Grsec/PaX: 2.9-{2.6.32.59,3.2.13,3.3.0}-201203251921

1 files changed, 86204 insertions, 0 deletions

diff --git a/3.2.13/4420_grsecurity-2.9-3.2.13-201203251921.patch b/3.2.13/4420_grsecurity-2.9-3.2.13-201203251921.patch
new file mode 100644
index 0000000..373d86b
--- /dev/null
+++ b/3.2.13/4420_grsecurity-2.9-3.2.13-201203251921.patch
@@ -0,0 +1,86204 @@
+diff --git a/Documentation/dontdiff b/Documentation/dontdiff
+index dfa6fc6..df93044 100644
+--- a/Documentation/dontdiff
++++ b/Documentation/dontdiff
+@@ -2,9 +2,11 @@
+ *.aux
+ *.bin
+ *.bz2
++*.c.[012]*.*
+ *.cis
+ *.cpio
+ *.csp
++*.dbg
+ *.dsp
+ *.dvi
+ *.elf
+@@ -14,6 +16,7 @@
+ *.gcov
+ *.gen.S
+ *.gif
++*.gmo
+ *.grep
+ *.grp
+ *.gz
+@@ -48,9 +51,11 @@
+ *.tab.h
+ *.tex
+ *.ver
++*.vim
+ *.xml
+ *.xz
+ *_MODULES
++*_reg_safe.h
+ *_vga16.c
+ *~
+ \#*#
+@@ -70,6 +75,7 @@ Kerntypes
+ Module.markers
+ Module.symvers
+ PENDING
++PERF*
+ SCCS
+ System.map*
+ TAGS
+@@ -93,19 +99,24 @@ bounds.h
+ bsetup
+ btfixupprep
+ build
++builtin-policy.h
+ bvmlinux
+ bzImage*
+ capability_names.h
+ capflags.c
+ classlist.h*
++clut_vga16.c
++common-cmds.h
+ comp*.log
+ compile.h*
+ conf
+ config
+ config-*
+ config_data.h*
++config.c
+ config.mak
+ config.mak.autogen
++config.tmp
+ conmakehash
+ consolemap_deftbl.c*
+ cpustr.h
+@@ -116,9 +127,11 @@ devlist.h*
+ dnotify_test
+ docproc
+ dslm
++dtc-lexer.lex.c
+ elf2ecoff
+ elfconfig.h*
+ evergreen_reg_safe.h
++exception_policy.conf
+ fixdep
+ flask.h
+ fore200e_mkfirm
+@@ -126,12 +139,15 @@ fore200e_pca_fw.c*
+ gconf
+ gconf.glade.h
+ gen-devlist
++gen-kdb_cmds.c
+ gen_crc32table
+ gen_init_cpio
+ generated
+ genheaders
+ genksyms
+ *_gray256.c
++hash
++hid-example
+ hpet_example
+ hugepage-mmap
+ hugepage-shm
+@@ -146,7 +162,7 @@ int32.c
+ int4.c
+ int8.c
+ kallsyms
+-kconfig
++kern_constants.h
+ keywords.c
+ ksym.c*
+ ksym.h*
+@@ -154,7 +170,7 @@ kxgettext
+ lkc_defs.h
+ lex.c
+ lex.*.c
+-linux
++lib1funcs.S
+ logo_*.c
+ logo_*_clut224.c
+ logo_*_mono.c
+@@ -166,14 +182,15 @@ machtypes.h
+ map
+ map_hugetlb
+ maui_boot.h
+-media
+ mconf
++mdp
+ miboot*
+ mk_elfconfig
+ mkboot
+ mkbugboot
+ mkcpustr
+ mkdep
++mkpiggy
+ mkprep
+ mkregtable
+ mktables
+@@ -209,6 +226,7 @@ r300_reg_safe.h
+ r420_reg_safe.h
+ r600_reg_safe.h
+ recordmcount
++regdb.c
+ relocs
+ rlim_names.h
+ rn50_reg_safe.h
+@@ -219,6 +237,7 @@ setup
+ setup.bin
+ setup.elf
+ sImage
++slabinfo
+ sm_tbl*
+ split-include
+ syscalltab.h
+@@ -229,6 +248,7 @@ tftpboot.img
+ timeconst.h
+ times.h*
+ trix_boot.h
++user_constants.h
+ utsrelease.h*
+ vdso-syms.lds
+ vdso.lds
+@@ -246,7 +266,9 @@ vmlinux
+ vmlinux-*
+ vmlinux.aout
+ vmlinux.bin.all
++vmlinux.bin.bz2
+ vmlinux.lds
++vmlinux.relocs
+ vmlinuz
+ voffset.h
+ vsyscall.lds
+@@ -254,9 +276,11 @@ vsyscall_32.lds
+ wanxlfw.inc
+ uImage
+ unifdef
++utsrelease.h
+ wakeup.bin
+ wakeup.elf
+ wakeup.lds
+ zImage*
+ zconf.hash.c
++zconf.lex.c
+ zoffset.h
+diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
+index 81c287f..d456d02 100644
+--- a/Documentation/kernel-parameters.txt
++++ b/Documentation/kernel-parameters.txt
+@@ -1935,6 +1935,13 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
+ 			the specified number of seconds.  This is to be used if
+ 			your oopses keep scrolling off the screen.
+ 
++	pax_nouderef	[X86] disables UDEREF.  Most likely needed under certain
++			virtualization environments that don't cope well with the
++			expand down segment used by UDEREF on X86-32 or the frequent
++			page table updates on X86-64.
++
++	pax_softmode=	0/1 to disable/enable PaX softmode on boot already.
++
+ 	pcbit=		[HW,ISDN]
+ 
+ 	pcd.		[PARIDE]
+diff --git a/Makefile b/Makefile
+index 172e041..4aede2e 100644
+--- a/Makefile
++++ b/Makefile
+@@ -245,8 +245,9 @@ CONFIG_SHELL := $(shell if [ -x "$$BASH" ]; then echo $$BASH; \
+ 
+ HOSTCC       = gcc
+ HOSTCXX      = g++
+-HOSTCFLAGS   = -Wall -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer
+-HOSTCXXFLAGS = -O2
++HOSTCFLAGS   = -Wall -W -Wmissing-prototypes -Wstrict-prototypes -Wno-unused-parameter -Wno-missing-field-initializers -O2 -fomit-frame-pointer -fno-delete-null-pointer-checks
++HOSTCLFAGS  += $(call cc-option, -Wno-empty-body)
++HOSTCXXFLAGS = -O2 -Wall -W -fno-delete-null-pointer-checks
+ 
+ # Decide whether to build built-in, modular, or both.
+ # Normally, just do built-in.
+@@ -407,8 +408,8 @@ export RCS_TAR_IGNORE := --exclude SCCS --exclude BitKeeper --exclude .svn --exc
+ # Rules shared between *config targets and build targets
+ 
+ # Basic helpers built in scripts/
+-PHONY += scripts_basic
+-scripts_basic:
++PHONY += scripts_basic gcc-plugins
++scripts_basic: gcc-plugins
+ 	$(Q)$(MAKE) $(build)=scripts/basic
+ 	$(Q)rm -f .tmp_quiet_recordmcount
+ 
+@@ -564,6 +565,53 @@ else
+ KBUILD_CFLAGS	+= -O2
+ endif
+ 
++ifndef DISABLE_PAX_PLUGINS
++ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-plugin.sh "$(HOSTCC)" "$(CC)"), y)
++ifndef DISABLE_PAX_CONSTIFY_PLUGIN
++CONSTIFY_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/constify_plugin.so -DCONSTIFY_PLUGIN
++endif
++ifdef CONFIG_PAX_MEMORY_STACKLEAK
++STACKLEAK_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/stackleak_plugin.so -DSTACKLEAK_PLUGIN
++STACKLEAK_PLUGIN_CFLAGS += -fplugin-arg-stackleak_plugin-track-lowest-sp=100
++endif
++ifdef CONFIG_KALLOCSTAT_PLUGIN
++KALLOCSTAT_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/kallocstat_plugin.so
++endif
++ifdef CONFIG_PAX_KERNEXEC_PLUGIN
++KERNEXEC_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/kernexec_plugin.so
++KERNEXEC_PLUGIN_CFLAGS += -fplugin-arg-kernexec_plugin-method=$(CONFIG_PAX_KERNEXEC_PLUGIN_METHOD) -DKERNEXEC_PLUGIN
++KERNEXEC_PLUGIN_AFLAGS := -DKERNEXEC_PLUGIN
++endif
++ifdef CONFIG_CHECKER_PLUGIN
++ifeq ($(call cc-ifversion, -ge, 0406, y), y)
++CHECKER_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/checker_plugin.so -DCHECKER_PLUGIN
++endif
++endif
++COLORIZE_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/colorize_plugin.so
++ifdef CONFIG_PAX_SIZE_OVERFLOW
++SIZE_OVERFLOW_PLUGIN_CFLAGS := -fplugin=$(objtree)/tools/gcc/size_overflow_plugin.so -DSIZE_OVERFLOW_PLUGIN
++endif
++GCC_PLUGINS_CFLAGS := $(CONSTIFY_PLUGIN_CFLAGS) $(STACKLEAK_PLUGIN_CFLAGS) $(KALLOCSTAT_PLUGIN_CFLAGS)
++GCC_PLUGINS_CFLAGS += $(KERNEXEC_PLUGIN_CFLAGS) $(CHECKER_PLUGIN_CFLAGS) $(COLORIZE_PLUGIN_CFLAGS) $(SIZE_OVERFLOW_PLUGIN_CFLAGS)
++GCC_PLUGINS_AFLAGS := $(KERNEXEC_PLUGIN_AFLAGS)
++export CONSTIFY_PLUGIN STACKLEAK_PLUGIN KERNEXEC_PLUGIN CHECKER_PLUGIN SIZE_OVERFLOW_PLUGIN
++ifeq ($(KBUILD_EXTMOD),)
++gcc-plugins:
++	$(Q)$(MAKE) $(build)=tools/gcc
++else
++gcc-plugins: ;
++endif
++else
++gcc-plugins:
++ifeq ($(call cc-ifversion, -ge, 0405, y), y)
++	$(error Your gcc installation does not support plugins.  If the necessary headers for plugin support are missing, they should be installed.  On Debian, apt-get install gcc-<ver>-plugin-dev.  If you choose to ignore this error and lessen the improvements provided by this patch, re-run make with the DISABLE_PAX_PLUGINS=y argument.))
++else
++	$(Q)echo "warning, your gcc version does not support plugins, you should upgrade it to gcc 4.5 at least"
++endif
++	$(Q)echo "PAX_MEMORY_STACKLEAK and other features will be less secure"
++endif
++endif
++
+ include $(srctree)/arch/$(SRCARCH)/Makefile
+ 
+ ifneq ($(CONFIG_FRAME_WARN),0)
+@@ -708,7 +756,7 @@ export mod_strip_cmd
+ 
+ 
+ ifeq ($(KBUILD_EXTMOD),)
+-core-y		+= kernel/ mm/ fs/ ipc/ security/ crypto/ block/
++core-y		+= kernel/ mm/ fs/ ipc/ security/ crypto/ block/ grsecurity/
+ 
+ vmlinux-dirs	:= $(patsubst %/,%,$(filter %/, $(init-y) $(init-m) \
+ 		     $(core-y) $(core-m) $(drivers-y) $(drivers-m) \
+@@ -932,6 +980,8 @@ vmlinux.o: $(modpost-init) $(vmlinux-main) FORCE
+ 
+ # The actual objects are generated when descending, 
+ # make sure no implicit rule kicks in
++$(sort $(vmlinux-init) $(vmlinux-main)) $(vmlinux-lds): KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++$(sort $(vmlinux-init) $(vmlinux-main)) $(vmlinux-lds): KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
+ $(sort $(vmlinux-init) $(vmlinux-main)) $(vmlinux-lds): $(vmlinux-dirs) ;
+ 
+ # Handle descending into subdirectories listed in $(vmlinux-dirs)
+@@ -941,7 +991,7 @@ $(sort $(vmlinux-init) $(vmlinux-main)) $(vmlinux-lds): $(vmlinux-dirs) ;
+ # Error messages still appears in the original language
+ 
+ PHONY += $(vmlinux-dirs)
+-$(vmlinux-dirs): prepare scripts
++$(vmlinux-dirs): gcc-plugins prepare scripts
+ 	$(Q)$(MAKE) $(build)=$@
+ 
+ # Store (new) KERNELRELASE string in include/config/kernel.release
+@@ -985,6 +1035,7 @@ prepare0: archprepare FORCE
+ 	$(Q)$(MAKE) $(build)=.
+ 
+ # All the preparing..
++prepare: KBUILD_CFLAGS := $(filter-out $(GCC_PLUGINS_CFLAGS),$(KBUILD_CFLAGS))
+ prepare: prepare0
+ 
+ # Generate some files
+@@ -1086,6 +1137,8 @@ all: modules
+ #	using awk while concatenating to the final file.
+ 
+ PHONY += modules
++modules: KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++modules: KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
+ modules: $(vmlinux-dirs) $(if $(KBUILD_BUILTIN),vmlinux) modules.builtin
+ 	$(Q)$(AWK) '!x[$$0]++' $(vmlinux-dirs:%=$(objtree)/%/modules.order) > $(objtree)/modules.order
+ 	@$(kecho) '  Building modules, stage 2.';
+@@ -1101,7 +1154,7 @@ modules.builtin: $(vmlinux-dirs:%=%/modules.builtin)
+ 
+ # Target to prepare building external modules
+ PHONY += modules_prepare
+-modules_prepare: prepare scripts
++modules_prepare: gcc-plugins prepare scripts
+ 
+ # Target to install modules
+ PHONY += modules_install
+@@ -1198,6 +1251,7 @@ distclean: mrproper
+ 		\( -name '*.orig' -o -name '*.rej' -o -name '*~' \
+ 		-o -name '*.bak' -o -name '#*#' -o -name '.*.orig' \
+ 		-o -name '.*.rej' \
++		-o -name '.*.rej' -o -name '*.so' \
+ 		-o -name '*%' -o -name '.*.cmd' -o -name 'core' \) \
+ 		-type f -print | xargs rm -f
+ 
+@@ -1358,6 +1412,8 @@ PHONY += $(module-dirs) modules
+ $(module-dirs): crmodverdir $(objtree)/Module.symvers
+ 	$(Q)$(MAKE) $(build)=$(patsubst _module_%,%,$@)
+ 
++modules: KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++modules: KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
+ modules: $(module-dirs)
+ 	@$(kecho) '  Building modules, stage 2.';
+ 	$(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost
+@@ -1484,17 +1540,21 @@ else
+         target-dir = $(if $(KBUILD_EXTMOD),$(dir $<),$(dir $@))
+ endif
+ 
+-%.s: %.c prepare scripts FORCE
++%.s: KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++%.s: KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
++%.s: %.c gcc-plugins prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+ %.i: %.c prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+-%.o: %.c prepare scripts FORCE
++%.o: KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++%.o: KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
++%.o: %.c gcc-plugins prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+ %.lst: %.c prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+-%.s: %.S prepare scripts FORCE
++%.s: %.S gcc-plugins prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+-%.o: %.S prepare scripts FORCE
++%.o: %.S gcc-plugins prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+ %.symtypes: %.c prepare scripts FORCE
+ 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+@@ -1504,11 +1564,15 @@ endif
+ 	$(cmd_crmodverdir)
+ 	$(Q)$(MAKE) KBUILD_MODULES=$(if $(CONFIG_MODULES),1) \
+ 	$(build)=$(build-dir)
+-%/: prepare scripts FORCE
++%/: KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++%/: KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
++%/: gcc-plugins prepare scripts FORCE
+ 	$(cmd_crmodverdir)
+ 	$(Q)$(MAKE) KBUILD_MODULES=$(if $(CONFIG_MODULES),1) \
+ 	$(build)=$(build-dir)
+-%.ko: prepare scripts FORCE
++%.ko: KBUILD_CFLAGS += $(GCC_PLUGINS_CFLAGS)
++%.ko: KBUILD_AFLAGS += $(GCC_PLUGINS_AFLAGS)
++%.ko: gcc-plugins prepare scripts FORCE
+ 	$(cmd_crmodverdir)
+ 	$(Q)$(MAKE) KBUILD_MODULES=$(if $(CONFIG_MODULES),1)   \
+ 	$(build)=$(build-dir) $(@:.ko=.o)
+diff --git a/arch/alpha/include/asm/atomic.h b/arch/alpha/include/asm/atomic.h
+index 640f909..48b6597 100644
+--- a/arch/alpha/include/asm/atomic.h
++++ b/arch/alpha/include/asm/atomic.h
+@@ -250,6 +250,16 @@ static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
+ #define atomic_dec(v) atomic_sub(1,(v))
+ #define atomic64_dec(v) atomic64_sub(1,(v))
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ #define smp_mb__before_atomic_dec()	smp_mb()
+ #define smp_mb__after_atomic_dec()	smp_mb()
+ #define smp_mb__before_atomic_inc()	smp_mb()
+diff --git a/arch/alpha/include/asm/cache.h b/arch/alpha/include/asm/cache.h
+index ad368a9..fbe0f25 100644
+--- a/arch/alpha/include/asm/cache.h
++++ b/arch/alpha/include/asm/cache.h
+@@ -4,19 +4,19 @@
+ #ifndef __ARCH_ALPHA_CACHE_H
+ #define __ARCH_ALPHA_CACHE_H
+ 
++#include <linux/const.h>
+ 
+ /* Bytes per L1 (data) cache line. */
+ #if defined(CONFIG_ALPHA_GENERIC) || defined(CONFIG_ALPHA_EV6)
+-# define L1_CACHE_BYTES     64
+ # define L1_CACHE_SHIFT     6
+ #else
+ /* Both EV4 and EV5 are write-through, read-allocate,
+    direct-mapped, physical.
+ */
+-# define L1_CACHE_BYTES     32
+ # define L1_CACHE_SHIFT     5
+ #endif
+ 
++#define L1_CACHE_BYTES     (_AC(1,UL) << L1_CACHE_SHIFT)
+ #define SMP_CACHE_BYTES    L1_CACHE_BYTES
+ 
+ #endif
+diff --git a/arch/alpha/include/asm/elf.h b/arch/alpha/include/asm/elf.h
+index da5449e..7418343 100644
+--- a/arch/alpha/include/asm/elf.h
++++ b/arch/alpha/include/asm/elf.h
+@@ -90,6 +90,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+ 
+ #define ELF_ET_DYN_BASE		(TASK_UNMAPPED_BASE + 0x1000000)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(current->personality & ADDR_LIMIT_32BIT ? 0x10000 : 0x120000000UL)
++
++#define PAX_DELTA_MMAP_LEN	(current->personality & ADDR_LIMIT_32BIT ? 14 : 28)
++#define PAX_DELTA_STACK_LEN	(current->personality & ADDR_LIMIT_32BIT ? 14 : 19)
++#endif
++
+ /* $0 is set by ld.so to a pointer to a function which might be 
+    registered using atexit.  This provides a mean for the dynamic
+    linker to call DT_FINI functions for shared libraries that have
+diff --git a/arch/alpha/include/asm/pgtable.h b/arch/alpha/include/asm/pgtable.h
+index de98a73..bd4f1f8 100644
+--- a/arch/alpha/include/asm/pgtable.h
++++ b/arch/alpha/include/asm/pgtable.h
+@@ -101,6 +101,17 @@ struct vm_area_struct;
+ #define PAGE_SHARED	__pgprot(_PAGE_VALID | __ACCESS_BITS)
+ #define PAGE_COPY	__pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW)
+ #define PAGE_READONLY	__pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW)
++
++#ifdef CONFIG_PAX_PAGEEXEC
++# define PAGE_SHARED_NOEXEC	__pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOE)
++# define PAGE_COPY_NOEXEC	__pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE)
++# define PAGE_READONLY_NOEXEC	__pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE)
++#else
++# define PAGE_SHARED_NOEXEC	PAGE_SHARED
++# define PAGE_COPY_NOEXEC	PAGE_COPY
++# define PAGE_READONLY_NOEXEC	PAGE_READONLY
++#endif
++
+ #define PAGE_KERNEL	__pgprot(_PAGE_VALID | _PAGE_ASM | _PAGE_KRE | _PAGE_KWE)
+ 
+ #define _PAGE_NORMAL(x) __pgprot(_PAGE_VALID | __ACCESS_BITS | (x))
+diff --git a/arch/alpha/kernel/module.c b/arch/alpha/kernel/module.c
+index 2fd00b7..cfd5069 100644
+--- a/arch/alpha/kernel/module.c
++++ b/arch/alpha/kernel/module.c
+@@ -160,7 +160,7 @@ apply_relocate_add(Elf64_Shdr *sechdrs, const char *strtab,
+ 
+ 	/* The small sections were sorted to the end of the segment.
+ 	   The following should definitely cover them.  */
+-	gp = (u64)me->module_core + me->core_size - 0x8000;
++	gp = (u64)me->module_core_rw + me->core_size_rw - 0x8000;
+ 	got = sechdrs[me->arch.gotsecindex].sh_addr;
+ 
+ 	for (i = 0; i < n; i++) {
+diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c
+index 01e8715..be0e80f 100644
+--- a/arch/alpha/kernel/osf_sys.c
++++ b/arch/alpha/kernel/osf_sys.c
+@@ -1147,7 +1147,7 @@ arch_get_unmapped_area_1(unsigned long addr, unsigned long len,
+ 		/* At this point:  (!vma || addr < vma->vm_end). */
+ 		if (limit - len < addr)
+ 			return -ENOMEM;
+-		if (!vma || addr + len <= vma->vm_start)
++		if (check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 		addr = vma->vm_end;
+ 		vma = vma->vm_next;
+@@ -1183,6 +1183,10 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 	   merely specific addresses, but regions of memory -- perhaps
+ 	   this feature should be incorporated into all ports?  */
+ 
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(current->mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
+ 	if (addr) {
+ 		addr = arch_get_unmapped_area_1 (PAGE_ALIGN(addr), len, limit);
+ 		if (addr != (unsigned long) -ENOMEM)
+@@ -1190,8 +1194,8 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 	}
+ 
+ 	/* Next, try allocating at TASK_UNMAPPED_BASE.  */
+-	addr = arch_get_unmapped_area_1 (PAGE_ALIGN(TASK_UNMAPPED_BASE),
+-					 len, limit);
++	addr = arch_get_unmapped_area_1 (PAGE_ALIGN(current->mm->mmap_base), len, limit);
++
+ 	if (addr != (unsigned long) -ENOMEM)
+ 		return addr;
+ 
+diff --git a/arch/alpha/mm/fault.c b/arch/alpha/mm/fault.c
+index fadd5f8..904e73a 100644
+--- a/arch/alpha/mm/fault.c
++++ b/arch/alpha/mm/fault.c
+@@ -54,6 +54,124 @@ __load_new_mm_context(struct mm_struct *next_mm)
+ 	__reload_thread(pcb);
+ }
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++/*
++ * PaX: decide what to do with offenders (regs->pc = fault address)
++ *
++ * returns 1 when task should be killed
++ *         2 when patched PLT trampoline was detected
++ *         3 when unpatched PLT trampoline was detected
++ */
++static int pax_handle_fetch_fault(struct pt_regs *regs)
++{
++
++#ifdef CONFIG_PAX_EMUPLT
++	int err;
++
++	do { /* PaX: patched PLT emulation #1 */
++		unsigned int ldah, ldq, jmp;
++
++		err = get_user(ldah, (unsigned int *)regs->pc);
++		err |= get_user(ldq, (unsigned int *)(regs->pc+4));
++		err |= get_user(jmp, (unsigned int *)(regs->pc+8));
++
++		if (err)
++			break;
++
++		if ((ldah & 0xFFFF0000U) == 0x277B0000U &&
++		    (ldq & 0xFFFF0000U) == 0xA77B0000U &&
++		    jmp == 0x6BFB0000U)
++		{
++			unsigned long r27, addr;
++			unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16;
++			unsigned long addrl = ldq | 0xFFFFFFFFFFFF0000UL;
++
++			addr = regs->r27 + ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL);
++			err = get_user(r27, (unsigned long *)addr);
++			if (err)
++				break;
++
++			regs->r27 = r27;
++			regs->pc = r27;
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: patched PLT emulation #2 */
++		unsigned int ldah, lda, br;
++
++		err = get_user(ldah, (unsigned int *)regs->pc);
++		err |= get_user(lda, (unsigned int *)(regs->pc+4));
++		err |= get_user(br, (unsigned int *)(regs->pc+8));
++
++		if (err)
++			break;
++
++		if ((ldah & 0xFFFF0000U) == 0x277B0000U &&
++		    (lda & 0xFFFF0000U) == 0xA77B0000U &&
++		    (br & 0xFFE00000U) == 0xC3E00000U)
++		{
++			unsigned long addr = br | 0xFFFFFFFFFFE00000UL;
++			unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16;
++			unsigned long addrl = lda | 0xFFFFFFFFFFFF0000UL;
++
++			regs->r27 += ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL);
++			regs->pc += 12 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2);
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: unpatched PLT emulation */
++		unsigned int br;
++
++		err = get_user(br, (unsigned int *)regs->pc);
++
++		if (!err && (br & 0xFFE00000U) == 0xC3800000U) {
++			unsigned int br2, ldq, nop, jmp;
++			unsigned long addr = br | 0xFFFFFFFFFFE00000UL, resolver;
++
++			addr = regs->pc + 4 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2);
++			err = get_user(br2, (unsigned int *)addr);
++			err |= get_user(ldq, (unsigned int *)(addr+4));
++			err |= get_user(nop, (unsigned int *)(addr+8));
++			err |= get_user(jmp, (unsigned int *)(addr+12));
++			err |= get_user(resolver, (unsigned long *)(addr+16));
++
++			if (err)
++				break;
++
++			if (br2 == 0xC3600000U &&
++			    ldq == 0xA77B000CU &&
++			    nop == 0x47FF041FU &&
++			    jmp == 0x6B7B0000U)
++			{
++				regs->r28 = regs->pc+4;
++				regs->r27 = addr+16;
++				regs->pc = resolver;
++				return 3;
++			}
++		}
++	} while (0);
++#endif
++
++	return 1;
++}
++
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 5; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
+ 
+ /*
+  * This routine handles page faults.  It determines the address,
+@@ -131,8 +249,29 @@ do_page_fault(unsigned long address, unsigned long mmcsr,
+  good_area:
+ 	si_code = SEGV_ACCERR;
+ 	if (cause < 0) {
+-		if (!(vma->vm_flags & VM_EXEC))
++		if (!(vma->vm_flags & VM_EXEC)) {
++
++#ifdef CONFIG_PAX_PAGEEXEC
++			if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->pc)
++				goto bad_area;
++
++			up_read(&mm->mmap_sem);
++			switch (pax_handle_fetch_fault(regs)) {
++
++#ifdef CONFIG_PAX_EMUPLT
++			case 2:
++			case 3:
++				return;
++#endif
++
++			}
++			pax_report_fault(regs, (void *)regs->pc, (void *)rdusp());
++			do_group_exit(SIGKILL);
++#else
+ 			goto bad_area;
++#endif
++
++		}
+ 	} else if (!cause) {
+ 		/* Allow reads even for write-only mappings */
+ 		if (!(vma->vm_flags & (VM_READ | VM_WRITE)))
+diff --git a/arch/arm/include/asm/atomic.h b/arch/arm/include/asm/atomic.h
+index 86976d0..8e07f84 100644
+--- a/arch/arm/include/asm/atomic.h
++++ b/arch/arm/include/asm/atomic.h
+@@ -15,6 +15,10 @@
+ #include <linux/types.h>
+ #include <asm/system.h>
+ 
++#ifdef CONFIG_GENERIC_ATOMIC64
++#include <asm-generic/atomic64.h>
++#endif
++
+ #define ATOMIC_INIT(i)	{ (i) }
+ 
+ #ifdef __KERNEL__
+@@ -25,7 +29,15 @@
+  * atomic_set() is the clrex or dummy strex done on every exception return.
+  */
+ #define atomic_read(v)	(*(volatile int *)&(v)->counter)
++static inline int atomic_read_unchecked(const atomic_unchecked_t *v)
++{
++	return v->counter;
++}
+ #define atomic_set(v,i)	(((v)->counter) = (i))
++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i)
++{
++	v->counter = i;
++}
+ 
+ #if __LINUX_ARM_ARCH__ >= 6
+ 
+@@ -40,6 +52,35 @@ static inline void atomic_add(int i, atomic_t *v)
+ 	int result;
+ 
+ 	__asm__ __volatile__("@ atomic_add\n"
++"1:	ldrex	%1, [%3]\n"
++"	adds	%0, %1, %4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
++"	strex	%1, %0, [%3]\n"
++"	teq	%1, #0\n"
++"	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
++	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
++	: "r" (&v->counter), "Ir" (i)
++	: "cc");
++}
++
++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v)
++{
++	unsigned long tmp;
++	int result;
++
++	__asm__ __volatile__("@ atomic_add_unchecked\n"
+ "1:	ldrex	%0, [%3]\n"
+ "	add	%0, %0, %4\n"
+ "	strex	%1, %0, [%3]\n"
+@@ -58,6 +99,42 @@ static inline int atomic_add_return(int i, atomic_t *v)
+ 	smp_mb();
+ 
+ 	__asm__ __volatile__("@ atomic_add_return\n"
++"1:	ldrex	%1, [%3]\n"
++"	adds	%0, %1, %4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"	mov	%0, %1\n"
++"2:	bkpt 0xf103\n"
++"3:\n"
++#endif
++
++"	strex	%1, %0, [%3]\n"
++"	teq	%1, #0\n"
++"	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
++	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
++	: "r" (&v->counter), "Ir" (i)
++	: "cc");
++
++	smp_mb();
++
++	return result;
++}
++
++static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v)
++{
++	unsigned long tmp;
++	int result;
++
++	smp_mb();
++
++	__asm__ __volatile__("@ atomic_add_return_unchecked\n"
+ "1:	ldrex	%0, [%3]\n"
+ "	add	%0, %0, %4\n"
+ "	strex	%1, %0, [%3]\n"
+@@ -78,6 +155,35 @@ static inline void atomic_sub(int i, atomic_t *v)
+ 	int result;
+ 
+ 	__asm__ __volatile__("@ atomic_sub\n"
++"1:	ldrex	%1, [%3]\n"
++"	subs	%0, %1, %4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
++"	strex	%1, %0, [%3]\n"
++"	teq	%1, #0\n"
++"	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
++	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
++	: "r" (&v->counter), "Ir" (i)
++	: "cc");
++}
++
++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v)
++{
++	unsigned long tmp;
++	int result;
++
++	__asm__ __volatile__("@ atomic_sub_unchecked\n"
+ "1:	ldrex	%0, [%3]\n"
+ "	sub	%0, %0, %4\n"
+ "	strex	%1, %0, [%3]\n"
+@@ -96,11 +202,25 @@ static inline int atomic_sub_return(int i, atomic_t *v)
+ 	smp_mb();
+ 
+ 	__asm__ __volatile__("@ atomic_sub_return\n"
+-"1:	ldrex	%0, [%3]\n"
+-"	sub	%0, %0, %4\n"
++"1:	ldrex	%1, [%3]\n"
++"	sub	%0, %1, %4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"	mov	%0, %1\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
+ "	strex	%1, %0, [%3]\n"
+ "	teq	%1, #0\n"
+ "	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
+ 	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
+ 	: "r" (&v->counter), "Ir" (i)
+ 	: "cc");
+@@ -132,6 +252,28 @@ static inline int atomic_cmpxchg(atomic_t *ptr, int old, int new)
+ 	return oldval;
+ }
+ 
++static inline int atomic_cmpxchg_unchecked(atomic_unchecked_t *ptr, int old, int new)
++{
++	unsigned long oldval, res;
++
++	smp_mb();
++
++	do {
++		__asm__ __volatile__("@ atomic_cmpxchg_unchecked\n"
++		"ldrex	%1, [%3]\n"
++		"mov	%0, #0\n"
++		"teq	%1, %4\n"
++		"strexeq %0, %5, [%3]\n"
++		    : "=&r" (res), "=&r" (oldval), "+Qo" (ptr->counter)
++		    : "r" (&ptr->counter), "Ir" (old), "r" (new)
++		    : "cc");
++	} while (res);
++
++	smp_mb();
++
++	return oldval;
++}
++
+ static inline void atomic_clear_mask(unsigned long mask, unsigned long *addr)
+ {
+ 	unsigned long tmp, tmp2;
+@@ -165,7 +307,9 @@ static inline int atomic_add_return(int i, atomic_t *v)
+ 
+ 	return val;
+ }
++#define atomic_add_return_unchecked(i, v) atomic_add_return(i, v)
+ #define atomic_add(i, v)	(void) atomic_add_return(i, v)
++#define atomic_add_unchecked(i, v)	(void) atomic_add_return_unchecked(i, v)
+ 
+ static inline int atomic_sub_return(int i, atomic_t *v)
+ {
+@@ -179,7 +323,9 @@ static inline int atomic_sub_return(int i, atomic_t *v)
+ 
+ 	return val;
+ }
++#define atomic_sub_return_unchecked(i, v) atomic_sub_return(i, v)
+ #define atomic_sub(i, v)	(void) atomic_sub_return(i, v)
++#define atomic_sub_unchecked(i, v)	(void) atomic_sub_return_unchecked(i, v)
+ 
+ static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+ {
+@@ -194,6 +340,7 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+ 
+ 	return ret;
+ }
++#define atomic_cmpxchg_unchecked(v, o, n) atomic_cmpxchg(v, o, n)
+ 
+ static inline void atomic_clear_mask(unsigned long mask, unsigned long *addr)
+ {
+@@ -207,6 +354,10 @@ static inline void atomic_clear_mask(unsigned long mask, unsigned long *addr)
+ #endif /* __LINUX_ARM_ARCH__ */
+ 
+ #define atomic_xchg(v, new) (xchg(&((v)->counter), new))
++static inline int atomic_xchg_unchecked(atomic_unchecked_t *v, int new)
++{
++	return xchg(&v->counter, new);
++}
+ 
+ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+ {
+@@ -219,11 +370,27 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+ }
+ 
+ #define atomic_inc(v)		atomic_add(1, v)
++static inline void atomic_inc_unchecked(atomic_unchecked_t *v)
++{
++	atomic_add_unchecked(1, v);
++}
+ #define atomic_dec(v)		atomic_sub(1, v)
++static inline void atomic_dec_unchecked(atomic_unchecked_t *v)
++{
++	atomic_sub_unchecked(1, v);
++}
+ 
+ #define atomic_inc_and_test(v)	(atomic_add_return(1, v) == 0)
++static inline int atomic_inc_and_test_unchecked(atomic_unchecked_t *v)
++{
++	return atomic_add_return_unchecked(1, v) == 0;
++}
+ #define atomic_dec_and_test(v)	(atomic_sub_return(1, v) == 0)
+ #define atomic_inc_return(v)    (atomic_add_return(1, v))
++static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v)
++{
++	return atomic_add_return_unchecked(1, v);
++}
+ #define atomic_dec_return(v)    (atomic_sub_return(1, v))
+ #define atomic_sub_and_test(i, v) (atomic_sub_return(i, v) == 0)
+ 
+@@ -239,6 +406,14 @@ typedef struct {
+ 	u64 __aligned(8) counter;
+ } atomic64_t;
+ 
++#ifdef CONFIG_PAX_REFCOUNT
++typedef struct {
++	u64 __aligned(8) counter;
++} atomic64_unchecked_t;
++#else
++typedef atomic64_t atomic64_unchecked_t;
++#endif
++
+ #define ATOMIC64_INIT(i) { (i) }
+ 
+ static inline u64 atomic64_read(atomic64_t *v)
+@@ -254,6 +429,19 @@ static inline u64 atomic64_read(atomic64_t *v)
+ 	return result;
+ }
+ 
++static inline u64 atomic64_read_unchecked(atomic64_unchecked_t *v)
++{
++	u64 result;
++
++	__asm__ __volatile__("@ atomic64_read_unchecked\n"
++"	ldrexd	%0, %H0, [%1]"
++	: "=&r" (result)
++	: "r" (&v->counter), "Qo" (v->counter)
++	);
++
++	return result;
++}
++
+ static inline void atomic64_set(atomic64_t *v, u64 i)
+ {
+ 	u64 tmp;
+@@ -268,6 +456,20 @@ static inline void atomic64_set(atomic64_t *v, u64 i)
+ 	: "cc");
+ }
+ 
++static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, u64 i)
++{
++	u64 tmp;
++
++	__asm__ __volatile__("@ atomic64_set_unchecked\n"
++"1:	ldrexd	%0, %H0, [%2]\n"
++"	strexd	%0, %3, %H3, [%2]\n"
++"	teq	%0, #0\n"
++"	bne	1b"
++	: "=&r" (tmp), "=Qo" (v->counter)
++	: "r" (&v->counter), "r" (i)
++	: "cc");
++}
++
+ static inline void atomic64_add(u64 i, atomic64_t *v)
+ {
+ 	u64 result;
+@@ -276,6 +478,36 @@ static inline void atomic64_add(u64 i, atomic64_t *v)
+ 	__asm__ __volatile__("@ atomic64_add\n"
+ "1:	ldrexd	%0, %H0, [%3]\n"
+ "	adds	%0, %0, %4\n"
++"	adcs	%H0, %H0, %H4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
++"	strexd	%1, %0, %H0, [%3]\n"
++"	teq	%1, #0\n"
++"	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
++	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
++	: "r" (&v->counter), "r" (i)
++	: "cc");
++}
++
++static inline void atomic64_add_unchecked(u64 i, atomic64_unchecked_t *v)
++{
++	u64 result;
++	unsigned long tmp;
++
++	__asm__ __volatile__("@ atomic64_add_unchecked\n"
++"1:	ldrexd	%0, %H0, [%3]\n"
++"	adds	%0, %0, %4\n"
+ "	adc	%H0, %H0, %H4\n"
+ "	strexd	%1, %0, %H0, [%3]\n"
+ "	teq	%1, #0\n"
+@@ -287,12 +519,49 @@ static inline void atomic64_add(u64 i, atomic64_t *v)
+ 
+ static inline u64 atomic64_add_return(u64 i, atomic64_t *v)
+ {
+-	u64 result;
+-	unsigned long tmp;
++	u64 result, tmp;
+ 
+ 	smp_mb();
+ 
+ 	__asm__ __volatile__("@ atomic64_add_return\n"
++"1:	ldrexd	%1, %H1, [%3]\n"
++"	adds	%0, %1, %4\n"
++"	adcs	%H0, %H1, %H4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"	mov	%0, %1\n"
++"	mov	%H0, %H1\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
++"	strexd	%1, %0, %H0, [%3]\n"
++"	teq	%1, #0\n"
++"	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
++	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
++	: "r" (&v->counter), "r" (i)
++	: "cc");
++
++	smp_mb();
++
++	return result;
++}
++
++static inline u64 atomic64_add_return_unchecked(u64 i, atomic64_unchecked_t *v)
++{
++	u64 result;
++	unsigned long tmp;
++
++	smp_mb();
++
++	__asm__ __volatile__("@ atomic64_add_return_unchecked\n"
+ "1:	ldrexd	%0, %H0, [%3]\n"
+ "	adds	%0, %0, %4\n"
+ "	adc	%H0, %H0, %H4\n"
+@@ -316,6 +585,36 @@ static inline void atomic64_sub(u64 i, atomic64_t *v)
+ 	__asm__ __volatile__("@ atomic64_sub\n"
+ "1:	ldrexd	%0, %H0, [%3]\n"
+ "	subs	%0, %0, %4\n"
++"	sbcs	%H0, %H0, %H4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
++"	strexd	%1, %0, %H0, [%3]\n"
++"	teq	%1, #0\n"
++"	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
++	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
++	: "r" (&v->counter), "r" (i)
++	: "cc");
++}
++
++static inline void atomic64_sub_unchecked(u64 i, atomic64_unchecked_t *v)
++{
++	u64 result;
++	unsigned long tmp;
++
++	__asm__ __volatile__("@ atomic64_sub_unchecked\n"
++"1:	ldrexd	%0, %H0, [%3]\n"
++"	subs	%0, %0, %4\n"
+ "	sbc	%H0, %H0, %H4\n"
+ "	strexd	%1, %0, %H0, [%3]\n"
+ "	teq	%1, #0\n"
+@@ -327,18 +626,32 @@ static inline void atomic64_sub(u64 i, atomic64_t *v)
+ 
+ static inline u64 atomic64_sub_return(u64 i, atomic64_t *v)
+ {
+-	u64 result;
+-	unsigned long tmp;
++	u64 result, tmp;
+ 
+ 	smp_mb();
+ 
+ 	__asm__ __volatile__("@ atomic64_sub_return\n"
+-"1:	ldrexd	%0, %H0, [%3]\n"
+-"	subs	%0, %0, %4\n"
+-"	sbc	%H0, %H0, %H4\n"
++"1:	ldrexd	%1, %H1, [%3]\n"
++"	subs	%0, %1, %4\n"
++"	sbc	%H0, %H1, %H4\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"	mov	%0, %1\n"
++"	mov	%H0, %H1\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
+ "	strexd	%1, %0, %H0, [%3]\n"
+ "	teq	%1, #0\n"
+ "	bne	1b"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"\n4:\n"
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
+ 	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
+ 	: "r" (&v->counter), "r" (i)
+ 	: "cc");
+@@ -372,6 +685,30 @@ static inline u64 atomic64_cmpxchg(atomic64_t *ptr, u64 old, u64 new)
+ 	return oldval;
+ }
+ 
++static inline u64 atomic64_cmpxchg_unchecked(atomic64_unchecked_t *ptr, u64 old, u64 new)
++{
++	u64 oldval;
++	unsigned long res;
++
++	smp_mb();
++
++	do {
++		__asm__ __volatile__("@ atomic64_cmpxchg_unchecked\n"
++		"ldrexd		%1, %H1, [%3]\n"
++		"mov		%0, #0\n"
++		"teq		%1, %4\n"
++		"teqeq		%H1, %H4\n"
++		"strexdeq	%0, %5, %H5, [%3]"
++		: "=&r" (res), "=&r" (oldval), "+Qo" (ptr->counter)
++		: "r" (&ptr->counter), "r" (old), "r" (new)
++		: "cc");
++	} while (res);
++
++	smp_mb();
++
++	return oldval;
++}
++
+ static inline u64 atomic64_xchg(atomic64_t *ptr, u64 new)
+ {
+ 	u64 result;
+@@ -395,21 +732,34 @@ static inline u64 atomic64_xchg(atomic64_t *ptr, u64 new)
+ 
+ static inline u64 atomic64_dec_if_positive(atomic64_t *v)
+ {
+-	u64 result;
+-	unsigned long tmp;
++	u64 result, tmp;
+ 
+ 	smp_mb();
+ 
+ 	__asm__ __volatile__("@ atomic64_dec_if_positive\n"
+-"1:	ldrexd	%0, %H0, [%3]\n"
+-"	subs	%0, %0, #1\n"
+-"	sbc	%H0, %H0, #0\n"
++"1:	ldrexd	%1, %H1, [%3]\n"
++"	subs	%0, %1, #1\n"
++"	sbc	%H0, %H1, #0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"	mov	%0, %1\n"
++"	mov	%H0, %H1\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
+ "	teq	%H0, #0\n"
+-"	bmi	2f\n"
++"	bmi	4f\n"
+ "	strexd	%1, %0, %H0, [%3]\n"
+ "	teq	%1, #0\n"
+ "	bne	1b\n"
+-"2:"
++"4:\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
+ 	: "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
+ 	: "r" (&v->counter)
+ 	: "cc");
+@@ -432,13 +782,25 @@ static inline int atomic64_add_unless(atomic64_t *v, u64 a, u64 u)
+ "	teq	%0, %5\n"
+ "	teqeq	%H0, %H5\n"
+ "	moveq	%1, #0\n"
+-"	beq	2f\n"
++"	beq	4f\n"
+ "	adds	%0, %0, %6\n"
+ "	adc	%H0, %H0, %H6\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	bvc	3f\n"
++"2:	bkpt	0xf103\n"
++"3:\n"
++#endif
++
+ "	strexd	%2, %0, %H0, [%4]\n"
+ "	teq	%2, #0\n"
+ "	bne	1b\n"
+-"2:"
++"4:\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++	_ASM_EXTABLE(2b, 4b)
++#endif
++
+ 	: "=&r" (val), "+r" (ret), "=&r" (tmp), "+Qo" (v->counter)
+ 	: "r" (&v->counter), "r" (u), "r" (a)
+ 	: "cc");
+@@ -451,10 +813,13 @@ static inline int atomic64_add_unless(atomic64_t *v, u64 a, u64 u)
+ 
+ #define atomic64_add_negative(a, v)	(atomic64_add_return((a), (v)) < 0)
+ #define atomic64_inc(v)			atomic64_add(1LL, (v))
++#define atomic64_inc_unchecked(v)	atomic64_add_unchecked(1LL, (v))
+ #define atomic64_inc_return(v)		atomic64_add_return(1LL, (v))
++#define atomic64_inc_return_unchecked(v)	atomic64_add_return_unchecked(1LL, (v))
+ #define atomic64_inc_and_test(v)	(atomic64_inc_return(v) == 0)
+ #define atomic64_sub_and_test(a, v)	(atomic64_sub_return((a), (v)) == 0)
+ #define atomic64_dec(v)			atomic64_sub(1LL, (v))
++#define atomic64_dec_unchecked(v)	atomic64_sub_unchecked(1LL, (v))
+ #define atomic64_dec_return(v)		atomic64_sub_return(1LL, (v))
+ #define atomic64_dec_and_test(v)	(atomic64_dec_return((v)) == 0)
+ #define atomic64_inc_not_zero(v)	atomic64_add_unless((v), 1LL, 0LL)
+diff --git a/arch/arm/include/asm/cache.h b/arch/arm/include/asm/cache.h
+index 75fe66b..2255c86 100644
+--- a/arch/arm/include/asm/cache.h
++++ b/arch/arm/include/asm/cache.h
+@@ -4,8 +4,10 @@
+ #ifndef __ASMARM_CACHE_H
+ #define __ASMARM_CACHE_H
+ 
++#include <linux/const.h>
++
+ #define L1_CACHE_SHIFT		CONFIG_ARM_L1_CACHE_SHIFT
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ /*
+  * Memory returned by kmalloc() may be used for DMA, so we must make
+diff --git a/arch/arm/include/asm/cacheflush.h b/arch/arm/include/asm/cacheflush.h
+index d5d8d5c..ad92c96 100644
+--- a/arch/arm/include/asm/cacheflush.h
++++ b/arch/arm/include/asm/cacheflush.h
+@@ -108,7 +108,7 @@ struct cpu_cache_fns {
+ 	void (*dma_unmap_area)(const void *, size_t, int);
+ 
+ 	void (*dma_flush_range)(const void *, const void *);
+-};
++} __no_const;
+ 
+ /*
+  * Select the calling method
+diff --git a/arch/arm/include/asm/elf.h b/arch/arm/include/asm/elf.h
+index 0e9ce8d..6ef1e03 100644
+--- a/arch/arm/include/asm/elf.h
++++ b/arch/arm/include/asm/elf.h
+@@ -116,7 +116,14 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs);
+    the loader.  We need to make sure that it is out of the way of the program
+    that it will "exec", and that there is sufficient room for the brk.  */
+ 
+-#define ELF_ET_DYN_BASE	(2 * TASK_SIZE / 3)
++#define ELF_ET_DYN_BASE		(TASK_SIZE / 3 * 2)
++
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	0x00008000UL
++
++#define PAX_DELTA_MMAP_LEN	((current->personality == PER_LINUX_32BIT) ? 16 : 10)
++#define PAX_DELTA_STACK_LEN	((current->personality == PER_LINUX_32BIT) ? 16 : 10)
++#endif
+ 
+ /* When the program starts, a1 contains a pointer to a function to be 
+    registered with atexit, as per the SVR4 ABI.  A value of 0 means we 
+@@ -126,10 +133,6 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs);
+ extern void elf_set_personality(const struct elf32_hdr *);
+ #define SET_PERSONALITY(ex)	elf_set_personality(&(ex))
+ 
+-struct mm_struct;
+-extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+-#define arch_randomize_brk arch_randomize_brk
+-
+ extern int vectors_user_mapping(void);
+ #define arch_setup_additional_pages(bprm, uses_interp) vectors_user_mapping()
+ #define ARCH_HAS_SETUP_ADDITIONAL_PAGES
+diff --git a/arch/arm/include/asm/kmap_types.h b/arch/arm/include/asm/kmap_types.h
+index e51b1e8..32a3113 100644
+--- a/arch/arm/include/asm/kmap_types.h
++++ b/arch/arm/include/asm/kmap_types.h
+@@ -21,6 +21,7 @@ enum km_type {
+ 	KM_L1_CACHE,
+ 	KM_L2_CACHE,
+ 	KM_KDB,
++	KM_CLEARPAGE,
+ 	KM_TYPE_NR
+ };
+ 
+diff --git a/arch/arm/include/asm/outercache.h b/arch/arm/include/asm/outercache.h
+index 53426c6..c7baff3 100644
+--- a/arch/arm/include/asm/outercache.h
++++ b/arch/arm/include/asm/outercache.h
+@@ -35,7 +35,7 @@ struct outer_cache_fns {
+ #endif
+ 	void (*set_debug)(unsigned long);
+ 	void (*resume)(void);
+-};
++} __no_const;
+ 
+ #ifdef CONFIG_OUTER_CACHE
+ 
+diff --git a/arch/arm/include/asm/page.h b/arch/arm/include/asm/page.h
+index ca94653..6ac0d56 100644
+--- a/arch/arm/include/asm/page.h
++++ b/arch/arm/include/asm/page.h
+@@ -123,7 +123,7 @@ struct cpu_user_fns {
+ 	void (*cpu_clear_user_highpage)(struct page *page, unsigned long vaddr);
+ 	void (*cpu_copy_user_highpage)(struct page *to, struct page *from,
+ 			unsigned long vaddr, struct vm_area_struct *vma);
+-};
++} __no_const;
+ 
+ #ifdef MULTI_USER
+ extern struct cpu_user_fns cpu_user;
+diff --git a/arch/arm/include/asm/system.h b/arch/arm/include/asm/system.h
+index 984014b..a6d914f 100644
+--- a/arch/arm/include/asm/system.h
++++ b/arch/arm/include/asm/system.h
+@@ -90,6 +90,8 @@ void hook_ifault_code(int nr, int (*fn)(unsigned long, unsigned int,
+ 
+ #define xchg(ptr,x) \
+ 	((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
++#define xchg_unchecked(ptr,x) \
++	((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
+ 
+ extern asmlinkage void c_backtrace(unsigned long fp, int pmode);
+ 
+@@ -101,7 +103,7 @@ extern int __pure cpu_architecture(void);
+ extern void cpu_init(void);
+ 
+ void arm_machine_restart(char mode, const char *cmd);
+-extern void (*arm_pm_restart)(char str, const char *cmd);
++extern void (*arm_pm_restart)(char str, const char *cmd) __noreturn;
+ 
+ #define UDBG_UNDEFINED	(1 << 0)
+ #define UDBG_SYSCALL	(1 << 1)
+@@ -526,6 +528,13 @@ static inline unsigned long long __cmpxchg64_mb(volatile void *ptr,
+ 
+ #endif	/* __LINUX_ARM_ARCH__ >= 6 */
+ 
++#define _ASM_EXTABLE(from, to)		\
++"	.pushsection __ex_table,\"a\"\n"\
++"	.align	3\n"			\
++"	.long	" #from ", " #to"\n"	\
++"	.popsection"
++
++
+ #endif /* __ASSEMBLY__ */
+ 
+ #define arch_align_stack(x) (x)
+diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h
+index b293616..96310e5 100644
+--- a/arch/arm/include/asm/uaccess.h
++++ b/arch/arm/include/asm/uaccess.h
+@@ -22,6 +22,8 @@
+ #define VERIFY_READ 0
+ #define VERIFY_WRITE 1
+ 
++extern void check_object_size(const void *ptr, unsigned long n, bool to);
++
+ /*
+  * The exception table consists of pairs of addresses: the first is the
+  * address of an instruction that is allowed to fault, and the second is
+@@ -387,8 +389,23 @@ do {									\
+ 
+ 
+ #ifdef CONFIG_MMU
+-extern unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n);
+-extern unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n);
++extern unsigned long __must_check ___copy_from_user(void *to, const void __user *from, unsigned long n);
++extern unsigned long __must_check ___copy_to_user(void __user *to, const void *from, unsigned long n);
++
++static inline unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n)
++{
++	if (!__builtin_constant_p(n))
++		check_object_size(to, n, false);
++	return ___copy_from_user(to, from, n);
++}
++
++static inline unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n)
++{
++	if (!__builtin_constant_p(n))
++		check_object_size(from, n, true);
++	return ___copy_to_user(to, from, n);
++}
++
+ extern unsigned long __must_check __copy_to_user_std(void __user *to, const void *from, unsigned long n);
+ extern unsigned long __must_check __clear_user(void __user *addr, unsigned long n);
+ extern unsigned long __must_check __clear_user_std(void __user *addr, unsigned long n);
+@@ -403,6 +420,9 @@ extern unsigned long __must_check __strnlen_user(const char __user *s, long n);
+ 
+ static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	if (access_ok(VERIFY_READ, from, n))
+ 		n = __copy_from_user(to, from, n);
+ 	else /* security hole - plug it */
+@@ -412,6 +432,9 @@ static inline unsigned long __must_check copy_from_user(void *to, const void __u
+ 
+ static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	if (access_ok(VERIFY_WRITE, to, n))
+ 		n = __copy_to_user(to, from, n);
+ 	return n;
+diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
+index 5b0bce6..becd81c 100644
+--- a/arch/arm/kernel/armksyms.c
++++ b/arch/arm/kernel/armksyms.c
+@@ -95,8 +95,8 @@ EXPORT_SYMBOL(__strncpy_from_user);
+ #ifdef CONFIG_MMU
+ EXPORT_SYMBOL(copy_page);
+ 
+-EXPORT_SYMBOL(__copy_from_user);
+-EXPORT_SYMBOL(__copy_to_user);
++EXPORT_SYMBOL(___copy_from_user);
++EXPORT_SYMBOL(___copy_to_user);
+ EXPORT_SYMBOL(__clear_user);
+ 
+ EXPORT_SYMBOL(__get_user_1);
+diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
+index 3d0c6fb..9d326fa 100644
+--- a/arch/arm/kernel/process.c
++++ b/arch/arm/kernel/process.c
+@@ -28,7 +28,6 @@
+ #include <linux/tick.h>
+ #include <linux/utsname.h>
+ #include <linux/uaccess.h>
+-#include <linux/random.h>
+ #include <linux/hw_breakpoint.h>
+ #include <linux/cpuidle.h>
+ 
+@@ -92,7 +91,7 @@ static int __init hlt_setup(char *__unused)
+ __setup("nohlt", nohlt_setup);
+ __setup("hlt", hlt_setup);
+ 
+-void arm_machine_restart(char mode, const char *cmd)
++__noreturn void arm_machine_restart(char mode, const char *cmd)
+ {
+ 	/* Disable interrupts first */
+ 	local_irq_disable();
+@@ -134,7 +133,7 @@ void arm_machine_restart(char mode, const char *cmd)
+ void (*pm_power_off)(void);
+ EXPORT_SYMBOL(pm_power_off);
+ 
+-void (*arm_pm_restart)(char str, const char *cmd) = arm_machine_restart;
++void (*arm_pm_restart)(char str, const char *cmd) __noreturn = arm_machine_restart;
+ EXPORT_SYMBOL_GPL(arm_pm_restart);
+ 
+ static void do_nothing(void *unused)
+@@ -248,6 +247,7 @@ void machine_power_off(void)
+ 	machine_shutdown();
+ 	if (pm_power_off)
+ 		pm_power_off();
++	BUG();
+ }
+ 
+ void machine_restart(char *cmd)
+@@ -484,12 +484,6 @@ unsigned long get_wchan(struct task_struct *p)
+ 	return 0;
+ }
+ 
+-unsigned long arch_randomize_brk(struct mm_struct *mm)
+-{
+-	unsigned long range_end = mm->brk + 0x02000000;
+-	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
+-}
+-
+ #ifdef CONFIG_MMU
+ /*
+  * The vectors page is always readable from user space for the
+diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
+index 8fc2c8f..064c150 100644
+--- a/arch/arm/kernel/setup.c
++++ b/arch/arm/kernel/setup.c
+@@ -108,13 +108,13 @@ struct processor processor __read_mostly;
+ struct cpu_tlb_fns cpu_tlb __read_mostly;
+ #endif
+ #ifdef MULTI_USER
+-struct cpu_user_fns cpu_user __read_mostly;
++struct cpu_user_fns cpu_user __read_only;
+ #endif
+ #ifdef MULTI_CACHE
+-struct cpu_cache_fns cpu_cache __read_mostly;
++struct cpu_cache_fns cpu_cache __read_only;
+ #endif
+ #ifdef CONFIG_OUTER_CACHE
+-struct outer_cache_fns outer_cache __read_mostly;
++struct outer_cache_fns outer_cache __read_only;
+ EXPORT_SYMBOL(outer_cache);
+ #endif
+ 
+diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
+index 99a5727..a3d5bb1 100644
+--- a/arch/arm/kernel/traps.c
++++ b/arch/arm/kernel/traps.c
+@@ -259,6 +259,8 @@ static int __die(const char *str, int err, struct thread_info *thread, struct pt
+ 
+ static DEFINE_RAW_SPINLOCK(die_lock);
+ 
++extern void gr_handle_kernel_exploit(void);
++
+ /*
+  * This function is protected against re-entrancy.
+  */
+@@ -288,6 +290,9 @@ void die(const char *str, struct pt_regs *regs, int err)
+ 		panic("Fatal exception in interrupt");
+ 	if (panic_on_oops)
+ 		panic("Fatal exception");
++
++	gr_handle_kernel_exploit();
++
+ 	if (ret != NOTIFY_STOP)
+ 		do_exit(SIGSEGV);
+ }
+diff --git a/arch/arm/lib/copy_from_user.S b/arch/arm/lib/copy_from_user.S
+index 66a477a..bee61d3 100644
+--- a/arch/arm/lib/copy_from_user.S
++++ b/arch/arm/lib/copy_from_user.S
+@@ -16,7 +16,7 @@
+ /*
+  * Prototype:
+  *
+- *	size_t __copy_from_user(void *to, const void *from, size_t n)
++ *	size_t ___copy_from_user(void *to, const void *from, size_t n)
+  *
+  * Purpose:
+  *
+@@ -84,11 +84,11 @@
+ 
+ 	.text
+ 
+-ENTRY(__copy_from_user)
++ENTRY(___copy_from_user)
+ 
+ #include "copy_template.S"
+ 
+-ENDPROC(__copy_from_user)
++ENDPROC(___copy_from_user)
+ 
+ 	.pushsection .fixup,"ax"
+ 	.align 0
+diff --git a/arch/arm/lib/copy_page.S b/arch/arm/lib/copy_page.S
+index 6ee2f67..d1cce76 100644
+--- a/arch/arm/lib/copy_page.S
++++ b/arch/arm/lib/copy_page.S
+@@ -10,6 +10,7 @@
+  *  ASM optimised string functions
+  */
+ #include <linux/linkage.h>
++#include <linux/const.h>
+ #include <asm/assembler.h>
+ #include <asm/asm-offsets.h>
+ #include <asm/cache.h>
+diff --git a/arch/arm/lib/copy_to_user.S b/arch/arm/lib/copy_to_user.S
+index d066df6..df28194 100644
+--- a/arch/arm/lib/copy_to_user.S
++++ b/arch/arm/lib/copy_to_user.S
+@@ -16,7 +16,7 @@
+ /*
+  * Prototype:
+  *
+- *	size_t __copy_to_user(void *to, const void *from, size_t n)
++ *	size_t ___copy_to_user(void *to, const void *from, size_t n)
+  *
+  * Purpose:
+  *
+@@ -88,11 +88,11 @@
+ 	.text
+ 
+ ENTRY(__copy_to_user_std)
+-WEAK(__copy_to_user)
++WEAK(___copy_to_user)
+ 
+ #include "copy_template.S"
+ 
+-ENDPROC(__copy_to_user)
++ENDPROC(___copy_to_user)
+ ENDPROC(__copy_to_user_std)
+ 
+ 	.pushsection .fixup,"ax"
+diff --git a/arch/arm/lib/uaccess.S b/arch/arm/lib/uaccess.S
+index d0ece2a..5ae2f39 100644
+--- a/arch/arm/lib/uaccess.S
++++ b/arch/arm/lib/uaccess.S
+@@ -20,7 +20,7 @@
+ 
+ #define PAGE_SHIFT 12
+ 
+-/* Prototype: int __copy_to_user(void *to, const char *from, size_t n)
++/* Prototype: int ___copy_to_user(void *to, const char *from, size_t n)
+  * Purpose  : copy a block to user memory from kernel memory
+  * Params   : to   - user memory
+  *          : from - kernel memory
+@@ -40,7 +40,7 @@ USER(		T(strgtb) r3, [r0], #1)			@ May fault
+ 		sub	r2, r2, ip
+ 		b	.Lc2u_dest_aligned
+ 
+-ENTRY(__copy_to_user)
++ENTRY(___copy_to_user)
+ 		stmfd	sp!, {r2, r4 - r7, lr}
+ 		cmp	r2, #4
+ 		blt	.Lc2u_not_enough
+@@ -278,14 +278,14 @@ USER(		T(strgeb) r3, [r0], #1)			@ May fault
+ 		ldrgtb	r3, [r1], #0
+ USER(		T(strgtb) r3, [r0], #1)			@ May fault
+ 		b	.Lc2u_finished
+-ENDPROC(__copy_to_user)
++ENDPROC(___copy_to_user)
+ 
+ 		.pushsection .fixup,"ax"
+ 		.align	0
+ 9001:		ldmfd	sp!, {r0, r4 - r7, pc}
+ 		.popsection
+ 
+-/* Prototype: unsigned long __copy_from_user(void *to,const void *from,unsigned long n);
++/* Prototype: unsigned long ___copy_from_user(void *to,const void *from,unsigned long n);
+  * Purpose  : copy a block from user memory to kernel memory
+  * Params   : to   - kernel memory
+  *          : from - user memory
+@@ -304,7 +304,7 @@ USER(		T(ldrgtb) r3, [r1], #1)			@ May fault
+ 		sub	r2, r2, ip
+ 		b	.Lcfu_dest_aligned
+ 
+-ENTRY(__copy_from_user)
++ENTRY(___copy_from_user)
+ 		stmfd	sp!, {r0, r2, r4 - r7, lr}
+ 		cmp	r2, #4
+ 		blt	.Lcfu_not_enough
+@@ -544,7 +544,7 @@ USER(		T(ldrgeb) r3, [r1], #1)			@ May fault
+ USER(		T(ldrgtb) r3, [r1], #1)			@ May fault
+ 		strgtb	r3, [r0], #1
+ 		b	.Lcfu_finished
+-ENDPROC(__copy_from_user)
++ENDPROC(___copy_from_user)
+ 
+ 		.pushsection .fixup,"ax"
+ 		.align	0
+diff --git a/arch/arm/lib/uaccess_with_memcpy.c b/arch/arm/lib/uaccess_with_memcpy.c
+index 025f742..8432b08 100644
+--- a/arch/arm/lib/uaccess_with_memcpy.c
++++ b/arch/arm/lib/uaccess_with_memcpy.c
+@@ -104,7 +104,7 @@ out:
+ }
+ 
+ unsigned long
+-__copy_to_user(void __user *to, const void *from, unsigned long n)
++___copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
+ 	/*
+ 	 * This test is stubbed out of the main function above to keep
+diff --git a/arch/arm/mach-omap2/board-n8x0.c b/arch/arm/mach-omap2/board-n8x0.c
+index e9d5f4a..f099699 100644
+--- a/arch/arm/mach-omap2/board-n8x0.c
++++ b/arch/arm/mach-omap2/board-n8x0.c
+@@ -593,7 +593,7 @@ static int n8x0_menelaus_late_init(struct device *dev)
+ }
+ #endif
+ 
+-static struct menelaus_platform_data n8x0_menelaus_platform_data __initdata = {
++static struct menelaus_platform_data n8x0_menelaus_platform_data __initconst = {
+ 	.late_init = n8x0_menelaus_late_init,
+ };
+ 
+diff --git a/arch/arm/mach-ux500/mbox-db5500.c b/arch/arm/mach-ux500/mbox-db5500.c
+index 2b2d51c..0127490 100644
+--- a/arch/arm/mach-ux500/mbox-db5500.c
++++ b/arch/arm/mach-ux500/mbox-db5500.c
+@@ -168,7 +168,7 @@ static ssize_t mbox_read_fifo(struct device *dev,
+ 	return sprintf(buf, "0x%X\n", mbox_value);
+ }
+ 
+-static DEVICE_ATTR(fifo, S_IWUGO | S_IRUGO, mbox_read_fifo, mbox_write_fifo);
++static DEVICE_ATTR(fifo, S_IWUSR | S_IRUGO, mbox_read_fifo, mbox_write_fifo);
+ 
+ static int mbox_show(struct seq_file *s, void *data)
+ {
+diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
+index aa33949..d366075 100644
+--- a/arch/arm/mm/fault.c
++++ b/arch/arm/mm/fault.c
+@@ -183,6 +183,13 @@ __do_user_fault(struct task_struct *tsk, unsigned long addr,
+ 	}
+ #endif
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++	if (fsr & FSR_LNX_PF) {
++		pax_report_fault(regs, (void *)regs->ARM_pc, (void *)regs->ARM_sp);
++		do_group_exit(SIGKILL);
++	}
++#endif
++
+ 	tsk->thread.address = addr;
+ 	tsk->thread.error_code = fsr;
+ 	tsk->thread.trap_no = 14;
+@@ -384,6 +391,33 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
+ }
+ #endif					/* CONFIG_MMU */
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 20; i++) {
++		unsigned char c;
++		if (get_user(c, (__force unsigned char __user *)pc+i))
++			printk(KERN_CONT "?? ");
++		else
++			printk(KERN_CONT "%02x ", c);
++	}
++	printk("\n");
++
++	printk(KERN_ERR "PAX: bytes at SP-4: ");
++	for (i = -1; i < 20; i++) {
++		unsigned long c;
++		if (get_user(c, (__force unsigned long __user *)sp+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08lx ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ /*
+  * First Level Translation Fault Handler
+  *
+@@ -628,6 +662,20 @@ do_PrefetchAbort(unsigned long addr, unsigned int ifsr, struct pt_regs *regs)
+ 	const struct fsr_info *inf = ifsr_info + fsr_fs(ifsr);
+ 	struct siginfo info;
+ 
++#ifdef CONFIG_PAX_REFCOUNT
++	if (fsr_fs(ifsr) == 2) {
++		unsigned int bkpt;
++
++		if (!probe_kernel_address((unsigned int *)addr, bkpt) && bkpt == 0xe12f1073) {
++			current->thread.error_code = ifsr;
++			current->thread.trap_no = 0;
++			pax_report_refcount_overflow(regs);
++			fixup_exception(regs);
++			return;
++		}
++	}
++#endif
++
+ 	if (!inf->fn(addr, ifsr | FSR_LNX_PF, regs))
+ 		return;
+ 
+diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
+index 44b628e..623ee2a 100644
+--- a/arch/arm/mm/mmap.c
++++ b/arch/arm/mm/mmap.c
+@@ -54,6 +54,10 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 	if (len > TASK_SIZE)
+ 		return -ENOMEM;
+ 
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
+ 	if (addr) {
+ 		if (do_align)
+ 			addr = COLOUR_ALIGN(addr, pgoff);
+@@ -61,15 +65,14 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 			addr = PAGE_ALIGN(addr);
+ 
+ 		vma = find_vma(mm, addr);
+-		if (TASK_SIZE - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 	if (len > mm->cached_hole_size) {
+-	        start_addr = addr = mm->free_area_cache;
++		start_addr = addr = mm->free_area_cache;
+ 	} else {
+-	        start_addr = addr = TASK_UNMAPPED_BASE;
+-	        mm->cached_hole_size = 0;
++		start_addr = addr = mm->mmap_base;
++		mm->cached_hole_size = 0;
+ 	}
+ 	/* 8 bits of randomness in 20 address space bits */
+ 	if ((current->flags & PF_RANDOMIZE) &&
+@@ -89,14 +92,14 @@ full_search:
+ 			 * Start a new search - just in case we missed
+ 			 * some holes.
+ 			 */
+-			if (start_addr != TASK_UNMAPPED_BASE) {
+-				start_addr = addr = TASK_UNMAPPED_BASE;
++			if (start_addr != mm->mmap_base) {
++				start_addr = addr = mm->mmap_base;
+ 				mm->cached_hole_size = 0;
+ 				goto full_search;
+ 			}
+ 			return -ENOMEM;
+ 		}
+-		if (!vma || addr + len <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr, len)) {
+ 			/*
+ 			 * Remember the place where we stopped the search:
+ 			 */
+diff --git a/arch/arm/plat-samsung/include/plat/dma-ops.h b/arch/arm/plat-samsung/include/plat/dma-ops.h
+index 4c1a363..df311d0 100644
+--- a/arch/arm/plat-samsung/include/plat/dma-ops.h
++++ b/arch/arm/plat-samsung/include/plat/dma-ops.h
+@@ -41,7 +41,7 @@ struct samsung_dma_ops {
+ 	int (*started)(unsigned ch);
+ 	int (*flush)(unsigned ch);
+ 	int (*stop)(unsigned ch);
+-};
++} __no_const;
+ 
+ extern void *samsung_dmadev_get_ops(void);
+ extern void *s3c_dma_get_ops(void);
+diff --git a/arch/arm/plat-samsung/include/plat/ehci.h b/arch/arm/plat-samsung/include/plat/ehci.h
+index 5f28cae..3d23723 100644
+--- a/arch/arm/plat-samsung/include/plat/ehci.h
++++ b/arch/arm/plat-samsung/include/plat/ehci.h
+@@ -14,7 +14,7 @@
+ struct s5p_ehci_platdata {
+ 	int (*phy_init)(struct platform_device *pdev, int type);
+ 	int (*phy_exit)(struct platform_device *pdev, int type);
+-};
++} __no_const;
+ 
+ extern void s5p_ehci_set_platdata(struct s5p_ehci_platdata *pd);
+ 
+diff --git a/arch/avr32/include/asm/cache.h b/arch/avr32/include/asm/cache.h
+index c3a58a1..78fbf54 100644
+--- a/arch/avr32/include/asm/cache.h
++++ b/arch/avr32/include/asm/cache.h
+@@ -1,8 +1,10 @@
+ #ifndef __ASM_AVR32_CACHE_H
+ #define __ASM_AVR32_CACHE_H
+ 
++#include <linux/const.h>
++
+ #define L1_CACHE_SHIFT 5
+-#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ /*
+  * Memory returned by kmalloc() may be used for DMA, so we must make
+diff --git a/arch/avr32/include/asm/elf.h b/arch/avr32/include/asm/elf.h
+index 3b3159b..425ea94 100644
+--- a/arch/avr32/include/asm/elf.h
++++ b/arch/avr32/include/asm/elf.h
+@@ -84,8 +84,14 @@ typedef struct user_fpu_struct elf_fpregset_t;
+    the loader.  We need to make sure that it is out of the way of the program
+    that it will "exec", and that there is sufficient room for the brk.  */
+ 
+-#define ELF_ET_DYN_BASE         (2 * TASK_SIZE / 3)
++#define ELF_ET_DYN_BASE		(TASK_SIZE / 3 * 2)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	0x00001000UL
++
++#define PAX_DELTA_MMAP_LEN	15
++#define PAX_DELTA_STACK_LEN	15
++#endif
+ 
+ /* This yields a mask that user programs can use to figure out what
+    instruction set this CPU supports.  This could be done in user space,
+diff --git a/arch/avr32/include/asm/kmap_types.h b/arch/avr32/include/asm/kmap_types.h
+index b7f5c68..556135c 100644
+--- a/arch/avr32/include/asm/kmap_types.h
++++ b/arch/avr32/include/asm/kmap_types.h
+@@ -22,7 +22,8 @@ D(10)	KM_IRQ0,
+ D(11)	KM_IRQ1,
+ D(12)	KM_SOFTIRQ0,
+ D(13)	KM_SOFTIRQ1,
+-D(14)	KM_TYPE_NR
++D(14)	KM_CLEARPAGE,
++D(15)	KM_TYPE_NR
+ };
+ 
+ #undef D
+diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c
+index f7040a1..db9f300 100644
+--- a/arch/avr32/mm/fault.c
++++ b/arch/avr32/mm/fault.c
+@@ -41,6 +41,23 @@ static inline int notify_page_fault(struct pt_regs *regs, int trap)
+ 
+ int exception_trace = 1;
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 20; i++) {
++		unsigned char c;
++		if (get_user(c, (unsigned char *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%02x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ /*
+  * This routine handles page faults. It determines the address and the
+  * problem, and then passes it off to one of the appropriate routines.
+@@ -156,6 +173,16 @@ bad_area:
+ 	up_read(&mm->mmap_sem);
+ 
+ 	if (user_mode(regs)) {
++
++#ifdef CONFIG_PAX_PAGEEXEC
++		if (mm->pax_flags & MF_PAX_PAGEEXEC) {
++			if (ecr == ECR_PROTECTION_X || ecr == ECR_TLB_MISS_X) {
++				pax_report_fault(regs, (void *)regs->pc, (void *)regs->sp);
++				do_group_exit(SIGKILL);
++			}
++		}
++#endif
++
+ 		if (exception_trace && printk_ratelimit())
+ 			printk("%s%s[%d]: segfault at %08lx pc %08lx "
+ 			       "sp %08lx ecr %lu\n",
+diff --git a/arch/blackfin/include/asm/cache.h b/arch/blackfin/include/asm/cache.h
+index 568885a..f8008df 100644
+--- a/arch/blackfin/include/asm/cache.h
++++ b/arch/blackfin/include/asm/cache.h
+@@ -7,6 +7,7 @@
+ #ifndef __ARCH_BLACKFIN_CACHE_H
+ #define __ARCH_BLACKFIN_CACHE_H
+ 
++#include <linux/const.h>
+ #include <linux/linkage.h>	/* for asmlinkage */
+ 
+ /*
+@@ -14,7 +15,7 @@
+  * Blackfin loads 32 bytes for cache
+  */
+ #define L1_CACHE_SHIFT	5
+-#define L1_CACHE_BYTES	(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES	(_AC(1,UL) << L1_CACHE_SHIFT)
+ #define SMP_CACHE_BYTES	L1_CACHE_BYTES
+ 
+ #define ARCH_DMA_MINALIGN	L1_CACHE_BYTES
+diff --git a/arch/cris/include/arch-v10/arch/cache.h b/arch/cris/include/arch-v10/arch/cache.h
+index aea2718..3639a60 100644
+--- a/arch/cris/include/arch-v10/arch/cache.h
++++ b/arch/cris/include/arch-v10/arch/cache.h
+@@ -1,8 +1,9 @@
+ #ifndef _ASM_ARCH_CACHE_H
+ #define _ASM_ARCH_CACHE_H
+ 
++#include <linux/const.h>
+ /* Etrax 100LX have 32-byte cache-lines. */
+-#define L1_CACHE_BYTES 32
+ #define L1_CACHE_SHIFT 5
++#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #endif /* _ASM_ARCH_CACHE_H */
+diff --git a/arch/cris/include/arch-v32/arch/cache.h b/arch/cris/include/arch-v32/arch/cache.h
+index 1de779f..336fad3 100644
+--- a/arch/cris/include/arch-v32/arch/cache.h
++++ b/arch/cris/include/arch-v32/arch/cache.h
+@@ -1,11 +1,12 @@
+ #ifndef _ASM_CRIS_ARCH_CACHE_H
+ #define _ASM_CRIS_ARCH_CACHE_H
+ 
++#include <linux/const.h>
+ #include <arch/hwregs/dma.h>
+ 
+ /* A cache-line is 32 bytes. */
+-#define L1_CACHE_BYTES 32
+ #define L1_CACHE_SHIFT 5
++#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define __read_mostly __attribute__((__section__(".data.read_mostly")))
+ 
+diff --git a/arch/frv/include/asm/atomic.h b/arch/frv/include/asm/atomic.h
+index 0d8a7d6..d0c9ff5 100644
+--- a/arch/frv/include/asm/atomic.h
++++ b/arch/frv/include/asm/atomic.h
+@@ -241,6 +241,16 @@ extern uint32_t __xchg_32(uint32_t i, volatile void *v);
+ #define atomic64_cmpxchg(v, old, new)	(__cmpxchg_64(old, new, &(v)->counter))
+ #define atomic64_xchg(v, new)		(__xchg_64(new, &(v)->counter))
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ static __inline__ int __atomic_add_unless(atomic_t *v, int a, int u)
+ {
+ 	int c, old;
+diff --git a/arch/frv/include/asm/cache.h b/arch/frv/include/asm/cache.h
+index 2797163..c2a401d 100644
+--- a/arch/frv/include/asm/cache.h
++++ b/arch/frv/include/asm/cache.h
+@@ -12,10 +12,11 @@
+ #ifndef __ASM_CACHE_H
+ #define __ASM_CACHE_H
+ 
++#include <linux/const.h>
+ 
+ /* bytes per L1 cache line */
+ #define L1_CACHE_SHIFT		(CONFIG_FRV_L1_CACHE_SHIFT)
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define __cacheline_aligned	__attribute__((aligned(L1_CACHE_BYTES)))
+ #define ____cacheline_aligned	__attribute__((aligned(L1_CACHE_BYTES)))
+diff --git a/arch/frv/include/asm/kmap_types.h b/arch/frv/include/asm/kmap_types.h
+index f8e16b2..c73ff79 100644
+--- a/arch/frv/include/asm/kmap_types.h
++++ b/arch/frv/include/asm/kmap_types.h
+@@ -23,6 +23,7 @@ enum km_type {
+ 	KM_IRQ1,
+ 	KM_SOFTIRQ0,
+ 	KM_SOFTIRQ1,
++	KM_CLEARPAGE,
+ 	KM_TYPE_NR
+ };
+ 
+diff --git a/arch/frv/mm/elf-fdpic.c b/arch/frv/mm/elf-fdpic.c
+index 385fd30..6c3d97e 100644
+--- a/arch/frv/mm/elf-fdpic.c
++++ b/arch/frv/mm/elf-fdpic.c
+@@ -73,8 +73,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 	if (addr) {
+ 		addr = PAGE_ALIGN(addr);
+ 		vma = find_vma(current->mm, addr);
+-		if (TASK_SIZE - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			goto success;
+ 	}
+ 
+@@ -89,7 +88,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 			for (; vma; vma = vma->vm_next) {
+ 				if (addr > limit)
+ 					break;
+-				if (addr + len <= vma->vm_start)
++				if (check_heap_stack_gap(vma, addr, len))
+ 					goto success;
+ 				addr = vma->vm_end;
+ 			}
+@@ -104,7 +103,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 		for (; vma; vma = vma->vm_next) {
+ 			if (addr > limit)
+ 				break;
+-			if (addr + len <= vma->vm_start)
++			if (check_heap_stack_gap(vma, addr, len))
+ 				goto success;
+ 			addr = vma->vm_end;
+ 		}
+diff --git a/arch/h8300/include/asm/cache.h b/arch/h8300/include/asm/cache.h
+index c635028..6d9445a 100644
+--- a/arch/h8300/include/asm/cache.h
++++ b/arch/h8300/include/asm/cache.h
+@@ -1,8 +1,10 @@
+ #ifndef __ARCH_H8300_CACHE_H
+ #define __ARCH_H8300_CACHE_H
+ 
++#include <linux/const.h>
++
+ /* bytes per L1 cache line */
+-#define        L1_CACHE_BYTES  4
++#define        L1_CACHE_BYTES  _AC(4,UL)
+ 
+ /* m68k-elf-gcc  2.95.2 doesn't like these */
+ 
+diff --git a/arch/hexagon/include/asm/cache.h b/arch/hexagon/include/asm/cache.h
+index 0f01de2..d37d309 100644
+--- a/arch/hexagon/include/asm/cache.h
++++ b/arch/hexagon/include/asm/cache.h
+@@ -21,9 +21,11 @@
+ #ifndef __ASM_CACHE_H
+ #define __ASM_CACHE_H
+ 
++#include <linux/const.h>
++
+ /* Bytes per L1 cache line */
+-#define L1_CACHE_SHIFT		(5)
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_SHIFT		5
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define __cacheline_aligned	__aligned(L1_CACHE_BYTES)
+ #define ____cacheline_aligned	__aligned(L1_CACHE_BYTES)
+diff --git a/arch/ia64/include/asm/atomic.h b/arch/ia64/include/asm/atomic.h
+index 3fad89e..3047da5 100644
+--- a/arch/ia64/include/asm/atomic.h
++++ b/arch/ia64/include/asm/atomic.h
+@@ -209,6 +209,16 @@ atomic64_add_negative (__s64 i, atomic64_t *v)
+ #define atomic64_inc(v)			atomic64_add(1, (v))
+ #define atomic64_dec(v)			atomic64_sub(1, (v))
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ /* Atomic operations are already serializing */
+ #define smp_mb__before_atomic_dec()	barrier()
+ #define smp_mb__after_atomic_dec()	barrier()
+diff --git a/arch/ia64/include/asm/cache.h b/arch/ia64/include/asm/cache.h
+index 988254a..e1ee885 100644
+--- a/arch/ia64/include/asm/cache.h
++++ b/arch/ia64/include/asm/cache.h
+@@ -1,6 +1,7 @@
+ #ifndef _ASM_IA64_CACHE_H
+ #define _ASM_IA64_CACHE_H
+ 
++#include <linux/const.h>
+ 
+ /*
+  * Copyright (C) 1998-2000 Hewlett-Packard Co
+@@ -9,7 +10,7 @@
+ 
+ /* Bytes per L1 (data) cache line.  */
+ #define L1_CACHE_SHIFT		CONFIG_IA64_L1_CACHE_SHIFT
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #ifdef CONFIG_SMP
+ # define SMP_CACHE_SHIFT	L1_CACHE_SHIFT
+diff --git a/arch/ia64/include/asm/elf.h b/arch/ia64/include/asm/elf.h
+index b5298eb..67c6e62 100644
+--- a/arch/ia64/include/asm/elf.h
++++ b/arch/ia64/include/asm/elf.h
+@@ -42,6 +42,13 @@
+  */
+ #define ELF_ET_DYN_BASE		(TASK_UNMAPPED_BASE + 0x800000000UL)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(current->personality == PER_LINUX32 ? 0x08048000UL : 0x4000000000000000UL)
++
++#define PAX_DELTA_MMAP_LEN	(current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13)
++#define PAX_DELTA_STACK_LEN	(current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13)
++#endif
++
+ #define PT_IA_64_UNWIND		0x70000001
+ 
+ /* IA-64 relocations: */
+diff --git a/arch/ia64/include/asm/pgtable.h b/arch/ia64/include/asm/pgtable.h
+index 1a97af3..7529d31 100644
+--- a/arch/ia64/include/asm/pgtable.h
++++ b/arch/ia64/include/asm/pgtable.h
+@@ -12,7 +12,7 @@
+  *	David Mosberger-Tang <davidm@hpl.hp.com>
+  */
+ 
+-
++#include <linux/const.h>
+ #include <asm/mman.h>
+ #include <asm/page.h>
+ #include <asm/processor.h>
+@@ -143,6 +143,17 @@
+ #define PAGE_READONLY	__pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R)
+ #define PAGE_COPY	__pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R)
+ #define PAGE_COPY_EXEC	__pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RX)
++
++#ifdef CONFIG_PAX_PAGEEXEC
++# define PAGE_SHARED_NOEXEC	__pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RW)
++# define PAGE_READONLY_NOEXEC	__pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R)
++# define PAGE_COPY_NOEXEC	__pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R)
++#else
++# define PAGE_SHARED_NOEXEC	PAGE_SHARED
++# define PAGE_READONLY_NOEXEC	PAGE_READONLY
++# define PAGE_COPY_NOEXEC	PAGE_COPY
++#endif
++
+ #define PAGE_GATE	__pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_X_RX)
+ #define PAGE_KERNEL	__pgprot(__DIRTY_BITS  | _PAGE_PL_0 | _PAGE_AR_RWX)
+ #define PAGE_KERNELRX	__pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_RX)
+diff --git a/arch/ia64/include/asm/spinlock.h b/arch/ia64/include/asm/spinlock.h
+index b77768d..e0795eb 100644
+--- a/arch/ia64/include/asm/spinlock.h
++++ b/arch/ia64/include/asm/spinlock.h
+@@ -72,7 +72,7 @@ static __always_inline void __ticket_spin_unlock(arch_spinlock_t *lock)
+ 	unsigned short	*p = (unsigned short *)&lock->lock + 1, tmp;
+ 
+ 	asm volatile ("ld2.bias %0=[%1]" : "=r"(tmp) : "r"(p));
+-	ACCESS_ONCE(*p) = (tmp + 2) & ~1;
++	ACCESS_ONCE_RW(*p) = (tmp + 2) & ~1;
+ }
+ 
+ static __always_inline void __ticket_spin_unlock_wait(arch_spinlock_t *lock)
+diff --git a/arch/ia64/include/asm/uaccess.h b/arch/ia64/include/asm/uaccess.h
+index 449c8c0..432a3d2 100644
+--- a/arch/ia64/include/asm/uaccess.h
++++ b/arch/ia64/include/asm/uaccess.h
+@@ -257,7 +257,7 @@ __copy_from_user (void *to, const void __user *from, unsigned long count)
+ 	const void *__cu_from = (from);							\
+ 	long __cu_len = (n);								\
+ 											\
+-	if (__access_ok(__cu_to, __cu_len, get_fs()))					\
++	if (__cu_len > 0  && __cu_len <= INT_MAX && __access_ok(__cu_to, __cu_len, get_fs()))			\
+ 		__cu_len = __copy_user(__cu_to, (__force void __user *) __cu_from, __cu_len);	\
+ 	__cu_len;									\
+ })
+@@ -269,7 +269,7 @@ __copy_from_user (void *to, const void __user *from, unsigned long count)
+ 	long __cu_len = (n);								\
+ 											\
+ 	__chk_user_ptr(__cu_from);							\
+-	if (__access_ok(__cu_from, __cu_len, get_fs()))					\
++	if (__cu_len > 0 && __cu_len <= INT_MAX  && __access_ok(__cu_from, __cu_len, get_fs()))			\
+ 		__cu_len = __copy_user((__force void __user *) __cu_to, __cu_from, __cu_len);	\
+ 	__cu_len;									\
+ })
+diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c
+index 24603be..948052d 100644
+--- a/arch/ia64/kernel/module.c
++++ b/arch/ia64/kernel/module.c
+@@ -307,8 +307,7 @@ plt_target (struct plt_entry *plt)
+ void
+ module_free (struct module *mod, void *module_region)
+ {
+-	if (mod && mod->arch.init_unw_table &&
+-	    module_region == mod->module_init) {
++	if (mod && mod->arch.init_unw_table && module_region == mod->module_init_rx) {
+ 		unw_remove_unwind_table(mod->arch.init_unw_table);
+ 		mod->arch.init_unw_table = NULL;
+ 	}
+@@ -494,15 +493,39 @@ module_frob_arch_sections (Elf_Ehdr *ehdr, Elf_Shdr *sechdrs, char *secstrings,
+ }
+ 
+ static inline int
++in_init_rx (const struct module *mod, uint64_t addr)
++{
++	return addr - (uint64_t) mod->module_init_rx < mod->init_size_rx;
++}
++
++static inline int
++in_init_rw (const struct module *mod, uint64_t addr)
++{
++	return addr - (uint64_t) mod->module_init_rw < mod->init_size_rw;
++}
++
++static inline int
+ in_init (const struct module *mod, uint64_t addr)
+ {
+-	return addr - (uint64_t) mod->module_init < mod->init_size;
++	return in_init_rx(mod, addr) || in_init_rw(mod, addr);
++}
++
++static inline int
++in_core_rx (const struct module *mod, uint64_t addr)
++{
++	return addr - (uint64_t) mod->module_core_rx < mod->core_size_rx;
++}
++
++static inline int
++in_core_rw (const struct module *mod, uint64_t addr)
++{
++	return addr - (uint64_t) mod->module_core_rw < mod->core_size_rw;
+ }
+ 
+ static inline int
+ in_core (const struct module *mod, uint64_t addr)
+ {
+-	return addr - (uint64_t) mod->module_core < mod->core_size;
++	return in_core_rx(mod, addr) || in_core_rw(mod, addr);
+ }
+ 
+ static inline int
+@@ -685,7 +708,14 @@ do_reloc (struct module *mod, uint8_t r_type, Elf64_Sym *sym, uint64_t addend,
+ 		break;
+ 
+ 	      case RV_BDREL:
+-		val -= (uint64_t) (in_init(mod, val) ? mod->module_init : mod->module_core);
++		if (in_init_rx(mod, val))
++			val -= (uint64_t) mod->module_init_rx;
++		else if (in_init_rw(mod, val))
++			val -= (uint64_t) mod->module_init_rw;
++		else if (in_core_rx(mod, val))
++			val -= (uint64_t) mod->module_core_rx;
++		else if (in_core_rw(mod, val))
++			val -= (uint64_t) mod->module_core_rw;
+ 		break;
+ 
+ 	      case RV_LTV:
+@@ -820,15 +850,15 @@ apply_relocate_add (Elf64_Shdr *sechdrs, const char *strtab, unsigned int symind
+ 		 *     addresses have been selected...
