looks like this:
+SECTIONS
+{
+       .init.ramfs : { *(.data) }
+}
+
+  The above example is for i386 - the parameters vary from architectures.
+  Eventually look up LDFLAGS_BLOB in an older version of the
+  arch/$(ARCH)/Makefile to see the flags used before .incbin was introduced.
+
+  Using .incbin has the advantage over ld that the correct flags are set
+  in the ELF header, as required by certain architectures.
+*/
+
+.section .init.ramfs,"a"
+.incbin "usr/initramfs_data.cpio.lzma"