+ 		 */
+ 		uint64_t gp;
+-		if (mod->core_size > MAX_LTOFF)
++		if (mod->core_size_rx + mod->core_size_rw > MAX_LTOFF)
+ 			/*
+ 			 * This takes advantage of fact that SHF_ARCH_SMALL gets allocated
+ 			 * at the end of the module.
+ 			 */
+-			gp = mod->core_size - MAX_LTOFF / 2;
++			gp = mod->core_size_rx + mod->core_size_rw - MAX_LTOFF / 2;
+ 		else
+-			gp = mod->core_size / 2;
+-		gp = (uint64_t) mod->module_core + ((gp + 7) & -8);
++			gp = (mod->core_size_rx + mod->core_size_rw) / 2;
++		gp = (uint64_t) mod->module_core_rx + ((gp + 7) & -8);
+ 		mod->arch.gp = gp;
+ 		DEBUGP("%s: placing gp at 0x%lx\n", __func__, gp);
+ 	}
+diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c
+index 609d500..7dde2a8 100644
+--- a/arch/ia64/kernel/sys_ia64.c
++++ b/arch/ia64/kernel/sys_ia64.c
+@@ -43,6 +43,13 @@ arch_get_unmapped_area (struct file *filp, unsigned long addr, unsigned long len
+ 	if (REGION_NUMBER(addr) == RGN_HPAGE)
+ 		addr = 0;
+ #endif
++
++#ifdef CONFIG_PAX_RANDMMAP
++	if (mm->pax_flags & MF_PAX_RANDMMAP)
++		addr = mm->free_area_cache;
++	else
++#endif
++
+ 	if (!addr)
+ 		addr = mm->free_area_cache;
+ 
+@@ -61,14 +68,14 @@ arch_get_unmapped_area (struct file *filp, unsigned long addr, unsigned long len
+ 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
+ 		/* At this point:  (!vma || addr < vma->vm_end). */
+ 		if (TASK_SIZE - len < addr || RGN_MAP_LIMIT - len < REGION_OFFSET(addr)) {
+-			if (start_addr != TASK_UNMAPPED_BASE) {
++			if (start_addr != mm->mmap_base) {
+ 				/* Start a new search --- just in case we missed some holes.  */
+-				addr = TASK_UNMAPPED_BASE;
++				addr = mm->mmap_base;
+ 				goto full_search;
+ 			}
+ 			return -ENOMEM;
+ 		}
+-		if (!vma || addr + len <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr, len)) {
+ 			/* Remember the address where we stopped this search:  */
+ 			mm->free_area_cache = addr + len;
+ 			return addr;
+diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
+index 53c0ba0..2accdde 100644
+--- a/arch/ia64/kernel/vmlinux.lds.S
++++ b/arch/ia64/kernel/vmlinux.lds.S
+@@ -199,7 +199,7 @@ SECTIONS {
+ 	/* Per-cpu data: */
+ 	. = ALIGN(PERCPU_PAGE_SIZE);
+ 	PERCPU_VADDR(SMP_CACHE_BYTES, PERCPU_ADDR, :percpu)
+-	__phys_per_cpu_start = __per_cpu_load;
++	__phys_per_cpu_start = per_cpu_load;
+ 	/*
+ 	 * ensure percpu data fits
+ 	 * into percpu page size
+diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c
+index 20b3593..1ce77f0 100644
+--- a/arch/ia64/mm/fault.c
++++ b/arch/ia64/mm/fault.c
+@@ -73,6 +73,23 @@ mapped_kernel_page_is_present (unsigned long address)
+ 	return pte_present(pte);
+ }
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 8; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ void __kprobes
+ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *regs)
+ {
+@@ -146,9 +163,23 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re
+ 	mask = (  (((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT)
+ 		| (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT));
+ 
+-	if ((vma->vm_flags & mask) != mask)
++	if ((vma->vm_flags & mask) != mask) {
++
++#ifdef CONFIG_PAX_PAGEEXEC
++		if (!(vma->vm_flags & VM_EXEC) && (mask & VM_EXEC)) {
++			if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->cr_iip)
++				goto bad_area;
++
++			up_read(&mm->mmap_sem);
++			pax_report_fault(regs, (void *)regs->cr_iip, (void *)regs->r12);
++			do_group_exit(SIGKILL);
++		}
++#endif
++
+ 		goto bad_area;
+ 
++	}
++
+ 	/*
+ 	 * If for any reason at all we couldn't handle the fault, make
+ 	 * sure we exit gracefully rather than endlessly redo the
+diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
+index 5ca674b..e0e1b70 100644
+--- a/arch/ia64/mm/hugetlbpage.c
++++ b/arch/ia64/mm/hugetlbpage.c
+@@ -171,7 +171,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, u
+ 		/* At this point:  (!vmm || addr < vmm->vm_end). */
+ 		if (REGION_OFFSET(addr) + len > RGN_MAP_LIMIT)
+ 			return -ENOMEM;
+-		if (!vmm || (addr + len) <= vmm->vm_start)
++		if (check_heap_stack_gap(vmm, addr, len))
+ 			return addr;
+ 		addr = ALIGN(vmm->vm_end, HPAGE_SIZE);
+ 	}
+diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
+index 00cb0e2..2ad8024 100644
+--- a/arch/ia64/mm/init.c
++++ b/arch/ia64/mm/init.c
+@@ -120,6 +120,19 @@ ia64_init_addr_space (void)
+ 		vma->vm_start = current->thread.rbs_bot & PAGE_MASK;
+ 		vma->vm_end = vma->vm_start + PAGE_SIZE;
+ 		vma->vm_flags = VM_DATA_DEFAULT_FLAGS|VM_GROWSUP|VM_ACCOUNT;
++
++#ifdef CONFIG_PAX_PAGEEXEC
++		if (current->mm->pax_flags & MF_PAX_PAGEEXEC) {
++			vma->vm_flags &= ~VM_EXEC;
++
++#ifdef CONFIG_PAX_MPROTECT
++			if (current->mm->pax_flags & MF_PAX_MPROTECT)
++				vma->vm_flags &= ~VM_MAYEXEC;
++#endif
++
++		}
++#endif
++
+ 		vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
+ 		down_write(&current->mm->mmap_sem);
+ 		if (insert_vm_struct(current->mm, vma)) {
+diff --git a/arch/m32r/include/asm/cache.h b/arch/m32r/include/asm/cache.h
+index 40b3ee9..8c2c112 100644
+--- a/arch/m32r/include/asm/cache.h
++++ b/arch/m32r/include/asm/cache.h
+@@ -1,8 +1,10 @@
+ #ifndef _ASM_M32R_CACHE_H
+ #define _ASM_M32R_CACHE_H
+ 
++#include <linux/const.h>
++
+ /* L1 cache line size */
+ #define L1_CACHE_SHIFT		4
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #endif  /* _ASM_M32R_CACHE_H */
+diff --git a/arch/m32r/lib/usercopy.c b/arch/m32r/lib/usercopy.c
+index 82abd15..d95ae5d 100644
+--- a/arch/m32r/lib/usercopy.c
++++ b/arch/m32r/lib/usercopy.c
+@@ -14,6 +14,9 @@
+ unsigned long
+ __generic_copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	prefetch(from);
+ 	if (access_ok(VERIFY_WRITE, to, n))
+ 		__copy_user(to,from,n);
+@@ -23,6 +26,9 @@ __generic_copy_to_user(void __user *to, const void *from, unsigned long n)
+ unsigned long
+ __generic_copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	prefetchw(to);
+ 	if (access_ok(VERIFY_READ, from, n))
+ 		__copy_user_zeroing(to,from,n);
+diff --git a/arch/m68k/include/asm/cache.h b/arch/m68k/include/asm/cache.h
+index 0395c51..5f26031 100644
+--- a/arch/m68k/include/asm/cache.h
++++ b/arch/m68k/include/asm/cache.h
+@@ -4,9 +4,11 @@
+ #ifndef __ARCH_M68K_CACHE_H
+ #define __ARCH_M68K_CACHE_H
+ 
++#include <linux/const.h>
++
+ /* bytes per L1 cache line */
+ #define        L1_CACHE_SHIFT  4
+-#define        L1_CACHE_BYTES  (1<< L1_CACHE_SHIFT)
++#define        L1_CACHE_BYTES  (_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define ARCH_DMA_MINALIGN	L1_CACHE_BYTES
+ 
+diff --git a/arch/microblaze/include/asm/cache.h b/arch/microblaze/include/asm/cache.h
+index 4efe96a..60e8699 100644
+--- a/arch/microblaze/include/asm/cache.h
++++ b/arch/microblaze/include/asm/cache.h
+@@ -13,11 +13,12 @@
+ #ifndef _ASM_MICROBLAZE_CACHE_H
+ #define _ASM_MICROBLAZE_CACHE_H
+ 
++#include <linux/const.h>
+ #include <asm/registers.h>
+ 
+ #define L1_CACHE_SHIFT 5
+ /* word-granular cache in microblaze */
+-#define L1_CACHE_BYTES	(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES	(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define SMP_CACHE_BYTES	L1_CACHE_BYTES
+ 
+diff --git a/arch/mips/include/asm/atomic.h b/arch/mips/include/asm/atomic.h
+index 1d93f81..67794d0 100644
+--- a/arch/mips/include/asm/atomic.h
++++ b/arch/mips/include/asm/atomic.h
+@@ -21,6 +21,10 @@
+ #include <asm/war.h>
+ #include <asm/system.h>
+ 
++#ifdef CONFIG_GENERIC_ATOMIC64
++#include <asm-generic/atomic64.h>
++#endif
++
+ #define ATOMIC_INIT(i)    { (i) }
+ 
+ /*
+@@ -765,6 +769,16 @@ static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
+  */
+ #define atomic64_add_negative(i, v) (atomic64_add_return(i, (v)) < 0)
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ #endif /* CONFIG_64BIT */
+ 
+ /*
+diff --git a/arch/mips/include/asm/cache.h b/arch/mips/include/asm/cache.h
+index b4db69f..8f3b093 100644
+--- a/arch/mips/include/asm/cache.h
++++ b/arch/mips/include/asm/cache.h
+@@ -9,10 +9,11 @@
+ #ifndef _ASM_CACHE_H
+ #define _ASM_CACHE_H
+ 
++#include <linux/const.h>
+ #include <kmalloc.h>
+ 
+ #define L1_CACHE_SHIFT		CONFIG_MIPS_L1_CACHE_SHIFT
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define SMP_CACHE_SHIFT		L1_CACHE_SHIFT
+ #define SMP_CACHE_BYTES		L1_CACHE_BYTES
+diff --git a/arch/mips/include/asm/elf.h b/arch/mips/include/asm/elf.h
+index 455c0ac..ad65fbe 100644
+--- a/arch/mips/include/asm/elf.h
++++ b/arch/mips/include/asm/elf.h
+@@ -372,13 +372,16 @@ extern const char *__elf_platform;
+ #define ELF_ET_DYN_BASE         (TASK_SIZE / 3 * 2)
+ #endif
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(TASK_IS_32BIT_ADDR ? 0x00400000UL : 0x00400000UL)
++
++#define PAX_DELTA_MMAP_LEN	(TASK_IS_32BIT_ADDR ? 27-PAGE_SHIFT : 36-PAGE_SHIFT)
++#define PAX_DELTA_STACK_LEN	(TASK_IS_32BIT_ADDR ? 27-PAGE_SHIFT : 36-PAGE_SHIFT)
++#endif
++
+ #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
+ struct linux_binprm;
+ extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+ 				       int uses_interp);
+ 
+-struct mm_struct;
+-extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+-#define arch_randomize_brk arch_randomize_brk
+-
+ #endif /* _ASM_ELF_H */
+diff --git a/arch/mips/include/asm/page.h b/arch/mips/include/asm/page.h
+index e59cd1a..8e329d6 100644
+--- a/arch/mips/include/asm/page.h
++++ b/arch/mips/include/asm/page.h
+@@ -93,7 +93,7 @@ extern void copy_user_highpage(struct page *to, struct page *from,
+   #ifdef CONFIG_CPU_MIPS32
+     typedef struct { unsigned long pte_low, pte_high; } pte_t;
+     #define pte_val(x)    ((x).pte_low | ((unsigned long long)(x).pte_high << 32))
+-    #define __pte(x)      ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; })
++    #define __pte(x)      ({ pte_t __pte = {(x), (x) >> 32}; __pte; })
+   #else
+      typedef struct { unsigned long long pte; } pte_t;
+      #define pte_val(x)	((x).pte)
+diff --git a/arch/mips/include/asm/system.h b/arch/mips/include/asm/system.h
+index 6018c80..7c37203 100644
+--- a/arch/mips/include/asm/system.h
++++ b/arch/mips/include/asm/system.h
+@@ -230,6 +230,6 @@ extern void per_cpu_trap_init(void);
+  */
+ #define __ARCH_WANT_UNLOCKED_CTXSW
+ 
+-extern unsigned long arch_align_stack(unsigned long sp);
++#define arch_align_stack(x) ((x) & ~0xfUL)
+ 
+ #endif /* _ASM_SYSTEM_H */
+diff --git a/arch/mips/kernel/binfmt_elfn32.c b/arch/mips/kernel/binfmt_elfn32.c
+index 9fdd8bc..4bd7f1a 100644
+--- a/arch/mips/kernel/binfmt_elfn32.c
++++ b/arch/mips/kernel/binfmt_elfn32.c
+@@ -50,6 +50,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+ #undef ELF_ET_DYN_BASE
+ #define ELF_ET_DYN_BASE         (TASK32_SIZE / 3 * 2)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(TASK_IS_32BIT_ADDR ? 0x00400000UL : 0x00400000UL)
++
++#define PAX_DELTA_MMAP_LEN	(TASK_IS_32BIT_ADDR ? 27-PAGE_SHIFT : 36-PAGE_SHIFT)
++#define PAX_DELTA_STACK_LEN	(TASK_IS_32BIT_ADDR ? 27-PAGE_SHIFT : 36-PAGE_SHIFT)
++#endif
++
+ #include <asm/processor.h>
+ #include <linux/module.h>
+ #include <linux/elfcore.h>
+diff --git a/arch/mips/kernel/binfmt_elfo32.c b/arch/mips/kernel/binfmt_elfo32.c
+index ff44823..97f8906 100644
+--- a/arch/mips/kernel/binfmt_elfo32.c
++++ b/arch/mips/kernel/binfmt_elfo32.c
+@@ -52,6 +52,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+ #undef ELF_ET_DYN_BASE
+ #define ELF_ET_DYN_BASE         (TASK32_SIZE / 3 * 2)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(TASK_IS_32BIT_ADDR ? 0x00400000UL : 0x00400000UL)
++
++#define PAX_DELTA_MMAP_LEN	(TASK_IS_32BIT_ADDR ? 27-PAGE_SHIFT : 36-PAGE_SHIFT)
++#define PAX_DELTA_STACK_LEN	(TASK_IS_32BIT_ADDR ? 27-PAGE_SHIFT : 36-PAGE_SHIFT)
++#endif
++
+ #include <asm/processor.h>
+ 
+ /*
+diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
+index c47f96e..661d418 100644
+--- a/arch/mips/kernel/process.c
++++ b/arch/mips/kernel/process.c
+@@ -481,15 +481,3 @@ unsigned long get_wchan(struct task_struct *task)
+ out:
+ 	return pc;
+ }
+-
+-/*
+- * Don't forget that the stack pointer must be aligned on a 8 bytes
+- * boundary for 32-bits ABI and 16 bytes for 64-bits ABI.
+- */
+-unsigned long arch_align_stack(unsigned long sp)
+-{
+-	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+-		sp -= get_random_int() & ~PAGE_MASK;
+-
+-	return sp & ALMASK;
+-}
+diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c
+index 937cf33..adb39bb 100644
+--- a/arch/mips/mm/fault.c
++++ b/arch/mips/mm/fault.c
+@@ -28,6 +28,23 @@
+ #include <asm/highmem.h>		/* For VMALLOC_END */
+ #include <linux/kdebug.h>
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 5; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ /*
+  * This routine handles page faults.  It determines the address,
+  * and the problem, and then passes it off to one of the appropriate
+diff --git a/arch/mips/mm/mmap.c b/arch/mips/mm/mmap.c
+index 302d779..7d35bf8 100644
+--- a/arch/mips/mm/mmap.c
++++ b/arch/mips/mm/mmap.c
+@@ -95,6 +95,11 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp,
+ 		do_color_align = 1;
+ 
+ 	/* requesting a specific address */
++
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(current->mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
+ 	if (addr) {
+ 		if (do_color_align)
+ 			addr = COLOUR_ALIGN(addr, pgoff);
+@@ -102,8 +107,7 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp,
+ 			addr = PAGE_ALIGN(addr);
+ 
+ 		vma = find_vma(mm, addr);
+-		if (TASK_SIZE - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (TASK_SIZE - len >= addr && check_heap_stack_gap(vmm, addr, len))
+ 			return addr;
+ 	}
+ 
+@@ -118,7 +122,7 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp,
+ 			/* At this point:  (!vma || addr < vma->vm_end). */
+ 			if (TASK_SIZE - len < addr)
+ 				return -ENOMEM;
+-			if (!vma || addr + len <= vma->vm_start)
++			if (check_heap_stack_gap(vmm, addr, len))
+ 				return addr;
+ 			addr = vma->vm_end;
+ 			if (do_color_align)
+@@ -145,7 +149,7 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp,
+ 		/* make sure it can fit in the remaining address space */
+ 		if (likely(addr > len)) {
+ 			vma = find_vma(mm, addr - len);
+-			if (!vma || addr <= vma->vm_start) {
++			if (check_heap_stack_gap(vmm, addr - len, len))
+ 				/* cache the address as a hint for next time */
+ 				return mm->free_area_cache = addr - len;
+ 			}
+@@ -165,7 +169,7 @@ static unsigned long arch_get_unmapped_area_common(struct file *filp,
+ 			 * return with success:
+ 			 */
+ 			vma = find_vma(mm, addr);
+-			if (likely(!vma || addr + len <= vma->vm_start)) {
++			if (check_heap_stack_gap(vmm, addr, len)) {
+ 				/* cache the address as a hint for next time */
+ 				return mm->free_area_cache = addr;
+ 			}
+@@ -242,30 +246,3 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
+ 		mm->unmap_area = arch_unmap_area_topdown;
+ 	}
+ }
+-
+-static inline unsigned long brk_rnd(void)
+-{
+-	unsigned long rnd = get_random_int();
+-
+-	rnd = rnd << PAGE_SHIFT;
+-	/* 8MB for 32bit, 256MB for 64bit */
+-	if (TASK_IS_32BIT_ADDR)
+-		rnd = rnd & 0x7ffffful;
+-	else
+-		rnd = rnd & 0xffffffful;
+-
+-	return rnd;
+-}
+-
+-unsigned long arch_randomize_brk(struct mm_struct *mm)
+-{
+-	unsigned long base = mm->brk;
+-	unsigned long ret;
+-
+-	ret = PAGE_ALIGN(base + brk_rnd());
+-
+-	if (ret < mm->brk)
+-		return mm->brk;
+-
+-	return ret;
+-}
+diff --git a/arch/mn10300/proc-mn103e010/include/proc/cache.h b/arch/mn10300/proc-mn103e010/include/proc/cache.h
+index 967d144..db12197 100644
+--- a/arch/mn10300/proc-mn103e010/include/proc/cache.h
++++ b/arch/mn10300/proc-mn103e010/include/proc/cache.h
+@@ -11,12 +11,14 @@
+ #ifndef _ASM_PROC_CACHE_H
+ #define _ASM_PROC_CACHE_H
+ 
++#include <linux/const.h>
++
+ /* L1 cache */
+ 
+ #define L1_CACHE_NWAYS		4	/* number of ways in caches */
+ #define L1_CACHE_NENTRIES	256	/* number of entries in each way */
+-#define L1_CACHE_BYTES		16	/* bytes per entry */
+ #define L1_CACHE_SHIFT		4	/* shift for bytes per entry */
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)	/* bytes per entry */
+ #define L1_CACHE_WAYDISP	0x1000	/* displacement of one way from the next */
+ 
+ #define L1_CACHE_TAG_VALID	0x00000001	/* cache tag valid bit */
+diff --git a/arch/mn10300/proc-mn2ws0050/include/proc/cache.h b/arch/mn10300/proc-mn2ws0050/include/proc/cache.h
+index bcb5df2..84fabd2 100644
+--- a/arch/mn10300/proc-mn2ws0050/include/proc/cache.h
++++ b/arch/mn10300/proc-mn2ws0050/include/proc/cache.h
+@@ -16,13 +16,15 @@
+ #ifndef _ASM_PROC_CACHE_H
+ #define _ASM_PROC_CACHE_H
+ 
++#include <linux/const.h>
++
+ /*
+  * L1 cache
+  */
+ #define L1_CACHE_NWAYS		4		/* number of ways in caches */
+ #define L1_CACHE_NENTRIES	128		/* number of entries in each way */
+-#define L1_CACHE_BYTES		32		/* bytes per entry */
+ #define L1_CACHE_SHIFT		5		/* shift for bytes per entry */
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)	/* bytes per entry */
+ #define L1_CACHE_WAYDISP	0x1000		/* distance from one way to the next */
+ 
+ #define L1_CACHE_TAG_VALID	0x00000001	/* cache tag valid bit */
+diff --git a/arch/openrisc/include/asm/cache.h b/arch/openrisc/include/asm/cache.h
+index 4ce7a01..449202a 100644
+--- a/arch/openrisc/include/asm/cache.h
++++ b/arch/openrisc/include/asm/cache.h
+@@ -19,11 +19,13 @@
+ #ifndef __ASM_OPENRISC_CACHE_H
+ #define __ASM_OPENRISC_CACHE_H
+ 
++#include <linux/const.h>
++
+ /* FIXME: How can we replace these with values from the CPU...
+  * they shouldn't be hard-coded!
+  */
+ 
+-#define L1_CACHE_BYTES 16
+ #define L1_CACHE_SHIFT 4
++#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #endif /* __ASM_OPENRISC_CACHE_H */
+diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h
+index 4054b31..a10c105 100644
+--- a/arch/parisc/include/asm/atomic.h
++++ b/arch/parisc/include/asm/atomic.h
+@@ -335,6 +335,16 @@ static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
+ 
+ #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ #endif /* !CONFIG_64BIT */
+ 
+ 
+diff --git a/arch/parisc/include/asm/cache.h b/arch/parisc/include/asm/cache.h
+index 47f11c7..3420df2 100644
+--- a/arch/parisc/include/asm/cache.h
++++ b/arch/parisc/include/asm/cache.h
+@@ -5,6 +5,7 @@
+ #ifndef __ARCH_PARISC_CACHE_H
+ #define __ARCH_PARISC_CACHE_H
+ 
++#include <linux/const.h>
+ 
+ /*
+  * PA 2.0 processors have 64-byte cachelines; PA 1.1 processors have
+@@ -15,13 +16,13 @@
+  * just ruin performance.
+  */
+ #ifdef CONFIG_PA20
+-#define L1_CACHE_BYTES 64
+ #define L1_CACHE_SHIFT 6
+ #else
+-#define L1_CACHE_BYTES 32
+ #define L1_CACHE_SHIFT 5
+ #endif
+ 
++#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT)
++
+ #ifndef __ASSEMBLY__
+ 
+ #define SMP_CACHE_BYTES L1_CACHE_BYTES
+diff --git a/arch/parisc/include/asm/elf.h b/arch/parisc/include/asm/elf.h
+index 19f6cb1..6c78cf2 100644
+--- a/arch/parisc/include/asm/elf.h
++++ b/arch/parisc/include/asm/elf.h
+@@ -342,6 +342,13 @@ struct pt_regs;	/* forward declaration... */
+ 
+ #define ELF_ET_DYN_BASE         (TASK_UNMAPPED_BASE + 0x01000000)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	0x10000UL
++
++#define PAX_DELTA_MMAP_LEN	16
++#define PAX_DELTA_STACK_LEN	16
++#endif
++
+ /* This yields a mask that user programs can use to figure out what
+    instruction set this CPU supports.  This could be done in user space,
+    but it's not easy, and we've already done it here.  */
+diff --git a/arch/parisc/include/asm/pgtable.h b/arch/parisc/include/asm/pgtable.h
+index 22dadeb..f6c2be4 100644
+--- a/arch/parisc/include/asm/pgtable.h
++++ b/arch/parisc/include/asm/pgtable.h
+@@ -210,6 +210,17 @@ struct vm_area_struct;
+ #define PAGE_EXECREAD   __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_EXEC |_PAGE_ACCESSED)
+ #define PAGE_COPY       PAGE_EXECREAD
+ #define PAGE_RWX        __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_EXEC |_PAGE_ACCESSED)
++
++#ifdef CONFIG_PAX_PAGEEXEC
++# define PAGE_SHARED_NOEXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_ACCESSED)
++# define PAGE_COPY_NOEXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED)
++# define PAGE_READONLY_NOEXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED)
++#else
++# define PAGE_SHARED_NOEXEC	PAGE_SHARED
++# define PAGE_COPY_NOEXEC	PAGE_COPY
++# define PAGE_READONLY_NOEXEC	PAGE_READONLY
++#endif
++
+ #define PAGE_KERNEL	__pgprot(_PAGE_KERNEL)
+ #define PAGE_KERNEL_EXEC	__pgprot(_PAGE_KERNEL_EXEC)
+ #define PAGE_KERNEL_RWX	__pgprot(_PAGE_KERNEL_RWX)
+diff --git a/arch/parisc/kernel/module.c b/arch/parisc/kernel/module.c
+index 5e34ccf..672bc9c 100644
+--- a/arch/parisc/kernel/module.c
++++ b/arch/parisc/kernel/module.c
+@@ -98,16 +98,38 @@
+ 
+ /* three functions to determine where in the module core
+  * or init pieces the location is */
++static inline int in_init_rx(struct module *me, void *loc)
++{
++	return (loc >= me->module_init_rx &&
++		loc < (me->module_init_rx + me->init_size_rx));
++}
++
++static inline int in_init_rw(struct module *me, void *loc)
++{
++	return (loc >= me->module_init_rw &&
++		loc < (me->module_init_rw + me->init_size_rw));
++}
++
+ static inline int in_init(struct module *me, void *loc)
+ {
+-	return (loc >= me->module_init &&
+-		loc <= (me->module_init + me->init_size));
++	return in_init_rx(me, loc) || in_init_rw(me, loc);
++}
++
++static inline int in_core_rx(struct module *me, void *loc)
++{
++	return (loc >= me->module_core_rx &&
++		loc < (me->module_core_rx + me->core_size_rx));
++}
++
++static inline int in_core_rw(struct module *me, void *loc)
++{
++	return (loc >= me->module_core_rw &&
++		loc < (me->module_core_rw + me->core_size_rw));
+ }
+ 
+ static inline int in_core(struct module *me, void *loc)
+ {
+-	return (loc >= me->module_core &&
+-		loc <= (me->module_core + me->core_size));
++	return in_core_rx(me, loc) || in_core_rw(me, loc);
+ }
+ 
+ static inline int in_local(struct module *me, void *loc)
+@@ -373,13 +395,13 @@ int module_frob_arch_sections(CONST Elf_Ehdr *hdr,
+ 	}
+ 
+ 	/* align things a bit */
+-	me->core_size = ALIGN(me->core_size, 16);
+-	me->arch.got_offset = me->core_size;
+-	me->core_size += gots * sizeof(struct got_entry);
++	me->core_size_rw = ALIGN(me->core_size_rw, 16);
++	me->arch.got_offset = me->core_size_rw;
++	me->core_size_rw += gots * sizeof(struct got_entry);
+ 
+-	me->core_size = ALIGN(me->core_size, 16);
+-	me->arch.fdesc_offset = me->core_size;
+-	me->core_size += fdescs * sizeof(Elf_Fdesc);
++	me->core_size_rw = ALIGN(me->core_size_rw, 16);
++	me->arch.fdesc_offset = me->core_size_rw;
++	me->core_size_rw += fdescs * sizeof(Elf_Fdesc);
+ 
+ 	me->arch.got_max = gots;
+ 	me->arch.fdesc_max = fdescs;
+@@ -397,7 +419,7 @@ static Elf64_Word get_got(struct module *me, unsigned long value, long addend)
+ 
+ 	BUG_ON(value == 0);
+ 
+-	got = me->module_core + me->arch.got_offset;
++	got = me->module_core_rw + me->arch.got_offset;
+ 	for (i = 0; got[i].addr; i++)
+ 		if (got[i].addr == value)
+ 			goto out;
+@@ -415,7 +437,7 @@ static Elf64_Word get_got(struct module *me, unsigned long value, long addend)
+ #ifdef CONFIG_64BIT
+ static Elf_Addr get_fdesc(struct module *me, unsigned long value)
+ {
+-	Elf_Fdesc *fdesc = me->module_core + me->arch.fdesc_offset;
++	Elf_Fdesc *fdesc = me->module_core_rw + me->arch.fdesc_offset;
+ 
+ 	if (!value) {
+ 		printk(KERN_ERR "%s: zero OPD requested!\n", me->name);
+@@ -433,7 +455,7 @@ static Elf_Addr get_fdesc(struct module *me, unsigned long value)
+ 
+ 	/* Create new one */
+ 	fdesc->addr = value;
+-	fdesc->gp = (Elf_Addr)me->module_core + me->arch.got_offset;
++	fdesc->gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset;
+ 	return (Elf_Addr)fdesc;
+ }
+ #endif /* CONFIG_64BIT */
+@@ -845,7 +867,7 @@ register_unwind_table(struct module *me,
+ 
+ 	table = (unsigned char *)sechdrs[me->arch.unwind_section].sh_addr;
+ 	end = table + sechdrs[me->arch.unwind_section].sh_size;
+-	gp = (Elf_Addr)me->module_core + me->arch.got_offset;
++	gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset;
+ 
+ 	DEBUGP("register_unwind_table(), sect = %d at 0x%p - 0x%p (gp=0x%lx)\n",
+ 	       me->arch.unwind_section, table, end, gp);
+diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c
+index c9b9322..02d8940 100644
+--- a/arch/parisc/kernel/sys_parisc.c
++++ b/arch/parisc/kernel/sys_parisc.c
+@@ -43,7 +43,7 @@ static unsigned long get_unshared_area(unsigned long addr, unsigned long len)
+ 		/* At this point:  (!vma || addr < vma->vm_end). */
+ 		if (TASK_SIZE - len < addr)
+ 			return -ENOMEM;
+-		if (!vma || addr + len <= vma->vm_start)
++		if (check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 		addr = vma->vm_end;
+ 	}
+@@ -79,7 +79,7 @@ static unsigned long get_shared_area(struct address_space *mapping,
+ 		/* At this point:  (!vma || addr < vma->vm_end). */
+ 		if (TASK_SIZE - len < addr)
+ 			return -ENOMEM;
+-		if (!vma || addr + len <= vma->vm_start)
++		if (check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 		addr = DCACHE_ALIGN(vma->vm_end - offset) + offset;
+ 		if (addr < vma->vm_end) /* handle wraparound */
+@@ -98,7 +98,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 	if (flags & MAP_FIXED)
+ 		return addr;
+ 	if (!addr)
+-		addr = TASK_UNMAPPED_BASE;
++		addr = current->mm->mmap_base;
+ 
+ 	if (filp) {
+ 		addr = get_shared_area(filp->f_mapping, addr, len, pgoff);
+diff --git a/arch/parisc/kernel/traps.c b/arch/parisc/kernel/traps.c
+index f19e660..414fe24 100644
+--- a/arch/parisc/kernel/traps.c
++++ b/arch/parisc/kernel/traps.c
+@@ -733,9 +733,7 @@ void notrace handle_interruption(int code, struct pt_regs *regs)
+ 
+ 			down_read(&current->mm->mmap_sem);
+ 			vma = find_vma(current->mm,regs->iaoq[0]);
+-			if (vma && (regs->iaoq[0] >= vma->vm_start)
+-				&& (vma->vm_flags & VM_EXEC)) {
+-
++			if (vma && (regs->iaoq[0] >= vma->vm_start)) {
+ 				fault_address = regs->iaoq[0];
+ 				fault_space = regs->iasq[0];
+ 
+diff --git a/arch/parisc/mm/fault.c b/arch/parisc/mm/fault.c
+index 18162ce..94de376 100644
+--- a/arch/parisc/mm/fault.c
++++ b/arch/parisc/mm/fault.c
+@@ -15,6 +15,7 @@
+ #include <linux/sched.h>
+ #include <linux/interrupt.h>
+ #include <linux/module.h>
++#include <linux/unistd.h>
+ 
+ #include <asm/uaccess.h>
+ #include <asm/traps.h>
+@@ -52,7 +53,7 @@ DEFINE_PER_CPU(struct exception_data, exception_data);
+ static unsigned long
+ parisc_acctyp(unsigned long code, unsigned int inst)
+ {
+-	if (code == 6 || code == 16)
++	if (code == 6 || code == 7 || code == 16)
+ 	    return VM_EXEC;
+ 
+ 	switch (inst & 0xf0000000) {
+@@ -138,6 +139,116 @@ parisc_acctyp(unsigned long code, unsigned int inst)
+ 			}
+ #endif
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++/*
++ * PaX: decide what to do with offenders (instruction_pointer(regs) = fault address)
++ *
++ * returns 1 when task should be killed
++ *         2 when rt_sigreturn trampoline was detected
++ *         3 when unpatched PLT trampoline was detected
++ */
++static int pax_handle_fetch_fault(struct pt_regs *regs)
++{
++
++#ifdef CONFIG_PAX_EMUPLT
++	int err;
++
++	do { /* PaX: unpatched PLT emulation */
++		unsigned int bl, depwi;
++
++		err = get_user(bl, (unsigned int *)instruction_pointer(regs));
++		err |= get_user(depwi, (unsigned int *)(instruction_pointer(regs)+4));
++
++		if (err)
++			break;
++
++		if (bl == 0xEA9F1FDDU && depwi == 0xD6801C1EU) {
++			unsigned int ldw, bv, ldw2, addr = instruction_pointer(regs)-12;
++
++			err = get_user(ldw, (unsigned int *)addr);
++			err |= get_user(bv, (unsigned int *)(addr+4));
++			err |= get_user(ldw2, (unsigned int *)(addr+8));
++
++			if (err)
++				break;
++
++			if (ldw == 0x0E801096U &&
++			    bv == 0xEAC0C000U &&
++			    ldw2 == 0x0E881095U)
++			{
++				unsigned int resolver, map;
++
++				err = get_user(resolver, (unsigned int *)(instruction_pointer(regs)+8));
++				err |= get_user(map, (unsigned int *)(instruction_pointer(regs)+12));
++				if (err)
++					break;
++
++				regs->gr[20] = instruction_pointer(regs)+8;
++				regs->gr[21] = map;
++				regs->gr[22] = resolver;
++				regs->iaoq[0] = resolver | 3UL;
++				regs->iaoq[1] = regs->iaoq[0] + 4;
++				return 3;
++			}
++		}
++	} while (0);
++#endif
++
++#ifdef CONFIG_PAX_EMUTRAMP
++
++#ifndef CONFIG_PAX_EMUSIGRT
++	if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP))
++		return 1;
++#endif
++
++	do { /* PaX: rt_sigreturn emulation */
++		unsigned int ldi1, ldi2, bel, nop;
++
++		err = get_user(ldi1, (unsigned int *)instruction_pointer(regs));
++		err |= get_user(ldi2, (unsigned int *)(instruction_pointer(regs)+4));
++		err |= get_user(bel, (unsigned int *)(instruction_pointer(regs)+8));
++		err |= get_user(nop, (unsigned int *)(instruction_pointer(regs)+12));
++
++		if (err)
++			break;
++
++		if ((ldi1 == 0x34190000U || ldi1 == 0x34190002U) &&
++		    ldi2 == 0x3414015AU &&
++		    bel == 0xE4008200U &&
++		    nop == 0x08000240U)
++		{
++			regs->gr[25] = (ldi1 & 2) >> 1;
++			regs->gr[20] = __NR_rt_sigreturn;
++			regs->gr[31] = regs->iaoq[1] + 16;
++			regs->sr[0] = regs->iasq[1];
++			regs->iaoq[0] = 0x100UL;
++			regs->iaoq[1] = regs->iaoq[0] + 4;
++			regs->iasq[0] = regs->sr[2];
++			regs->iasq[1] = regs->sr[2];
++			return 2;
++		}
++	} while (0);
++#endif
++
++	return 1;
++}
++
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 5; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ int fixup_exception(struct pt_regs *regs)
+ {
+ 	const struct exception_table_entry *fix;
+@@ -192,8 +303,33 @@ good_area:
+ 
+ 	acc_type = parisc_acctyp(code,regs->iir);
+ 
+-	if ((vma->vm_flags & acc_type) != acc_type)
++	if ((vma->vm_flags & acc_type) != acc_type) {
++
++#ifdef CONFIG_PAX_PAGEEXEC
++		if ((mm->pax_flags & MF_PAX_PAGEEXEC) && (acc_type & VM_EXEC) &&
++		    (address & ~3UL) == instruction_pointer(regs))
++		{
++			up_read(&mm->mmap_sem);
++			switch (pax_handle_fetch_fault(regs)) {
++
++#ifdef CONFIG_PAX_EMUPLT
++			case 3:
++				return;
++#endif
++
++#ifdef CONFIG_PAX_EMUTRAMP
++			case 2:
++				return;
++#endif
++
++			}
++			pax_report_fault(regs, (void *)instruction_pointer(regs), (void *)regs->gr[30]);
++			do_group_exit(SIGKILL);
++		}
++#endif
++
+ 		goto bad_area;
++	}
+ 
+ 	/*
+ 	 * If for any reason at all we couldn't handle the fault, make
+diff --git a/arch/powerpc/include/asm/atomic.h b/arch/powerpc/include/asm/atomic.h
+index 02e41b5..ec6e26c 100644
+--- a/arch/powerpc/include/asm/atomic.h
++++ b/arch/powerpc/include/asm/atomic.h
+@@ -469,6 +469,16 @@ static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
+ 
+ #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ #endif /* __powerpc64__ */
+ 
+ #endif /* __KERNEL__ */
+diff --git a/arch/powerpc/include/asm/cache.h b/arch/powerpc/include/asm/cache.h
+index 4b50941..5605819 100644
+--- a/arch/powerpc/include/asm/cache.h
++++ b/arch/powerpc/include/asm/cache.h
+@@ -3,6 +3,7 @@
+ 
+ #ifdef __KERNEL__
+ 
++#include <linux/const.h>
+ 
+ /* bytes per L1 cache line */
+ #if defined(CONFIG_8xx) || defined(CONFIG_403GCX)
+@@ -22,7 +23,7 @@
+ #define L1_CACHE_SHIFT		7
+ #endif
+ 
+-#define	L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define	L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define	SMP_CACHE_BYTES		L1_CACHE_BYTES
+ 
+diff --git a/arch/powerpc/include/asm/elf.h b/arch/powerpc/include/asm/elf.h
+index 3bf9cca..e7457d0 100644
+--- a/arch/powerpc/include/asm/elf.h
++++ b/arch/powerpc/include/asm/elf.h
+@@ -178,8 +178,19 @@ typedef elf_fpreg_t elf_vsrreghalf_t32[ELF_NVSRHALFREG];
+    the loader.  We need to make sure that it is out of the way of the program
+    that it will "exec", and that there is sufficient room for the brk.  */
+ 
+-extern unsigned long randomize_et_dyn(unsigned long base);
+-#define ELF_ET_DYN_BASE		(randomize_et_dyn(0x20000000))
++#define ELF_ET_DYN_BASE		(0x20000000)
++
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(0x10000000UL)
++
++#ifdef __powerpc64__
++#define PAX_DELTA_MMAP_LEN	(is_32bit_task() ? 16 : 28)
++#define PAX_DELTA_STACK_LEN	(is_32bit_task() ? 16 : 28)
++#else
++#define PAX_DELTA_MMAP_LEN	15
++#define PAX_DELTA_STACK_LEN	15
++#endif
++#endif
+ 
+ /*
+  * Our registers are always unsigned longs, whether we're a 32 bit
+@@ -274,9 +285,6 @@ extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+ 	(0x7ff >> (PAGE_SHIFT - 12)) : \
+ 	(0x3ffff >> (PAGE_SHIFT - 12)))
+ 
+-extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+-#define arch_randomize_brk arch_randomize_brk
+-
+ #endif /* __KERNEL__ */
+ 
+ /*
+diff --git a/arch/powerpc/include/asm/kmap_types.h b/arch/powerpc/include/asm/kmap_types.h
+index bca8fdc..61e9580 100644
+--- a/arch/powerpc/include/asm/kmap_types.h
++++ b/arch/powerpc/include/asm/kmap_types.h
+@@ -27,6 +27,7 @@ enum km_type {
+ 	KM_PPC_SYNC_PAGE,
+ 	KM_PPC_SYNC_ICACHE,
+ 	KM_KDB,
++	KM_CLEARPAGE,
+ 	KM_TYPE_NR
+ };
+ 
+diff --git a/arch/powerpc/include/asm/mman.h b/arch/powerpc/include/asm/mman.h
+index d4a7f64..451de1c 100644
+--- a/arch/powerpc/include/asm/mman.h
++++ b/arch/powerpc/include/asm/mman.h
+@@ -44,7 +44,7 @@ static inline unsigned long arch_calc_vm_prot_bits(unsigned long prot)
+ }
+ #define arch_calc_vm_prot_bits(prot) arch_calc_vm_prot_bits(prot)
+ 
+-static inline pgprot_t arch_vm_get_page_prot(unsigned long vm_flags)
++static inline pgprot_t arch_vm_get_page_prot(vm_flags_t vm_flags)
+ {
+ 	return (vm_flags & VM_SAO) ? __pgprot(_PAGE_SAO) : __pgprot(0);
+ }
+diff --git a/arch/powerpc/include/asm/page.h b/arch/powerpc/include/asm/page.h
+index dd9c4fd..a2ced87 100644
+--- a/arch/powerpc/include/asm/page.h
++++ b/arch/powerpc/include/asm/page.h
+@@ -141,8 +141,9 @@ extern phys_addr_t kernstart_addr;
+  * and needs to be executable.  This means the whole heap ends
+  * up being executable.
+  */
+-#define VM_DATA_DEFAULT_FLAGS32	(VM_READ | VM_WRITE | VM_EXEC | \
+-				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
++#define VM_DATA_DEFAULT_FLAGS32 \
++	(((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \
++	 VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+ 
+ #define VM_DATA_DEFAULT_FLAGS64	(VM_READ | VM_WRITE | \
+ 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+@@ -170,6 +171,9 @@ extern phys_addr_t kernstart_addr;
+ #define is_kernel_addr(x)	((x) >= PAGE_OFFSET)
+ #endif
+ 
++#define ktla_ktva(addr)		(addr)
++#define ktva_ktla(addr)		(addr)
++
+ /*
+  * Use the top bit of the higher-level page table entries to indicate whether
+  * the entries we point to contain hugepages.  This works because we know that
+diff --git a/arch/powerpc/include/asm/page_64.h b/arch/powerpc/include/asm/page_64.h
+index fb40ede..d3ce956 100644
+--- a/arch/powerpc/include/asm/page_64.h
++++ b/arch/powerpc/include/asm/page_64.h
+@@ -144,15 +144,18 @@ do {						\
+  * stack by default, so in the absence of a PT_GNU_STACK program header
+  * we turn execute permission off.
+  */
+-#define VM_STACK_DEFAULT_FLAGS32	(VM_READ | VM_WRITE | VM_EXEC | \
+-					 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
++#define VM_STACK_DEFAULT_FLAGS32 \
++	(((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \
++	 VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+ 
+ #define VM_STACK_DEFAULT_FLAGS64	(VM_READ | VM_WRITE | \
+ 					 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+ 
++#ifndef CONFIG_PAX_PAGEEXEC
+ #define VM_STACK_DEFAULT_FLAGS \
+ 	(is_32bit_task() ? \
+ 	 VM_STACK_DEFAULT_FLAGS32 : VM_STACK_DEFAULT_FLAGS64)
++#endif
+ 
+ #include <asm-generic/getorder.h>
+ 
+diff --git a/arch/powerpc/include/asm/pgtable.h b/arch/powerpc/include/asm/pgtable.h
+index 88b0bd9..e32bc67 100644
+--- a/arch/powerpc/include/asm/pgtable.h
++++ b/arch/powerpc/include/asm/pgtable.h
+@@ -2,6 +2,7 @@
+ #define _ASM_POWERPC_PGTABLE_H
+ #ifdef __KERNEL__
+ 
++#include <linux/const.h>
+ #ifndef __ASSEMBLY__
+ #include <asm/processor.h>		/* For TASK_SIZE */
+ #include <asm/mmu.h>
+diff --git a/arch/powerpc/include/asm/pte-hash32.h b/arch/powerpc/include/asm/pte-hash32.h
+index 4aad413..85d86bf 100644
+--- a/arch/powerpc/include/asm/pte-hash32.h
++++ b/arch/powerpc/include/asm/pte-hash32.h
+@@ -21,6 +21,7 @@
+ #define _PAGE_FILE	0x004	/* when !present: nonlinear file mapping */
+ #define _PAGE_USER	0x004	/* usermode access allowed */
+ #define _PAGE_GUARDED	0x008	/* G: prohibit speculative access */
++#define _PAGE_EXEC	_PAGE_GUARDED
+ #define _PAGE_COHERENT	0x010	/* M: enforce memory coherence (SMP systems) */
+ #define _PAGE_NO_CACHE	0x020	/* I: cache inhibit */
+ #define _PAGE_WRITETHRU	0x040	/* W: cache write-through */
+diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h
+index 559da19..7e5835c 100644
+--- a/arch/powerpc/include/asm/reg.h
++++ b/arch/powerpc/include/asm/reg.h
+@@ -212,6 +212,7 @@
+ #define SPRN_DBCR	0x136	/* e300 Data Breakpoint Control Reg */
+ #define SPRN_DSISR	0x012	/* Data Storage Interrupt Status Register */
+ #define   DSISR_NOHPTE		0x40000000	/* no translation found */
++#define   DSISR_GUARDED		0x10000000	/* fetch from guarded storage */
+ #define   DSISR_PROTFAULT	0x08000000	/* protection fault */
+ #define   DSISR_ISSTORE		0x02000000	/* access was a store */
+ #define   DSISR_DABRMATCH	0x00400000	/* hit data breakpoint */
+diff --git a/arch/powerpc/include/asm/system.h b/arch/powerpc/include/asm/system.h
+index e30a13d..2b7d994 100644
+--- a/arch/powerpc/include/asm/system.h
++++ b/arch/powerpc/include/asm/system.h
+@@ -530,7 +530,7 @@ __cmpxchg_local(volatile void *ptr, unsigned long old, unsigned long new,
+ #define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n))
+ #endif
+ 
+-extern unsigned long arch_align_stack(unsigned long sp);
++#define arch_align_stack(x) ((x) & ~0xfUL)
+ 
+ /* Used in very early kernel initialization. */
+ extern unsigned long reloc_offset(void);
+diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h
+index bd0fb84..a42a14b 100644
+--- a/arch/powerpc/include/asm/uaccess.h
++++ b/arch/powerpc/include/asm/uaccess.h
+@@ -13,6 +13,8 @@
+ #define VERIFY_READ	0
+ #define VERIFY_WRITE	1
+ 
++extern void check_object_size(const void *ptr, unsigned long n, bool to);
++
+ /*
+  * The fs value determines whether argument validity checking should be
+  * performed or not.  If get_fs() == USER_DS, checking is performed, with
+@@ -327,52 +329,6 @@ do {								\
+ extern unsigned long __copy_tofrom_user(void __user *to,
+ 		const void __user *from, unsigned long size);
+ 
+-#ifndef __powerpc64__
+-
+-static inline unsigned long copy_from_user(void *to,
+-		const void __user *from, unsigned long n)
+-{
+-	unsigned long over;
+-
+-	if (access_ok(VERIFY_READ, from, n))
+-		return __copy_tofrom_user((__force void __user *)to, from, n);
+-	if ((unsigned long)from < TASK_SIZE) {
+-		over = (unsigned long)from + n - TASK_SIZE;
+-		return __copy_tofrom_user((__force void __user *)to, from,
+-				n - over) + over;
+-	}
+-	return n;
+-}
+-
+-static inline unsigned long copy_to_user(void __user *to,
+-		const void *from, unsigned long n)
+-{
+-	unsigned long over;
+-
+-	if (access_ok(VERIFY_WRITE, to, n))
+-		return __copy_tofrom_user(to, (__force void __user *)from, n);
+-	if ((unsigned long)to < TASK_SIZE) {
+-		over = (unsigned long)to + n - TASK_SIZE;
+-		return __copy_tofrom_user(to, (__force void __user *)from,
+-				n - over) + over;
+-	}
+-	return n;
+-}
+-
+-#else /* __powerpc64__ */
+-
+-#define __copy_in_user(to, from, size) \
+-	__copy_tofrom_user((to), (from), (size))
+-
+-extern unsigned long copy_from_user(void *to, const void __user *from,
+-				    unsigned long n);
+-extern unsigned long copy_to_user(void __user *to, const void *from,
+-				  unsigned long n);
+-extern unsigned long copy_in_user(void __user *to, const void __user *from,
+-				  unsigned long n);
+-
+-#endif /* __powerpc64__ */
+-
+ static inline unsigned long __copy_from_user_inatomic(void *to,
+ 		const void __user *from, unsigned long n)
+ {
+@@ -396,6 +352,10 @@ static inline unsigned long __copy_from_user_inatomic(void *to,
+ 		if (ret == 0)
+ 			return 0;
+ 	}
++
++	if (!__builtin_constant_p(n))
++		check_object_size(to, n, false);
++
+ 	return __copy_tofrom_user((__force void __user *)to, from, n);
+ }
+ 
+@@ -422,6 +382,10 @@ static inline unsigned long __copy_to_user_inatomic(void __user *to,
+ 		if (ret == 0)
+ 			return 0;
+ 	}
++
++	if (!__builtin_constant_p(n))
++		check_object_size(from, n, true);
++
+ 	return __copy_tofrom_user(to, (__force const void __user *)from, n);
+ }
+ 
+@@ -439,6 +403,92 @@ static inline unsigned long __copy_to_user(void __user *to,
+ 	return __copy_to_user_inatomic(to, from, size);
+ }
+ 
++#ifndef __powerpc64__
++
++static inline unsigned long __must_check copy_from_user(void *to,
++		const void __user *from, unsigned long n)
++{
++	unsigned long over;
++
++	if ((long)n < 0)
++		return n;
++
++	if (access_ok(VERIFY_READ, from, n)) {
++		if (!__builtin_constant_p(n))
++			check_object_size(to, n, false);
++		return __copy_tofrom_user((__force void __user *)to, from, n);
++	}
++	if ((unsigned long)from < TASK_SIZE) {
++		over = (unsigned long)from + n - TASK_SIZE;
++		if (!__builtin_constant_p(n - over))
++			check_object_size(to, n - over, false);
++		return __copy_tofrom_user((__force void __user *)to, from,
++				n - over) + over;
++	}
++	return n;
++}
++
++static inline unsigned long __must_check copy_to_user(void __user *to,
++		const void *from, unsigned long n)
++{
++	unsigned long over;
++
++	if ((long)n < 0)
++		return n;
++
++	if (access_ok(VERIFY_WRITE, to, n)) {
++		if (!__builtin_constant_p(n))
++			check_object_size(from, n, true);
++		return __copy_tofrom_user(to, (__force void __user *)from, n);
++	}
++	if ((unsigned long)to < TASK_SIZE) {
++		over = (unsigned long)to + n - TASK_SIZE;
++		if (!__builtin_constant_p(n))
++			check_object_size(from, n - over, true);
++		return __copy_tofrom_user(to, (__force void __user *)from,
++				n - over) + over;
++	}
++	return n;
++}
++
++#else /* __powerpc64__ */
++
++#define __copy_in_user(to, from, size) \
++	__copy_tofrom_user((to), (from), (size))
++
++static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n)
++{
++	if ((long)n < 0 || n > INT_MAX)
++		return n;
++
++	if (!__builtin_constant_p(n))
++		check_object_size(to, n, false);
++
++	if (likely(access_ok(VERIFY_READ, from, n)))
++		n = __copy_from_user(to, from, n);
++	else
++		memset(to, 0, n);
++	return n;
++}
++
++static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n)
++{
++	if ((long)n < 0 || n > INT_MAX)
++		return n;
++
++	if (likely(access_ok(VERIFY_WRITE, to, n))) {
++		if (!__builtin_constant_p(n))
++			check_object_size(from, n, true);
++		n = __copy_to_user(to, from, n);
++	}
++	return n;
++}
++
++extern unsigned long copy_in_user(void __user *to, const void __user *from,
++				  unsigned long n);
++
++#endif /* __powerpc64__ */
++
+ extern unsigned long __clear_user(void __user *addr, unsigned long size);
+ 
+ static inline unsigned long clear_user(void __user *addr, unsigned long size)
+diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
+index 429983c..7af363b 100644
+--- a/arch/powerpc/kernel/exceptions-64e.S
++++ b/arch/powerpc/kernel/exceptions-64e.S
+@@ -587,6 +587,7 @@ storage_fault_common:
+ 	std	r14,_DAR(r1)
+ 	std	r15,_DSISR(r1)
+ 	addi	r3,r1,STACK_FRAME_OVERHEAD
++	bl	.save_nvgprs
+ 	mr	r4,r14
+ 	mr	r5,r15
+ 	ld	r14,PACA_EXGEN+EX_R14(r13)
+@@ -596,8 +597,7 @@ storage_fault_common:
+ 	cmpdi	r3,0
+ 	bne-	1f
+ 	b	.ret_from_except_lite
+-1:	bl	.save_nvgprs
+-	mr	r5,r3
++1:	mr	r5,r3
+ 	addi	r3,r1,STACK_FRAME_OVERHEAD
+ 	ld	r4,_DAR(r1)
+ 	bl	.bad_page_fault
+diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
+index cf9c69b..ebc9640 100644
+--- a/arch/powerpc/kernel/exceptions-64s.S
++++ b/arch/powerpc/kernel/exceptions-64s.S
+@@ -1004,10 +1004,10 @@ handle_page_fault:
+ 11:	ld	r4,_DAR(r1)
+ 	ld	r5,_DSISR(r1)
+ 	addi	r3,r1,STACK_FRAME_OVERHEAD
++	bl	.save_nvgprs
+ 	bl	.do_page_fault
+ 	cmpdi	r3,0
+ 	beq+	13f
+-	bl	.save_nvgprs
+ 	mr	r5,r3
+ 	addi	r3,r1,STACK_FRAME_OVERHEAD
+ 	lwz	r4,_DAR(r1)
+diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
+index 745c1e7..59d97a6 100644
+--- a/arch/powerpc/kernel/irq.c
++++ b/arch/powerpc/kernel/irq.c
+@@ -547,9 +547,6 @@ struct irq_host *irq_alloc_host(struct device_node *of_node,
+ 	host->ops = ops;
+ 	host->of_node = of_node_get(of_node);
+ 
+-	if (host->ops->match == NULL)
+-		host->ops->match = default_irq_host_match;
+-
+ 	raw_spin_lock_irqsave(&irq_big_lock, flags);
+ 
+ 	/* If it's a legacy controller, check for duplicates and
+@@ -622,7 +619,12 @@ struct irq_host *irq_find_host(struct device_node *node)
+ 	 */
+ 	raw_spin_lock_irqsave(&irq_big_lock, flags);
+ 	list_for_each_entry(h, &irq_hosts, link)
+-		if (h->ops->match(h, node)) {
++		if (h->ops->match) {
++			if (h->ops->match(h, node)) {
++				found = h;
++				break;
++			}
++		} else if (default_irq_host_match(h, node)) {
+ 			found = h;
+ 			break;
+ 		}
+diff --git a/arch/powerpc/kernel/module_32.c b/arch/powerpc/kernel/module_32.c
+index 0b6d796..d760ddb 100644
+--- a/arch/powerpc/kernel/module_32.c
++++ b/arch/powerpc/kernel/module_32.c
+@@ -162,7 +162,7 @@ int module_frob_arch_sections(Elf32_Ehdr *hdr,
+ 			me->arch.core_plt_section = i;
+ 	}
+ 	if (!me->arch.core_plt_section || !me->arch.init_plt_section) {
+-		printk("Module doesn't contain .plt or .init.plt sections.\n");
++		printk("Module %s doesn't contain .plt or .init.plt sections.\n", me->name);
+ 		return -ENOEXEC;
+ 	}
+ 
+@@ -192,11 +192,16 @@ static uint32_t do_plt_call(void *location,
+ 
+ 	DEBUGP("Doing plt for call to 0x%x at 0x%x\n", val, (unsigned int)location);
+ 	/* Init, or core PLT? */
+-	if (location >= mod->module_core
+-	    && location < mod->module_core + mod->core_size)
++	if ((location >= mod->module_core_rx && location < mod->module_core_rx + mod->core_size_rx) ||
++	    (location >= mod->module_core_rw && location < mod->module_core_rw + mod->core_size_rw))
+ 		entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr;
+-	else
++	else if ((location >= mod->module_init_rx && location < mod->module_init_rx + mod->init_size_rx) ||
++		 (location >= mod->module_init_rw && location < mod->module_init_rw + mod->init_size_rw))
+ 		entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr;
++	else {
++		printk(KERN_ERR "%s: invalid R_PPC_REL24 entry found\n", mod->name);
++		return ~0UL;
++	}
+ 
+ 	/* Find this entry, or if that fails, the next avail. entry */
+ 	while (entry->jump[0]) {
+diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
+index 6457574..08b28d3 100644
+--- a/arch/powerpc/kernel/process.c
++++ b/arch/powerpc/kernel/process.c
+@@ -660,8 +660,8 @@ void show_regs(struct pt_regs * regs)
+ 	 * Lookup NIP late so we have the best change of getting the
+ 	 * above info out without failing
+ 	 */
+-	printk("NIP ["REG"] %pS\n", regs->nip, (void *)regs->nip);
+-	printk("LR ["REG"] %pS\n", regs->link, (void *)regs->link);
++	printk("NIP ["REG"] %pA\n", regs->nip, (void *)regs->nip);
++	printk("LR ["REG"] %pA\n", regs->link, (void *)regs->link);
+ #endif
+ 	show_stack(current, (unsigned long *) regs->gpr[1]);
+ 	if (!user_mode(regs))
+@@ -1165,10 +1165,10 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)
+ 		newsp = stack[0];
+ 		ip = stack[STACK_FRAME_LR_SAVE];
+ 		if (!firstframe || ip != lr) {
+-			printk("["REG"] ["REG"] %pS", sp, ip, (void *)ip);
++			printk("["REG"] ["REG"] %pA", sp, ip, (void *)ip);
+ #ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ 			if ((ip == rth || ip == mrth) && curr_frame >= 0) {
+-				printk(" (%pS)",
++				printk(" (%pA)",
+ 				       (void *)current->ret_stack[curr_frame].ret);
+ 				curr_frame--;
+ 			}
+@@ -1188,7 +1188,7 @@ void show_stack(struct task_struct *tsk, unsigned long *stack)
+ 			struct pt_regs *regs = (struct pt_regs *)
+ 				(sp + STACK_FRAME_OVERHEAD);
+ 			lr = regs->link;
+-			printk("--- Exception: %lx at %pS\n    LR = %pS\n",
++			printk("--- Exception: %lx at %pA\n    LR = %pA\n",
+ 			       regs->trap, (void *)regs->nip, (void *)lr);
+ 			firstframe = 1;
+ 		}
+@@ -1263,58 +1263,3 @@ void thread_info_cache_init(void)
+ }
+ 
+ #endif /* THREAD_SHIFT < PAGE_SHIFT */
+-
+-unsigned long arch_align_stack(unsigned long sp)
+-{
+-	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+-		sp -= get_random_int() & ~PAGE_MASK;
+-	return sp & ~0xf;
+-}
+-
+-static inline unsigned long brk_rnd(void)
+-{
+-        unsigned long rnd = 0;
+-
+-	/* 8MB for 32bit, 1GB for 64bit */
+-	if (is_32bit_task())
+-		rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT)));
+-	else
+-		rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT)));
+-
+-	return rnd << PAGE_SHIFT;
+-}
+-
+-unsigned long arch_randomize_brk(struct mm_struct *mm)
+-{
+-	unsigned long base = mm->brk;
+-	unsigned long ret;
+-
+-#ifdef CONFIG_PPC_STD_MMU_64
+-	/*
+-	 * If we are using 1TB segments and we are allowed to randomise
+-	 * the heap, we can put it above 1TB so it is backed by a 1TB
+-	 * segment. Otherwise the heap will be in the bottom 1TB
+-	 * which always uses 256MB segments and this may result in a
+-	 * performance penalty.
+-	 */
+-	if (!is_32bit_task() && (mmu_highuser_ssize == MMU_SEGSIZE_1T))
+-		base = max_t(unsigned long, mm->brk, 1UL << SID_SHIFT_1T);
+-#endif
+-
+-	ret = PAGE_ALIGN(base + brk_rnd());
+-
+-	if (ret < mm->brk)
+-		return mm->brk;
+-
+-	return ret;
+-}
+-
+-unsigned long randomize_et_dyn(unsigned long base)
+-{
+-	unsigned long ret = PAGE_ALIGN(base + brk_rnd());
+-
+-	if (ret < base)
+-		return base;
+-
+-	return ret;
+-}
+diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c
+index 836a5a1..27289a3 100644
+--- a/arch/powerpc/kernel/signal_32.c
++++ b/arch/powerpc/kernel/signal_32.c
+@@ -859,7 +859,7 @@ int handle_rt_signal32(unsigned long sig, struct k_sigaction *ka,
+ 	/* Save user registers on the stack */
+ 	frame = &rt_sf->uc.uc_mcontext;
+ 	addr = frame;
+-	if (vdso32_rt_sigtramp && current->mm->context.vdso_base) {
++	if (vdso32_rt_sigtramp && current->mm->context.vdso_base != ~0UL) {
+ 		if (save_user_regs(regs, frame, 0, 1))
+ 			goto badframe;
+ 		regs->link = current->mm->context.vdso_base + vdso32_rt_sigtramp;
+diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c
+index a50b5ec..547078a 100644
+--- a/arch/powerpc/kernel/signal_64.c
++++ b/arch/powerpc/kernel/signal_64.c
+@@ -429,7 +429,7 @@ int handle_rt_signal64(int signr, struct k_sigaction *ka, siginfo_t *info,
+ 	current->thread.fpscr.val = 0;
+ 
+ 	/* Set up to return from userspace. */
+-	if (vdso64_rt_sigtramp && current->mm->context.vdso_base) {
++	if (vdso64_rt_sigtramp && current->mm->context.vdso_base != ~0UL) {
+ 		regs->link = current->mm->context.vdso_base + vdso64_rt_sigtramp;
+ 	} else {
+ 		err |= setup_trampoline(__NR_rt_sigreturn, &frame->tramp[0]);
+diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
+index 5459d14..10f8070 100644
+--- a/arch/powerpc/kernel/traps.c
++++ b/arch/powerpc/kernel/traps.c
+@@ -98,6 +98,8 @@ static void pmac_backlight_unblank(void)
+ static inline void pmac_backlight_unblank(void) { }
+ #endif
+ 
++extern void gr_handle_kernel_exploit(void);
++
+ int die(const char *str, struct pt_regs *regs, long err)
+ {
+ 	static struct {
+@@ -171,6 +173,8 @@ int die(const char *str, struct pt_regs *regs, long err)
+ 	if (panic_on_oops)
+ 		panic("Fatal exception");
+ 
++	gr_handle_kernel_exploit();
++
+ 	oops_exit();
+ 	do_exit(err);
+ 
+diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c
+index 7d14bb6..1305601 100644
+--- a/arch/powerpc/kernel/vdso.c
++++ b/arch/powerpc/kernel/vdso.c
+@@ -35,6 +35,7 @@
+ #include <asm/firmware.h>
+ #include <asm/vdso.h>
+ #include <asm/vdso_datapage.h>
++#include <asm/mman.h>
+ 
+ #include "setup.h"
+ 
+@@ -219,7 +220,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
+ 	vdso_base = VDSO32_MBASE;
+ #endif
+ 
+-	current->mm->context.vdso_base = 0;
++	current->mm->context.vdso_base = ~0UL;
+ 
+ 	/* vDSO has a problem and was disabled, just don't "enable" it for the
+ 	 * process
+@@ -239,7 +240,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
+ 	vdso_base = get_unmapped_area(NULL, vdso_base,
+ 				      (vdso_pages << PAGE_SHIFT) +
+ 				      ((VDSO_ALIGNMENT - 1) & PAGE_MASK),
+-				      0, 0);
++				      0, MAP_PRIVATE | MAP_EXECUTABLE);
+ 	if (IS_ERR_VALUE(vdso_base)) {
+ 		rc = vdso_base;
+ 		goto fail_mmapsem;
+diff --git a/arch/powerpc/lib/usercopy_64.c b/arch/powerpc/lib/usercopy_64.c
+index 5eea6f3..5d10396 100644
+--- a/arch/powerpc/lib/usercopy_64.c
++++ b/arch/powerpc/lib/usercopy_64.c
+@@ -9,22 +9,6 @@
+ #include <linux/module.h>
+ #include <asm/uaccess.h>
+ 
+-unsigned long copy_from_user(void *to, const void __user *from, unsigned long n)
+-{
+-	if (likely(access_ok(VERIFY_READ, from, n)))
+-		n = __copy_from_user(to, from, n);
+-	else
+-		memset(to, 0, n);
+-	return n;
+-}
+-
+-unsigned long copy_to_user(void __user *to, const void *from, unsigned long n)
+-{
+-	if (likely(access_ok(VERIFY_WRITE, to, n)))
+-		n = __copy_to_user(to, from, n);
+-	return n;
+-}
+-
+ unsigned long copy_in_user(void __user *to, const void __user *from,
+ 			   unsigned long n)
+ {
+@@ -35,7 +19,5 @@ unsigned long copy_in_user(void __user *to, const void __user *from,
+ 	return n;
+ }
+ 
+-EXPORT_SYMBOL(copy_from_user);
+-EXPORT_SYMBOL(copy_to_user);
+ EXPORT_SYMBOL(copy_in_user);
+ 
+diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
+index 5efe8c9..db9ceef 100644
+--- a/arch/powerpc/mm/fault.c
++++ b/arch/powerpc/mm/fault.c
+@@ -32,6 +32,10 @@
+ #include <linux/perf_event.h>
+ #include <linux/magic.h>
+ #include <linux/ratelimit.h>
++#include <linux/slab.h>
++#include <linux/pagemap.h>
++#include <linux/compiler.h>
++#include <linux/unistd.h>
+ 
+ #include <asm/firmware.h>
+ #include <asm/page.h>
+@@ -43,6 +47,7 @@
+ #include <asm/tlbflush.h>
+ #include <asm/siginfo.h>
+ #include <mm/mmu_decl.h>
++#include <asm/ptrace.h>
+ 
+ #ifdef CONFIG_KPROBES
+ static inline int notify_page_fault(struct pt_regs *regs)
+@@ -66,6 +71,33 @@ static inline int notify_page_fault(struct pt_regs *regs)
+ }
+ #endif
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++/*
++ * PaX: decide what to do with offenders (regs->nip = fault address)
++ *
++ * returns 1 when task should be killed
++ */
++static int pax_handle_fetch_fault(struct pt_regs *regs)
++{
++	return 1;
++}
++
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 5; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int __user *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ /*
+  * Check whether the instruction at regs->nip is a store using
+  * an update addressing form which will update r1.
+@@ -136,7 +168,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
+ 	 * indicate errors in DSISR but can validly be set in SRR1.
+ 	 */
+ 	if (trap == 0x400)
+-		error_code &= 0x48200000;
++		error_code &= 0x58200000;
+ 	else
+ 		is_write = error_code & DSISR_ISSTORE;
+ #else
+@@ -259,7 +291,7 @@ good_area:
+          * "undefined".  Of those that can be set, this is the only
+          * one which seems bad.
+          */
+-	if (error_code & 0x10000000)
++	if (error_code & DSISR_GUARDED)
+                 /* Guarded storage error. */
+ 		goto bad_area;
+ #endif /* CONFIG_8xx */
+@@ -274,7 +306,7 @@ good_area:
+ 		 * processors use the same I/D cache coherency mechanism
+ 		 * as embedded.
+ 		 */
+-		if (error_code & DSISR_PROTFAULT)
++		if (error_code & (DSISR_PROTFAULT | DSISR_GUARDED))
+ 			goto bad_area;
+ #endif /* CONFIG_PPC_STD_MMU */
+ 
+@@ -343,6 +375,23 @@ bad_area:
+ bad_area_nosemaphore:
+ 	/* User mode accesses cause a SIGSEGV */
+ 	if (user_mode(regs)) {
++
++#ifdef CONFIG_PAX_PAGEEXEC
++		if (mm->pax_flags & MF_PAX_PAGEEXEC) {
++#ifdef CONFIG_PPC_STD_MMU
++			if (is_exec && (error_code & (DSISR_PROTFAULT | DSISR_GUARDED))) {
++#else
++			if (is_exec && regs->nip == address) {
++#endif
++				switch (pax_handle_fetch_fault(regs)) {
++				}
++
++				pax_report_fault(regs, (void *)regs->nip, (void *)regs->gpr[PT_R1]);
++				do_group_exit(SIGKILL);
++			}
++		}
++#endif
++
+ 		_exception(SIGSEGV, regs, code, address);
+ 		return 0;
+ 	}
+diff --git a/arch/powerpc/mm/mmap_64.c b/arch/powerpc/mm/mmap_64.c
+index 5a783d8..c23e14b 100644
+--- a/arch/powerpc/mm/mmap_64.c
++++ b/arch/powerpc/mm/mmap_64.c
+@@ -99,10 +99,22 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
+ 	 */
+ 	if (mmap_is_legacy()) {
+ 		mm->mmap_base = TASK_UNMAPPED_BASE;
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base += mm->delta_mmap;
++#endif
++
+ 		mm->get_unmapped_area = arch_get_unmapped_area;
+ 		mm->unmap_area = arch_unmap_area;
+ 	} else {
+ 		mm->mmap_base = mmap_base();
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base -= mm->delta_mmap + mm->delta_stack;
++#endif
++
+ 		mm->get_unmapped_area = arch_get_unmapped_area_topdown;
+ 		mm->unmap_area = arch_unmap_area_topdown;
+ 	}
+diff --git a/arch/powerpc/mm/slice.c b/arch/powerpc/mm/slice.c
+index 73709f7..6b90313 100644
+--- a/arch/powerpc/mm/slice.c
++++ b/arch/powerpc/mm/slice.c
+@@ -98,7 +98,7 @@ static int slice_area_is_free(struct mm_struct *mm, unsigned long addr,
+ 	if ((mm->task_size - len) < addr)
+ 		return 0;
+ 	vma = find_vma(mm, addr);
+-	return (!vma || (addr + len) <= vma->vm_start);
++	return check_heap_stack_gap(vma, addr, len);
+ }
+ 
+ static int slice_low_has_vma(struct mm_struct *mm, unsigned long slice)
+@@ -256,7 +256,7 @@ full_search:
+ 				addr = _ALIGN_UP(addr + 1,  1ul << SLICE_HIGH_SHIFT);
+ 			continue;
+ 		}
+-		if (!vma || addr + len <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr, len)) {
+ 			/*
+ 			 * Remember the place where we stopped the search:
+ 			 */
+@@ -313,10 +313,14 @@ static unsigned long slice_find_area_topdown(struct mm_struct *mm,
+ 		}
+ 	}
+ 
+-	addr = mm->mmap_base;
+-	while (addr > len) {
++	if (mm->mmap_base < len)
++		addr = -ENOMEM;
++	else
++		addr = mm->mmap_base - len;
++
++	while (!IS_ERR_VALUE(addr)) {
+ 		/* Go down by chunk size */
+-		addr = _ALIGN_DOWN(addr - len, 1ul << pshift);
++		addr = _ALIGN_DOWN(addr, 1ul << pshift);
+ 
+ 		/* Check for hit with different page size */
+ 		mask = slice_range_to_mask(addr, len);
+@@ -336,7 +340,7 @@ static unsigned long slice_find_area_topdown(struct mm_struct *mm,
+ 		 * return with success:
+ 		 */
+ 		vma = find_vma(mm, addr);
+-		if (!vma || (addr + len) <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr, len)) {
+ 			/* remember the address as a hint for next time */
+ 			if (use_cache)
+ 				mm->free_area_cache = addr;
+@@ -348,7 +352,7 @@ static unsigned long slice_find_area_topdown(struct mm_struct *mm,
+ 		        mm->cached_hole_size = vma->vm_start - addr;
+ 
+ 		/* try just below the current vma->vm_start */
+-		addr = vma->vm_start;
++		addr = skip_heap_stack_gap(vma, len);
+ 	}
+ 
+ 	/*
+@@ -426,6 +430,11 @@ unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len,
+ 	if (fixed && addr > (mm->task_size - len))
+ 		return -EINVAL;
+ 
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!fixed && (mm->pax_flags & MF_PAX_RANDMMAP))
++		addr = 0;
++#endif
++
+ 	/* If hint, make sure it matches our alignment restrictions */
+ 	if (!fixed && addr) {
+ 		addr = _ALIGN_UP(addr, 1ul << pshift);
+diff --git a/arch/s390/include/asm/atomic.h b/arch/s390/include/asm/atomic.h
+index 8517d2a..d2738d4 100644
+--- a/arch/s390/include/asm/atomic.h
++++ b/arch/s390/include/asm/atomic.h
+@@ -326,6 +326,16 @@ static inline long long atomic64_dec_if_positive(atomic64_t *v)
+ #define atomic64_dec_and_test(_v)	(atomic64_sub_return(1, _v) == 0)
+ #define atomic64_inc_not_zero(v)	atomic64_add_unless((v), 1, 0)
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ #define smp_mb__before_atomic_dec()	smp_mb()
+ #define smp_mb__after_atomic_dec()	smp_mb()
+ #define smp_mb__before_atomic_inc()	smp_mb()
+diff --git a/arch/s390/include/asm/cache.h b/arch/s390/include/asm/cache.h
+index 2a30d5a..5e5586f 100644
+--- a/arch/s390/include/asm/cache.h
++++ b/arch/s390/include/asm/cache.h
+@@ -11,8 +11,10 @@
+ #ifndef __ARCH_S390_CACHE_H
+ #define __ARCH_S390_CACHE_H
+ 
+-#define L1_CACHE_BYTES     256
++#include <linux/const.h>
++
+ #define L1_CACHE_SHIFT     8
++#define L1_CACHE_BYTES     (_AC(1,UL) << L1_CACHE_SHIFT)
+ #define NET_SKB_PAD	   32
+ 
+ #define __read_mostly __attribute__((__section__(".data..read_mostly")))
+diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h
+index 547f1a6..0b22b53 100644
+--- a/arch/s390/include/asm/elf.h
++++ b/arch/s390/include/asm/elf.h
+@@ -162,8 +162,14 @@ extern unsigned int vdso_enabled;
+    the loader.  We need to make sure that it is out of the way of the program
+    that it will "exec", and that there is sufficient room for the brk.  */
+ 
+-extern unsigned long randomize_et_dyn(unsigned long base);
+-#define ELF_ET_DYN_BASE		(randomize_et_dyn(STACK_TOP / 3 * 2))
++#define ELF_ET_DYN_BASE		(STACK_TOP / 3 * 2)
++
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(test_thread_flag(TIF_31BIT) ? 0x10000UL : 0x80000000UL)
++
++#define PAX_DELTA_MMAP_LEN	(test_thread_flag(TIF_31BIT) ? 15 : 26)
++#define PAX_DELTA_STACK_LEN	(test_thread_flag(TIF_31BIT) ? 15 : 26)
++#endif
+ 
+ /* This yields a mask that user programs can use to figure out what
+    instruction set this CPU supports. */
+@@ -211,7 +217,4 @@ struct linux_binprm;
+ #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
+ int arch_setup_additional_pages(struct linux_binprm *, int);
+ 
+-extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+-#define arch_randomize_brk arch_randomize_brk
+-
+ #endif
+diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
+index ef573c1..75a1ce6 100644
+--- a/arch/s390/include/asm/system.h
++++ b/arch/s390/include/asm/system.h
+@@ -262,7 +262,7 @@ extern void (*_machine_restart)(char *command);
+ extern void (*_machine_halt)(void);
+ extern void (*_machine_power_off)(void);
+ 
+-extern unsigned long arch_align_stack(unsigned long sp);
++#define arch_align_stack(x) ((x) & ~0xfUL)
+ 
+ static inline int tprot(unsigned long addr)
+ {
+diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h
+index 2b23885..e136e31 100644
+--- a/arch/s390/include/asm/uaccess.h
++++ b/arch/s390/include/asm/uaccess.h
+@@ -235,6 +235,10 @@ static inline unsigned long __must_check
+ copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
+ 	might_fault();
++
++	if ((long)n < 0)
++		return n;
++
+ 	if (access_ok(VERIFY_WRITE, to, n))
+ 		n = __copy_to_user(to, from, n);
+ 	return n;
+@@ -260,6 +264,9 @@ copy_to_user(void __user *to, const void *from, unsigned long n)
+ static inline unsigned long __must_check
+ __copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	if (__builtin_constant_p(n) && (n <= 256))
+ 		return uaccess.copy_from_user_small(n, from, to);
+ 	else
+@@ -294,6 +301,10 @@ copy_from_user(void *to, const void __user *from, unsigned long n)
+ 	unsigned int sz = __compiletime_object_size(to);
+ 
+ 	might_fault();
++
++	if ((long)n < 0)
++		return n;
++
+ 	if (unlikely(sz != -1 && sz < n)) {
+ 		copy_from_user_overflow();
+ 		return n;
+diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c
+index dfcb343..eda788a 100644
+--- a/arch/s390/kernel/module.c
++++ b/arch/s390/kernel/module.c
+@@ -161,11 +161,11 @@ module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
+ 
+ 	/* Increase core size by size of got & plt and set start
+ 	   offsets for got and plt. */
+-	me->core_size = ALIGN(me->core_size, 4);
+-	me->arch.got_offset = me->core_size;
+-	me->core_size += me->arch.got_size;
+-	me->arch.plt_offset = me->core_size;
+-	me->core_size += me->arch.plt_size;
++	me->core_size_rw = ALIGN(me->core_size_rw, 4);
++	me->arch.got_offset = me->core_size_rw;
++	me->core_size_rw += me->arch.got_size;
++	me->arch.plt_offset = me->core_size_rx;
++	me->core_size_rx += me->arch.plt_size;
+ 	return 0;
+ }
+ 
+@@ -242,7 +242,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
+ 		if (info->got_initialized == 0) {
+ 			Elf_Addr *gotent;
+ 
+-			gotent = me->module_core + me->arch.got_offset +
++			gotent = me->module_core_rw + me->arch.got_offset +
+ 				info->got_offset;
+ 			*gotent = val;
+ 			info->got_initialized = 1;
+@@ -266,7 +266,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
+ 		else if (r_type == R_390_GOTENT ||
+ 			 r_type == R_390_GOTPLTENT)
+ 			*(unsigned int *) loc =
+-				(val + (Elf_Addr) me->module_core - loc) >> 1;
++				(val + (Elf_Addr) me->module_core_rw - loc) >> 1;
+ 		else if (r_type == R_390_GOT64 ||
+ 			 r_type == R_390_GOTPLT64)
+ 			*(unsigned long *) loc = val;
+@@ -280,7 +280,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
+ 	case R_390_PLTOFF64:	/* 16 bit offset from GOT to PLT. */
+ 		if (info->plt_initialized == 0) {
+ 			unsigned int *ip;
+-			ip = me->module_core + me->arch.plt_offset +
++			ip = me->module_core_rx + me->arch.plt_offset +
+ 				info->plt_offset;
+ #ifndef CONFIG_64BIT
+ 			ip[0] = 0x0d105810; /* basr 1,0; l 1,6(1); br 1 */
+@@ -305,7 +305,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
+ 			       val - loc + 0xffffUL < 0x1ffffeUL) ||
+ 			      (r_type == R_390_PLT32DBL &&
+ 			       val - loc + 0xffffffffULL < 0x1fffffffeULL)))
+-				val = (Elf_Addr) me->module_core +
++				val = (Elf_Addr) me->module_core_rx +
+ 					me->arch.plt_offset +
+ 					info->plt_offset;
+ 			val += rela->r_addend - loc;
+@@ -327,7 +327,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
+ 	case R_390_GOTOFF32:	/* 32 bit offset to GOT.  */
+ 	case R_390_GOTOFF64:	/* 64 bit offset to GOT. */
+ 		val = val + rela->r_addend -
+-			((Elf_Addr) me->module_core + me->arch.got_offset);
++			((Elf_Addr) me->module_core_rw + me->arch.got_offset);
+ 		if (r_type == R_390_GOTOFF16)
+ 			*(unsigned short *) loc = val;
+ 		else if (r_type == R_390_GOTOFF32)
+@@ -337,7 +337,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
+ 		break;
+ 	case R_390_GOTPC:	/* 32 bit PC relative offset to GOT. */
+ 	case R_390_GOTPCDBL:	/* 32 bit PC rel. off. to GOT shifted by 1. */
+-		val = (Elf_Addr) me->module_core + me->arch.got_offset +
++		val = (Elf_Addr) me->module_core_rw + me->arch.got_offset +
+ 			rela->r_addend - loc;
+ 		if (r_type == R_390_GOTPC)
+ 			*(unsigned int *) loc = val;
+diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
+index 53088e2..9f44a36 100644
+--- a/arch/s390/kernel/process.c
++++ b/arch/s390/kernel/process.c
+@@ -320,39 +320,3 @@ unsigned long get_wchan(struct task_struct *p)
+ 	}
+ 	return 0;
+ }
+-
+-unsigned long arch_align_stack(unsigned long sp)
+-{
+-	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+-		sp -= get_random_int() & ~PAGE_MASK;
+-	return sp & ~0xf;
+-}
+-
+-static inline unsigned long brk_rnd(void)
+-{
+-	/* 8MB for 32bit, 1GB for 64bit */
+-	if (is_32bit_task())
+-		return (get_random_int() & 0x7ffUL) << PAGE_SHIFT;
+-	else
+-		return (get_random_int() & 0x3ffffUL) << PAGE_SHIFT;
+-}
+-
+-unsigned long arch_randomize_brk(struct mm_struct *mm)
+-{
+-	unsigned long ret = PAGE_ALIGN(mm->brk + brk_rnd());
+-
+-	if (ret < mm->brk)
+-		return mm->brk;
+-	return ret;
+-}
+-
+-unsigned long randomize_et_dyn(unsigned long base)
+-{
+-	unsigned long ret = PAGE_ALIGN(base + brk_rnd());
+-
+-	if (!(current->flags & PF_RANDOMIZE))
+-		return base;
+-	if (ret < base)
+-		return base;
+-	return ret;
+-}
+diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c
+index a0155c0..34cc491 100644
+--- a/arch/s390/mm/mmap.c
++++ b/arch/s390/mm/mmap.c
+@@ -92,10 +92,22 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
+ 	 */
+ 	if (mmap_is_legacy()) {
+ 		mm->mmap_base = TASK_UNMAPPED_BASE;
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base += mm->delta_mmap;
++#endif
++
+ 		mm->get_unmapped_area = arch_get_unmapped_area;
+ 		mm->unmap_area = arch_unmap_area;
+ 	} else {
+ 		mm->mmap_base = mmap_base();
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base -= mm->delta_mmap + mm->delta_stack;
++#endif
++
+ 		mm->get_unmapped_area = arch_get_unmapped_area_topdown;
+ 		mm->unmap_area = arch_unmap_area_topdown;
+ 	}
+@@ -167,10 +179,22 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
+ 	 */
+ 	if (mmap_is_legacy()) {
+ 		mm->mmap_base = TASK_UNMAPPED_BASE;
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base += mm->delta_mmap;
++#endif
++
+ 		mm->get_unmapped_area = s390_get_unmapped_area;
+ 		mm->unmap_area = arch_unmap_area;
+ 	} else {
+ 		mm->mmap_base = mmap_base();
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base -= mm->delta_mmap + mm->delta_stack;
++#endif
++
+ 		mm->get_unmapped_area = s390_get_unmapped_area_topdown;
+ 		mm->unmap_area = arch_unmap_area_topdown;
+ 	}
+diff --git a/arch/score/include/asm/cache.h b/arch/score/include/asm/cache.h
+index ae3d59f..f65f075 100644
+--- a/arch/score/include/asm/cache.h
++++ b/arch/score/include/asm/cache.h
+@@ -1,7 +1,9 @@
+ #ifndef _ASM_SCORE_CACHE_H
+ #define _ASM_SCORE_CACHE_H
+ 
++#include <linux/const.h>
++
+ #define L1_CACHE_SHIFT		4
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #endif /* _ASM_SCORE_CACHE_H */
+diff --git a/arch/score/include/asm/system.h b/arch/score/include/asm/system.h
+index 589d5c7..669e274 100644
+--- a/arch/score/include/asm/system.h
++++ b/arch/score/include/asm/system.h
+@@ -17,7 +17,7 @@ do {								\
+ #define finish_arch_switch(prev)	do {} while (0)
+ 
+ typedef void (*vi_handler_t)(void);
+-extern unsigned long arch_align_stack(unsigned long sp);
++#define arch_align_stack(x) (x)
+ 
+ #define mb()		barrier()
+ #define rmb()		barrier()
+diff --git a/arch/score/kernel/process.c b/arch/score/kernel/process.c
+index 25d0803..d6c8e36 100644
+--- a/arch/score/kernel/process.c
++++ b/arch/score/kernel/process.c
+@@ -161,8 +161,3 @@ unsigned long get_wchan(struct task_struct *task)
+ 
+ 	return task_pt_regs(task)->cp0_epc;
+ }
+-
+-unsigned long arch_align_stack(unsigned long sp)
+-{
+-	return sp;
+-}
+diff --git a/arch/sh/include/asm/cache.h b/arch/sh/include/asm/cache.h
+index ef9e555..331bd29 100644
+--- a/arch/sh/include/asm/cache.h
++++ b/arch/sh/include/asm/cache.h
+@@ -9,10 +9,11 @@
+ #define __ASM_SH_CACHE_H
+ #ifdef __KERNEL__
+ 
++#include <linux/const.h>
+ #include <linux/init.h>
+ #include <cpu/cache.h>
+ 
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define __read_mostly __attribute__((__section__(".data..read_mostly")))
+ 
+diff --git a/arch/sh/mm/mmap.c b/arch/sh/mm/mmap.c
+index afeb710..d1d1289 100644
+--- a/arch/sh/mm/mmap.c
++++ b/arch/sh/mm/mmap.c
+@@ -74,8 +74,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 			addr = PAGE_ALIGN(addr);
+ 
+ 		vma = find_vma(mm, addr);
+-		if (TASK_SIZE - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 
+@@ -106,7 +105,7 @@ full_search:
+ 			}
+ 			return -ENOMEM;
+ 		}
+-		if (likely(!vma || addr + len <= vma->vm_start)) {
++		if (likely(check_heap_stack_gap(vma, addr, len))) {
+ 			/*
+ 			 * Remember the place where we stopped the search:
+ 			 */
+@@ -157,8 +156,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 			addr = PAGE_ALIGN(addr);
+ 
+ 		vma = find_vma(mm, addr);
+-		if (TASK_SIZE - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 
+@@ -179,7 +177,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	/* make sure it can fit in the remaining address space */
+ 	if (likely(addr > len)) {
+ 		vma = find_vma(mm, addr-len);
+-		if (!vma || addr <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr - len, len)) {
+ 			/* remember the address as a hint for next time */
+ 			return (mm->free_area_cache = addr-len);
+ 		}
+@@ -188,18 +186,18 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	if (unlikely(mm->mmap_base < len))
+ 		goto bottomup;
+ 
+-	addr = mm->mmap_base-len;
+-	if (do_colour_align)
+-		addr = COLOUR_ALIGN_DOWN(addr, pgoff);
++	addr = mm->mmap_base - len;
+ 
+ 	do {
++		if (do_colour_align)
++			addr = COLOUR_ALIGN_DOWN(addr, pgoff);
+ 		/*
+ 		 * Lookup failure means no vma is above this address,
+ 		 * else if new region fits below vma->vm_start,
+ 		 * return with success:
+ 		 */
+ 		vma = find_vma(mm, addr);
+-		if (likely(!vma || addr+len <= vma->vm_start)) {
++		if (likely(check_heap_stack_gap(vma, addr, len))) {
+ 			/* remember the address as a hint for next time */
+ 			return (mm->free_area_cache = addr);
+ 		}
+@@ -209,10 +207,8 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 		        mm->cached_hole_size = vma->vm_start - addr;
+ 
+ 		/* try just below the current vma->vm_start */
+-		addr = vma->vm_start-len;
+-		if (do_colour_align)
+-			addr = COLOUR_ALIGN_DOWN(addr, pgoff);
+-	} while (likely(len < vma->vm_start));
++		addr = skip_heap_stack_gap(vma, len);
++	} while (!IS_ERR_VALUE(addr));
+ 
+ bottomup:
+ 	/*
+diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
+index f92602e..27060b2 100644
+--- a/arch/sparc/Kconfig
++++ b/arch/sparc/Kconfig
+@@ -31,6 +31,7 @@ config SPARC
+ 
+ config SPARC32
+ 	def_bool !64BIT
++	select GENERIC_ATOMIC64
+ 
+ config SPARC64
+ 	def_bool 64BIT
+diff --git a/arch/sparc/Makefile b/arch/sparc/Makefile
+index eddcfb3..b117d90 100644
+--- a/arch/sparc/Makefile
++++ b/arch/sparc/Makefile
+@@ -75,7 +75,7 @@ drivers-$(CONFIG_OPROFILE)	+= arch/sparc/oprofile/
+ # Export what is needed by arch/sparc/boot/Makefile
+ export VMLINUX_INIT VMLINUX_MAIN
+ VMLINUX_INIT := $(head-y) $(init-y)
+-VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/
++VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ grsecurity/
+ VMLINUX_MAIN += $(patsubst %/, %/lib.a, $(libs-y)) $(libs-y)
+ VMLINUX_MAIN += $(drivers-y) $(net-y)
+ 
+diff --git a/arch/sparc/include/asm/atomic_32.h b/arch/sparc/include/asm/atomic_32.h
+index 5c3c8b6..ba822fa 100644
+--- a/arch/sparc/include/asm/atomic_32.h
++++ b/arch/sparc/include/asm/atomic_32.h
+@@ -13,6 +13,8 @@
+ 
+ #include <linux/types.h>
+ 
++#include <asm-generic/atomic64.h>
++
+ #ifdef __KERNEL__
+ 
+ #include <asm/system.h>
+diff --git a/arch/sparc/include/asm/atomic_64.h b/arch/sparc/include/asm/atomic_64.h
+index 9f421df..b81fc12 100644
+--- a/arch/sparc/include/asm/atomic_64.h
++++ b/arch/sparc/include/asm/atomic_64.h
+@@ -14,18 +14,40 @@
+ #define ATOMIC64_INIT(i)	{ (i) }
+ 
+ #define atomic_read(v)		(*(volatile int *)&(v)->counter)
++static inline int atomic_read_unchecked(const atomic_unchecked_t *v)
++{
++	return v->counter;
++}
+ #define atomic64_read(v)	(*(volatile long *)&(v)->counter)
++static inline long atomic64_read_unchecked(const atomic64_unchecked_t *v)
++{
++	return v->counter;
++}
+ 
+ #define atomic_set(v, i)	(((v)->counter) = i)
++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i)
++{
++	v->counter = i;
++}
+ #define atomic64_set(v, i)	(((v)->counter) = i)
++static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long i)
++{
++	v->counter = i;
++}
+ 
+ extern void atomic_add(int, atomic_t *);
++extern void atomic_add_unchecked(int, atomic_unchecked_t *);
+ extern void atomic64_add(long, atomic64_t *);
++extern void atomic64_add_unchecked(long, atomic64_unchecked_t *);
+ extern void atomic_sub(int, atomic_t *);
++extern void atomic_sub_unchecked(int, atomic_unchecked_t *);
+ extern void atomic64_sub(long, atomic64_t *);
++extern void atomic64_sub_unchecked(long, atomic64_unchecked_t *);
+ 
+ extern int atomic_add_ret(int, atomic_t *);
++extern int atomic_add_ret_unchecked(int, atomic_unchecked_t *);
+ extern long atomic64_add_ret(long, atomic64_t *);
++extern long atomic64_add_ret_unchecked(long, atomic64_unchecked_t *);
+ extern int atomic_sub_ret(int, atomic_t *);
+ extern long atomic64_sub_ret(long, atomic64_t *);
+ 
+@@ -33,13 +55,29 @@ extern long atomic64_sub_ret(long, atomic64_t *);
+ #define atomic64_dec_return(v) atomic64_sub_ret(1, v)
+ 
+ #define atomic_inc_return(v) atomic_add_ret(1, v)
++static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v)
++{
++	return atomic_add_ret_unchecked(1, v);
++}
+ #define atomic64_inc_return(v) atomic64_add_ret(1, v)
++static inline long atomic64_inc_return_unchecked(atomic64_unchecked_t *v)
++{
++	return atomic64_add_ret_unchecked(1, v);
++}
+ 
+ #define atomic_sub_return(i, v) atomic_sub_ret(i, v)
+ #define atomic64_sub_return(i, v) atomic64_sub_ret(i, v)
+ 
+ #define atomic_add_return(i, v) atomic_add_ret(i, v)
++static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v)
++{
++	return atomic_add_ret_unchecked(i, v);
++}
+ #define atomic64_add_return(i, v) atomic64_add_ret(i, v)
++static inline long atomic64_add_return_unchecked(long i, atomic64_unchecked_t *v)
++{
++	return atomic64_add_ret_unchecked(i, v);
++}
+ 
+ /*
+  * atomic_inc_and_test - increment and test
+@@ -50,6 +88,10 @@ extern long atomic64_sub_ret(long, atomic64_t *);
+  * other cases.
+  */
+ #define atomic_inc_and_test(v) (atomic_inc_return(v) == 0)
++static inline int atomic_inc_and_test_unchecked(atomic_unchecked_t *v)
++{
++	return atomic_inc_return_unchecked(v) == 0;
++}
+ #define atomic64_inc_and_test(v) (atomic64_inc_return(v) == 0)
+ 
+ #define atomic_sub_and_test(i, v) (atomic_sub_ret(i, v) == 0)
+@@ -59,25 +101,60 @@ extern long atomic64_sub_ret(long, atomic64_t *);
+ #define atomic64_dec_and_test(v) (atomic64_sub_ret(1, v) == 0)
+ 
+ #define atomic_inc(v) atomic_add(1, v)
++static inline void atomic_inc_unchecked(atomic_unchecked_t *v)
++{
++	atomic_add_unchecked(1, v);
++}
+ #define atomic64_inc(v) atomic64_add(1, v)
++static inline void atomic64_inc_unchecked(atomic64_unchecked_t *v)
++{
++	atomic64_add_unchecked(1, v);
++}
+ 
+ #define atomic_dec(v) atomic_sub(1, v)
++static inline void atomic_dec_unchecked(atomic_unchecked_t *v)
++{
++	atomic_sub_unchecked(1, v);
++}
+ #define atomic64_dec(v) atomic64_sub(1, v)
++static inline void atomic64_dec_unchecked(atomic64_unchecked_t *v)
++{
++	atomic64_sub_unchecked(1, v);
++}
+ 
+ #define atomic_add_negative(i, v) (atomic_add_ret(i, v) < 0)
+ #define atomic64_add_negative(i, v) (atomic64_add_ret(i, v) < 0)
+ 
+ #define atomic_cmpxchg(v, o, n) (cmpxchg(&((v)->counter), (o), (n)))
++static inline int atomic_cmpxchg_unchecked(atomic_unchecked_t *v, int old, int new)
++{
++	return cmpxchg(&v->counter, old, new);
++}
+ #define atomic_xchg(v, new) (xchg(&((v)->counter), new))
++static inline int atomic_xchg_unchecked(atomic_unchecked_t *v, int new)
++{
++	return xchg(&v->counter, new);
++}
+ 
+ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+ {
+-	int c, old;
++	int c, old, new;
+ 	c = atomic_read(v);
+ 	for (;;) {
+-		if (unlikely(c == (u)))
++		if (unlikely(c == u))
+ 			break;
+-		old = atomic_cmpxchg((v), c, c + (a));
++
++		asm volatile("addcc %2, %0, %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++			     "tvs %%icc, 6\n"
++#endif
++
++			     : "=r" (new)
++			     : "0" (c), "ir" (a)
++			     : "cc");
++
++		old = atomic_cmpxchg(v, c, new);
+ 		if (likely(old == c))
+ 			break;
+ 		c = old;
+@@ -89,20 +166,35 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+ #define atomic64_cmpxchg(v, o, n) \
+ 	((__typeof__((v)->counter))cmpxchg(&((v)->counter), (o), (n)))
+ #define atomic64_xchg(v, new) (xchg(&((v)->counter), new))
++static inline long atomic64_xchg_unchecked(atomic64_unchecked_t *v, long new)
++{
++	return xchg(&v->counter, new);
++}
+ 
+ static inline long atomic64_add_unless(atomic64_t *v, long a, long u)
+ {
+-	long c, old;
++	long c, old, new;
+ 	c = atomic64_read(v);
+ 	for (;;) {
+-		if (unlikely(c == (u)))
++		if (unlikely(c == u))
+ 			break;
+-		old = atomic64_cmpxchg((v), c, c + (a));
++
++		asm volatile("addcc %2, %0, %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++			     "tvs %%xcc, 6\n"
++#endif
++
++			     : "=r" (new)
++			     : "0" (c), "ir" (a)
++			     : "cc");
++
++		old = atomic64_cmpxchg(v, c, new);
+ 		if (likely(old == c))
+ 			break;
+ 		c = old;
+ 	}
+-	return c != (u);
++	return c != u;
+ }
+ 
+ #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
+diff --git a/arch/sparc/include/asm/cache.h b/arch/sparc/include/asm/cache.h
+index 69358b5..9d0d492 100644
+--- a/arch/sparc/include/asm/cache.h
++++ b/arch/sparc/include/asm/cache.h
+@@ -7,10 +7,12 @@
+ #ifndef _SPARC_CACHE_H
+ #define _SPARC_CACHE_H
+ 
++#include <linux/const.h>
++
+ #define ARCH_SLAB_MINALIGN	__alignof__(unsigned long long)
+ 
+ #define L1_CACHE_SHIFT 5
+-#define L1_CACHE_BYTES 32
++#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #ifdef CONFIG_SPARC32
+ #define SMP_CACHE_BYTES_SHIFT 5
+diff --git a/arch/sparc/include/asm/elf_32.h b/arch/sparc/include/asm/elf_32.h
+index 4269ca6..e3da77f 100644
+--- a/arch/sparc/include/asm/elf_32.h
++++ b/arch/sparc/include/asm/elf_32.h
+@@ -114,6 +114,13 @@ typedef struct {
+ 
+ #define ELF_ET_DYN_BASE         (TASK_UNMAPPED_BASE)
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	0x10000UL
++
++#define PAX_DELTA_MMAP_LEN	16
++#define PAX_DELTA_STACK_LEN	16
++#endif
++
+ /* This yields a mask that user programs can use to figure out what
+    instruction set this cpu supports.  This can NOT be done in userspace
+    on Sparc.  */
+diff --git a/arch/sparc/include/asm/elf_64.h b/arch/sparc/include/asm/elf_64.h
+index 7df8b7f..4946269 100644
+--- a/arch/sparc/include/asm/elf_64.h
++++ b/arch/sparc/include/asm/elf_64.h
+@@ -180,6 +180,13 @@ typedef struct {
+ #define ELF_ET_DYN_BASE		0x0000010000000000UL
+ #define COMPAT_ELF_ET_DYN_BASE	0x0000000070000000UL
+ 
++#ifdef CONFIG_PAX_ASLR
++#define PAX_ELF_ET_DYN_BASE	(test_thread_flag(TIF_32BIT) ? 0x10000UL : 0x100000UL)
++
++#define PAX_DELTA_MMAP_LEN	(test_thread_flag(TIF_32BIT) ? 14 : 28)
++#define PAX_DELTA_STACK_LEN	(test_thread_flag(TIF_32BIT) ? 15 : 29)
++#endif
++
+ extern unsigned long sparc64_elf_hwcap;
+ #define ELF_HWCAP	sparc64_elf_hwcap
+ 
+diff --git a/arch/sparc/include/asm/page_32.h b/arch/sparc/include/asm/page_32.h
+index 156707b..aefa786 100644
+--- a/arch/sparc/include/asm/page_32.h
++++ b/arch/sparc/include/asm/page_32.h
+@@ -8,6 +8,8 @@
+ #ifndef _SPARC_PAGE_H
+ #define _SPARC_PAGE_H
+ 
++#include <linux/const.h>
++
+ #define PAGE_SHIFT   12
+ 
+ #ifndef __ASSEMBLY__
+diff --git a/arch/sparc/include/asm/pgtable_32.h b/arch/sparc/include/asm/pgtable_32.h
+index a790cc6..091ed94 100644
+--- a/arch/sparc/include/asm/pgtable_32.h
++++ b/arch/sparc/include/asm/pgtable_32.h
+@@ -45,6 +45,13 @@ BTFIXUPDEF_SIMM13(user_ptrs_per_pgd)
+ BTFIXUPDEF_INT(page_none)
+ BTFIXUPDEF_INT(page_copy)
+ BTFIXUPDEF_INT(page_readonly)
++
++#ifdef CONFIG_PAX_PAGEEXEC
++BTFIXUPDEF_INT(page_shared_noexec)
++BTFIXUPDEF_INT(page_copy_noexec)
++BTFIXUPDEF_INT(page_readonly_noexec)
++#endif
++
+ BTFIXUPDEF_INT(page_kernel)
+ 
+ #define PMD_SHIFT		SUN4C_PMD_SHIFT
+@@ -66,6 +73,16 @@ extern pgprot_t PAGE_SHARED;
+ #define PAGE_COPY      __pgprot(BTFIXUP_INT(page_copy))
+ #define PAGE_READONLY  __pgprot(BTFIXUP_INT(page_readonly))
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++extern pgprot_t PAGE_SHARED_NOEXEC;
++# define PAGE_COPY_NOEXEC	__pgprot(BTFIXUP_INT(page_copy_noexec))
++# define PAGE_READONLY_NOEXEC	__pgprot(BTFIXUP_INT(page_readonly_noexec))
++#else
++# define PAGE_SHARED_NOEXEC	PAGE_SHARED
++# define PAGE_COPY_NOEXEC	PAGE_COPY
++# define PAGE_READONLY_NOEXEC	PAGE_READONLY
++#endif
++
+ extern unsigned long page_kernel;
+ 
+ #ifdef MODULE
+diff --git a/arch/sparc/include/asm/pgtsrmmu.h b/arch/sparc/include/asm/pgtsrmmu.h
+index f6ae2b2..b03ffc7 100644
+--- a/arch/sparc/include/asm/pgtsrmmu.h
++++ b/arch/sparc/include/asm/pgtsrmmu.h
+@@ -115,6 +115,13 @@
+ 				    SRMMU_EXEC | SRMMU_REF)
+ #define SRMMU_PAGE_RDONLY  __pgprot(SRMMU_VALID | SRMMU_CACHE | \
+ 				    SRMMU_EXEC | SRMMU_REF)
++
++#ifdef CONFIG_PAX_PAGEEXEC
++#define SRMMU_PAGE_SHARED_NOEXEC	__pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_WRITE | SRMMU_REF)
++#define SRMMU_PAGE_COPY_NOEXEC	__pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF)
++#define SRMMU_PAGE_RDONLY_NOEXEC	__pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF)
++#endif
++
+ #define SRMMU_PAGE_KERNEL  __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_PRIV | \
+ 				    SRMMU_DIRTY | SRMMU_REF)
+ 
+diff --git a/arch/sparc/include/asm/spinlock_64.h b/arch/sparc/include/asm/spinlock_64.h
+index 9689176..63c18ea 100644
+--- a/arch/sparc/include/asm/spinlock_64.h
++++ b/arch/sparc/include/asm/spinlock_64.h
+@@ -92,14 +92,19 @@ static inline void arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long fla
+ 
+ /* Multi-reader locks, these are much saner than the 32-bit Sparc ones... */
+ 
+-static void inline arch_read_lock(arch_rwlock_t *lock)
++static inline void arch_read_lock(arch_rwlock_t *lock)
+ {
+ 	unsigned long tmp1, tmp2;
+ 
+ 	__asm__ __volatile__ (
+ "1:	ldsw		[%2], %0\n"
+ "	brlz,pn		%0, 2f\n"
+-"4:	 add		%0, 1, %1\n"
++"4:	 addcc		%0, 1, %1\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	tvs		%%icc, 6\n"
++#endif
++
+ "	cas		[%2], %0, %1\n"
+ "	cmp		%0, %1\n"
+ "	bne,pn		%%icc, 1b\n"
+@@ -112,10 +117,10 @@ static void inline arch_read_lock(arch_rwlock_t *lock)
+ "	.previous"
+ 	: "=&r" (tmp1), "=&r" (tmp2)
+ 	: "r" (lock)
+-	: "memory");
++	: "memory", "cc");
+ }
+ 
+-static int inline arch_read_trylock(arch_rwlock_t *lock)
++static inline int arch_read_trylock(arch_rwlock_t *lock)
+ {
+ 	int tmp1, tmp2;
+ 
+@@ -123,7 +128,12 @@ static int inline arch_read_trylock(arch_rwlock_t *lock)
+ "1:	ldsw		[%2], %0\n"
+ "	brlz,a,pn	%0, 2f\n"
+ "	 mov		0, %0\n"
+-"	add		%0, 1, %1\n"
++"	addcc		%0, 1, %1\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	tvs		%%icc, 6\n"
++#endif
++
+ "	cas		[%2], %0, %1\n"
+ "	cmp		%0, %1\n"
+ "	bne,pn		%%icc, 1b\n"
+@@ -136,13 +146,18 @@ static int inline arch_read_trylock(arch_rwlock_t *lock)
+ 	return tmp1;
+ }
+ 
+-static void inline arch_read_unlock(arch_rwlock_t *lock)
++static inline void arch_read_unlock(arch_rwlock_t *lock)
+ {
+ 	unsigned long tmp1, tmp2;
+ 
+ 	__asm__ __volatile__(
+ "1:	lduw	[%2], %0\n"
+-"	sub	%0, 1, %1\n"
++"	subcc	%0, 1, %1\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++"	tvs	%%icc, 6\n"
++#endif
++
+ "	cas	[%2], %0, %1\n"
+ "	cmp	%0, %1\n"
+ "	bne,pn	%%xcc, 1b\n"
+@@ -152,7 +167,7 @@ static void inline arch_read_unlock(arch_rwlock_t *lock)
+ 	: "memory");
+ }
+ 
+-static void inline arch_write_lock(arch_rwlock_t *lock)
++static inline void arch_write_lock(arch_rwlock_t *lock)
+ {
+ 	unsigned long mask, tmp1, tmp2;
+ 
+@@ -177,7 +192,7 @@ static void inline arch_write_lock(arch_rwlock_t *lock)
+ 	: "memory");
+ }
+ 
+-static void inline arch_write_unlock(arch_rwlock_t *lock)
++static inline void arch_write_unlock(arch_rwlock_t *lock)
+ {
+ 	__asm__ __volatile__(
+ "	stw		%%g0, [%0]"
+@@ -186,7 +201,7 @@ static void inline arch_write_unlock(arch_rwlock_t *lock)
+ 	: "memory");
+ }
+ 
+-static int inline arch_write_trylock(arch_rwlock_t *lock)
++static inline int arch_write_trylock(arch_rwlock_t *lock)
+ {
+ 	unsigned long mask, tmp1, tmp2, result;
+ 
+diff --git a/arch/sparc/include/asm/thread_info_32.h b/arch/sparc/include/asm/thread_info_32.h
+index fa57532..e1a4c53 100644
+--- a/arch/sparc/include/asm/thread_info_32.h
++++ b/arch/sparc/include/asm/thread_info_32.h
+@@ -50,6 +50,8 @@ struct thread_info {
+ 	unsigned long		w_saved;
+ 
+ 	struct restart_block	restart_block;
++
++	unsigned long		lowest_stack;
+ };
+ 
+ /*
+diff --git a/arch/sparc/include/asm/thread_info_64.h b/arch/sparc/include/asm/thread_info_64.h
+index 60d86be..952dea1 100644
+--- a/arch/sparc/include/asm/thread_info_64.h
++++ b/arch/sparc/include/asm/thread_info_64.h
+@@ -63,6 +63,8 @@ struct thread_info {
+ 	struct pt_regs		*kern_una_regs;
+ 	unsigned int		kern_una_insn;
+ 
++	unsigned long		lowest_stack;
++
+ 	unsigned long		fpregs[0] __attribute__ ((aligned(64)));
+ };
+ 
+diff --git a/arch/sparc/include/asm/uaccess.h b/arch/sparc/include/asm/uaccess.h
+index e88fbe5..96b0ce5 100644
+--- a/arch/sparc/include/asm/uaccess.h
++++ b/arch/sparc/include/asm/uaccess.h
+@@ -1,5 +1,13 @@
+ #ifndef ___ASM_SPARC_UACCESS_H
+ #define ___ASM_SPARC_UACCESS_H
++
++#ifdef __KERNEL__
++#ifndef __ASSEMBLY__
++#include <linux/types.h>
++extern void check_object_size(const void *ptr, unsigned long n, bool to);
++#endif
++#endif
++
+ #if defined(__sparc__) && defined(__arch64__)
+ #include <asm/uaccess_64.h>
+ #else
+diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h
+index 8303ac4..07f333d 100644
+--- a/arch/sparc/include/asm/uaccess_32.h
++++ b/arch/sparc/include/asm/uaccess_32.h
+@@ -249,27 +249,46 @@ extern unsigned long __copy_user(void __user *to, const void __user *from, unsig
+ 
+ static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
+-	if (n && __access_ok((unsigned long) to, n))
++	if ((long)n < 0)
++		return n;
++
++	if (n && __access_ok((unsigned long) to, n)) {
++		if (!__builtin_constant_p(n))
++			check_object_size(from, n, true);
+ 		return __copy_user(to, (__force void __user *) from, n);
+-	else
++	} else
+ 		return n;
+ }
+ 
+ static inline unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
++	if (!__builtin_constant_p(n))
++		check_object_size(from, n, true);
++
+ 	return __copy_user(to, (__force void __user *) from, n);
+ }
+ 
+ static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
+-	if (n && __access_ok((unsigned long) from, n))
++	if ((long)n < 0)
++		return n;
++
++	if (n && __access_ok((unsigned long) from, n)) {
++		if (!__builtin_constant_p(n))
++			check_object_size(to, n, false);
+ 		return __copy_user((__force void __user *) to, from, n);
+-	else
++	} else
+ 		return n;
+ }
+ 
+ static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	return __copy_user((__force void __user *) to, from, n);
+ }
+ 
+diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
+index 3e1449f..5293a0e 100644
+--- a/arch/sparc/include/asm/uaccess_64.h
++++ b/arch/sparc/include/asm/uaccess_64.h
+@@ -10,6 +10,7 @@
+ #include <linux/compiler.h>
+ #include <linux/string.h>
+ #include <linux/thread_info.h>
++#include <linux/kernel.h>
+ #include <asm/asi.h>
+ #include <asm/system.h>
+ #include <asm/spitfire.h>
+@@ -213,8 +214,15 @@ extern unsigned long copy_from_user_fixup(void *to, const void __user *from,
+ static inline unsigned long __must_check
+ copy_from_user(void *to, const void __user *from, unsigned long size)
+ {
+-	unsigned long ret = ___copy_from_user(to, from, size);
++	unsigned long ret;
+ 
++	if ((long)size < 0 || size > INT_MAX)
++		return size;
++
++	if (!__builtin_constant_p(size))
++		check_object_size(to, size, false);
++
++	ret = ___copy_from_user(to, from, size);
+ 	if (unlikely(ret))
+ 		ret = copy_from_user_fixup(to, from, size);
+ 
+@@ -230,8 +238,15 @@ extern unsigned long copy_to_user_fixup(void __user *to, const void *from,
+ static inline unsigned long __must_check
+ copy_to_user(void __user *to, const void *from, unsigned long size)
+ {
+-	unsigned long ret = ___copy_to_user(to, from, size);
++	unsigned long ret;
+ 
++	if ((long)size < 0 || size > INT_MAX)
++		return size;
++
++	if (!__builtin_constant_p(size))
++		check_object_size(from, size, true);
++
++	ret = ___copy_to_user(to, from, size);
+ 	if (unlikely(ret))
+ 		ret = copy_to_user_fixup(to, from, size);
+ 	return ret;
+diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile
+index cb85458..e063f17 100644
+--- a/arch/sparc/kernel/Makefile
++++ b/arch/sparc/kernel/Makefile
+@@ -3,7 +3,7 @@
+ #
+ 
+ asflags-y := -ansi
+-ccflags-y := -Werror
++#ccflags-y := -Werror
+ 
+ extra-y     := head_$(BITS).o
+ extra-y     += init_task.o
+diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c
+index f793742..4d880af 100644
+--- a/arch/sparc/kernel/process_32.c
++++ b/arch/sparc/kernel/process_32.c
+@@ -204,7 +204,7 @@ void __show_backtrace(unsigned long fp)
+ 		       rw->ins[4], rw->ins[5],
+ 		       rw->ins[6],
+ 		       rw->ins[7]);
+-		printk("%pS\n", (void *) rw->ins[7]);
++		printk("%pA\n", (void *) rw->ins[7]);
+ 		rw = (struct reg_window32 *) rw->ins[6];
+ 	}
+ 	spin_unlock_irqrestore(&sparc_backtrace_lock, flags);
+@@ -271,14 +271,14 @@ void show_regs(struct pt_regs *r)
+ 
+         printk("PSR: %08lx PC: %08lx NPC: %08lx Y: %08lx    %s\n",
+ 	       r->psr, r->pc, r->npc, r->y, print_tainted());
+-	printk("PC: <%pS>\n", (void *) r->pc);
++	printk("PC: <%pA>\n", (void *) r->pc);
+ 	printk("%%G: %08lx %08lx  %08lx %08lx  %08lx %08lx  %08lx %08lx\n",
+ 	       r->u_regs[0], r->u_regs[1], r->u_regs[2], r->u_regs[3],
+ 	       r->u_regs[4], r->u_regs[5], r->u_regs[6], r->u_regs[7]);
+ 	printk("%%O: %08lx %08lx  %08lx %08lx  %08lx %08lx  %08lx %08lx\n",
+ 	       r->u_regs[8], r->u_regs[9], r->u_regs[10], r->u_regs[11],
+ 	       r->u_regs[12], r->u_regs[13], r->u_regs[14], r->u_regs[15]);
+-	printk("RPC: <%pS>\n", (void *) r->u_regs[15]);
++	printk("RPC: <%pA>\n", (void *) r->u_regs[15]);
+ 
+ 	printk("%%L: %08lx %08lx  %08lx %08lx  %08lx %08lx  %08lx %08lx\n",
+ 	       rw->locals[0], rw->locals[1], rw->locals[2], rw->locals[3],
+@@ -313,7 +313,7 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp)
+ 		rw = (struct reg_window32 *) fp;
+ 		pc = rw->ins[7];
+ 		printk("[%08lx : ", pc);
+-		printk("%pS ] ", (void *) pc);
++		printk("%pA ] ", (void *) pc);
+ 		fp = rw->ins[6];
+ 	} while (++count < 16);
+ 	printk("\n");
+diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
+index 3739a06..48b2ff0 100644
+--- a/arch/sparc/kernel/process_64.c
++++ b/arch/sparc/kernel/process_64.c
+@@ -180,14 +180,14 @@ static void show_regwindow(struct pt_regs *regs)
+ 	printk("i4: %016lx i5: %016lx i6: %016lx i7: %016lx\n",
+ 	       rwk->ins[4], rwk->ins[5], rwk->ins[6], rwk->ins[7]);
+ 	if (regs->tstate & TSTATE_PRIV)
+-		printk("I7: <%pS>\n", (void *) rwk->ins[7]);
++		printk("I7: <%pA>\n", (void *) rwk->ins[7]);
+ }
+ 
+ void show_regs(struct pt_regs *regs)
+ {
+ 	printk("TSTATE: %016lx TPC: %016lx TNPC: %016lx Y: %08x    %s\n", regs->tstate,
+ 	       regs->tpc, regs->tnpc, regs->y, print_tainted());
+-	printk("TPC: <%pS>\n", (void *) regs->tpc);
++	printk("TPC: <%pA>\n", (void *) regs->tpc);
+ 	printk("g0: %016lx g1: %016lx g2: %016lx g3: %016lx\n",
+ 	       regs->u_regs[0], regs->u_regs[1], regs->u_regs[2],
+ 	       regs->u_regs[3]);
+@@ -200,7 +200,7 @@ void show_regs(struct pt_regs *regs)
+ 	printk("o4: %016lx o5: %016lx sp: %016lx ret_pc: %016lx\n",
+ 	       regs->u_regs[12], regs->u_regs[13], regs->u_regs[14],
+ 	       regs->u_regs[15]);
+-	printk("RPC: <%pS>\n", (void *) regs->u_regs[15]);
++	printk("RPC: <%pA>\n", (void *) regs->u_regs[15]);
+ 	show_regwindow(regs);
+ 	show_stack(current, (unsigned long *) regs->u_regs[UREG_FP]);
+ }
+@@ -285,7 +285,7 @@ void arch_trigger_all_cpu_backtrace(void)
+ 		       ((tp && tp->task) ? tp->task->pid : -1));
+ 
+ 		if (gp->tstate & TSTATE_PRIV) {
+-			printk("             TPC[%pS] O7[%pS] I7[%pS] RPC[%pS]\n",
++			printk("             TPC[%pA] O7[%pA] I7[%pA] RPC[%pA]\n",
+ 			       (void *) gp->tpc,
+ 			       (void *) gp->o7,
+ 			       (void *) gp->i7,
+diff --git a/arch/sparc/kernel/sys_sparc_32.c b/arch/sparc/kernel/sys_sparc_32.c
+index 42b282f..28ce9f2 100644
+--- a/arch/sparc/kernel/sys_sparc_32.c
++++ b/arch/sparc/kernel/sys_sparc_32.c
+@@ -56,7 +56,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 	if (ARCH_SUN4C && len > 0x20000000)
+ 		return -ENOMEM;
+ 	if (!addr)
+-		addr = TASK_UNMAPPED_BASE;
++		addr = current->mm->mmap_base;
+ 
+ 	if (flags & MAP_SHARED)
+ 		addr = COLOUR_ALIGN(addr);
+@@ -71,7 +71,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 		}
+ 		if (TASK_SIZE - PAGE_SIZE - len < addr)
+ 			return -ENOMEM;
+-		if (!vmm || addr + len <= vmm->vm_start)
++		if (check_heap_stack_gap(vmm, addr, len))
+ 			return addr;
+ 		addr = vmm->vm_end;
+ 		if (flags & MAP_SHARED)
+diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c
+index 441521a..b767073 100644
+--- a/arch/sparc/kernel/sys_sparc_64.c
++++ b/arch/sparc/kernel/sys_sparc_64.c
+@@ -124,7 +124,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 		/* We do not accept a shared mapping if it would violate
+ 		 * cache aliasing constraints.
+ 		 */
+-		if ((flags & MAP_SHARED) &&
++		if ((filp || (flags & MAP_SHARED)) &&
+ 		    ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1)))
+ 			return -EINVAL;
+ 		return addr;
+@@ -139,6 +139,10 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 	if (filp || (flags & MAP_SHARED))
+ 		do_color_align = 1;
+ 
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
+ 	if (addr) {
+ 		if (do_color_align)
+ 			addr = COLOUR_ALIGN(addr, pgoff);
+@@ -146,15 +150,14 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
+ 			addr = PAGE_ALIGN(addr);
+ 
+ 		vma = find_vma(mm, addr);
+-		if (task_size - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 
+ 	if (len > mm->cached_hole_size) {
+-	        start_addr = addr = mm->free_area_cache;
++		start_addr = addr = mm->free_area_cache;
+ 	} else {
+-	        start_addr = addr = TASK_UNMAPPED_BASE;
++		start_addr = addr = mm->mmap_base;
+ 	        mm->cached_hole_size = 0;
+ 	}
+ 
+@@ -174,14 +177,14 @@ full_search:
+ 			vma = find_vma(mm, VA_EXCLUDE_END);
+ 		}
+ 		if (unlikely(task_size < addr)) {
+-			if (start_addr != TASK_UNMAPPED_BASE) {
+-				start_addr = addr = TASK_UNMAPPED_BASE;
++			if (start_addr != mm->mmap_base) {
++				start_addr = addr = mm->mmap_base;
+ 				mm->cached_hole_size = 0;
+ 				goto full_search;
+ 			}
+ 			return -ENOMEM;
+ 		}
+-		if (likely(!vma || addr + len <= vma->vm_start)) {
++		if (likely(check_heap_stack_gap(vma, addr, len))) {
+ 			/*
+ 			 * Remember the place where we stopped the search:
+ 			 */
+@@ -215,7 +218,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 		/* We do not accept a shared mapping if it would violate
+ 		 * cache aliasing constraints.
+ 		 */
+-		if ((flags & MAP_SHARED) &&
++		if ((filp || (flags & MAP_SHARED)) &&
+ 		    ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1)))
+ 			return -EINVAL;
+ 		return addr;
+@@ -236,8 +239,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 			addr = PAGE_ALIGN(addr);
+ 
+ 		vma = find_vma(mm, addr);
+-		if (task_size - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 
+@@ -258,7 +260,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	/* make sure it can fit in the remaining address space */
+ 	if (likely(addr > len)) {
+ 		vma = find_vma(mm, addr-len);
+-		if (!vma || addr <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr - len, len)) {
+ 			/* remember the address as a hint for next time */
+ 			return (mm->free_area_cache = addr-len);
+ 		}
+@@ -267,18 +269,18 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	if (unlikely(mm->mmap_base < len))
+ 		goto bottomup;
+ 
+-	addr = mm->mmap_base-len;
+-	if (do_color_align)
+-		addr = COLOUR_ALIGN_DOWN(addr, pgoff);
++	addr = mm->mmap_base - len;
+ 
+ 	do {
++		if (do_color_align)
++			addr = COLOUR_ALIGN_DOWN(addr, pgoff);
+ 		/*
+ 		 * Lookup failure means no vma is above this address,
+ 		 * else if new region fits below vma->vm_start,
+ 		 * return with success:
+ 		 */
+ 		vma = find_vma(mm, addr);
+-		if (likely(!vma || addr+len <= vma->vm_start)) {
++		if (likely(check_heap_stack_gap(vma, addr, len))) {
+ 			/* remember the address as a hint for next time */
+ 			return (mm->free_area_cache = addr);
+ 		}
+@@ -288,10 +290,8 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+  		        mm->cached_hole_size = vma->vm_start - addr;
+ 
+ 		/* try just below the current vma->vm_start */
+-		addr = vma->vm_start-len;
+-		if (do_color_align)
+-			addr = COLOUR_ALIGN_DOWN(addr, pgoff);
+-	} while (likely(len < vma->vm_start));
++		addr = skip_heap_stack_gap(vma, len);
++	} while (!IS_ERR_VALUE(addr));
+ 
+ bottomup:
+ 	/*
+@@ -390,6 +390,12 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
+ 	    gap == RLIM_INFINITY ||
+ 	    sysctl_legacy_va_layout) {
+ 		mm->mmap_base = TASK_UNMAPPED_BASE + random_factor;
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base += mm->delta_mmap;
++#endif
++
+ 		mm->get_unmapped_area = arch_get_unmapped_area;
+ 		mm->unmap_area = arch_unmap_area;
+ 	} else {
+@@ -402,6 +408,12 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
+ 			gap = (task_size / 6 * 5);
+ 
+ 		mm->mmap_base = PAGE_ALIGN(task_size - gap - random_factor);
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			mm->mmap_base -= mm->delta_mmap + mm->delta_stack;
++#endif
++
+ 		mm->get_unmapped_area = arch_get_unmapped_area_topdown;
+ 		mm->unmap_area = arch_unmap_area_topdown;
+ 	}
+diff --git a/arch/sparc/kernel/traps_32.c b/arch/sparc/kernel/traps_32.c
+index 591f20c..0f1b925 100644
+--- a/arch/sparc/kernel/traps_32.c
++++ b/arch/sparc/kernel/traps_32.c
+@@ -45,6 +45,8 @@ static void instruction_dump(unsigned long *pc)
+ #define __SAVE __asm__ __volatile__("save %sp, -0x40, %sp\n\t")
+ #define __RESTORE __asm__ __volatile__("restore %g0, %g0, %g0\n\t")
+ 
++extern void gr_handle_kernel_exploit(void);
++
+ void die_if_kernel(char *str, struct pt_regs *regs)
+ {
+ 	static int die_counter;
+@@ -77,15 +79,17 @@ void die_if_kernel(char *str, struct pt_regs *regs)
+ 		      count++ < 30				&&
+                       (((unsigned long) rw) >= PAGE_OFFSET)	&&
+ 		      !(((unsigned long) rw) & 0x7)) {
+-			printk("Caller[%08lx]: %pS\n", rw->ins[7],
++			printk("Caller[%08lx]: %pA\n", rw->ins[7],
+ 			       (void *) rw->ins[7]);
+ 			rw = (struct reg_window32 *)rw->ins[6];
+ 		}
+ 	}
+ 	printk("Instruction DUMP:");
+ 	instruction_dump ((unsigned long *) regs->pc);
+-	if(regs->psr & PSR_PS)
++	if(regs->psr & PSR_PS) {
++		gr_handle_kernel_exploit();
+ 		do_exit(SIGKILL);
++	}
+ 	do_exit(SIGSEGV);
+ }
+ 
+diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c
+index 0cbdaa4..438e4c9 100644
+--- a/arch/sparc/kernel/traps_64.c
++++ b/arch/sparc/kernel/traps_64.c
+@@ -75,7 +75,7 @@ static void dump_tl1_traplog(struct tl1_traplog *p)
+ 		       i + 1,
+ 		       p->trapstack[i].tstate, p->trapstack[i].tpc,
+ 		       p->trapstack[i].tnpc, p->trapstack[i].tt);
+-		printk("TRAPLOG: TPC<%pS>\n", (void *) p->trapstack[i].tpc);
++		printk("TRAPLOG: TPC<%pA>\n", (void *) p->trapstack[i].tpc);
+ 	}
+ }
+ 
+@@ -95,6 +95,12 @@ void bad_trap(struct pt_regs *regs, long lvl)
+ 
+ 	lvl -= 0x100;
+ 	if (regs->tstate & TSTATE_PRIV) {
++
++#ifdef CONFIG_PAX_REFCOUNT
++		if (lvl == 6)
++			pax_report_refcount_overflow(regs);
++#endif
++
+ 		sprintf(buffer, "Kernel bad sw trap %lx", lvl);
+ 		die_if_kernel(buffer, regs);
+ 	}
+@@ -113,11 +119,16 @@ void bad_trap(struct pt_regs *regs, long lvl)
+ void bad_trap_tl1(struct pt_regs *regs, long lvl)
+ {
+ 	char buffer[32];
+-	
++
+ 	if (notify_die(DIE_TRAP_TL1, "bad trap tl1", regs,
+ 		       0, lvl, SIGTRAP) == NOTIFY_STOP)
+ 		return;
+ 
++#ifdef CONFIG_PAX_REFCOUNT
++	if (lvl == 6)
++		pax_report_refcount_overflow(regs);
++#endif
++
+ 	dump_tl1_traplog((struct tl1_traplog *)(regs + 1));
+ 
+ 	sprintf (buffer, "Bad trap %lx at tl>0", lvl);
+@@ -1141,7 +1152,7 @@ static void cheetah_log_errors(struct pt_regs *regs, struct cheetah_err_info *in
+ 	       regs->tpc, regs->tnpc, regs->u_regs[UREG_I7], regs->tstate);
+ 	printk("%s" "ERROR(%d): ",
+ 	       (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id());
+-	printk("TPC<%pS>\n", (void *) regs->tpc);
++	printk("TPC<%pA>\n", (void *) regs->tpc);
+ 	printk("%s" "ERROR(%d): M_SYND(%lx),  E_SYND(%lx)%s%s\n",
+ 	       (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id(),
+ 	       (afsr & CHAFSR_M_SYNDROME) >> CHAFSR_M_SYNDROME_SHIFT,
+@@ -1748,7 +1759,7 @@ void cheetah_plus_parity_error(int type, struct pt_regs *regs)
+ 		       smp_processor_id(),
+ 		       (type & 0x1) ? 'I' : 'D',
+ 		       regs->tpc);
+-		printk(KERN_EMERG "TPC<%pS>\n", (void *) regs->tpc);
++		printk(KERN_EMERG "TPC<%pA>\n", (void *) regs->tpc);
+ 		panic("Irrecoverable Cheetah+ parity error.");
+ 	}
+ 
+@@ -1756,7 +1767,7 @@ void cheetah_plus_parity_error(int type, struct pt_regs *regs)
+ 	       smp_processor_id(),
+ 	       (type & 0x1) ? 'I' : 'D',
+ 	       regs->tpc);
+-	printk(KERN_WARNING "TPC<%pS>\n", (void *) regs->tpc);
++	printk(KERN_WARNING "TPC<%pA>\n", (void *) regs->tpc);
+ }
+ 
+ struct sun4v_error_entry {
+@@ -1963,9 +1974,9 @@ void sun4v_itlb_error_report(struct pt_regs *regs, int tl)
+ 
+ 	printk(KERN_EMERG "SUN4V-ITLB: Error at TPC[%lx], tl %d\n",
+ 	       regs->tpc, tl);
+-	printk(KERN_EMERG "SUN4V-ITLB: TPC<%pS>\n", (void *) regs->tpc);
++	printk(KERN_EMERG "SUN4V-ITLB: TPC<%pA>\n", (void *) regs->tpc);
+ 	printk(KERN_EMERG "SUN4V-ITLB: O7[%lx]\n", regs->u_regs[UREG_I7]);
+-	printk(KERN_EMERG "SUN4V-ITLB: O7<%pS>\n",
++	printk(KERN_EMERG "SUN4V-ITLB: O7<%pA>\n",
+ 	       (void *) regs->u_regs[UREG_I7]);
+ 	printk(KERN_EMERG "SUN4V-ITLB: vaddr[%lx] ctx[%lx] "
+ 	       "pte[%lx] error[%lx]\n",
+@@ -1987,9 +1998,9 @@ void sun4v_dtlb_error_report(struct pt_regs *regs, int tl)
+ 
+ 	printk(KERN_EMERG "SUN4V-DTLB: Error at TPC[%lx], tl %d\n",
+ 	       regs->tpc, tl);
+-	printk(KERN_EMERG "SUN4V-DTLB: TPC<%pS>\n", (void *) regs->tpc);
++	printk(KERN_EMERG "SUN4V-DTLB: TPC<%pA>\n", (void *) regs->tpc);
+ 	printk(KERN_EMERG "SUN4V-DTLB: O7[%lx]\n", regs->u_regs[UREG_I7]);
+-	printk(KERN_EMERG "SUN4V-DTLB: O7<%pS>\n",
++	printk(KERN_EMERG "SUN4V-DTLB: O7<%pA>\n",
+ 	       (void *) regs->u_regs[UREG_I7]);
+ 	printk(KERN_EMERG "SUN4V-DTLB: vaddr[%lx] ctx[%lx] "
+ 	       "pte[%lx] error[%lx]\n",
+@@ -2195,13 +2206,13 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp)
+ 			fp = (unsigned long)sf->fp + STACK_BIAS;
+ 		}
+ 
+-		printk(" [%016lx] %pS\n", pc, (void *) pc);
++		printk(" [%016lx] %pA\n", pc, (void *) pc);
+ #ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ 		if ((pc + 8UL) == (unsigned long) &return_to_handler) {
+ 			int index = tsk->curr_ret_stack;
+ 			if (tsk->ret_stack && index >= graph) {
+ 				pc = tsk->ret_stack[index - graph].ret;
+-				printk(" [%016lx] %pS\n", pc, (void *) pc);
++				printk(" [%016lx] %pA\n", pc, (void *) pc);
+ 				graph++;
+ 			}
+ 		}
+@@ -2226,6 +2237,8 @@ static inline struct reg_window *kernel_stack_up(struct reg_window *rw)
+ 	return (struct reg_window *) (fp + STACK_BIAS);
+ }
+ 
++extern void gr_handle_kernel_exploit(void);
++
+ void die_if_kernel(char *str, struct pt_regs *regs)
+ {
+ 	static int die_counter;
+@@ -2254,7 +2267,7 @@ void die_if_kernel(char *str, struct pt_regs *regs)
+ 		while (rw &&
+ 		       count++ < 30 &&
+ 		       kstack_valid(tp, (unsigned long) rw)) {
+-			printk("Caller[%016lx]: %pS\n", rw->ins[7],
++			printk("Caller[%016lx]: %pA\n", rw->ins[7],
+ 			       (void *) rw->ins[7]);
+ 
+ 			rw = kernel_stack_up(rw);
+@@ -2267,8 +2280,10 @@ void die_if_kernel(char *str, struct pt_regs *regs)
+ 		}
+ 		user_instruction_dump ((unsigned int __user *) regs->tpc);
+ 	}
+-	if (regs->tstate & TSTATE_PRIV)
++	if (regs->tstate & TSTATE_PRIV) {
++		gr_handle_kernel_exploit();
+ 		do_exit(SIGKILL);
++	}
+ 	do_exit(SIGSEGV);
+ }
+ EXPORT_SYMBOL(die_if_kernel);
+diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c
+index 76e4ac1..78f8bb1 100644
+--- a/arch/sparc/kernel/unaligned_64.c
++++ b/arch/sparc/kernel/unaligned_64.c
+@@ -279,7 +279,7 @@ static void log_unaligned(struct pt_regs *regs)
+ 	static DEFINE_RATELIMIT_STATE(ratelimit, 5 * HZ, 5);
+ 
+ 	if (__ratelimit(&ratelimit)) {
+-		printk("Kernel unaligned access at TPC[%lx] %pS\n",
++		printk("Kernel unaligned access at TPC[%lx] %pA\n",
+ 		       regs->tpc, (void *) regs->tpc);
+ 	}
+ }
+diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile
+index a3fc437..fea9957 100644
+--- a/arch/sparc/lib/Makefile
++++ b/arch/sparc/lib/Makefile
+@@ -2,7 +2,7 @@
+ #
+ 
+ asflags-y := -ansi -DST_DIV0=0x02
+-ccflags-y := -Werror
++#ccflags-y := -Werror
+ 
+ lib-$(CONFIG_SPARC32) += mul.o rem.o sdiv.o udiv.o umul.o urem.o ashrdi3.o
+ lib-$(CONFIG_SPARC32) += memcpy.o memset.o
+diff --git a/arch/sparc/lib/atomic_64.S b/arch/sparc/lib/atomic_64.S
+index 59186e0..f747d7a 100644
+--- a/arch/sparc/lib/atomic_64.S
++++ b/arch/sparc/lib/atomic_64.S
+@@ -18,7 +18,12 @@
+ atomic_add: /* %o0 = increment, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	lduw	[%o1], %g1
+-	add	%g1, %o0, %g7
++	addcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%icc, 6
++#endif
++
+ 	cas	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%icc, BACKOFF_LABEL(2f, 1b)
+@@ -28,12 +33,32 @@ atomic_add: /* %o0 = increment, %o1 = atomic_ptr */
+ 2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ 	.size	atomic_add, .-atomic_add
+ 
++	.globl	atomic_add_unchecked
++	.type	atomic_add_unchecked,#function
++atomic_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */
++	BACKOFF_SETUP(%o2)
++1:	lduw	[%o1], %g1
++	add	%g1, %o0, %g7
++	cas	[%o1], %g1, %g7
++	cmp	%g1, %g7
++	bne,pn	%icc, 2f
++	 nop
++	retl
++	 nop
++2:	BACKOFF_SPIN(%o2, %o3, 1b)
++	.size	atomic_add_unchecked, .-atomic_add_unchecked
++
+ 	.globl	atomic_sub
+ 	.type	atomic_sub,#function
+ atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	lduw	[%o1], %g1
+-	sub	%g1, %o0, %g7
++	subcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%icc, 6
++#endif
++
+ 	cas	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%icc, BACKOFF_LABEL(2f, 1b)
+@@ -43,12 +68,32 @@ atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */
+ 2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ 	.size	atomic_sub, .-atomic_sub
+ 
++	.globl	atomic_sub_unchecked
++	.type	atomic_sub_unchecked,#function
++atomic_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */
++	BACKOFF_SETUP(%o2)
++1:	lduw	[%o1], %g1
++	sub	%g1, %o0, %g7
++	cas	[%o1], %g1, %g7
++	cmp	%g1, %g7
++	bne,pn	%icc, 2f
++	 nop
++	retl
++	 nop
++2:	BACKOFF_SPIN(%o2, %o3, 1b)
++	.size	atomic_sub_unchecked, .-atomic_sub_unchecked
++
+ 	.globl	atomic_add_ret
+ 	.type	atomic_add_ret,#function
+ atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	lduw	[%o1], %g1
+-	add	%g1, %o0, %g7
++	addcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%icc, 6
++#endif
++
+ 	cas	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%icc, BACKOFF_LABEL(2f, 1b)
+@@ -58,12 +103,33 @@ atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
+ 2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ 	.size	atomic_add_ret, .-atomic_add_ret
+ 
++	.globl	atomic_add_ret_unchecked
++	.type	atomic_add_ret_unchecked,#function
++atomic_add_ret_unchecked: /* %o0 = increment, %o1 = atomic_ptr */
++	BACKOFF_SETUP(%o2)
++1:	lduw	[%o1], %g1
++	addcc	%g1, %o0, %g7
++	cas	[%o1], %g1, %g7
++	cmp	%g1, %g7
++	bne,pn	%icc, 2f
++	 add	%g7, %o0, %g7
++	sra	%g7, 0, %o0
++	retl
++	 nop
++2:	BACKOFF_SPIN(%o2, %o3, 1b)
++	.size	atomic_add_ret_unchecked, .-atomic_add_ret_unchecked
++
+ 	.globl	atomic_sub_ret
+ 	.type	atomic_sub_ret,#function
+ atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	lduw	[%o1], %g1
+-	sub	%g1, %o0, %g7
++	subcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%icc, 6
++#endif
++
+ 	cas	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%icc, BACKOFF_LABEL(2f, 1b)
+@@ -78,7 +144,12 @@ atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
+ atomic64_add: /* %o0 = increment, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	ldx	[%o1], %g1
+-	add	%g1, %o0, %g7
++	addcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%xcc, 6
++#endif
++
+ 	casx	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%xcc, BACKOFF_LABEL(2f, 1b)
+@@ -88,12 +159,32 @@ atomic64_add: /* %o0 = increment, %o1 = atomic_ptr */
+ 2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ 	.size	atomic64_add, .-atomic64_add
+ 
++	.globl	atomic64_add_unchecked
++	.type	atomic64_add_unchecked,#function
++atomic64_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */
++	BACKOFF_SETUP(%o2)
++1:	ldx	[%o1], %g1
++	addcc	%g1, %o0, %g7
++	casx	[%o1], %g1, %g7
++	cmp	%g1, %g7
++	bne,pn	%xcc, 2f
++	 nop
++	retl
++	 nop
++2:	BACKOFF_SPIN(%o2, %o3, 1b)
++	.size	atomic64_add_unchecked, .-atomic64_add_unchecked
++
+ 	.globl	atomic64_sub
+ 	.type	atomic64_sub,#function
+ atomic64_sub: /* %o0 = decrement, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	ldx	[%o1], %g1
+-	sub	%g1, %o0, %g7
++	subcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%xcc, 6
++#endif
++
+ 	casx	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%xcc, BACKOFF_LABEL(2f, 1b)
+@@ -103,12 +194,32 @@ atomic64_sub: /* %o0 = decrement, %o1 = atomic_ptr */
+ 2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ 	.size	atomic64_sub, .-atomic64_sub
+ 
++	.globl	atomic64_sub_unchecked
++	.type	atomic64_sub_unchecked,#function
++atomic64_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */
++	BACKOFF_SETUP(%o2)
++1:	ldx	[%o1], %g1
++	subcc	%g1, %o0, %g7
++	casx	[%o1], %g1, %g7
++	cmp	%g1, %g7
++	bne,pn	%xcc, 2f
++	 nop
++	retl
++	 nop
++2:	BACKOFF_SPIN(%o2, %o3, 1b)
++	.size	atomic64_sub_unchecked, .-atomic64_sub_unchecked
++
+ 	.globl	atomic64_add_ret
+ 	.type	atomic64_add_ret,#function
+ atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	ldx	[%o1], %g1
+-	add	%g1, %o0, %g7
++	addcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%xcc, 6
++#endif
++
+ 	casx	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%xcc, BACKOFF_LABEL(2f, 1b)
+@@ -118,12 +229,33 @@ atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */
+ 2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ 	.size	atomic64_add_ret, .-atomic64_add_ret
+ 
++	.globl	atomic64_add_ret_unchecked
++	.type	atomic64_add_ret_unchecked,#function
++atomic64_add_ret_unchecked: /* %o0 = increment, %o1 = atomic_ptr */
++	BACKOFF_SETUP(%o2)
++1:	ldx	[%o1], %g1
++	addcc	%g1, %o0, %g7
++	casx	[%o1], %g1, %g7
++	cmp	%g1, %g7
++	bne,pn	%xcc, 2f
++	 add	%g7, %o0, %g7
++	mov	%g7, %o0
++	retl
++	 nop
++2:	BACKOFF_SPIN(%o2, %o3, 1b)
++	.size	atomic64_add_ret_unchecked, .-atomic64_add_ret_unchecked
++
+ 	.globl	atomic64_sub_ret
+ 	.type	atomic64_sub_ret,#function
+ atomic64_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */
+ 	BACKOFF_SETUP(%o2)
+ 1:	ldx	[%o1], %g1
+-	sub	%g1, %o0, %g7
++	subcc	%g1, %o0, %g7
++
++#ifdef CONFIG_PAX_REFCOUNT
++	tvs	%xcc, 6
++#endif
++
+ 	casx	[%o1], %g1, %g7
+ 	cmp	%g1, %g7
+ 	bne,pn	%xcc, BACKOFF_LABEL(2f, 1b)
+diff --git a/arch/sparc/lib/ksyms.c b/arch/sparc/lib/ksyms.c
+index 1b30bb3..b4a16c7 100644
+--- a/arch/sparc/lib/ksyms.c
++++ b/arch/sparc/lib/ksyms.c
+@@ -142,12 +142,18 @@ EXPORT_SYMBOL(__downgrade_write);
+ 
+ /* Atomic counter implementation. */
+ EXPORT_SYMBOL(atomic_add);
++EXPORT_SYMBOL(atomic_add_unchecked);
+ EXPORT_SYMBOL(atomic_add_ret);
++EXPORT_SYMBOL(atomic_add_ret_unchecked);
+ EXPORT_SYMBOL(atomic_sub);
++EXPORT_SYMBOL(atomic_sub_unchecked);
+ EXPORT_SYMBOL(atomic_sub_ret);
+ EXPORT_SYMBOL(atomic64_add);
++EXPORT_SYMBOL(atomic64_add_unchecked);
+ EXPORT_SYMBOL(atomic64_add_ret);
++EXPORT_SYMBOL(atomic64_add_ret_unchecked);
+ EXPORT_SYMBOL(atomic64_sub);
++EXPORT_SYMBOL(atomic64_sub_unchecked);
+ EXPORT_SYMBOL(atomic64_sub_ret);
+ 
+ /* Atomic bit operations. */
+diff --git a/arch/sparc/mm/Makefile b/arch/sparc/mm/Makefile
+index 301421c..e2535d1 100644
+--- a/arch/sparc/mm/Makefile
++++ b/arch/sparc/mm/Makefile
+@@ -2,7 +2,7 @@
+ #
+ 
+ asflags-y := -ansi
+-ccflags-y := -Werror
++#ccflags-y := -Werror
+ 
+ obj-$(CONFIG_SPARC64)   += ultra.o tlb.o tsb.o gup.o
+ obj-y                   += fault_$(BITS).o
+diff --git a/arch/sparc/mm/fault_32.c b/arch/sparc/mm/fault_32.c
+index 8023fd7..c8e89e9 100644
+--- a/arch/sparc/mm/fault_32.c
++++ b/arch/sparc/mm/fault_32.c
+@@ -21,6 +21,9 @@
+ #include <linux/perf_event.h>
+ #include <linux/interrupt.h>
+ #include <linux/kdebug.h>
++#include <linux/slab.h>
++#include <linux/pagemap.h>
++#include <linux/compiler.h>
+ 
+ #include <asm/system.h>
+ #include <asm/page.h>
+@@ -208,6 +211,268 @@ static unsigned long compute_si_addr(struct pt_regs *regs, int text_fault)
+ 	return safe_compute_effective_address(regs, insn);
+ }
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++#ifdef CONFIG_PAX_DLRESOLVE
++static void pax_emuplt_close(struct vm_area_struct *vma)
++{
++	vma->vm_mm->call_dl_resolve = 0UL;
++}
++
++static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
++{
++	unsigned int *kaddr;
++
++	vmf->page = alloc_page(GFP_HIGHUSER);
++	if (!vmf->page)
++		return VM_FAULT_OOM;
++
++	kaddr = kmap(vmf->page);
++	memset(kaddr, 0, PAGE_SIZE);
++	kaddr[0] = 0x9DE3BFA8U; /* save */
++	flush_dcache_page(vmf->page);
++	kunmap(vmf->page);
++	return VM_FAULT_MAJOR;
++}
++
++static const struct vm_operations_struct pax_vm_ops = {
++	.close = pax_emuplt_close,
++	.fault = pax_emuplt_fault
++};
++
++static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr)
++{
++	int ret;
++
++	INIT_LIST_HEAD(&vma->anon_vma_chain);
++	vma->vm_mm = current->mm;
++	vma->vm_start = addr;
++	vma->vm_end = addr + PAGE_SIZE;
++	vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC;
++	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
++	vma->vm_ops = &pax_vm_ops;
++
++	ret = insert_vm_struct(current->mm, vma);
++	if (ret)
++		return ret;
++
++	++current->mm->total_vm;
++	return 0;
++}
++#endif
++
++/*
++ * PaX: decide what to do with offenders (regs->pc = fault address)
++ *
++ * returns 1 when task should be killed
++ *         2 when patched PLT trampoline was detected
++ *         3 when unpatched PLT trampoline was detected
++ */
++static int pax_handle_fetch_fault(struct pt_regs *regs)
++{
++
++#ifdef CONFIG_PAX_EMUPLT
++	int err;
++
++	do { /* PaX: patched PLT emulation #1 */
++		unsigned int sethi1, sethi2, jmpl;
++
++		err = get_user(sethi1, (unsigned int *)regs->pc);
++		err |= get_user(sethi2, (unsigned int *)(regs->pc+4));
++		err |= get_user(jmpl, (unsigned int *)(regs->pc+8));
++
++		if (err)
++			break;
++
++		if ((sethi1 & 0xFFC00000U) == 0x03000000U &&
++		    (sethi2 & 0xFFC00000U) == 0x03000000U &&
++		    (jmpl & 0xFFFFE000U) == 0x81C06000U)
++		{
++			unsigned int addr;
++
++			regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10;
++			addr = regs->u_regs[UREG_G1];
++			addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U);
++			regs->pc = addr;
++			regs->npc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	{ /* PaX: patched PLT emulation #2 */
++		unsigned int ba;
++
++		err = get_user(ba, (unsigned int *)regs->pc);
++
++		if (!err && (ba & 0xFFC00000U) == 0x30800000U) {
++			unsigned int addr;
++
++			addr = regs->pc + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2);
++			regs->pc = addr;
++			regs->npc = addr+4;
++			return 2;
++		}
++	}
++
++	do { /* PaX: patched PLT emulation #3 */
++		unsigned int sethi, jmpl, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->pc);
++		err |= get_user(jmpl, (unsigned int *)(regs->pc+4));
++		err |= get_user(nop, (unsigned int *)(regs->pc+8));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    (jmpl & 0xFFFFE000U) == 0x81C06000U &&
++		    nop == 0x01000000U)
++		{
++			unsigned int addr;
++
++			addr = (sethi & 0x003FFFFFU) << 10;
++			regs->u_regs[UREG_G1] = addr;
++			addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U);
++			regs->pc = addr;
++			regs->npc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: unpatched PLT emulation step 1 */
++		unsigned int sethi, ba, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->pc);
++		err |= get_user(ba, (unsigned int *)(regs->pc+4));
++		err |= get_user(nop, (unsigned int *)(regs->pc+8));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) &&
++		    nop == 0x01000000U)
++		{
++			unsigned int addr, save, call;
++
++			if ((ba & 0xFFC00000U) == 0x30800000U)
++				addr = regs->pc + 4 + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2);
++			else
++				addr = regs->pc + 4 + ((((ba | 0xFFF80000U) ^ 0x00040000U) + 0x00040000U) << 2);
++
++			err = get_user(save, (unsigned int *)addr);
++			err |= get_user(call, (unsigned int *)(addr+4));
++			err |= get_user(nop, (unsigned int *)(addr+8));
++			if (err)
++				break;
++
++#ifdef CONFIG_PAX_DLRESOLVE
++			if (save == 0x9DE3BFA8U &&
++			    (call & 0xC0000000U) == 0x40000000U &&
++			    nop == 0x01000000U)
++			{
++				struct vm_area_struct *vma;
++				unsigned long call_dl_resolve;
++
++				down_read(&current->mm->mmap_sem);
++				call_dl_resolve = current->mm->call_dl_resolve;
++				up_read(&current->mm->mmap_sem);
++				if (likely(call_dl_resolve))
++					goto emulate;
++
++				vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
++
++				down_write(&current->mm->mmap_sem);
++				if (current->mm->call_dl_resolve) {
++					call_dl_resolve = current->mm->call_dl_resolve;
++					up_write(&current->mm->mmap_sem);
++					if (vma)
++						kmem_cache_free(vm_area_cachep, vma);
++					goto emulate;
++				}
++
++				call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE);
++				if (!vma || (call_dl_resolve & ~PAGE_MASK)) {
++					up_write(&current->mm->mmap_sem);
++					if (vma)
++						kmem_cache_free(vm_area_cachep, vma);
++					return 1;
++				}
++
++				if (pax_insert_vma(vma, call_dl_resolve)) {
++					up_write(&current->mm->mmap_sem);
++					kmem_cache_free(vm_area_cachep, vma);
++					return 1;
++				}
++
++				current->mm->call_dl_resolve = call_dl_resolve;
++				up_write(&current->mm->mmap_sem);
++
++emulate:
++				regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10;
++				regs->pc = call_dl_resolve;
++				regs->npc = addr+4;
++				return 3;
++			}
++#endif
++
++			/* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */
++			if ((save & 0xFFC00000U) == 0x05000000U &&
++			    (call & 0xFFFFE000U) == 0x85C0A000U &&
++			    nop == 0x01000000U)
++			{
++				regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10;
++				regs->u_regs[UREG_G2] = addr + 4;
++				addr = (save & 0x003FFFFFU) << 10;
++				addr += (((call | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U);
++				regs->pc = addr;
++				regs->npc = addr+4;
++				return 3;
++			}
++		}
++	} while (0);
++
++	do { /* PaX: unpatched PLT emulation step 2 */
++		unsigned int save, call, nop;
++
++		err = get_user(save, (unsigned int *)(regs->pc-4));
++		err |= get_user(call, (unsigned int *)regs->pc);
++		err |= get_user(nop, (unsigned int *)(regs->pc+4));
++		if (err)
++			break;
++
++		if (save == 0x9DE3BFA8U &&
++		    (call & 0xC0000000U) == 0x40000000U &&
++		    nop == 0x01000000U)
++		{
++			unsigned int dl_resolve = regs->pc + ((((call | 0xC0000000U) ^ 0x20000000U) + 0x20000000U) << 2);
++
++			regs->u_regs[UREG_RETPC] = regs->pc;
++			regs->pc = dl_resolve;
++			regs->npc = dl_resolve+4;
++			return 3;
++		}
++	} while (0);
++#endif
++
++	return 1;
++}
++
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 8; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ static noinline void do_fault_siginfo(int code, int sig, struct pt_regs *regs,
+ 				      int text_fault)
+ {
+@@ -280,6 +545,24 @@ good_area:
+ 		if(!(vma->vm_flags & VM_WRITE))
+ 			goto bad_area;
+ 	} else {
++
++#ifdef CONFIG_PAX_PAGEEXEC
++		if ((mm->pax_flags & MF_PAX_PAGEEXEC) && text_fault && !(vma->vm_flags & VM_EXEC)) {
++			up_read(&mm->mmap_sem);
++			switch (pax_handle_fetch_fault(regs)) {
++
++#ifdef CONFIG_PAX_EMUPLT
++			case 2:
++			case 3:
++				return;
++#endif
++
++			}
++			pax_report_fault(regs, (void *)regs->pc, (void *)regs->u_regs[UREG_FP]);
++			do_group_exit(SIGKILL);
++		}
++#endif
++
+ 		/* Allow reads even for write-only mappings */
+ 		if(!(vma->vm_flags & (VM_READ | VM_EXEC)))
+ 			goto bad_area;
+diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c
+index 504c062..6fcb9c6 100644
+--- a/arch/sparc/mm/fault_64.c
++++ b/arch/sparc/mm/fault_64.c
+@@ -21,6 +21,9 @@
+ #include <linux/kprobes.h>
+ #include <linux/kdebug.h>
+ #include <linux/percpu.h>
++#include <linux/slab.h>
++#include <linux/pagemap.h>
++#include <linux/compiler.h>
+ 
+ #include <asm/page.h>
+ #include <asm/pgtable.h>
+@@ -74,7 +77,7 @@ static void __kprobes bad_kernel_pc(struct pt_regs *regs, unsigned long vaddr)
+ 	printk(KERN_CRIT "OOPS: Bogus kernel PC [%016lx] in fault handler\n",
+ 	       regs->tpc);
+ 	printk(KERN_CRIT "OOPS: RPC [%016lx]\n", regs->u_regs[15]);
+-	printk("OOPS: RPC <%pS>\n", (void *) regs->u_regs[15]);
++	printk("OOPS: RPC <%pA>\n", (void *) regs->u_regs[15]);
+ 	printk(KERN_CRIT "OOPS: Fault was to vaddr[%lx]\n", vaddr);
+ 	dump_stack();
+ 	unhandled_fault(regs->tpc, current, regs);
+@@ -272,6 +275,457 @@ static void noinline __kprobes bogus_32bit_fault_address(struct pt_regs *regs,
+ 	show_regs(regs);
+ }
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++#ifdef CONFIG_PAX_DLRESOLVE
++static void pax_emuplt_close(struct vm_area_struct *vma)
++{
++	vma->vm_mm->call_dl_resolve = 0UL;
++}
++
++static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
++{
++	unsigned int *kaddr;
++
++	vmf->page = alloc_page(GFP_HIGHUSER);
++	if (!vmf->page)
++		return VM_FAULT_OOM;
++
++	kaddr = kmap(vmf->page);
++	memset(kaddr, 0, PAGE_SIZE);
++	kaddr[0] = 0x9DE3BFA8U; /* save */
++	flush_dcache_page(vmf->page);
++	kunmap(vmf->page);
++	return VM_FAULT_MAJOR;
++}
++
++static const struct vm_operations_struct pax_vm_ops = {
++	.close = pax_emuplt_close,
++	.fault = pax_emuplt_fault
++};
++
++static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr)
++{
++	int ret;
++
++	INIT_LIST_HEAD(&vma->anon_vma_chain);
++	vma->vm_mm = current->mm;
++	vma->vm_start = addr;
++	vma->vm_end = addr + PAGE_SIZE;
++	vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC;
++	vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
++	vma->vm_ops = &pax_vm_ops;
++
++	ret = insert_vm_struct(current->mm, vma);
++	if (ret)
++		return ret;
++
++	++current->mm->total_vm;
++	return 0;
++}
++#endif
++
++/*
++ * PaX: decide what to do with offenders (regs->tpc = fault address)
++ *
++ * returns 1 when task should be killed
++ *         2 when patched PLT trampoline was detected
++ *         3 when unpatched PLT trampoline was detected
++ */
++static int pax_handle_fetch_fault(struct pt_regs *regs)
++{
++
++#ifdef CONFIG_PAX_EMUPLT
++	int err;
++
++	do { /* PaX: patched PLT emulation #1 */
++		unsigned int sethi1, sethi2, jmpl;
++
++		err = get_user(sethi1, (unsigned int *)regs->tpc);
++		err |= get_user(sethi2, (unsigned int *)(regs->tpc+4));
++		err |= get_user(jmpl, (unsigned int *)(regs->tpc+8));
++
++		if (err)
++			break;
++
++		if ((sethi1 & 0xFFC00000U) == 0x03000000U &&
++		    (sethi2 & 0xFFC00000U) == 0x03000000U &&
++		    (jmpl & 0xFFFFE000U) == 0x81C06000U)
++		{
++			unsigned long addr;
++
++			regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10;
++			addr = regs->u_regs[UREG_G1];
++			addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL);
++
++			if (test_thread_flag(TIF_32BIT))
++				addr &= 0xFFFFFFFFUL;
++
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	{ /* PaX: patched PLT emulation #2 */
++		unsigned int ba;
++
++		err = get_user(ba, (unsigned int *)regs->tpc);
++
++		if (!err && (ba & 0xFFC00000U) == 0x30800000U) {
++			unsigned long addr;
++
++			addr = regs->tpc + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2);
++
++			if (test_thread_flag(TIF_32BIT))
++				addr &= 0xFFFFFFFFUL;
++
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	}
++
++	do { /* PaX: patched PLT emulation #3 */
++		unsigned int sethi, jmpl, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->tpc);
++		err |= get_user(jmpl, (unsigned int *)(regs->tpc+4));
++		err |= get_user(nop, (unsigned int *)(regs->tpc+8));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    (jmpl & 0xFFFFE000U) == 0x81C06000U &&
++		    nop == 0x01000000U)
++		{
++			unsigned long addr;
++
++			addr = (sethi & 0x003FFFFFU) << 10;
++			regs->u_regs[UREG_G1] = addr;
++			addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL);
++
++			if (test_thread_flag(TIF_32BIT))
++				addr &= 0xFFFFFFFFUL;
++
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: patched PLT emulation #4 */
++		unsigned int sethi, mov1, call, mov2;
++
++		err = get_user(sethi, (unsigned int *)regs->tpc);
++		err |= get_user(mov1, (unsigned int *)(regs->tpc+4));
++		err |= get_user(call, (unsigned int *)(regs->tpc+8));
++		err |= get_user(mov2, (unsigned int *)(regs->tpc+12));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    mov1 == 0x8210000FU &&
++		    (call & 0xC0000000U) == 0x40000000U &&
++		    mov2 == 0x9E100001U)
++		{
++			unsigned long addr;
++
++			regs->u_regs[UREG_G1] = regs->u_regs[UREG_RETPC];
++			addr = regs->tpc + 4 + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2);
++
++			if (test_thread_flag(TIF_32BIT))
++				addr &= 0xFFFFFFFFUL;
++
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: patched PLT emulation #5 */
++		unsigned int sethi, sethi1, sethi2, or1, or2, sllx, jmpl, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->tpc);
++		err |= get_user(sethi1, (unsigned int *)(regs->tpc+4));
++		err |= get_user(sethi2, (unsigned int *)(regs->tpc+8));
++		err |= get_user(or1, (unsigned int *)(regs->tpc+12));
++		err |= get_user(or2, (unsigned int *)(regs->tpc+16));
++		err |= get_user(sllx, (unsigned int *)(regs->tpc+20));
++		err |= get_user(jmpl, (unsigned int *)(regs->tpc+24));
++		err |= get_user(nop, (unsigned int *)(regs->tpc+28));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    (sethi1 & 0xFFC00000U) == 0x03000000U &&
++		    (sethi2 & 0xFFC00000U) == 0x0B000000U &&
++		    (or1 & 0xFFFFE000U) == 0x82106000U &&
++		    (or2 & 0xFFFFE000U) == 0x8A116000U &&
++		    sllx == 0x83287020U &&
++		    jmpl == 0x81C04005U &&
++		    nop == 0x01000000U)
++		{
++			unsigned long addr;
++
++			regs->u_regs[UREG_G1] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU);
++			regs->u_regs[UREG_G1] <<= 32;
++			regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU);
++			addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5];
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: patched PLT emulation #6 */
++		unsigned int sethi, sethi1, sethi2, sllx, or,  jmpl, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->tpc);
++		err |= get_user(sethi1, (unsigned int *)(regs->tpc+4));
++		err |= get_user(sethi2, (unsigned int *)(regs->tpc+8));
++		err |= get_user(sllx, (unsigned int *)(regs->tpc+12));
++		err |= get_user(or, (unsigned int *)(regs->tpc+16));
++		err |= get_user(jmpl, (unsigned int *)(regs->tpc+20));
++		err |= get_user(nop, (unsigned int *)(regs->tpc+24));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    (sethi1 & 0xFFC00000U) == 0x03000000U &&
++		    (sethi2 & 0xFFC00000U) == 0x0B000000U &&
++		    sllx == 0x83287020U &&
++		    (or & 0xFFFFE000U) == 0x8A116000U &&
++		    jmpl == 0x81C04005U &&
++		    nop == 0x01000000U)
++		{
++			unsigned long addr;
++
++			regs->u_regs[UREG_G1] = (sethi1 & 0x003FFFFFU) << 10;
++			regs->u_regs[UREG_G1] <<= 32;
++			regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or & 0x3FFU);
++			addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5];
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++	do { /* PaX: unpatched PLT emulation step 1 */
++		unsigned int sethi, ba, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->tpc);
++		err |= get_user(ba, (unsigned int *)(regs->tpc+4));
++		err |= get_user(nop, (unsigned int *)(regs->tpc+8));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) &&
++		    nop == 0x01000000U)
++		{
++			unsigned long addr;
++			unsigned int save, call;
++			unsigned int sethi1, sethi2, or1, or2, sllx, add, jmpl;
++
++			if ((ba & 0xFFC00000U) == 0x30800000U)
++				addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2);
++			else
++				addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2);
++
++			if (test_thread_flag(TIF_32BIT))
++				addr &= 0xFFFFFFFFUL;
++
++			err = get_user(save, (unsigned int *)addr);
++			err |= get_user(call, (unsigned int *)(addr+4));
++			err |= get_user(nop, (unsigned int *)(addr+8));
++			if (err)
++				break;
++
++#ifdef CONFIG_PAX_DLRESOLVE
++			if (save == 0x9DE3BFA8U &&
++			    (call & 0xC0000000U) == 0x40000000U &&
++			    nop == 0x01000000U)
++			{
++				struct vm_area_struct *vma;
++				unsigned long call_dl_resolve;
++
++				down_read(&current->mm->mmap_sem);
++				call_dl_resolve = current->mm->call_dl_resolve;
++				up_read(&current->mm->mmap_sem);
++				if (likely(call_dl_resolve))
++					goto emulate;
++
++				vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
++
++				down_write(&current->mm->mmap_sem);
++				if (current->mm->call_dl_resolve) {
++					call_dl_resolve = current->mm->call_dl_resolve;
++					up_write(&current->mm->mmap_sem);
++					if (vma)
++						kmem_cache_free(vm_area_cachep, vma);
++					goto emulate;
++				}
++
++				call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE);
++				if (!vma || (call_dl_resolve & ~PAGE_MASK)) {
++					up_write(&current->mm->mmap_sem);
++					if (vma)
++						kmem_cache_free(vm_area_cachep, vma);
++					return 1;
++				}
++
++				if (pax_insert_vma(vma, call_dl_resolve)) {
++					up_write(&current->mm->mmap_sem);
++					kmem_cache_free(vm_area_cachep, vma);
++					return 1;
++				}
++
++				current->mm->call_dl_resolve = call_dl_resolve;
++				up_write(&current->mm->mmap_sem);
++
++emulate:
++				regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10;
++				regs->tpc = call_dl_resolve;
++				regs->tnpc = addr+4;
++				return 3;
++			}
++#endif
++
++			/* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */
++			if ((save & 0xFFC00000U) == 0x05000000U &&
++			    (call & 0xFFFFE000U) == 0x85C0A000U &&
++			    nop == 0x01000000U)
++			{
++				regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10;
++				regs->u_regs[UREG_G2] = addr + 4;
++				addr = (save & 0x003FFFFFU) << 10;
++				addr += (((call | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL);
++
++				if (test_thread_flag(TIF_32BIT))
++					addr &= 0xFFFFFFFFUL;
++
++				regs->tpc = addr;
++				regs->tnpc = addr+4;
++				return 3;
++			}
++
++			/* PaX: 64-bit PLT stub */
++			err = get_user(sethi1, (unsigned int *)addr);
++			err |= get_user(sethi2, (unsigned int *)(addr+4));
++			err |= get_user(or1, (unsigned int *)(addr+8));
++			err |= get_user(or2, (unsigned int *)(addr+12));
++			err |= get_user(sllx, (unsigned int *)(addr+16));
++			err |= get_user(add, (unsigned int *)(addr+20));
++			err |= get_user(jmpl, (unsigned int *)(addr+24));
++			err |= get_user(nop, (unsigned int *)(addr+28));
++			if (err)
++				break;
++
++			if ((sethi1 & 0xFFC00000U) == 0x09000000U &&
++			    (sethi2 & 0xFFC00000U) == 0x0B000000U &&
++			    (or1 & 0xFFFFE000U) == 0x88112000U &&
++			    (or2 & 0xFFFFE000U) == 0x8A116000U &&
++			    sllx == 0x89293020U &&
++			    add == 0x8A010005U &&
++			    jmpl == 0x89C14000U &&
++			    nop == 0x01000000U)
++			{
++				regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10;
++				regs->u_regs[UREG_G4] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU);
++				regs->u_regs[UREG_G4] <<= 32;
++				regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU);
++				regs->u_regs[UREG_G5] += regs->u_regs[UREG_G4];
++				regs->u_regs[UREG_G4] = addr + 24;
++				addr = regs->u_regs[UREG_G5];
++				regs->tpc = addr;
++				regs->tnpc = addr+4;
++				return 3;
++			}
++		}
++	} while (0);
++
++#ifdef CONFIG_PAX_DLRESOLVE
++	do { /* PaX: unpatched PLT emulation step 2 */
++		unsigned int save, call, nop;
++
++		err = get_user(save, (unsigned int *)(regs->tpc-4));
++		err |= get_user(call, (unsigned int *)regs->tpc);
++		err |= get_user(nop, (unsigned int *)(regs->tpc+4));
++		if (err)
++			break;
++
++		if (save == 0x9DE3BFA8U &&
++		    (call & 0xC0000000U) == 0x40000000U &&
++		    nop == 0x01000000U)
++		{
++			unsigned long dl_resolve = regs->tpc + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2);
++
++			if (test_thread_flag(TIF_32BIT))
++				dl_resolve &= 0xFFFFFFFFUL;
++
++			regs->u_regs[UREG_RETPC] = regs->tpc;
++			regs->tpc = dl_resolve;
++			regs->tnpc = dl_resolve+4;
++			return 3;
++		}
++	} while (0);
++#endif
++
++	do { /* PaX: patched PLT emulation #7, must be AFTER the unpatched PLT emulation */
++		unsigned int sethi, ba, nop;
++
++		err = get_user(sethi, (unsigned int *)regs->tpc);
++		err |= get_user(ba, (unsigned int *)(regs->tpc+4));
++		err |= get_user(nop, (unsigned int *)(regs->tpc+8));
++
++		if (err)
++			break;
++
++		if ((sethi & 0xFFC00000U) == 0x03000000U &&
++		    (ba & 0xFFF00000U) == 0x30600000U &&
++		    nop == 0x01000000U)
++		{
++			unsigned long addr;
++
++			addr = (sethi & 0x003FFFFFU) << 10;
++			regs->u_regs[UREG_G1] = addr;
++			addr = regs->tpc + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2);
++
++			if (test_thread_flag(TIF_32BIT))
++				addr &= 0xFFFFFFFFUL;
++
++			regs->tpc = addr;
++			regs->tnpc = addr+4;
++			return 2;
++		}
++	} while (0);
++
++#endif
++
++	return 1;
++}
++
++void pax_report_insns(struct pt_regs *regs, void *pc, void *sp)
++{
++	unsigned long i;
++
++	printk(KERN_ERR "PAX: bytes at PC: ");
++	for (i = 0; i < 8; i++) {
++		unsigned int c;
++		if (get_user(c, (unsigned int *)pc+i))
++			printk(KERN_CONT "???????? ");
++		else
++			printk(KERN_CONT "%08x ", c);
++	}
++	printk("\n");
++}
++#endif
++
+ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
+ {
+ 	struct mm_struct *mm = current->mm;
+@@ -340,6 +794,29 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
+ 	if (!vma)
+ 		goto bad_area;
+ 
++#ifdef CONFIG_PAX_PAGEEXEC
++	/* PaX: detect ITLB misses on non-exec pages */
++	if ((mm->pax_flags & MF_PAX_PAGEEXEC) && vma->vm_start <= address &&
++	    !(vma->vm_flags & VM_EXEC) && (fault_code & FAULT_CODE_ITLB))
++	{
++		if (address != regs->tpc)
++			goto good_area;
++
++		up_read(&mm->mmap_sem);
++		switch (pax_handle_fetch_fault(regs)) {
++
++#ifdef CONFIG_PAX_EMUPLT
++		case 2:
++		case 3:
++			return;
++#endif
++
++		}
++		pax_report_fault(regs, (void *)regs->tpc, (void *)(regs->u_regs[UREG_FP] + STACK_BIAS));
++		do_group_exit(SIGKILL);
++	}
++#endif
++
+ 	/* Pure DTLB misses do not tell us whether the fault causing
+ 	 * load/store/atomic was a write or not, it only says that there
+ 	 * was no match.  So in such a case we (carefully) read the
+diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c
+index 07e1453..0a7d9e9 100644
+--- a/arch/sparc/mm/hugetlbpage.c
++++ b/arch/sparc/mm/hugetlbpage.c
+@@ -67,7 +67,7 @@ full_search:
+ 			}
+ 			return -ENOMEM;
+ 		}
+-		if (likely(!vma || addr + len <= vma->vm_start)) {
++		if (likely(check_heap_stack_gap(vma, addr, len))) {
+ 			/*
+ 			 * Remember the place where we stopped the search:
+ 			 */
+@@ -106,7 +106,7 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	/* make sure it can fit in the remaining address space */
+ 	if (likely(addr > len)) {
+ 		vma = find_vma(mm, addr-len);
+-		if (!vma || addr <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr - len, len)) {
+ 			/* remember the address as a hint for next time */
+ 			return (mm->free_area_cache = addr-len);
+ 		}
+@@ -115,16 +115,17 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	if (unlikely(mm->mmap_base < len))
+ 		goto bottomup;
+ 
+-	addr = (mm->mmap_base-len) & HPAGE_MASK;
++	addr = mm->mmap_base - len;
+ 
+ 	do {
++		addr &= HPAGE_MASK;
+ 		/*
+ 		 * Lookup failure means no vma is above this address,
+ 		 * else if new region fits below vma->vm_start,
+ 		 * return with success:
+ 		 */
+ 		vma = find_vma(mm, addr);
+-		if (likely(!vma || addr+len <= vma->vm_start)) {
++		if (likely(check_heap_stack_gap(vma, addr, len))) {
+ 			/* remember the address as a hint for next time */
+ 			return (mm->free_area_cache = addr);
+ 		}
+@@ -134,8 +135,8 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+  		        mm->cached_hole_size = vma->vm_start - addr;
+ 
+ 		/* try just below the current vma->vm_start */
+-		addr = (vma->vm_start-len) & HPAGE_MASK;
+-	} while (likely(len < vma->vm_start));
++		addr = skip_heap_stack_gap(vma, len);
++	} while (!IS_ERR_VALUE(addr));
+ 
+ bottomup:
+ 	/*
+@@ -181,8 +182,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
+ 	if (addr) {
+ 		addr = ALIGN(addr, HPAGE_SIZE);
+ 		vma = find_vma(mm, addr);
+-		if (task_size - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 	if (mm->get_unmapped_area == arch_get_unmapped_area)
+diff --git a/arch/sparc/mm/init_32.c b/arch/sparc/mm/init_32.c
+index 7b00de6..78239f4 100644
+--- a/arch/sparc/mm/init_32.c
++++ b/arch/sparc/mm/init_32.c
+@@ -316,6 +316,9 @@ extern void device_scan(void);
+ pgprot_t PAGE_SHARED __read_mostly;
+ EXPORT_SYMBOL(PAGE_SHARED);
+ 
++pgprot_t PAGE_SHARED_NOEXEC __read_mostly;
++EXPORT_SYMBOL(PAGE_SHARED_NOEXEC);
++
+ void __init paging_init(void)
+ {
+ 	switch(sparc_cpu_model) {
+@@ -344,17 +347,17 @@ void __init paging_init(void)
+ 
+ 	/* Initialize the protection map with non-constant, MMU dependent values. */
+ 	protection_map[0] = PAGE_NONE;
+-	protection_map[1] = PAGE_READONLY;
+-	protection_map[2] = PAGE_COPY;
+-	protection_map[3] = PAGE_COPY;
++	protection_map[1] = PAGE_READONLY_NOEXEC;
++	protection_map[2] = PAGE_COPY_NOEXEC;
++	protection_map[3] = PAGE_COPY_NOEXEC;
+ 	protection_map[4] = PAGE_READONLY;
+ 	protection_map[5] = PAGE_READONLY;
+ 	protection_map[6] = PAGE_COPY;
+ 	protection_map[7] = PAGE_COPY;
+ 	protection_map[8] = PAGE_NONE;
+-	protection_map[9] = PAGE_READONLY;
+-	protection_map[10] = PAGE_SHARED;
+-	protection_map[11] = PAGE_SHARED;
++	protection_map[9] = PAGE_READONLY_NOEXEC;
++	protection_map[10] = PAGE_SHARED_NOEXEC;
++	protection_map[11] = PAGE_SHARED_NOEXEC;
+ 	protection_map[12] = PAGE_READONLY;
+ 	protection_map[13] = PAGE_READONLY;
+ 	protection_map[14] = PAGE_SHARED;
+diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
+index cbef74e..c38fead 100644
+--- a/arch/sparc/mm/srmmu.c
++++ b/arch/sparc/mm/srmmu.c
+@@ -2200,6 +2200,13 @@ void __init ld_mmu_srmmu(void)
+ 	PAGE_SHARED = pgprot_val(SRMMU_PAGE_SHARED);
+ 	BTFIXUPSET_INT(page_copy, pgprot_val(SRMMU_PAGE_COPY));
+ 	BTFIXUPSET_INT(page_readonly, pgprot_val(SRMMU_PAGE_RDONLY));
++
++#ifdef CONFIG_PAX_PAGEEXEC
++	PAGE_SHARED_NOEXEC = pgprot_val(SRMMU_PAGE_SHARED_NOEXEC);
++	BTFIXUPSET_INT(page_copy_noexec, pgprot_val(SRMMU_PAGE_COPY_NOEXEC));
++	BTFIXUPSET_INT(page_readonly_noexec, pgprot_val(SRMMU_PAGE_RDONLY_NOEXEC));
++#endif
++
+ 	BTFIXUPSET_INT(page_kernel, pgprot_val(SRMMU_PAGE_KERNEL));
+ 	page_kernel = pgprot_val(SRMMU_PAGE_KERNEL);
+ 
+diff --git a/arch/tile/include/asm/atomic_64.h b/arch/tile/include/asm/atomic_64.h
+index 27fe667..36d474c 100644
+--- a/arch/tile/include/asm/atomic_64.h
++++ b/arch/tile/include/asm/atomic_64.h
+@@ -142,6 +142,16 @@ static inline long atomic64_add_unless(atomic64_t *v, long a, long u)
+ 
+ #define atomic64_inc_not_zero(v)	atomic64_add_unless((v), 1, 0)
+ 
++#define atomic64_read_unchecked(v)		atomic64_read(v)
++#define atomic64_set_unchecked(v, i)		atomic64_set((v), (i))
++#define atomic64_add_unchecked(a, v)		atomic64_add((a), (v))
++#define atomic64_add_return_unchecked(a, v)	atomic64_add_return((a), (v))
++#define atomic64_sub_unchecked(a, v)		atomic64_sub((a), (v))
++#define atomic64_inc_unchecked(v)		atomic64_inc(v)
++#define atomic64_inc_return_unchecked(v)	atomic64_inc_return(v)
++#define atomic64_dec_unchecked(v)		atomic64_dec(v)
++#define atomic64_cmpxchg_unchecked(v, o, n)	atomic64_cmpxchg((v), (o), (n))
++
+ /* Atomic dec and inc don't implement barrier, so provide them if needed. */
+ #define smp_mb__before_atomic_dec()	smp_mb()
+ #define smp_mb__after_atomic_dec()	smp_mb()
+diff --git a/arch/tile/include/asm/cache.h b/arch/tile/include/asm/cache.h
+index 392e533..536b092 100644
+--- a/arch/tile/include/asm/cache.h
++++ b/arch/tile/include/asm/cache.h
+@@ -15,11 +15,12 @@
+ #ifndef _ASM_TILE_CACHE_H
+ #define _ASM_TILE_CACHE_H
+ 
++#include <linux/const.h>
+ #include <arch/chip.h>
+ 
+ /* bytes per L1 data cache line */
+ #define L1_CACHE_SHIFT		CHIP_L1D_LOG_LINE_SIZE()
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ /* bytes per L2 cache line */
+ #define L2_CACHE_SHIFT		CHIP_L2_LOG_LINE_SIZE()
+diff --git a/arch/um/Makefile b/arch/um/Makefile
+index 7730af6..cce5b19 100644
+--- a/arch/um/Makefile
++++ b/arch/um/Makefile
+@@ -61,6 +61,10 @@ USER_CFLAGS = $(patsubst $(KERNEL_DEFINES),,$(patsubst -D__KERNEL__,,\
+ 	$(patsubst -I%,,$(KBUILD_CFLAGS)))) $(ARCH_INCLUDE) $(MODE_INCLUDE) \
+ 	$(filter -I%,$(CFLAGS)) -D_FILE_OFFSET_BITS=64 -idirafter include
+ 
++ifdef CONSTIFY_PLUGIN
++USER_CFLAGS	+= $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify
++endif
++
+ #This will adjust *FLAGS accordingly to the platform.
+ include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS)
+ 
+diff --git a/arch/um/include/asm/cache.h b/arch/um/include/asm/cache.h
+index 19e1bdd..3665b77 100644
+--- a/arch/um/include/asm/cache.h
++++ b/arch/um/include/asm/cache.h
+@@ -1,6 +1,7 @@
+ #ifndef __UM_CACHE_H
+ #define __UM_CACHE_H
+ 
++#include <linux/const.h>
+ 
+ #if defined(CONFIG_UML_X86) && !defined(CONFIG_64BIT)
+ # define L1_CACHE_SHIFT		(CONFIG_X86_L1_CACHE_SHIFT)
+@@ -12,6 +13,6 @@
+ # define L1_CACHE_SHIFT		5
+ #endif
+ 
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #endif
+diff --git a/arch/um/include/asm/kmap_types.h b/arch/um/include/asm/kmap_types.h
+index 6c03acd..a5e0215 100644
+--- a/arch/um/include/asm/kmap_types.h
++++ b/arch/um/include/asm/kmap_types.h
+@@ -23,6 +23,7 @@ enum km_type {
+ 	KM_IRQ1,
+ 	KM_SOFTIRQ0,
+ 	KM_SOFTIRQ1,
++	KM_CLEARPAGE,
+ 	KM_TYPE_NR
+ };
+ 
+diff --git a/arch/um/include/asm/page.h b/arch/um/include/asm/page.h
+index 7cfc3ce..cbd1a58 100644
+--- a/arch/um/include/asm/page.h
++++ b/arch/um/include/asm/page.h
+@@ -14,6 +14,9 @@
+ #define PAGE_SIZE	(_AC(1, UL) << PAGE_SHIFT)
+ #define PAGE_MASK	(~(PAGE_SIZE-1))
+ 
++#define ktla_ktva(addr)			(addr)
++#define ktva_ktla(addr)			(addr)
++
+ #ifndef __ASSEMBLY__
+ 
+ struct page;
+diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c
+index c533835..84db18e 100644
+--- a/arch/um/kernel/process.c
++++ b/arch/um/kernel/process.c
+@@ -406,22 +406,6 @@ int singlestepping(void * t)
+ 	return 2;
+ }
+ 
+-/*
+- * Only x86 and x86_64 have an arch_align_stack().
+- * All other arches have "#define arch_align_stack(x) (x)"
+- * in their asm/system.h
+- * As this is included in UML from asm-um/system-generic.h,
+- * we can use it to behave as the subarch does.
+- */
+-#ifndef arch_align_stack
+-unsigned long arch_align_stack(unsigned long sp)
+-{
+-	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+-		sp -= get_random_int() % 8192;
+-	return sp & ~0xf;
+-}
+-#endif
+-
+ unsigned long get_wchan(struct task_struct *p)
+ {
+ 	unsigned long stack_page, sp, ip;
+diff --git a/arch/unicore32/include/asm/cache.h b/arch/unicore32/include/asm/cache.h
+index ad8f795..2c7eec6 100644
+--- a/arch/unicore32/include/asm/cache.h
++++ b/arch/unicore32/include/asm/cache.h
+@@ -12,8 +12,10 @@
+ #ifndef __UNICORE_CACHE_H__
+ #define __UNICORE_CACHE_H__
+ 
+-#define L1_CACHE_SHIFT		(5)
+-#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
++#include <linux/const.h>
++
++#define L1_CACHE_SHIFT		5
++#define L1_CACHE_BYTES		(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ /*
+  * Memory returned by kmalloc() may be used for DMA, so we must make
+diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
+index efb4294..61bc18c 100644
+--- a/arch/x86/Kconfig
++++ b/arch/x86/Kconfig
+@@ -235,7 +235,7 @@ config X86_HT
+ 
+ config X86_32_LAZY_GS
+ 	def_bool y
+-	depends on X86_32 && !CC_STACKPROTECTOR
++	depends on X86_32 && !CC_STACKPROTECTOR && !PAX_MEMORY_UDEREF
+ 
+ config ARCH_HWEIGHT_CFLAGS
+ 	string
+@@ -1022,7 +1022,7 @@ choice
+ 
+ config NOHIGHMEM
+ 	bool "off"
+-	depends on !X86_NUMAQ
++	depends on !X86_NUMAQ && !(PAX_PAGEEXEC && PAX_ENABLE_PAE)
+ 	---help---
+ 	  Linux can use up to 64 Gigabytes of physical memory on x86 systems.
+ 	  However, the address space of 32-bit x86 processors is only 4
+@@ -1059,7 +1059,7 @@ config NOHIGHMEM
+ 
+ config HIGHMEM4G
+ 	bool "4GB"
+-	depends on !X86_NUMAQ
++	depends on !X86_NUMAQ && !(PAX_PAGEEXEC && PAX_ENABLE_PAE)
+ 	---help---
+ 	  Select this if you have a 32-bit processor and between 1 and 4
+ 	  gigabytes of physical RAM.
+@@ -1113,7 +1113,7 @@ config PAGE_OFFSET
+ 	hex
+ 	default 0xB0000000 if VMSPLIT_3G_OPT
+ 	default 0x80000000 if VMSPLIT_2G
+-	default 0x78000000 if VMSPLIT_2G_OPT
++	default 0x70000000 if VMSPLIT_2G_OPT
+ 	default 0x40000000 if VMSPLIT_1G
+ 	default 0xC0000000
+ 	depends on X86_32
+@@ -1496,6 +1496,7 @@ config SECCOMP
+ 
+ config CC_STACKPROTECTOR
+ 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
++	depends on X86_64 || !PAX_MEMORY_UDEREF
+ 	---help---
+ 	  This option turns on the -fstack-protector GCC feature. This
+ 	  feature puts, at the beginning of functions, a canary value on
+@@ -1553,6 +1554,7 @@ config KEXEC_JUMP
+ config PHYSICAL_START
+ 	hex "Physical address where the kernel is loaded" if (EXPERT || CRASH_DUMP)
+ 	default "0x1000000"
++	range 0x400000 0x40000000
+ 	---help---
+ 	  This gives the physical address where the kernel is loaded.
+ 
+@@ -1616,6 +1618,7 @@ config X86_NEED_RELOCS
+ config PHYSICAL_ALIGN
+ 	hex "Alignment value to which kernel should be aligned" if X86_32
+ 	default "0x1000000"
++	range 0x400000 0x1000000 if PAX_KERNEXEC
+ 	range 0x2000 0x1000000
+ 	---help---
+ 	  This value puts the alignment restrictions on physical address
+@@ -1647,9 +1650,10 @@ config HOTPLUG_CPU
+ 	  Say N if you want to disable CPU hotplug.
+ 
+ config COMPAT_VDSO
+-	def_bool y
++	def_bool n
+ 	prompt "Compat VDSO support"
+ 	depends on X86_32 || IA32_EMULATION
++	depends on !PAX_NOEXEC && !PAX_MEMORY_UDEREF
+ 	---help---
+ 	  Map the 32-bit VDSO to the predictable old-style address too.
+ 
+diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu
+index e3ca7e0..b30b28a 100644
+--- a/arch/x86/Kconfig.cpu
++++ b/arch/x86/Kconfig.cpu
+@@ -341,7 +341,7 @@ config X86_PPRO_FENCE
+ 
+ config X86_F00F_BUG
+ 	def_bool y
+-	depends on M586MMX || M586TSC || M586 || M486 || M386
++	depends on (M586MMX || M586TSC || M586 || M486 || M386) && !PAX_KERNEXEC
+ 
+ config X86_INVD_BUG
+ 	def_bool y
+@@ -365,7 +365,7 @@ config X86_POPAD_OK
+ 
+ config X86_ALIGNMENT_16
+ 	def_bool y
+-	depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1
++	depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK8 || MK7 || MK6 || MCORE2 || MPENTIUM4 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1
+ 
+ config X86_INTEL_USERCOPY
+ 	def_bool y
+@@ -411,7 +411,7 @@ config X86_CMPXCHG64
+ # generates cmov.
+ config X86_CMOV
+ 	def_bool y
+-	depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM || MGEODE_LX)
++	depends on (MK8 || MK7 || MCORE2 || MPSC || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM || MGEODE_LX)
+ 
+ config X86_MINIMUM_CPU_FAMILY
+ 	int
+diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
+index bf56e17..05f9891 100644
+--- a/arch/x86/Kconfig.debug
++++ b/arch/x86/Kconfig.debug
+@@ -81,7 +81,7 @@ config X86_PTDUMP
+ config DEBUG_RODATA
+ 	bool "Write protect kernel read-only data structures"
+ 	default y
+-	depends on DEBUG_KERNEL
++	depends on DEBUG_KERNEL && BROKEN
+ 	---help---
+ 	  Mark the kernel read-only data as write-protected in the pagetables,
+ 	  in order to catch accidental (and incorrect) writes to such const
+@@ -99,7 +99,7 @@ config DEBUG_RODATA_TEST
+ 
+ config DEBUG_SET_MODULE_RONX
+ 	bool "Set loadable kernel module data as NX and text as RO"
+-	depends on MODULES
++	depends on MODULES && BROKEN
+ 	---help---
+ 	  This option helps catch unintended modifications to loadable
+ 	  kernel module's text and read-only data. It also prevents execution
+diff --git a/arch/x86/Makefile b/arch/x86/Makefile
+index b02e509..2631e48 100644
+--- a/arch/x86/Makefile
++++ b/arch/x86/Makefile
+@@ -46,6 +46,7 @@ else
+         UTS_MACHINE := x86_64
+         CHECKFLAGS += -D__x86_64__ -m64
+ 
++        biarch := $(call cc-option,-m64)
+         KBUILD_AFLAGS += -m64
+         KBUILD_CFLAGS += -m64
+ 
+@@ -195,3 +196,12 @@ define archhelp
+   echo  '                  FDARGS="..."  arguments for the booted kernel'
+   echo  '                  FDINITRD=file initrd for the booted kernel'
+ endef
++
++define OLD_LD
++
++*** ${VERSION}.${PATCHLEVEL} PaX kernels no longer build correctly with old versions of binutils.
++*** Please upgrade your binutils to 2.18 or newer
++endef
++
++archprepare:
++	$(if $(LDFLAGS_BUILD_ID),,$(error $(OLD_LD)))
+diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile
+index 95365a8..52f857b 100644
+--- a/arch/x86/boot/Makefile
++++ b/arch/x86/boot/Makefile
+@@ -63,6 +63,9 @@ KBUILD_CFLAGS	:= $(LINUXINCLUDE) -g -Os -D_SETUP -D__KERNEL__ \
+ 		   $(call cc-option, -fno-stack-protector) \
+ 		   $(call cc-option, -mpreferred-stack-boundary=2)
+ KBUILD_CFLAGS	+= $(call cc-option, -m32)
++ifdef CONSTIFY_PLUGIN
++KBUILD_CFLAGS	+= $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify
++endif
+ KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
+ GCOV_PROFILE := n
+ 
+diff --git a/arch/x86/boot/bitops.h b/arch/x86/boot/bitops.h
+index 878e4b9..20537ab 100644
+--- a/arch/x86/boot/bitops.h
++++ b/arch/x86/boot/bitops.h
+@@ -26,7 +26,7 @@ static inline int variable_test_bit(int nr, const void *addr)
+ 	u8 v;
+ 	const u32 *p = (const u32 *)addr;
+ 
+-	asm("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr));
++	asm volatile("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr));
+ 	return v;
+ }
+ 
+@@ -37,7 +37,7 @@ static inline int variable_test_bit(int nr, const void *addr)
+ 
+ static inline void set_bit(int nr, void *addr)
+ {
+-	asm("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr));
++	asm volatile("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr));
+ }
+ 
+ #endif /* BOOT_BITOPS_H */
+diff --git a/arch/x86/boot/boot.h b/arch/x86/boot/boot.h
+index c7093bd..d4247ffe0 100644
+--- a/arch/x86/boot/boot.h
++++ b/arch/x86/boot/boot.h
+@@ -85,7 +85,7 @@ static inline void io_delay(void)
+ static inline u16 ds(void)
+ {
+ 	u16 seg;
+-	asm("movw %%ds,%0" : "=rm" (seg));
++	asm volatile("movw %%ds,%0" : "=rm" (seg));
+ 	return seg;
+ }
+ 
+@@ -181,7 +181,7 @@ static inline void wrgs32(u32 v, addr_t addr)
+ static inline int memcmp(const void *s1, const void *s2, size_t len)
+ {
+ 	u8 diff;
+-	asm("repe; cmpsb; setnz %0"
++	asm volatile("repe; cmpsb; setnz %0"
+ 	    : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len));
+ 	return diff;
+ }
+diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
+index 09664ef..edc5d03 100644
+--- a/arch/x86/boot/compressed/Makefile
++++ b/arch/x86/boot/compressed/Makefile
+@@ -14,6 +14,9 @@ cflags-$(CONFIG_X86_64) := -mcmodel=small
+ KBUILD_CFLAGS += $(cflags-y)
+ KBUILD_CFLAGS += $(call cc-option,-ffreestanding)
+ KBUILD_CFLAGS += $(call cc-option,-fno-stack-protector)
++ifdef CONSTIFY_PLUGIN
++KBUILD_CFLAGS += $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify
++endif
+ 
+ KBUILD_AFLAGS  := $(KBUILD_CFLAGS) -D__ASSEMBLY__
+ GCOV_PROFILE := n
+diff --git a/arch/x86/boot/compressed/head_32.S b/arch/x86/boot/compressed/head_32.S
+index 67a655a..b924059 100644
+--- a/arch/x86/boot/compressed/head_32.S
++++ b/arch/x86/boot/compressed/head_32.S
+@@ -76,7 +76,7 @@ ENTRY(startup_32)
+ 	notl	%eax
+ 	andl    %eax, %ebx
+ #else
+-	movl	$LOAD_PHYSICAL_ADDR, %ebx
++	movl	$____LOAD_PHYSICAL_ADDR, %ebx
+ #endif
+ 
+ 	/* Target address to relocate to for decompression */
+@@ -162,7 +162,7 @@ relocated:
+  * and where it was actually loaded.
+  */
+ 	movl	%ebp, %ebx
+-	subl	$LOAD_PHYSICAL_ADDR, %ebx
++	subl	$____LOAD_PHYSICAL_ADDR, %ebx
+ 	jz	2f	/* Nothing to be done if loaded at compiled addr. */
+ /*
+  * Process relocations.
+@@ -170,8 +170,7 @@ relocated:
+ 
+ 1:	subl	$4, %edi
+ 	movl	(%edi), %ecx
+-	testl	%ecx, %ecx
+-	jz	2f
++	jecxz	2f
+ 	addl	%ebx, -__PAGE_OFFSET(%ebx, %ecx)
+ 	jmp	1b
+ 2:
+diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
+index 35af09d..99c9676 100644
+--- a/arch/x86/boot/compressed/head_64.S
++++ b/arch/x86/boot/compressed/head_64.S
+@@ -91,7 +91,7 @@ ENTRY(startup_32)
+ 	notl	%eax
+ 	andl	%eax, %ebx
+ #else
+-	movl	$LOAD_PHYSICAL_ADDR, %ebx
++	movl	$____LOAD_PHYSICAL_ADDR, %ebx
+ #endif
+ 
+ 	/* Target address to relocate to for decompression */
+@@ -233,7 +233,7 @@ ENTRY(startup_64)
+ 	notq	%rax
+ 	andq	%rax, %rbp
+ #else
+-	movq	$LOAD_PHYSICAL_ADDR, %rbp
++	movq	$____LOAD_PHYSICAL_ADDR, %rbp
+ #endif
+ 
+ 	/* Target address to relocate to for decompression */
+diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
+index 3a19d04..7c1d55a 100644
+--- a/arch/x86/boot/compressed/misc.c
++++ b/arch/x86/boot/compressed/misc.c
+@@ -310,7 +310,7 @@ static void parse_elf(void *output)
+ 		case PT_LOAD:
+ #ifdef CONFIG_RELOCATABLE
+ 			dest = output;
+-			dest += (phdr->p_paddr - LOAD_PHYSICAL_ADDR);
++			dest += (phdr->p_paddr - ____LOAD_PHYSICAL_ADDR);
+ #else
+ 			dest = (void *)(phdr->p_paddr);
+ #endif
+@@ -363,7 +363,7 @@ asmlinkage void decompress_kernel(void *rmode, memptr heap,
+ 		error("Destination address too large");
+ #endif
+ #ifndef CONFIG_RELOCATABLE
+-	if ((unsigned long)output != LOAD_PHYSICAL_ADDR)
++	if ((unsigned long)output != ____LOAD_PHYSICAL_ADDR)
+ 		error("Wrong destination address");
+ #endif
+ 
+diff --git a/arch/x86/boot/compressed/relocs.c b/arch/x86/boot/compressed/relocs.c
+index 89bbf4e..869908e 100644
+--- a/arch/x86/boot/compressed/relocs.c
++++ b/arch/x86/boot/compressed/relocs.c
+@@ -13,8 +13,11 @@
+ 
+ static void die(char *fmt, ...);
+ 
++#include "../../../../include/generated/autoconf.h"
++
+ #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+ static Elf32_Ehdr ehdr;
++static Elf32_Phdr *phdr;
+ static unsigned long reloc_count, reloc_idx;
+ static unsigned long *relocs;
+ 
+@@ -270,9 +273,39 @@ static void read_ehdr(FILE *fp)
+ 	}
+ }
+ 
++static void read_phdrs(FILE *fp)
++{
++	unsigned int i;
++
++	phdr = calloc(ehdr.e_phnum, sizeof(Elf32_Phdr));
++	if (!phdr) {
++		die("Unable to allocate %d program headers\n",
++		    ehdr.e_phnum);
++	}
++	if (fseek(fp, ehdr.e_phoff, SEEK_SET) < 0) {
++		die("Seek to %d failed: %s\n",
++			ehdr.e_phoff, strerror(errno));
++	}
++	if (fread(phdr, sizeof(*phdr), ehdr.e_phnum, fp) != ehdr.e_phnum) {
++		die("Cannot read ELF program headers: %s\n",
++			strerror(errno));
++	}
++	for(i = 0; i < ehdr.e_phnum; i++) {
++		phdr[i].p_type      = elf32_to_cpu(phdr[i].p_type);
++		phdr[i].p_offset    = elf32_to_cpu(phdr[i].p_offset);
++		phdr[i].p_vaddr     = elf32_to_cpu(phdr[i].p_vaddr);
++		phdr[i].p_paddr     = elf32_to_cpu(phdr[i].p_paddr);
++		phdr[i].p_filesz    = elf32_to_cpu(phdr[i].p_filesz);
++		phdr[i].p_memsz     = elf32_to_cpu(phdr[i].p_memsz);
++		phdr[i].p_flags     = elf32_to_cpu(phdr[i].p_flags);
++		phdr[i].p_align     = elf32_to_cpu(phdr[i].p_align);
++	}
++
++}
++
+ static void read_shdrs(FILE *fp)
+ {
+-	int i;
++	unsigned int i;
+ 	Elf32_Shdr shdr;
+ 
+ 	secs = calloc(ehdr.e_shnum, sizeof(struct section));
+@@ -307,7 +340,7 @@ static void read_shdrs(FILE *fp)
+ 
+ static void read_strtabs(FILE *fp)
+ {
+-	int i;
++	unsigned int i;
+ 	for (i = 0; i < ehdr.e_shnum; i++) {
+ 		struct section *sec = &secs[i];
+ 		if (sec->shdr.sh_type != SHT_STRTAB) {
+@@ -332,7 +365,7 @@ static void read_strtabs(FILE *fp)
+ 
+ static void read_symtabs(FILE *fp)
+ {
+-	int i,j;
++	unsigned int i,j;
+ 	for (i = 0; i < ehdr.e_shnum; i++) {
+ 		struct section *sec = &secs[i];
+ 		if (sec->shdr.sh_type != SHT_SYMTAB) {
+@@ -365,7 +398,9 @@ static void read_symtabs(FILE *fp)
+ 
+ static void read_relocs(FILE *fp)
+ {
+-	int i,j;
++	unsigned int i,j;
++	uint32_t base;
++
+ 	for (i = 0; i < ehdr.e_shnum; i++) {
+ 		struct section *sec = &secs[i];
+ 		if (sec->shdr.sh_type != SHT_REL) {
+@@ -385,9 +420,18 @@ static void read_relocs(FILE *fp)
+ 			die("Cannot read symbol table: %s\n",
+ 				strerror(errno));
+ 		}
++		base = 0;
++		for (j = 0; j < ehdr.e_phnum; j++) {
++			if (phdr[j].p_type != PT_LOAD )
++				continue;
++			if (secs[sec->shdr.sh_info].shdr.sh_offset < phdr[j].p_offset || secs[sec->shdr.sh_info].shdr.sh_offset >= phdr[j].p_offset + phdr[j].p_filesz)
++				continue;
++			base = CONFIG_PAGE_OFFSET + phdr[j].p_paddr - phdr[j].p_vaddr;
++			break;
++		}
+ 		for (j = 0; j < sec->shdr.sh_size/sizeof(Elf32_Rel); j++) {
+ 			Elf32_Rel *rel = &sec->reltab[j];
+-			rel->r_offset = elf32_to_cpu(rel->r_offset);
++			rel->r_offset = elf32_to_cpu(rel->r_offset) + base;
+ 			rel->r_info   = elf32_to_cpu(rel->r_info);
+ 		}
+ 	}
+@@ -396,14 +440,14 @@ static void read_relocs(FILE *fp)
+ 
+ static void print_absolute_symbols(void)
+ {
+-	int i;
++	unsigned int i;
+ 	printf("Absolute symbols\n");
+ 	printf(" Num:    Value Size  Type       Bind        Visibility  Name\n");
+ 	for (i = 0; i < ehdr.e_shnum; i++) {
+ 		struct section *sec = &secs[i];
+ 		char *sym_strtab;
+ 		Elf32_Sym *sh_symtab;
+-		int j;
++		unsigned int j;
+ 
+ 		if (sec->shdr.sh_type != SHT_SYMTAB) {
+ 			continue;
+@@ -431,14 +475,14 @@ static void print_absolute_symbols(void)
+ 
+ static void print_absolute_relocs(void)
+ {
+-	int i, printed = 0;
++	unsigned int i, printed = 0;
+ 
+ 	for (i = 0; i < ehdr.e_shnum; i++) {
+ 		struct section *sec = &secs[i];
+ 		struct section *sec_applies, *sec_symtab;
+ 		char *sym_strtab;
+ 		Elf32_Sym *sh_symtab;
+-		int j;
++		unsigned int j;
+ 		if (sec->shdr.sh_type != SHT_REL) {
+ 			continue;
+ 		}
+@@ -499,13 +543,13 @@ static void print_absolute_relocs(void)
+ 
+ static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym))
+ {
+-	int i;
++	unsigned int i;
+ 	/* Walk through the relocations */
+ 	for (i = 0; i < ehdr.e_shnum; i++) {
+ 		char *sym_strtab;
+ 		Elf32_Sym *sh_symtab;
+ 		struct section *sec_applies, *sec_symtab;
+-		int j;
++		unsigned int j;
+ 		struct section *sec = &secs[i];
+ 
+ 		if (sec->shdr.sh_type != SHT_REL) {
+@@ -530,6 +574,22 @@ static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym))
+ 			    !is_rel_reloc(sym_name(sym_strtab, sym))) {
+ 				continue;
+ 			}
++			/* Don't relocate actual per-cpu variables, they are absolute indices, not addresses */
++			if (!strcmp(sec_name(sym->st_shndx), ".data..percpu") && strcmp(sym_name(sym_strtab, sym), "__per_cpu_load"))
++				continue;
++
++#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_X86_32)
++			/* Don't relocate actual code, they are relocated implicitly by the base address of KERNEL_CS */
++			if (!strcmp(sec_name(sym->st_shndx), ".module.text") && !strcmp(sym_name(sym_strtab, sym), "_etext"))
++				continue;
++			if (!strcmp(sec_name(sym->st_shndx), ".init.text"))
++				continue;
++			if (!strcmp(sec_name(sym->st_shndx), ".exit.text"))
++				continue;
++			if (!strcmp(sec_name(sym->st_shndx), ".text") && strcmp(sym_name(sym_strtab, sym), "__LOAD_PHYSICAL_ADDR"))
++				continue;
++#endif
++
+ 			switch (r_type) {
+ 			case R_386_NONE:
+ 			case R_386_PC32:
+@@ -571,7 +631,7 @@ static int cmp_relocs(const void *va, const void *vb)
+ 
+ static void emit_relocs(int as_text)
+ {
+-	int i;
++	unsigned int i;
+ 	/* Count how many relocations I have and allocate space for them. */
+ 	reloc_count = 0;
+ 	walk_relocs(count_reloc);
+@@ -665,6 +725,7 @@ int main(int argc, char **argv)
+ 			fname, strerror(errno));
+ 	}
+ 	read_ehdr(fp);
++	read_phdrs(fp);
+ 	read_shdrs(fp);
+ 	read_strtabs(fp);
+ 	read_symtabs(fp);
+diff --git a/arch/x86/boot/cpucheck.c b/arch/x86/boot/cpucheck.c
+index 4d3ff03..e4972ff 100644
+--- a/arch/x86/boot/cpucheck.c
++++ b/arch/x86/boot/cpucheck.c
+@@ -74,7 +74,7 @@ static int has_fpu(void)
+ 	u16 fcw = -1, fsw = -1;
+ 	u32 cr0;
+ 
+-	asm("movl %%cr0,%0" : "=r" (cr0));
++	asm volatile("movl %%cr0,%0" : "=r" (cr0));
+ 	if (cr0 & (X86_CR0_EM|X86_CR0_TS)) {
+ 		cr0 &= ~(X86_CR0_EM|X86_CR0_TS);
+ 		asm volatile("movl %0,%%cr0" : : "r" (cr0));
+@@ -90,7 +90,7 @@ static int has_eflag(u32 mask)
+ {
+ 	u32 f0, f1;
+ 
+-	asm("pushfl ; "
++	asm volatile("pushfl ; "
+ 	    "pushfl ; "
+ 	    "popl %0 ; "
+ 	    "movl %0,%1 ; "
+@@ -115,7 +115,7 @@ static void get_flags(void)
+ 		set_bit(X86_FEATURE_FPU, cpu.flags);
+ 
+ 	if (has_eflag(X86_EFLAGS_ID)) {
+-		asm("cpuid"
++		asm volatile("cpuid"
+ 		    : "=a" (max_intel_level),
+ 		      "=b" (cpu_vendor[0]),
+ 		      "=d" (cpu_vendor[1]),
+@@ -124,7 +124,7 @@ static void get_flags(void)
+ 
+ 		if (max_intel_level >= 0x00000001 &&
+ 		    max_intel_level <= 0x0000ffff) {
+-			asm("cpuid"
++			asm volatile("cpuid"
+ 			    : "=a" (tfms),
+ 			      "=c" (cpu.flags[4]),
+ 			      "=d" (cpu.flags[0])
+@@ -136,7 +136,7 @@ static void get_flags(void)
+ 				cpu.model += ((tfms >> 16) & 0xf) << 4;
+ 		}
+ 
+-		asm("cpuid"
++		asm volatile("cpuid"
+ 		    : "=a" (max_amd_level)
+ 		    : "a" (0x80000000)
+ 		    : "ebx", "ecx", "edx");
+@@ -144,7 +144,7 @@ static void get_flags(void)
+ 		if (max_amd_level >= 0x80000001 &&
+ 		    max_amd_level <= 0x8000ffff) {
+ 			u32 eax = 0x80000001;
+-			asm("cpuid"
++			asm volatile("cpuid"
+ 			    : "+a" (eax),
+ 			      "=c" (cpu.flags[6]),
+ 			      "=d" (cpu.flags[1])
+@@ -203,9 +203,9 @@ int check_cpu(int *cpu_level_ptr, int *req_level_ptr, u32 **err_flags_ptr)
+ 		u32 ecx = MSR_K7_HWCR;
+ 		u32 eax, edx;
+ 
+-		asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
++		asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
+ 		eax &= ~(1 << 15);
+-		asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
++		asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
+ 
+ 		get_flags();	/* Make sure it really did something */
+ 		err = check_flags();
+@@ -218,9 +218,9 @@ int check_cpu(int *cpu_level_ptr, int *req_level_ptr, u32 **err_flags_ptr)
+ 		u32 ecx = MSR_VIA_FCR;
+ 		u32 eax, edx;
+ 
+-		asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
++		asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
+ 		eax |= (1<<1)|(1<<7);
+-		asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
++		asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
+ 
+ 		set_bit(X86_FEATURE_CX8, cpu.flags);
+ 		err = check_flags();
+@@ -231,12 +231,12 @@ int check_cpu(int *cpu_level_ptr, int *req_level_ptr, u32 **err_flags_ptr)
+ 		u32 eax, edx;
+ 		u32 level = 1;
+ 
+-		asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
+-		asm("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx));
+-		asm("cpuid"
++		asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
++		asm volatile("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx));
++		asm volatile("cpuid"
+ 		    : "+a" (level), "=d" (cpu.flags[0])
+ 		    : : "ecx", "ebx");
+-		asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
++		asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
+ 
+ 		err = check_flags();
+ 	}
+diff --git a/arch/x86/boot/header.S b/arch/x86/boot/header.S
+index bdb4d45..0476680 100644
+--- a/arch/x86/boot/header.S
++++ b/arch/x86/boot/header.S
+@@ -224,7 +224,7 @@ setup_data:		.quad 0			# 64-bit physical pointer to
+ 						# single linked list of
+ 						# struct setup_data
+ 
+-pref_address:		.quad LOAD_PHYSICAL_ADDR	# preferred load addr
++pref_address:		.quad ____LOAD_PHYSICAL_ADDR	# preferred load addr
+ 
+ #define ZO_INIT_SIZE	(ZO__end - ZO_startup_32 + ZO_z_extract_offset)
+ #define VO_INIT_SIZE	(VO__end - VO__text)
+diff --git a/arch/x86/boot/memory.c b/arch/x86/boot/memory.c
+index db75d07..8e6d0af 100644
+--- a/arch/x86/boot/memory.c
++++ b/arch/x86/boot/memory.c
+@@ -19,7 +19,7 @@
+ 
+ static int detect_memory_e820(void)
+ {
+-	int count = 0;
++	unsigned int count = 0;
+ 	struct biosregs ireg, oreg;
+ 	struct e820entry *desc = boot_params.e820_map;
+ 	static struct e820entry buf; /* static so it is zeroed */
+diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c
+index 11e8c6e..fdbb1ed 100644
+--- a/arch/x86/boot/video-vesa.c
++++ b/arch/x86/boot/video-vesa.c
+@@ -200,6 +200,7 @@ static void vesa_store_pm_info(void)
+ 
+ 	boot_params.screen_info.vesapm_seg = oreg.es;
+ 	boot_params.screen_info.vesapm_off = oreg.di;
++	boot_params.screen_info.vesapm_size = oreg.cx;
+ }
+ 
+ /*
+diff --git a/arch/x86/boot/video.c b/arch/x86/boot/video.c
+index 43eda28..5ab5fdb 100644
+--- a/arch/x86/boot/video.c
++++ b/arch/x86/boot/video.c
+@@ -96,7 +96,7 @@ static void store_mode_params(void)
+ static unsigned int get_entry(void)
+ {
+ 	char entry_buf[4];
+-	int i, len = 0;
++	unsigned int i, len = 0;
+ 	int key;
+ 	unsigned int v;
+ 
+diff --git a/arch/x86/crypto/aes-x86_64-asm_64.S b/arch/x86/crypto/aes-x86_64-asm_64.S
+index 5b577d5..3c1fed4 100644
+--- a/arch/x86/crypto/aes-x86_64-asm_64.S
++++ b/arch/x86/crypto/aes-x86_64-asm_64.S
+@@ -8,6 +8,8 @@
+  * including this sentence is retained in full.
+  */
+ 
++#include <asm/alternative-asm.h>
++
+ .extern crypto_ft_tab
+ .extern crypto_it_tab
+ .extern crypto_fl_tab
+@@ -71,6 +73,8 @@ FUNC:	movq	r1,r2;			\
+ 	je	B192;			\
+ 	leaq	32(r9),r9;
+ 
++#define ret	pax_force_retaddr 0, 1; ret
++
+ #define epilogue(r1,r2,r3,r4,r5,r6,r7,r8,r9) \
+ 	movq	r1,r2;			\
+ 	movq	r3,r4;			\
+diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S
+index be6d9e3..21fbbca 100644
+--- a/arch/x86/crypto/aesni-intel_asm.S
++++ b/arch/x86/crypto/aesni-intel_asm.S
+@@ -31,6 +31,7 @@
+ 
+ #include <linux/linkage.h>
+ #include <asm/inst.h>
++#include <asm/alternative-asm.h>
+ 
+ #ifdef __x86_64__
+ .data
+@@ -1436,7 +1437,9 @@ _return_T_done_decrypt:
+ 	pop	%r14
+ 	pop	%r13
+ 	pop	%r12
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_gcm_dec)
+ 
+ 
+ /*****************************************************************************
+@@ -1699,7 +1702,9 @@ _return_T_done_encrypt:
+ 	pop	%r14
+ 	pop	%r13
+ 	pop	%r12
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_gcm_enc)
+ 
+ #endif
+ 
+@@ -1714,6 +1719,7 @@ _key_expansion_256a:
+ 	pxor %xmm1, %xmm0
+ 	movaps %xmm0, (TKEYP)
+ 	add $0x10, TKEYP
++	pax_force_retaddr_bts
+ 	ret
+ 
+ .align 4
+@@ -1738,6 +1744,7 @@ _key_expansion_192a:
+ 	shufps $0b01001110, %xmm2, %xmm1
+ 	movaps %xmm1, 0x10(TKEYP)
+ 	add $0x20, TKEYP
++	pax_force_retaddr_bts
+ 	ret
+ 
+ .align 4
+@@ -1757,6 +1764,7 @@ _key_expansion_192b:
+ 
+ 	movaps %xmm0, (TKEYP)
+ 	add $0x10, TKEYP
++	pax_force_retaddr_bts
+ 	ret
+ 
+ .align 4
+@@ -1769,6 +1777,7 @@ _key_expansion_256b:
+ 	pxor %xmm1, %xmm2
+ 	movaps %xmm2, (TKEYP)
+ 	add $0x10, TKEYP
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -1881,7 +1890,9 @@ ENTRY(aesni_set_key)
+ #ifndef __x86_64__
+ 	popl KEYP
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_set_key)
+ 
+ /*
+  * void aesni_enc(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src)
+@@ -1902,7 +1913,9 @@ ENTRY(aesni_enc)
+ 	popl KLEN
+ 	popl KEYP
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_enc)
+ 
+ /*
+  * _aesni_enc1:		internal ABI
+@@ -1959,6 +1972,7 @@ _aesni_enc1:
+ 	AESENC KEY STATE
+ 	movaps 0x70(TKEYP), KEY
+ 	AESENCLAST KEY STATE
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -2067,6 +2081,7 @@ _aesni_enc4:
+ 	AESENCLAST KEY STATE2
+ 	AESENCLAST KEY STATE3
+ 	AESENCLAST KEY STATE4
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -2089,7 +2104,9 @@ ENTRY(aesni_dec)
+ 	popl KLEN
+ 	popl KEYP
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_dec)
+ 
+ /*
+  * _aesni_dec1:		internal ABI
+@@ -2146,6 +2163,7 @@ _aesni_dec1:
+ 	AESDEC KEY STATE
+ 	movaps 0x70(TKEYP), KEY
+ 	AESDECLAST KEY STATE
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -2254,6 +2272,7 @@ _aesni_dec4:
+ 	AESDECLAST KEY STATE2
+ 	AESDECLAST KEY STATE3
+ 	AESDECLAST KEY STATE4
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -2311,7 +2330,9 @@ ENTRY(aesni_ecb_enc)
+ 	popl KEYP
+ 	popl LEN
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_ecb_enc)
+ 
+ /*
+  * void aesni_ecb_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+@@ -2369,7 +2390,9 @@ ENTRY(aesni_ecb_dec)
+ 	popl KEYP
+ 	popl LEN
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_ecb_dec)
+ 
+ /*
+  * void aesni_cbc_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+@@ -2410,7 +2433,9 @@ ENTRY(aesni_cbc_enc)
+ 	popl LEN
+ 	popl IVP
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_cbc_enc)
+ 
+ /*
+  * void aesni_cbc_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+@@ -2498,7 +2523,9 @@ ENTRY(aesni_cbc_dec)
+ 	popl LEN
+ 	popl IVP
+ #endif
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_cbc_dec)
+ 
+ #ifdef __x86_64__
+ .align 16
+@@ -2524,6 +2551,7 @@ _aesni_inc_init:
+ 	mov $1, TCTR_LOW
+ 	MOVQ_R64_XMM TCTR_LOW INC
+ 	MOVQ_R64_XMM CTR TCTR_LOW
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -2552,6 +2580,7 @@ _aesni_inc:
+ .Linc_low:
+ 	movaps CTR, IV
+ 	PSHUFB_XMM BSWAP_MASK IV
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -2612,5 +2641,7 @@ ENTRY(aesni_ctr_enc)
+ .Lctr_enc_ret:
+ 	movups IV, (IVP)
+ .Lctr_enc_just_ret:
++	pax_force_retaddr 0, 1
+ 	ret
++ENDPROC(aesni_ctr_enc)
+ #endif
+diff --git a/arch/x86/crypto/blowfish-x86_64-asm_64.S b/arch/x86/crypto/blowfish-x86_64-asm_64.S
+index 391d245..67f35c2 100644
+--- a/arch/x86/crypto/blowfish-x86_64-asm_64.S
++++ b/arch/x86/crypto/blowfish-x86_64-asm_64.S
+@@ -20,6 +20,8 @@
+  *
+  */
+ 
++#include <asm/alternative-asm.h>
++
+ .file "blowfish-x86_64-asm.S"
+ .text
+ 
+@@ -151,9 +153,11 @@ __blowfish_enc_blk:
+ 	jnz __enc_xor;
+ 
+ 	write_block();
++	pax_force_retaddr 0, 1
+ 	ret;
+ __enc_xor:
+ 	xor_block();
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+ .align 8
+@@ -188,6 +192,7 @@ blowfish_dec_blk:
+ 
+ 	movq %r11, %rbp;
+ 
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+ /**********************************************************************
+@@ -342,6 +347,7 @@ __blowfish_enc_blk_4way:
+ 
+ 	popq %rbx;
+ 	popq %rbp;
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+ __enc_xor4:
+@@ -349,6 +355,7 @@ __enc_xor4:
+ 
+ 	popq %rbx;
+ 	popq %rbp;
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+ .align 8
+@@ -386,5 +393,6 @@ blowfish_dec_blk_4way:
+ 	popq %rbx;
+ 	popq %rbp;
+ 
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+diff --git a/arch/x86/crypto/salsa20-x86_64-asm_64.S b/arch/x86/crypto/salsa20-x86_64-asm_64.S
+index 6214a9b..1f4fc9a 100644
+--- a/arch/x86/crypto/salsa20-x86_64-asm_64.S
++++ b/arch/x86/crypto/salsa20-x86_64-asm_64.S
+@@ -1,3 +1,5 @@
++#include <asm/alternative-asm.h>
++
+ # enter ECRYPT_encrypt_bytes
+ .text
+ .p2align 5
+@@ -790,6 +792,7 @@ ECRYPT_encrypt_bytes:
+ 	add	%r11,%rsp
+ 	mov	%rdi,%rax
+ 	mov	%rsi,%rdx
++	pax_force_retaddr 0, 1
+ 	ret
+ #   bytesatleast65:
+ ._bytesatleast65:
+@@ -891,6 +894,7 @@ ECRYPT_keysetup:
+ 	add	%r11,%rsp
+ 	mov	%rdi,%rax
+ 	mov	%rsi,%rdx
++	pax_force_retaddr
+ 	ret
+ # enter ECRYPT_ivsetup
+ .text
+@@ -917,4 +921,5 @@ ECRYPT_ivsetup:
+ 	add	%r11,%rsp
+ 	mov	%rdi,%rax
+ 	mov	%rsi,%rdx
++	pax_force_retaddr
+ 	ret
+diff --git a/arch/x86/crypto/sha1_ssse3_asm.S b/arch/x86/crypto/sha1_ssse3_asm.S
+index b2c2f57..8470cab 100644
+--- a/arch/x86/crypto/sha1_ssse3_asm.S
++++ b/arch/x86/crypto/sha1_ssse3_asm.S
+@@ -28,6 +28,8 @@
+  * (at your option) any later version.
+  */
+ 
++#include <asm/alternative-asm.h>
++
+ #define CTX	%rdi	// arg1
+ #define BUF	%rsi	// arg2
+ #define CNT	%rdx	// arg3
+@@ -104,6 +106,7 @@
+ 	pop	%r12
+ 	pop	%rbp
+ 	pop	%rbx
++	pax_force_retaddr 0, 1
+ 	ret
+ 
+ 	.size	\name, .-\name
+diff --git a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
+index 5b012a2..36d5364 100644
+--- a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
++++ b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
+@@ -20,6 +20,8 @@
+  *
+  */
+ 
++#include <asm/alternative-asm.h>
++
+ .file "twofish-x86_64-asm-3way.S"
+ .text
+ 
+@@ -260,6 +262,7 @@ __twofish_enc_blk_3way:
+ 	popq %r13;
+ 	popq %r14;
+ 	popq %r15;
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+ __enc_xor3:
+@@ -271,6 +274,7 @@ __enc_xor3:
+ 	popq %r13;
+ 	popq %r14;
+ 	popq %r15;
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+ .global twofish_dec_blk_3way
+@@ -312,5 +316,6 @@ twofish_dec_blk_3way:
+ 	popq %r13;
+ 	popq %r14;
+ 	popq %r15;
++	pax_force_retaddr 0, 1
+ 	ret;
+ 
+diff --git a/arch/x86/crypto/twofish-x86_64-asm_64.S b/arch/x86/crypto/twofish-x86_64-asm_64.S
+index 7bcf3fc..f53832f 100644
+--- a/arch/x86/crypto/twofish-x86_64-asm_64.S
++++ b/arch/x86/crypto/twofish-x86_64-asm_64.S
+@@ -21,6 +21,7 @@
+ .text
+ 
+ #include <asm/asm-offsets.h>
++#include <asm/alternative-asm.h>
+ 
+ #define a_offset	0
+ #define b_offset	4
+@@ -268,6 +269,7 @@ twofish_enc_blk:
+ 
+ 	popq	R1
+ 	movq	$1,%rax
++	pax_force_retaddr 0, 1
+ 	ret
+ 
+ twofish_dec_blk:
+@@ -319,4 +321,5 @@ twofish_dec_blk:
+ 
+ 	popq	R1
+ 	movq	$1,%rax
++	pax_force_retaddr 0, 1
+ 	ret
+diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
+index fd84387..887aa7e 100644
+--- a/arch/x86/ia32/ia32_aout.c
++++ b/arch/x86/ia32/ia32_aout.c
+@@ -162,6 +162,8 @@ static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file,
+ 	unsigned long dump_start, dump_size;
+ 	struct user32 dump;
+ 
++	memset(&dump, 0, sizeof(dump));
++
+ 	fs = get_fs();
+ 	set_fs(KERNEL_DS);
+ 	has_dumped = 1;
+@@ -315,6 +317,13 @@ static int load_aout_binary(struct linux_binprm *bprm, struct pt_regs *regs)
+ 	current->mm->free_area_cache = TASK_UNMAPPED_BASE;
+ 	current->mm->cached_hole_size = 0;
+ 
++	retval = setup_arg_pages(bprm, IA32_STACK_TOP, EXSTACK_DEFAULT);
++	if (retval < 0) {
++		/* Someone check-me: is this error path enough? */
++		send_sig(SIGKILL, current, 0);
++		return retval;
++	}
++
+ 	install_exec_creds(bprm);
+ 	current->flags &= ~PF_FORKNOEXEC;
+ 
+@@ -410,13 +419,6 @@ beyond_if:
+ 
+ 	set_brk(current->mm->start_brk, current->mm->brk);
+ 
+-	retval = setup_arg_pages(bprm, IA32_STACK_TOP, EXSTACK_DEFAULT);
+-	if (retval < 0) {
+-		/* Someone check-me: is this error path enough? */
+-		send_sig(SIGKILL, current, 0);
+-		return retval;
+-	}
+-
+ 	current->mm->start_stack =
+ 		(unsigned long)create_aout_tables((char __user *)bprm->p, bprm);
+ 	/* start thread */
+diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
+index 6557769..ef6ae89 100644
+--- a/arch/x86/ia32/ia32_signal.c
++++ b/arch/x86/ia32/ia32_signal.c
+@@ -169,7 +169,7 @@ asmlinkage long sys32_sigaltstack(const stack_ia32_t __user *uss_ptr,
+ 	}
+ 	seg = get_fs();
+ 	set_fs(KERNEL_DS);
+-	ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss, regs->sp);
++	ret = do_sigaltstack(uss_ptr ? (const stack_t __force_user *)&uss : NULL, (stack_t __force_user *)&uoss, regs->sp);
+ 	set_fs(seg);
+ 	if (ret >= 0 && uoss_ptr)  {
+ 		if (!access_ok(VERIFY_WRITE, uoss_ptr, sizeof(stack_ia32_t)))
+@@ -370,7 +370,7 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
+  */
+ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs,
+ 				 size_t frame_size,
+-				 void **fpstate)
++				 void __user **fpstate)
+ {
+ 	unsigned long sp;
+ 
+@@ -391,7 +391,7 @@ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs,
+ 
+ 	if (used_math()) {
+ 		sp = sp - sig_xstate_ia32_size;
+-		*fpstate = (struct _fpstate_ia32 *) sp;
++		*fpstate = (struct _fpstate_ia32 __user *) sp;
+ 		if (save_i387_xstate_ia32(*fpstate) < 0)
+ 			return (void __user *) -1L;
+ 	}
+@@ -399,7 +399,7 @@ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs,
+ 	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;
++	sp = ((sp - 12) & -16ul) - 4;
+ 	return (void __user *) sp;
+ }
+ 
+@@ -457,7 +457,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
+ 		 * These are actually not used anymore, but left because some
+ 		 * gdb versions depend on them as a marker.
+ 		 */
+-		put_user_ex(*((u64 *)&code), (u64 *)frame->retcode);
++		put_user_ex(*((const u64 *)&code), (u64 __user *)frame->retcode);
+ 	} put_user_catch(err);
+ 
+ 	if (err)
+@@ -499,7 +499,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+ 		0xb8,
+ 		__NR_ia32_rt_sigreturn,
+ 		0x80cd,
+-		0,
++		0
+ 	};
+ 
+ 	frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate);
+@@ -529,16 +529,18 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+ 
+ 		if (ka->sa.sa_flags & SA_RESTORER)
+ 			restorer = ka->sa.sa_restorer;
++		else if (current->mm->context.vdso)
++			/* Return stub is in 32bit vsyscall page */
++			restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
+ 		else
+-			restorer = VDSO32_SYMBOL(current->mm->context.vdso,
+-						 rt_sigreturn);
++			restorer = &frame->retcode;
+ 		put_user_ex(ptr_to_compat(restorer), &frame->pretcode);
+ 
+ 		/*
+ 		 * Not actually used anymore, but left because some gdb
+ 		 * versions need it.
+ 		 */
+-		put_user_ex(*((u64 *)&code), (u64 *)frame->retcode);
++		put_user_ex(*((const u64 *)&code), (u64 __user *)frame->retcode);
+ 	} put_user_catch(err);
+ 
+ 	if (err)
+diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
+index a6253ec..4ad2120 100644
+--- a/arch/x86/ia32/ia32entry.S
++++ b/arch/x86/ia32/ia32entry.S
+@@ -13,7 +13,9 @@
+ #include <asm/thread_info.h>	
+ #include <asm/segment.h>
+ #include <asm/irqflags.h>
++#include <asm/pgtable.h>
+ #include <linux/linkage.h>
++#include <asm/alternative-asm.h>
+ 
+ /* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this.  */
+ #include <linux/elf-em.h>
+@@ -95,6 +97,32 @@ ENTRY(native_irq_enable_sysexit)
+ ENDPROC(native_irq_enable_sysexit)
+ #endif
+ 
++	.macro pax_enter_kernel_user
++	pax_set_fptr_mask
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	call pax_enter_kernel_user
++#endif
++	.endm
++
++	.macro pax_exit_kernel_user
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	call pax_exit_kernel_user
++#endif
++#ifdef CONFIG_PAX_RANDKSTACK
++	pushq %rax
++	pushq %r11
++	call pax_randomize_kstack
++	popq %r11
++	popq %rax
++#endif
++	.endm
++
++.macro pax_erase_kstack
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++	call pax_erase_kstack
++#endif
++.endm
++
+ /*
+  * 32bit SYSENTER instruction entry.
+  *
+@@ -121,12 +149,6 @@ ENTRY(ia32_sysenter_target)
+ 	CFI_REGISTER	rsp,rbp
+ 	SWAPGS_UNSAFE_STACK
+ 	movq	PER_CPU_VAR(kernel_stack), %rsp
+-	addq	$(KERNEL_STACK_OFFSET),%rsp
+-	/*
+-	 * No need to follow this irqs on/off section: the syscall
+-	 * disabled irqs, here we enable it straight after entry:
+-	 */
+-	ENABLE_INTERRUPTS(CLBR_NONE)
+  	movl	%ebp,%ebp		/* zero extension */
+ 	pushq_cfi $__USER32_DS
+ 	/*CFI_REL_OFFSET ss,0*/
+@@ -134,25 +156,39 @@ ENTRY(ia32_sysenter_target)
+ 	CFI_REL_OFFSET rsp,0
+ 	pushfq_cfi
+ 	/*CFI_REL_OFFSET rflags,0*/
+-	movl	8*3-THREAD_SIZE+TI_sysenter_return(%rsp), %r10d
+-	CFI_REGISTER rip,r10
++	orl	$X86_EFLAGS_IF,(%rsp)
++	GET_THREAD_INFO(%r11)
++	movl	TI_sysenter_return(%r11), %r11d
++	CFI_REGISTER rip,r11
+ 	pushq_cfi $__USER32_CS
+ 	/*CFI_REL_OFFSET cs,0*/
+ 	movl	%eax, %eax
+-	pushq_cfi %r10
++	pushq_cfi %r11
+ 	CFI_REL_OFFSET rip,0
+ 	pushq_cfi %rax
+ 	cld
+ 	SAVE_ARGS 0,1,0
++	pax_enter_kernel_user
++	/*
++	 * No need to follow this irqs on/off section: the syscall
++	 * disabled irqs, here we enable it straight after entry:
++	 */
++	ENABLE_INTERRUPTS(CLBR_NONE)
+  	/* no need to do an access_ok check here because rbp has been
+  	   32bit zero extended */ 
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	mov $PAX_USER_SHADOW_BASE,%r11
++	add %r11,%rbp
++#endif
++
+ 1:	movl	(%rbp),%ebp
+  	.section __ex_table,"a"
+  	.quad 1b,ia32_badarg
+  	.previous	
+-	GET_THREAD_INFO(%r10)
+-	orl    $TS_COMPAT,TI_status(%r10)
+-	testl  $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
++	GET_THREAD_INFO(%r11)
++	orl    $TS_COMPAT,TI_status(%r11)
++	testl  $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r11)
+ 	CFI_REMEMBER_STATE
+ 	jnz  sysenter_tracesys
+ 	cmpq	$(IA32_NR_syscalls-1),%rax
+@@ -162,13 +198,15 @@ sysenter_do_call:
+ sysenter_dispatch:
+ 	call	*ia32_sys_call_table(,%rax,8)
+ 	movq	%rax,RAX-ARGOFFSET(%rsp)
+-	GET_THREAD_INFO(%r10)
++	GET_THREAD_INFO(%r11)
+ 	DISABLE_INTERRUPTS(CLBR_NONE)
+ 	TRACE_IRQS_OFF
+-	testl	$_TIF_ALLWORK_MASK,TI_flags(%r10)
++	testl	$_TIF_ALLWORK_MASK,TI_flags(%r11)
+ 	jnz	sysexit_audit
+ sysexit_from_sys_call:
+-	andl    $~TS_COMPAT,TI_status(%r10)
++	pax_exit_kernel_user
++	pax_erase_kstack
++	andl    $~TS_COMPAT,TI_status(%r11)
+ 	/* clear IF, that popfq doesn't enable interrupts early */
+ 	andl  $~0x200,EFLAGS-R11(%rsp) 
+ 	movl	RIP-R11(%rsp),%edx		/* User %eip */
+@@ -194,6 +232,9 @@ sysexit_from_sys_call:
+ 	movl %eax,%esi			/* 2nd arg: syscall number */
+ 	movl $AUDIT_ARCH_I386,%edi	/* 1st arg: audit arch */
+ 	call audit_syscall_entry
++
++	pax_erase_kstack
++
+ 	movl RAX-ARGOFFSET(%rsp),%eax	/* reload syscall number */
+ 	cmpq $(IA32_NR_syscalls-1),%rax
+ 	ja ia32_badsys
+@@ -205,7 +246,7 @@ sysexit_from_sys_call:
+ 	.endm
+ 
+ 	.macro auditsys_exit exit
+-	testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
++	testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),TI_flags(%r11)
+ 	jnz ia32_ret_from_sys_call
+ 	TRACE_IRQS_ON
+ 	sti
+@@ -215,12 +256,12 @@ sysexit_from_sys_call:
+ 	movzbl %al,%edi		/* zero-extend that into %edi */
+ 	inc %edi /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */
+ 	call audit_syscall_exit
+-	GET_THREAD_INFO(%r10)
++	GET_THREAD_INFO(%r11)
+ 	movl RAX-ARGOFFSET(%rsp),%eax	/* reload syscall return value */
+ 	movl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),%edi
+ 	cli
+ 	TRACE_IRQS_OFF
+-	testl %edi,TI_flags(%r10)
++	testl %edi,TI_flags(%r11)
+ 	jz \exit
+ 	CLEAR_RREGS -ARGOFFSET
+ 	jmp int_with_check
+@@ -238,7 +279,7 @@ sysexit_audit:
+ 
+ sysenter_tracesys:
+ #ifdef CONFIG_AUDITSYSCALL
+-	testl	$(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
++	testl	$(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r11)
+ 	jz	sysenter_auditsys
+ #endif
+ 	SAVE_REST
+@@ -246,6 +287,9 @@ sysenter_tracesys:
+ 	movq	$-ENOSYS,RAX(%rsp)/* ptrace can change this for a bad syscall */
+ 	movq	%rsp,%rdi        /* &pt_regs -> arg1 */
+ 	call	syscall_trace_enter
++
++	pax_erase_kstack
++
+ 	LOAD_ARGS32 ARGOFFSET  /* reload args from stack in case ptrace changed it */
+ 	RESTORE_REST
+ 	cmpq	$(IA32_NR_syscalls-1),%rax
+@@ -277,19 +321,20 @@ ENDPROC(ia32_sysenter_target)
+ ENTRY(ia32_cstar_target)
+ 	CFI_STARTPROC32	simple
+ 	CFI_SIGNAL_FRAME
+-	CFI_DEF_CFA	rsp,KERNEL_STACK_OFFSET
++	CFI_DEF_CFA	rsp,0
+ 	CFI_REGISTER	rip,rcx
+ 	/*CFI_REGISTER	rflags,r11*/
+ 	SWAPGS_UNSAFE_STACK
+ 	movl	%esp,%r8d
+ 	CFI_REGISTER	rsp,r8
+ 	movq	PER_CPU_VAR(kernel_stack),%rsp
++	SAVE_ARGS 8*6,0,0
++	pax_enter_kernel_user
+ 	/*
+ 	 * No need to follow this irqs on/off section: the syscall
+ 	 * disabled irqs and here we enable it straight after entry:
+ 	 */
+ 	ENABLE_INTERRUPTS(CLBR_NONE)
+-	SAVE_ARGS 8,0,0
+ 	movl 	%eax,%eax	/* zero extension */
+ 	movq	%rax,ORIG_RAX-ARGOFFSET(%rsp)
+ 	movq	%rcx,RIP-ARGOFFSET(%rsp)
+@@ -305,13 +350,19 @@ ENTRY(ia32_cstar_target)
+ 	/* no need to do an access_ok check here because r8 has been
+ 	   32bit zero extended */ 
+ 	/* hardware stack frame is complete now */	
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	mov $PAX_USER_SHADOW_BASE,%r11
++	add %r11,%r8
++#endif
++
+ 1:	movl	(%r8),%r9d
+ 	.section __ex_table,"a"
+ 	.quad 1b,ia32_badarg
+ 	.previous	
+-	GET_THREAD_INFO(%r10)
+-	orl   $TS_COMPAT,TI_status(%r10)
+-	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
++	GET_THREAD_INFO(%r11)
++	orl   $TS_COMPAT,TI_status(%r11)
++	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r11)
+ 	CFI_REMEMBER_STATE
+ 	jnz   cstar_tracesys
+ 	cmpq $IA32_NR_syscalls-1,%rax
+@@ -321,13 +372,15 @@ cstar_do_call:
+ cstar_dispatch:
+ 	call *ia32_sys_call_table(,%rax,8)
+ 	movq %rax,RAX-ARGOFFSET(%rsp)
+-	GET_THREAD_INFO(%r10)
++	GET_THREAD_INFO(%r11)
+ 	DISABLE_INTERRUPTS(CLBR_NONE)
+ 	TRACE_IRQS_OFF
+-	testl $_TIF_ALLWORK_MASK,TI_flags(%r10)
++	testl $_TIF_ALLWORK_MASK,TI_flags(%r11)
+ 	jnz sysretl_audit
+ sysretl_from_sys_call:
+-	andl $~TS_COMPAT,TI_status(%r10)
++	pax_exit_kernel_user
++	pax_erase_kstack
++	andl $~TS_COMPAT,TI_status(%r11)
+ 	RESTORE_ARGS 0,-ARG_SKIP,0,0,0
+ 	movl RIP-ARGOFFSET(%rsp),%ecx
+ 	CFI_REGISTER rip,rcx
+@@ -355,7 +408,7 @@ sysretl_audit:
+ 
+ cstar_tracesys:
+ #ifdef CONFIG_AUDITSYSCALL
+-	testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
++	testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r11)
+ 	jz cstar_auditsys
+ #endif
+ 	xchgl %r9d,%ebp
+@@ -364,6 +417,9 @@ cstar_tracesys:
+ 	movq $-ENOSYS,RAX(%rsp)	/* ptrace can change this for a bad syscall */
+ 	movq %rsp,%rdi        /* &pt_regs -> arg1 */
+ 	call syscall_trace_enter
++
++	pax_erase_kstack
++
+ 	LOAD_ARGS32 ARGOFFSET, 1  /* reload args from stack in case ptrace changed it */
+ 	RESTORE_REST
+ 	xchgl %ebp,%r9d
+@@ -409,20 +465,21 @@ ENTRY(ia32_syscall)
+ 	CFI_REL_OFFSET	rip,RIP-RIP
+ 	PARAVIRT_ADJUST_EXCEPTION_FRAME
+ 	SWAPGS
+-	/*
+-	 * No need to follow this irqs on/off section: the syscall
+-	 * disabled irqs and here we enable it straight after entry:
+-	 */
+-	ENABLE_INTERRUPTS(CLBR_NONE)
+ 	movl %eax,%eax
+ 	pushq_cfi %rax
+ 	cld
+ 	/* note the registers are not zero extended to the sf.
+ 	   this could be a problem. */
+ 	SAVE_ARGS 0,1,0
+-	GET_THREAD_INFO(%r10)
+-	orl   $TS_COMPAT,TI_status(%r10)
+-	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
++	pax_enter_kernel_user
++	/*
++	 * No need to follow this irqs on/off section: the syscall
++	 * disabled irqs and here we enable it straight after entry:
++	 */
++	ENABLE_INTERRUPTS(CLBR_NONE)
++	GET_THREAD_INFO(%r11)
++	orl   $TS_COMPAT,TI_status(%r11)
++	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r11)
+ 	jnz ia32_tracesys
+ 	cmpq $(IA32_NR_syscalls-1),%rax
+ 	ja ia32_badsys
+@@ -441,6 +498,9 @@ ia32_tracesys:
+ 	movq $-ENOSYS,RAX(%rsp)	/* ptrace can change this for a bad syscall */
+ 	movq %rsp,%rdi        /* &pt_regs -> arg1 */
+ 	call syscall_trace_enter
++
++	pax_erase_kstack
++
+ 	LOAD_ARGS32 ARGOFFSET  /* reload args from stack in case ptrace changed it */
+ 	RESTORE_REST
+ 	cmpq $(IA32_NR_syscalls-1),%rax
+@@ -455,6 +515,7 @@ ia32_badsys:
+ 
+ quiet_ni_syscall:
+ 	movq $-ENOSYS,%rax
++	pax_force_retaddr
+ 	ret
+ 	CFI_ENDPROC
+ 	
+diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
+index f6f5c53..b358b28 100644
+--- a/arch/x86/ia32/sys_ia32.c
++++ b/arch/x86/ia32/sys_ia32.c
+@@ -69,8 +69,8 @@ asmlinkage long sys32_ftruncate64(unsigned int fd, unsigned long offset_low,
+  */
+ static int cp_stat64(struct stat64 __user *ubuf, struct kstat *stat)
+ {
+-	typeof(ubuf->st_uid) uid = 0;
+-	typeof(ubuf->st_gid) gid = 0;
++	typeof(((struct stat64 *)0)->st_uid) uid = 0;
++	typeof(((struct stat64 *)0)->st_gid) gid = 0;
+ 	SET_UID(uid, stat->uid);
+ 	SET_GID(gid, stat->gid);
+ 	if (!access_ok(VERIFY_WRITE, ubuf, sizeof(struct stat64)) ||
+@@ -308,8 +308,8 @@ asmlinkage long sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
+ 	}
+ 	set_fs(KERNEL_DS);
+ 	ret = sys_rt_sigprocmask(how,
+-				 set ? (sigset_t __user *)&s : NULL,
+-				 oset ? (sigset_t __user *)&s : NULL,
++				 set ? (sigset_t __force_user *)&s : NULL,
++				 oset ? (sigset_t __force_user *)&s : NULL,
+ 				 sigsetsize);
+ 	set_fs(old_fs);
+ 	if (ret)
+@@ -332,7 +332,7 @@ asmlinkage long sys32_alarm(unsigned int seconds)
+ 	return alarm_setitimer(seconds);
+ }
+ 
+-asmlinkage long sys32_waitpid(compat_pid_t pid, unsigned int *stat_addr,
++asmlinkage long sys32_waitpid(compat_pid_t pid, unsigned int __user *stat_addr,
+ 			      int options)
+ {
+ 	return compat_sys_wait4(pid, stat_addr, options, NULL);
+@@ -353,7 +353,7 @@ asmlinkage long sys32_sched_rr_get_interval(compat_pid_t pid,
+ 	mm_segment_t old_fs = get_fs();
+ 
+ 	set_fs(KERNEL_DS);
+-	ret = sys_sched_rr_get_interval(pid, (struct timespec __user *)&t);
++	ret = sys_sched_rr_get_interval(pid, (struct timespec __force_user *)&t);
+ 	set_fs(old_fs);
+ 	if (put_compat_timespec(&t, interval))
+ 		return -EFAULT;
+@@ -369,7 +369,7 @@ asmlinkage long sys32_rt_sigpending(compat_sigset_t __user *set,
+ 	mm_segment_t old_fs = get_fs();
+ 
+ 	set_fs(KERNEL_DS);
+-	ret = sys_rt_sigpending((sigset_t __user *)&s, sigsetsize);
++	ret = sys_rt_sigpending((sigset_t __force_user *)&s, sigsetsize);
+ 	set_fs(old_fs);
+ 	if (!ret) {
+ 		switch (_NSIG_WORDS) {
+@@ -394,7 +394,7 @@ asmlinkage long sys32_rt_sigqueueinfo(int pid, int sig,
+ 	if (copy_siginfo_from_user32(&info, uinfo))
+ 		return -EFAULT;
+ 	set_fs(KERNEL_DS);
+-	ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __user *)&info);
++	ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __force_user *)&info);
+ 	set_fs(old_fs);
+ 	return ret;
+ }
+@@ -439,7 +439,7 @@ asmlinkage long sys32_sendfile(int out_fd, int in_fd,
+ 		return -EFAULT;
+ 
+ 	set_fs(KERNEL_DS);
+-	ret = sys_sendfile(out_fd, in_fd, offset ? (off_t __user *)&of : NULL,
++	ret = sys_sendfile(out_fd, in_fd, offset ? (off_t __force_user *)&of : NULL,
+ 			   count);
+ 	set_fs(old_fs);
+ 
+diff --git a/arch/x86/include/asm/alternative-asm.h b/arch/x86/include/asm/alternative-asm.h
+index 091508b..7692c6f 100644
+--- a/arch/x86/include/asm/alternative-asm.h
++++ b/arch/x86/include/asm/alternative-asm.h
+@@ -4,10 +4,10 @@
+ 
+ #ifdef CONFIG_SMP
+ 	.macro LOCK_PREFIX
+-1:	lock
++672:	lock
+ 	.section .smp_locks,"a"
+ 	.balign 4
+-	.long 1b - .
++	.long 672b - .
+ 	.previous
+ 	.endm
+ #else
+@@ -15,6 +15,45 @@
+ 	.endm
+ #endif
+ 
++#ifdef KERNEXEC_PLUGIN
++	.macro pax_force_retaddr_bts rip=0
++	btsq $63,\rip(%rsp)
++	.endm
++#ifdef CONFIG_PAX_KERNEXEC_PLUGIN_METHOD_BTS
++	.macro pax_force_retaddr rip=0, reload=0
++	btsq $63,\rip(%rsp)
++	.endm
++	.macro pax_force_fptr ptr
++	btsq $63,\ptr
++	.endm
++	.macro pax_set_fptr_mask
++	.endm
++#endif
++#ifdef CONFIG_PAX_KERNEXEC_PLUGIN_METHOD_OR
++	.macro pax_force_retaddr rip=0, reload=0
++	.if \reload
++	pax_set_fptr_mask
++	.endif
++	orq %r10,\rip(%rsp)
++	.endm
++	.macro pax_force_fptr ptr
++	orq %r10,\ptr
++	.endm
++	.macro pax_set_fptr_mask
++	movabs $0x8000000000000000,%r10
++	.endm
++#endif
++#else
++	.macro pax_force_retaddr rip=0, reload=0
++	.endm
++	.macro pax_force_fptr ptr
++	.endm
++	.macro pax_force_retaddr_bts rip=0
++	.endm
++	.macro pax_set_fptr_mask
++	.endm
++#endif
++
+ .macro altinstruction_entry orig alt feature orig_len alt_len
+ 	.long \orig - .
+ 	.long \alt - .
+diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h
+index 37ad100..7d47faa 100644
+--- a/arch/x86/include/asm/alternative.h
++++ b/arch/x86/include/asm/alternative.h
+@@ -89,7 +89,7 @@ static inline int alternatives_text_reserved(void *start, void *end)
+       ".section .discard,\"aw\",@progbits\n"				\
+       "	 .byte 0xff + (664f-663f) - (662b-661b)\n" /* rlen <= slen */	\
+       ".previous\n"							\
+-      ".section .altinstr_replacement, \"ax\"\n"			\
++      ".section .altinstr_replacement, \"a\"\n"			\
+       "663:\n\t" newinstr "\n664:\n"		/* replacement     */	\
+       ".previous"
+ 
+diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
+index 1a6c09a..fec2432 100644
+--- a/arch/x86/include/asm/apic.h
++++ b/arch/x86/include/asm/apic.h
+@@ -45,7 +45,7 @@ static inline void generic_apic_probe(void)
+ 
+ #ifdef CONFIG_X86_LOCAL_APIC
+ 
+-extern unsigned int apic_verbosity;
++extern int apic_verbosity;
+ extern int local_apic_timer_c2_ok;
+ 
+ extern int disable_apic;
+diff --git a/arch/x86/include/asm/apm.h b/arch/x86/include/asm/apm.h
+index 20370c6..a2eb9b0 100644
+--- a/arch/x86/include/asm/apm.h
++++ b/arch/x86/include/asm/apm.h
+@@ -34,7 +34,7 @@ static inline void apm_bios_call_asm(u32 func, u32 ebx_in, u32 ecx_in,
+ 	__asm__ __volatile__(APM_DO_ZERO_SEGS
+ 		"pushl %%edi\n\t"
+ 		"pushl %%ebp\n\t"
+-		"lcall *%%cs:apm_bios_entry\n\t"
++		"lcall *%%ss:apm_bios_entry\n\t"
+ 		"setc %%al\n\t"
+ 		"popl %%ebp\n\t"
+ 		"popl %%edi\n\t"
+@@ -58,7 +58,7 @@ static inline u8 apm_bios_call_simple_asm(u32 func, u32 ebx_in,
+ 	__asm__ __volatile__(APM_DO_ZERO_SEGS
+ 		"pushl %%edi\n\t"
+ 		"pushl %%ebp\n\t"
+-		"lcall *%%cs:apm_bios_entry\n\t"
++		"lcall *%%ss:apm_bios_entry\n\t"
+ 		"setc %%bl\n\t"
+ 		"popl %%ebp\n\t"
+ 		"popl %%edi\n\t"
+diff --git a/arch/x86/include/asm/atomic.h b/arch/x86/include/asm/atomic.h
+index 58cb6d4..ca9010d 100644
+--- a/arch/x86/include/asm/atomic.h
++++ b/arch/x86/include/asm/atomic.h
+@@ -22,7 +22,18 @@
+  */
+ static inline int atomic_read(const atomic_t *v)
+ {
+-	return (*(volatile int *)&(v)->counter);
++	return (*(volatile const int *)&(v)->counter);
++}
++
++/**
++ * atomic_read_unchecked - read atomic variable
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically reads the value of @v.
++ */
++static inline int atomic_read_unchecked(const atomic_unchecked_t *v)
++{
++	return (*(volatile const int *)&(v)->counter);
+ }
+ 
+ /**
+@@ -38,6 +49,18 @@ static inline void atomic_set(atomic_t *v, int i)
+ }
+ 
+ /**
++ * atomic_set_unchecked - set atomic variable
++ * @v: pointer of type atomic_unchecked_t
++ * @i: required value
++ *
++ * Atomically sets the value of @v to @i.
++ */
++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i)
++{
++	v->counter = i;
++}
++
++/**
+  * atomic_add - add integer to atomic variable
+  * @i: integer value to add
+  * @v: pointer of type atomic_t
+@@ -46,7 +69,29 @@ static inline void atomic_set(atomic_t *v, int i)
+  */
+ static inline void atomic_add(int i, atomic_t *v)
+ {
+-	asm volatile(LOCK_PREFIX "addl %1,%0"
++	asm volatile(LOCK_PREFIX "addl %1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "subl %1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "+m" (v->counter)
++		     : "ir" (i));
++}
++
++/**
++ * atomic_add_unchecked - add integer to atomic variable
++ * @i: integer value to add
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically adds @i to @v.
++ */
++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v)
++{
++	asm volatile(LOCK_PREFIX "addl %1,%0\n"
+ 		     : "+m" (v->counter)
+ 		     : "ir" (i));
+ }
+@@ -60,7 +105,29 @@ static inline void atomic_add(int i, atomic_t *v)
+  */
+ static inline void atomic_sub(int i, atomic_t *v)
+ {
+-	asm volatile(LOCK_PREFIX "subl %1,%0"
++	asm volatile(LOCK_PREFIX "subl %1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "addl %1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "+m" (v->counter)
++		     : "ir" (i));
++}
++
++/**
++ * atomic_sub_unchecked - subtract integer from atomic variable
++ * @i: integer value to subtract
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically subtracts @i from @v.
++ */
++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v)
++{
++	asm volatile(LOCK_PREFIX "subl %1,%0\n"
+ 		     : "+m" (v->counter)
+ 		     : "ir" (i));
+ }
+@@ -78,7 +145,16 @@ static inline int atomic_sub_and_test(int i, atomic_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "subl %2,%0; sete %1"
++	asm volatile(LOCK_PREFIX "subl %2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "addl %2,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "+m" (v->counter), "=qm" (c)
+ 		     : "ir" (i) : "memory");
+ 	return c;
+@@ -92,7 +168,27 @@ static inline int atomic_sub_and_test(int i, atomic_t *v)
+  */
+ static inline void atomic_inc(atomic_t *v)
+ {
+-	asm volatile(LOCK_PREFIX "incl %0"
++	asm volatile(LOCK_PREFIX "incl %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "decl %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "+m" (v->counter));
++}
++
++/**
++ * atomic_inc_unchecked - increment atomic variable
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically increments @v by 1.
++ */
++static inline void atomic_inc_unchecked(atomic_unchecked_t *v)
++{
++	asm volatile(LOCK_PREFIX "incl %0\n"
+ 		     : "+m" (v->counter));
+ }
+ 
+@@ -104,7 +200,27 @@ static inline void atomic_inc(atomic_t *v)
+  */
+ static inline void atomic_dec(atomic_t *v)
+ {
+-	asm volatile(LOCK_PREFIX "decl %0"
++	asm volatile(LOCK_PREFIX "decl %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "incl %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "+m" (v->counter));
++}
++
++/**
++ * atomic_dec_unchecked - decrement atomic variable
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically decrements @v by 1.
++ */
++static inline void atomic_dec_unchecked(atomic_unchecked_t *v)
++{
++	asm volatile(LOCK_PREFIX "decl %0\n"
+ 		     : "+m" (v->counter));
+ }
+ 
+@@ -120,7 +236,16 @@ static inline int atomic_dec_and_test(atomic_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "decl %0; sete %1"
++	asm volatile(LOCK_PREFIX "decl %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "incl %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "+m" (v->counter), "=qm" (c)
+ 		     : : "memory");
+ 	return c != 0;
+@@ -138,7 +263,35 @@ static inline int atomic_inc_and_test(atomic_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "incl %0; sete %1"
++	asm volatile(LOCK_PREFIX "incl %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "decl %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
++		     : "+m" (v->counter), "=qm" (c)
++		     : : "memory");
++	return c != 0;
++}
++
++/**
++ * atomic_inc_and_test_unchecked - increment and test
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically increments @v by 1
++ * and returns true if the result is zero, or false for all
++ * other cases.
++ */
++static inline int atomic_inc_and_test_unchecked(atomic_unchecked_t *v)
++{
++	unsigned char c;
++
++	asm volatile(LOCK_PREFIX "incl %0\n"
++		     "sete %1\n"
+ 		     : "+m" (v->counter), "=qm" (c)
+ 		     : : "memory");
+ 	return c != 0;
+@@ -157,7 +310,16 @@ static inline int atomic_add_negative(int i, atomic_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "addl %2,%0; sets %1"
++	asm volatile(LOCK_PREFIX "addl %2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "subl %2,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sets %1\n"
+ 		     : "+m" (v->counter), "=qm" (c)
+ 		     : "ir" (i) : "memory");
+ 	return c;
+@@ -179,7 +341,7 @@ static inline int atomic_add_return(int i, atomic_t *v)
+ 		goto no_xadd;
+ #endif
+ 	/* Modern 486+ processor */
+-	return i + xadd(&v->counter, i);
++	return i + xadd_check_overflow(&v->counter, i);
+ 
+ #ifdef CONFIG_M386
+ no_xadd: /* Legacy 386 processor */
+@@ -192,6 +354,34 @@ no_xadd: /* Legacy 386 processor */
+ }
+ 
+ /**
++ * atomic_add_return_unchecked - add integer and return
++ * @i: integer value to add
++ * @v: pointer of type atomic_unchecked_t
++ *
++ * Atomically adds @i to @v and returns @i + @v
++ */
++static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v)
++{
++#ifdef CONFIG_M386
++	int __i;
++	unsigned long flags;
++	if (unlikely(boot_cpu_data.x86 <= 3))
++		goto no_xadd;
++#endif
++	/* Modern 486+ processor */
++	return i + xadd(&v->counter, i);
++
++#ifdef CONFIG_M386
++no_xadd: /* Legacy 386 processor */
++	raw_local_irq_save(flags);
++	__i = atomic_read_unchecked(v);
++	atomic_set_unchecked(v, i + __i);
++	raw_local_irq_restore(flags);
++	return i + __i;
++#endif
++}
++
++/**
+  * atomic_sub_return - subtract integer and return
+  * @v: pointer of type atomic_t
+  * @i: integer value to subtract
+@@ -204,6 +394,10 @@ static inline int atomic_sub_return(int i, atomic_t *v)
+ }
+ 
+ #define atomic_inc_return(v)  (atomic_add_return(1, v))
++static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v)
++{
++	return atomic_add_return_unchecked(1, v);
++}
+ #define atomic_dec_return(v)  (atomic_sub_return(1, v))
+ 
+ static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+@@ -211,11 +405,21 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+ 	return cmpxchg(&v->counter, old, new);
+ }
+ 
++static inline int atomic_cmpxchg_unchecked(atomic_unchecked_t *v, int old, int new)
++{
++	return cmpxchg(&v->counter, old, new);
++}
++
+ static inline int atomic_xchg(atomic_t *v, int new)
+ {
+ 	return xchg(&v->counter, new);
+ }
+ 
++static inline int atomic_xchg_unchecked(atomic_unchecked_t *v, int new)
++{
++	return xchg(&v->counter, new);
++}
++
+ /**
+  * __atomic_add_unless - add unless the number is already a given value
+  * @v: pointer of type atomic_t
+@@ -227,12 +431,25 @@ static inline int atomic_xchg(atomic_t *v, int new)
+  */
+ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+ {
+-	int c, old;
++	int c, old, new;
+ 	c = atomic_read(v);
+ 	for (;;) {
+-		if (unlikely(c == (u)))
++		if (unlikely(c == u))
+ 			break;
+-		old = atomic_cmpxchg((v), c, c + (a));
++
++		asm volatile("addl %2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++			     "jno 0f\n"
++			     "subl %2,%0\n"
++			     "int $4\n0:\n"
++			     _ASM_EXTABLE(0b, 0b)
++#endif
++
++			     : "=r" (new)
++			     : "0" (c), "ir" (a));
++
++		old = atomic_cmpxchg(v, c, new);
+ 		if (likely(old == c))
+ 			break;
+ 		c = old;
+@@ -240,6 +457,48 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+ 	return c;
+ }
+ 
++/**
++ * atomic_inc_not_zero_hint - increment if not null
++ * @v: pointer of type atomic_t
++ * @hint: probable value of the atomic before the increment
++ *
++ * This version of atomic_inc_not_zero() gives a hint of probable
++ * value of the atomic. This helps processor to not read the memory
++ * before doing the atomic read/modify/write cycle, lowering
++ * number of bus transactions on some arches.
++ *
++ * Returns: 0 if increment was not done, 1 otherwise.
++ */
++#define atomic_inc_not_zero_hint atomic_inc_not_zero_hint
++static inline int atomic_inc_not_zero_hint(atomic_t *v, int hint)
++{
++	int val, c = hint, new;
++
++	/* sanity test, should be removed by compiler if hint is a constant */
++	if (!hint)
++		return __atomic_add_unless(v, 1, 0);
++
++	do {
++		asm volatile("incl %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++			     "jno 0f\n"
++			     "decl %0\n"
++			     "int $4\n0:\n"
++			     _ASM_EXTABLE(0b, 0b)
++#endif
++
++			     : "=r" (new)
++			     : "0" (c));
++
++		val = atomic_cmpxchg(v, c, new);
++		if (val == c)
++			return 1;
++		c = val;
++	} while (c);
++
++	return 0;
++}
+ 
+ /*
+  * atomic_dec_if_positive - decrement by 1 if old value positive
+diff --git a/arch/x86/include/asm/atomic64_32.h b/arch/x86/include/asm/atomic64_32.h
+index 24098aa..1e37723 100644
+--- a/arch/x86/include/asm/atomic64_32.h
++++ b/arch/x86/include/asm/atomic64_32.h
+@@ -12,6 +12,14 @@ typedef struct {
+ 	u64 __aligned(8) counter;
+ } atomic64_t;
+ 
++#ifdef CONFIG_PAX_REFCOUNT
++typedef struct {
++	u64 __aligned(8) counter;
++} atomic64_unchecked_t;
++#else
++typedef atomic64_t atomic64_unchecked_t;
++#endif
++
+ #define ATOMIC64_INIT(val)	{ (val) }
+ 
+ #ifdef CONFIG_X86_CMPXCHG64
+@@ -38,6 +46,21 @@ static inline long long atomic64_cmpxchg(atomic64_t *v, long long o, long long n
+ }
+ 
+ /**
++ * atomic64_cmpxchg_unchecked - cmpxchg atomic64 variable
++ * @p: pointer to type atomic64_unchecked_t
++ * @o: expected value
++ * @n: new value
++ *
++ * Atomically sets @v to @n if it was equal to @o and returns
++ * the old value.
++ */
++
++static inline long long atomic64_cmpxchg_unchecked(atomic64_unchecked_t *v, long long o, long long n)
++{
++	return cmpxchg64(&v->counter, o, n);
++}
++
++/**
+  * atomic64_xchg - xchg atomic64 variable
+  * @v: pointer to type atomic64_t
+  * @n: value to assign
+@@ -77,6 +100,24 @@ static inline void atomic64_set(atomic64_t *v, long long i)
+ }
+ 
+ /**
++ * atomic64_set_unchecked - set atomic64 variable
++ * @v: pointer to type atomic64_unchecked_t
++ * @n: value to assign
++ *
++ * Atomically sets the value of @v to @n.
++ */
++static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long long i)
++{
++	unsigned high = (unsigned)(i >> 32);
++	unsigned low = (unsigned)i;
++	asm volatile(ATOMIC64_ALTERNATIVE(set)
++		     : "+b" (low), "+c" (high)
++		     : "S" (v)
++		     : "eax", "edx", "memory"
++		     );
++}
++
++/**
+  * atomic64_read - read atomic64 variable
+  * @v: pointer to type atomic64_t
+  *
+@@ -93,6 +134,22 @@ static inline long long atomic64_read(atomic64_t *v)
+  }
+ 
+ /**
++ * atomic64_read_unchecked - read atomic64 variable
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically reads the value of @v and returns it.
++ */
++static inline long long atomic64_read_unchecked(atomic64_unchecked_t *v)
++{
++	long long r;
++	asm volatile(ATOMIC64_ALTERNATIVE(read_unchecked)
++		     : "=A" (r), "+c" (v)
++		     : : "memory"
++		     );
++	return r;
++ }
++
++/**
+  * atomic64_add_return - add and return
+  * @i: integer value to add
+  * @v: pointer to type atomic64_t
+@@ -108,6 +165,22 @@ static inline long long atomic64_add_return(long long i, atomic64_t *v)
+ 	return i;
+ }
+ 
++/**
++ * atomic64_add_return_unchecked - add and return
++ * @i: integer value to add
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically adds @i to @v and returns @i + *@v
++ */
++static inline long long atomic64_add_return_unchecked(long long i, atomic64_unchecked_t *v)
++{
++	asm volatile(ATOMIC64_ALTERNATIVE(add_return_unchecked)
++		     : "+A" (i), "+c" (v)
++		     : : "memory"
++		     );
++	return i;
++}
++
+ /*
+  * Other variants with different arithmetic operators:
+  */
+@@ -131,6 +204,17 @@ static inline long long atomic64_inc_return(atomic64_t *v)
+ 	return a;
+ }
+ 
++static inline long long atomic64_inc_return_unchecked(atomic64_unchecked_t *v)
++{
++	long long a;
++	asm volatile(ATOMIC64_ALTERNATIVE(inc_return_unchecked)
++		     : "=A" (a)
++		     : "S" (v)
++		     : "memory", "ecx"
++		     );
++	return a;
++}
++
+ static inline long long atomic64_dec_return(atomic64_t *v)
+ {
+ 	long long a;
+@@ -159,6 +243,22 @@ static inline long long atomic64_add(long long i, atomic64_t *v)
+ }
+ 
+ /**
++ * atomic64_add_unchecked - add integer to atomic64 variable
++ * @i: integer value to add
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically adds @i to @v.
++ */
++static inline long long atomic64_add_unchecked(long long i, atomic64_unchecked_t *v)
++{
++	asm volatile(ATOMIC64_ALTERNATIVE_(add_unchecked, add_return_unchecked)
++		     : "+A" (i), "+c" (v)
++		     : : "memory"
++		     );
++	return i;
++}
++
++/**
+  * atomic64_sub - subtract the atomic64 variable
+  * @i: integer value to subtract
+  * @v: pointer to type atomic64_t
+diff --git a/arch/x86/include/asm/atomic64_64.h b/arch/x86/include/asm/atomic64_64.h
+index 0e1cbfc..5623683 100644
+--- a/arch/x86/include/asm/atomic64_64.h
++++ b/arch/x86/include/asm/atomic64_64.h
+@@ -18,7 +18,19 @@
+  */
+ static inline long atomic64_read(const atomic64_t *v)
+ {
+-	return (*(volatile long *)&(v)->counter);
++	return (*(volatile const long *)&(v)->counter);
++}
++
++/**
++ * atomic64_read_unchecked - read atomic64 variable
++ * @v: pointer of type atomic64_unchecked_t
++ *
++ * Atomically reads the value of @v.
++ * Doesn't imply a read memory barrier.
++ */
++static inline long atomic64_read_unchecked(const atomic64_unchecked_t *v)
++{
++	return (*(volatile const long *)&(v)->counter);
+ }
+ 
+ /**
+@@ -34,6 +46,18 @@ static inline void atomic64_set(atomic64_t *v, long i)
+ }
+ 
+ /**
++ * atomic64_set_unchecked - set atomic64 variable
++ * @v: pointer to type atomic64_unchecked_t
++ * @i: required value
++ *
++ * Atomically sets the value of @v to @i.
++ */
++static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long i)
++{
++	v->counter = i;
++}
++
++/**
+  * atomic64_add - add integer to atomic64 variable
+  * @i: integer value to add
+  * @v: pointer to type atomic64_t
+@@ -42,6 +66,28 @@ static inline void atomic64_set(atomic64_t *v, long i)
+  */
+ static inline void atomic64_add(long i, atomic64_t *v)
+ {
++	asm volatile(LOCK_PREFIX "addq %1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "subq %1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "=m" (v->counter)
++		     : "er" (i), "m" (v->counter));
++}
++
++/**
++ * atomic64_add_unchecked - add integer to atomic64 variable
++ * @i: integer value to add
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically adds @i to @v.
++ */
++static inline void atomic64_add_unchecked(long i, atomic64_unchecked_t *v)
++{
+ 	asm volatile(LOCK_PREFIX "addq %1,%0"
+ 		     : "=m" (v->counter)
+ 		     : "er" (i), "m" (v->counter));
+@@ -56,7 +102,29 @@ static inline void atomic64_add(long i, atomic64_t *v)
+  */
+ static inline void atomic64_sub(long i, atomic64_t *v)
+ {
+-	asm volatile(LOCK_PREFIX "subq %1,%0"
++	asm volatile(LOCK_PREFIX "subq %1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "addq %1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "=m" (v->counter)
++		     : "er" (i), "m" (v->counter));
++}
++
++/**
++ * atomic64_sub_unchecked - subtract the atomic64 variable
++ * @i: integer value to subtract
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically subtracts @i from @v.
++ */
++static inline void atomic64_sub_unchecked(long i, atomic64_unchecked_t *v)
++{
++	asm volatile(LOCK_PREFIX "subq %1,%0\n"
+ 		     : "=m" (v->counter)
+ 		     : "er" (i), "m" (v->counter));
+ }
+@@ -74,7 +142,16 @@ static inline int atomic64_sub_and_test(long i, atomic64_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "subq %2,%0; sete %1"
++	asm volatile(LOCK_PREFIX "subq %2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "addq %2,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "=m" (v->counter), "=qm" (c)
+ 		     : "er" (i), "m" (v->counter) : "memory");
+ 	return c;
+@@ -88,6 +165,27 @@ static inline int atomic64_sub_and_test(long i, atomic64_t *v)
+  */
+ static inline void atomic64_inc(atomic64_t *v)
+ {
++	asm volatile(LOCK_PREFIX "incq %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "decq %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "=m" (v->counter)
++		     : "m" (v->counter));
++}
++
++/**
++ * atomic64_inc_unchecked - increment atomic64 variable
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically increments @v by 1.
++ */
++static inline void atomic64_inc_unchecked(atomic64_unchecked_t *v)
++{
+ 	asm volatile(LOCK_PREFIX "incq %0"
+ 		     : "=m" (v->counter)
+ 		     : "m" (v->counter));
+@@ -101,7 +199,28 @@ static inline void atomic64_inc(atomic64_t *v)
+  */
+ static inline void atomic64_dec(atomic64_t *v)
+ {
+-	asm volatile(LOCK_PREFIX "decq %0"
++	asm volatile(LOCK_PREFIX "decq %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "incq %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     : "=m" (v->counter)
++		     : "m" (v->counter));
++}
++
++/**
++ * atomic64_dec_unchecked - decrement atomic64 variable
++ * @v: pointer to type atomic64_t
++ *
++ * Atomically decrements @v by 1.
++ */
++static inline void atomic64_dec_unchecked(atomic64_unchecked_t *v)
++{
++	asm volatile(LOCK_PREFIX "decq %0\n"
+ 		     : "=m" (v->counter)
+ 		     : "m" (v->counter));
+ }
+@@ -118,7 +237,16 @@ static inline int atomic64_dec_and_test(atomic64_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "decq %0; sete %1"
++	asm volatile(LOCK_PREFIX "decq %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "incq %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "=m" (v->counter), "=qm" (c)
+ 		     : "m" (v->counter) : "memory");
+ 	return c != 0;
+@@ -136,7 +264,16 @@ static inline int atomic64_inc_and_test(atomic64_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "incq %0; sete %1"
++	asm volatile(LOCK_PREFIX "incq %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "decq %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "=m" (v->counter), "=qm" (c)
+ 		     : "m" (v->counter) : "memory");
+ 	return c != 0;
+@@ -155,7 +292,16 @@ static inline int atomic64_add_negative(long i, atomic64_t *v)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(LOCK_PREFIX "addq %2,%0; sets %1"
++	asm volatile(LOCK_PREFIX "addq %2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX "subq %2,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sets %1\n"
+ 		     : "=m" (v->counter), "=qm" (c)
+ 		     : "er" (i), "m" (v->counter) : "memory");
+ 	return c;
+@@ -170,6 +316,18 @@ static inline int atomic64_add_negative(long i, atomic64_t *v)
+  */
+ static inline long atomic64_add_return(long i, atomic64_t *v)
+ {
++	return i + xadd_check_overflow(&v->counter, i);
++}
++
++/**
++ * atomic64_add_return_unchecked - add and return
++ * @i: integer value to add
++ * @v: pointer to type atomic64_unchecked_t
++ *
++ * Atomically adds @i to @v and returns @i + @v
++ */
++static inline long atomic64_add_return_unchecked(long i, atomic64_unchecked_t *v)
++{
+ 	return i + xadd(&v->counter, i);
+ }
+ 
+@@ -179,6 +337,10 @@ static inline long atomic64_sub_return(long i, atomic64_t *v)
+ }
+ 
+ #define atomic64_inc_return(v)  (atomic64_add_return(1, (v)))
++static inline long atomic64_inc_return_unchecked(atomic64_unchecked_t *v)
++{
++	return atomic64_add_return_unchecked(1, v);
++}
+ #define atomic64_dec_return(v)  (atomic64_sub_return(1, (v)))
+ 
+ static inline long atomic64_cmpxchg(atomic64_t *v, long old, long new)
+@@ -186,6 +348,11 @@ static inline long atomic64_cmpxchg(atomic64_t *v, long old, long new)
+ 	return cmpxchg(&v->counter, old, new);
+ }
+ 
++static inline long atomic64_cmpxchg_unchecked(atomic64_unchecked_t *v, long old, long new)
++{
++	return cmpxchg(&v->counter, old, new);
++}
++
+ static inline long atomic64_xchg(atomic64_t *v, long new)
+ {
+ 	return xchg(&v->counter, new);
+@@ -202,17 +369,30 @@ static inline long atomic64_xchg(atomic64_t *v, long new)
+  */
+ static inline int atomic64_add_unless(atomic64_t *v, long a, long u)
+ {
+-	long c, old;
++	long c, old, new;
+ 	c = atomic64_read(v);
+ 	for (;;) {
+-		if (unlikely(c == (u)))
++		if (unlikely(c == u))
+ 			break;
+-		old = atomic64_cmpxchg((v), c, c + (a));
++
++		asm volatile("add %2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++			     "jno 0f\n"
++			     "sub %2,%0\n"
++			     "int $4\n0:\n"
++			     _ASM_EXTABLE(0b, 0b)
++#endif
++
++			     : "=r" (new)
++			     : "0" (c), "ir" (a));
++
++		old = atomic64_cmpxchg(v, c, new);
+ 		if (likely(old == c))
+ 			break;
+ 		c = old;
+ 	}
+-	return c != (u);
++	return c != u;
+ }
+ 
+ #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
+diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h
+index 1775d6e..b65017f 100644
+--- a/arch/x86/include/asm/bitops.h
++++ b/arch/x86/include/asm/bitops.h
+@@ -38,7 +38,7 @@
+  * a mask operation on a byte.
+  */
+ #define IS_IMMEDIATE(nr)		(__builtin_constant_p(nr))
+-#define CONST_MASK_ADDR(nr, addr)	BITOP_ADDR((void *)(addr) + ((nr)>>3))
++#define CONST_MASK_ADDR(nr, addr)	BITOP_ADDR((volatile void *)(addr) + ((nr)>>3))
+ #define CONST_MASK(nr)			(1 << ((nr) & 7))
+ 
+ /**
+diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
+index 5e1a2ee..c9f9533 100644
+--- a/arch/x86/include/asm/boot.h
++++ b/arch/x86/include/asm/boot.h
+@@ -11,10 +11,15 @@
+ #include <asm/pgtable_types.h>
+ 
+ /* Physical address where kernel should be loaded. */
+-#define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \
++#define ____LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \
+ 				+ (CONFIG_PHYSICAL_ALIGN - 1)) \
+ 				& ~(CONFIG_PHYSICAL_ALIGN - 1))
+ 
++#ifndef __ASSEMBLY__
++extern unsigned char __LOAD_PHYSICAL_ADDR[];
++#define LOAD_PHYSICAL_ADDR ((unsigned long)__LOAD_PHYSICAL_ADDR)
++#endif
++
+ /* Minimum kernel alignment, as a power of two */
+ #ifdef CONFIG_X86_64
+ #define MIN_KERNEL_ALIGN_LG2	PMD_SHIFT
+diff --git a/arch/x86/include/asm/cache.h b/arch/x86/include/asm/cache.h
+index 48f99f1..d78ebf9 100644
+--- a/arch/x86/include/asm/cache.h
++++ b/arch/x86/include/asm/cache.h
+@@ -5,12 +5,13 @@
+ 
+ /* L1 cache line size */
+ #define L1_CACHE_SHIFT	(CONFIG_X86_L1_CACHE_SHIFT)
+-#define L1_CACHE_BYTES	(1 << L1_CACHE_SHIFT)
++#define L1_CACHE_BYTES	(_AC(1,UL) << L1_CACHE_SHIFT)
+ 
+ #define __read_mostly __attribute__((__section__(".data..read_mostly")))
++#define __read_only __attribute__((__section__(".data..read_only")))
+ 
+ #define INTERNODE_CACHE_SHIFT CONFIG_X86_INTERNODE_CACHE_SHIFT
+-#define INTERNODE_CACHE_BYTES (1 << INTERNODE_CACHE_SHIFT)
++#define INTERNODE_CACHE_BYTES (_AC(1,UL) << INTERNODE_CACHE_SHIFT)
+ 
+ #ifdef CONFIG_X86_VSMP
+ #ifdef CONFIG_SMP
+diff --git a/arch/x86/include/asm/cacheflush.h b/arch/x86/include/asm/cacheflush.h
+index 4e12668..501d239 100644
+--- a/arch/x86/include/asm/cacheflush.h
++++ b/arch/x86/include/asm/cacheflush.h
+@@ -26,7 +26,7 @@ static inline unsigned long get_page_memtype(struct page *pg)
+ 	unsigned long pg_flags = pg->flags & _PGMT_MASK;
+ 
+ 	if (pg_flags == _PGMT_DEFAULT)
+-		return -1;
++		return ~0UL;
+ 	else if (pg_flags == _PGMT_WC)
+ 		return _PAGE_CACHE_WC;
+ 	else if (pg_flags == _PGMT_UC_MINUS)
+diff --git a/arch/x86/include/asm/checksum_32.h b/arch/x86/include/asm/checksum_32.h
+index 46fc474..b02b0f9 100644
+--- a/arch/x86/include/asm/checksum_32.h
++++ b/arch/x86/include/asm/checksum_32.h
+@@ -31,6 +31,14 @@ asmlinkage __wsum csum_partial_copy_generic(const void *src, void *dst,
+ 					    int len, __wsum sum,
+ 					    int *src_err_ptr, int *dst_err_ptr);
+ 
++asmlinkage __wsum csum_partial_copy_generic_to_user(const void *src, void *dst,
++						  int len, __wsum sum,
++						  int *src_err_ptr, int *dst_err_ptr);
++
++asmlinkage __wsum csum_partial_copy_generic_from_user(const void *src, void *dst,
++						  int len, __wsum sum,
++						  int *src_err_ptr, int *dst_err_ptr);
++
+ /*
+  *	Note: when you get a NULL pointer exception here this means someone
+  *	passed in an incorrect kernel address to one of these functions.
+@@ -50,7 +58,7 @@ static inline __wsum csum_partial_copy_from_user(const void __user *src,
+ 						 int *err_ptr)
+ {
+ 	might_sleep();
+-	return csum_partial_copy_generic((__force void *)src, dst,
++	return csum_partial_copy_generic_from_user((__force void *)src, dst,
+ 					 len, sum, err_ptr, NULL);
+ }
+ 
+@@ -178,7 +186,7 @@ static inline __wsum csum_and_copy_to_user(const void *src,
+ {
+ 	might_sleep();
+ 	if (access_ok(VERIFY_WRITE, dst, len))
+-		return csum_partial_copy_generic(src, (__force void *)dst,
++		return csum_partial_copy_generic_to_user(src, (__force void *)dst,
+ 						 len, sum, NULL, err_ptr);
+ 
+ 	if (len)
+diff --git a/arch/x86/include/asm/cmpxchg.h b/arch/x86/include/asm/cmpxchg.h
+index 5d3acdf..6447a02 100644
+--- a/arch/x86/include/asm/cmpxchg.h
++++ b/arch/x86/include/asm/cmpxchg.h
+@@ -14,6 +14,8 @@ extern void __cmpxchg_wrong_size(void)
+ 	__compiletime_error("Bad argument size for cmpxchg");
+ extern void __xadd_wrong_size(void)
+ 	__compiletime_error("Bad argument size for xadd");
++extern void __xadd_check_overflow_wrong_size(void)
++	__compiletime_error("Bad argument size for xadd_check_overflow");
+ 
+ /*
+  * Constants for operation sizes. On 32-bit, the 64-bit size it set to
+@@ -195,6 +197,34 @@ extern void __xadd_wrong_size(void)
+ 		__ret;							\
+ 	})
+ 
++#define __xadd_check_overflow(ptr, inc, lock)				\
++	({								\
++	        __typeof__ (*(ptr)) __ret = (inc);			\
++		switch (sizeof(*(ptr))) {				\
++		case __X86_CASE_L:					\
++			asm volatile (lock "xaddl %0, %1\n"		\
++				      "jno 0f\n"			\
++				      "mov %0,%1\n"			\
++				      "int $4\n0:\n"			\
++				      _ASM_EXTABLE(0b, 0b)		\
++				      : "+r" (__ret), "+m" (*(ptr))	\
++				      : : "memory", "cc");		\
++			break;						\
++		case __X86_CASE_Q:					\
++			asm volatile (lock "xaddq %q0, %1\n"		\
++				      "jno 0f\n"			\
++				      "mov %0,%1\n"			\
++				      "int $4\n0:\n"			\
++				      _ASM_EXTABLE(0b, 0b)		\
++				      : "+r" (__ret), "+m" (*(ptr))	\
++				      : : "memory", "cc");		\
++			break;						\
++		default:						\
++			__xadd_check_overflow_wrong_size();		\
++		}							\
++		__ret;							\
++	})
++
+ /*
+  * xadd() adds "inc" to "*ptr" and atomically returns the previous
+  * value of "*ptr".
+@@ -207,4 +237,6 @@ extern void __xadd_wrong_size(void)
+ #define xadd_sync(ptr, inc)	__xadd((ptr), (inc), "lock; ")
+ #define xadd_local(ptr, inc)	__xadd((ptr), (inc), "")
+ 
++#define xadd_check_overflow(ptr, inc)	__xadd_check_overflow((ptr), (inc), LOCK_PREFIX)
++
+ #endif	/* ASM_X86_CMPXCHG_H */
+diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h
+index f3444f7..051a196 100644
+--- a/arch/x86/include/asm/cpufeature.h
++++ b/arch/x86/include/asm/cpufeature.h
+@@ -363,7 +363,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit)
+ 			     ".section .discard,\"aw\",@progbits\n"
+ 			     " .byte 0xff + (4f-3f) - (2b-1b)\n" /* size check */
+ 			     ".previous\n"
+-			     ".section .altinstr_replacement,\"ax\"\n"
++			     ".section .altinstr_replacement,\"a\"\n"
+ 			     "3: movb $1,%0\n"
+ 			     "4:\n"
+ 			     ".previous\n"
+diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h
+index 41935fa..3b40db8 100644
+--- a/arch/x86/include/asm/desc.h
++++ b/arch/x86/include/asm/desc.h
+@@ -4,6 +4,7 @@
+ #include <asm/desc_defs.h>
+ #include <asm/ldt.h>
+ #include <asm/mmu.h>
++#include <asm/pgtable.h>
+ 
+ #include <linux/smp.h>
+ 
+@@ -16,6 +17,7 @@ static inline void fill_ldt(struct desc_struct *desc, const struct user_desc *in
+ 
+ 	desc->type		= (info->read_exec_only ^ 1) << 1;
+ 	desc->type	       |= info->contents << 2;
++	desc->type	       |= info->seg_not_present ^ 1;
+ 
+ 	desc->s			= 1;
+ 	desc->dpl		= 0x3;
+@@ -34,17 +36,12 @@ static inline void fill_ldt(struct desc_struct *desc, const struct user_desc *in
+ }
+ 
+ extern struct desc_ptr idt_descr;
+-extern gate_desc idt_table[];
+-
+-struct gdt_page {
+-	struct desc_struct gdt[GDT_ENTRIES];
+-} __attribute__((aligned(PAGE_SIZE)));
+-
+-DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page);
++extern gate_desc idt_table[256];
+ 
++extern struct desc_struct cpu_gdt_table[NR_CPUS][PAGE_SIZE / sizeof(struct desc_struct)];
+ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
+ {
+-	return per_cpu(gdt_page, cpu).gdt;
++	return cpu_gdt_table[cpu];
+ }
+ 
+ #ifdef CONFIG_X86_64
+@@ -69,8 +66,14 @@ static inline void pack_gate(gate_desc *gate, unsigned char type,
+ 			     unsigned long base, unsigned dpl, unsigned flags,
+ 			     unsigned short seg)
+ {
+-	gate->a = (seg << 16) | (base & 0xffff);
+-	gate->b = (base & 0xffff0000) | (((0x80 | type | (dpl << 5)) & 0xff) << 8);
++	gate->gate.offset_low	= base;
++	gate->gate.seg		= seg;
++	gate->gate.reserved	= 0;
++	gate->gate.type		= type;
++	gate->gate.s		= 0;
++	gate->gate.dpl		= dpl;
++	gate->gate.p		= 1;
++	gate->gate.offset_high	= base >> 16;
+ }
+ 
+ #endif
+@@ -115,12 +118,16 @@ static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries)
+ 
+ static inline void native_write_idt_entry(gate_desc *idt, int entry, const gate_desc *gate)
+ {
++	pax_open_kernel();
+ 	memcpy(&idt[entry], gate, sizeof(*gate));
++	pax_close_kernel();
+ }
+ 
+ static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry, const void *desc)
+ {
++	pax_open_kernel();
+ 	memcpy(&ldt[entry], desc, 8);
++	pax_close_kernel();
+ }
+ 
+ static inline void
+@@ -134,7 +141,9 @@ native_write_gdt_entry(struct desc_struct *gdt, int entry, const void *desc, int
+ 	default:	size = sizeof(*gdt);		break;
+ 	}
+ 
++	pax_open_kernel();
+ 	memcpy(&gdt[entry], desc, size);
++	pax_close_kernel();
+ }
+ 
+ static inline void pack_descriptor(struct desc_struct *desc, unsigned long base,
+@@ -207,7 +216,9 @@ static inline void native_set_ldt(const void *addr, unsigned int entries)
+ 
+ static inline void native_load_tr_desc(void)
+ {
++	pax_open_kernel();
+ 	asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8));
++	pax_close_kernel();
+ }
+ 
+ static inline void native_load_gdt(const struct desc_ptr *dtr)
+@@ -244,8 +255,10 @@ static inline void native_load_tls(struct thread_struct *t, unsigned int cpu)
+ 	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
+ 	unsigned int i;
+ 
++	pax_open_kernel();
+ 	for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++)
+ 		gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i];
++	pax_close_kernel();
+ }
+ 
+ #define _LDT_empty(info)				\
+@@ -307,7 +320,7 @@ static inline void set_desc_limit(struct desc_struct *desc, unsigned long limit)
+ 	desc->limit = (limit >> 16) & 0xf;
+ }
+ 
+-static inline void _set_gate(int gate, unsigned type, void *addr,
++static inline void _set_gate(int gate, unsigned type, const void *addr,
+ 			     unsigned dpl, unsigned ist, unsigned seg)
+ {
+ 	gate_desc s;
+@@ -326,7 +339,7 @@ static inline void _set_gate(int gate, unsigned type, void *addr,
+  * Pentium F0 0F bugfix can have resulted in the mapped
+  * IDT being write-protected.
+  */
+-static inline void set_intr_gate(unsigned int n, void *addr)
++static inline void set_intr_gate(unsigned int n, const void *addr)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+ 	_set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS);
+@@ -356,19 +369,19 @@ static inline void alloc_intr_gate(unsigned int n, void *addr)
+ /*
+  * This routine sets up an interrupt gate at directory privilege level 3.
+  */
+-static inline void set_system_intr_gate(unsigned int n, void *addr)
++static inline void set_system_intr_gate(unsigned int n, const void *addr)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+ 	_set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS);
+ }
+ 
+-static inline void set_system_trap_gate(unsigned int n, void *addr)
++static inline void set_system_trap_gate(unsigned int n, const void *addr)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+ 	_set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS);
+ }
+ 
+-static inline void set_trap_gate(unsigned int n, void *addr)
++static inline void set_trap_gate(unsigned int n, const void *addr)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+ 	_set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS);
+@@ -377,19 +390,31 @@ static inline void set_trap_gate(unsigned int n, void *addr)
+ static inline void set_task_gate(unsigned int n, unsigned int gdt_entry)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+-	_set_gate(n, GATE_TASK, (void *)0, 0, 0, (gdt_entry<<3));
++	_set_gate(n, GATE_TASK, (const void *)0, 0, 0, (gdt_entry<<3));
+ }
+ 
+-static inline void set_intr_gate_ist(int n, void *addr, unsigned ist)
++static inline void set_intr_gate_ist(int n, const void *addr, unsigned ist)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+ 	_set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS);
+ }
+ 
+-static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist)
++static inline void set_system_intr_gate_ist(int n, const void *addr, unsigned ist)
+ {
+ 	BUG_ON((unsigned)n > 0xFF);
+ 	_set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS);
+ }
+ 
++#ifdef CONFIG_X86_32
++static inline void set_user_cs(unsigned long base, unsigned long limit, int cpu)
++{
++	struct desc_struct d;
++
++	if (likely(limit))
++		limit = (limit - 1UL) >> PAGE_SHIFT;
++	pack_descriptor(&d, base, limit, 0xFB, 0xC);
++	write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_DEFAULT_USER_CS, &d, DESCTYPE_S);
++}
++#endif
++
+ #endif /* _ASM_X86_DESC_H */
+diff --git a/arch/x86/include/asm/desc_defs.h b/arch/x86/include/asm/desc_defs.h
+index 278441f..b95a174 100644
+--- a/arch/x86/include/asm/desc_defs.h
++++ b/arch/x86/include/asm/desc_defs.h
+@@ -31,6 +31,12 @@ struct desc_struct {
+ 			unsigned base1: 8, type: 4, s: 1, dpl: 2, p: 1;
+ 			unsigned limit: 4, avl: 1, l: 1, d: 1, g: 1, base2: 8;
+ 		};
++		struct {
++			u16 offset_low;
++			u16 seg;
++			unsigned reserved: 8, type: 4, s: 1, dpl: 2, p: 1;
++			unsigned offset_high: 16;
++		} gate;
+ 	};
+ } __attribute__((packed));
+ 
+diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h
+index 908b969..a1f4eb4 100644
+--- a/arch/x86/include/asm/e820.h
++++ b/arch/x86/include/asm/e820.h
+@@ -69,7 +69,7 @@ struct e820map {
+ #define ISA_START_ADDRESS	0xa0000
+ #define ISA_END_ADDRESS		0x100000
+ 
+-#define BIOS_BEGIN		0x000a0000
++#define BIOS_BEGIN		0x000c0000
+ #define BIOS_END		0x00100000
+ 
+ #define BIOS_ROM_BASE		0xffe00000
+diff --git a/arch/x86/include/asm/elf.h b/arch/x86/include/asm/elf.h
+index 5f962df..7289f09 100644
+--- a/arch/x86/include/asm/elf.h
++++ b/arch/x86/include/asm/elf.h
+@@ -238,7 +238,25 @@ extern int force_personality32;
+    the loader.  We need to make sure that it is out of the way of the program
+    that it will "exec", and that there is sufficient room for the brk.  */
+ 
++#ifdef CONFIG_PAX_SEGMEXEC
++#define ELF_ET_DYN_BASE		((current->mm->pax_flags & MF_PAX_SEGMEXEC) ? SEGMEXEC_TASK_SIZE/3*2 : TASK_SIZE/3*2)
++#else
+ #define ELF_ET_DYN_BASE		(TASK_SIZE / 3 * 2)
++#endif
++
++#ifdef CONFIG_PAX_ASLR
++#ifdef CONFIG_X86_32
++#define PAX_ELF_ET_DYN_BASE	0x10000000UL
++
++#define PAX_DELTA_MMAP_LEN	(current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16)
++#define PAX_DELTA_STACK_LEN	(current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16)
++#else
++#define PAX_ELF_ET_DYN_BASE	0x400000UL
++
++#define PAX_DELTA_MMAP_LEN	((test_thread_flag(TIF_IA32)) ? 16 : TASK_SIZE_MAX_SHIFT - PAGE_SHIFT - 3)
++#define PAX_DELTA_STACK_LEN	((test_thread_flag(TIF_IA32)) ? 16 : TASK_SIZE_MAX_SHIFT - PAGE_SHIFT - 3)
++#endif
++#endif
+ 
+ /* This yields a mask that user programs can use to figure out what
+    instruction set this CPU supports.  This could be done in user space,
+@@ -291,9 +309,7 @@ do {									\
+ 
+ #define ARCH_DLINFO							\
+ do {									\
+-	if (vdso_enabled)						\
+-		NEW_AUX_ENT(AT_SYSINFO_EHDR,				\
+-			    (unsigned long)current->mm->context.vdso);	\
++	NEW_AUX_ENT(AT_SYSINFO_EHDR, current->mm->context.vdso);	\
+ } while (0)
+ 
+ #define AT_SYSINFO		32
+@@ -304,7 +320,7 @@ do {									\
+ 
+ #endif /* !CONFIG_X86_32 */
+ 
+-#define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
++#define VDSO_CURRENT_BASE	(current->mm->context.vdso)
+ 
+ #define VDSO_ENTRY							\
+ 	((unsigned long)VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
+@@ -318,9 +334,6 @@ extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+ extern int syscall32_setup_pages(struct linux_binprm *, int exstack);
+ #define compat_arch_setup_additional_pages	syscall32_setup_pages
+ 
+-extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+-#define arch_randomize_brk arch_randomize_brk
+-
+ /*
+  * True on X86_32 or when emulating IA32 on X86_64
+  */
+diff --git a/arch/x86/include/asm/emergency-restart.h b/arch/x86/include/asm/emergency-restart.h
+index cc70c1c..d96d011 100644
+--- a/arch/x86/include/asm/emergency-restart.h
++++ b/arch/x86/include/asm/emergency-restart.h
+@@ -15,6 +15,6 @@ enum reboot_type {
+ 
+ extern enum reboot_type reboot_type;
+ 
+-extern void machine_emergency_restart(void);
++extern void machine_emergency_restart(void) __noreturn;
+ 
+ #endif /* _ASM_X86_EMERGENCY_RESTART_H */
+diff --git a/arch/x86/include/asm/floppy.h b/arch/x86/include/asm/floppy.h
+index dbe82a5..c6d8a00 100644
+--- a/arch/x86/include/asm/floppy.h
++++ b/arch/x86/include/asm/floppy.h
+@@ -157,6 +157,7 @@ static unsigned long dma_mem_alloc(unsigned long size)
+ }
+ 
+ 
++static unsigned long vdma_mem_alloc(unsigned long size) __size_overflow(1);
+ static unsigned long vdma_mem_alloc(unsigned long size)
+ {
+ 	return (unsigned long)vmalloc(size);
+diff --git a/arch/x86/include/asm/futex.h b/arch/x86/include/asm/futex.h
+index d09bb03..4ea4194 100644
+--- a/arch/x86/include/asm/futex.h
++++ b/arch/x86/include/asm/futex.h
+@@ -12,16 +12,18 @@
+ #include <asm/system.h>
+ 
+ #define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg)	\
++	typecheck(u32 __user *, uaddr);				\
+ 	asm volatile("1:\t" insn "\n"				\
+ 		     "2:\t.section .fixup,\"ax\"\n"		\
+ 		     "3:\tmov\t%3, %1\n"			\
+ 		     "\tjmp\t2b\n"				\
+ 		     "\t.previous\n"				\
+ 		     _ASM_EXTABLE(1b, 3b)			\
+-		     : "=r" (oldval), "=r" (ret), "+m" (*uaddr)	\
++		     : "=r" (oldval), "=r" (ret), "+m" (*(u32 __user *)____m(uaddr))\
+ 		     : "i" (-EFAULT), "0" (oparg), "1" (0))
+ 
+ #define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg)	\
++	typecheck(u32 __user *, uaddr);				\
+ 	asm volatile("1:\tmovl	%2, %0\n"			\
+ 		     "\tmovl\t%0, %3\n"				\
+ 		     "\t" insn "\n"				\
+@@ -34,7 +36,7 @@
+ 		     _ASM_EXTABLE(1b, 4b)			\
+ 		     _ASM_EXTABLE(2b, 4b)			\
+ 		     : "=&a" (oldval), "=&r" (ret),		\
+-		       "+m" (*uaddr), "=&r" (tem)		\
++		       "+m" (*(u32 __user *)____m(uaddr)), "=&r" (tem)	\
+ 		     : "r" (oparg), "i" (-EFAULT), "1" (0))
+ 
+ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
+@@ -61,10 +63,10 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
+ 
+ 	switch (op) {
+ 	case FUTEX_OP_SET:
+-		__futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg);
++		__futex_atomic_op1(__copyuser_seg"xchgl %0, %2", ret, oldval, uaddr, oparg);
+ 		break;
+ 	case FUTEX_OP_ADD:
+-		__futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval,
++		__futex_atomic_op1(LOCK_PREFIX __copyuser_seg"xaddl %0, %2", ret, oldval,
+ 				   uaddr, oparg);
+ 		break;
+ 	case FUTEX_OP_OR:
+@@ -123,13 +125,13 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+ 	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
+ 		return -EFAULT;
+ 
+-	asm volatile("1:\t" LOCK_PREFIX "cmpxchgl %4, %2\n"
++	asm volatile("1:\t" LOCK_PREFIX __copyuser_seg"cmpxchgl %4, %2\n"
+ 		     "2:\t.section .fixup, \"ax\"\n"
+ 		     "3:\tmov     %3, %0\n"
+ 		     "\tjmp     2b\n"
+ 		     "\t.previous\n"
+ 		     _ASM_EXTABLE(1b, 3b)
+-		     : "+r" (ret), "=a" (oldval), "+m" (*uaddr)
++		     : "+r" (ret), "=a" (oldval), "+m" (*(u32 __user *)____m(uaddr))
+ 		     : "i" (-EFAULT), "r" (newval), "1" (oldval)
+ 		     : "memory"
+ 	);
+diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
+index eb92a6e..b98b2f4 100644
+--- a/arch/x86/include/asm/hw_irq.h
++++ b/arch/x86/include/asm/hw_irq.h
+@@ -136,8 +136,8 @@ extern void setup_ioapic_dest(void);
+ extern void enable_IO_APIC(void);
+ 
+ /* Statistics */
+-extern atomic_t irq_err_count;
+-extern atomic_t irq_mis_count;
++extern atomic_unchecked_t irq_err_count;
++extern atomic_unchecked_t irq_mis_count;
+ 
+ /* EISA */
+ extern void eisa_set_level_irq(unsigned int irq);
+diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
+index a850b4d..bae26dc 100644
+--- a/arch/x86/include/asm/i387.h
++++ b/arch/x86/include/asm/i387.h
+@@ -92,6 +92,11 @@ static inline int fxrstor_checking(struct i387_fxsave_struct *fx)
+ {
+ 	int err;
+ 
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++	if ((unsigned long)fx < PAX_USER_SHADOW_BASE)
++		fx = (struct i387_fxsave_struct __user *)((void *)fx + PAX_USER_SHADOW_BASE);
++#endif
++
+ 	/* See comment in fxsave() below. */
+ #ifdef CONFIG_AS_FXSAVEQ
+ 	asm volatile("1:  fxrstorq %[fx]\n\t"
+@@ -121,6 +126,11 @@ static inline int fxsave_user(struct i387_fxsave_struct __user *fx)
+ {
+ 	int err;
+ 
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++	if ((unsigned long)fx < PAX_USER_SHADOW_BASE)
++		fx = (struct i387_fxsave_struct __user *)((void __user *)fx + PAX_USER_SHADOW_BASE);
++#endif
++
+ 	/*
+ 	 * Clear the bytes not touched by the fxsave and reserved
+ 	 * for the SW usage.
+@@ -424,7 +434,7 @@ static inline bool interrupted_kernel_fpu_idle(void)
+ static inline bool interrupted_user_mode(void)
+ {
+ 	struct pt_regs *regs = get_irq_regs();
+-	return regs && user_mode_vm(regs);
++	return regs && user_mode(regs);
+ }
+ 
+ /*
+diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h
+index d8e8eef..99f81ae 100644
+--- a/arch/x86/include/asm/io.h
++++ b/arch/x86/include/asm/io.h
+@@ -194,6 +194,17 @@ extern void set_iounmap_nonlazy(void);
+ 
+ #include <linux/vmalloc.h>
+ 
++#define ARCH_HAS_VALID_PHYS_ADDR_RANGE
++static inline int valid_phys_addr_range(unsigned long addr, size_t count)
++{
++	return ((addr + count + PAGE_SIZE - 1) >> PAGE_SHIFT) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0;
++}
++
++static inline int valid_mmap_phys_addr_range(unsigned long pfn, size_t count)
++{
++	return (pfn + (count >> PAGE_SHIFT)) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0;
++}
++
+ /*
+  * Convert a virtual cached pointer to an uncached pointer
+  */
+diff --git a/arch/x86/include/asm/irqflags.h b/arch/x86/include/asm/irqflags.h
+index bba3cf8..06bc8da 100644
+--- a/arch/x86/include/asm/irqflags.h
++++ b/arch/x86/include/asm/irqflags.h
+@@ -141,6 +141,11 @@ static inline notrace unsigned long arch_local_irq_save(void)
+ 	sti;					\
+ 	sysexit
+ 
++#define GET_CR0_INTO_RDI		mov %cr0, %rdi
++#define SET_RDI_INTO_CR0		mov %rdi, %cr0
++#define GET_CR3_INTO_RDI		mov %cr3, %rdi
++#define SET_RDI_INTO_CR3		mov %rdi, %cr3
++
+ #else
+ #define INTERRUPT_RETURN		iret
+ #define ENABLE_INTERRUPTS_SYSEXIT	sti; sysexit
+diff --git a/arch/x86/include/asm/kprobes.h b/arch/x86/include/asm/kprobes.h
+index 5478825..839e88c 100644
+--- a/arch/x86/include/asm/kprobes.h
++++ b/arch/x86/include/asm/kprobes.h
+@@ -37,13 +37,8 @@ typedef u8 kprobe_opcode_t;
+ #define RELATIVEJUMP_SIZE 5
+ #define RELATIVECALL_OPCODE 0xe8
+ #define RELATIVE_ADDR_SIZE 4
+-#define MAX_STACK_SIZE 64
+-#define MIN_STACK_SIZE(ADDR)					       \
+-	(((MAX_STACK_SIZE) < (((unsigned long)current_thread_info()) + \
+-			      THREAD_SIZE - (unsigned long)(ADDR)))    \
+-	 ? (MAX_STACK_SIZE)					       \
+-	 : (((unsigned long)current_thread_info()) +		       \
+-	    THREAD_SIZE - (unsigned long)(ADDR)))
++#define MAX_STACK_SIZE 64UL
++#define MIN_STACK_SIZE(ADDR)	min(MAX_STACK_SIZE, current->thread.sp0 - (unsigned long)(ADDR))
+ 
+ #define flush_insn_slot(p)	do { } while (0)
+ 
+diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
+index b4973f4..a42170a 100644
+--- a/arch/x86/include/asm/kvm_host.h
++++ b/arch/x86/include/asm/kvm_host.h
+@@ -459,7 +459,7 @@ struct kvm_arch {
+ 	unsigned int n_requested_mmu_pages;
+ 	unsigned int n_max_mmu_pages;
+ 	unsigned int indirect_shadow_pages;
+-	atomic_t invlpg_counter;
++	atomic_unchecked_t invlpg_counter;
+ 	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
+ 	/*
+ 	 * Hash table of struct kvm_mmu_page.
+@@ -638,7 +638,7 @@ struct kvm_x86_ops {
+ 	int (*check_intercept)(struct kvm_vcpu *vcpu,
+ 			       struct x86_instruction_info *info,
+ 			       enum x86_intercept_stage stage);
+-};
++} __do_const;
+ 
+ struct kvm_arch_async_pf {
+ 	u32 token;
+@@ -667,9 +667,9 @@ void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);
+ int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3);
+ 
+ int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
+-			  const void *val, int bytes);
++			  const void *val, int bytes) __size_overflow(2);
+ int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes,
+-		  gpa_t addr, unsigned long *ret);
++		  gpa_t addr, unsigned long *ret) __size_overflow(2,3);
+ u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn);
+ 
+ extern bool tdp_enabled;
+@@ -730,7 +730,7 @@ void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
+ int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr);
+ 
+ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata);
+-int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data);
++int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) __size_overflow(3);
+ 
+ unsigned long kvm_get_rflags(struct kvm_vcpu *vcpu);
+ void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);
+@@ -755,7 +755,7 @@ int fx_init(struct kvm_vcpu *vcpu);
+ void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
+ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
+ 		       const u8 *new, int bytes,
+-		       bool guest_initiated);
++		       bool guest_initiated) __size_overflow(2);
+ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva);
+ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
+ int kvm_mmu_load(struct kvm_vcpu *vcpu);
+diff --git a/arch/x86/include/asm/local.h b/arch/x86/include/asm/local.h
+index 9cdae5d..300d20f 100644
+--- a/arch/x86/include/asm/local.h
++++ b/arch/x86/include/asm/local.h
+@@ -18,26 +18,58 @@ typedef struct {
+ 
+ static inline void local_inc(local_t *l)
+ {
+-	asm volatile(_ASM_INC "%0"
++	asm volatile(_ASM_INC "%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_DEC "%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+m" (l->a.counter));
+ }
+ 
+ static inline void local_dec(local_t *l)
+ {
+-	asm volatile(_ASM_DEC "%0"
++	asm volatile(_ASM_DEC "%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_INC "%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+m" (l->a.counter));
+ }
+ 
+ static inline void local_add(long i, local_t *l)
+ {
+-	asm volatile(_ASM_ADD "%1,%0"
++	asm volatile(_ASM_ADD "%1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_SUB "%1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+m" (l->a.counter)
+ 		     : "ir" (i));
+ }
+ 
+ static inline void local_sub(long i, local_t *l)
+ {
+-	asm volatile(_ASM_SUB "%1,%0"
++	asm volatile(_ASM_SUB "%1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_ADD "%1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+m" (l->a.counter)
+ 		     : "ir" (i));
+ }
+@@ -55,7 +87,16 @@ static inline int local_sub_and_test(long i, local_t *l)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(_ASM_SUB "%2,%0; sete %1"
++	asm volatile(_ASM_SUB "%2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_ADD "%2,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "+m" (l->a.counter), "=qm" (c)
+ 		     : "ir" (i) : "memory");
+ 	return c;
+@@ -73,7 +114,16 @@ static inline int local_dec_and_test(local_t *l)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(_ASM_DEC "%0; sete %1"
++	asm volatile(_ASM_DEC "%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_INC "%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "+m" (l->a.counter), "=qm" (c)
+ 		     : : "memory");
+ 	return c != 0;
+@@ -91,7 +141,16 @@ static inline int local_inc_and_test(local_t *l)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(_ASM_INC "%0; sete %1"
++	asm volatile(_ASM_INC "%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_DEC "%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sete %1\n"
+ 		     : "+m" (l->a.counter), "=qm" (c)
+ 		     : : "memory");
+ 	return c != 0;
+@@ -110,7 +169,16 @@ static inline int local_add_negative(long i, local_t *l)
+ {
+ 	unsigned char c;
+ 
+-	asm volatile(_ASM_ADD "%2,%0; sets %1"
++	asm volatile(_ASM_ADD "%2,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_SUB "%2,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
++		     "sets %1\n"
+ 		     : "+m" (l->a.counter), "=qm" (c)
+ 		     : "ir" (i) : "memory");
+ 	return c;
+@@ -133,7 +201,15 @@ static inline long local_add_return(long i, local_t *l)
+ #endif
+ 	/* Modern 486+ processor */
+ 	__i = i;
+-	asm volatile(_ASM_XADD "%0, %1;"
++	asm volatile(_ASM_XADD "%0, %1\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     _ASM_MOV "%0,%1\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+r" (i), "+m" (l->a.counter)
+ 		     : : "memory");
+ 	return i + __i;
+diff --git a/arch/x86/include/asm/mman.h b/arch/x86/include/asm/mman.h
+index 593e51d..fa69c9a 100644
+--- a/arch/x86/include/asm/mman.h
++++ b/arch/x86/include/asm/mman.h
+@@ -5,4 +5,14 @@
+ 
+ #include <asm-generic/mman.h>
+ 
++#ifdef __KERNEL__
++#ifndef __ASSEMBLY__
++#ifdef CONFIG_X86_32
++#define arch_mmap_check	i386_mmap_check
++int i386_mmap_check(unsigned long addr, unsigned long len,
++		unsigned long flags);
++#endif
++#endif
++#endif
++
+ #endif /* _ASM_X86_MMAN_H */
+diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h
+index 5f55e69..e20bfb1 100644
+--- a/arch/x86/include/asm/mmu.h
++++ b/arch/x86/include/asm/mmu.h
+@@ -9,7 +9,7 @@
+  * we put the segment information here.
+  */
+ typedef struct {
+-	void *ldt;
++	struct desc_struct *ldt;
+ 	int size;
+ 
+ #ifdef CONFIG_X86_64
+@@ -18,7 +18,19 @@ typedef struct {
+ #endif
+ 
+ 	struct mutex lock;
+-	void *vdso;
++	unsigned long vdso;
++
++#ifdef CONFIG_X86_32
++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)
++	unsigned long user_cs_base;
++	unsigned long user_cs_limit;
++
++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP)
++	cpumask_t cpu_user_cs_mask;
++#endif
++
++#endif
++#endif
+ } mm_context_t;
+ 
+ #ifdef CONFIG_SMP
+diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
+index 6902152..399f3a2 100644
+--- a/arch/x86/include/asm/mmu_context.h
++++ b/arch/x86/include/asm/mmu_context.h
+@@ -24,6 +24,18 @@ void destroy_context(struct mm_struct *mm);
+ 
+ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+ {
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++	unsigned int i;
++	pgd_t *pgd;
++
++	pax_open_kernel();
++	pgd = get_cpu_pgd(smp_processor_id());
++	for (i = USER_PGD_PTRS; i < 2 * USER_PGD_PTRS; ++i)
++		set_pgd_batched(pgd+i, native_make_pgd(0));
++	pax_close_kernel();
++#endif
++
+ #ifdef CONFIG_SMP
+ 	if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK)
+ 		percpu_write(cpu_tlbstate.state, TLBSTATE_LAZY);
+@@ -34,16 +46,30 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ 			     struct task_struct *tsk)
+ {
+ 	unsigned cpu = smp_processor_id();
++#if defined(CONFIG_X86_32) && defined(CONFIG_SMP) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC))
++	int tlbstate = TLBSTATE_OK;
++#endif
+ 
+ 	if (likely(prev != next)) {
+ #ifdef CONFIG_SMP
++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC))
++		tlbstate = percpu_read(cpu_tlbstate.state);
++#endif
+ 		percpu_write(cpu_tlbstate.state, TLBSTATE_OK);
+ 		percpu_write(cpu_tlbstate.active_mm, next);
+ #endif
+ 		cpumask_set_cpu(cpu, mm_cpumask(next));
+ 
+ 		/* Re-load page tables */
++#ifdef CONFIG_PAX_PER_CPU_PGD
++		pax_open_kernel();
++		__clone_user_pgds(get_cpu_pgd(cpu), next->pgd, USER_PGD_PTRS);
++		__shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd, USER_PGD_PTRS);
++		pax_close_kernel();
++		load_cr3(get_cpu_pgd(cpu));
++#else
+ 		load_cr3(next->pgd);
++#endif
+ 
+ 		/* stop flush ipis for the previous mm */
+ 		cpumask_clear_cpu(cpu, mm_cpumask(prev));
+@@ -53,9 +79,38 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ 		 */
+ 		if (unlikely(prev->context.ldt != next->context.ldt))
+ 			load_LDT_nolock(&next->context);
+-	}
++
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP)
++		if (!(__supported_pte_mask & _PAGE_NX)) {
++			smp_mb__before_clear_bit();
++			cpu_clear(cpu, prev->context.cpu_user_cs_mask);
++			smp_mb__after_clear_bit();
++			cpu_set(cpu, next->context.cpu_user_cs_mask);
++		}
++#endif
++
++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC))
++		if (unlikely(prev->context.user_cs_base != next->context.user_cs_base ||
++			     prev->context.user_cs_limit != next->context.user_cs_limit))
++			set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu);
+ #ifdef CONFIG_SMP
++		else if (unlikely(tlbstate != TLBSTATE_OK))
++			set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu);
++#endif
++#endif
++
++	}
+ 	else {
++
++#ifdef CONFIG_PAX_PER_CPU_PGD
++		pax_open_kernel();
++		__clone_user_pgds(get_cpu_pgd(cpu), next->pgd, USER_PGD_PTRS);
++		__shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd, USER_PGD_PTRS);
++		pax_close_kernel();
++		load_cr3(get_cpu_pgd(cpu));
++#endif
++
++#ifdef CONFIG_SMP
+ 		percpu_write(cpu_tlbstate.state, TLBSTATE_OK);
+ 		BUG_ON(percpu_read(cpu_tlbstate.active_mm) != next);
+ 
+@@ -64,11 +119,28 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ 			 * tlb flush IPI delivery. We must reload CR3
+ 			 * to make sure to use no freed page tables.
+ 			 */
++
++#ifndef CONFIG_PAX_PER_CPU_PGD
+ 			load_cr3(next->pgd);
++#endif
++
+ 			load_LDT_nolock(&next->context);
++
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC)
++			if (!(__supported_pte_mask & _PAGE_NX))
++				cpu_set(cpu, next->context.cpu_user_cs_mask);
++#endif
++
++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC))
++#ifdef CONFIG_PAX_PAGEEXEC
++			if (!((next->pax_flags & MF_PAX_PAGEEXEC) && (__supported_pte_mask & _PAGE_NX)))
++#endif
++				set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu);
++#endif
++
+ 		}
++#endif
+ 	}
+-#endif
+ }
+ 
+ #define activate_mm(prev, next)			\
+diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h
+index 9eae775..c914fea 100644
+--- a/arch/x86/include/asm/module.h
++++ b/arch/x86/include/asm/module.h
+@@ -5,6 +5,7 @@
+ 
+ #ifdef CONFIG_X86_64
+ /* X86_64 does not define MODULE_PROC_FAMILY */
++#define MODULE_PROC_FAMILY ""
+ #elif defined CONFIG_M386
+ #define MODULE_PROC_FAMILY "386 "
+ #elif defined CONFIG_M486
+@@ -59,8 +60,20 @@
+ #error unknown processor family
+ #endif
+ 
+-#ifdef CONFIG_X86_32
+-# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
++#ifdef CONFIG_PAX_KERNEXEC_PLUGIN_METHOD_BTS
++#define MODULE_PAX_KERNEXEC "KERNEXEC_BTS "
++#elif defined(CONFIG_PAX_KERNEXEC_PLUGIN_METHOD_OR)
++#define MODULE_PAX_KERNEXEC "KERNEXEC_OR "
++#else
++#define MODULE_PAX_KERNEXEC ""
+ #endif
+ 
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++#define MODULE_PAX_UDEREF "UDEREF "
++#else
++#define MODULE_PAX_UDEREF ""
++#endif
++
++#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_PAX_KERNEXEC MODULE_PAX_UDEREF
++
+ #endif /* _ASM_X86_MODULE_H */
+diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h
+index 7639dbf..e08a58c 100644
+--- a/arch/x86/include/asm/page_64_types.h
++++ b/arch/x86/include/asm/page_64_types.h
+@@ -56,7 +56,7 @@ void copy_page(void *to, void *from);
+ 
+ /* duplicated to the one in bootmem.h */
+ extern unsigned long max_pfn;
+-extern unsigned long phys_base;
++extern const unsigned long phys_base;
+ 
+ extern unsigned long __phys_addr(unsigned long);
+ #define __phys_reloc_hide(x)	(x)
+diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
+index a7d2db9..edb023e 100644
+--- a/arch/x86/include/asm/paravirt.h
++++ b/arch/x86/include/asm/paravirt.h
+@@ -667,6 +667,18 @@ static inline void set_pgd(pgd_t *pgdp, pgd_t pgd)
+ 			    val);
+ }
+ 
++static inline void set_pgd_batched(pgd_t *pgdp, pgd_t pgd)
++{
++	pgdval_t val = native_pgd_val(pgd);
++
++	if (sizeof(pgdval_t) > sizeof(long))
++		PVOP_VCALL3(pv_mmu_ops.set_pgd_batched, pgdp,
++			    val, (u64)val >> 32);
++	else
++		PVOP_VCALL2(pv_mmu_ops.set_pgd_batched, pgdp,
++			    val);
++}
++
+ static inline void pgd_clear(pgd_t *pgdp)
+ {
+ 	set_pgd(pgdp, __pgd(0));
+@@ -748,6 +760,21 @@ static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx,
+ 	pv_mmu_ops.set_fixmap(idx, phys, flags);
+ }
+ 
++#ifdef CONFIG_PAX_KERNEXEC
++static inline unsigned long pax_open_kernel(void)
++{
++	return PVOP_CALL0(unsigned long, pv_mmu_ops.pax_open_kernel);
++}
++
++static inline unsigned long pax_close_kernel(void)
++{
++	return PVOP_CALL0(unsigned long, pv_mmu_ops.pax_close_kernel);
++}
++#else
++static inline unsigned long pax_open_kernel(void) { return 0; }
++static inline unsigned long pax_close_kernel(void) { return 0; }
++#endif
++
+ #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS)
+ 
+ static inline int arch_spin_is_locked(struct arch_spinlock *lock)
+@@ -964,7 +991,7 @@ extern void default_banner(void);
+ 
+ #define PARA_PATCH(struct, off)        ((PARAVIRT_PATCH_##struct + (off)) / 4)
+ #define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .long, 4)
+-#define PARA_INDIRECT(addr)	*%cs:addr
++#define PARA_INDIRECT(addr)	*%ss:addr
+ #endif
+ 
+ #define INTERRUPT_RETURN						\
+@@ -1041,6 +1068,21 @@ extern void default_banner(void);
+ 	PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_sysexit),	\
+ 		  CLBR_NONE,						\
+ 		  jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_irq_enable_sysexit))
++
++#define GET_CR0_INTO_RDI				\
++	call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0);	\
++	mov %rax,%rdi
++
++#define SET_RDI_INTO_CR0				\
++	call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0)
++
++#define GET_CR3_INTO_RDI				\
++	call PARA_INDIRECT(pv_mmu_ops+PV_MMU_read_cr3);	\
++	mov %rax,%rdi
++
++#define SET_RDI_INTO_CR3				\
++	call PARA_INDIRECT(pv_mmu_ops+PV_MMU_write_cr3)
++
+ #endif	/* CONFIG_X86_32 */
+ 
+ #endif /* __ASSEMBLY__ */
+diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h
+index 8e8b9a4..f07d725 100644
+--- a/arch/x86/include/asm/paravirt_types.h
++++ b/arch/x86/include/asm/paravirt_types.h
+@@ -84,20 +84,20 @@ struct pv_init_ops {
+ 	 */
+ 	unsigned (*patch)(u8 type, u16 clobber, void *insnbuf,
+ 			  unsigned long addr, unsigned len);
+-};
++} __no_const;
+ 
+ 
+ struct pv_lazy_ops {
+ 	/* Set deferred update mode, used for batching operations. */
+ 	void (*enter)(void);
+ 	void (*leave)(void);
+-};
++} __no_const;
+ 
+ struct pv_time_ops {
+ 	unsigned long long (*sched_clock)(void);
+ 	unsigned long long (*steal_clock)(int cpu);
+ 	unsigned long (*get_tsc_khz)(void);
+-};
++} __no_const;
+ 
+ struct pv_cpu_ops {
+ 	/* hooks for various privileged instructions */
+@@ -193,7 +193,7 @@ struct pv_cpu_ops {
+ 
+ 	void (*start_context_switch)(struct task_struct *prev);
+ 	void (*end_context_switch)(struct task_struct *next);
+-};
++} __no_const;
+ 
+ struct pv_irq_ops {
+ 	/*
+@@ -224,7 +224,7 @@ struct pv_apic_ops {
+ 				 unsigned long start_eip,
+ 				 unsigned long start_esp);
+ #endif
+-};
++} __no_const;
+ 
+ struct pv_mmu_ops {
+ 	unsigned long (*read_cr2)(void);
+@@ -313,6 +313,7 @@ struct pv_mmu_ops {
+ 	struct paravirt_callee_save make_pud;
+ 
+ 	void (*set_pgd)(pgd_t *pudp, pgd_t pgdval);
++	void (*set_pgd_batched)(pgd_t *pudp, pgd_t pgdval);
+ #endif	/* PAGETABLE_LEVELS == 4 */
+ #endif	/* PAGETABLE_LEVELS >= 3 */
+ 
+@@ -324,6 +325,12 @@ struct pv_mmu_ops {
+ 	   an mfn.  We can tell which is which from the index. */
+ 	void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx,
+ 			   phys_addr_t phys, pgprot_t flags);
++
++#ifdef CONFIG_PAX_KERNEXEC
++	unsigned long (*pax_open_kernel)(void);
++	unsigned long (*pax_close_kernel)(void);
++#endif
++
+ };
+ 
+ struct arch_spinlock;
+@@ -334,7 +341,7 @@ struct pv_lock_ops {
+ 	void (*spin_lock_flags)(struct arch_spinlock *lock, unsigned long flags);
+ 	int (*spin_trylock)(struct arch_spinlock *lock);
+ 	void (*spin_unlock)(struct arch_spinlock *lock);
+-};
++} __no_const;
+ 
+ /* This contains all the paravirt structures: we get a convenient
+  * number for each function using the offset which we use to indicate
+diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h
+index b4389a4..b7ff22c 100644
+--- a/arch/x86/include/asm/pgalloc.h
++++ b/arch/x86/include/asm/pgalloc.h
+@@ -63,6 +63,13 @@ static inline void pmd_populate_kernel(struct mm_struct *mm,
+ 				       pmd_t *pmd, pte_t *pte)
+ {
+ 	paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT);
++	set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE));
++}
++
++static inline void pmd_populate_user(struct mm_struct *mm,
++				       pmd_t *pmd, pte_t *pte)
++{
++	paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT);
+ 	set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE));
+ }
+ 
+diff --git a/arch/x86/include/asm/pgtable-2level.h b/arch/x86/include/asm/pgtable-2level.h
+index 98391db..8f6984e 100644
+--- a/arch/x86/include/asm/pgtable-2level.h
++++ b/arch/x86/include/asm/pgtable-2level.h
+@@ -18,7 +18,9 @@ static inline void native_set_pte(pte_t *ptep , pte_t pte)
+ 
+ static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
+ {
++	pax_open_kernel();
+ 	*pmdp = pmd;
++	pax_close_kernel();
+ }
+ 
+ static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
+diff --git a/arch/x86/include/asm/pgtable-3level.h b/arch/x86/include/asm/pgtable-3level.h
+index effff47..f9e4035 100644
+--- a/arch/x86/include/asm/pgtable-3level.h
++++ b/arch/x86/include/asm/pgtable-3level.h
+@@ -38,12 +38,16 @@ static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
+ 
+ static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
+ {
++	pax_open_kernel();
+ 	set_64bit((unsigned long long *)(pmdp), native_pmd_val(pmd));
++	pax_close_kernel();
+ }
+ 
+ static inline void native_set_pud(pud_t *pudp, pud_t pud)
+ {
++	pax_open_kernel();
+ 	set_64bit((unsigned long long *)(pudp), native_pud_val(pud));
++	pax_close_kernel();
+ }
+ 
+ /*
+diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
+index 18601c8..3d716d1 100644
+--- a/arch/x86/include/asm/pgtable.h
++++ b/arch/x86/include/asm/pgtable.h
+@@ -44,6 +44,7 @@ extern struct mm_struct *pgd_page_get_mm(struct page *page);
+ 
+ #ifndef __PAGETABLE_PUD_FOLDED
+ #define set_pgd(pgdp, pgd)		native_set_pgd(pgdp, pgd)
++#define set_pgd_batched(pgdp, pgd)	native_set_pgd_batched(pgdp, pgd)
+ #define pgd_clear(pgd)			native_pgd_clear(pgd)
+ #endif
+ 
+@@ -81,12 +82,51 @@ extern struct mm_struct *pgd_page_get_mm(struct page *page);
+ 
+ #define arch_end_context_switch(prev)	do {} while(0)
+ 
++#define pax_open_kernel()	native_pax_open_kernel()
++#define pax_close_kernel()	native_pax_close_kernel()
+ #endif	/* CONFIG_PARAVIRT */
+ 
++#define  __HAVE_ARCH_PAX_OPEN_KERNEL
++#define  __HAVE_ARCH_PAX_CLOSE_KERNEL
++
++#ifdef CONFIG_PAX_KERNEXEC
++static inline unsigned long native_pax_open_kernel(void)
++{
++	unsigned long cr0;
++
++	preempt_disable();
++	barrier();
++	cr0 = read_cr0() ^ X86_CR0_WP;
++	BUG_ON(unlikely(cr0 & X86_CR0_WP));
++	write_cr0(cr0);
++	return cr0 ^ X86_CR0_WP;
++}
++
++static inline unsigned long native_pax_close_kernel(void)
++{
++	unsigned long cr0;
++
++	cr0 = read_cr0() ^ X86_CR0_WP;
++	BUG_ON(unlikely(!(cr0 & X86_CR0_WP)));
++	write_cr0(cr0);
++	barrier();
++	preempt_enable_no_resched();
++	return cr0 ^ X86_CR0_WP;
++}
++#else
++static inline unsigned long native_pax_open_kernel(void) { return 0; }
++static inline unsigned long native_pax_close_kernel(void) { return 0; }
++#endif
++
+ /*
+  * The following only work if pte_present() is true.
+  * Undefined behaviour if not..
+  */
++static inline int pte_user(pte_t pte)
++{
++	return pte_val(pte) & _PAGE_USER;
++}
++
+ static inline int pte_dirty(pte_t pte)
+ {
+ 	return pte_flags(pte) & _PAGE_DIRTY;
+@@ -196,9 +236,29 @@ static inline pte_t pte_wrprotect(pte_t pte)
+ 	return pte_clear_flags(pte, _PAGE_RW);
+ }
+ 
++static inline pte_t pte_mkread(pte_t pte)
++{
++	return __pte(pte_val(pte) | _PAGE_USER);
++}
++
+ static inline pte_t pte_mkexec(pte_t pte)
+ {
+-	return pte_clear_flags(pte, _PAGE_NX);
++#ifdef CONFIG_X86_PAE
++	if (__supported_pte_mask & _PAGE_NX)
++		return pte_clear_flags(pte, _PAGE_NX);
++	else
++#endif
++		return pte_set_flags(pte, _PAGE_USER);
++}
++
++static inline pte_t pte_exprotect(pte_t pte)
++{
++#ifdef CONFIG_X86_PAE
++	if (__supported_pte_mask & _PAGE_NX)
++		return pte_set_flags(pte, _PAGE_NX);
++	else
++#endif
++		return pte_clear_flags(pte, _PAGE_USER);
+ }
+ 
+ static inline pte_t pte_mkdirty(pte_t pte)
+@@ -390,6 +450,15 @@ pte_t *populate_extra_pte(unsigned long vaddr);
+ #endif
+ 
+ #ifndef __ASSEMBLY__
++
++#ifdef CONFIG_PAX_PER_CPU_PGD
++extern pgd_t cpu_pgd[NR_CPUS][PTRS_PER_PGD];
++static inline pgd_t *get_cpu_pgd(unsigned int cpu)
++{
++	return cpu_pgd[cpu];
++}
++#endif
++
+ #include <linux/mm_types.h>
+ 
+ static inline int pte_none(pte_t pte)
+@@ -560,7 +629,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
+ 
+ static inline int pgd_bad(pgd_t pgd)
+ {
+-	return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE;
++	return (pgd_flags(pgd) & ~(_PAGE_USER | _PAGE_NX)) != _KERNPG_TABLE;
+ }
+ 
+ static inline int pgd_none(pgd_t pgd)
+@@ -583,7 +652,12 @@ static inline int pgd_none(pgd_t pgd)
+  * pgd_offset() returns a (pgd_t *)
+  * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
+  */
+-#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address)))
++#define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address))
++
++#ifdef CONFIG_PAX_PER_CPU_PGD
++#define pgd_offset_cpu(cpu, address) (get_cpu_pgd(cpu) + pgd_index(address))
++#endif
++
+ /*
+  * a shortcut which implies the use of the kernel's pgd, instead
+  * of a process's
+@@ -594,6 +668,20 @@ static inline int pgd_none(pgd_t pgd)
+ #define KERNEL_PGD_BOUNDARY	pgd_index(PAGE_OFFSET)
+ #define KERNEL_PGD_PTRS		(PTRS_PER_PGD - KERNEL_PGD_BOUNDARY)
+ 
++#ifdef CONFIG_X86_32
++#define USER_PGD_PTRS		KERNEL_PGD_BOUNDARY
++#else
++#define TASK_SIZE_MAX_SHIFT CONFIG_TASK_SIZE_MAX_SHIFT
++#define USER_PGD_PTRS		(_AC(1,UL) << (TASK_SIZE_MAX_SHIFT - PGDIR_SHIFT))
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++#define PAX_USER_SHADOW_BASE	(_AC(1,UL) << TASK_SIZE_MAX_SHIFT)
++#else
++#define PAX_USER_SHADOW_BASE	(_AC(0,UL))
++#endif
++
++#endif
++
+ #ifndef __ASSEMBLY__
+ 
+ extern int direct_gbpages;
+@@ -758,11 +846,23 @@ static inline void pmdp_set_wrprotect(struct mm_struct *mm,
+  * dst and src can be on the same page, but the range must not overlap,
+  * and must not cross a page boundary.
+  */
+-static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
++static inline void clone_pgd_range(pgd_t *dst, const pgd_t *src, int count)
+ {
+-       memcpy(dst, src, count * sizeof(pgd_t));
++	pax_open_kernel();
++	while (count--)
++		*dst++ = *src++;
++	pax_close_kernel();
+ }
+ 
++#ifdef CONFIG_PAX_PER_CPU_PGD
++extern void __clone_user_pgds(pgd_t *dst, const pgd_t *src, int count);
++#endif
++
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++extern void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count);
++#else
++static inline void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count) {}
++#endif
+ 
+ #include <asm-generic/pgtable.h>
+ #endif	/* __ASSEMBLY__ */
+diff --git a/arch/x86/include/asm/pgtable_32.h b/arch/x86/include/asm/pgtable_32.h
+index 0c92113..34a77c6 100644
+--- a/arch/x86/include/asm/pgtable_32.h
++++ b/arch/x86/include/asm/pgtable_32.h
+@@ -25,9 +25,6 @@
+ struct mm_struct;
+ struct vm_area_struct;
+ 
+-extern pgd_t swapper_pg_dir[1024];
+-extern pgd_t initial_page_table[1024];
+-
+ static inline void pgtable_cache_init(void) { }
+ static inline void check_pgt_cache(void) { }
+ void paging_init(void);
+@@ -48,6 +45,12 @@ extern void set_pmd_pfn(unsigned long, unsigned long, pgprot_t);
+ # include <asm/pgtable-2level.h>
+ #endif
+ 
++extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
++extern pgd_t initial_page_table[PTRS_PER_PGD];
++#ifdef CONFIG_X86_PAE
++extern pmd_t swapper_pm_dir[PTRS_PER_PGD][PTRS_PER_PMD];
++#endif
++
+ #if defined(CONFIG_HIGHPTE)
+ #define pte_offset_map(dir, address)					\
+ 	((pte_t *)kmap_atomic(pmd_page(*(dir))) +		\
+@@ -62,7 +65,9 @@ extern void set_pmd_pfn(unsigned long, unsigned long, pgprot_t);
+ /* Clear a kernel PTE and flush it from the TLB */
+ #define kpte_clear_flush(ptep, vaddr)		\
+ do {						\
++	pax_open_kernel();			\
+ 	pte_clear(&init_mm, (vaddr), (ptep));	\
++	pax_close_kernel();			\
+ 	__flush_tlb_one((vaddr));		\
+ } while (0)
+ 
+@@ -74,6 +79,9 @@ do {						\
+ 
+ #endif /* !__ASSEMBLY__ */
+ 
++#define HAVE_ARCH_UNMAPPED_AREA
++#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN
++
+ /*
+  * kern_addr_valid() is (1) for FLATMEM and (0) for
+  * SPARSEMEM and DISCONTIGMEM
+diff --git a/arch/x86/include/asm/pgtable_32_types.h b/arch/x86/include/asm/pgtable_32_types.h
+index ed5903b..c7fe163 100644
+--- a/arch/x86/include/asm/pgtable_32_types.h
++++ b/arch/x86/include/asm/pgtable_32_types.h
+@@ -8,7 +8,7 @@
+  */
+ #ifdef CONFIG_X86_PAE
+ # include <asm/pgtable-3level_types.h>
+-# define PMD_SIZE	(1UL << PMD_SHIFT)
++# define PMD_SIZE	(_AC(1, UL) << PMD_SHIFT)
+ # define PMD_MASK	(~(PMD_SIZE - 1))
+ #else
+ # include <asm/pgtable-2level_types.h>
+@@ -46,6 +46,19 @@ extern bool __vmalloc_start_set; /* set once high_memory is set */
+ # define VMALLOC_END	(FIXADDR_START - 2 * PAGE_SIZE)
+ #endif
+ 
++#ifdef CONFIG_PAX_KERNEXEC
++#ifndef __ASSEMBLY__
++extern unsigned char MODULES_EXEC_VADDR[];
++extern unsigned char MODULES_EXEC_END[];
++#endif
++#include <asm/boot.h>
++#define ktla_ktva(addr)		(addr + LOAD_PHYSICAL_ADDR + PAGE_OFFSET)
++#define ktva_ktla(addr)		(addr - LOAD_PHYSICAL_ADDR - PAGE_OFFSET)
++#else
++#define ktla_ktva(addr)		(addr)
++#define ktva_ktla(addr)		(addr)
++#endif
++
+ #define MODULES_VADDR	VMALLOC_START
+ #define MODULES_END	VMALLOC_END
+ #define MODULES_LEN	(MODULES_VADDR - MODULES_END)
+diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
+index 975f709..107976d 100644
+--- a/arch/x86/include/asm/pgtable_64.h
++++ b/arch/x86/include/asm/pgtable_64.h
+@@ -16,10 +16,14 @@
+ 
+ extern pud_t level3_kernel_pgt[512];
+ extern pud_t level3_ident_pgt[512];
++extern pud_t level3_vmalloc_start_pgt[512];
++extern pud_t level3_vmalloc_end_pgt[512];
++extern pud_t level3_vmemmap_pgt[512];
++extern pud_t level2_vmemmap_pgt[512];
+ extern pmd_t level2_kernel_pgt[512];
+ extern pmd_t level2_fixmap_pgt[512];
+-extern pmd_t level2_ident_pgt[512];
+-extern pgd_t init_level4_pgt[];
++extern pmd_t level2_ident_pgt[512*2];
++extern pgd_t init_level4_pgt[512];
+ 
+ #define swapper_pg_dir init_level4_pgt
+ 
+@@ -61,7 +65,9 @@ static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
+ 
+ static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
+ {
++	pax_open_kernel();
+ 	*pmdp = pmd;
++	pax_close_kernel();
+ }
+ 
+ static inline void native_pmd_clear(pmd_t *pmd)
+@@ -107,6 +113,13 @@ static inline void native_pud_clear(pud_t *pud)
+ 
+ static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd)
+ {
++	pax_open_kernel();
++	*pgdp = pgd;
++	pax_close_kernel();
++}
++
++static inline void native_set_pgd_batched(pgd_t *pgdp, pgd_t pgd)
++{
+ 	*pgdp = pgd;
+ }
+ 
+diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h
+index 766ea16..5b96cb3 100644
+--- a/arch/x86/include/asm/pgtable_64_types.h
++++ b/arch/x86/include/asm/pgtable_64_types.h
+@@ -59,5 +59,10 @@ typedef struct { pteval_t pte; } pte_t;
+ #define MODULES_VADDR    _AC(0xffffffffa0000000, UL)
+ #define MODULES_END      _AC(0xffffffffff000000, UL)
+ #define MODULES_LEN   (MODULES_END - MODULES_VADDR)
++#define MODULES_EXEC_VADDR MODULES_VADDR
++#define MODULES_EXEC_END MODULES_END
++
++#define ktla_ktva(addr)		(addr)
++#define ktva_ktla(addr)		(addr)
+ 
+ #endif /* _ASM_X86_PGTABLE_64_DEFS_H */
+diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
+index 013286a..8b42f4f 100644
+--- a/arch/x86/include/asm/pgtable_types.h
++++ b/arch/x86/include/asm/pgtable_types.h
+@@ -16,13 +16,12 @@
+ #define _PAGE_BIT_PSE		7	/* 4 MB (or 2MB) page */
+ #define _PAGE_BIT_PAT		7	/* on 4KB pages */
+ #define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
+-#define _PAGE_BIT_UNUSED1	9	/* available for programmer */
++#define _PAGE_BIT_SPECIAL	9	/* special mappings, no associated struct page */
+ #define _PAGE_BIT_IOMAP		10	/* flag used to indicate IO mapping */
+ #define _PAGE_BIT_HIDDEN	11	/* hidden by kmemcheck */
+ #define _PAGE_BIT_PAT_LARGE	12	/* On 2MB or 1GB pages */
+-#define _PAGE_BIT_SPECIAL	_PAGE_BIT_UNUSED1
+-#define _PAGE_BIT_CPA_TEST	_PAGE_BIT_UNUSED1
+-#define _PAGE_BIT_SPLITTING	_PAGE_BIT_UNUSED1 /* only valid on a PSE pmd */
++#define _PAGE_BIT_CPA_TEST	_PAGE_BIT_SPECIAL
++#define _PAGE_BIT_SPLITTING	_PAGE_BIT_SPECIAL /* only valid on a PSE pmd */
+ #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
+ 
+ /* If _PAGE_BIT_PRESENT is clear, we use these: */
+@@ -40,7 +39,6 @@
+ #define _PAGE_DIRTY	(_AT(pteval_t, 1) << _PAGE_BIT_DIRTY)
+ #define _PAGE_PSE	(_AT(pteval_t, 1) << _PAGE_BIT_PSE)
+ #define _PAGE_GLOBAL	(_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL)
+-#define _PAGE_UNUSED1	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1)
+ #define _PAGE_IOMAP	(_AT(pteval_t, 1) << _PAGE_BIT_IOMAP)
+ #define _PAGE_PAT	(_AT(pteval_t, 1) << _PAGE_BIT_PAT)
+ #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE)
+@@ -57,8 +55,10 @@
+ 
+ #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
+ #define _PAGE_NX	(_AT(pteval_t, 1) << _PAGE_BIT_NX)
+-#else
++#elif defined(CONFIG_KMEMCHECK)
+ #define _PAGE_NX	(_AT(pteval_t, 0))
++#else
++#define _PAGE_NX	(_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN)
+ #endif
+ 
+ #define _PAGE_FILE	(_AT(pteval_t, 1) << _PAGE_BIT_FILE)
+@@ -96,6 +96,9 @@
+ #define PAGE_READONLY_EXEC	__pgprot(_PAGE_PRESENT | _PAGE_USER |	\
+ 					 _PAGE_ACCESSED)
+ 
++#define PAGE_READONLY_NOEXEC PAGE_READONLY
++#define PAGE_SHARED_NOEXEC PAGE_SHARED
++
+ #define __PAGE_KERNEL_EXEC						\
+ 	(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_GLOBAL)
+ #define __PAGE_KERNEL		(__PAGE_KERNEL_EXEC | _PAGE_NX)
+@@ -106,7 +109,7 @@
+ #define __PAGE_KERNEL_WC		(__PAGE_KERNEL | _PAGE_CACHE_WC)
+ #define __PAGE_KERNEL_NOCACHE		(__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT)
+ #define __PAGE_KERNEL_UC_MINUS		(__PAGE_KERNEL | _PAGE_PCD)
+-#define __PAGE_KERNEL_VSYSCALL		(__PAGE_KERNEL_RX | _PAGE_USER)
++#define __PAGE_KERNEL_VSYSCALL		(__PAGE_KERNEL_RO | _PAGE_USER)
+ #define __PAGE_KERNEL_VVAR		(__PAGE_KERNEL_RO | _PAGE_USER)
+ #define __PAGE_KERNEL_VVAR_NOCACHE	(__PAGE_KERNEL_VVAR | _PAGE_PCD | _PAGE_PWT)
+ #define __PAGE_KERNEL_LARGE		(__PAGE_KERNEL | _PAGE_PSE)
+@@ -168,8 +171,8 @@
+  * bits are combined, this will alow user to access the high address mapped
+  * VDSO in the presence of CONFIG_COMPAT_VDSO
+  */
+-#define PTE_IDENT_ATTR	 0x003		/* PRESENT+RW */
+-#define PDE_IDENT_ATTR	 0x067		/* PRESENT+RW+USER+DIRTY+ACCESSED */
++#define PTE_IDENT_ATTR	 0x063		/* PRESENT+RW+DIRTY+ACCESSED */
++#define PDE_IDENT_ATTR	 0x063		/* PRESENT+RW+DIRTY+ACCESSED */
+ #define PGD_IDENT_ATTR	 0x001		/* PRESENT (no other attributes) */
+ #endif
+ 
+@@ -207,7 +210,17 @@ static inline pgdval_t pgd_flags(pgd_t pgd)
+ {
+ 	return native_pgd_val(pgd) & PTE_FLAGS_MASK;
+ }
++#endif
+ 
++#if PAGETABLE_LEVELS == 3
++#include <asm-generic/pgtable-nopud.h>
++#endif
++
++#if PAGETABLE_LEVELS == 2
++#include <asm-generic/pgtable-nopmd.h>
++#endif
++
++#ifndef __ASSEMBLY__
+ #if PAGETABLE_LEVELS > 3
+ typedef struct { pudval_t pud; } pud_t;
+ 
+@@ -221,8 +234,6 @@ static inline pudval_t native_pud_val(pud_t pud)
+ 	return pud.pud;
+ }
+ #else
+-#include <asm-generic/pgtable-nopud.h>
+-
+ static inline pudval_t native_pud_val(pud_t pud)
+ {
+ 	return native_pgd_val(pud.pgd);
+@@ -242,8 +253,6 @@ static inline pmdval_t native_pmd_val(pmd_t pmd)
+ 	return pmd.pmd;
+ }
+ #else
+-#include <asm-generic/pgtable-nopmd.h>
+-
+ static inline pmdval_t native_pmd_val(pmd_t pmd)
+ {
+ 	return native_pgd_val(pmd.pud.pgd);
+@@ -283,7 +292,6 @@ typedef struct page *pgtable_t;
+ 
+ extern pteval_t __supported_pte_mask;
+ extern void set_nx(void);
+-extern int nx_enabled;
+ 
+ #define pgprot_writecombine	pgprot_writecombine
+ extern pgprot_t pgprot_writecombine(pgprot_t prot);
+diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
+index bb3ee36..781a6b8 100644
+--- a/arch/x86/include/asm/processor.h
++++ b/arch/x86/include/asm/processor.h
+@@ -268,7 +268,7 @@ struct tss_struct {
+ 
+ } ____cacheline_aligned;
+ 
+-DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss);
++extern struct tss_struct init_tss[NR_CPUS];
+ 
+ /*
+  * Save the original ist values for checking stack pointers during debugging
+@@ -861,11 +861,18 @@ static inline void spin_lock_prefetch(const void *x)
+  */
+ #define TASK_SIZE		PAGE_OFFSET
+ #define TASK_SIZE_MAX		TASK_SIZE
++
++#ifdef CONFIG_PAX_SEGMEXEC
++#define SEGMEXEC_TASK_SIZE	(TASK_SIZE / 2)
++#define STACK_TOP		((current->mm->pax_flags & MF_PAX_SEGMEXEC)?SEGMEXEC_TASK_SIZE:TASK_SIZE)
++#else
+ #define STACK_TOP		TASK_SIZE
+-#define STACK_TOP_MAX		STACK_TOP
++#endif
++
++#define STACK_TOP_MAX		TASK_SIZE
+ 
+ #define INIT_THREAD  {							  \
+-	.sp0			= sizeof(init_stack) + (long)&init_stack, \
++	.sp0			= sizeof(init_stack) + (long)&init_stack - 8, \
+ 	.vm86_info		= NULL,					  \
+ 	.sysenter_cs		= __KERNEL_CS,				  \
+ 	.io_bitmap_ptr		= NULL,					  \
+@@ -879,7 +886,7 @@ static inline void spin_lock_prefetch(const void *x)
+  */
+ #define INIT_TSS  {							  \
+ 	.x86_tss = {							  \
+-		.sp0		= sizeof(init_stack) + (long)&init_stack, \
++		.sp0		= sizeof(init_stack) + (long)&init_stack - 8, \
+ 		.ss0		= __KERNEL_DS,				  \
+ 		.ss1		= __KERNEL_CS,				  \
+ 		.io_bitmap_base	= INVALID_IO_BITMAP_OFFSET,		  \
+@@ -890,11 +897,7 @@ static inline void spin_lock_prefetch(const void *x)
+ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+ 
+ #define THREAD_SIZE_LONGS      (THREAD_SIZE/sizeof(unsigned long))
+-#define KSTK_TOP(info)                                                 \
+-({                                                                     \
+-       unsigned long *__ptr = (unsigned long *)(info);                 \
+-       (unsigned long)(&__ptr[THREAD_SIZE_LONGS]);                     \
+-})
++#define KSTK_TOP(info)         ((container_of(info, struct task_struct, tinfo))->thread.sp0)
+ 
+ /*
+  * The below -8 is to reserve 8 bytes on top of the ring0 stack.
+@@ -909,7 +912,7 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+ #define task_pt_regs(task)                                             \
+ ({                                                                     \
+        struct pt_regs *__regs__;                                       \
+-       __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \
++       __regs__ = (struct pt_regs *)((task)->thread.sp0);              \
+        __regs__ - 1;                                                   \
+ })
+ 
+@@ -919,13 +922,13 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+ /*
+  * User space process size. 47bits minus one guard page.
+  */
+-#define TASK_SIZE_MAX	((1UL << 47) - PAGE_SIZE)
++#define TASK_SIZE_MAX	((1UL << TASK_SIZE_MAX_SHIFT) - PAGE_SIZE)
+ 
+ /* This decides where the kernel will search for a free chunk of vm
+  * space during mmap's.
+  */
+ #define IA32_PAGE_OFFSET	((current->personality & ADDR_LIMIT_3GB) ? \
+-					0xc0000000 : 0xFFFFe000)
++					0xc0000000 : 0xFFFFf000)
+ 
+ #define TASK_SIZE		(test_thread_flag(TIF_IA32) ? \
+ 					IA32_PAGE_OFFSET : TASK_SIZE_MAX)
+@@ -936,11 +939,11 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
+ #define STACK_TOP_MAX		TASK_SIZE_MAX
+ 
+ #define INIT_THREAD  { \
+-	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
++	.sp0 = (unsigned long)&init_stack + sizeof(init_stack) - 16 \
+ }
+ 
+ #define INIT_TSS  { \
+-	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \
++	.x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) - 16 \
+ }
+ 
+ /*
+@@ -962,6 +965,10 @@ extern void start_thread(struct pt_regs *regs, unsigned long new_ip,
+  */
+ #define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
+ 
++#ifdef CONFIG_PAX_SEGMEXEC
++#define SEGMEXEC_TASK_UNMAPPED_BASE	(PAGE_ALIGN(SEGMEXEC_TASK_SIZE / 3))
++#endif
++
+ #define KSTK_EIP(task)		(task_pt_regs(task)->ip)
+ 
+ /* Get/set a process' ability to use the timestamp counter instruction */
+diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h
+index 3566454..4bdfb8c 100644
+--- a/arch/x86/include/asm/ptrace.h
++++ b/arch/x86/include/asm/ptrace.h
+@@ -156,28 +156,29 @@ static inline unsigned long regs_return_value(struct pt_regs *regs)
+ }
+ 
+ /*
+- * user_mode_vm(regs) determines whether a register set came from user mode.
++ * user_mode(regs) determines whether a register set came from user mode.
+  * This is true if V8086 mode was enabled OR if the register set was from
+  * protected mode with RPL-3 CS value.  This tricky test checks that with
+  * one comparison.  Many places in the kernel can bypass this full check
+- * if they have already ruled out V8086 mode, so user_mode(regs) can be used.
++ * if they have already ruled out V8086 mode, so user_mode_novm(regs) can
++ * be used.
+  */
+-static inline int user_mode(struct pt_regs *regs)
++static inline int user_mode_novm(struct pt_regs *regs)
+ {
+ #ifdef CONFIG_X86_32
+ 	return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL;
+ #else
+-	return !!(regs->cs & 3);
++	return !!(regs->cs & SEGMENT_RPL_MASK);
+ #endif
+ }
+ 
+-static inline int user_mode_vm(struct pt_regs *regs)
++static inline int user_mode(struct pt_regs *regs)
+ {
+ #ifdef CONFIG_X86_32
+ 	return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >=
+ 		USER_RPL;
+ #else
+-	return user_mode(regs);
++	return user_mode_novm(regs);
+ #endif
+ }
+ 
+@@ -193,15 +194,16 @@ static inline int v8086_mode(struct pt_regs *regs)
+ #ifdef CONFIG_X86_64
+ static inline bool user_64bit_mode(struct pt_regs *regs)
+ {
++	unsigned long cs = regs->cs & 0xffff;
+ #ifndef CONFIG_PARAVIRT
+ 	/*
+ 	 * On non-paravirt systems, this is the only long mode CPL 3
+ 	 * selector.  We do not allow long mode selectors in the LDT.
+ 	 */
+-	return regs->cs == __USER_CS;
++	return cs == __USER_CS;
+ #else
+ 	/* Headers are too twisted for this to go in paravirt.h. */
+-	return regs->cs == __USER_CS || regs->cs == pv_info.extra_user_64bit_cs;
++	return cs == __USER_CS || cs == pv_info.extra_user_64bit_cs;
+ #endif
+ }
+ #endif
+diff --git a/arch/x86/include/asm/reboot.h b/arch/x86/include/asm/reboot.h
+index 92f29706..a79cbbb 100644
+--- a/arch/x86/include/asm/reboot.h
++++ b/arch/x86/include/asm/reboot.h
+@@ -6,19 +6,19 @@
+ struct pt_regs;
+ 
+ struct machine_ops {
+-	void (*restart)(char *cmd);
+-	void (*halt)(void);
+-	void (*power_off)(void);
++	void (* __noreturn restart)(char *cmd);
++	void (* __noreturn halt)(void);
++	void (* __noreturn power_off)(void);
+ 	void (*shutdown)(void);
+ 	void (*crash_shutdown)(struct pt_regs *);
+-	void (*emergency_restart)(void);
+-};
++	void (* __noreturn emergency_restart)(void);
++} __no_const;
+ 
+ extern struct machine_ops machine_ops;
+ 
+ void native_machine_crash_shutdown(struct pt_regs *regs);
+ void native_machine_shutdown(void);
+-void machine_real_restart(unsigned int type);
++void machine_real_restart(unsigned int type) __noreturn;
+ /* These must match dispatch_table in reboot_32.S */
+ #define MRR_BIOS	0
+ #define MRR_APM		1
+diff --git a/arch/x86/include/asm/rwsem.h b/arch/x86/include/asm/rwsem.h
+index 2dbe4a7..ce1db00 100644
+--- a/arch/x86/include/asm/rwsem.h
++++ b/arch/x86/include/asm/rwsem.h
+@@ -64,6 +64,14 @@ static inline void __down_read(struct rw_semaphore *sem)
+ {
+ 	asm volatile("# beginning down_read\n\t"
+ 		     LOCK_PREFIX _ASM_INC "(%1)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX _ASM_DEC "(%1)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     /* adds 0x00000001 */
+ 		     "  jns        1f\n"
+ 		     "  call call_rwsem_down_read_failed\n"
+@@ -85,6 +93,14 @@ static inline int __down_read_trylock(struct rw_semaphore *sem)
+ 		     "1:\n\t"
+ 		     "  mov          %1,%2\n\t"
+ 		     "  add          %3,%2\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     "sub %3,%2\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     "  jle	     2f\n\t"
+ 		     LOCK_PREFIX "  cmpxchg  %2,%0\n\t"
+ 		     "  jnz	     1b\n\t"
+@@ -104,6 +120,14 @@ static inline void __down_write_nested(struct rw_semaphore *sem, int subclass)
+ 	long tmp;
+ 	asm volatile("# beginning down_write\n\t"
+ 		     LOCK_PREFIX "  xadd      %1,(%2)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     "mov %1,(%2)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     /* adds 0xffff0001, returns the old value */
+ 		     "  test      %1,%1\n\t"
+ 		     /* was the count 0 before? */
+@@ -141,6 +165,14 @@ static inline void __up_read(struct rw_semaphore *sem)
+ 	long tmp;
+ 	asm volatile("# beginning __up_read\n\t"
+ 		     LOCK_PREFIX "  xadd      %1,(%2)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     "mov %1,(%2)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     /* subtracts 1, returns the old value */
+ 		     "  jns        1f\n\t"
+ 		     "  call call_rwsem_wake\n" /* expects old value in %edx */
+@@ -159,6 +191,14 @@ static inline void __up_write(struct rw_semaphore *sem)
+ 	long tmp;
+ 	asm volatile("# beginning __up_write\n\t"
+ 		     LOCK_PREFIX "  xadd      %1,(%2)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     "mov %1,(%2)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     /* subtracts 0xffff0001, returns the old value */
+ 		     "  jns        1f\n\t"
+ 		     "  call call_rwsem_wake\n" /* expects old value in %edx */
+@@ -176,6 +216,14 @@ static inline void __downgrade_write(struct rw_semaphore *sem)
+ {
+ 	asm volatile("# beginning __downgrade_write\n\t"
+ 		     LOCK_PREFIX _ASM_ADD "%2,(%1)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX _ASM_SUB "%2,(%1)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     /*
+ 		      * transitions 0xZZZZ0001 -> 0xYYYY0001 (i386)
+ 		      *     0xZZZZZZZZ00000001 -> 0xYYYYYYYY00000001 (x86_64)
+@@ -194,7 +242,15 @@ static inline void __downgrade_write(struct rw_semaphore *sem)
+  */
+ static inline void rwsem_atomic_add(long delta, struct rw_semaphore *sem)
+ {
+-	asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0"
++	asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX _ASM_SUB "%1,%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+m" (sem->count)
+ 		     : "er" (delta));
+ }
+@@ -204,7 +260,7 @@ static inline void rwsem_atomic_add(long delta, struct rw_semaphore *sem)
+  */
+ static inline long rwsem_atomic_update(long delta, struct rw_semaphore *sem)
+ {
+-	return delta + xadd(&sem->count, delta);
++	return delta + xadd_check_overflow(&sem->count, delta);
+ }
+ 
+ #endif /* __KERNEL__ */
+diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h
+index 5e64171..f58957e 100644
+--- a/arch/x86/include/asm/segment.h
++++ b/arch/x86/include/asm/segment.h
+@@ -64,10 +64,15 @@
+  *  26 - ESPFIX small SS
+  *  27 - per-cpu			[ offset to per-cpu data area ]
+  *  28 - stack_canary-20		[ for stack protector ]
+- *  29 - unused
+- *  30 - unused
++ *  29 - PCI BIOS CS
++ *  30 - PCI BIOS DS
+  *  31 - TSS for double fault handler
+  */
++#define GDT_ENTRY_KERNEXEC_EFI_CS	(1)
++#define GDT_ENTRY_KERNEXEC_EFI_DS	(2)
++#define __KERNEXEC_EFI_CS	(GDT_ENTRY_KERNEXEC_EFI_CS*8)
++#define __KERNEXEC_EFI_DS	(GDT_ENTRY_KERNEXEC_EFI_DS*8)
++
+ #define GDT_ENTRY_TLS_MIN	6
+ #define GDT_ENTRY_TLS_MAX 	(GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1)
+ 
+@@ -79,6 +84,8 @@
+ 
+ #define GDT_ENTRY_KERNEL_CS		(GDT_ENTRY_KERNEL_BASE+0)
+ 
++#define GDT_ENTRY_KERNEXEC_KERNEL_CS	(4)
++
+ #define GDT_ENTRY_KERNEL_DS		(GDT_ENTRY_KERNEL_BASE+1)
+ 
+ #define GDT_ENTRY_TSS			(GDT_ENTRY_KERNEL_BASE+4)
+@@ -104,6 +111,12 @@
+ #define __KERNEL_STACK_CANARY		0
+ #endif
+ 
++#define GDT_ENTRY_PCIBIOS_CS		(GDT_ENTRY_KERNEL_BASE+17)
++#define __PCIBIOS_CS (GDT_ENTRY_PCIBIOS_CS * 8)
++
++#define GDT_ENTRY_PCIBIOS_DS		(GDT_ENTRY_KERNEL_BASE+18)
++#define __PCIBIOS_DS (GDT_ENTRY_PCIBIOS_DS * 8)
++
+ #define GDT_ENTRY_DOUBLEFAULT_TSS	31
+ 
+ /*
+@@ -141,7 +154,7 @@
+  */
+ 
+ /* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */
+-#define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8)
++#define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xFFFCU) == PNP_CS32 || ((x) & 0xFFFCU) == PNP_CS16)
+ 
+ 
+ #else
+@@ -165,6 +178,8 @@
+ #define __USER32_CS   (GDT_ENTRY_DEFAULT_USER32_CS*8+3)
+ #define __USER32_DS	__USER_DS
+ 
++#define GDT_ENTRY_KERNEXEC_KERNEL_CS 7
++
+ #define GDT_ENTRY_TSS 8	/* needs two entries */
+ #define GDT_ENTRY_LDT 10 /* needs two entries */
+ #define GDT_ENTRY_TLS_MIN 12
+@@ -185,6 +200,7 @@
+ #endif
+ 
+ #define __KERNEL_CS	(GDT_ENTRY_KERNEL_CS*8)
++#define __KERNEXEC_KERNEL_CS	(GDT_ENTRY_KERNEXEC_KERNEL_CS*8)
+ #define __KERNEL_DS	(GDT_ENTRY_KERNEL_DS*8)
+ #define __USER_DS	(GDT_ENTRY_DEFAULT_USER_DS*8+3)
+ #define __USER_CS	(GDT_ENTRY_DEFAULT_USER_CS*8+3)
+diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
+index 73b11bc..d4a3b63 100644
+--- a/arch/x86/include/asm/smp.h
++++ b/arch/x86/include/asm/smp.h
+@@ -36,7 +36,7 @@ DECLARE_PER_CPU(cpumask_var_t, cpu_core_map);
+ /* cpus sharing the last level cache: */
+ DECLARE_PER_CPU(cpumask_var_t, cpu_llc_shared_map);
+ DECLARE_PER_CPU(u16, cpu_llc_id);
+-DECLARE_PER_CPU(int, cpu_number);
++DECLARE_PER_CPU(unsigned int, cpu_number);
+ 
+ static inline struct cpumask *cpu_sibling_mask(int cpu)
+ {
+@@ -77,7 +77,7 @@ struct smp_ops {
+ 
+ 	void (*send_call_func_ipi)(const struct cpumask *mask);
+ 	void (*send_call_func_single_ipi)(int cpu);
+-};
++} __no_const;
+ 
+ /* Globals due to paravirt */
+ extern void set_cpu_sibling_map(int cpu);
+@@ -192,14 +192,8 @@ extern unsigned disabled_cpus __cpuinitdata;
+ extern int safe_smp_processor_id(void);
+ 
+ #elif defined(CONFIG_X86_64_SMP)
+-#define raw_smp_processor_id() (percpu_read(cpu_number))
+-
+-#define stack_smp_processor_id()					\
+-({								\
+-	struct thread_info *ti;						\
+-	__asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK));	\
+-	ti->cpu;							\
+-})
++#define raw_smp_processor_id()		(percpu_read(cpu_number))
++#define stack_smp_processor_id()	raw_smp_processor_id()
+ #define safe_smp_processor_id()		smp_processor_id()
+ 
+ #endif
+diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h
+index 972c260..43ab1fd 100644
+--- a/arch/x86/include/asm/spinlock.h
++++ b/arch/x86/include/asm/spinlock.h
+@@ -188,6 +188,14 @@ static inline int arch_write_can_lock(arch_rwlock_t *lock)
+ static inline void arch_read_lock(arch_rwlock_t *rw)
+ {
+ 	asm volatile(LOCK_PREFIX READ_LOCK_SIZE(dec) " (%0)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX READ_LOCK_SIZE(inc) " (%0)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     "jns 1f\n"
+ 		     "call __read_lock_failed\n\t"
+ 		     "1:\n"
+@@ -197,6 +205,14 @@ static inline void arch_read_lock(arch_rwlock_t *rw)
+ static inline void arch_write_lock(arch_rwlock_t *rw)
+ {
+ 	asm volatile(LOCK_PREFIX WRITE_LOCK_SUB(%1) "(%0)\n\t"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX WRITE_LOCK_ADD(%1) "(%0)\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     "jz 1f\n"
+ 		     "call __write_lock_failed\n\t"
+ 		     "1:\n"
+@@ -226,13 +242,29 @@ static inline int arch_write_trylock(arch_rwlock_t *lock)
+ 
+ static inline void arch_read_unlock(arch_rwlock_t *rw)
+ {
+-	asm volatile(LOCK_PREFIX READ_LOCK_SIZE(inc) " %0"
++	asm volatile(LOCK_PREFIX READ_LOCK_SIZE(inc) " %0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX READ_LOCK_SIZE(dec) " %0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     :"+m" (rw->lock) : : "memory");
+ }
+ 
+ static inline void arch_write_unlock(arch_rwlock_t *rw)
+ {
+-	asm volatile(LOCK_PREFIX WRITE_LOCK_ADD(%1) "%0"
++	asm volatile(LOCK_PREFIX WRITE_LOCK_ADD(%1) "%0\n"
++
++#ifdef CONFIG_PAX_REFCOUNT
++		     "jno 0f\n"
++		     LOCK_PREFIX WRITE_LOCK_SUB(%1) "%0\n"
++		     "int $4\n0:\n"
++		     _ASM_EXTABLE(0b, 0b)
++#endif
++
+ 		     : "+m" (rw->write) : "i" (RW_LOCK_BIAS) : "memory");
+ }
+ 
+diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h
+index 1575177..cb23f52 100644
+--- a/arch/x86/include/asm/stackprotector.h
++++ b/arch/x86/include/asm/stackprotector.h
+@@ -48,7 +48,7 @@
+  * head_32 for boot CPU and setup_per_cpu_areas() for others.
+  */
+ #define GDT_STACK_CANARY_INIT						\
+-	[GDT_ENTRY_STACK_CANARY] = GDT_ENTRY_INIT(0x4090, 0, 0x18),
++	[GDT_ENTRY_STACK_CANARY] = GDT_ENTRY_INIT(0x4090, 0, 0x17),
+ 
+ /*
+  * Initialize the stackprotector canary value.
+@@ -113,7 +113,7 @@ static inline void setup_stack_canary_segment(int cpu)
+ 
+ static inline void load_stack_canary_segment(void)
+ {
+-#ifdef CONFIG_X86_32
++#if defined(CONFIG_X86_32) && !defined(CONFIG_PAX_MEMORY_UDEREF)
+ 	asm volatile ("mov %0, %%gs" : : "r" (0));
+ #endif
+ }
+diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h
+index 70bbe39..4ae2bd4 100644
+--- a/arch/x86/include/asm/stacktrace.h
++++ b/arch/x86/include/asm/stacktrace.h
+@@ -11,28 +11,20 @@
+ 
+ extern int kstack_depth_to_print;
+ 
+-struct thread_info;
++struct task_struct;
+ struct stacktrace_ops;
+ 
+-typedef unsigned long (*walk_stack_t)(struct thread_info *tinfo,
+-				      unsigned long *stack,
+-				      unsigned long bp,
+-				      const struct stacktrace_ops *ops,
+-				      void *data,
+-				      unsigned long *end,
+-				      int *graph);
++typedef unsigned long walk_stack_t(struct task_struct *task,
++				   void *stack_start,
++				   unsigned long *stack,
++				   unsigned long bp,
++				   const struct stacktrace_ops *ops,
++				   void *data,
++				   unsigned long *end,
++				   int *graph);
+ 
+-extern unsigned long
+-print_context_stack(struct thread_info *tinfo,
+-		    unsigned long *stack, unsigned long bp,
+-		    const struct stacktrace_ops *ops, void *data,
+-		    unsigned long *end, int *graph);
+-
+-extern unsigned long
+-print_context_stack_bp(struct thread_info *tinfo,
+-		       unsigned long *stack, unsigned long bp,
+-		       const struct stacktrace_ops *ops, void *data,
+-		       unsigned long *end, int *graph);
++extern walk_stack_t print_context_stack;
++extern walk_stack_t print_context_stack_bp;
+ 
+ /* Generic stack tracer with callbacks */
+ 
+@@ -40,7 +32,7 @@ struct stacktrace_ops {
+ 	void (*address)(void *data, unsigned long address, int reliable);
+ 	/* On negative return stop dumping */
+ 	int (*stack)(void *data, char *name);
+-	walk_stack_t	walk_stack;
++	walk_stack_t	*walk_stack;
+ };
+ 
+ void dump_trace(struct task_struct *tsk, struct pt_regs *regs,
+diff --git a/arch/x86/include/asm/sys_ia32.h b/arch/x86/include/asm/sys_ia32.h
+index cb23852..2dde194 100644
+--- a/arch/x86/include/asm/sys_ia32.h
++++ b/arch/x86/include/asm/sys_ia32.h
+@@ -40,7 +40,7 @@ asmlinkage long sys32_rt_sigprocmask(int, compat_sigset_t __user *,
+ 				     compat_sigset_t __user *, unsigned int);
+ asmlinkage long sys32_alarm(unsigned int);
+ 
+-asmlinkage long sys32_waitpid(compat_pid_t, unsigned int *, int);
++asmlinkage long sys32_waitpid(compat_pid_t, unsigned int __user *, int);
+ asmlinkage long sys32_sysfs(int, u32, u32);
+ 
+ asmlinkage long sys32_sched_rr_get_interval(compat_pid_t,
+diff --git a/arch/x86/include/asm/syscalls.h b/arch/x86/include/asm/syscalls.h
+index f1d8b44..a4de8b7 100644
+--- a/arch/x86/include/asm/syscalls.h
++++ b/arch/x86/include/asm/syscalls.h
+@@ -30,7 +30,7 @@ long sys_clone(unsigned long, unsigned long, void __user *,
+ 	       void __user *, struct pt_regs *);
+ 
+ /* kernel/ldt.c */
+-asmlinkage int sys_modify_ldt(int, void __user *, unsigned long);
++asmlinkage int sys_modify_ldt(int, void __user *, unsigned long) __size_overflow(3);
+ 
+ /* kernel/signal.c */
+ long sys_rt_sigreturn(struct pt_regs *);
+diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h
+index 2d2f01c..f985723 100644
+--- a/arch/x86/include/asm/system.h
++++ b/arch/x86/include/asm/system.h
+@@ -129,7 +129,7 @@ do {									\
+ 	     "call __switch_to\n\t"					  \
+ 	     "movq "__percpu_arg([current_task])",%%rsi\n\t"		  \
+ 	     __switch_canary						  \
+-	     "movq %P[thread_info](%%rsi),%%r8\n\t"			  \
++	     "movq "__percpu_arg([thread_info])",%%r8\n\t"		  \
+ 	     "movq %%rax,%%rdi\n\t" 					  \
+ 	     "testl  %[_tif_fork],%P[ti_flags](%%r8)\n\t"		  \
+ 	     "jnz   ret_from_fork\n\t"					  \
+@@ -140,7 +140,7 @@ do {									\
+ 	       [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \
+ 	       [ti_flags] "i" (offsetof(struct thread_info, flags)),	  \
+ 	       [_tif_fork] "i" (_TIF_FORK),			  	  \
+-	       [thread_info] "i" (offsetof(struct task_struct, stack)),   \
++	       [thread_info] "m" (current_tinfo),			  \
+ 	       [current_task] "m" (current_task)			  \
+ 	       __switch_canary_iparam					  \
+ 	     : "memory", "cc" __EXTRA_CLOBBER)
+@@ -200,7 +200,7 @@ static inline unsigned long get_limit(unsigned long segment)
+ {
+ 	unsigned long __limit;
+ 	asm("lsll %1,%0" : "=r" (__limit) : "r" (segment));
+-	return __limit + 1;
++	return __limit;
+ }
+ 
+ static inline void native_clts(void)
+@@ -397,13 +397,13 @@ void enable_hlt(void);
+ 
+ void cpu_idle_wait(void);
+ 
+-extern unsigned long arch_align_stack(unsigned long sp);
++#define arch_align_stack(x) ((x) & ~0xfUL)
+ extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
+ 
+ void default_idle(void);
+ bool set_pm_idle_to_default(void);
+ 
+-void stop_this_cpu(void *dummy);
++void stop_this_cpu(void *dummy) __noreturn;
+ 
+ /*
+  * Force strict CPU ordering.
+diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
+index d7ef849..6af292e 100644
+--- a/arch/x86/include/asm/thread_info.h
++++ b/arch/x86/include/asm/thread_info.h
+@@ -10,6 +10,7 @@
+ #include <linux/compiler.h>
+ #include <asm/page.h>
+ #include <asm/types.h>
++#include <asm/percpu.h>
+ 
+ /*
+  * low level task data that entry.S needs immediate access to
+@@ -24,7 +25,6 @@ struct exec_domain;
+ #include <linux/atomic.h>
+ 
+ struct thread_info {
+-	struct task_struct	*task;		/* main task structure */
+ 	struct exec_domain	*exec_domain;	/* execution domain */
+ 	__u32			flags;		/* low level flags */
+ 	__u32			status;		/* thread synchronous flags */
+@@ -34,18 +34,12 @@ struct thread_info {
+ 	mm_segment_t		addr_limit;
+ 	struct restart_block    restart_block;
+ 	void __user		*sysenter_return;
+-#ifdef CONFIG_X86_32
+-	unsigned long           previous_esp;   /* ESP of the previous stack in
+-						   case of nested (IRQ) stacks
+-						*/
+-	__u8			supervisor_stack[0];
+-#endif
++	unsigned long		lowest_stack;
+ 	int			uaccess_err;
+ };
+ 
+-#define INIT_THREAD_INFO(tsk)			\
++#define INIT_THREAD_INFO			\
+ {						\
+-	.task		= &tsk,			\
+ 	.exec_domain	= &default_exec_domain,	\
+ 	.flags		= 0,			\
+ 	.cpu		= 0,			\
+@@ -56,7 +50,7 @@ struct thread_info {
+ 	},					\
+ }
+ 
+-#define init_thread_info	(init_thread_union.thread_info)
++#define init_thread_info	(init_thread_union.stack)
+ #define init_stack		(init_thread_union.stack)
+ 
+ #else /* !__ASSEMBLY__ */
+@@ -170,45 +164,40 @@ struct thread_info {
+ 	ret;								\
+ })
+ 
+-#ifdef CONFIG_X86_32
+-
+-#define STACK_WARN	(THREAD_SIZE/8)
+-/*
+- * macros/functions for gaining access to the thread information structure
+- *
+- * preempt_count needs to be 1 initially, until the scheduler is functional.
+- */
+-#ifndef __ASSEMBLY__
+-
+-
+-/* how to get the current stack pointer from C */
+-register unsigned long current_stack_pointer asm("esp") __used;
+-
+-/* how to get the thread information struct from C */
+-static inline struct thread_info *current_thread_info(void)
+-{
+-	return (struct thread_info *)
+-		(current_stack_pointer & ~(THREAD_SIZE - 1));
+-}
+-
+-#else /* !__ASSEMBLY__ */
+-
++#ifdef __ASSEMBLY__
+ /* how to get the thread information struct from ASM */
+ #define GET_THREAD_INFO(reg)	 \
+-	movl $-THREAD_SIZE, reg; \
+-	andl %esp, reg
++	mov PER_CPU_VAR(current_tinfo), reg
+ 
+ /* use this one if reg already contains %esp */
+-#define GET_THREAD_INFO_WITH_ESP(reg) \
+-	andl $-THREAD_SIZE, reg
++#define GET_THREAD_INFO_WITH_ESP(reg) GET_THREAD_INFO(reg)
++#else
++/* how to get the thread information struct from C */
++DECLARE_PER_CPU(struct thread_info *, current_tinfo);
++
++static __always_inline struct thread_info *current_thread_info(void)
++{
++	return percpu_read_stable(current_tinfo);
++}
++#endif
++
++#ifdef CONFIG_X86_32
++
++#define STACK_WARN	(THREAD_SIZE/8)
++/*
++ * macros/functions for gaining access to the thread information structure
++ *
++ * preempt_count needs to be 1 initially, until the scheduler is functional.
++ */
++#ifndef __ASSEMBLY__
++
++/* how to get the current stack pointer from C */
++register unsigned long current_stack_pointer asm("esp") __used;
+ 
+ #endif
+ 
+ #else /* X86_32 */
+ 
+-#include <asm/percpu.h>
+-#define KERNEL_STACK_OFFSET (5*8)
+-
+ /*
+  * macros/functions for gaining access to the thread information structure
+  * preempt_count needs to be 1 initially, until the scheduler is functional.
+@@ -216,21 +205,8 @@ static inline struct thread_info *current_thread_info(void)
+ #ifndef __ASSEMBLY__
+ DECLARE_PER_CPU(unsigned long, kernel_stack);
+ 
+-static inline struct thread_info *current_thread_info(void)
+-{
+-	struct thread_info *ti;
+-	ti = (void *)(percpu_read_stable(kernel_stack) +
+-		      KERNEL_STACK_OFFSET - THREAD_SIZE);
+-	return ti;
+-}
+-
+-#else /* !__ASSEMBLY__ */
+-
+-/* how to get the thread information struct from ASM */
+-#define GET_THREAD_INFO(reg) \
+-	movq PER_CPU_VAR(kernel_stack),reg ; \
+-	subq $(THREAD_SIZE-KERNEL_STACK_OFFSET),reg
+-
++/* how to get the current stack pointer from C */
++register unsigned long current_stack_pointer asm("rsp") __used;
+ #endif
+ 
+ #endif /* !X86_32 */
+@@ -264,5 +240,16 @@ extern void arch_task_cache_init(void);
+ extern void free_thread_info(struct thread_info *ti);
+ extern int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src);
+ #define arch_task_cache_init arch_task_cache_init
++
++#define __HAVE_THREAD_FUNCTIONS
++#define task_thread_info(task)	(&(task)->tinfo)
++#define task_stack_page(task)	((task)->stack)
++#define setup_thread_stack(p, org) do {} while (0)
++#define end_of_stack(p) ((unsigned long *)task_stack_page(p) + 1)
++
++#define __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
++extern struct task_struct *alloc_task_struct_node(int node);
++extern void free_task_struct(struct task_struct *);
++
+ #endif
+ #endif /* _ASM_X86_THREAD_INFO_H */
+diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
+index 36361bf..324f262 100644
+--- a/arch/x86/include/asm/uaccess.h
++++ b/arch/x86/include/asm/uaccess.h
+@@ -7,12 +7,15 @@
+ #include <linux/compiler.h>
+ #include <linux/thread_info.h>
+ #include <linux/string.h>
++#include <linux/sched.h>
+ #include <asm/asm.h>
+ #include <asm/page.h>
+ 
+ #define VERIFY_READ 0
+ #define VERIFY_WRITE 1
+ 
++extern void check_object_size(const void *ptr, unsigned long n, bool to);
++
+ /*
+  * The fs value determines whether argument validity checking should be
+  * performed or not.  If get_fs() == USER_DS, checking is performed, with
+@@ -28,7 +31,12 @@
+ 
+ #define get_ds()	(KERNEL_DS)
+ #define get_fs()	(current_thread_info()->addr_limit)
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF)
++void __set_fs(mm_segment_t x);
++void set_fs(mm_segment_t x);
++#else
+ #define set_fs(x)	(current_thread_info()->addr_limit = (x))
++#endif
+ 
+ #define segment_eq(a, b)	((a).seg == (b).seg)
+ 
+@@ -76,7 +84,33 @@
+  * checks that the pointer is in the user space range - after calling
+  * this function, memory access functions may still return -EFAULT.
+  */
+-#define access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0))
++#define __access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0))
++#define access_ok(type, addr, size)					\
++({									\
++	long __size = size;						\
++	unsigned long __addr = (unsigned long)addr;			\
++	unsigned long __addr_ao = __addr & PAGE_MASK;			\
++	unsigned long __end_ao = __addr + __size - 1;			\
++	bool __ret_ao = __range_not_ok(__addr, __size) == 0;		\
++	if (__ret_ao && unlikely((__end_ao ^ __addr_ao) & PAGE_MASK)) {	\
++		while(__addr_ao <= __end_ao) {				\
++			char __c_ao;					\
++			__addr_ao += PAGE_SIZE;				\
++			if (__size > PAGE_SIZE)				\
++				cond_resched();				\
++			if (__get_user(__c_ao, (char __user *)__addr))	\
++				break;					\
++			if (type != VERIFY_WRITE) {			\
++				__addr = __addr_ao;			\
++				continue;				\
++			}						\
++			if (__put_user(__c_ao, (char __user *)__addr))	\
++				break;					\
++			__addr = __addr_ao;				\
++		}							\
++	}								\
++	__ret_ao;							\
++})
+ 
+ /*
+  * The exception table consists of pairs of addresses: the first is the
+@@ -182,12 +216,20 @@ extern int __get_user_bad(void);
+ 	asm volatile("call __put_user_" #size : "=a" (__ret_pu)	\
+ 		     : "0" ((typeof(*(ptr)))(x)), "c" (ptr) : "ebx")
+ 
+-
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF)
++#define __copyuser_seg "gs;"
++#define __COPYUSER_SET_ES "pushl %%gs; popl %%es\n"
++#define __COPYUSER_RESTORE_ES "pushl %%ss; popl %%es\n"
++#else
++#define __copyuser_seg
++#define __COPYUSER_SET_ES
++#define __COPYUSER_RESTORE_ES
++#endif
+ 
+ #ifdef CONFIG_X86_32
+ #define __put_user_asm_u64(x, addr, err, errret)			\
+-	asm volatile("1:	movl %%eax,0(%2)\n"			\
+-		     "2:	movl %%edx,4(%2)\n"			\
++	asm volatile("1:	"__copyuser_seg"movl %%eax,0(%2)\n"	\
++		     "2:	"__copyuser_seg"movl %%edx,4(%2)\n"	\
+ 		     "3:\n"						\
+ 		     ".section .fixup,\"ax\"\n"				\
+ 		     "4:	movl %3,%0\n"				\
+@@ -199,8 +241,8 @@ extern int __get_user_bad(void);
+ 		     : "A" (x), "r" (addr), "i" (errret), "0" (err))
+ 
+ #define __put_user_asm_ex_u64(x, addr)					\
+-	asm volatile("1:	movl %%eax,0(%1)\n"			\
+-		     "2:	movl %%edx,4(%1)\n"			\
++	asm volatile("1:	"__copyuser_seg"movl %%eax,0(%1)\n"	\
++		     "2:	"__copyuser_seg"movl %%edx,4(%1)\n"	\
+ 		     "3:\n"						\
+ 		     _ASM_EXTABLE(1b, 2b - 1b)				\
+ 		     _ASM_EXTABLE(2b, 3b - 2b)				\
+@@ -252,7 +294,7 @@ extern void __put_user_8(void);
+ 	__typeof__(*(ptr)) __pu_val;				\
+ 	__chk_user_ptr(ptr);					\
+ 	might_fault();						\
+-	__pu_val = x;						\
++	__pu_val = (x);						\
+ 	switch (sizeof(*(ptr))) {				\
+ 	case 1:							\
+ 		__put_user_x(1, __pu_val, ptr, __ret_pu);	\
+@@ -373,7 +415,7 @@ do {									\
+ } while (0)
+ 
+ #define __get_user_asm(x, addr, err, itype, rtype, ltype, errret)	\
+-	asm volatile("1:	mov"itype" %2,%"rtype"1\n"		\
++	asm volatile("1:	"__copyuser_seg"mov"itype" %2,%"rtype"1\n"\
+ 		     "2:\n"						\
+ 		     ".section .fixup,\"ax\"\n"				\
+ 		     "3:	mov %3,%0\n"				\
+@@ -381,7 +423,7 @@ do {									\
+ 		     "	jmp 2b\n"					\
+ 		     ".previous\n"					\
+ 		     _ASM_EXTABLE(1b, 3b)				\
+-		     : "=r" (err), ltype(x)				\
++		     : "=r" (err), ltype (x)				\
+ 		     : "m" (__m(addr)), "i" (errret), "0" (err))
+ 
+ #define __get_user_size_ex(x, ptr, size)				\
+@@ -406,7 +448,7 @@ do {									\
+ } while (0)
+ 
+ #define __get_user_asm_ex(x, addr, itype, rtype, ltype)			\
+-	asm volatile("1:	mov"itype" %1,%"rtype"0\n"		\
++	asm volatile("1:	"__copyuser_seg"mov"itype" %1,%"rtype"0\n"\
+ 		     "2:\n"						\
+ 		     _ASM_EXTABLE(1b, 2b - 1b)				\
+ 		     : ltype(x) : "m" (__m(addr)))
+@@ -423,13 +465,24 @@ do {									\
+ 	int __gu_err;							\
+ 	unsigned long __gu_val;						\
+ 	__get_user_size(__gu_val, (ptr), (size), __gu_err, -EFAULT);	\
+-	(x) = (__force __typeof__(*(ptr)))__gu_val;			\
++	(x) = (__typeof__(*(ptr)))__gu_val;				\
+ 	__gu_err;							\
+ })
+ 
+ /* FIXME: this hack is definitely wrong -AK */
+ struct __large_struct { unsigned long buf[100]; };
+-#define __m(x) (*(struct __large_struct __user *)(x))
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++#define ____m(x)					\
++({							\
++	unsigned long ____x = (unsigned long)(x);	\
++	if (____x < PAX_USER_SHADOW_BASE)		\
++		____x += PAX_USER_SHADOW_BASE;		\
++	(void __user *)____x;				\
++})
++#else
++#define ____m(x) (x)
++#endif
++#define __m(x) (*(struct __large_struct __user *)____m(x))
+ 
+ /*
+  * Tell gcc we read from memory instead of writing: this is because
+@@ -437,7 +490,7 @@ struct __large_struct { unsigned long buf[100]; };
+  * aliasing issues.
+  */
+ #define __put_user_asm(x, addr, err, itype, rtype, ltype, errret)	\
+-	asm volatile("1:	mov"itype" %"rtype"1,%2\n"		\
++	asm volatile("1:	"__copyuser_seg"mov"itype" %"rtype"1,%2\n"\
+ 		     "2:\n"						\
+ 		     ".section .fixup,\"ax\"\n"				\
+ 		     "3:	mov %3,%0\n"				\
+@@ -445,10 +498,10 @@ struct __large_struct { unsigned long buf[100]; };
+ 		     ".previous\n"					\
+ 		     _ASM_EXTABLE(1b, 3b)				\
+ 		     : "=r"(err)					\
+-		     : ltype(x), "m" (__m(addr)), "i" (errret), "0" (err))
++		     : ltype (x), "m" (__m(addr)), "i" (errret), "0" (err))
+ 
+ #define __put_user_asm_ex(x, addr, itype, rtype, ltype)			\
+-	asm volatile("1:	mov"itype" %"rtype"0,%1\n"		\
++	asm volatile("1:	"__copyuser_seg"mov"itype" %"rtype"0,%1\n"\
+ 		     "2:\n"						\
+ 		     _ASM_EXTABLE(1b, 2b - 1b)				\
+ 		     : : ltype(x), "m" (__m(addr)))
+@@ -487,8 +540,12 @@ struct __large_struct { unsigned long buf[100]; };
+  * On error, the variable @x is set to zero.
+  */
+ 
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++#define __get_user(x, ptr)	get_user((x), (ptr))
++#else
+ #define __get_user(x, ptr)						\
+ 	__get_user_nocheck((x), (ptr), sizeof(*(ptr)))
++#endif
+ 
+ /**
+  * __put_user: - Write a simple value into user space, with less checking.
+@@ -510,8 +567,12 @@ struct __large_struct { unsigned long buf[100]; };
+  * Returns zero on success, or -EFAULT on error.
+  */
+ 
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++#define __put_user(x, ptr)	put_user((x), (ptr))
++#else
+ #define __put_user(x, ptr)						\
+ 	__put_user_nocheck((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr)))
++#endif
+ 
+ #define __get_user_unaligned __get_user
+ #define __put_user_unaligned __put_user
+@@ -529,7 +590,7 @@ struct __large_struct { unsigned long buf[100]; };
+ #define get_user_ex(x, ptr)	do {					\
+ 	unsigned long __gue_val;					\
+ 	__get_user_size_ex((__gue_val), (ptr), (sizeof(*(ptr))));	\
+-	(x) = (__force __typeof__(*(ptr)))__gue_val;			\
++	(x) = (__typeof__(*(ptr)))__gue_val;				\
+ } while (0)
+ 
+ #ifdef CONFIG_X86_WP_WORKS_OK
+diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h
+index 566e803..7183d0b 100644
+--- a/arch/x86/include/asm/uaccess_32.h
++++ b/arch/x86/include/asm/uaccess_32.h
+@@ -11,15 +11,15 @@
+ #include <asm/page.h>
+ 
+ unsigned long __must_check __copy_to_user_ll
+-		(void __user *to, const void *from, unsigned long n);
++		(void __user *to, const void *from, unsigned long n) __size_overflow(3);
+ unsigned long __must_check __copy_from_user_ll
+-		(void *to, const void __user *from, unsigned long n);
++		(void *to, const void __user *from, unsigned long n) __size_overflow(3);
+ unsigned long __must_check __copy_from_user_ll_nozero
+-		(void *to, const void __user *from, unsigned long n);
++		(void *to, const void __user *from, unsigned long n) __size_overflow(3);
+ unsigned long __must_check __copy_from_user_ll_nocache
+-		(void *to, const void __user *from, unsigned long n);
++		(void *to, const void __user *from, unsigned long n) __size_overflow(3);
+ unsigned long __must_check __copy_from_user_ll_nocache_nozero
+-		(void *to, const void __user *from, unsigned long n);
++		(void *to, const void __user *from, unsigned long n) __size_overflow(3);
+ 
+ /**
+  * __copy_to_user_inatomic: - Copy a block of data into user space, with less checking.
+@@ -41,8 +41,13 @@ unsigned long __must_check __copy_from_user_ll_nocache_nozero
+  */
+ 
+ static __always_inline unsigned long __must_check
++__copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) __size_overflow(3);
++static __always_inline unsigned long __must_check
+ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	if (__builtin_constant_p(n)) {
+ 		unsigned long ret;
+ 
+@@ -61,6 +66,8 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
+ 			return ret;
+ 		}
+ 	}
++	if (!__builtin_constant_p(n))
++		check_object_size(from, n, true);
+ 	return __copy_to_user_ll(to, from, n);
+ }
+ 
+@@ -79,15 +86,23 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
+  * On success, this will be zero.
+  */
+ static __always_inline unsigned long __must_check
++__copy_to_user(void __user *to, const void *from, unsigned long n) __size_overflow(3);
++static __always_inline unsigned long __must_check
+ __copy_to_user(void __user *to, const void *from, unsigned long n)
+ {
+ 	might_fault();
++
+ 	return __copy_to_user_inatomic(to, from, n);
+ }
+ 
+ static __always_inline unsigned long
++__copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) __size_overflow(3);
++static __always_inline unsigned long
+ __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n)
+ {
++	if ((long)n < 0)
++		return n;
++
+ 	/* Avoid zeroing the tail if the copy fails..
+ 	 * If 'n' is constant and 1, 2, or 4, we do still zero on a failure,
+ 	 * but as the zeroing behaviour is only significant when n is not
+@@ -134,9 +149,15 @@ __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n)
+  * for explanation of why this is needed.
+  */
+ static __always_inline unsigned long
++__copy_from_user(void *to, const void __user *from, unsigned long n) __size_overflow(3);
++static __always_inline unsigned long
+ __copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
+ 	might_fault();
++
++	if ((long)n < 0)
++		return n;
++
+ 	if (__builtin_constant_p(n)) {
+ 		unsigned long ret;
+ 
+@@ -152,13 +173,21 @@ __copy_from_user(void *to, const void __user *from, unsigned long n)
+ 			return ret;
+ 		}
+ 	}
++	if (!__builtin_constant_p(n))
++		check_object_size(to, n, false);
+ 	return __copy_from_user_ll(to, from, n);
+ }
+ 
+ static __always_inline unsigned long __copy_from_user_nocache(void *to,
++				const void __user *from, unsigned long n) __size_overflow(3);
++static __always_inline unsigned long __copy_from_user_nocache(void *to,
+ 				const void __user *from, unsigned long n)
+ {
+ 	might_fault();
++
++	if ((long)n < 0)
++		return n;
++
+ 	if (__builtin_constant_p(n)) {
+ 		unsigned long ret;
+ 
+@@ -179,17 +208,24 @@ 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) __size_overflow(3);
++static __always_inline unsigned long
++__copy_from_user_inatomic_nocache(void *to, const void __user *from,
+ 				  unsigned long n)
+ {
+-       return __copy_from_user_ll_nocache_nozero(to, from, n);
++	if ((long)n < 0)
++		return n;
++
++	return __copy_from_user_ll_nocache_nozero(to, from, n);
+ }
+ 
+-unsigned long __must_check copy_to_user(void __user *to,
+-					const void *from, unsigned long n);
+-unsigned long __must_check _copy_from_user(void *to,
+-					  const void __user *from,
+-					  unsigned long n);
+-
++extern void copy_to_user_overflow(void)
++#ifdef CONFIG_DEBUG_STRICT_USER_COPY_CHECKS
++	__compiletime_error("copy_to_user() buffer size is not provably correct")
++#else
++	__compiletime_warning("copy_to_user() buffer size is not provably correct")
++#endif
++;
+ 
+ extern void copy_from_user_overflow(void)
+ #ifdef CONFIG_DEBUG_STRICT_USER_COPY_CHECKS
+@@ -199,17 +235,65 @@ extern void copy_from_user_overflow(void)
+ #endif
+ ;
+ 
+-static inline unsigned long __must_check copy_from_user(void *to,
+-					  const void __user *from,
+-					  unsigned long n)
++/**
++ * copy_to_user: - Copy a block of data into user space.
++ * @to:   Destination address, in user space.
++ * @from: Source address, in kernel space.
++ * @n:    Number of bytes to copy.
++ *
++ * Context: User context only.  This function may sleep.
++ *
++ * Copy data from kernel space to user space.
++ *
++ * Returns number of bytes that could not be copied.
++ * On success, this will be zero.
++ */
++static inline unsigned long __must_check
++copy_to_user(void __user *to, const void *from, unsigned long n) __size_overflow(3);
++static inline unsigned long __must_check
++copy_to_user(void __user *to, const void *from, unsigned long n)
++{
++	int sz = __compiletime_object_size(from);
++
++	if (unlikely(sz != -1 && sz < n))
++		copy_to_user_overflow();
++	else if (access_ok(VERIFY_WRITE, to, n))
++		n = __copy_to_user(to, from, n);
++	return n;
++}
++
++/**
++ * copy_from_user: - Copy a block of data from user space.
++ * @to:   Destination address, in kernel space.
++ * @from: Source address, in user space.
++ * @n:    Number of bytes to copy.
++ *
++ * Context: User context only.  This function may sleep.
++ *
++ * Copy data from user space to kernel space.
++ *
++ * Returns number of bytes that could not be copied.
++ * On success, this will be zero.
++ *
++ * If some data could not be copied, this function will pad the copied
++ * data to the requested size using zero bytes.
++ */
++static inline unsigned long __must_check
++copy_from_user(void *to, const void __user *from, unsigned long n) __size_overflow(3);
++static inline unsigned long __must_check
++copy_from_user(void *to, const void __user *from, unsigned long n)
+ {
+ 	int sz = __compiletime_object_size(to);
+ 
+-	if (likely(sz == -1 || sz >= n))
+-		n = _copy_from_user(to, from, n);
+-	else
++	if (unlikely(sz != -1 && sz < n))
+ 		copy_from_user_overflow();
+-
++	else if (access_ok(VERIFY_READ, from, n))
++		n = __copy_from_user(to, from, n);
++	else if ((long)n > 0) {
++		if (!__builtin_constant_p(n))
++			check_object_size(to, n, false);
++		memset(to, 0, n);
++	}
+ 	return n;
+ }
+ 
+@@ -235,7 +319,7 @@ long __must_check __strncpy_from_user(char *dst,
+ #define strlen_user(str) strnlen_user(str, LONG_MAX)
+ 
+ long strnlen_user(const char __user *str, long n);
+-unsigned long __must_check clear_user(void __user *mem, unsigned long len);
+-unsigned long __must_check __clear_user(void __user *mem, unsigned long len);
++unsigned long __must_check clear_user(void __user *mem, unsigned long len) __size_overflow(2);
++unsigned long __must_check __clear_user(void __user *mem, unsigned long len) __size_overflow(2);
+ 
+ #endif /* _ASM_X86_UACCESS_32_H */
+diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h
+index 1c66d30..e294b5f 100644
+--- a/arch/x86/include/asm/uaccess_64.h
++++ b/arch/x86/include/asm/uaccess_64.h
+@@ -10,6 +10,9 @@
+ #include <asm/alternative.h>
+ #include <asm/cpufeature.h>
+ #include <asm/page.h>
++#include <asm/pgtable.h>
++
++#define set_fs(x)	(current_thread_info()->addr_limit = (x))
+ 
+ /*
+  * Copy To/From Userspace
+@@ -17,12 +20,14 @@
+ 
+ /* Handles exceptions in both to and from, but doesn't do access_ok */
+ __must_check unsigned long
+-copy_user_generic_string(void *to, const void *from, unsigned len);
++copy_user_generic_string(void *to, const void *from, unsigned long len) __size_overflow(3);
+ __must_check unsigned long
+-copy_user_generic_unrolled(void *to, const void *from, unsigned len);
++copy_user_generic_unrolled(void *to, const void *from, unsigned long len) __size_overflow(3);
+ 
+ static __always_inline __must_check unsigned long
+-copy_user_generic(void *to, const void *from, unsigned len)
++copy_user_generic(void *to, const void *from, unsigned long len) __size_overflow(3);
++static __always_inline __must_check unsigned long
++copy_user_generic(void *to, const void *from, unsigned long len)
+ {
+ 	unsigned ret;
+ 
+@@ -32,142 +37,237 @@ copy_user_generic(void *to, const void *from, unsigned len)
+ 			 ASM_OUTPUT2("=a" (ret), "=D" (to), "=S" (from),
+ 				     "=d" (len)),
+ 			 "1" (to), "2" (from), "3" (len)
+-			 : "memory", "rcx", "r8", "r9", "r10", "r11");
++			 : "memory", "rcx", "r8", "r9", "r11");
+ 	return ret;
+ }
+ 
++static __always_inline __must_check unsigned long
++__copy_to_user(void __user *to, const void *from, unsigned long len) __size_overflow(3);
++static __always_inline __must_check unsigned long
++__copy_from_user(void *to, const void __user *from, unsigned long len) __size_overflow(3);
+ __must_check unsigned long
+-_copy_to_user(void __user *to, const void *from, unsigned len);
+-__must_check unsigned long
+-_copy_from_user(void *to, const void __user *from, unsigned len);
+-__must_check unsigned long
+-copy_in_user(void __user *to, const void __user *from, unsigned len);
++copy_in_user(void __user *to, const void __user *from, unsigned long len) __size_overflow(3);
+ 
+ static inline unsigned long __must_check copy_from_user(void *to,
+ 					  const void __user *from,
++					  unsigned long n) __size_overflow(3);
++static inline unsigned long __must_check copy_from_user(void *to,
++					  const void __user *from,
+ 					  unsigned long n)
+ {
+-	int sz = __compiletime_object_size(to);
+-
+ 	might_fault();
+-	if (likely(sz == -1 || sz >= n))
+-		n = _copy_from_user(to, from, n);
+-#ifdef CONFIG_DEBUG_VM
+-	else
+-		WARN(1, "Buffer overflow detected!\n");
+-#endif
++
++	if (access_ok(VERIFY_READ, from, n))
++		n = __copy_from_user(to, from, n);
++	else if (n < INT_MAX) {
++		if (!__builtin_constant_p(n))
++			check_object_size(to, n, false);
++		memset(to, 0, n);
++	}
+ 	return n;
+ }
+ 
+ static __always_inline __must_check
+-int copy_to_user(void __user *dst, const void *src, unsigned size)
++int copy_to_user(void __user *dst, const void *src, unsigned long size) __size_overflow(3);
++static __always_inline __must_check
++int copy_to_user(void __user *dst, const void *src, unsigned long size)
+ {
+ 	might_fault();
+ 
+-	return _copy_to_user(dst, src, size);
++	if (access_ok(VERIFY_WRITE, dst, size))
++		size = __copy_to_user(dst, src, size);
++	return size;
+ }
+ 
+ static __always_inline __must_check
+-int __copy_from_user(void *dst, const void __user *src, unsigned size)
++unsigned long __copy_from_user(void *dst, const void __user *src, unsigned long size) __size_overflow(3);
++static __always_inline __must_check
++unsigned long __copy_from_user(void *dst, const void __user *src, unsigned long size)
+ {
+-	int ret = 0;
++	int sz = __compiletime_object_size(dst);
++	unsigned ret = 0;
+ 
+ 	might_fault();
+-	if (!__builtin_constant_p(size))
+-		return copy_user_generic(dst, (__force void *)src, size);
++
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_READ, src, size))
++		return size;
++#endif
++
++	if (unlikely(sz != -1 && sz < size)) {
++#ifdef CONFIG_DEBUG_VM
++		WARN(1, "Buffer overflow detected!\n");
++#endif
++		return size;
++	}
++
++	if (!__builtin_constant_p(size)) {
++		check_object_size(dst, size, false);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		if ((unsigned long)src < PAX_USER_SHADOW_BASE)
++			src += PAX_USER_SHADOW_BASE;
++#endif
++
++		return copy_user_generic(dst, (__force_kernel const void *)src, size);
++	}
+ 	switch (size) {
+-	case 1:__get_user_asm(*(u8 *)dst, (u8 __user *)src,
++	case 1:__get_user_asm(*(u8 *)dst, (const u8 __user *)src,
+ 			      ret, "b", "b", "=q", 1);
+ 		return ret;
+-	case 2:__get_user_asm(*(u16 *)dst, (u16 __user *)src,
++	case 2:__get_user_asm(*(u16 *)dst, (const u16 __user *)src,
+ 			      ret, "w", "w", "=r", 2);
+ 		return ret;
+-	case 4:__get_user_asm(*(u32 *)dst, (u32 __user *)src,
++	case 4:__get_user_asm(*(u32 *)dst, (const u32 __user *)src,
+ 			      ret, "l", "k", "=r", 4);
+ 		return ret;
+-	case 8:__get_user_asm(*(u64 *)dst, (u64 __user *)src,
++	case 8:__get_user_asm(*(u64 *)dst, (const u64 __user *)src,
+ 			      ret, "q", "", "=r", 8);
+ 		return ret;
+ 	case 10:
+-		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
++		__get_user_asm(*(u64 *)dst, (const u64 __user *)src,
+ 			       ret, "q", "", "=r", 10);
+ 		if (unlikely(ret))
+ 			return ret;
+ 		__get_user_asm(*(u16 *)(8 + (char *)dst),
+-			       (u16 __user *)(8 + (char __user *)src),
++			       (const u16 __user *)(8 + (const char __user *)src),
+ 			       ret, "w", "w", "=r", 2);
+ 		return ret;
+ 	case 16:
+-		__get_user_asm(*(u64 *)dst, (u64 __user *)src,
++		__get_user_asm(*(u64 *)dst, (const u64 __user *)src,
+ 			       ret, "q", "", "=r", 16);
+ 		if (unlikely(ret))
+ 			return ret;
+ 		__get_user_asm(*(u64 *)(8 + (char *)dst),
+-			       (u64 __user *)(8 + (char __user *)src),
++			       (const u64 __user *)(8 + (const char __user *)src),
+ 			       ret, "q", "", "=r", 8);
+ 		return ret;
+ 	default:
+-		return copy_user_generic(dst, (__force void *)src, size);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		if ((unsigned long)src < PAX_USER_SHADOW_BASE)
++			src += PAX_USER_SHADOW_BASE;
++#endif
++
++		return copy_user_generic(dst, (__force_kernel const void *)src, size);
+ 	}
+ }
+ 
+ static __always_inline __must_check
+-int __copy_to_user(void __user *dst, const void *src, unsigned size)
++unsigned long __copy_to_user(void __user *dst, const void *src, unsigned long size) __size_overflow(3);
++static __always_inline __must_check
++unsigned long __copy_to_user(void __user *dst, const void *src, unsigned long size)
+ {
+-	int ret = 0;
++	int sz = __compiletime_object_size(src);
++	unsigned ret = 0;
+ 
+ 	might_fault();
+-	if (!__builtin_constant_p(size))
+-		return copy_user_generic((__force void *)dst, src, size);
++
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_WRITE, dst, size))
++		return size;
++#endif
++
++	if (unlikely(sz != -1 && sz < size)) {
++#ifdef CONFIG_DEBUG_VM
++		WARN(1, "Buffer overflow detected!\n");
++#endif
++		return size;
++	}
++
++	if (!__builtin_constant_p(size)) {
++		check_object_size(src, size, true);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		if ((unsigned long)dst < PAX_USER_SHADOW_BASE)
++			dst += PAX_USER_SHADOW_BASE;
++#endif
++
++		return copy_user_generic((__force_kernel void *)dst, src, size);
++	}
+ 	switch (size) {
+-	case 1:__put_user_asm(*(u8 *)src, (u8 __user *)dst,
++	case 1:__put_user_asm(*(const u8 *)src, (u8 __user *)dst,
+ 			      ret, "b", "b", "iq", 1);
+ 		return ret;
+-	case 2:__put_user_asm(*(u16 *)src, (u16 __user *)dst,
++	case 2:__put_user_asm(*(const u16 *)src, (u16 __user *)dst,
+ 			      ret, "w", "w", "ir", 2);
+ 		return ret;
+-	case 4:__put_user_asm(*(u32 *)src, (u32 __user *)dst,
++	case 4:__put_user_asm(*(const u32 *)src, (u32 __user *)dst,
+ 			      ret, "l", "k", "ir", 4);
+ 		return ret;
+-	case 8:__put_user_asm(*(u64 *)src, (u64 __user *)dst,
++	case 8:__put_user_asm(*(const u64 *)src, (u64 __user *)dst,
+ 			      ret, "q", "", "er", 8);
+ 		return ret;
+ 	case 10:
+-		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
++		__put_user_asm(*(const u64 *)src, (u64 __user *)dst,
+ 			       ret, "q", "", "er", 10);
+ 		if (unlikely(ret))
+ 			return ret;
+ 		asm("":::"memory");
+-		__put_user_asm(4[(u16 *)src], 4 + (u16 __user *)dst,
++		__put_user_asm(4[(const u16 *)src], 4 + (u16 __user *)dst,
+ 			       ret, "w", "w", "ir", 2);
+ 		return ret;
+ 	case 16:
+-		__put_user_asm(*(u64 *)src, (u64 __user *)dst,
++		__put_user_asm(*(const u64 *)src, (u64 __user *)dst,
+ 			       ret, "q", "", "er", 16);
+ 		if (unlikely(ret))
+ 			return ret;
+ 		asm("":::"memory");
+-		__put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst,
++		__put_user_asm(1[(const u64 *)src], 1 + (u64 __user *)dst,
+ 			       ret, "q", "", "er", 8);
+ 		return ret;
+ 	default:
+-		return copy_user_generic((__force void *)dst, src, size);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		if ((unsigned long)dst < PAX_USER_SHADOW_BASE)
++			dst += PAX_USER_SHADOW_BASE;
++#endif
++
++		return copy_user_generic((__force_kernel void *)dst, src, size);
+ 	}
+ }
+ 
+ static __always_inline __must_check
+-int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
++unsigned long __copy_in_user(void __user *dst, const void __user *src, unsigned long size) __size_overflow(3);
++static __always_inline __must_check
++unsigned long __copy_in_user(void __user *dst, const void __user *src, unsigned long size)
+ {
+-	int ret = 0;
++	unsigned ret = 0;
+ 
+ 	might_fault();
+-	if (!__builtin_constant_p(size))
+-		return copy_user_generic((__force void *)dst,
+-					 (__force void *)src, size);
++
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_READ, src, size))
++		return size;
++	if (!__access_ok(VERIFY_WRITE, dst, size))
++		return size;
++#endif
++
++	if (!__builtin_constant_p(size)) {
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		if ((unsigned long)src < PAX_USER_SHADOW_BASE)
++			src += PAX_USER_SHADOW_BASE;
++		if ((unsigned long)dst < PAX_USER_SHADOW_BASE)
++			dst += PAX_USER_SHADOW_BASE;
++#endif
++
++		return copy_user_generic((__force_kernel void *)dst,
++					 (__force_kernel const void *)src, size);
++	}
+ 	switch (size) {
+ 	case 1: {
+ 		u8 tmp;
+-		__get_user_asm(tmp, (u8 __user *)src,
++		__get_user_asm(tmp, (const u8 __user *)src,
+ 			       ret, "b", "b", "=q", 1);
+ 		if (likely(!ret))
+ 			__put_user_asm(tmp, (u8 __user *)dst,
+@@ -176,7 +276,7 @@ int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
+ 	}
+ 	case 2: {
+ 		u16 tmp;
+-		__get_user_asm(tmp, (u16 __user *)src,
++		__get_user_asm(tmp, (const u16 __user *)src,
+ 			       ret, "w", "w", "=r", 2);
+ 		if (likely(!ret))
+ 			__put_user_asm(tmp, (u16 __user *)dst,
+@@ -186,7 +286,7 @@ int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
+ 
+ 	case 4: {
+ 		u32 tmp;
+-		__get_user_asm(tmp, (u32 __user *)src,
++		__get_user_asm(tmp, (const u32 __user *)src,
+ 			       ret, "l", "k", "=r", 4);
+ 		if (likely(!ret))
+ 			__put_user_asm(tmp, (u32 __user *)dst,
+@@ -195,7 +295,7 @@ int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
+ 	}
+ 	case 8: {
+ 		u64 tmp;
+-		__get_user_asm(tmp, (u64 __user *)src,
++		__get_user_asm(tmp, (const u64 __user *)src,
+ 			       ret, "q", "", "=r", 8);
+ 		if (likely(!ret))
+ 			__put_user_asm(tmp, (u64 __user *)dst,
+@@ -203,8 +303,16 @@ int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
+ 		return ret;
+ 	}
+ 	default:
+-		return copy_user_generic((__force void *)dst,
+-					 (__force void *)src, size);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		if ((unsigned long)src < PAX_USER_SHADOW_BASE)
++			src += PAX_USER_SHADOW_BASE;
++		if ((unsigned long)dst < PAX_USER_SHADOW_BASE)
++			dst += PAX_USER_SHADOW_BASE;
++#endif
++
++		return copy_user_generic((__force_kernel void *)dst,
++					 (__force_kernel const void *)src, size);
+ 	}
+ }
+ 
+@@ -215,39 +323,83 @@ __strncpy_from_user(char *dst, const char __user *src, long count);
+ __must_check long strnlen_user(const char __user *str, long n);
+ __must_check long __strnlen_user(const char __user *str, long n);
+ __must_check long strlen_user(const char __user *str);
+-__must_check unsigned long clear_user(void __user *mem, unsigned long len);
+-__must_check unsigned long __clear_user(void __user *mem, unsigned long len);
++__must_check unsigned long clear_user(void __user *mem, unsigned long len) __size_overflow(2);
++__must_check unsigned long __clear_user(void __user *mem, unsigned long len) __size_overflow(2);
+ 
+ static __must_check __always_inline int
+-__copy_from_user_inatomic(void *dst, const void __user *src, unsigned size)
++__copy_from_user_inatomic(void *dst, const void __user *src, unsigned long size) __size_overflow(3);
++static __must_check __always_inline int
++__copy_from_user_inatomic(void *dst, const void __user *src, unsigned long size)
+ {
+-	return copy_user_generic(dst, (__force const void *)src, size);
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_READ, src, size))
++		return size;
++
++	if ((unsigned long)src < PAX_USER_SHADOW_BASE)
++		src += PAX_USER_SHADOW_BASE;
++#endif
++
++	return copy_user_generic(dst, (__force_kernel const void *)src, size);
+ }
+ 
+-static __must_check __always_inline int
+-__copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
++static __must_check __always_inline unsigned long
++__copy_to_user_inatomic(void __user *dst, const void *src, unsigned long size) __size_overflow(3);
++static __must_check __always_inline unsigned long
++__copy_to_user_inatomic(void __user *dst, const void *src, unsigned long size)
+ {
+-	return copy_user_generic((__force void *)dst, src, size);
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_WRITE, dst, size))
++		return size;
++
++	if ((unsigned long)dst < PAX_USER_SHADOW_BASE)
++		dst += PAX_USER_SHADOW_BASE;
++#endif
++
++	return copy_user_generic((__force_kernel void *)dst, src, size);
+ }
+ 
+-extern long __copy_user_nocache(void *dst, const void __user *src,
+-				unsigned size, int zerorest);
++extern unsigned long __copy_user_nocache(void *dst, const void __user *src,
++				unsigned long size, int zerorest) __size_overflow(3);
+ 
+-static inline int
+-__copy_from_user_nocache(void *dst, const void __user *src, unsigned size)
++static inline unsigned long __copy_from_user_nocache(void *dst, const void __user *src, unsigned long size) __size_overflow(3);
++static inline unsigned long __copy_from_user_nocache(void *dst, const void __user *src, unsigned long size)
+ {
+ 	might_sleep();
++
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_READ, src, size))
++		return size;
++#endif
++
+ 	return __copy_user_nocache(dst, src, size, 1);
+ }
+ 
+-static inline int
+-__copy_from_user_inatomic_nocache(void *dst, const void __user *src,
+-				  unsigned size)
++static inline unsigned long __copy_from_user_inatomic_nocache(void *dst, const void __user *src,
++				  unsigned long size) __size_overflow(3);
++static inline unsigned long __copy_from_user_inatomic_nocache(void *dst, const void __user *src,
++				  unsigned long size)
+ {
++	if (size > INT_MAX)
++		return size;
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	if (!__access_ok(VERIFY_READ, src, size))
++		return size;
++#endif
++
+ 	return __copy_user_nocache(dst, src, size, 0);
+ }
+ 
+-unsigned long
+-copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest);
++extern unsigned long
++copy_user_handle_tail(char __user *to, char __user *from, unsigned long len, unsigned zerorest) __size_overflow(3);
+ 
+ #endif /* _ASM_X86_UACCESS_64_H */
+diff --git a/arch/x86/include/asm/vdso.h b/arch/x86/include/asm/vdso.h
+index bb05228..d763d5b 100644
+--- a/arch/x86/include/asm/vdso.h
++++ b/arch/x86/include/asm/vdso.h
+@@ -11,7 +11,7 @@ extern const char VDSO32_PRELINK[];
+ #define VDSO32_SYMBOL(base, name)					\
+ ({									\
+ 	extern const char VDSO32_##name[];				\
+-	(void *)(VDSO32_##name - VDSO32_PRELINK + (unsigned long)(base)); \
++	(void __user *)(VDSO32_##name - VDSO32_PRELINK + (unsigned long)(base)); \
+ })
+ #endif
+ 
+diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
+index 1971e65..1e3559b 100644
+--- a/arch/x86/include/asm/x86_init.h
++++ b/arch/x86/include/asm/x86_init.h
+@@ -28,7 +28,7 @@ struct x86_init_mpparse {
+ 	void (*mpc_oem_bus_info)(struct mpc_bus *m, char *name);
+ 	void (*find_smp_config)(void);
+ 	void (*get_smp_config)(unsigned int early);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_resources - platform specific resource related ops
+@@ -42,7 +42,7 @@ struct x86_init_resources {
+ 	void (*probe_roms)(void);
+ 	void (*reserve_resources)(void);
+ 	char *(*memory_setup)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_irqs - platform specific interrupt setup
+@@ -55,7 +55,7 @@ struct x86_init_irqs {
+ 	void (*pre_vector_init)(void);
+ 	void (*intr_init)(void);
+ 	void (*trap_init)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_oem - oem platform specific customizing functions
+@@ -65,7 +65,7 @@ struct x86_init_irqs {
+ struct x86_init_oem {
+ 	void (*arch_setup)(void);
+ 	void (*banner)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_mapping - platform specific initial kernel pagetable setup
+@@ -76,7 +76,7 @@ struct x86_init_oem {
+  */
+ struct x86_init_mapping {
+ 	void (*pagetable_reserve)(u64 start, u64 end);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_paging - platform specific paging functions
+@@ -86,7 +86,7 @@ struct x86_init_mapping {
+ struct x86_init_paging {
+ 	void (*pagetable_setup_start)(pgd_t *base);
+ 	void (*pagetable_setup_done)(pgd_t *base);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_timers - platform specific timer setup
+@@ -101,7 +101,7 @@ struct x86_init_timers {
+ 	void (*tsc_pre_init)(void);
+ 	void (*timer_init)(void);
+ 	void (*wallclock_init)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_iommu - platform specific iommu setup
+@@ -109,7 +109,7 @@ struct x86_init_timers {
+  */
+ struct x86_init_iommu {
+ 	int (*iommu_init)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_pci - platform specific pci init functions
+@@ -123,7 +123,7 @@ struct x86_init_pci {
+ 	int (*init)(void);
+ 	void (*init_irq)(void);
+ 	void (*fixup_irqs)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_init_ops - functions for platform specific setup
+@@ -139,7 +139,7 @@ struct x86_init_ops {
+ 	struct x86_init_timers		timers;
+ 	struct x86_init_iommu		iommu;
+ 	struct x86_init_pci		pci;
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_cpuinit_ops - platform specific cpu hotplug setups
+@@ -147,7 +147,7 @@ struct x86_init_ops {
+  */
+ struct x86_cpuinit_ops {
+ 	void (*setup_percpu_clockev)(void);
+-};
++} __no_const;
+ 
+ /**
+  * struct x86_platform_ops - platform specific runtime functions
+@@ -169,7 +169,7 @@ struct x86_platform_ops {
+ 	void (*nmi_init)(void);
+ 	unsigned char (*get_nmi_reason)(void);
+ 	int (*i8042_detect)(void);
+-};
++} __no_const;
+ 
+ struct pci_dev;
+ 
+@@ -177,7 +177,7 @@ struct x86_msi_ops {
+ 	int (*setup_msi_irqs)(struct pci_dev *dev, int nvec, int type);
+ 	void (*teardown_msi_irq)(unsigned int irq);
+ 	void (*teardown_msi_irqs)(struct pci_dev *dev);
+-};
++} __no_const;
+ 
+ extern struct x86_init_ops x86_init;
+ extern struct x86_cpuinit_ops x86_cpuinit;
+diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h
+index c6ce245..ffbdab7 100644
+--- a/arch/x86/include/asm/xsave.h
++++ b/arch/x86/include/asm/xsave.h
+@@ -65,6 +65,11 @@ static inline int xsave_user(struct xsave_struct __user *buf)
+ {
+ 	int err;
+ 
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++	if ((unsigned long)buf < PAX_USER_SHADOW_BASE)
++		buf = (struct xsave_struct __user *)((void __user*)buf + PAX_USER_SHADOW_BASE);
++#endif
++
+ 	/*
+ 	 * Clear the xsave header first, so that reserved fields are
+ 	 * initialized to zero.
+@@ -96,10 +101,15 @@ static inline int xsave_user(struct xsave_struct __user *buf)
+ static inline int xrestore_user(struct xsave_struct __user *buf, u64 mask)
+ {
+ 	int err;
+-	struct xsave_struct *xstate = ((__force struct xsave_struct *)buf);
++	struct xsave_struct *xstate = ((__force_kernel struct xsave_struct *)buf);
+ 	u32 lmask = mask;
+ 	u32 hmask = mask >> 32;
+ 
++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF)
++	if ((unsigned long)xstate < PAX_USER_SHADOW_BASE)
++		xstate = (struct xsave_struct *)((void *)xstate + PAX_USER_SHADOW_BASE);
++#endif
++
+ 	__asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n"
+ 			     "2:\n"
+ 			     ".section .fixup,\"ax\"\n"
+diff --git a/arch/x86/kernel/acpi/realmode/Makefile b/arch/x86/kernel/acpi/realmode/Makefile
+index 6a564ac..9b1340c 100644
+--- a/arch/x86/kernel/acpi/realmode/Makefile
++++ b/arch/x86/kernel/acpi/realmode/Makefile
+@@ -41,6 +41,9 @@ KBUILD_CFLAGS	:= $(LINUXINCLUDE) -g -Os -D_SETUP -D_WAKEUP -D__KERNEL__ \
+ 		   $(call cc-option, -fno-stack-protector) \
+ 		   $(call cc-option, -mpreferred-stack-boundary=2)
+ KBUILD_CFLAGS	+= $(call cc-option, -m32)
++ifdef CONSTIFY_PLUGIN
++KBUILD_CFLAGS	+= $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify
++endif
+ KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
+ GCOV_PROFILE := n
+ 
+diff --git a/arch/x86/kernel/acpi/realmode/wakeup.S b/arch/x86/kernel/acpi/realmode/wakeup.S
+index b4fd836..4358fe3 100644
+--- a/arch/x86/kernel/acpi/realmode/wakeup.S
++++ b/arch/x86/kernel/acpi/realmode/wakeup.S
+@@ -108,6 +108,9 @@ wakeup_code:
+ 	/* Do any other stuff... */
+ 
+ #ifndef CONFIG_64BIT
++	/* Recheck NX bit overrides (64bit path does this in trampoline */
++	call	verify_cpu
++
+ 	/* This could also be done in C code... */
+ 	movl	pmode_cr3, %eax
+ 	movl	%eax, %cr3
+@@ -131,6 +134,7 @@ wakeup_code:
+ 	movl	pmode_cr0, %eax
+ 	movl	%eax, %cr0
+ 	jmp	pmode_return
++# include "../../verify_cpu.S"
+ #else
+ 	pushw	$0
+ 	pushw	trampoline_segment
+diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
+index 103b6ab..2004d0a 100644
+--- a/arch/x86/kernel/acpi/sleep.c
++++ b/arch/x86/kernel/acpi/sleep.c
+@@ -94,8 +94,12 @@ int acpi_suspend_lowlevel(void)
+ 	header->trampoline_segment = trampoline_address() >> 4;
+ #ifdef CONFIG_SMP
+ 	stack_start = (unsigned long)temp_stack + sizeof(temp_stack);
++
++	pax_open_kernel();
+ 	early_gdt_descr.address =
+ 			(unsigned long)get_cpu_gdt_table(smp_processor_id());
++	pax_close_kernel();
++
+ 	initial_gs = per_cpu_offset(smp_processor_id());
+ #endif
+ 	initial_code = (unsigned long)wakeup_long64;
+diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S
+index 13ab720..95d5442 100644
+--- a/arch/x86/kernel/acpi/wakeup_32.S
++++ b/arch/x86/kernel/acpi/wakeup_32.S
+@@ -30,13 +30,11 @@ wakeup_pmode_return:
+ 	# and restore the stack ... but you need gdt for this to work
+ 	movl	saved_context_esp, %esp
+ 
+-	movl	%cs:saved_magic, %eax
+-	cmpl	$0x12345678, %eax
++	cmpl	$0x12345678, saved_magic
+ 	jne	bogus_magic
+ 
+ 	# jump to place where we left off
+-	movl	saved_eip, %eax
+-	jmp	*%eax
++	jmp	*(saved_eip)
+ 
+ bogus_magic:
+ 	jmp	bogus_magic
+diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
+index 1f84794..e23f862 100644
+--- a/arch/x86/kernel/alternative.c
++++ b/arch/x86/kernel/alternative.c
+@@ -276,6 +276,13 @@ void __init_or_module apply_alternatives(struct alt_instr *start,
+ 	 */
+ 	for (a = start; a < end; a++) {
+ 		instr = (u8 *)&a->instr_offset + a->instr_offset;
++
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC)
++		instr += ____LOAD_PHYSICAL_ADDR - LOAD_PHYSICAL_ADDR;
++		if (instr < (u8 *)_text || (u8 *)_einittext <= instr)
++			instr -= ____LOAD_PHYSICAL_ADDR - LOAD_PHYSICAL_ADDR;
++#endif
++
+ 		replacement = (u8 *)&a->repl_offset + a->repl_offset;
+ 		BUG_ON(a->replacementlen > a->instrlen);
+ 		BUG_ON(a->instrlen > sizeof(insnbuf));
+@@ -307,10 +314,16 @@ static void alternatives_smp_lock(const s32 *start, const s32 *end,
+ 	for (poff = start; poff < end; poff++) {
+ 		u8 *ptr = (u8 *)poff + *poff;
+ 
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC)
++		ptr += ____LOAD_PHYSICAL_ADDR - LOAD_PHYSICAL_ADDR;
++		if (ptr < (u8 *)_text || (u8 *)_einittext <= ptr)
++			ptr -= ____LOAD_PHYSICAL_ADDR - LOAD_PHYSICAL_ADDR;
++#endif
++
+ 		if (!*poff || ptr < text || ptr >= text_end)
+ 			continue;
+ 		/* turn DS segment override prefix into lock prefix */
+-		if (*ptr == 0x3e)
++		if (*ktla_ktva(ptr) == 0x3e)
+ 			text_poke(ptr, ((unsigned char []){0xf0}), 1);
+ 	};
+ 	mutex_unlock(&text_mutex);
+@@ -328,10 +341,16 @@ static void alternatives_smp_unlock(const s32 *start, const s32 *end,
+ 	for (poff = start; poff < end; poff++) {
+ 		u8 *ptr = (u8 *)poff + *poff;
+ 
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC)
++		ptr += ____LOAD_PHYSICAL_ADDR - LOAD_PHYSICAL_ADDR;
++		if (ptr < (u8 *)_text || (u8 *)_einittext <= ptr)
++			ptr -= ____LOAD_PHYSICAL_ADDR - LOAD_PHYSICAL_ADDR;
++#endif
++
+ 		if (!*poff || ptr < text || ptr >= text_end)
+ 			continue;
+ 		/* turn lock prefix into DS segment override prefix */
+-		if (*ptr == 0xf0)
++		if (*ktla_ktva(ptr) == 0xf0)
+ 			text_poke(ptr, ((unsigned char []){0x3E}), 1);
+ 	};
+ 	mutex_unlock(&text_mutex);
+@@ -500,7 +519,7 @@ void __init_or_module apply_paravirt(struct paravirt_patch_site *start,
+ 
+ 		BUG_ON(p->len > MAX_PATCH_LEN);
+ 		/* prep the buffer with the original instructions */
+-		memcpy(insnbuf, p->instr, p->len);
++		memcpy(insnbuf, ktla_ktva(p->instr), p->len);
+ 		used = pv_init_ops.patch(p->instrtype, p->clobbers, insnbuf,
+ 					 (unsigned long)p->instr, p->len);
+ 
+@@ -568,7 +587,7 @@ void __init alternative_instructions(void)
+ 	if (smp_alt_once)
+ 		free_init_pages("SMP alternatives",
+ 				(unsigned long)__smp_locks,
+-				(unsigned long)__smp_locks_end);
++				PAGE_ALIGN((unsigned long)__smp_locks_end));
+ 
+ 	restart_nmi();
+ }
+@@ -585,13 +604,17 @@ void __init alternative_instructions(void)
+  * instructions. And on the local CPU you need to be protected again NMI or MCE
+  * handlers seeing an inconsistent instruction while you patch.
+  */
+-void *__init_or_module text_poke_early(void *addr, const void *opcode,
++void *__kprobes text_poke_early(void *addr, const void *opcode,
+ 					      size_t len)
+ {
+ 	unsigned long flags;
+ 	local_irq_save(flags);
+-	memcpy(addr, opcode, len);
++
++	pax_open_kernel();
++	memcpy(ktla_ktva(addr), opcode, len);
+ 	sync_core();
++	pax_close_kernel();
++
+ 	local_irq_restore(flags);
+ 	/* Could also do a CLFLUSH here to speed up CPU recovery; but
+ 	   that causes hangs on some VIA CPUs. */
+@@ -613,36 +636,22 @@ void *__init_or_module text_poke_early(void *addr, const void *opcode,
+  */
+ void *__kprobes text_poke(void *addr, const void *opcode, size_t len)
+ {
+-	unsigned long flags;
+-	char *vaddr;
++	unsigned char *vaddr = ktla_ktva(addr);
+ 	struct page *pages[2];
+-	int i;
++	size_t i;
+ 
+ 	if (!core_kernel_text((unsigned long)addr)) {
+-		pages[0] = vmalloc_to_page(addr);
+-		pages[1] = vmalloc_to_page(addr + PAGE_SIZE);
++		pages[0] = vmalloc_to_page(vaddr);
++		pages[1] = vmalloc_to_page(vaddr + PAGE_SIZE);
+ 	} else {
+-		pages[0] = virt_to_page(addr);
++		pages[0] = virt_to_page(vaddr);
+ 		WARN_ON(!PageReserved(pages[0]));
+-		pages[1] = virt_to_page(addr + PAGE_SIZE);
++		pages[1] = virt_to_page(vaddr + PAGE_SIZE);
+ 	}
+ 	BUG_ON(!pages[0]);
+-	local_irq_save(flags);
+-	set_fixmap(FIX_TEXT_POKE0, page_to_phys(pages[0]));
+-	if (pages[1])
+-		set_fixmap(FIX_TEXT_POKE1, page_to_phys(pages[1]));
+-	vaddr = (char *)fix_to_virt(FIX_TEXT_POKE0);
+-	memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len);
+-	clear_fixmap(FIX_TEXT_POKE0);
+-	if (pages[1])
+-		clear_fixmap(FIX_TEXT_POKE1);
+-	local_flush_tlb();
+-	sync_core();
+-	/* Could also do a CLFLUSH here to speed up CPU recovery; but
+-	   that causes hangs on some VIA CPUs. */
++	text_poke_early(addr, opcode, len);
+ 	for (i = 0; i < len; i++)
+-		BUG_ON(((char *)addr)[i] != ((char *)opcode)[i]);
+-	local_irq_restore(flags);
++		BUG_ON((vaddr)[i] != ((const unsigned char *)opcode)[i]);
+ 	return addr;
+ }
+ 
+diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
+index f98d84c..e402a69 100644
+--- a/arch/x86/kernel/apic/apic.c
++++ b/arch/x86/kernel/apic/apic.c
+@@ -174,7 +174,7 @@ int first_system_vector = 0xfe;
+ /*
+  * Debug level, exported for io_apic.c
+  */
+-unsigned int apic_verbosity;
++int apic_verbosity;
+ 
+ int pic_mode;
+ 
+@@ -1853,7 +1853,7 @@ void smp_error_interrupt(struct pt_regs *regs)
+ 	apic_write(APIC_ESR, 0);
+ 	v1 = apic_read(APIC_ESR);
+ 	ack_APIC_irq();
+-	atomic_inc(&irq_err_count);
++	atomic_inc_unchecked(&irq_err_count);
+ 
+ 	apic_printk(APIC_DEBUG, KERN_DEBUG "APIC error on CPU%d: %02x(%02x)",
+ 		    smp_processor_id(), v0 , v1);
+diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
+index 6d939d7..0697fcc 100644
+--- a/arch/x86/kernel/apic/io_apic.c
++++ b/arch/x86/kernel/apic/io_apic.c
+@@ -1096,7 +1096,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin,
+ }
+ EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
+ 
+-void lock_vector_lock(void)
++void lock_vector_lock(void) __acquires(vector_lock)
+ {
+ 	/* Used to the online set of cpus does not change
+ 	 * during assign_irq_vector.
+@@ -1104,7 +1104,7 @@ void lock_vector_lock(void)
+ 	raw_spin_lock(&vector_lock);
+ }
+ 
+-void unlock_vector_lock(void)
++void unlock_vector_lock(void) __releases(vector_lock)
+ {
+ 	raw_spin_unlock(&vector_lock);
+ }
+@@ -2510,7 +2510,7 @@ static void ack_apic_edge(struct irq_data *data)
+ 	ack_APIC_irq();
+ }
+ 
+-atomic_t irq_mis_count;
++atomic_unchecked_t irq_mis_count;
+ 
+ static void ack_apic_level(struct irq_data *data)
+ {
+@@ -2576,7 +2576,7 @@ static void ack_apic_level(struct irq_data *data)
+ 	 * at the cpu.
+ 	 */
+ 	if (!(v & (1 << (i & 0x1f)))) {
+-		atomic_inc(&irq_mis_count);
++		atomic_inc_unchecked(&irq_mis_count);
+ 
+ 		eoi_ioapic_irq(irq, cfg);
+ 	}
+diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
+index a46bd38..6b906d7 100644
+--- a/arch/x86/kernel/apm_32.c
++++ b/arch/x86/kernel/apm_32.c
+@@ -411,7 +411,7 @@ static DEFINE_MUTEX(apm_mutex);
+  * This is for buggy BIOS's that refer to (real mode) segment 0x40
+  * even though they are called in protected mode.
+  */
+-static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092,
++static const struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4093,
+ 			(unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1);
+ 
+ static const char driver_version[] = "1.16ac";	/* no spaces */
+@@ -589,7 +589,10 @@ static long __apm_bios_call(void *_call)
+ 	BUG_ON(cpu != 0);
+ 	gdt = get_cpu_gdt_table(cpu);
+ 	save_desc_40 = gdt[0x40 / 8];
++
++	pax_open_kernel();
+ 	gdt[0x40 / 8] = bad_bios_desc;
++	pax_close_kernel();
+ 
+ 	apm_irq_save(flags);
+ 	APM_DO_SAVE_SEGS;
+@@ -598,7 +601,11 @@ static long __apm_bios_call(void *_call)
+ 			  &call->esi);
+ 	APM_DO_RESTORE_SEGS;
+ 	apm_irq_restore(flags);
++
++	pax_open_kernel();
+ 	gdt[0x40 / 8] = save_desc_40;
++	pax_close_kernel();
++
+ 	put_cpu();
+ 
+ 	return call->eax & 0xff;
+@@ -665,7 +672,10 @@ static long __apm_bios_call_simple(void *_call)
+ 	BUG_ON(cpu != 0);
+ 	gdt = get_cpu_gdt_table(cpu);
+ 	save_desc_40 = gdt[0x40 / 8];
++
++	pax_open_kernel();
+ 	gdt[0x40 / 8] = bad_bios_desc;
++	pax_close_kernel();
+ 
+ 	apm_irq_save(flags);
+ 	APM_DO_SAVE_SEGS;
+@@ -673,7 +683,11 @@ static long __apm_bios_call_simple(void *_call)
+ 					 &call->eax);
+ 	APM_DO_RESTORE_SEGS;
+ 	apm_irq_restore(flags);
++
++	pax_open_kernel();
+ 	gdt[0x40 / 8] = save_desc_40;
++	pax_close_kernel();
++
+ 	put_cpu();
+ 	return error;
+ }
+@@ -2347,12 +2361,15 @@ static int __init apm_init(void)
+ 	 * code to that CPU.
+ 	 */
+ 	gdt = get_cpu_gdt_table(0);
++
++	pax_open_kernel();
+ 	set_desc_base(&gdt[APM_CS >> 3],
+ 		 (unsigned long)__va((unsigned long)apm_info.bios.cseg << 4));
+ 	set_desc_base(&gdt[APM_CS_16 >> 3],
+ 		 (unsigned long)__va((unsigned long)apm_info.bios.cseg_16 << 4));
+ 	set_desc_base(&gdt[APM_DS >> 3],
+ 		 (unsigned long)__va((unsigned long)apm_info.bios.dseg << 4));
++	pax_close_kernel();
+ 
+ 	proc_create("apm", 0, NULL, &apm_file_ops);
+ 
+diff --git a/arch/x86/kernel/asm-offsets.c b/arch/x86/kernel/asm-offsets.c
+index 4f13faf..87db5d2 100644
+--- a/arch/x86/kernel/asm-offsets.c
++++ b/arch/x86/kernel/asm-offsets.c
+@@ -33,6 +33,8 @@ void common(void) {
+ 	OFFSET(TI_status, thread_info, status);
+ 	OFFSET(TI_addr_limit, thread_info, addr_limit);
+ 	OFFSET(TI_preempt_count, thread_info, preempt_count);
++	OFFSET(TI_lowest_stack, thread_info, lowest_stack);
++	DEFINE(TI_task_thread_sp0, offsetof(struct task_struct, thread.sp0) - offsetof(struct task_struct, tinfo));
+ 
+ 	BLANK();
+ 	OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx);
+@@ -53,8 +55,26 @@ void common(void) {
+ 	OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit);
+ 	OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0);
+ 	OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2);
++
++#ifdef CONFIG_PAX_KERNEXEC
++	OFFSET(PV_CPU_write_cr0, pv_cpu_ops, write_cr0);
+ #endif
+ 
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	OFFSET(PV_MMU_read_cr3, pv_mmu_ops, read_cr3);
++	OFFSET(PV_MMU_write_cr3, pv_mmu_ops, write_cr3);
++#ifdef CONFIG_X86_64
++	OFFSET(PV_MMU_set_pgd_batched, pv_mmu_ops, set_pgd_batched);
++#endif
++#endif
++
++#endif
++
++	BLANK();
++	DEFINE(PAGE_SIZE_asm, PAGE_SIZE);
++	DEFINE(PAGE_SHIFT_asm, PAGE_SHIFT);
++	DEFINE(THREAD_SIZE_asm, THREAD_SIZE);
++
+ #ifdef CONFIG_XEN
+ 	BLANK();
+ 	OFFSET(XEN_vcpu_info_mask, vcpu_info, evtchn_upcall_mask);
+diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
+index e72a119..6e2955d 100644
+--- a/arch/x86/kernel/asm-offsets_64.c
++++ b/arch/x86/kernel/asm-offsets_64.c
+@@ -69,6 +69,7 @@ int main(void)
+ 	BLANK();
+ #undef ENTRY
+ 
++	DEFINE(TSS_size, sizeof(struct tss_struct));
+ 	OFFSET(TSS_ist, tss_struct, x86_tss.ist);
+ 	BLANK();
+ 
+diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
+index 25f24dc..4094a7f 100644
+--- a/arch/x86/kernel/cpu/Makefile
++++ b/arch/x86/kernel/cpu/Makefile
+@@ -8,10 +8,6 @@ CFLAGS_REMOVE_common.o = -pg
+ CFLAGS_REMOVE_perf_event.o = -pg
+ endif
+ 
+-# Make sure load_percpu_segment has no stackprotector
+-nostackp := $(call cc-option, -fno-stack-protector)
+-CFLAGS_common.o		:= $(nostackp)
+-
+ obj-y			:= intel_cacheinfo.o scattered.o topology.o
+ obj-y			+= proc.o capflags.o powerflags.o common.o
+ obj-y			+= vmware.o hypervisor.o sched.o mshyperv.o
+diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
+index 0bab2b1..d0a1bf8 100644
+--- a/arch/x86/kernel/cpu/amd.c
++++ b/arch/x86/kernel/cpu/amd.c
+@@ -664,7 +664,7 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c,
+ 							unsigned int size)
+ {
+ 	/* AMD errata T13 (order #21922) */
+-	if ((c->x86 == 6)) {
++	if (c->x86 == 6) {
+ 		/* Duron Rev A0 */
+ 		if (c->x86_model == 3 && c->x86_mask == 0)
+ 			size = 64;
+diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
+index aa003b1..47ea638 100644
+--- a/arch/x86/kernel/cpu/common.c
++++ b/arch/x86/kernel/cpu/common.c
+@@ -84,60 +84,6 @@ static const struct cpu_dev __cpuinitconst default_cpu = {
+ 
+ static const struct cpu_dev *this_cpu __cpuinitdata = &default_cpu;
+ 
+-DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
+-#ifdef CONFIG_X86_64
+-	/*
+-	 * We need valid kernel segments for data and code in long mode too
+-	 * IRET will check the segment types  kkeil 2000/10/28
+-	 * Also sysret mandates a special GDT layout
+-	 *
+-	 * TLS descriptors are currently at a different place compared to i386.
+-	 * Hopefully nobody expects them at a fixed place (Wine?)
+-	 */
+-	[GDT_ENTRY_KERNEL32_CS]		= GDT_ENTRY_INIT(0xc09b, 0, 0xfffff),
+-	[GDT_ENTRY_KERNEL_CS]		= GDT_ENTRY_INIT(0xa09b, 0, 0xfffff),
+-	[GDT_ENTRY_KERNEL_DS]		= GDT_ENTRY_INIT(0xc093, 0, 0xfffff),
+-	[GDT_ENTRY_DEFAULT_USER32_CS]	= GDT_ENTRY_INIT(0xc0fb, 0, 0xfffff),
+-	[GDT_ENTRY_DEFAULT_USER_DS]	= GDT_ENTRY_INIT(0xc0f3, 0, 0xfffff),
+-	[GDT_ENTRY_DEFAULT_USER_CS]	= GDT_ENTRY_INIT(0xa0fb, 0, 0xfffff),
+-#else
+-	[GDT_ENTRY_KERNEL_CS]		= GDT_ENTRY_INIT(0xc09a, 0, 0xfffff),
+-	[GDT_ENTRY_KERNEL_DS]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
+-	[GDT_ENTRY_DEFAULT_USER_CS]	= GDT_ENTRY_INIT(0xc0fa, 0, 0xfffff),
+-	[GDT_ENTRY_DEFAULT_USER_DS]	= GDT_ENTRY_INIT(0xc0f2, 0, 0xfffff),
+-	/*
+-	 * Segments used for calling PnP BIOS have byte granularity.
+-	 * They code segments and data segments have fixed 64k limits,
+-	 * the transfer segment sizes are set at run time.
+-	 */
+-	/* 32-bit code */
+-	[GDT_ENTRY_PNPBIOS_CS32]	= GDT_ENTRY_INIT(0x409a, 0, 0xffff),
+-	/* 16-bit code */
+-	[GDT_ENTRY_PNPBIOS_CS16]	= GDT_ENTRY_INIT(0x009a, 0, 0xffff),
+-	/* 16-bit data */
+-	[GDT_ENTRY_PNPBIOS_DS]		= GDT_ENTRY_INIT(0x0092, 0, 0xffff),
+-	/* 16-bit data */
+-	[GDT_ENTRY_PNPBIOS_TS1]		= GDT_ENTRY_INIT(0x0092, 0, 0),
+-	/* 16-bit data */
+-	[GDT_ENTRY_PNPBIOS_TS2]		= GDT_ENTRY_INIT(0x0092, 0, 0),
+-	/*
+-	 * The APM segments have byte granularity and their bases
+-	 * are set at run time.  All have 64k limits.
+-	 */
+-	/* 32-bit code */
+-	[GDT_ENTRY_APMBIOS_BASE]	= GDT_ENTRY_INIT(0x409a, 0, 0xffff),
+-	/* 16-bit code */
+-	[GDT_ENTRY_APMBIOS_BASE+1]	= GDT_ENTRY_INIT(0x009a, 0, 0xffff),
+-	/* data */
+-	[GDT_ENTRY_APMBIOS_BASE+2]	= GDT_ENTRY_INIT(0x4092, 0, 0xffff),
+-
+-	[GDT_ENTRY_ESPFIX_SS]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
+-	[GDT_ENTRY_PERCPU]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
+-	GDT_STACK_CANARY_INIT
+-#endif
+-} };
+-EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
+-
+ static int __init x86_xsave_setup(char *s)
+ {
+ 	setup_clear_cpu_cap(X86_FEATURE_XSAVE);
+@@ -372,7 +318,7 @@ void switch_to_new_gdt(int cpu)
+ {
+ 	struct desc_ptr gdt_descr;
+ 
+-	gdt_descr.address = (long)get_cpu_gdt_table(cpu);
++	gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
+ 	gdt_descr.size = GDT_SIZE - 1;
+ 	load_gdt(&gdt_descr);
+ 	/* Reload the per-cpu base */
+@@ -844,6 +790,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+ 	/* Filter out anything that depends on CPUID levels we don't have */
+ 	filter_cpuid_features(c, true);
+ 
++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF))
++	setup_clear_cpu_cap(X86_FEATURE_SEP);
++#endif
++
+ 	/* If the model name is still unset, do table lookup. */
+ 	if (!c->x86_model_id[0]) {
+ 		const char *p;
+@@ -1024,6 +974,9 @@ static __init int setup_disablecpuid(char *arg)
+ }
+ __setup("clearcpuid=", setup_disablecpuid);
+ 
++DEFINE_PER_CPU(struct thread_info *, current_tinfo) = &init_task.tinfo;
++EXPORT_PER_CPU_SYMBOL(current_tinfo);
++
+ #ifdef CONFIG_X86_64
+ struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table };
+ 
+@@ -1039,7 +992,7 @@ DEFINE_PER_CPU(struct task_struct *, current_task) ____cacheline_aligned =
+ EXPORT_PER_CPU_SYMBOL(current_task);
+ 
+ DEFINE_PER_CPU(unsigned long, kernel_stack) =
+-	(unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE;
++	(unsigned long)&init_thread_union - 16 + THREAD_SIZE;
+ EXPORT_PER_CPU_SYMBOL(kernel_stack);
+ 
+ DEFINE_PER_CPU(char *, irq_stack_ptr) =
+@@ -1104,7 +1057,7 @@ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
+ {
+ 	memset(regs, 0, sizeof(struct pt_regs));
+ 	regs->fs = __KERNEL_PERCPU;
+-	regs->gs = __KERNEL_STACK_CANARY;
++	savesegment(gs, regs->gs);
+ 
+ 	return regs;
+ }
+@@ -1159,7 +1112,7 @@ void __cpuinit cpu_init(void)
+ 	int i;
+ 
+ 	cpu = stack_smp_processor_id();
+-	t = &per_cpu(init_tss, cpu);
++	t = init_tss + cpu;
+ 	oist = &per_cpu(orig_ist, cpu);
+ 
+ #ifdef CONFIG_NUMA
+@@ -1185,7 +1138,7 @@ void __cpuinit cpu_init(void)
+ 	switch_to_new_gdt(cpu);
+ 	loadsegment(fs, 0);
+ 
+-	load_idt((const struct desc_ptr *)&idt_descr);
++	load_idt(&idt_descr);
+ 
+ 	memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
+ 	syscall_init();
+@@ -1194,7 +1147,6 @@ void __cpuinit cpu_init(void)
+ 	wrmsrl(MSR_KERNEL_GS_BASE, 0);
+ 	barrier();
+ 
+-	x86_configure_nx();
+ 	if (cpu != 0)
+ 		enable_x2apic();
+ 
+@@ -1248,7 +1200,7 @@ void __cpuinit cpu_init(void)
+ {
+ 	int cpu = smp_processor_id();
+ 	struct task_struct *curr = current;
+-	struct tss_struct *t = &per_cpu(init_tss, cpu);
++	struct tss_struct *t = init_tss + cpu;
+ 	struct thread_struct *thread = &curr->thread;
+ 
+ 	if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) {
+diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
+index 5231312..a78a987 100644
+--- a/arch/x86/kernel/cpu/intel.c
++++ b/arch/x86/kernel/cpu/intel.c
+@@ -174,7 +174,7 @@ static void __cpuinit trap_init_f00f_bug(void)
+ 	 * Update the IDT descriptor and reload the IDT so that
+ 	 * it uses the read-only mapped virtual address.
+ 	 */
+-	idt_descr.address = fix_to_virt(FIX_F00F_IDT);
++	idt_descr.address = (struct desc_struct *)fix_to_virt(FIX_F00F_IDT);
+ 	load_idt(&idt_descr);
+ }
+ #endif
+diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
+index 319882e..993534e 100644
+--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
++++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
+@@ -173,6 +173,8 @@ static void raise_mce(struct mce *m)
+ 
+ /* Error injection interface */
+ static ssize_t mce_write(struct file *filp, const char __user *ubuf,
++			 size_t usize, loff_t *off) __size_overflow(3);
++static ssize_t mce_write(struct file *filp, const char __user *ubuf,
+ 			 size_t usize, loff_t *off)
+ {
+ 	struct mce m;
+diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
+index 2af127d..8ff7ac0 100644
+--- a/arch/x86/kernel/cpu/mcheck/mce.c
++++ b/arch/x86/kernel/cpu/mcheck/mce.c
+@@ -42,6 +42,7 @@
+ #include <asm/processor.h>
+ #include <asm/mce.h>
+ #include <asm/msr.h>
++#include <asm/local.h>
+ 
+ #include "mce-internal.h"
+ 
+@@ -202,7 +203,7 @@ static void print_mce(struct mce *m)
+ 			!(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
+ 				m->cs, m->ip);
+ 
+-		if (m->cs == __KERNEL_CS)
++		if (m->cs == __KERNEL_CS || m->cs == __KERNEXEC_KERNEL_CS)
+ 			print_symbol("{%s}", m->ip);
+ 		pr_cont("\n");
+ 	}
+@@ -235,10 +236,10 @@ static void print_mce(struct mce *m)
+ 
+ #define PANIC_TIMEOUT 5 /* 5 seconds */
+ 
+-static atomic_t mce_paniced;
++static atomic_unchecked_t mce_paniced;
+ 
+ static int fake_panic;
+-static atomic_t mce_fake_paniced;
++static atomic_unchecked_t mce_fake_paniced;
+ 
+ /* Panic in progress. Enable interrupts and wait for final IPI */
+ static void wait_for_panic(void)
+@@ -262,7 +263,7 @@ static void mce_panic(char *msg, struct mce *final, char *exp)
+ 		/*
+ 		 * Make sure only one CPU runs in machine check panic
+ 		 */
+-		if (atomic_inc_return(&mce_paniced) > 1)
++		if (atomic_inc_return_unchecked(&mce_paniced) > 1)
+ 			wait_for_panic();
+ 		barrier();
+ 
+@@ -270,7 +271,7 @@ static void mce_panic(char *msg, struct mce *final, char *exp)
+ 		console_verbose();
+ 	} else {
+ 		/* Don't log too much for fake panic */
+-		if (atomic_inc_return(&mce_fake_paniced) > 1)
++		if (atomic_inc_return_unchecked(&mce_fake_paniced) > 1)
+ 			return;
+ 	}
+ 	/* First print corrected ones that are still unlogged */
+@@ -610,7 +611,7 @@ static int mce_timed_out(u64 *t)
+ 	 * might have been modified by someone else.
+ 	 */
+ 	rmb();
+-	if (atomic_read(&mce_paniced))
++	if (atomic_read_unchecked(&mce_paniced))
+ 		wait_for_panic();
+ 	if (!monarch_timeout)
+ 		goto out;
+@@ -1398,7 +1399,7 @@ static void unexpected_machine_check(struct pt_regs *regs, long error_code)
+ }
+ 
+ /* Call the installed machine check handler for this CPU setup. */
+-void (*machine_check_vector)(struct pt_regs *, long error_code) =
++void (*machine_check_vector)(struct pt_regs *, long error_code) __read_only =
+ 						unexpected_machine_check;
+ 
+ /*
+@@ -1421,7 +1422,9 @@ void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c)
+ 		return;
+ 	}
+ 
++	pax_open_kernel();
+ 	machine_check_vector = do_machine_check;
++	pax_close_kernel();
+ 
+ 	__mcheck_cpu_init_generic();
+ 	__mcheck_cpu_init_vendor(c);
+@@ -1435,7 +1438,7 @@ void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c)
+  */
+ 
+ static DEFINE_SPINLOCK(mce_chrdev_state_lock);
+-static int mce_chrdev_open_count;	/* #times opened */
++static local_t mce_chrdev_open_count;	/* #times opened */
+ static int mce_chrdev_open_exclu;	/* already open exclusive? */
+ 
+ static int mce_chrdev_open(struct inode *inode, struct file *file)
+@@ -1443,7 +1446,7 @@ static int mce_chrdev_open(struct inode *inode, struct file *file)
+ 	spin_lock(&mce_chrdev_state_lock);
+ 
+ 	if (mce_chrdev_open_exclu ||
+-	    (mce_chrdev_open_count && (file->f_flags & O_EXCL))) {
++	    (local_read(&mce_chrdev_open_count) && (file->f_flags & O_EXCL))) {
+ 		spin_unlock(&mce_chrdev_state_lock);
+ 
+ 		return -EBUSY;
+@@ -1451,7 +1454,7 @@ static int mce_chrdev_open(struct inode *inode, struct file *file)
+ 
+ 	if (file->f_flags & O_EXCL)
+ 		mce_chrdev_open_exclu = 1;
+-	mce_chrdev_open_count++;
++	local_inc(&mce_chrdev_open_count);
+ 
+ 	spin_unlock(&mce_chrdev_state_lock);
+ 
+@@ -1462,7 +1465,7 @@ static int mce_chrdev_release(struct inode *inode, struct file *file)
+ {
+ 	spin_lock(&mce_chrdev_state_lock);
+ 
+-	mce_chrdev_open_count--;
++	local_dec(&mce_chrdev_open_count);
+ 	mce_chrdev_open_exclu = 0;
+ 
+ 	spin_unlock(&mce_chrdev_state_lock);
+@@ -2171,7 +2174,7 @@ struct dentry *mce_get_debugfs_dir(void)
+ static void mce_reset(void)
+ {
+ 	cpu_missing = 0;
+-	atomic_set(&mce_fake_paniced, 0);
++	atomic_set_unchecked(&mce_fake_paniced, 0);
+ 	atomic_set(&mce_executing, 0);
+ 	atomic_set(&mce_callin, 0);
+ 	atomic_set(&global_nwo, 0);
+diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c
+index 5c0e653..0882b0a 100644
+--- a/arch/x86/kernel/cpu/mcheck/p5.c
++++ b/arch/x86/kernel/cpu/mcheck/p5.c
+@@ -12,6 +12,7 @@
+ #include <asm/system.h>
+ #include <asm/mce.h>
+ #include <asm/msr.h>
++#include <asm/pgtable.h>
+ 
+ /* By default disabled */
+ int mce_p5_enabled __read_mostly;
+@@ -50,7 +51,9 @@ void intel_p5_mcheck_init(struct cpuinfo_x86 *c)
+ 	if (!cpu_has(c, X86_FEATURE_MCE))
+ 		return;
+ 
++	pax_open_kernel();
+ 	machine_check_vector = pentium_machine_check;
++	pax_close_kernel();
+ 	/* Make sure the vector pointer is visible before we enable MCEs: */
+ 	wmb();
+ 
+diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c
+index 54060f5..c1a7577 100644
+--- a/arch/x86/kernel/cpu/mcheck/winchip.c
++++ b/arch/x86/kernel/cpu/mcheck/winchip.c
+@@ -11,6 +11,7 @@
+ #include <asm/system.h>
+ #include <asm/mce.h>
+ #include <asm/msr.h>
++#include <asm/pgtable.h>
+ 
+ /* Machine check handler for WinChip C6: */
+ static void winchip_machine_check(struct pt_regs *regs, long error_code)
+@@ -24,7 +25,9 @@ void winchip_mcheck_init(struct cpuinfo_x86 *c)
+ {
+ 	u32 lo, hi;
+ 
++	pax_open_kernel();
+ 	machine_check_vector = winchip_machine_check;
++	pax_close_kernel();
+ 	/* Make sure the vector pointer is visible before we enable MCEs: */
+ 	wmb();
+ 
+diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c
+index 7928963..1b16001 100644
+--- a/arch/x86/kernel/cpu/mtrr/if.c
++++ b/arch/x86/kernel/cpu/mtrr/if.c
+@@ -91,6 +91,8 @@ mtrr_file_del(unsigned long base, unsigned long size,
+  *    "base=%Lx size=%Lx type=%s" or "disable=%d"
+  */
+ static ssize_t
++mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos) __size_overflow(3);
++static ssize_t
+ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos)
+ {
+ 	int i, err;
+diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
+index 6b96110..0da73eb 100644
+--- a/arch/x86/kernel/cpu/mtrr/main.c
++++ b/arch/x86/kernel/cpu/mtrr/main.c
+@@ -62,7 +62,7 @@ static DEFINE_MUTEX(mtrr_mutex);
+ u64 size_or_mask, size_and_mask;
+ static bool mtrr_aps_delayed_init;
+ 
+-static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM];
++static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM] __read_only;
+ 
+ const struct mtrr_ops *mtrr_if;
+ 
+diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
+index df5e41f..816c719 100644
+--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
++++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
+@@ -25,7 +25,7 @@ struct mtrr_ops {
+ 	int	(*validate_add_page)(unsigned long base, unsigned long size,
+ 				     unsigned int type);
+ 	int	(*have_wrcomb)(void);
+-};
++} __do_const;
+ 
+ extern int generic_get_free_region(unsigned long base, unsigned long size,
+ 				   int replace_reg);
+diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
+index 2bda212..78cc605 100644
+--- a/arch/x86/kernel/cpu/perf_event.c
++++ b/arch/x86/kernel/cpu/perf_event.c
+@@ -1529,7 +1529,7 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
+ 			break;
+ 
+ 		perf_callchain_store(entry, frame.return_address);
+-		fp = frame.next_frame;
++		fp = (const void __force_user *)frame.next_frame;
+ 	}
+ }
+ 
+diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
+index 13ad899..f642b9a 100644
+--- a/arch/x86/kernel/crash.c
++++ b/arch/x86/kernel/crash.c
+@@ -36,10 +36,8 @@ static void kdump_nmi_callback(int cpu, struct pt_regs *regs)
+ {
+ #ifdef CONFIG_X86_32
+ 	struct pt_regs fixed_regs;
+-#endif
+ 
+-#ifdef CONFIG_X86_32
+-	if (!user_mode_vm(regs)) {
++	if (!user_mode(regs)) {
+ 		crash_fixup_ss_esp(&fixed_regs, regs);
+ 		regs = &fixed_regs;
+ 	}
+diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c
+index 37250fe..bf2ec74 100644
+--- a/arch/x86/kernel/doublefault_32.c
++++ b/arch/x86/kernel/doublefault_32.c
+@@ -11,7 +11,7 @@
+ 
+ #define DOUBLEFAULT_STACKSIZE (1024)
+ static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE];
+-#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE)
++#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE-2)
+ 
+ #define ptr_ok(x) ((x) > PAGE_OFFSET && (x) < PAGE_OFFSET + MAXMEM)
+ 
+@@ -21,7 +21,7 @@ static void doublefault_fn(void)
+ 	unsigned long gdt, tss;
+ 
+ 	store_gdt(&gdt_desc);
+-	gdt = gdt_desc.address;
++	gdt = (unsigned long)gdt_desc.address;
+ 
+ 	printk(KERN_EMERG "PANIC: double fault, gdt at %08lx [%d bytes]\n", gdt, gdt_desc.size);
+ 
+@@ -58,10 +58,10 @@ struct tss_struct doublefault_tss __cacheline_aligned = {
+ 		/* 0x2 bit is always set */
+ 		.flags		= X86_EFLAGS_SF | 0x2,
+ 		.sp		= STACK_START,
+-		.es		= __USER_DS,
++		.es		= __KERNEL_DS,
+ 		.cs		= __KERNEL_CS,
+ 		.ss		= __KERNEL_DS,
+-		.ds		= __USER_DS,
++		.ds		= __KERNEL_DS,
+ 		.fs		= __KERNEL_PERCPU,
+ 
+ 		.__cr3		= __pa_nodebug(swapper_pg_dir),
+diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
+index 1aae78f..aab3a3d 100644
+--- a/arch/x86/kernel/dumpstack.c
++++ b/arch/x86/kernel/dumpstack.c
+@@ -2,6 +2,9 @@
+  *  Copyright (C) 1991, 1992  Linus Torvalds
+  *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+  */
++#ifdef CONFIG_GRKERNSEC_HIDESYM
++#define __INCLUDED_BY_HIDESYM 1
++#endif
+ #include <linux/kallsyms.h>
+ #include <linux/kprobes.h>
+ #include <linux/uaccess.h>
+@@ -35,9 +38,8 @@ void printk_address(unsigned long address, int reliable)
+ static void
+ print_ftrace_graph_addr(unsigned long addr, void *data,
+ 			const struct stacktrace_ops *ops,
+-			struct thread_info *tinfo, int *graph)
++			struct task_struct *task, int *graph)
+ {
+-	struct task_struct *task = tinfo->task;
+ 	unsigned long ret_addr;
+ 	int index = task->curr_ret_stack;
+ 
+@@ -58,7 +60,7 @@ print_ftrace_graph_addr(unsigned long addr, void *data,
+ static inline void
+ print_ftrace_graph_addr(unsigned long addr, void *data,
+ 			const struct stacktrace_ops *ops,
+-			struct thread_info *tinfo, int *graph)
++			struct task_struct *task, int *graph)
+ { }
+ #endif
+ 
+@@ -69,10 +71,8 @@ print_ftrace_graph_addr(unsigned long addr, void *data,
+  * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
+  */
+ 
+-static inline int valid_stack_ptr(struct thread_info *tinfo,
+-			void *p, unsigned int size, void *end)
++static inline int valid_stack_ptr(void *t, void *p, unsigned int size, void *end)
+ {
+-	void *t = tinfo;
+ 	if (end) {
+ 		if (p < end && p >= (end-THREAD_SIZE))
+ 			return 1;
+@@ -83,14 +83,14 @@ static inline int valid_stack_ptr(struct thread_info *tinfo,
+ }
+ 
+ unsigned long
+-print_context_stack(struct thread_info *tinfo,
++print_context_stack(struct task_struct *task, void *stack_start,
+ 		unsigned long *stack, unsigned long bp,
+ 		const struct stacktrace_ops *ops, void *data,
+ 		unsigned long *end, int *graph)
+ {
+ 	struct stack_frame *frame = (struct stack_frame *)bp;
+ 
+-	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
++	while (valid_stack_ptr(stack_start, stack, sizeof(*stack), end)) {
+ 		unsigned long addr;
+ 
+ 		addr = *stack;
+@@ -102,7 +102,7 @@ print_context_stack(struct thread_info *tinfo,
+ 			} else {
+ 				ops->address(data, addr, 0);
+ 			}
+-			print_ftrace_graph_addr(addr, data, ops, tinfo, graph);
++			print_ftrace_graph_addr(addr, data, ops, task, graph);
+ 		}
+ 		stack++;
+ 	}
+@@ -111,7 +111,7 @@ print_context_stack(struct thread_info *tinfo,
+ EXPORT_SYMBOL_GPL(print_context_stack);
+ 
+ unsigned long
+-print_context_stack_bp(struct thread_info *tinfo,
++print_context_stack_bp(struct task_struct *task, void *stack_start,
+ 		       unsigned long *stack, unsigned long bp,
+ 		       const struct stacktrace_ops *ops, void *data,
+ 		       unsigned long *end, int *graph)
+@@ -119,7 +119,7 @@ print_context_stack_bp(struct thread_info *tinfo,
+ 	struct stack_frame *frame = (struct stack_frame *)bp;
+ 	unsigned long *ret_addr = &frame->return_address;
+ 
+-	while (valid_stack_ptr(tinfo, ret_addr, sizeof(*ret_addr), end)) {
++	while (valid_stack_ptr(stack_start, ret_addr, sizeof(*ret_addr), end)) {
+ 		unsigned long addr = *ret_addr;
+ 
+ 		if (!__kernel_text_address(addr))
+@@ -128,7 +128,7 @@ print_context_stack_bp(struct thread_info *tinfo,
+ 		ops->address(data, addr, 1);
+ 		frame = frame->next_frame;
+ 		ret_addr = &frame->return_address;
+-		print_ftrace_graph_addr(addr, data, ops, tinfo, graph);
++		print_ftrace_graph_addr(addr, data, ops, task, graph);
+ 	}
+ 
+ 	return (unsigned long)frame;
+@@ -186,7 +186,7 @@ void dump_stack(void)
+ 
+ 	bp = stack_frame(current, NULL);
+ 	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
+-		current->pid, current->comm, print_tainted(),
++		task_pid_nr(current), current->comm, print_tainted(),
+ 		init_utsname()->release,
+ 		(int)strcspn(init_utsname()->version, " "),
+ 		init_utsname()->version);
+@@ -222,6 +222,8 @@ unsigned __kprobes long oops_begin(void)
+ }
+ EXPORT_SYMBOL_GPL(oops_begin);
+ 
++extern void gr_handle_kernel_exploit(void);
++
+ void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+ {
+ 	if (regs && kexec_should_crash(current))
+@@ -243,7 +245,10 @@ void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+ 		panic("Fatal exception in interrupt");
+ 	if (panic_on_oops)
+ 		panic("Fatal exception");
+-	do_exit(signr);
++
++	gr_handle_kernel_exploit();
++
++	do_group_exit(signr);
+ }
+ 
+ int __kprobes __die(const char *str, struct pt_regs *regs, long err)
+@@ -269,7 +274,7 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err)
+ 
+ 	show_registers(regs);
+ #ifdef CONFIG_X86_32
+-	if (user_mode_vm(regs)) {
++	if (user_mode(regs)) {
+ 		sp = regs->sp;
+ 		ss = regs->ss & 0xffff;
+ 	} else {
+@@ -297,7 +302,7 @@ void die(const char *str, struct pt_regs *regs, long err)
+ 	unsigned long flags = oops_begin();
+ 	int sig = SIGSEGV;
+ 
+-	if (!user_mode_vm(regs))
++	if (!user_mode(regs))
+ 		report_bug(regs->ip, regs);
+ 
+ 	if (__die(str, regs, err))
+diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c
+index c99f9ed..2a15d80 100644
+--- a/arch/x86/kernel/dumpstack_32.c
++++ b/arch/x86/kernel/dumpstack_32.c
+@@ -38,15 +38,13 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
+ 		bp = stack_frame(task, regs);
+ 
+ 	for (;;) {
+-		struct thread_info *context;
++		void *stack_start = (void *)((unsigned long)stack & ~(THREAD_SIZE-1));
+ 
+-		context = (struct thread_info *)
+-			((unsigned long)stack & (~(THREAD_SIZE - 1)));
+-		bp = ops->walk_stack(context, stack, bp, ops, data, NULL, &graph);
++		bp = ops->walk_stack(task, stack_start, stack, bp, ops, data, NULL, &graph);
+ 
+-		stack = (unsigned long *)context->previous_esp;
+-		if (!stack)
++		if (stack_start == task_stack_page(task))
+ 			break;
++		stack = *(unsigned long **)stack_start;
+ 		if (ops->stack(data, "IRQ") < 0)
+ 			break;
+ 		touch_nmi_watchdog();
+@@ -96,21 +94,22 @@ void show_registers(struct pt_regs *regs)
+ 	 * When in-kernel, we also print out the stack and code at the
+ 	 * time of the fault..
+ 	 */
+-	if (!user_mode_vm(regs)) {
++	if (!user_mode(regs)) {
+ 		unsigned int code_prologue = code_bytes * 43 / 64;
+ 		unsigned int code_len = code_bytes;
+ 		unsigned char c;
+ 		u8 *ip;
++		unsigned long cs_base = get_desc_base(&get_cpu_gdt_table(smp_processor_id())[(0xffff & regs->cs) >> 3]);
+ 
+ 		printk(KERN_EMERG "Stack:\n");
+ 		show_stack_log_lvl(NULL, regs, &regs->sp, 0, KERN_EMERG);
+ 
+ 		printk(KERN_EMERG "Code: ");
+ 
+-		ip = (u8 *)regs->ip - code_prologue;
++		ip = (u8 *)regs->ip - code_prologue + cs_base;
+ 		if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
+ 			/* try starting at IP */
+-			ip = (u8 *)regs->ip;
++			ip = (u8 *)regs->ip + cs_base;
+ 			code_len = code_len - code_prologue + 1;
+ 		}
+ 		for (i = 0; i < code_len; i++, ip++) {
+@@ -119,7 +118,7 @@ void show_registers(struct pt_regs *regs)
+ 				printk(KERN_CONT " Bad EIP value.");
+ 				break;
+ 			}
+-			if (ip == (u8 *)regs->ip)
++			if (ip == (u8 *)regs->ip + cs_base)
+ 				printk(KERN_CONT "<%02x> ", c);
+ 			else
+ 				printk(KERN_CONT "%02x ", c);
+@@ -132,6 +131,7 @@ int is_valid_bugaddr(unsigned long ip)
+ {
+ 	unsigned short ud2;
+ 
++	ip = ktla_ktva(ip);
+ 	if (ip < PAGE_OFFSET)
+ 		return 0;
+ 	if (probe_kernel_address((unsigned short *)ip, ud2))
+@@ -139,3 +139,15 @@ int is_valid_bugaddr(unsigned long ip)
+ 
+ 	return ud2 == 0x0b0f;
+ }
++
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++void pax_check_alloca(unsigned long size)
++{
++	unsigned long sp = (unsigned long)&sp, stack_left;
++
++	/* all kernel stacks are of the same size */
++	stack_left = sp & (THREAD_SIZE - 1);
++	BUG_ON(stack_left < 256 || size >= stack_left - 256);
++}
++EXPORT_SYMBOL(pax_check_alloca);
++#endif
+diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c
+index 6d728d9..279514e 100644
+--- a/arch/x86/kernel/dumpstack_64.c
++++ b/arch/x86/kernel/dumpstack_64.c
+@@ -119,9 +119,9 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
+ 	unsigned long *irq_stack_end =
+ 		(unsigned long *)per_cpu(irq_stack_ptr, cpu);
+ 	unsigned used = 0;
+-	struct thread_info *tinfo;
+ 	int graph = 0;
+ 	unsigned long dummy;
++	void *stack_start;
+ 
+ 	if (!task)
+ 		task = current;
+@@ -142,10 +142,10 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
+ 	 * current stack address. If the stacks consist of nested
+ 	 * exceptions
+ 	 */
+-	tinfo = task_thread_info(task);
+ 	for (;;) {
+ 		char *id;
+ 		unsigned long *estack_end;
++
+ 		estack_end = in_exception_stack(cpu, (unsigned long)stack,
+ 						&used, &id);
+ 
+@@ -153,7 +153,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
+ 			if (ops->stack(data, id) < 0)
+ 				break;
+ 
+-			bp = ops->walk_stack(tinfo, stack, bp, ops,
++			bp = ops->walk_stack(task, estack_end - EXCEPTION_STKSZ, stack, bp, ops,
+ 					     data, estack_end, &graph);
+ 			ops->stack(data, "<EOE>");
+ 			/*
+@@ -172,7 +172,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
+ 			if (in_irq_stack(stack, irq_stack, irq_stack_end)) {
+ 				if (ops->stack(data, "IRQ") < 0)
+ 					break;
+-				bp = ops->walk_stack(tinfo, stack, bp,
++				bp = ops->walk_stack(task, irq_stack, stack, bp,
+ 					ops, data, irq_stack_end, &graph);
+ 				/*
+ 				 * We link to the next stack (which would be
+@@ -191,7 +191,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
+ 	/*
+ 	 * This handles the process stack:
+ 	 */
+-	bp = ops->walk_stack(tinfo, stack, bp, ops, data, NULL, &graph);
++	stack_start = (void *)((unsigned long)stack & ~(THREAD_SIZE-1));
++	bp = ops->walk_stack(task, stack_start, stack, bp, ops, data, NULL, &graph);
+ 	put_cpu();
+ }
+ EXPORT_SYMBOL(dump_trace);
+@@ -305,3 +306,50 @@ int is_valid_bugaddr(unsigned long ip)
+ 
+ 	return ud2 == 0x0b0f;
+ }
++
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++void pax_check_alloca(unsigned long size)
++{
++	unsigned long sp = (unsigned long)&sp, stack_start, stack_end;
++	unsigned cpu, used;
++	char *id;
++
++	/* check the process stack first */
++	stack_start = (unsigned long)task_stack_page(current);
++	stack_end = stack_start + THREAD_SIZE;
++	if (likely(stack_start <= sp && sp < stack_end)) {
++		unsigned long stack_left = sp & (THREAD_SIZE - 1);
++		BUG_ON(stack_left < 256 || size >= stack_left - 256);
++		return;
++	}
++
++	cpu = get_cpu();
++
++	/* check the irq stacks */
++	stack_end = (unsigned long)per_cpu(irq_stack_ptr, cpu);
++	stack_start = stack_end - IRQ_STACK_SIZE;
++	if (stack_start <= sp && sp < stack_end) {
++		unsigned long stack_left = sp & (IRQ_STACK_SIZE - 1);
++		put_cpu();
++		BUG_ON(stack_left < 256 || size >= stack_left - 256);
++		return;
++	}
++
++	/* check the exception stacks */
++	used = 0;
++	stack_end = (unsigned long)in_exception_stack(cpu, sp, &used, &id);
++	stack_start = stack_end - EXCEPTION_STKSZ;
++	if (stack_end && stack_start <= sp && sp < stack_end) {
++		unsigned long stack_left = sp & (EXCEPTION_STKSZ - 1);
++		put_cpu();
++		BUG_ON(stack_left < 256 || size >= stack_left - 256);
++		return;
++	}
++
++	put_cpu();
++
++	/* unknown stack */
++	BUG();
++}
++EXPORT_SYMBOL(pax_check_alloca);
++#endif
+diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
+index cd28a35..c72ed9a 100644
+--- a/arch/x86/kernel/early_printk.c
++++ b/arch/x86/kernel/early_printk.c
+@@ -7,6 +7,7 @@
+ #include <linux/pci_regs.h>
+ #include <linux/pci_ids.h>
+ #include <linux/errno.h>
++#include <linux/sched.h>
+ #include <asm/io.h>
+ #include <asm/processor.h>
+ #include <asm/fcntl.h>
+diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
+index f3f6f53..b0cbdf9 100644
+--- a/arch/x86/kernel/entry_32.S
++++ b/arch/x86/kernel/entry_32.S
+@@ -99,12 +99,6 @@
+ #endif
+ .endm
+ 
+-#ifdef CONFIG_VM86
+-#define resume_userspace_sig	check_userspace
+-#else
+-#define resume_userspace_sig	resume_userspace
+-#endif
+-
+ /*
+  * User gs save/restore
+  *
+@@ -186,13 +180,146 @@
+ 	/*CFI_REL_OFFSET gs, PT_GS*/
+ .endm
+ .macro SET_KERNEL_GS reg
++
++#ifdef CONFIG_CC_STACKPROTECTOR
+ 	movl $(__KERNEL_STACK_CANARY), \reg
++#elif defined(CONFIG_PAX_MEMORY_UDEREF)
++	movl $(__USER_DS), \reg
++#else
++	xorl \reg, \reg
++#endif
++
+ 	movl \reg, %gs
+ .endm
+ 
+ #endif	/* CONFIG_X86_32_LAZY_GS */
+ 
+-.macro SAVE_ALL
++.macro pax_enter_kernel
++#ifdef CONFIG_PAX_KERNEXEC
++	call pax_enter_kernel
++#endif
++.endm
++
++.macro pax_exit_kernel
++#ifdef CONFIG_PAX_KERNEXEC
++	call pax_exit_kernel
++#endif
++.endm
++
++#ifdef CONFIG_PAX_KERNEXEC
++ENTRY(pax_enter_kernel)
++#ifdef CONFIG_PARAVIRT
++	pushl %eax
++	pushl %ecx
++	call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0)
++	mov %eax, %esi
++#else
++	mov %cr0, %esi
++#endif
++	bts $16, %esi
++	jnc 1f
++	mov %cs, %esi
++	cmp $__KERNEL_CS, %esi
++	jz 3f
++	ljmp $__KERNEL_CS, $3f
++1:	ljmp $__KERNEXEC_KERNEL_CS, $2f
++2:
++#ifdef CONFIG_PARAVIRT
++	mov %esi, %eax
++	call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0)
++#else
++	mov %esi, %cr0
++#endif
++3:
++#ifdef CONFIG_PARAVIRT
++	popl %ecx
++	popl %eax
++#endif
++	ret
++ENDPROC(pax_enter_kernel)
++
++ENTRY(pax_exit_kernel)
++#ifdef CONFIG_PARAVIRT
++	pushl %eax
++	pushl %ecx
++#endif
++	mov %cs, %esi
++	cmp $__KERNEXEC_KERNEL_CS, %esi
++	jnz 2f
++#ifdef CONFIG_PARAVIRT
++	call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0);
++	mov %eax, %esi
++#else
++	mov %cr0, %esi
++#endif
++	btr $16, %esi
++	ljmp $__KERNEL_CS, $1f
++1:
++#ifdef CONFIG_PARAVIRT
++	mov %esi, %eax
++	call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0);
++#else
++	mov %esi, %cr0
++#endif
++2:
++#ifdef CONFIG_PARAVIRT
++	popl %ecx
++	popl %eax
++#endif
++	ret
++ENDPROC(pax_exit_kernel)
++#endif
++
++.macro pax_erase_kstack
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++	call pax_erase_kstack
++#endif
++.endm
++
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++/*
++ * ebp: thread_info
++ * ecx, edx: can be clobbered
++ */
++ENTRY(pax_erase_kstack)
++	pushl %edi
++	pushl %eax
++
++	mov TI_lowest_stack(%ebp), %edi
++	mov $-0xBEEF, %eax
++	std
++
++1:	mov %edi, %ecx
++	and $THREAD_SIZE_asm - 1, %ecx
++	shr $2, %ecx
++	repne scasl
++	jecxz 2f
++
++	cmp $2*16, %ecx
++	jc 2f
++
++	mov $2*16, %ecx
++	repe scasl
++	jecxz 2f
++	jne 1b
++
++2:	cld
++	mov %esp, %ecx
++	sub %edi, %ecx
++	shr $2, %ecx
++	rep stosl
++
++	mov TI_task_thread_sp0(%ebp), %edi
++	sub $128, %edi
++	mov %edi, TI_lowest_stack(%ebp)
++
++	popl %eax
++	popl %edi
++	ret
++ENDPROC(pax_erase_kstack)
++#endif
++
++.macro __SAVE_ALL _DS
+ 	cld
+ 	PUSH_GS
+ 	pushl_cfi %fs
+@@ -215,7 +342,7 @@
+ 	CFI_REL_OFFSET ecx, 0
+ 	pushl_cfi %ebx
+ 	CFI_REL_OFFSET ebx, 0
+-	movl $(__USER_DS), %edx
++	movl $\_DS, %edx
+ 	movl %edx, %ds
+ 	movl %edx, %es
+ 	movl $(__KERNEL_PERCPU), %edx
+@@ -223,6 +350,15 @@
+ 	SET_KERNEL_GS %edx
+ .endm
+ 
++.macro SAVE_ALL
++#if defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF)
++	__SAVE_ALL __KERNEL_DS
++	pax_enter_kernel
++#else
++	__SAVE_ALL __USER_DS
++#endif
++.endm
++
+ .macro RESTORE_INT_REGS
+ 	popl_cfi %ebx
+ 	CFI_RESTORE ebx
+@@ -308,7 +444,7 @@ ENTRY(ret_from_fork)
+ 	popfl_cfi
+ 	jmp syscall_exit
+ 	CFI_ENDPROC
+-END(ret_from_fork)
++ENDPROC(ret_from_fork)
+ 
+ /*
+  * Interrupt exit functions should be protected against kprobes
+@@ -328,12 +464,29 @@ ret_from_exception:
+ 	preempt_stop(CLBR_ANY)
+ ret_from_intr:
+ 	GET_THREAD_INFO(%ebp)
+-check_userspace:
++resume_userspace_sig:
++#ifdef CONFIG_VM86
+ 	movl PT_EFLAGS(%esp), %eax	# mix EFLAGS and CS
+ 	movb PT_CS(%esp), %al
+ 	andl $(X86_EFLAGS_VM | SEGMENT_RPL_MASK), %eax
++#else
++	/*
++	 * We can be coming here from a syscall done in the kernel space,
++	 * e.g. a failed kernel_execve().
++	 */
++	movl PT_CS(%esp), %eax
++	andl $SEGMENT_RPL_MASK, %eax
++#endif
+ 	cmpl $USER_RPL, %eax
++
++#ifdef CONFIG_PAX_KERNEXEC
++	jae resume_userspace
++
++	PAX_EXIT_KERNEL
++	jmp resume_kernel
++#else
+ 	jb resume_kernel		# not returning to v8086 or userspace
++#endif
+ 
+ ENTRY(resume_userspace)
+ 	LOCKDEP_SYS_EXIT
+@@ -345,8 +498,8 @@ ENTRY(resume_userspace)
+ 	andl $_TIF_WORK_MASK, %ecx	# is there any work to be done on
+ 					# int/exception return?
+ 	jne work_pending
+-	jmp restore_all
+-END(ret_from_exception)
++	jmp restore_all_pax
++ENDPROC(ret_from_exception)
+ 
+ #ifdef CONFIG_PREEMPT
+ ENTRY(resume_kernel)
+@@ -361,7 +514,7 @@ need_resched:
+ 	jz restore_all
+ 	call preempt_schedule_irq
+ 	jmp need_resched
+-END(resume_kernel)
++ENDPROC(resume_kernel)
+ #endif
+ 	CFI_ENDPROC
+ /*
+@@ -395,23 +548,34 @@ sysenter_past_esp:
+ 	/*CFI_REL_OFFSET cs, 0*/
+ 	/*
+ 	 * Push current_thread_info()->sysenter_return to the stack.
+-	 * A tiny bit of offset fixup is necessary - 4*4 means the 4 words
+-	 * pushed above; +8 corresponds to copy_thread's esp0 setting.
+ 	 */
+-	pushl_cfi ((TI_sysenter_return)-THREAD_SIZE+8+4*4)(%esp)
++	pushl_cfi $0
+ 	CFI_REL_OFFSET eip, 0
+ 
+ 	pushl_cfi %eax
+ 	SAVE_ALL
++	GET_THREAD_INFO(%ebp)
++	movl TI_sysenter_return(%ebp),%ebp
++	movl %ebp,PT_EIP(%esp)
+ 	ENABLE_INTERRUPTS(CLBR_NONE)
+ 
+ /*
+  * Load the potential sixth argument from user stack.
+  * Careful about security.
+  */
++	movl PT_OLDESP(%esp),%ebp
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	mov PT_OLDSS(%esp),%ds
++1:	movl %ds:(%ebp),%ebp
++	push %ss
++	pop %ds
++#else
+ 	cmpl $__PAGE_OFFSET-3,%ebp
+ 	jae syscall_fault
+ 1:	movl (%ebp),%ebp
++#endif
++
+ 	movl %ebp,PT_EBP(%esp)
+ .section __ex_table,"a"
+ 	.align 4
+@@ -434,12 +598,24 @@ sysenter_do_call:
+ 	testl $_TIF_ALLWORK_MASK, %ecx
+ 	jne sysexit_audit
+ sysenter_exit:
++
++#ifdef CONFIG_PAX_RANDKSTACK
++	pushl_cfi %eax
++	movl %esp, %eax
++	call pax_randomize_kstack
++	popl_cfi %eax
++#endif
++
++	pax_erase_kstack
++
+ /* if something modifies registers it must also disable sysexit */
+ 	movl PT_EIP(%esp), %edx
+ 	movl PT_OLDESP(%esp), %ecx
+ 	xorl %ebp,%ebp
+ 	TRACE_IRQS_ON
+ 1:	mov  PT_FS(%esp), %fs
++2:	mov  PT_DS(%esp), %ds
++3:	mov  PT_ES(%esp), %es
+ 	PTGS_TO_GS
+ 	ENABLE_INTERRUPTS_SYSEXIT
+ 
+@@ -456,6 +632,9 @@ sysenter_audit:
+ 	movl %eax,%edx			/* 2nd arg: syscall number */
+ 	movl $AUDIT_ARCH_I386,%eax	/* 1st arg: audit arch */
+ 	call audit_syscall_entry
++
++	pax_erase_kstack
++
+ 	pushl_cfi %ebx
+ 	movl PT_EAX(%esp),%eax		/* reload syscall number */
+ 	jmp sysenter_do_call
+@@ -482,11 +661,17 @@ sysexit_audit:
+ 
+ 	CFI_ENDPROC
+ .pushsection .fixup,"ax"
+-2:	movl $0,PT_FS(%esp)
++4:	movl $0,PT_FS(%esp)
++	jmp 1b
++5:	movl $0,PT_DS(%esp)
++	jmp 1b
++6:	movl $0,PT_ES(%esp)
+ 	jmp 1b
+ .section __ex_table,"a"
+ 	.align 4
+-	.long 1b,2b
++	.long 1b,4b
++	.long 2b,5b
++	.long 3b,6b
+ .popsection
+ 	PTGS_TO_GS_EX
+ ENDPROC(ia32_sysenter_target)
+@@ -519,6 +704,15 @@ syscall_exit:
+ 	testl $_TIF_ALLWORK_MASK, %ecx	# current->work
+ 	jne syscall_exit_work
+ 
++restore_all_pax:
++
++#ifdef CONFIG_PAX_RANDKSTACK
++	movl %esp, %eax
++	call pax_randomize_kstack
++#endif
++
++	pax_erase_kstack
++
+ restore_all:
+ 	TRACE_IRQS_IRET
+ restore_all_notrace:
+@@ -578,14 +772,34 @@ ldt_ss:
+  * compensating for the offset by changing to the ESPFIX segment with
+  * a base address that matches for the difference.
+  */
+-#define GDT_ESPFIX_SS PER_CPU_VAR(gdt_page) + (GDT_ENTRY_ESPFIX_SS * 8)
++#define GDT_ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)(%ebx)
+ 	mov %esp, %edx			/* load kernel esp */
+ 	mov PT_OLDESP(%esp), %eax	/* load userspace esp */
+ 	mov %dx, %ax			/* eax: new kernel esp */
+ 	sub %eax, %edx			/* offset (low word is 0) */
++#ifdef CONFIG_SMP
++	movl PER_CPU_VAR(cpu_number), %ebx
++	shll $PAGE_SHIFT_asm, %ebx
++	addl $cpu_gdt_table, %ebx
++#else
++	movl $cpu_gdt_table, %ebx
++#endif
+ 	shr $16, %edx
+-	mov %dl, GDT_ESPFIX_SS + 4 /* bits 16..23 */
+-	mov %dh, GDT_ESPFIX_SS + 7 /* bits 24..31 */
++
++#ifdef CONFIG_PAX_KERNEXEC
++	mov %cr0, %esi
++	btr $16, %esi
++	mov %esi, %cr0
++#endif
++
++	mov %dl, 4 + GDT_ESPFIX_SS /* bits 16..23 */
++	mov %dh, 7 + GDT_ESPFIX_SS /* bits 24..31 */
++
++#ifdef CONFIG_PAX_KERNEXEC
++	bts $16, %esi
++	mov %esi, %cr0
++#endif
++
+ 	pushl_cfi $__ESPFIX_SS
+ 	pushl_cfi %eax			/* new kernel esp */
+ 	/* Disable interrupts, but do not irqtrace this section: we
+@@ -614,34 +828,28 @@ work_resched:
+ 	movl TI_flags(%ebp), %ecx
+ 	andl $_TIF_WORK_MASK, %ecx	# is there any work to be done other
+ 					# than syscall tracing?
+-	jz restore_all
++	jz restore_all_pax
+ 	testb $_TIF_NEED_RESCHED, %cl
+ 	jnz work_resched
+ 
+ work_notifysig:				# deal with pending signals and
+ 					# notify-resume requests
++	movl %esp, %eax
+ #ifdef CONFIG_VM86
+ 	testl $X86_EFLAGS_VM, PT_EFLAGS(%esp)
+-	movl %esp, %eax
+-	jne work_notifysig_v86		# returning to kernel-space or
++	jz 1f				# returning to kernel-space or
+ 					# vm86-space
+-	xorl %edx, %edx
+-	call do_notify_resume
+-	jmp resume_userspace_sig
+ 
+-	ALIGN
+-work_notifysig_v86:
+ 	pushl_cfi %ecx			# save ti_flags for do_notify_resume
+ 	call save_v86_state		# %eax contains pt_regs pointer
+ 	popl_cfi %ecx
+ 	movl %eax, %esp
+-#else
+-	movl %esp, %eax
++1:
+ #endif
+ 	xorl %edx, %edx
+ 	call do_notify_resume
+ 	jmp resume_userspace_sig
+-END(work_pending)
++ENDPROC(work_pending)
+ 
+ 	# perform syscall exit tracing
+ 	ALIGN
+@@ -649,11 +857,14 @@ syscall_trace_entry:
+ 	movl $-ENOSYS,PT_EAX(%esp)
+ 	movl %esp, %eax
+ 	call syscall_trace_enter
++
++	pax_erase_kstack
++
+ 	/* What it returned is what we'll actually use.  */
+ 	cmpl $(nr_syscalls), %eax
+ 	jnae syscall_call
+ 	jmp syscall_exit
+-END(syscall_trace_entry)
++ENDPROC(syscall_trace_entry)
+ 
+ 	# perform syscall exit tracing
+ 	ALIGN
+@@ -666,20 +877,24 @@ syscall_exit_work:
+ 	movl %esp, %eax
+ 	call syscall_trace_leave
+ 	jmp resume_userspace
+-END(syscall_exit_work)
++ENDPROC(syscall_exit_work)
+ 	CFI_ENDPROC
+ 
+ 	RING0_INT_FRAME			# can't unwind into user space anyway
+ syscall_fault:
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	push %ss
++	pop %ds
++#endif
+ 	GET_THREAD_INFO(%ebp)
+ 	movl $-EFAULT,PT_EAX(%esp)
+ 	jmp resume_userspace
+-END(syscall_fault)
++ENDPROC(syscall_fault)
+ 
+ syscall_badsys:
+ 	movl $-ENOSYS,PT_EAX(%esp)
+ 	jmp resume_userspace
+-END(syscall_badsys)
++ENDPROC(syscall_badsys)
+ 	CFI_ENDPROC
+ /*
+  * End of kprobes section
+@@ -753,6 +968,36 @@ ptregs_clone:
+ 	CFI_ENDPROC
+ ENDPROC(ptregs_clone)
+ 
++	ALIGN;
++ENTRY(kernel_execve)
++	CFI_STARTPROC
++	pushl_cfi %ebp
++	sub $PT_OLDSS+4,%esp
++	pushl_cfi %edi
++	pushl_cfi %ecx
++	pushl_cfi %eax
++	lea 3*4(%esp),%edi
++	mov $PT_OLDSS/4+1,%ecx
++	xorl %eax,%eax
++	rep stosl
++	popl_cfi %eax
++	popl_cfi %ecx
++	popl_cfi %edi
++	movl $X86_EFLAGS_IF,PT_EFLAGS(%esp)
++	pushl_cfi %esp
++	call sys_execve
++	add $4,%esp
++	CFI_ADJUST_CFA_OFFSET -4
++	GET_THREAD_INFO(%ebp)
++	test %eax,%eax
++	jz syscall_exit
++	add $PT_OLDSS+4,%esp
++	CFI_ADJUST_CFA_OFFSET -PT_OLDSS-4
++	popl_cfi %ebp
++	ret
++	CFI_ENDPROC
++ENDPROC(kernel_execve)
++
+ .macro FIXUP_ESPFIX_STACK
+ /*
+  * Switch back for ESPFIX stack to the normal zerobased stack
+@@ -762,8 +1007,15 @@ ENDPROC(ptregs_clone)
+  * normal stack and adjusts ESP with the matching offset.
+  */
+ 	/* fixup the stack */
+-	mov GDT_ESPFIX_SS + 4, %al /* bits 16..23 */
+-	mov GDT_ESPFIX_SS + 7, %ah /* bits 24..31 */
++#ifdef CONFIG_SMP
++	movl PER_CPU_VAR(cpu_number), %ebx
++	shll $PAGE_SHIFT_asm, %ebx
++	addl $cpu_gdt_table, %ebx
++#else
++	movl $cpu_gdt_table, %ebx
++#endif
++	mov 4 + GDT_ESPFIX_SS, %al /* bits 16..23 */
++	mov 7 + GDT_ESPFIX_SS, %ah /* bits 24..31 */
+ 	shl $16, %eax
+ 	addl %esp, %eax			/* the adjusted stack pointer */
+ 	pushl_cfi $__KERNEL_DS
+@@ -816,7 +1068,7 @@ vector=vector+1
+   .endr
+ 2:	jmp common_interrupt
+ .endr
+-END(irq_entries_start)
++ENDPROC(irq_entries_start)
+ 
+ .previous
+ END(interrupt)
+@@ -864,7 +1116,7 @@ ENTRY(coprocessor_error)
+ 	pushl_cfi $do_coprocessor_error
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(coprocessor_error)
++ENDPROC(coprocessor_error)
+ 
+ ENTRY(simd_coprocessor_error)
+ 	RING0_INT_FRAME
+@@ -885,7 +1137,7 @@ ENTRY(simd_coprocessor_error)
+ #endif
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(simd_coprocessor_error)
++ENDPROC(simd_coprocessor_error)
+ 
+ ENTRY(device_not_available)
+ 	RING0_INT_FRAME
+@@ -893,7 +1145,7 @@ ENTRY(device_not_available)
+ 	pushl_cfi $do_device_not_available
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(device_not_available)
++ENDPROC(device_not_available)
+ 
+ #ifdef CONFIG_PARAVIRT
+ ENTRY(native_iret)
+@@ -902,12 +1154,12 @@ ENTRY(native_iret)
+ 	.align 4
+ 	.long native_iret, iret_exc
+ .previous
+-END(native_iret)
++ENDPROC(native_iret)
+ 
+ ENTRY(native_irq_enable_sysexit)
+ 	sti
+ 	sysexit
+-END(native_irq_enable_sysexit)
++ENDPROC(native_irq_enable_sysexit)
+ #endif
+ 
+ ENTRY(overflow)
+@@ -916,7 +1168,7 @@ ENTRY(overflow)
+ 	pushl_cfi $do_overflow
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(overflow)
++ENDPROC(overflow)
+ 
+ ENTRY(bounds)
+ 	RING0_INT_FRAME
+@@ -924,7 +1176,7 @@ ENTRY(bounds)
+ 	pushl_cfi $do_bounds
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(bounds)
++ENDPROC(bounds)
+ 
+ ENTRY(invalid_op)
+ 	RING0_INT_FRAME
+@@ -932,7 +1184,7 @@ ENTRY(invalid_op)
+ 	pushl_cfi $do_invalid_op
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(invalid_op)
++ENDPROC(invalid_op)
+ 
+ ENTRY(coprocessor_segment_overrun)
+ 	RING0_INT_FRAME
+@@ -940,35 +1192,35 @@ ENTRY(coprocessor_segment_overrun)
+ 	pushl_cfi $do_coprocessor_segment_overrun
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(coprocessor_segment_overrun)
++ENDPROC(coprocessor_segment_overrun)
+ 
+ ENTRY(invalid_TSS)
+ 	RING0_EC_FRAME
+ 	pushl_cfi $do_invalid_TSS
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(invalid_TSS)
++ENDPROC(invalid_TSS)
+ 
+ ENTRY(segment_not_present)
+ 	RING0_EC_FRAME
+ 	pushl_cfi $do_segment_not_present
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(segment_not_present)
++ENDPROC(segment_not_present)
+ 
+ ENTRY(stack_segment)
+ 	RING0_EC_FRAME
+ 	pushl_cfi $do_stack_segment
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(stack_segment)
++ENDPROC(stack_segment)
+ 
+ ENTRY(alignment_check)
+ 	RING0_EC_FRAME
+ 	pushl_cfi $do_alignment_check
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(alignment_check)
++ENDPROC(alignment_check)
+ 
+ ENTRY(divide_error)
+ 	RING0_INT_FRAME
+@@ -976,7 +1228,7 @@ ENTRY(divide_error)
+ 	pushl_cfi $do_divide_error
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(divide_error)
++ENDPROC(divide_error)
+ 
+ #ifdef CONFIG_X86_MCE
+ ENTRY(machine_check)
+@@ -985,7 +1237,7 @@ ENTRY(machine_check)
+ 	pushl_cfi machine_check_vector
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(machine_check)
++ENDPROC(machine_check)
+ #endif
+ 
+ ENTRY(spurious_interrupt_bug)
+@@ -994,7 +1246,7 @@ ENTRY(spurious_interrupt_bug)
+ 	pushl_cfi $do_spurious_interrupt_bug
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(spurious_interrupt_bug)
++ENDPROC(spurious_interrupt_bug)
+ /*
+  * End of kprobes section
+  */
+@@ -1109,7 +1361,7 @@ BUILD_INTERRUPT3(xen_hvm_callback_vector, XEN_HVM_EVTCHN_CALLBACK,
+ 
+ ENTRY(mcount)
+ 	ret
+-END(mcount)
++ENDPROC(mcount)
+ 
+ ENTRY(ftrace_caller)
+ 	cmpl $0, function_trace_stop
+@@ -1138,7 +1390,7 @@ ftrace_graph_call:
+ .globl ftrace_stub
+ ftrace_stub:
+ 	ret
+-END(ftrace_caller)
++ENDPROC(ftrace_caller)
+ 
+ #else /* ! CONFIG_DYNAMIC_FTRACE */
+ 
+@@ -1174,7 +1426,7 @@ trace:
+ 	popl %ecx
+ 	popl %eax
+ 	jmp ftrace_stub
+-END(mcount)
++ENDPROC(mcount)
+ #endif /* CONFIG_DYNAMIC_FTRACE */
+ #endif /* CONFIG_FUNCTION_TRACER */
+ 
+@@ -1195,7 +1447,7 @@ ENTRY(ftrace_graph_caller)
+ 	popl %ecx
+ 	popl %eax
+ 	ret
+-END(ftrace_graph_caller)
++ENDPROC(ftrace_graph_caller)
+ 
+ .globl return_to_handler
+ return_to_handler:
+@@ -1209,7 +1461,6 @@ return_to_handler:
+ 	jmp *%ecx
+ #endif
+ 
+-.section .rodata,"a"
+ #include "syscall_table_32.S"
+ 
+ syscall_table_size=(.-sys_call_table)
+@@ -1255,15 +1506,18 @@ error_code:
+ 	movl $-1, PT_ORIG_EAX(%esp)	# no syscall to restart
+ 	REG_TO_PTGS %ecx
+ 	SET_KERNEL_GS %ecx
+-	movl $(__USER_DS), %ecx
++	movl $(__KERNEL_DS), %ecx
+ 	movl %ecx, %ds
+ 	movl %ecx, %es
++
++	pax_enter_kernel
++
+ 	TRACE_IRQS_OFF
+ 	movl %esp,%eax			# pt_regs pointer
+ 	call *%edi
+ 	jmp ret_from_exception
+ 	CFI_ENDPROC
+-END(page_fault)
++ENDPROC(page_fault)
+ 
+ /*
+  * Debug traps and NMI can happen at the one SYSENTER instruction
+@@ -1305,7 +1559,7 @@ debug_stack_correct:
+ 	call do_debug
+ 	jmp ret_from_exception
+ 	CFI_ENDPROC
+-END(debug)
++ENDPROC(debug)
+ 
+ /*
+  * NMI is doubly nasty. It can happen _while_ we're handling
+@@ -1342,6 +1596,9 @@ nmi_stack_correct:
+ 	xorl %edx,%edx		# zero error code
+ 	movl %esp,%eax		# pt_regs pointer
+ 	call do_nmi
++
++	pax_exit_kernel
++
+ 	jmp restore_all_notrace
+ 	CFI_ENDPROC
+ 
+@@ -1378,12 +1635,15 @@ nmi_espfix_stack:
+ 	FIXUP_ESPFIX_STACK		# %eax == %esp
+ 	xorl %edx,%edx			# zero error code
+ 	call do_nmi
++
++	pax_exit_kernel
++
+ 	RESTORE_REGS
+ 	lss 12+4(%esp), %esp		# back to espfix stack
+ 	CFI_ADJUST_CFA_OFFSET -24
+ 	jmp irq_return
+ 	CFI_ENDPROC
+-END(nmi)
++ENDPROC(nmi)
+ 
+ ENTRY(int3)
+ 	RING0_INT_FRAME
+@@ -1395,14 +1655,14 @@ ENTRY(int3)
+ 	call do_int3
+ 	jmp ret_from_exception
+ 	CFI_ENDPROC
+-END(int3)
++ENDPROC(int3)
+ 
+ ENTRY(general_protection)
+ 	RING0_EC_FRAME
+ 	pushl_cfi $do_general_protection
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(general_protection)
++ENDPROC(general_protection)
+ 
+ #ifdef CONFIG_KVM_GUEST
+ ENTRY(async_page_fault)
+@@ -1410,7 +1670,7 @@ ENTRY(async_page_fault)
+ 	pushl_cfi $do_async_page_fault
+ 	jmp error_code
+ 	CFI_ENDPROC
+-END(async_page_fault)
++ENDPROC(async_page_fault)
+ #endif
+ 
+ /*
+diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
+index faf8d5e..4f16a68 100644
+--- a/arch/x86/kernel/entry_64.S
++++ b/arch/x86/kernel/entry_64.S
+@@ -55,6 +55,8 @@
+ #include <asm/paravirt.h>
+ #include <asm/ftrace.h>
+ #include <asm/percpu.h>
++#include <asm/pgtable.h>
++#include <asm/alternative-asm.h>
+ 
+ /* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this.  */
+ #include <linux/elf-em.h>
+@@ -68,8 +70,9 @@
+ #ifdef CONFIG_FUNCTION_TRACER
+ #ifdef CONFIG_DYNAMIC_FTRACE
+ ENTRY(mcount)
++	pax_force_retaddr
+ 	retq
+-END(mcount)
++ENDPROC(mcount)
+ 
+ ENTRY(ftrace_caller)
+ 	cmpl $0, function_trace_stop
+@@ -92,8 +95,9 @@ GLOBAL(ftrace_graph_call)
+ #endif
+ 
+ GLOBAL(ftrace_stub)
++	pax_force_retaddr
+ 	retq
+-END(ftrace_caller)
++ENDPROC(ftrace_caller)
+ 
+ #else /* ! CONFIG_DYNAMIC_FTRACE */
+ ENTRY(mcount)
+@@ -112,6 +116,7 @@ ENTRY(mcount)
+ #endif
+ 
+ GLOBAL(ftrace_stub)
++	pax_force_retaddr
+ 	retq
+ 
+ trace:
+@@ -121,12 +126,13 @@ trace:
+ 	movq 8(%rbp), %rsi
+ 	subq $MCOUNT_INSN_SIZE, %rdi
+ 
++	pax_force_fptr ftrace_trace_function
+ 	call   *ftrace_trace_function
+ 
+ 	MCOUNT_RESTORE_FRAME
+ 
+ 	jmp ftrace_stub
+-END(mcount)
++ENDPROC(mcount)
+ #endif /* CONFIG_DYNAMIC_FTRACE */
+ #endif /* CONFIG_FUNCTION_TRACER */
+ 
+@@ -146,8 +152,9 @@ ENTRY(ftrace_graph_caller)
+ 
+ 	MCOUNT_RESTORE_FRAME
+ 
++	pax_force_retaddr
+ 	retq
+-END(ftrace_graph_caller)
++ENDPROC(ftrace_graph_caller)
+ 
+ GLOBAL(return_to_handler)
+ 	subq  $24, %rsp
+@@ -163,6 +170,7 @@ GLOBAL(return_to_handler)
+ 	movq 8(%rsp), %rdx
+ 	movq (%rsp), %rax
+ 	addq $24, %rsp
++	pax_force_fptr %rdi
+ 	jmp *%rdi
+ #endif
+ 
+@@ -178,6 +186,282 @@ ENTRY(native_usergs_sysret64)
+ ENDPROC(native_usergs_sysret64)
+ #endif /* CONFIG_PARAVIRT */
+ 
++	.macro ljmpq sel, off
++#if defined(CONFIG_MPSC) || defined(CONFIG_MCORE2) || defined (CONFIG_MATOM)
++	.byte 0x48; ljmp *1234f(%rip)
++	.pushsection .rodata
++	.align 16
++	1234: .quad \off; .word \sel
++	.popsection
++#else
++	pushq $\sel
++	pushq $\off
++	lretq
++#endif
++	.endm
++
++	.macro pax_enter_kernel
++	pax_set_fptr_mask
++#ifdef CONFIG_PAX_KERNEXEC
++	call pax_enter_kernel
++#endif
++	.endm
++
++	.macro pax_exit_kernel
++#ifdef CONFIG_PAX_KERNEXEC
++	call pax_exit_kernel
++#endif
++	.endm
++
++#ifdef CONFIG_PAX_KERNEXEC
++ENTRY(pax_enter_kernel)
++	pushq %rdi
++
++#ifdef CONFIG_PARAVIRT
++	PV_SAVE_REGS(CLBR_RDI)
++#endif
++
++	GET_CR0_INTO_RDI
++	bts $16,%rdi
++	jnc 3f
++	mov %cs,%edi
++	cmp $__KERNEL_CS,%edi
++	jnz 2f
++1:
++
++#ifdef CONFIG_PARAVIRT
++	PV_RESTORE_REGS(CLBR_RDI)
++#endif
++
++	popq %rdi
++	pax_force_retaddr
++	retq
++
++2:	ljmpq __KERNEL_CS,1f
++3:	ljmpq __KERNEXEC_KERNEL_CS,4f
++4:	SET_RDI_INTO_CR0
++	jmp 1b
++ENDPROC(pax_enter_kernel)
++
++ENTRY(pax_exit_kernel)
++	pushq %rdi
++
++#ifdef CONFIG_PARAVIRT
++	PV_SAVE_REGS(CLBR_RDI)
++#endif
++
++	mov %cs,%rdi
++	cmp $__KERNEXEC_KERNEL_CS,%edi
++	jz 2f
++1:
++
++#ifdef CONFIG_PARAVIRT
++	PV_RESTORE_REGS(CLBR_RDI);
++#endif
++
++	popq %rdi
++	pax_force_retaddr
++	retq
++
++2:	GET_CR0_INTO_RDI
++	btr $16,%rdi
++	ljmpq __KERNEL_CS,3f
++3:	SET_RDI_INTO_CR0
++	jmp 1b
++#ifdef CONFIG_PARAVIRT
++	PV_RESTORE_REGS(CLBR_RDI);
++#endif
++
++	popq %rdi
++	pax_force_retaddr
++	retq
++ENDPROC(pax_exit_kernel)
++#endif
++
++	.macro pax_enter_kernel_user
++	pax_set_fptr_mask
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	call pax_enter_kernel_user
++#endif
++	.endm
++
++	.macro pax_exit_kernel_user
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	call pax_exit_kernel_user
++#endif
++#ifdef CONFIG_PAX_RANDKSTACK
++	pushq %rax
++	call pax_randomize_kstack
++	popq %rax
++#endif
++	.endm
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++ENTRY(pax_enter_kernel_user)
++	pushq %rdi
++	pushq %rbx
++
++#ifdef CONFIG_PARAVIRT
++	PV_SAVE_REGS(CLBR_RDI)
++#endif
++
++	GET_CR3_INTO_RDI
++	mov %rdi,%rbx
++	add $__START_KERNEL_map,%rbx
++	sub phys_base(%rip),%rbx
++
++#ifdef CONFIG_PARAVIRT
++	pushq %rdi
++	cmpl $0, pv_info+PARAVIRT_enabled
++	jz 1f
++	i = 0
++	.rept USER_PGD_PTRS
++	mov i*8(%rbx),%rsi
++	mov $0,%sil
++	lea i*8(%rbx),%rdi
++	call PARA_INDIRECT(pv_mmu_ops+PV_MMU_set_pgd_batched)
++	i = i + 1
++	.endr
++	jmp 2f
++1:
++#endif
++
++	i = 0
++	.rept USER_PGD_PTRS
++	movb $0,i*8(%rbx)
++	i = i + 1
++	.endr
++
++#ifdef CONFIG_PARAVIRT
++2:	popq %rdi
++#endif
++	SET_RDI_INTO_CR3
++
++#ifdef CONFIG_PAX_KERNEXEC
++	GET_CR0_INTO_RDI
++	bts $16,%rdi
++	SET_RDI_INTO_CR0
++#endif
++
++#ifdef CONFIG_PARAVIRT
++	PV_RESTORE_REGS(CLBR_RDI)
++#endif
++
++	popq %rbx
++	popq %rdi
++	pax_force_retaddr
++	retq
++ENDPROC(pax_enter_kernel_user)
++
++ENTRY(pax_exit_kernel_user)
++	push %rdi
++
++#ifdef CONFIG_PARAVIRT
++	pushq %rbx
++	PV_SAVE_REGS(CLBR_RDI)
++#endif
++
++#ifdef CONFIG_PAX_KERNEXEC
++	GET_CR0_INTO_RDI
++	btr $16,%rdi
++	SET_RDI_INTO_CR0
++#endif
++
++	GET_CR3_INTO_RDI
++	add $__START_KERNEL_map,%rdi
++	sub phys_base(%rip),%rdi
++
++#ifdef CONFIG_PARAVIRT
++	cmpl $0, pv_info+PARAVIRT_enabled
++	jz 1f
++	mov %rdi,%rbx
++	i = 0
++	.rept USER_PGD_PTRS
++	mov i*8(%rbx),%rsi
++	mov $0x67,%sil
++	lea i*8(%rbx),%rdi
++	call PARA_INDIRECT(pv_mmu_ops+PV_MMU_set_pgd_batched)
++	i = i + 1
++	.endr
++	jmp 2f
++1:
++#endif
++
++	i = 0
++	.rept USER_PGD_PTRS
++	movb $0x67,i*8(%rdi)
++	i = i + 1
++	.endr
++
++#ifdef CONFIG_PARAVIRT
++2:	PV_RESTORE_REGS(CLBR_RDI)
++	popq %rbx
++#endif
++
++	popq %rdi
++	pax_force_retaddr
++	retq
++ENDPROC(pax_exit_kernel_user)
++#endif
++
++.macro pax_erase_kstack
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++	call pax_erase_kstack
++#endif
++.endm
++
++#ifdef CONFIG_PAX_MEMORY_STACKLEAK
++/*
++ * r11: thread_info
++ * rcx, rdx: can be clobbered
++ */
++ENTRY(pax_erase_kstack)
++	pushq %rdi
++	pushq %rax
++	pushq %r11
++
++	GET_THREAD_INFO(%r11)
++	mov TI_lowest_stack(%r11), %rdi
++	mov $-0xBEEF, %rax
++	std
++
++1:	mov %edi, %ecx
++	and $THREAD_SIZE_asm - 1, %ecx
++	shr $3, %ecx
++	repne scasq
++	jecxz 2f
++
++	cmp $2*8, %ecx
++	jc 2f
++
++	mov $2*8, %ecx
++	repe scasq
++	jecxz 2f
++	jne 1b
++
++2:	cld
++	mov %esp, %ecx
++	sub %edi, %ecx
++
++	cmp $THREAD_SIZE_asm, %rcx
++	jb 3f
++	ud2
++3:
++
++	shr $3, %ecx
++	rep stosq
++
++	mov TI_task_thread_sp0(%r11), %rdi
++	sub $256, %rdi
++	mov %rdi, TI_lowest_stack(%r11)
++
++	popq %r11
++	popq %rax
++	popq %rdi
++	pax_force_retaddr
++	ret
++ENDPROC(pax_erase_kstack)
++#endif
+ 
+ .macro TRACE_IRQS_IRETQ offset=ARGOFFSET
+ #ifdef CONFIG_TRACE_IRQFLAGS
+@@ -231,8 +515,8 @@ ENDPROC(native_usergs_sysret64)
+ 	.endm
+ 
+ 	.macro UNFAKE_STACK_FRAME
+-	addq $8*6, %rsp
+-	CFI_ADJUST_CFA_OFFSET	-(6*8)
++	addq $8*6 + ARG_SKIP, %rsp
++	CFI_ADJUST_CFA_OFFSET	-(6*8 + ARG_SKIP)
+ 	.endm
+ 
+ /*
+@@ -319,7 +603,7 @@ ENDPROC(native_usergs_sysret64)
+ 	movq %rsp, %rsi
+ 
+ 	leaq -RBP(%rsp),%rdi	/* arg1 for handler */
+-	testl $3, CS(%rdi)
++	testb $3, CS(%rdi)
+ 	je 1f
+ 	SWAPGS
+ 	/*
+@@ -355,9 +639,10 @@ ENTRY(save_rest)
+ 	movq_cfi r15, R15+16
+ 	movq %r11, 8(%rsp)	/* return address */
+ 	FIXUP_TOP_OF_STACK %r11, 16
++	pax_force_retaddr
+ 	ret
+ 	CFI_ENDPROC
+-END(save_rest)
++ENDPROC(save_rest)
+ 
+ /* save complete stack frame */
+ 	.pushsection .kprobes.text, "ax"
+@@ -386,9 +671,10 @@ ENTRY(save_paranoid)
+ 	js 1f	/* negative -> in kernel */
+ 	SWAPGS
+ 	xorl %ebx,%ebx
+-1:	ret
++1:	pax_force_retaddr_bts
++	ret
+ 	CFI_ENDPROC
+-END(save_paranoid)
++ENDPROC(save_paranoid)
+ 	.popsection
+ 
+ /*
+@@ -410,7 +696,7 @@ ENTRY(ret_from_fork)
+ 
+ 	RESTORE_REST
+ 
+-	testl $3, CS-ARGOFFSET(%rsp)		# from kernel_thread?
++	testb $3, CS-ARGOFFSET(%rsp)		# from kernel_thread?
+ 	je   int_ret_from_sys_call
+ 
+ 	testl $_TIF_IA32, TI_flags(%rcx)	# 32-bit compat task needs IRET
+@@ -420,7 +706,7 @@ ENTRY(ret_from_fork)
+ 	jmp ret_from_sys_call			# go to the SYSRET fastpath
+ 
+ 	CFI_ENDPROC
+-END(ret_from_fork)
++ENDPROC(ret_from_fork)
+ 
+ /*
+  * System call entry. Up to 6 arguments in registers are supported.
+@@ -456,7 +742,7 @@ END(ret_from_fork)
+ ENTRY(system_call)
+ 	CFI_STARTPROC	simple
+ 	CFI_SIGNAL_FRAME
+-	CFI_DEF_CFA	rsp,KERNEL_STACK_OFFSET
++	CFI_DEF_CFA	rsp,0
+ 	CFI_REGISTER	rip,rcx
+ 	/*CFI_REGISTER	rflags,r11*/
+ 	SWAPGS_UNSAFE_STACK
+@@ -469,12 +755,13 @@ ENTRY(system_call_after_swapgs)
+ 
+ 	movq	%rsp,PER_CPU_VAR(old_rsp)
+ 	movq	PER_CPU_VAR(kernel_stack),%rsp
++	SAVE_ARGS 8*6,0
++	pax_enter_kernel_user
+ 	/*
+ 	 * No need to follow this irqs off/on section - it's straight
+ 	 * and short:
+ 	 */
+ 	ENABLE_INTERRUPTS(CLBR_NONE)
+-	SAVE_ARGS 8,0
+ 	movq  %rax,ORIG_RAX-ARGOFFSET(%rsp)
+ 	movq  %rcx,RIP-ARGOFFSET(%rsp)
+ 	CFI_REL_OFFSET rip,RIP-ARGOFFSET
+@@ -484,7 +771,7 @@ ENTRY(system_call_after_swapgs)
+ system_call_fastpath:
+ 	cmpq $__NR_syscall_max,%rax
+ 	ja badsys
+-	movq %r10,%rcx
++	movq R10-ARGOFFSET(%rsp),%rcx
+ 	call *sys_call_table(,%rax,8)  # XXX:	 rip relative
+ 	movq %rax,RAX-ARGOFFSET(%rsp)
+ /*
+@@ -503,6 +790,8 @@ sysret_check:
+ 	andl %edi,%edx
+ 	jnz  sysret_careful
+ 	CFI_REMEMBER_STATE
++	pax_exit_kernel_user
++	pax_erase_kstack
+ 	/*
+ 	 * sysretq will re-enable interrupts:
+ 	 */
+@@ -554,14 +843,18 @@ badsys:
+ 	 * jump back to the normal fast path.
+ 	 */
+ auditsys:
+-	movq %r10,%r9			/* 6th arg: 4th syscall arg */
++	movq R10-ARGOFFSET(%rsp),%r9	/* 6th arg: 4th syscall arg */
+ 	movq %rdx,%r8			/* 5th arg: 3rd syscall arg */
+ 	movq %rsi,%rcx			/* 4th arg: 2nd syscall arg */
+ 	movq %rdi,%rdx			/* 3rd arg: 1st syscall arg */
+ 	movq %rax,%rsi			/* 2nd arg: syscall number */
+ 	movl $AUDIT_ARCH_X86_64,%edi	/* 1st arg: audit arch */
+ 	call audit_syscall_entry
++
++	pax_erase_kstack
++
+ 	LOAD_ARGS 0		/* reload call-clobbered registers */
++	pax_set_fptr_mask
+ 	jmp system_call_fastpath
+ 
+ 	/*
+@@ -591,16 +884,20 @@ tracesys:
+ 	FIXUP_TOP_OF_STACK %rdi
+ 	movq %rsp,%rdi
+ 	call syscall_trace_enter
++
++	pax_erase_kstack
++
+ 	/*
+ 	 * Reload arg registers from stack in case ptrace changed them.
+ 	 * We don't reload %rax because syscall_trace_enter() returned
+ 	 * the value it wants us to use in the table lookup.
+ 	 */
+ 	LOAD_ARGS ARGOFFSET, 1
++	pax_set_fptr_mask
+ 	RESTORE_REST
+ 	cmpq $__NR_syscall_max,%rax
+ 	ja   int_ret_from_sys_call	/* RAX(%rsp) set to -ENOSYS above */
+-	movq %r10,%rcx	/* fixup for C */
++	movq R10-ARGOFFSET(%rsp),%rcx	/* fixup for C */
+ 	call *sys_call_table(,%rax,8)
+ 	movq %rax,RAX-ARGOFFSET(%rsp)
+ 	/* Use IRET because user could have changed frame */
+@@ -612,7 +909,7 @@ tracesys:
+ GLOBAL(int_ret_from_sys_call)
+ 	DISABLE_INTERRUPTS(CLBR_NONE)
+ 	TRACE_IRQS_OFF
+-	testl $3,CS-ARGOFFSET(%rsp)
++	testb $3,CS-ARGOFFSET(%rsp)
+ 	je retint_restore_args
+ 	movl $_TIF_ALLWORK_MASK,%edi
+ 	/* edi:	mask to check */
+@@ -623,6 +920,7 @@ GLOBAL(int_with_check)
+ 	andl %edi,%edx
+ 	jnz   int_careful
+ 	andl    $~TS_COMPAT,TI_status(%rcx)
++	pax_erase_kstack
+ 	jmp   retint_swapgs
+ 
+ 	/* Either reschedule or signal or syscall exit tracking needed. */
+@@ -669,7 +967,7 @@ int_restore_rest:
+ 	TRACE_IRQS_OFF
+ 	jmp int_with_check
+ 	CFI_ENDPROC
+-END(system_call)
++ENDPROC(system_call)
+ 
+ /*
+  * Certain special system calls that need to save a complete full stack frame.
+@@ -685,7 +983,7 @@ ENTRY(\label)
+ 	call \func
+ 	jmp ptregscall_common
+ 	CFI_ENDPROC
+-END(\label)
++ENDPROC(\label)
+ 	.endm
+ 
+ 	PTREGSCALL stub_clone, sys_clone, %r8
+@@ -703,9 +1001,10 @@ ENTRY(ptregscall_common)
+ 	movq_cfi_restore R12+8, r12
+ 	movq_cfi_restore RBP+8, rbp
+ 	movq_cfi_restore RBX+8, rbx
++	pax_force_retaddr
+ 	ret $REST_SKIP		/* pop extended registers */
+ 	CFI_ENDPROC
+-END(ptregscall_common)
++ENDPROC(ptregscall_common)
+ 
+ ENTRY(stub_execve)
+ 	CFI_STARTPROC
+@@ -720,7 +1019,7 @@ ENTRY(stub_execve)
+ 	RESTORE_REST
+ 	jmp int_ret_from_sys_call
+ 	CFI_ENDPROC
+-END(stub_execve)
++ENDPROC(stub_execve)
+ 
+ /*
+  * sigreturn is special because it needs to restore all registers on return.
+@@ -738,7 +1037,7 @@ ENTRY(stub_rt_sigreturn)
+ 	RESTORE_REST
+ 	jmp int_ret_from_sys_call
+ 	CFI_ENDPROC
+-END(stub_rt_sigreturn)
++ENDPROC(stub_rt_sigreturn)
+ 
+ /*
+  * Build the entry stubs and pointer table with some assembler magic.
+@@ -773,7 +1072,7 @@ vector=vector+1
+ 2:	jmp common_interrupt
+ .endr
+ 	CFI_ENDPROC
+-END(irq_entries_start)
++ENDPROC(irq_entries_start)
+ 
+ .previous
+ END(interrupt)
+@@ -793,6 +1092,16 @@ END(interrupt)
+ 	subq $ORIG_RAX-RBP, %rsp
+ 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-RBP
+ 	SAVE_ARGS_IRQ
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rdi)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	call \func
+ 	.endm
+ 
+@@ -824,7 +1133,7 @@ ret_from_intr:
+ 
+ exit_intr:
+ 	GET_THREAD_INFO(%rcx)
+-	testl $3,CS-ARGOFFSET(%rsp)
++	testb $3,CS-ARGOFFSET(%rsp)
+ 	je retint_kernel
+ 
+ 	/* Interrupt came from user space */
+@@ -846,12 +1155,15 @@ retint_swapgs:		/* return to user-space */
+ 	 * The iretq could re-enable interrupts:
+ 	 */
+ 	DISABLE_INTERRUPTS(CLBR_ANY)
++	pax_exit_kernel_user
+ 	TRACE_IRQS_IRETQ
+ 	SWAPGS
+ 	jmp restore_args
+ 
+ retint_restore_args:	/* return to kernel space */
+ 	DISABLE_INTERRUPTS(CLBR_ANY)
++	pax_exit_kernel
++	pax_force_retaddr RIP-ARGOFFSET
+ 	/*
+ 	 * The iretq could re-enable interrupts:
+ 	 */
+@@ -940,7 +1252,7 @@ ENTRY(retint_kernel)
+ #endif
+ 
+ 	CFI_ENDPROC
+-END(common_interrupt)
++ENDPROC(common_interrupt)
+ /*
+  * End of kprobes section
+  */
+@@ -956,7 +1268,7 @@ ENTRY(\sym)
+ 	interrupt \do_sym
+ 	jmp ret_from_intr
+ 	CFI_ENDPROC
+-END(\sym)
++ENDPROC(\sym)
+ .endm
+ 
+ #ifdef CONFIG_SMP
+@@ -1021,12 +1333,22 @@ ENTRY(\sym)
+ 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
+ 	call error_entry
+ 	DEFAULT_FRAME 0
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rsp)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	movq %rsp,%rdi		/* pt_regs pointer */
+ 	xorl %esi,%esi		/* no error code */
+ 	call \do_sym
+ 	jmp error_exit		/* %ebx: no swapgs flag */
+ 	CFI_ENDPROC
+-END(\sym)
++ENDPROC(\sym)
+ .endm
+ 
+ .macro paranoidzeroentry sym do_sym
+@@ -1038,15 +1360,25 @@ ENTRY(\sym)
+ 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
+ 	call save_paranoid
+ 	TRACE_IRQS_OFF
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rsp)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	movq %rsp,%rdi		/* pt_regs pointer */
+ 	xorl %esi,%esi		/* no error code */
+ 	call \do_sym
+ 	jmp paranoid_exit	/* %ebx: no swapgs flag */
+ 	CFI_ENDPROC
+-END(\sym)
++ENDPROC(\sym)
+ .endm
+ 
+-#define INIT_TSS_IST(x) PER_CPU_VAR(init_tss) + (TSS_ist + ((x) - 1) * 8)
++#define INIT_TSS_IST(x) (TSS_ist + ((x) - 1) * 8)(%r12)
+ .macro paranoidzeroentry_ist sym do_sym ist
+ ENTRY(\sym)
+ 	INTR_FRAME
+@@ -1056,14 +1388,30 @@ ENTRY(\sym)
+ 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
+ 	call save_paranoid
+ 	TRACE_IRQS_OFF
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rsp)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	movq %rsp,%rdi		/* pt_regs pointer */
+ 	xorl %esi,%esi		/* no error code */
++#ifdef CONFIG_SMP
++	imul $TSS_size, PER_CPU_VAR(cpu_number), %r12d
++	lea init_tss(%r12), %r12
++#else
++	lea init_tss(%rip), %r12
++#endif
+ 	subq $EXCEPTION_STKSZ, INIT_TSS_IST(\ist)
+ 	call \do_sym
+ 	addq $EXCEPTION_STKSZ, INIT_TSS_IST(\ist)
+ 	jmp paranoid_exit	/* %ebx: no swapgs flag */
+ 	CFI_ENDPROC
+-END(\sym)
++ENDPROC(\sym)
+ .endm
+ 
+ .macro errorentry sym do_sym
+@@ -1074,13 +1422,23 @@ ENTRY(\sym)
+ 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
+ 	call error_entry
+ 	DEFAULT_FRAME 0
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rsp)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	movq %rsp,%rdi			/* pt_regs pointer */
+ 	movq ORIG_RAX(%rsp),%rsi	/* get error code */
+ 	movq $-1,ORIG_RAX(%rsp)		/* no syscall to restart */
+ 	call \do_sym
+ 	jmp error_exit			/* %ebx: no swapgs flag */
+ 	CFI_ENDPROC
+-END(\sym)
++ENDPROC(\sym)
+ .endm
+ 
+ 	/* error code is on the stack already */
+@@ -1093,13 +1451,23 @@ ENTRY(\sym)
+ 	call save_paranoid
+ 	DEFAULT_FRAME 0
+ 	TRACE_IRQS_OFF
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rsp)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	movq %rsp,%rdi			/* pt_regs pointer */
+ 	movq ORIG_RAX(%rsp),%rsi	/* get error code */
+ 	movq $-1,ORIG_RAX(%rsp)		/* no syscall to restart */
+ 	call \do_sym
+ 	jmp paranoid_exit		/* %ebx: no swapgs flag */
+ 	CFI_ENDPROC
+-END(\sym)
++ENDPROC(\sym)
+ .endm
+ 
+ zeroentry divide_error do_divide_error
+@@ -1129,9 +1497,10 @@ gs_change:
+ 2:	mfence		/* workaround */
+ 	SWAPGS
+ 	popfq_cfi
++	pax_force_retaddr
+ 	ret
+ 	CFI_ENDPROC
+-END(native_load_gs_index)
++ENDPROC(native_load_gs_index)
+ 
+ 	.section __ex_table,"a"
+ 	.align 8
+@@ -1153,13 +1522,14 @@ ENTRY(kernel_thread_helper)
+ 	 * Here we are in the child and the registers are set as they were
+ 	 * at kernel_thread() invocation in the parent.
+ 	 */
++	pax_force_fptr %rsi
+ 	call *%rsi
+ 	# exit
+ 	mov %eax, %edi
+ 	call do_exit
+ 	ud2			# padding for call trace
+ 	CFI_ENDPROC
+-END(kernel_thread_helper)
++ENDPROC(kernel_thread_helper)
+ 
+ /*
+  * execve(). This function needs to use IRET, not SYSRET, to set up all state properly.
+@@ -1186,11 +1556,11 @@ ENTRY(kernel_execve)
+ 	RESTORE_REST
+ 	testq %rax,%rax
+ 	je int_ret_from_sys_call
+-	RESTORE_ARGS
+ 	UNFAKE_STACK_FRAME
++	pax_force_retaddr
+ 	ret
+ 	CFI_ENDPROC
+-END(kernel_execve)
++ENDPROC(kernel_execve)
+ 
+ /* Call softirq on interrupt stack. Interrupts are off. */
+ ENTRY(call_softirq)
+@@ -1208,9 +1578,10 @@ ENTRY(call_softirq)
+ 	CFI_DEF_CFA_REGISTER	rsp
+ 	CFI_ADJUST_CFA_OFFSET   -8
+ 	decl PER_CPU_VAR(irq_count)
++	pax_force_retaddr
+ 	ret
+ 	CFI_ENDPROC
+-END(call_softirq)
++ENDPROC(call_softirq)
+ 
+ #ifdef CONFIG_XEN
+ zeroentry xen_hypervisor_callback xen_do_hypervisor_callback
+@@ -1248,7 +1619,7 @@ ENTRY(xen_do_hypervisor_callback)   # do_hypervisor_callback(struct *pt_regs)
+ 	decl PER_CPU_VAR(irq_count)
+ 	jmp  error_exit
+ 	CFI_ENDPROC
+-END(xen_do_hypervisor_callback)
++ENDPROC(xen_do_hypervisor_callback)
+ 
+ /*
+  * Hypervisor uses this for application faults while it executes.
+@@ -1307,7 +1678,7 @@ ENTRY(xen_failsafe_callback)
+ 	SAVE_ALL
+ 	jmp error_exit
+ 	CFI_ENDPROC
+-END(xen_failsafe_callback)
++ENDPROC(xen_failsafe_callback)
+ 
+ apicinterrupt XEN_HVM_EVTCHN_CALLBACK \
+ 	xen_hvm_callback_vector xen_evtchn_do_upcall
+@@ -1356,16 +1727,31 @@ ENTRY(paranoid_exit)
+ 	TRACE_IRQS_OFF
+ 	testl %ebx,%ebx				/* swapgs needed? */
+ 	jnz paranoid_restore
+-	testl $3,CS(%rsp)
++	testb $3,CS(%rsp)
+ 	jnz   paranoid_userspace
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	pax_exit_kernel
++	TRACE_IRQS_IRETQ 0
++	SWAPGS_UNSAFE_STACK
++	RESTORE_ALL 8
++	pax_force_retaddr_bts
++	jmp irq_return
++#endif
+ paranoid_swapgs:
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	pax_exit_kernel_user
++#else
++	pax_exit_kernel
++#endif
+ 	TRACE_IRQS_IRETQ 0
+ 	SWAPGS_UNSAFE_STACK
+ 	RESTORE_ALL 8
+ 	jmp irq_return
+ paranoid_restore:
++	pax_exit_kernel
+ 	TRACE_IRQS_IRETQ 0
+ 	RESTORE_ALL 8
++	pax_force_retaddr_bts
+ 	jmp irq_return
+ paranoid_userspace:
+ 	GET_THREAD_INFO(%rcx)
+@@ -1394,7 +1780,7 @@ paranoid_schedule:
+ 	TRACE_IRQS_OFF
+ 	jmp paranoid_userspace
+ 	CFI_ENDPROC
+-END(paranoid_exit)
++ENDPROC(paranoid_exit)
+ 
+ /*
+  * Exception entry point. This expects an error code/orig_rax on the stack.
+@@ -1421,12 +1807,13 @@ ENTRY(error_entry)
+ 	movq_cfi r14, R14+8
+ 	movq_cfi r15, R15+8
+ 	xorl %ebx,%ebx
+-	testl $3,CS+8(%rsp)
++	testb $3,CS+8(%rsp)
+ 	je error_kernelspace
+ error_swapgs:
+ 	SWAPGS
+ error_sti:
+ 	TRACE_IRQS_OFF
++	pax_force_retaddr_bts
+ 	ret
+ 
+ /*
+@@ -1453,7 +1840,7 @@ bstep_iret:
+ 	movq %rcx,RIP+8(%rsp)
+ 	jmp error_swapgs
+ 	CFI_ENDPROC
+-END(error_entry)
++ENDPROC(error_entry)
+ 
+ 
+ /* ebx:	no swapgs flag (1: don't need swapgs, 0: need it) */
+@@ -1473,7 +1860,7 @@ ENTRY(error_exit)
+ 	jnz retint_careful
+ 	jmp retint_swapgs
+ 	CFI_ENDPROC
+-END(error_exit)
++ENDPROC(error_exit)
+ 
+ 
+ 	/* runs on exception stack */
+@@ -1485,6 +1872,16 @@ ENTRY(nmi)
+ 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
+ 	call save_paranoid
+ 	DEFAULT_FRAME 0
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	testb $3, CS(%rsp)
++	jnz 1f
++	pax_enter_kernel
++	jmp 2f
++1:	pax_enter_kernel_user
++2:
++#else
++	pax_enter_kernel
++#endif
+ 	/* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */
+ 	movq %rsp,%rdi
+ 	movq $-1,%rsi
+@@ -1495,12 +1892,28 @@ ENTRY(nmi)
+ 	DISABLE_INTERRUPTS(CLBR_NONE)
+ 	testl %ebx,%ebx				/* swapgs needed? */
+ 	jnz nmi_restore
+-	testl $3,CS(%rsp)
++	testb $3,CS(%rsp)
+ 	jnz nmi_userspace
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	pax_exit_kernel
++	SWAPGS_UNSAFE_STACK
++	RESTORE_ALL 8
++	pax_force_retaddr_bts
++	jmp irq_return
++#endif
+ nmi_swapgs:
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	pax_exit_kernel_user
++#else
++	pax_exit_kernel
++#endif
+ 	SWAPGS_UNSAFE_STACK
++	RESTORE_ALL 8
++	jmp irq_return
+ nmi_restore:
++	pax_exit_kernel
+ 	RESTORE_ALL 8
++	pax_force_retaddr_bts
+ 	jmp irq_return
+ nmi_userspace:
+ 	GET_THREAD_INFO(%rcx)
+@@ -1529,14 +1942,14 @@ nmi_schedule:
+ 	jmp paranoid_exit
+ 	CFI_ENDPROC
+ #endif
+-END(nmi)
++ENDPROC(nmi)
+ 
+ ENTRY(ignore_sysret)
+ 	CFI_STARTPROC
+ 	mov $-ENOSYS,%eax
+ 	sysret
+ 	CFI_ENDPROC
+-END(ignore_sysret)
++ENDPROC(ignore_sysret)
+ 
+ /*
+  * End of kprobes section
+diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
+index c9a281f..ce2f317 100644
+--- a/arch/x86/kernel/ftrace.c
++++ b/arch/x86/kernel/ftrace.c
+@@ -126,7 +126,7 @@ static void *mod_code_ip;		/* holds the IP to write to */
+ static const void *mod_code_newcode;	/* holds the text to write to the IP */
+ 
+ static unsigned nmi_wait_count;
+-static atomic_t nmi_update_count = ATOMIC_INIT(0);
++static atomic_unchecked_t nmi_update_count = ATOMIC_INIT(0);
+ 
+ int ftrace_arch_read_dyn_info(char *buf, int size)
+ {
+@@ -134,7 +134,7 @@ int ftrace_arch_read_dyn_info(char *buf, int size)
+ 
+ 	r = snprintf(buf, size, "%u %u",
+ 		     nmi_wait_count,
+-		     atomic_read(&nmi_update_count));
++		     atomic_read_unchecked(&nmi_update_count));
+ 	return r;
+ }
+ 
+@@ -177,8 +177,10 @@ void ftrace_nmi_enter(void)
+ 
+ 	if (atomic_inc_return(&nmi_running) & MOD_CODE_WRITE_FLAG) {
+ 		smp_rmb();
++		pax_open_kernel();
+ 		ftrace_mod_code();
+-		atomic_inc(&nmi_update_count);
++		pax_close_kernel();
++		atomic_inc_unchecked(&nmi_update_count);
+ 	}
+ 	/* Must have previous changes seen before executions */
+ 	smp_mb();
+@@ -271,6 +273,8 @@ ftrace_modify_code(unsigned long ip, unsigned const char *old_code,
+ {
+ 	unsigned char replaced[MCOUNT_INSN_SIZE];
+ 
++	ip = ktla_ktva(ip);
++
+ 	/*
+ 	 * Note: Due to modules and __init, code can
+ 	 *  disappear and change, we need to protect against faulting
+@@ -327,7 +331,7 @@ int ftrace_update_ftrace_func(ftrace_func_t func)
+ 	unsigned char old[MCOUNT_INSN_SIZE], *new;
+ 	int ret;
+ 
+-	memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE);
++	memcpy(old, (void *)ktla_ktva((unsigned long)ftrace_call), MCOUNT_INSN_SIZE);
+ 	new = ftrace_call_replace(ip, (unsigned long)func);
+ 	ret = ftrace_modify_code(ip, old, new);
+ 
+@@ -353,6 +357,8 @@ static int ftrace_mod_jmp(unsigned long ip,
+ {
+ 	unsigned char code[MCOUNT_INSN_SIZE];
+ 
++	ip = ktla_ktva(ip);
++
+ 	if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE))
+ 		return -EFAULT;
+ 
+diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c
+index 3bb0850..55a56f4 100644
+--- a/arch/x86/kernel/head32.c
++++ b/arch/x86/kernel/head32.c
+@@ -19,6 +19,7 @@
+ #include <asm/io_apic.h>
+ #include <asm/bios_ebda.h>
+ #include <asm/tlbflush.h>
++#include <asm/boot.h>
+ 
+ static void __init i386_default_early_setup(void)
+ {
+@@ -33,7 +34,7 @@ void __init i386_start_kernel(void)
+ {
+ 	memblock_init();
+ 
+-	memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS");
++	memblock_x86_reserve_range(LOAD_PHYSICAL_ADDR, __pa_symbol(&__bss_stop), "TEXT DATA BSS");
+ 
+ #ifdef CONFIG_BLK_DEV_INITRD
+ 	/* Reserve INITRD */
+diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
+index ce0be7c..c41476e 100644
+--- a/arch/x86/kernel/head_32.S
++++ b/arch/x86/kernel/head_32.S
+@@ -25,6 +25,12 @@
+ /* Physical address */
+ #define pa(X) ((X) - __PAGE_OFFSET)
+ 
++#ifdef CONFIG_PAX_KERNEXEC
++#define ta(X) (X)
++#else
++#define ta(X) ((X) - __PAGE_OFFSET)
++#endif
++
+ /*
+  * References to members of the new_cpu_data structure.
+  */
+@@ -54,11 +60,7 @@
+  * and small than max_low_pfn, otherwise will waste some page table entries
+  */
+ 
+-#if PTRS_PER_PMD > 1
+-#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD)
+-#else
+-#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD)
+-#endif
++#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PTE)
+ 
+ /* Number of possible pages in the lowmem region */
+ LOWMEM_PAGES = (((1<<32) - __PAGE_OFFSET) >> PAGE_SHIFT)
+@@ -77,6 +79,12 @@ INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE
+ RESERVE_BRK(pagetables, INIT_MAP_SIZE)
+ 
+ /*
++ * Real beginning of normal "text" segment
++ */
++ENTRY(stext)
++ENTRY(_stext)
++
++/*
+  * 32-bit kernel entrypoint; only used by the boot CPU.  On entry,
+  * %esi points to the real-mode code as a 32-bit pointer.
+  * CS and DS must be 4 GB flat segments, but we don't depend on
+@@ -84,6 +92,13 @@ RESERVE_BRK(pagetables, INIT_MAP_SIZE)
+  * can.
+  */
+ __HEAD
++
++#ifdef CONFIG_PAX_KERNEXEC
++	jmp startup_32
++/* PaX: fill first page in .text with int3 to catch NULL derefs in kernel mode */
++.fill PAGE_SIZE-5,1,0xcc
++#endif
++
+ ENTRY(startup_32)
+ 	movl pa(stack_start),%ecx
+ 	
+@@ -105,6 +120,57 @@ ENTRY(startup_32)
+ 2:
+ 	leal -__PAGE_OFFSET(%ecx),%esp
+ 
++#ifdef CONFIG_SMP
++	movl $pa(cpu_gdt_table),%edi
++	movl $__per_cpu_load,%eax
++	movw %ax,__KERNEL_PERCPU + 2(%edi)
++	rorl $16,%eax
++	movb %al,__KERNEL_PERCPU + 4(%edi)
++	movb %ah,__KERNEL_PERCPU + 7(%edi)
++	movl $__per_cpu_end - 1,%eax
++	subl $__per_cpu_start,%eax
++	movw %ax,__KERNEL_PERCPU + 0(%edi)
++#endif
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	movl $NR_CPUS,%ecx
++	movl $pa(cpu_gdt_table),%edi
++1:
++	movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c09700),GDT_ENTRY_KERNEL_DS * 8 + 4(%edi)
++	movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c0fb00),GDT_ENTRY_DEFAULT_USER_CS * 8 + 4(%edi)
++	movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c0f300),GDT_ENTRY_DEFAULT_USER_DS * 8 + 4(%edi)
++	addl $PAGE_SIZE_asm,%edi
++	loop 1b
++#endif
++
++#ifdef CONFIG_PAX_KERNEXEC
++	movl $pa(boot_gdt),%edi
++	movl $__LOAD_PHYSICAL_ADDR,%eax
++	movw %ax,__BOOT_CS + 2(%edi)
++	rorl $16,%eax
++	movb %al,__BOOT_CS + 4(%edi)
++	movb %ah,__BOOT_CS + 7(%edi)
++	rorl $16,%eax
++
++	ljmp $(__BOOT_CS),$1f
++1:
++
++	movl $NR_CPUS,%ecx
++	movl $pa(cpu_gdt_table),%edi
++	addl $__PAGE_OFFSET,%eax
++1:
++	movw %ax,__KERNEL_CS + 2(%edi)
++	movw %ax,__KERNEXEC_KERNEL_CS + 2(%edi)
++	rorl $16,%eax
++	movb %al,__KERNEL_CS + 4(%edi)
++	movb %al,__KERNEXEC_KERNEL_CS + 4(%edi)
++	movb %ah,__KERNEL_CS + 7(%edi)
++	movb %ah,__KERNEXEC_KERNEL_CS + 7(%edi)
++	rorl $16,%eax
++	addl $PAGE_SIZE_asm,%edi
++	loop 1b
++#endif
++
+ /*
+  * Clear BSS first so that there are no surprises...
+  */
+@@ -195,8 +261,11 @@ ENTRY(startup_32)
+ 	movl %eax, pa(max_pfn_mapped)
+ 
+ 	/* Do early initialization of the fixmap area */
+-	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
+-	movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8)
++#ifdef CONFIG_COMPAT_VDSO
++	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(initial_pg_pmd+0x1000*KPMDS-8)
++#else
++	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,pa(initial_pg_pmd+0x1000*KPMDS-8)
++#endif
+ #else	/* Not PAE */
+ 
+ page_pde_offset = (__PAGE_OFFSET >> 20);
+@@ -226,8 +295,11 @@ page_pde_offset = (__PAGE_OFFSET >> 20);
+ 	movl %eax, pa(max_pfn_mapped)
+ 
+ 	/* Do early initialization of the fixmap area */
+-	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
+-	movl %eax,pa(initial_page_table+0xffc)
++#ifdef CONFIG_COMPAT_VDSO
++	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(initial_page_table+0xffc)
++#else
++	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,pa(initial_page_table+0xffc)
++#endif
+ #endif
+ 
+ #ifdef CONFIG_PARAVIRT
+@@ -241,9 +313,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20);
+ 	cmpl $num_subarch_entries, %eax
+ 	jae bad_subarch
+ 
+-	movl pa(subarch_entries)(,%eax,4), %eax
+-	subl $__PAGE_OFFSET, %eax
+-	jmp *%eax
++	jmp *pa(subarch_entries)(,%eax,4)
+ 
+ bad_subarch:
+ WEAK(lguest_entry)
+@@ -255,10 +325,10 @@ WEAK(xen_entry)
+ 	__INITDATA
+ 
+ subarch_entries:
+-	.long default_entry		/* normal x86/PC */
+-	.long lguest_entry		/* lguest hypervisor */
+-	.long xen_entry			/* Xen hypervisor */
+-	.long default_entry		/* Moorestown MID */
++	.long ta(default_entry)		/* normal x86/PC */
++	.long ta(lguest_entry)		/* lguest hypervisor */
++	.long ta(xen_entry)		/* Xen hypervisor */
++	.long ta(default_entry)		/* Moorestown MID */
+ num_subarch_entries = (. - subarch_entries) / 4
+ .previous
+ #else
+@@ -312,6 +382,7 @@ default_entry:
+ 	orl %edx,%eax
+ 	movl %eax,%cr4
+ 
++#ifdef CONFIG_X86_PAE
+ 	testb $X86_CR4_PAE, %al		# check if PAE is enabled
+ 	jz 6f
+ 
+@@ -340,6 +411,9 @@ default_entry:
+ 	/* Make changes effective */
+ 	wrmsr
+ 
++	btsl $_PAGE_BIT_NX-32,pa(__supported_pte_mask+4)
++#endif
++
+ 6:
+ 
+ /*
+@@ -443,7 +517,7 @@ is386:	movl $2,%ecx		# set MP
+ 1:	movl $(__KERNEL_DS),%eax	# reload all the segment registers
+ 	movl %eax,%ss			# after changing gdt.
+ 
+-	movl $(__USER_DS),%eax		# DS/ES contains default USER segment
++#	movl $(__KERNEL_DS),%eax	# DS/ES contains default KERNEL segment
+ 	movl %eax,%ds
+ 	movl %eax,%es
+ 
+@@ -457,15 +531,22 @@ is386:	movl $2,%ecx		# set MP
+ 	 */
+ 	cmpb $0,ready
+ 	jne 1f
+-	movl $gdt_page,%eax
++	movl $cpu_gdt_table,%eax
+ 	movl $stack_canary,%ecx
++#ifdef CONFIG_SMP
++	addl $__per_cpu_load,%ecx
++#endif
+ 	movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax)
+ 	shrl $16, %ecx
+ 	movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax)
+ 	movb %ch, 8 * GDT_ENTRY_STACK_CANARY + 7(%eax)
+ 1:
+-#endif
+ 	movl $(__KERNEL_STACK_CANARY),%eax
++#elif defined(CONFIG_PAX_MEMORY_UDEREF)
++	movl $(__USER_DS),%eax
++#else
++	xorl %eax,%eax
++#endif
+ 	movl %eax,%gs
+ 
+ 	xorl %eax,%eax			# Clear LDT
+@@ -558,22 +639,22 @@ early_page_fault:
+ 	jmp early_fault
+ 
+ early_fault:
+-	cld
+ #ifdef CONFIG_PRINTK
++	cmpl $1,%ss:early_recursion_flag
++	je hlt_loop
++	incl %ss:early_recursion_flag
++	cld
+ 	pusha
+ 	movl $(__KERNEL_DS),%eax
+ 	movl %eax,%ds
+ 	movl %eax,%es
+-	cmpl $2,early_recursion_flag
+-	je hlt_loop
+-	incl early_recursion_flag
+ 	movl %cr2,%eax
+ 	pushl %eax
+ 	pushl %edx		/* trapno */
+ 	pushl $fault_msg
+ 	call printk
++;	call dump_stack
+ #endif
+-	call dump_stack
+ hlt_loop:
+ 	hlt
+ 	jmp hlt_loop
+@@ -581,8 +662,11 @@ hlt_loop:
+ /* This is the default interrupt "handler" :-) */
+ 	ALIGN
+ ignore_int:
+-	cld
+ #ifdef CONFIG_PRINTK
++	cmpl $2,%ss:early_recursion_flag
++	je hlt_loop
++	incl %ss:early_recursion_flag
++	cld
+ 	pushl %eax
+ 	pushl %ecx
+ 	pushl %edx
+@@ -591,9 +675,6 @@ ignore_int:
+ 	movl $(__KERNEL_DS),%eax
+ 	movl %eax,%ds
+ 	movl %eax,%es
+-	cmpl $2,early_recursion_flag
+-	je hlt_loop
+-	incl early_recursion_flag
+ 	pushl 16(%esp)
+ 	pushl 24(%esp)
+ 	pushl 32(%esp)
+@@ -622,29 +703,43 @@ ENTRY(initial_code)
+ /*
+  * BSS section
+  */
+-__PAGE_ALIGNED_BSS
+-	.align PAGE_SIZE
+ #ifdef CONFIG_X86_PAE
++.section .initial_pg_pmd,"a",@progbits
+ initial_pg_pmd:
+ 	.fill 1024*KPMDS,4,0
+ #else
++.section .initial_page_table,"a",@progbits
+ ENTRY(initial_page_table)
+ 	.fill 1024,4,0
+ #endif
++.section .initial_pg_fixmap,"a",@progbits
+ initial_pg_fixmap:
+ 	.fill 1024,4,0
++.section .empty_zero_page,"a",@progbits
+ ENTRY(empty_zero_page)
+ 	.fill 4096,1,0
++.section .swapper_pg_dir,"a",@progbits
+ ENTRY(swapper_pg_dir)
++#ifdef CONFIG_X86_PAE
++	.fill 4,8,0
++#else
+ 	.fill 1024,4,0
++#endif
++
++/*
++ * The IDT has to be page-aligned to simplify the Pentium
++ * F0 0F bug workaround.. We have a special link segment
++ * for this.
++ */
++.section .idt,"a",@progbits
++ENTRY(idt_table)
++	.fill 256,8,0
+ 
+ /*
+  * This starts the data section.
+  */
+ #ifdef CONFIG_X86_PAE
+-__PAGE_ALIGNED_DATA
+-	/* Page-aligned for the benefit of paravirt? */
+-	.align PAGE_SIZE
++.section .initial_page_table,"a",@progbits
+ ENTRY(initial_page_table)
+ 	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
+ # if KPMDS == 3
+@@ -663,18 +758,27 @@ ENTRY(initial_page_table)
+ #  error "Kernel PMDs should be 1, 2 or 3"
+ # endif
+ 	.align PAGE_SIZE		/* needs to be page-sized too */
++
++#ifdef CONFIG_PAX_PER_CPU_PGD
++ENTRY(cpu_pgd)
++	.rept NR_CPUS
++	.fill	4,8,0
++	.endr
++#endif
++
+ #endif
+ 
+ .data
+ .balign 4
+ ENTRY(stack_start)
+-	.long init_thread_union+THREAD_SIZE
++	.long init_thread_union+THREAD_SIZE-8
+ 
++ready:	.byte 0
++
++.section .rodata,"a",@progbits
+ early_recursion_flag:
+ 	.long 0
+ 
+-ready:	.byte 0
+-
+ int_msg:
+ 	.asciz "Unknown interrupt or fault at: %p %p %p\n"
+ 
+@@ -707,7 +811,7 @@ fault_msg:
+ 	.word 0				# 32 bit align gdt_desc.address
+ boot_gdt_descr:
+ 	.word __BOOT_DS+7
+-	.long boot_gdt - __PAGE_OFFSET
++	.long pa(boot_gdt)
+ 
+ 	.word 0				# 32-bit align idt_desc.address
+ idt_descr:
+@@ -718,7 +822,7 @@ idt_descr:
+ 	.word 0				# 32 bit align gdt_desc.address
+ ENTRY(early_gdt_descr)
+ 	.word GDT_ENTRIES*8-1
+-	.long gdt_page			/* Overwritten for secondary CPUs */
++	.long cpu_gdt_table		/* Overwritten for secondary CPUs */
+ 
+ /*
+  * The boot_gdt must mirror the equivalent in setup.S and is
+@@ -727,5 +831,65 @@ ENTRY(early_gdt_descr)
+ 	.align L1_CACHE_BYTES
+ ENTRY(boot_gdt)
+ 	.fill GDT_ENTRY_BOOT_CS,8,0
+-	.quad 0x00cf9a000000ffff	/* kernel 4GB code at 0x00000000 */
+-	.quad 0x00cf92000000ffff	/* kernel 4GB data at 0x00000000 */
++	.quad 0x00cf9b000000ffff	/* kernel 4GB code at 0x00000000 */
++	.quad 0x00cf93000000ffff	/* kernel 4GB data at 0x00000000 */
++
++	.align PAGE_SIZE_asm
++ENTRY(cpu_gdt_table)
++	.rept NR_CPUS
++	.quad 0x0000000000000000	/* NULL descriptor */
++	.quad 0x0000000000000000	/* 0x0b reserved */
++	.quad 0x0000000000000000	/* 0x13 reserved */
++	.quad 0x0000000000000000	/* 0x1b reserved */
++
++#ifdef CONFIG_PAX_KERNEXEC
++	.quad 0x00cf9b000000ffff	/* 0x20 alternate kernel 4GB code at 0x00000000 */
++#else
++	.quad 0x0000000000000000	/* 0x20 unused */
++#endif
++
++	.quad 0x0000000000000000	/* 0x28 unused */
++	.quad 0x0000000000000000	/* 0x33 TLS entry 1 */
++	.quad 0x0000000000000000	/* 0x3b TLS entry 2 */
++	.quad 0x0000000000000000	/* 0x43 TLS entry 3 */
++	.quad 0x0000000000000000	/* 0x4b reserved */
++	.quad 0x0000000000000000	/* 0x53 reserved */
++	.quad 0x0000000000000000	/* 0x5b reserved */
++
++	.quad 0x00cf9b000000ffff	/* 0x60 kernel 4GB code at 0x00000000 */
++	.quad 0x00cf93000000ffff	/* 0x68 kernel 4GB data at 0x00000000 */
++	.quad 0x00cffb000000ffff	/* 0x73 user 4GB code at 0x00000000 */
++	.quad 0x00cff3000000ffff	/* 0x7b user 4GB data at 0x00000000 */
++
++	.quad 0x0000000000000000	/* 0x80 TSS descriptor */
++	.quad 0x0000000000000000	/* 0x88 LDT descriptor */
++
++	/*
++	 * Segments used for calling PnP BIOS have byte granularity.
++	 * The code segments and data segments have fixed 64k limits,
++	 * the transfer segment sizes are set at run time.
++	 */
++	.quad 0x00409b000000ffff	/* 0x90 32-bit code */
++	.quad 0x00009b000000ffff	/* 0x98 16-bit code */
++	.quad 0x000093000000ffff	/* 0xa0 16-bit data */
++	.quad 0x0000930000000000	/* 0xa8 16-bit data */
++	.quad 0x0000930000000000	/* 0xb0 16-bit data */
++
++	/*
++	 * The APM segments have byte granularity and their bases
++	 * are set at run time.  All have 64k limits.
++	 */
++	.quad 0x00409b000000ffff	/* 0xb8 APM CS    code */
++	.quad 0x00009b000000ffff	/* 0xc0 APM CS 16 code (16 bit) */
++	.quad 0x004093000000ffff	/* 0xc8 APM DS    data */
++
++	.quad 0x00c0930000000000	/* 0xd0 - ESPFIX SS */
++	.quad 0x0040930000000000	/* 0xd8 - PERCPU */
++	.quad 0x0040910000000017	/* 0xe0 - STACK_CANARY */
++	.quad 0x0000000000000000	/* 0xe8 - PCIBIOS_CS */
++	.quad 0x0000000000000000	/* 0xf0 - PCIBIOS_DS */
++	.quad 0x0000000000000000	/* 0xf8 - GDT entry 31: double-fault TSS */
++
++	/* Be sure this is zeroed to avoid false validations in Xen */
++	.fill PAGE_SIZE_asm - GDT_SIZE,1,0
++	.endr
+diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
+index e11e394..9aebc5d 100644
+--- a/arch/x86/kernel/head_64.S
++++ b/arch/x86/kernel/head_64.S
+@@ -19,6 +19,8 @@
+ #include <asm/cache.h>
+ #include <asm/processor-flags.h>
+ #include <asm/percpu.h>
++#include <asm/cpufeature.h>
++#include <asm/alternative-asm.h>
+ 
+ #ifdef CONFIG_PARAVIRT
+ #include <asm/asm-offsets.h>
+@@ -38,6 +40,12 @@ L4_PAGE_OFFSET = pgd_index(__PAGE_OFFSET)
+ L3_PAGE_OFFSET = pud_index(__PAGE_OFFSET)
+ L4_START_KERNEL = pgd_index(__START_KERNEL_map)
+ L3_START_KERNEL = pud_index(__START_KERNEL_map)
++L4_VMALLOC_START = pgd_index(VMALLOC_START)
++L3_VMALLOC_START = pud_index(VMALLOC_START)
++L4_VMALLOC_END = pgd_index(VMALLOC_END)
++L3_VMALLOC_END = pud_index(VMALLOC_END)
++L4_VMEMMAP_START = pgd_index(VMEMMAP_START)
++L3_VMEMMAP_START = pud_index(VMEMMAP_START)
+ 
+ 	.text
+ 	__HEAD
+@@ -85,35 +93,23 @@ startup_64:
+ 	 */
+ 	addq	%rbp, init_level4_pgt + 0(%rip)
+ 	addq	%rbp, init_level4_pgt + (L4_PAGE_OFFSET*8)(%rip)
++	addq	%rbp, init_level4_pgt + (L4_VMALLOC_START*8)(%rip)
++	addq	%rbp, init_level4_pgt + (L4_VMALLOC_END*8)(%rip)
++	addq	%rbp, init_level4_pgt + (L4_VMEMMAP_START*8)(%rip)
+ 	addq	%rbp, init_level4_pgt + (L4_START_KERNEL*8)(%rip)
+ 
+ 	addq	%rbp, level3_ident_pgt + 0(%rip)
++#ifndef CONFIG_XEN
++	addq	%rbp, level3_ident_pgt + 8(%rip)
++#endif
+ 
+-	addq	%rbp, level3_kernel_pgt + (510*8)(%rip)
+-	addq	%rbp, level3_kernel_pgt + (511*8)(%rip)
++	addq	%rbp, level3_vmemmap_pgt + (L3_VMEMMAP_START*8)(%rip)
++
++	addq	%rbp, level3_kernel_pgt + (L3_START_KERNEL*8)(%rip)
++	addq	%rbp, level3_kernel_pgt + (L3_START_KERNEL*8+8)(%rip)
+ 
+ 	addq	%rbp, level2_fixmap_pgt + (506*8)(%rip)
+-
+-	/* Add an Identity mapping if I am above 1G */
+-	leaq	_text(%rip), %rdi
+-	andq	$PMD_PAGE_MASK, %rdi
+-
+-	movq	%rdi, %rax
+-	shrq	$PUD_SHIFT, %rax
+-	andq	$(PTRS_PER_PUD - 1), %rax
+-	jz	ident_complete
+-
+-	leaq	(level2_spare_pgt - __START_KERNEL_map + _KERNPG_TABLE)(%rbp), %rdx
+-	leaq	level3_ident_pgt(%rip), %rbx
+-	movq	%rdx, 0(%rbx, %rax, 8)
+-
+-	movq	%rdi, %rax
+-	shrq	$PMD_SHIFT, %rax
+-	andq	$(PTRS_PER_PMD - 1), %rax
+-	leaq	__PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx
+-	leaq	level2_spare_pgt(%rip), %rbx
+-	movq	%rdx, 0(%rbx, %rax, 8)
+-ident_complete:
++	addq	%rbp, level2_fixmap_pgt + (507*8)(%rip)
+ 
+ 	/*
+ 	 * Fixup the kernel text+data virtual addresses. Note that
+@@ -160,8 +156,8 @@ ENTRY(secondary_startup_64)
+ 	 * after the boot processor executes this code.
+ 	 */
+ 
+-	/* Enable PAE mode and PGE */
+-	movl	$(X86_CR4_PAE | X86_CR4_PGE), %eax
++	/* Enable PAE mode and PSE/PGE */
++	movl	$(X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE), %eax
+ 	movq	%rax, %cr4
+ 
+ 	/* Setup early boot stage 4 level pagetables. */
+@@ -183,9 +179,17 @@ ENTRY(secondary_startup_64)
+ 	movl	$MSR_EFER, %ecx
+ 	rdmsr
+ 	btsl	$_EFER_SCE, %eax	/* Enable System Call */
+-	btl	$20,%edi		/* No Execute supported? */
++	btl	$(X86_FEATURE_NX & 31),%edi	/* No Execute supported? */
+ 	jnc     1f
+ 	btsl	$_EFER_NX, %eax
++	leaq	init_level4_pgt(%rip), %rdi
++#ifndef CONFIG_EFI
++	btsq	$_PAGE_BIT_NX, 8*L4_PAGE_OFFSET(%rdi)
++#endif
++	btsq	$_PAGE_BIT_NX, 8*L4_VMALLOC_START(%rdi)
++	btsq	$_PAGE_BIT_NX, 8*L4_VMALLOC_END(%rdi)
++	btsq	$_PAGE_BIT_NX, 8*L4_VMEMMAP_START(%rdi)
++	btsq	$_PAGE_BIT_NX, __supported_pte_mask(%rip)
+ 1:	wrmsr				/* Make changes effective */
+ 
+ 	/* Setup cr0 */
+@@ -247,6 +251,7 @@ ENTRY(secondary_startup_64)
+ 	 * jump.  In addition we need to ensure %cs is set so we make this
+ 	 * a far return.
+ 	 */
++	pax_set_fptr_mask
+ 	movq	initial_code(%rip),%rax
+ 	pushq	$0		# fake return address to stop unwinder
+ 	pushq	$__KERNEL_CS	# set correct cs
+@@ -269,7 +274,7 @@ ENTRY(secondary_startup_64)
+ bad_address:
+ 	jmp bad_address
+ 
+-	.section ".init.text","ax"
++	__INIT
+ #ifdef CONFIG_EARLY_PRINTK
+ 	.globl early_idt_handlers
+ early_idt_handlers:
+@@ -314,18 +319,23 @@ ENTRY(early_idt_handler)
+ #endif /* EARLY_PRINTK */
+ 1:	hlt
+ 	jmp 1b
++	.previous
+ 
+ #ifdef CONFIG_EARLY_PRINTK
++	__INITDATA
+ early_recursion_flag:
+ 	.long 0
++	.previous
+ 
++	.section .rodata,"a",@progbits
+ early_idt_msg:
+ 	.asciz "PANIC: early exception %02lx rip %lx:%lx error %lx cr2 %lx\n"
+ early_idt_ripmsg:
+ 	.asciz "RIP %s\n"
++	.previous
+ #endif /* CONFIG_EARLY_PRINTK */
+-	.previous
+ 
++	.section .rodata,"a",@progbits
+ #define NEXT_PAGE(name) \
+ 	.balign	PAGE_SIZE; \
+ ENTRY(name)
+@@ -338,7 +348,6 @@ ENTRY(name)
+ 	i = i + 1 ;					\
+ 	.endr
+ 
+-	.data
+ 	/*
+ 	 * This default setting generates an ident mapping at address 0x100000
+ 	 * and a mapping for the kernel that precisely maps virtual address
+@@ -349,13 +358,41 @@ NEXT_PAGE(init_level4_pgt)
+ 	.quad	level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
+ 	.org	init_level4_pgt + L4_PAGE_OFFSET*8, 0
+ 	.quad	level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
++	.org	init_level4_pgt + L4_VMALLOC_START*8, 0
++	.quad	level3_vmalloc_start_pgt - __START_KERNEL_map + _KERNPG_TABLE
++	.org	init_level4_pgt + L4_VMALLOC_END*8, 0
++	.quad	level3_vmalloc_end_pgt - __START_KERNEL_map + _KERNPG_TABLE
++	.org	init_level4_pgt + L4_VMEMMAP_START*8, 0
++	.quad	level3_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE
+ 	.org	init_level4_pgt + L4_START_KERNEL*8, 0
+ 	/* (2^48-(2*1024*1024*1024))/(2^39) = 511 */
+ 	.quad	level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE
+ 
++#ifdef CONFIG_PAX_PER_CPU_PGD
++NEXT_PAGE(cpu_pgd)
++	.rept NR_CPUS
++	.fill	512,8,0
++	.endr
++#endif
++
+ NEXT_PAGE(level3_ident_pgt)
+ 	.quad	level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
++#ifdef CONFIG_XEN
+ 	.fill	511,8,0
++#else
++	.quad	level2_ident_pgt + PAGE_SIZE - __START_KERNEL_map + _KERNPG_TABLE
++	.fill	510,8,0
++#endif
++
++NEXT_PAGE(level3_vmalloc_start_pgt)
++	.fill	512,8,0
++
++NEXT_PAGE(level3_vmalloc_end_pgt)
++	.fill	512,8,0
++
++NEXT_PAGE(level3_vmemmap_pgt)
++	.fill	L3_VMEMMAP_START,8,0
++	.quad	level2_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE
+ 
+ NEXT_PAGE(level3_kernel_pgt)
+ 	.fill	L3_START_KERNEL,8,0
+@@ -363,20 +400,23 @@ NEXT_PAGE(level3_kernel_pgt)
+ 	.quad	level2_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE
+ 	.quad	level2_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE
+ 
++NEXT_PAGE(level2_vmemmap_pgt)
++	.fill	512,8,0
++
+ NEXT_PAGE(level2_fixmap_pgt)
+-	.fill	506,8,0
+-	.quad	level1_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE
+-	/* 8MB reserved for vsyscalls + a 2MB hole = 4 + 1 entries */
+-	.fill	5,8,0
++	.fill	507,8,0
++	.quad	level1_vsyscall_pgt - __START_KERNEL_map + _PAGE_TABLE
++	/* 6MB reserved for vsyscalls + a 2MB hole = 3 + 1 entries */
++	.fill	4,8,0
+ 
+-NEXT_PAGE(level1_fixmap_pgt)
++NEXT_PAGE(level1_vsyscall_pgt)
+ 	.fill	512,8,0
+ 
+-NEXT_PAGE(level2_ident_pgt)
+-	/* Since I easily can, map the first 1G.
++	/* Since I easily can, map the first 2G.
+ 	 * Don't set NX because code runs from these pages.
+ 	 */
+-	PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD)
++NEXT_PAGE(level2_ident_pgt)
++	PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, 2*PTRS_PER_PMD)
+ 
+ NEXT_PAGE(level2_kernel_pgt)
+ 	/*
+@@ -389,33 +429,55 @@ NEXT_PAGE(level2_kernel_pgt)
+ 	 *  If you want to increase this then increase MODULES_VADDR
+ 	 *  too.)
+ 	 */
+-	PMDS(0, __PAGE_KERNEL_LARGE_EXEC,
+-		KERNEL_IMAGE_SIZE/PMD_SIZE)
+-
+-NEXT_PAGE(level2_spare_pgt)
+-	.fill   512, 8, 0
++	PMDS(0, __PAGE_KERNEL_LARGE_EXEC, KERNEL_IMAGE_SIZE/PMD_SIZE)
+ 
+ #undef PMDS
+ #undef NEXT_PAGE
+ 
+-	.data
++	.align PAGE_SIZE
++ENTRY(cpu_gdt_table)
++	.rept NR_CPUS
++	.quad	0x0000000000000000	/* NULL descriptor */
++	.quad	0x00cf9b000000ffff	/* __KERNEL32_CS */
++	.quad	0x00af9b000000ffff	/* __KERNEL_CS */
++	.quad	0x00cf93000000ffff	/* __KERNEL_DS */
++	.quad	0x00cffb000000ffff	/* __USER32_CS */
++	.quad	0x00cff3000000ffff	/* __USER_DS, __USER32_DS  */
++	.quad	0x00affb000000ffff	/* __USER_CS */
++
++#ifdef CONFIG_PAX_KERNEXEC
++	.quad	0x00af9b000000ffff	/* __KERNEXEC_KERNEL_CS */
++#else
++	.quad	0x0			/* unused */
++#endif
++
++	.quad	0,0			/* TSS */
++	.quad	0,0			/* LDT */
++	.quad	0,0,0			/* three TLS descriptors */
++	.quad	0x0000f40000000000	/* node/CPU stored in limit */
++	/* asm/segment.h:GDT_ENTRIES must match this */
++
++	/* zero the remaining page */
++	.fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0
++	.endr
++
+ 	.align 16
+ 	.globl early_gdt_descr
+ early_gdt_descr:
+ 	.word	GDT_ENTRIES*8-1
+ early_gdt_descr_base:
+-	.quad	INIT_PER_CPU_VAR(gdt_page)
++	.quad	cpu_gdt_table
+ 
+ ENTRY(phys_base)
+ 	/* This must match the first entry in level2_kernel_pgt */
+ 	.quad   0x0000000000000000
+ 
+ #include "../../x86/xen/xen-head.S"
+-	
+-	.section .bss, "aw", @nobits
++
++	.section .rodata,"a",@progbits
+ 	.align L1_CACHE_BYTES
+ ENTRY(idt_table)
+-	.skip IDT_ENTRIES * 16
++	.fill 512,8,0
+ 
+ 	__PAGE_ALIGNED_BSS
+ 	.align PAGE_SIZE
+diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c
+index 9c3bd4a..e1d9b35 100644
+--- a/arch/x86/kernel/i386_ksyms_32.c
++++ b/arch/x86/kernel/i386_ksyms_32.c
+@@ -20,8 +20,12 @@ extern void cmpxchg8b_emu(void);
+ EXPORT_SYMBOL(cmpxchg8b_emu);
+ #endif
+ 
++EXPORT_SYMBOL_GPL(cpu_gdt_table);
++
+ /* Networking helper routines. */
+ EXPORT_SYMBOL(csum_partial_copy_generic);
++EXPORT_SYMBOL(csum_partial_copy_generic_to_user);
++EXPORT_SYMBOL(csum_partial_copy_generic_from_user);
+ 
+ EXPORT_SYMBOL(__get_user_1);
+ EXPORT_SYMBOL(__get_user_2);
+@@ -36,3 +40,7 @@ EXPORT_SYMBOL(strstr);
+ 
+ EXPORT_SYMBOL(csum_partial);
+ EXPORT_SYMBOL(empty_zero_page);
++
++#ifdef CONFIG_PAX_KERNEXEC
++EXPORT_SYMBOL(__LOAD_PHYSICAL_ADDR);
++#endif
+diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
+index 739d859..d1d6be7 100644
+--- a/arch/x86/kernel/i387.c
++++ b/arch/x86/kernel/i387.c
+@@ -188,6 +188,9 @@ int xfpregs_active(struct task_struct *target, const struct user_regset *regset)
+ 
+ int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
+ 		unsigned int pos, unsigned int count,
++		void *kbuf, void __user *ubuf) __size_overflow(4);
++int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
++		unsigned int pos, unsigned int count,
+ 		void *kbuf, void __user *ubuf)
+ {
+ 	int ret;
+@@ -207,6 +210,9 @@ int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
+ 
+ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
+ 		unsigned int pos, unsigned int count,
++		const void *kbuf, const void __user *ubuf) __size_overflow(4);
++int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
++		unsigned int pos, unsigned int count,
+ 		const void *kbuf, const void __user *ubuf)
+ {
+ 	int ret;
+@@ -240,6 +246,9 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
+ 
+ int xstateregs_get(struct task_struct *target, const struct user_regset *regset,
+ 		unsigned int pos, unsigned int count,
++		void *kbuf, void __user *ubuf) __size_overflow(4);
++int xstateregs_get(struct task_struct *target, const struct user_regset *regset,
++		unsigned int pos, unsigned int count,
+ 		void *kbuf, void __user *ubuf)
+ {
+ 	int ret;
+@@ -269,6 +278,9 @@ int xstateregs_get(struct task_struct *target, const struct user_regset *regset,
+ 
+ int xstateregs_set(struct task_struct *target, const struct user_regset *regset,
+ 		  unsigned int pos, unsigned int count,
++		  const void *kbuf, const void __user *ubuf) __size_overflow(4);
++int xstateregs_set(struct task_struct *target, const struct user_regset *regset,
++		  unsigned int pos, unsigned int count,
+ 		  const void *kbuf, const void __user *ubuf)
+ {
+ 	int ret;
+@@ -439,6 +451,9 @@ static void convert_to_fxsr(struct task_struct *tsk,
+ 
+ int fpregs_get(struct task_struct *target, const struct user_regset *regset,
+ 	       unsigned int pos, unsigned int count,
++	       void *kbuf, void __user *ubuf) __size_overflow(3,4);
++int fpregs_get(struct task_struct *target, const struct user_regset *regset,
++	       unsigned int pos, unsigned int count,
+ 	       void *kbuf, void __user *ubuf)
+ {
+ 	struct user_i387_ia32_struct env;
+@@ -471,6 +486,9 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset,
+ 
+ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
+ 	       unsigned int pos, unsigned int count,
++	       const void *kbuf, const void __user *ubuf) __size_overflow(3,4);
++int fpregs_set(struct task_struct *target, const struct user_regset *regset,
++	       unsigned int pos, unsigned int count,
+ 	       const void *kbuf, const void __user *ubuf)
+ {
+ 	struct user_i387_ia32_struct env;
+@@ -619,6 +637,8 @@ static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf)
+ }
+ 
+ static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf,
++			       unsigned int size) __size_overflow(2);
++static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf,
+ 			       unsigned int size)
+ {
+ 	struct task_struct *tsk = current;
+diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c
+index 6104852..6114160 100644
+--- a/arch/x86/kernel/i8259.c
++++ b/arch/x86/kernel/i8259.c
+@@ -210,7 +210,7 @@ spurious_8259A_irq:
+ 			       "spurious 8259A interrupt: IRQ%d.\n", irq);
+ 			spurious_irq_mask |= irqmask;
+ 		}
+-		atomic_inc(&irq_err_count);
++		atomic_inc_unchecked(&irq_err_count);
+ 		/*
+ 		 * Theoretically we do not have to handle this IRQ,
+ 		 * but in Linux this does not cause problems and is
+diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
+index 43e9ccf..44ccf6f 100644
+--- a/arch/x86/kernel/init_task.c
++++ b/arch/x86/kernel/init_task.c
+@@ -20,8 +20,7 @@ static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+  * way process stacks are handled. This is done by having a special
+  * "init_task" linker map entry..
+  */
+-union thread_union init_thread_union __init_task_data =
+-	{ INIT_THREAD_INFO(init_task) };
++union thread_union init_thread_union __init_task_data;
+ 
+ /*
+  * Initial task structure.
+@@ -38,5 +37,5 @@ EXPORT_SYMBOL(init_task);
+  * section. Since TSS's are completely CPU-local, we want them
+  * on exact cacheline boundaries, to eliminate cacheline ping-pong.
+  */
+-DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS;
+-
++struct tss_struct init_tss[NR_CPUS] ____cacheline_internodealigned_in_smp = { [0 ... NR_CPUS-1] = INIT_TSS };
++EXPORT_SYMBOL(init_tss);
+diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c
+index 8c96897..be66bfa 100644
+--- a/arch/x86/kernel/ioport.c
++++ b/arch/x86/kernel/ioport.c
+@@ -6,6 +6,7 @@
+ #include <linux/sched.h>
+ #include <linux/kernel.h>
+ #include <linux/capability.h>
++#include <linux/security.h>
+ #include <linux/errno.h>
+ #include <linux/types.h>
+ #include <linux/ioport.h>
+@@ -28,6 +29,12 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
+ 
+ 	if ((from + num <= from) || (from + num > IO_BITMAP_BITS))
+ 		return -EINVAL;
++#ifdef CONFIG_GRKERNSEC_IO
++	if (turn_on && grsec_disable_privio) {
++		gr_handle_ioperm();
++		return -EPERM;
++	}
++#endif
+ 	if (turn_on && !capable(CAP_SYS_RAWIO))
+ 		return -EPERM;
+ 
+@@ -54,7 +61,7 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
+ 	 * because the ->io_bitmap_max value must match the bitmap
+ 	 * contents:
+ 	 */
+-	tss = &per_cpu(init_tss, get_cpu());
++	tss = init_tss + get_cpu();
+ 
+ 	if (turn_on)
+ 		bitmap_clear(t->io_bitmap_ptr, from, num);
+@@ -102,6 +109,12 @@ long sys_iopl(unsigned int level, struct pt_regs *regs)
+ 		return -EINVAL;
+ 	/* Trying to gain more privileges? */
+ 	if (level > old) {
++#ifdef CONFIG_GRKERNSEC_IO
++		if (grsec_disable_privio) {
++			gr_handle_iopl();
++			return -EPERM;
++		}
++#endif
+ 		if (!capable(CAP_SYS_RAWIO))
+ 			return -EPERM;
+ 	}
+diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
+index 429e0c9..17b3ece 100644
+--- a/arch/x86/kernel/irq.c
++++ b/arch/x86/kernel/irq.c
+@@ -18,7 +18,7 @@
+ #include <asm/mce.h>
+ #include <asm/hw_irq.h>
+ 
+-atomic_t irq_err_count;
++atomic_unchecked_t irq_err_count;
+ 
+ /* Function pointer for generic interrupt vector handling */
+ void (*x86_platform_ipi_callback)(void) = NULL;
+@@ -117,9 +117,9 @@ int arch_show_interrupts(struct seq_file *p, int prec)
+ 		seq_printf(p, "%10u ", per_cpu(mce_poll_count, j));
+ 	seq_printf(p, "  Machine check polls\n");
+ #endif
+-	seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read(&irq_err_count));
++	seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read_unchecked(&irq_err_count));
+ #if defined(CONFIG_X86_IO_APIC)
+-	seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read(&irq_mis_count));
++	seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read_unchecked(&irq_mis_count));
+ #endif
+ 	return 0;
+ }
+@@ -159,10 +159,10 @@ u64 arch_irq_stat_cpu(unsigned int cpu)
+ 
+ u64 arch_irq_stat(void)
+ {
+-	u64 sum = atomic_read(&irq_err_count);
++	u64 sum = atomic_read_unchecked(&irq_err_count);
+ 
+ #ifdef CONFIG_X86_IO_APIC
+-	sum += atomic_read(&irq_mis_count);
++	sum += atomic_read_unchecked(&irq_mis_count);
+ #endif
+ 	return sum;
+ }
+diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
+index 7209070..cbcd71a 100644
+--- a/arch/x86/kernel/irq_32.c
++++ b/arch/x86/kernel/irq_32.c
+@@ -36,7 +36,7 @@ static int check_stack_overflow(void)
+ 	__asm__ __volatile__("andl %%esp,%0" :
+ 			     "=r" (sp) : "0" (THREAD_SIZE - 1));
+ 
+-	return sp < (sizeof(struct thread_info) + STACK_WARN);
++	return sp < STACK_WARN;
+ }
+ 
+ static void print_stack_overflow(void)
+@@ -54,8 +54,8 @@ static inline void print_stack_overflow(void) { }
+  * per-CPU IRQ handling contexts (thread information and stack)
+  */
+ union irq_ctx {
+-	struct thread_info      tinfo;
+-	u32                     stack[THREAD_SIZE/sizeof(u32)];
++	unsigned long		previous_esp;
++	u32			stack[THREAD_SIZE/sizeof(u32)];
+ } __attribute__((aligned(THREAD_SIZE)));
+ 
+ static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx);
+@@ -75,10 +75,9 @@ static void call_on_stack(void *func, void *stack)
+ static inline int
+ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
+ {
+-	union irq_ctx *curctx, *irqctx;
++	union irq_ctx *irqctx;
+ 	u32 *isp, arg1, arg2;
+ 
+-	curctx = (union irq_ctx *) current_thread_info();
+ 	irqctx = __this_cpu_read(hardirq_ctx);
+ 
+ 	/*
+@@ -87,21 +86,16 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
+ 	 * handler) we can't do that and just have to keep using the
+ 	 * current stack (which is the irq stack already after all)
+ 	 */
+-	if (unlikely(curctx == irqctx))
++	if (unlikely((void *)current_stack_pointer - (void *)irqctx < THREAD_SIZE))
+ 		return 0;
+ 
+ 	/* build the stack frame on the IRQ stack */
+-	isp = (u32 *) ((char *)irqctx + sizeof(*irqctx));
+-	irqctx->tinfo.task = curctx->tinfo.task;
+-	irqctx->tinfo.previous_esp = current_stack_pointer;
++	isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8);
++	irqctx->previous_esp = current_stack_pointer;
+ 
+-	/*
+-	 * Copy the softirq bits in preempt_count so that the
+-	 * softirq checks work in the hardirq context.
+-	 */
+-	irqctx->tinfo.preempt_count =
+-		(irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) |
+-		(curctx->tinfo.preempt_count & SOFTIRQ_MASK);
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	__set_fs(MAKE_MM_SEG(0));
++#endif
+ 
+ 	if (unlikely(overflow))
+ 		call_on_stack(print_stack_overflow, isp);
+@@ -113,6 +107,11 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
+ 		     :  "0" (irq),   "1" (desc),  "2" (isp),
+ 			"D" (desc->handle_irq)
+ 		     : "memory", "cc", "ecx");
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	__set_fs(current_thread_info()->addr_limit);
++#endif
++
+ 	return 1;
+ }
+ 
+@@ -121,29 +120,11 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
+  */
+ void __cpuinit irq_ctx_init(int cpu)
+ {
+-	union irq_ctx *irqctx;
+-
+ 	if (per_cpu(hardirq_ctx, cpu))
+ 		return;
+ 
+-	irqctx = page_address(alloc_pages_node(cpu_to_node(cpu),
+-					       THREAD_FLAGS,
+-					       THREAD_ORDER));
+-	memset(&irqctx->tinfo, 0, sizeof(struct thread_info));
+-	irqctx->tinfo.cpu		= cpu;
+-	irqctx->tinfo.preempt_count	= HARDIRQ_OFFSET;
+-	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+-
+-	per_cpu(hardirq_ctx, cpu) = irqctx;
+-
+-	irqctx = page_address(alloc_pages_node(cpu_to_node(cpu),
+-					       THREAD_FLAGS,
+-					       THREAD_ORDER));
+-	memset(&irqctx->tinfo, 0, sizeof(struct thread_info));
+-	irqctx->tinfo.cpu		= cpu;
+-	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+-
+-	per_cpu(softirq_ctx, cpu) = irqctx;
++	per_cpu(hardirq_ctx, cpu) = page_address(alloc_pages_node(cpu_to_node(cpu), THREAD_FLAGS, THREAD_ORDER));
++	per_cpu(softirq_ctx, cpu) = page_address(alloc_pages_node(cpu_to_node(cpu), THREAD_FLAGS, THREAD_ORDER));
+ 
+ 	printk(KERN_DEBUG "CPU %u irqstacks, hard=%p soft=%p\n",
+ 	       cpu, per_cpu(hardirq_ctx, cpu),  per_cpu(softirq_ctx, cpu));
+@@ -152,7 +133,6 @@ void __cpuinit irq_ctx_init(int cpu)
+ asmlinkage void do_softirq(void)
+ {
+ 	unsigned long flags;
+-	struct thread_info *curctx;
+ 	union irq_ctx *irqctx;
+ 	u32 *isp;
+ 
+@@ -162,15 +142,22 @@ asmlinkage void do_softirq(void)
+ 	local_irq_save(flags);
+ 
+ 	if (local_softirq_pending()) {
+-		curctx = current_thread_info();
+ 		irqctx = __this_cpu_read(softirq_ctx);
+-		irqctx->tinfo.task = curctx->task;
+-		irqctx->tinfo.previous_esp = current_stack_pointer;
++		irqctx->previous_esp = current_stack_pointer;
+ 
+ 		/* build the stack frame on the softirq stack */
+-		isp = (u32 *) ((char *)irqctx + sizeof(*irqctx));
++		isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		__set_fs(MAKE_MM_SEG(0));
++#endif
+ 
+ 		call_on_stack(__do_softirq, isp);
++
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++		__set_fs(current_thread_info()->addr_limit);
++#endif
++
+ 		/*
+ 		 * Shouldn't happen, we returned above if in_interrupt():
+ 		 */
+diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
+index 69bca46..0bac999 100644
+--- a/arch/x86/kernel/irq_64.c
++++ b/arch/x86/kernel/irq_64.c
+@@ -38,7 +38,7 @@ static inline void stack_overflow_check(struct pt_regs *regs)
+ #ifdef CONFIG_DEBUG_STACKOVERFLOW
+ 	u64 curbase = (u64)task_stack_page(current);
+ 
+-	if (user_mode_vm(regs))
++	if (user_mode(regs))
+ 		return;
+ 
+ 	WARN_ONCE(regs->sp >= curbase &&
+diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
+index faba577..93b9e71 100644
+--- a/arch/x86/kernel/kgdb.c
++++ b/arch/x86/kernel/kgdb.c
+@@ -124,11 +124,11 @@ char *dbg_get_reg(int regno, void *mem, struct pt_regs *regs)
+ #ifdef CONFIG_X86_32
+ 	switch (regno) {
+ 	case GDB_SS:
+-		if (!user_mode_vm(regs))
++		if (!user_mode(regs))
+ 			*(unsigned long *)mem = __KERNEL_DS;
+ 		break;
+ 	case GDB_SP:
+-		if (!user_mode_vm(regs))
++		if (!user_mode(regs))
+ 			*(unsigned long *)mem = kernel_stack_pointer(regs);
+ 		break;
+ 	case GDB_GS:
+@@ -473,12 +473,12 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code,
+ 	case 'k':
+ 		/* clear the trace bit */
+ 		linux_regs->flags &= ~X86_EFLAGS_TF;
+-		atomic_set(&kgdb_cpu_doing_single_step, -1);
++		atomic_set_unchecked(&kgdb_cpu_doing_single_step, -1);
+ 
+ 		/* set the trace bit if we're stepping */
+ 		if (remcomInBuffer[0] == 's') {
+ 			linux_regs->flags |= X86_EFLAGS_TF;
+-			atomic_set(&kgdb_cpu_doing_single_step,
++			atomic_set_unchecked(&kgdb_cpu_doing_single_step,
+ 				   raw_smp_processor_id());
+ 		}
+ 
+@@ -543,7 +543,7 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
+ 
+ 	switch (cmd) {
+ 	case DIE_DEBUG:
+-		if (atomic_read(&kgdb_cpu_doing_single_step) != -1) {
++		if (atomic_read_unchecked(&kgdb_cpu_doing_single_step) != -1) {
+ 			if (user_mode(regs))
+ 				return single_step_cont(regs, args);
+ 			break;
+diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
+index 7da647d..56fe348 100644
+--- a/arch/x86/kernel/kprobes.c
++++ b/arch/x86/kernel/kprobes.c
+@@ -118,8 +118,11 @@ static void __kprobes __synthesize_relative_insn(void *from, void *to, u8 op)
+ 	} __attribute__((packed)) *insn;
+ 
+ 	insn = (struct __arch_relative_insn *)from;
++
++	pax_open_kernel();
+ 	insn->raddr = (s32)((long)(to) - ((long)(from) + 5));
+ 	insn->op = op;
++	pax_close_kernel();
+ }
+ 
+ /* Insert a jump instruction at address 'from', which jumps to address 'to'.*/
+@@ -156,7 +159,7 @@ static int __kprobes can_boost(kprobe_opcode_t *opcodes)
+ 	kprobe_opcode_t opcode;
+ 	kprobe_opcode_t *orig_opcodes = opcodes;
+ 
+-	if (search_exception_tables((unsigned long)opcodes))
++	if (search_exception_tables(ktva_ktla((unsigned long)opcodes)))
+ 		return 0;	/* Page fault may occur on this address. */
+ 
+ retry:
+@@ -317,7 +320,9 @@ static int __kprobes __copy_instruction(u8 *dest, u8 *src, int recover)
+ 		}
+ 	}
+ 	insn_get_length(&insn);
++	pax_open_kernel();
+ 	memcpy(dest, insn.kaddr, insn.length);
++	pax_close_kernel();
+ 
+ #ifdef CONFIG_X86_64
+ 	if (insn_rip_relative(&insn)) {
+@@ -341,7 +346,9 @@ static int __kprobes __copy_instruction(u8 *dest, u8 *src, int recover)
+ 			  (u8 *) dest;
+ 		BUG_ON((s64) (s32) newdisp != newdisp); /* Sanity check.  */
+ 		disp = (u8 *) dest + insn_offset_displacement(&insn);
++		pax_open_kernel();
+ 		*(s32 *) disp = (s32) newdisp;
++		pax_close_kernel();
+ 	}
+ #endif
+ 	return insn.length;
+@@ -355,12 +362,12 @@ static void __kprobes arch_copy_kprobe(struct kprobe *p)
+ 	 */
+ 	__copy_instruction(p->ainsn.insn, p->addr, 0);
+ 
+-	if (can_boost(p->addr))
++	if (can_boost(ktla_ktva(p->addr)))
+ 		p->ainsn.boostable = 0;
+ 	else
+ 		p->ainsn.boostable = -1;
+ 
+-	p->opcode = *p->addr;
++	p->opcode = *(ktla_ktva(p->addr));
+ }
+ 
+ int __kprobes arch_prepare_kprobe(struct kprobe *p)
+@@ -477,7 +484,7 @@ static void __kprobes setup_singlestep(struct kprobe *p, struct pt_regs *regs,
+ 		 * nor set current_kprobe, because it doesn't use single
+ 		 * stepping.
+ 		 */
+-		regs->ip = (unsigned long)p->ainsn.insn;
++		regs->ip = ktva_ktla((unsigned long)p->ainsn.insn);
+ 		preempt_enable_no_resched();
+ 		return;
+ 	}
+@@ -496,7 +503,7 @@ static void __kprobes setup_singlestep(struct kprobe *p, struct pt_regs *regs,
+ 	if (p->opcode == BREAKPOINT_INSTRUCTION)
+ 		regs->ip = (unsigned long)p->addr;
+ 	else
+-		regs->ip = (unsigned long)p->ainsn.insn;
++		regs->ip = ktva_ktla((unsigned long)p->ainsn.insn);
+ }
+ 
+ /*
+@@ -575,7 +582,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
+ 				setup_singlestep(p, regs, kcb, 0);
+ 			return 1;
+ 		}
+-	} else if (*addr != BREAKPOINT_INSTRUCTION) {
++	} else if (*(kprobe_opcode_t *)ktla_ktva((unsigned long)addr) != BREAKPOINT_INSTRUCTION) {
+ 		/*
+ 		 * The breakpoint instruction was removed right
+ 		 * after we hit it.  Another cpu has removed
+@@ -683,6 +690,9 @@ static void __used __kprobes kretprobe_trampoline_holder(void)
+ 			"	movq %rax, 152(%rsp)\n"
+ 			RESTORE_REGS_STRING
+ 			"	popfq\n"
++#ifdef KERNEXEC_PLUGIN
++			"	btsq $63,(%rsp)\n"
++#endif
+ #else
+ 			"	pushf\n"
+ 			SAVE_REGS_STRING
+@@ -820,7 +830,7 @@ static void __kprobes resume_execution(struct kprobe *p,
+ 		struct pt_regs *regs, struct kprobe_ctlblk *kcb)
+ {
+ 	unsigned long *tos = stack_addr(regs);
+-	unsigned long copy_ip = (unsigned long)p->ainsn.insn;
++	unsigned long copy_ip = ktva_ktla((unsigned long)p->ainsn.insn);
+ 	unsigned long orig_ip = (unsigned long)p->addr;
+ 	kprobe_opcode_t *insn = p->ainsn.insn;
+ 
+@@ -1002,7 +1012,7 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+ 	struct die_args *args = data;
+ 	int ret = NOTIFY_DONE;
+ 
+-	if (args->regs && user_mode_vm(args->regs))
++	if (args->regs && user_mode(args->regs))
+ 		return ret;
+ 
+ 	switch (val) {
+@@ -1384,7 +1394,7 @@ int __kprobes arch_prepare_optimized_kprobe(struct optimized_kprobe *op)
+ 	 * Verify if the address gap is in 2GB range, because this uses
+ 	 * a relative jump.
+ 	 */
+-	rel = (long)op->optinsn.insn - (long)op->kp.addr + RELATIVEJUMP_SIZE;
++	rel = (long)op->optinsn.insn - ktla_ktva((long)op->kp.addr) + RELATIVEJUMP_SIZE;
+ 	if (abs(rel) > 0x7fffffff)
+ 		return -ERANGE;
+ 
+@@ -1405,11 +1415,11 @@ int __kprobes arch_prepare_optimized_kprobe(struct optimized_kprobe *op)
+ 	synthesize_set_arg1(buf + TMPL_MOVE_IDX, (unsigned long)op);
+ 
+ 	/* Set probe function call */
+-	synthesize_relcall(buf + TMPL_CALL_IDX, optimized_callback);
++	synthesize_relcall(buf + TMPL_CALL_IDX, ktla_ktva(optimized_callback));
+ 
+ 	/* Set returning jmp instruction at the tail of out-of-line buffer */
+ 	synthesize_reljump(buf + TMPL_END_IDX + op->optinsn.size,
+-			   (u8 *)op->kp.addr + op->optinsn.size);
++			   (u8 *)ktla_ktva(op->kp.addr) + op->optinsn.size);
+ 
+ 	flush_icache_range((unsigned long) buf,
+ 			   (unsigned long) buf + TMPL_END_IDX +
+@@ -1431,7 +1441,7 @@ static void __kprobes setup_optimize_kprobe(struct text_poke_param *tprm,
+ 			((long)op->kp.addr + RELATIVEJUMP_SIZE));
+ 
+ 	/* Backup instructions which will be replaced by jump address */
+-	memcpy(op->optinsn.copied_insn, op->kp.addr + INT3_SIZE,
++	memcpy(op->optinsn.copied_insn, ktla_ktva(op->kp.addr) + INT3_SIZE,
+ 	       RELATIVE_ADDR_SIZE);
+ 
+ 	insn_buf[0] = RELATIVEJUMP_OPCODE;
+diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
+index a9c2116..a52d4fc 100644
+--- a/arch/x86/kernel/kvm.c
++++ b/arch/x86/kernel/kvm.c
+@@ -437,6 +437,7 @@ static void __init paravirt_ops_setup(void)
+ 		pv_mmu_ops.set_pud = kvm_set_pud;
+ #if PAGETABLE_LEVELS == 4
+ 		pv_mmu_ops.set_pgd = kvm_set_pgd;
++		pv_mmu_ops.set_pgd_batched = kvm_set_pgd;
+ #endif
+ #endif
+ 		pv_mmu_ops.flush_tlb_user = kvm_flush_tlb;
+diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
+index ea69726..8b497c9 100644
+--- a/arch/x86/kernel/ldt.c
++++ b/arch/x86/kernel/ldt.c
+@@ -67,13 +67,13 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
+ 	if (reload) {
+ #ifdef CONFIG_SMP
+ 		preempt_disable();
+-		load_LDT(pc);
++		load_LDT_nolock(pc);
+ 		if (!cpumask_equal(mm_cpumask(current->mm),
+ 				   cpumask_of(smp_processor_id())))
+ 			smp_call_function(flush_ldt, current->mm, 1);
+ 		preempt_enable();
+ #else
+-		load_LDT(pc);
++		load_LDT_nolock(pc);
+ #endif
+ 	}
+ 	if (oldsize) {
+@@ -95,7 +95,7 @@ static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
+ 		return err;
+ 
+ 	for (i = 0; i < old->size; i++)
+-		write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE);
++		write_ldt_entry(new->ldt, i, old->ldt + i);
+ 	return 0;
+ }
+ 
+@@ -116,6 +116,24 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
+ 		retval = copy_ldt(&mm->context, &old_mm->context);
+ 		mutex_unlock(&old_mm->context.lock);
+ 	}
++
++	if (tsk == current) {
++		mm->context.vdso = 0;
++
++#ifdef CONFIG_X86_32
++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)
++		mm->context.user_cs_base = 0UL;
++		mm->context.user_cs_limit = ~0UL;
++
++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP)
++		cpus_clear(mm->context.cpu_user_cs_mask);
++#endif
++
++#endif
++#endif
++
++	}
++
+ 	return retval;
+ }
+ 
+@@ -141,6 +159,7 @@ void destroy_context(struct mm_struct *mm)
+ 	}
+ }
+ 
++static int read_ldt(void __user *ptr, unsigned long bytecount) __size_overflow(2);
+ static int read_ldt(void __user *ptr, unsigned long bytecount)
+ {
+ 	int err;
+@@ -230,6 +249,13 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
+ 		}
+ 	}
+ 
++#ifdef CONFIG_PAX_SEGMEXEC
++	if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (ldt_info.contents & MODIFY_LDT_CONTENTS_CODE)) {
++		error = -EINVAL;
++		goto out_unlock;
++	}
++#endif
++
+ 	fill_ldt(&ldt, &ldt_info);
+ 	if (oldmode)
+ 		ldt.avl = 0;
+diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
+index a3fa43b..8966f4c 100644
+--- a/arch/x86/kernel/machine_kexec_32.c
++++ b/arch/x86/kernel/machine_kexec_32.c
+@@ -27,7 +27,7 @@
+ #include <asm/cacheflush.h>
+ #include <asm/debugreg.h>
+ 
+-static void set_idt(void *newidt, __u16 limit)
++static void set_idt(struct desc_struct *newidt, __u16 limit)
+ {
+ 	struct desc_ptr curidt;
+ 
+@@ -39,7 +39,7 @@ static void set_idt(void *newidt, __u16 limit)
+ }
+ 
+ 
+-static void set_gdt(void *newgdt, __u16 limit)
++static void set_gdt(struct desc_struct *newgdt, __u16 limit)
+ {
+ 	struct desc_ptr curgdt;
+ 
+@@ -217,7 +217,7 @@ void machine_kexec(struct kimage *image)
+ 	}
+ 
+ 	control_page = page_address(image->control_code_page);
+-	memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE);
++	memcpy(control_page, (void *)ktla_ktva((unsigned long)relocate_kernel), KEXEC_CONTROL_CODE_MAX_SIZE);
+ 
+ 	relocate_kernel_ptr = control_page;
+ 	page_list[PA_CONTROL_PAGE] = __pa(control_page);
+diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c
+index 3ca42d0..79d24cd 100644
+--- a/arch/x86/kernel/microcode_intel.c
++++ b/arch/x86/kernel/microcode_intel.c
+@@ -434,15 +434,16 @@ static enum ucode_state request_microcode_fw(int cpu, struct device *device)
+ 	return ret;
+ }
+ 
++static int get_ucode_user(void *to, const void *from, size_t n) __size_overflow(3);
+ static int get_ucode_user(void *to, const void *from, size_t n)
+ {
+-	return copy_from_user(to, from, n);
++	return copy_from_user(to, (const void __force_user *)from, n);
+ }
+ 
+ static enum ucode_state
+ request_microcode_user(int cpu, const void __user *buf, size_t size)
+ {
+-	return generic_load_microcode(cpu, (void *)buf, size, &get_ucode_user);
++	return generic_load_microcode(cpu, (__force_kernel void *)buf, size, &get_ucode_user);
+ }
+ 
+ static void microcode_fini_cpu(int cpu)
+diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
+index 925179f..267ac7a 100644
+--- a/arch/x86/kernel/module.c
++++ b/arch/x86/kernel/module.c
+@@ -36,15 +36,60 @@
+ #define DEBUGP(fmt...)
+ #endif
+ 
+-void *module_alloc(unsigned long size)
++static inline void *__module_alloc(unsigned long size, pgprot_t prot)
+ {
+-	if (PAGE_ALIGN(size) > MODULES_LEN)
++	if (size == 0 || PAGE_ALIGN(size) > MODULES_LEN)
+ 		return NULL;
+ 	return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END,
+-				GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC,
++				GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, prot,
+ 				-1, __builtin_return_address(0));
+ }
+ 
++void *module_alloc(unsigned long size)
++{
++
++#ifdef CONFIG_PAX_KERNEXEC
++	return __module_alloc(size, PAGE_KERNEL);
++#else
++	return __module_alloc(size, PAGE_KERNEL_EXEC);
++#endif
++
++}
++
++#ifdef CONFIG_PAX_KERNEXEC
++#ifdef CONFIG_X86_32
++void *module_alloc_exec(unsigned long size)
++{
++	struct vm_struct *area;
++
++	if (size == 0)
++		return NULL;
++
++	area = __get_vm_area(size, VM_ALLOC, (unsigned long)&MODULES_EXEC_VADDR, (unsigned long)&MODULES_EXEC_END);
++	return area ? area->addr : NULL;
++}
++EXPORT_SYMBOL(module_alloc_exec);
++
++void module_free_exec(struct module *mod, void *module_region)
++{
++	vunmap(module_region);
++}
++EXPORT_SYMBOL(module_free_exec);
++#else
++void module_free_exec(struct module *mod, void *module_region)
++{
++	module_free(mod, module_region);
++}
++EXPORT_SYMBOL(module_free_exec);
++
++void *module_alloc_exec(unsigned long size)
++{
++	return __module_alloc(size, PAGE_KERNEL_RX);
++}
++EXPORT_SYMBOL(module_alloc_exec);
++#endif
++#endif
++
+ #ifdef CONFIG_X86_32
+ int apply_relocate(Elf32_Shdr *sechdrs,
+ 		   const char *strtab,
+@@ -55,14 +100,16 @@ int apply_relocate(Elf32_Shdr *sechdrs,
+ 	unsigned int i;
+ 	Elf32_Rel *rel = (void *)sechdrs[relsec].sh_addr;
+ 	Elf32_Sym *sym;
+-	uint32_t *location;
++	uint32_t *plocation, location;
+ 
+ 	DEBUGP("Applying relocate section %u to %u\n", relsec,
+ 	       sechdrs[relsec].sh_info);
+ 	for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) {
+ 		/* This is where to make the change */
+-		location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr
+-			+ rel[i].r_offset;
++		plocation = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr + rel[i].r_offset;
++		location = (uint32_t)plocation;
++		if (sechdrs[sechdrs[relsec].sh_info].sh_flags & SHF_EXECINSTR)
++			plocation = ktla_ktva((void *)plocation);
+ 		/* This is the symbol it is referring to.  Note that all
+ 		   undefined symbols have been resolved.  */
+ 		sym = (Elf32_Sym *)sechdrs[symindex].sh_addr
+@@ -71,11 +118,15 @@ int apply_relocate(Elf32_Shdr *sechdrs,
+ 		switch (ELF32_R_TYPE(rel[i].r_info)) {
+ 		case R_386_32:
+ 			/* We add the value into the location given */
+-			*location += sym->st_value;
++			pax_open_kernel();
++			*plocation += sym->st_value;
++			pax_close_kernel();
+ 			break;
+ 		case R_386_PC32:
+ 			/* Add the value, subtract its postition */
+-			*location += sym->st_value - (uint32_t)location;
++			pax_open_kernel();
++			*plocation += sym->st_value - location;
++			pax_close_kernel();
+ 			break;
+ 		default:
+ 			printk(KERN_ERR "module %s: Unknown relocation: %u\n",
+@@ -120,21 +171,30 @@ int apply_relocate_add(Elf64_Shdr *sechdrs,
+ 		case R_X86_64_NONE:
+ 			break;
+ 		case R_X86_64_64:
++			pax_open_kernel();
+ 			*(u64 *)loc = val;
++			pax_close_kernel();
+ 			break;
+ 		case R_X86_64_32:
++			pax_open_kernel();
+ 			*(u32 *)loc = val;
++			pax_close_kernel();
+ 			if (val != *(u32 *)loc)
+ 				goto overflow;
+ 			break;
+ 		case R_X86_64_32S:
++			pax_open_kernel();
+ 			*(s32 *)loc = val;
++			pax_close_kernel();
+ 			if ((s64)val != *(s32 *)loc)
+ 				goto overflow;
+ 			break;
+ 		case R_X86_64_PC32:
+ 			val -= (u64)loc;
++			pax_open_kernel();
+ 			*(u32 *)loc = val;
++			pax_close_kernel();
++
+ #if 0
+ 			if ((s64)val != *(s32 *)loc)
+ 				goto overflow;
+diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
+index e88f37b..1353db6 100644
+--- a/arch/x86/kernel/nmi.c
++++ b/arch/x86/kernel/nmi.c
+@@ -408,6 +408,17 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
+ dotraplinkage notrace __kprobes void
+ do_nmi(struct pt_regs *regs, long error_code)
+ {
++
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC)
++	if (!user_mode(regs)) {
++		unsigned long cs = regs->cs & 0xFFFF;
++		unsigned long ip = ktva_ktla(regs->ip);
++
++		if ((cs == __KERNEL_CS || cs == __KERNEXEC_KERNEL_CS) && ip <= (unsigned long)_etext)
++			regs->ip = ip;
++	}
++#endif
++
+ 	nmi_enter();
+ 
+ 	inc_irq_stat(__nmi_count);
+diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c
+index 676b8c7..870ba04 100644
+--- a/arch/x86/kernel/paravirt-spinlocks.c
++++ b/arch/x86/kernel/paravirt-spinlocks.c
+@@ -13,7 +13,7 @@ default_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags)
+ 	arch_spin_lock(lock);
+ }
+ 
+-struct pv_lock_ops pv_lock_ops = {
++struct pv_lock_ops pv_lock_ops __read_only = {
+ #ifdef CONFIG_SMP
+ 	.spin_is_locked = __ticket_spin_is_locked,
+ 	.spin_is_contended = __ticket_spin_is_contended,
+diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
+index d90272e..6bb013b 100644
+--- a/arch/x86/kernel/paravirt.c
++++ b/arch/x86/kernel/paravirt.c
+@@ -53,6 +53,9 @@ u64 _paravirt_ident_64(u64 x)
+ {
+ 	return x;
+ }
++#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE)
++PV_CALLEE_SAVE_REGS_THUNK(_paravirt_ident_64);
++#endif
+ 
+ void __init default_banner(void)
+ {
+@@ -145,15 +148,19 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
+ 	if (opfunc == NULL)
+ 		/* If there's no function, patch it with a ud2a (BUG) */
+ 		ret = paravirt_patch_insns(insnbuf, len, ud2a, ud2a+sizeof(ud2a));
+-	else if (opfunc == _paravirt_nop)
++	else if (opfunc == (void *)_paravirt_nop)
+ 		/* If the operation is a nop, then nop the callsite */
+ 		ret = paravirt_patch_nop();
+ 
+ 	/* identity functions just return their single argument */
+-	else if (opfunc == _paravirt_ident_32)
++	else if (opfunc == (void *)_paravirt_ident_32)
+ 		ret = paravirt_patch_ident_32(insnbuf, len);
+-	else if (opfunc == _paravirt_ident_64)
++	else if (opfunc == (void *)_paravirt_ident_64)
+ 		ret = paravirt_patch_ident_64(insnbuf, len);
++#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE)
++	else if (opfunc == (void *)__raw_callee_save__paravirt_ident_64)
++		ret = paravirt_patch_ident_64(insnbuf, len);
++#endif
+ 
+ 	else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) ||
+ 		 type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) ||
+@@ -178,7 +185,7 @@ unsigned paravirt_patch_insns(void *insnbuf, unsigned len,
+ 	if (insn_len > len || start == NULL)
+ 		insn_len = len;
+ 	else
+-		memcpy(insnbuf, start, insn_len);
++		memcpy(insnbuf, ktla_ktva(start), insn_len);
+ 
+ 	return insn_len;
+ }
+@@ -302,7 +309,7 @@ void arch_flush_lazy_mmu_mode(void)
+ 	preempt_enable();
+ }
+ 
+-struct pv_info pv_info = {
++struct pv_info pv_info __read_only = {
+ 	.name = "bare hardware",
+ 	.paravirt_enabled = 0,
+ 	.kernel_rpl = 0,
+@@ -313,16 +320,16 @@ struct pv_info pv_info = {
+ #endif
+ };
+ 
+-struct pv_init_ops pv_init_ops = {
++struct pv_init_ops pv_init_ops __read_only = {
+ 	.patch = native_patch,
+ };
+ 
+-struct pv_time_ops pv_time_ops = {
++struct pv_time_ops pv_time_ops __read_only = {
+ 	.sched_clock = native_sched_clock,
+ 	.steal_clock = native_steal_clock,
+ };
+ 
+-struct pv_irq_ops pv_irq_ops = {
++struct pv_irq_ops pv_irq_ops __read_only = {
+ 	.save_fl = __PV_IS_CALLEE_SAVE(native_save_fl),
+ 	.restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl),
+ 	.irq_disable = __PV_IS_CALLEE_SAVE(native_irq_disable),
+@@ -334,7 +341,7 @@ struct pv_irq_ops pv_irq_ops = {
+ #endif
+ };
+ 
+-struct pv_cpu_ops pv_cpu_ops = {
++struct pv_cpu_ops pv_cpu_ops __read_only = {
+ 	.cpuid = native_cpuid,
+ 	.get_debugreg = native_get_debugreg,
+ 	.set_debugreg = native_set_debugreg,
+@@ -395,21 +402,26 @@ struct pv_cpu_ops pv_cpu_ops = {
+ 	.end_context_switch = paravirt_nop,
+ };
+ 
+-struct pv_apic_ops pv_apic_ops = {
++struct pv_apic_ops pv_apic_ops __read_only = {
+ #ifdef CONFIG_X86_LOCAL_APIC
+ 	.startup_ipi_hook = paravirt_nop,
+ #endif
+ };
+ 
+-#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_PAE)
++#ifdef CONFIG_X86_32
++#ifdef CONFIG_X86_PAE
++/* 64-bit pagetable entries */
++#define PTE_IDENT	PV_CALLEE_SAVE(_paravirt_ident_64)
++#else
+ /* 32-bit pagetable entries */
+ #define PTE_IDENT	__PV_IS_CALLEE_SAVE(_paravirt_ident_32)
++#endif
+ #else
+ /* 64-bit pagetable entries */
+ #define PTE_IDENT	__PV_IS_CALLEE_SAVE(_paravirt_ident_64)
+ #endif
+ 
+-struct pv_mmu_ops pv_mmu_ops = {
++struct pv_mmu_ops pv_mmu_ops __read_only = {
+ 
+ 	.read_cr2 = native_read_cr2,
+ 	.write_cr2 = native_write_cr2,
+@@ -459,6 +471,7 @@ struct pv_mmu_ops pv_mmu_ops = {
+ 	.make_pud = PTE_IDENT,
+ 
+ 	.set_pgd = native_set_pgd,
++	.set_pgd_batched = native_set_pgd_batched,
+ #endif
+ #endif /* PAGETABLE_LEVELS >= 3 */
+ 
+@@ -478,6 +491,12 @@ struct pv_mmu_ops pv_mmu_ops = {
+ 	},
+ 
+ 	.set_fixmap = native_set_fixmap,
++
++#ifdef CONFIG_PAX_KERNEXEC
++	.pax_open_kernel = native_pax_open_kernel,
++	.pax_close_kernel = native_pax_close_kernel,
++#endif
++
+ };
+ 
+ EXPORT_SYMBOL_GPL(pv_time_ops);
+diff --git a/arch/x86/kernel/pci-iommu_table.c b/arch/x86/kernel/pci-iommu_table.c
+index 35ccf75..7a15747 100644
+--- a/arch/x86/kernel/pci-iommu_table.c
++++ b/arch/x86/kernel/pci-iommu_table.c
+@@ -2,7 +2,7 @@
+ #include <asm/iommu_table.h>
+ #include <linux/string.h>
+ #include <linux/kallsyms.h>
+-
++#include <linux/sched.h>
+ 
+ #define DEBUG 1
+ 
+diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
+index ee5d4fb..426649b 100644
+--- a/arch/x86/kernel/process.c
++++ b/arch/x86/kernel/process.c
+@@ -48,16 +48,33 @@ void free_thread_xstate(struct task_struct *tsk)
+ 
+ void free_thread_info(struct thread_info *ti)
+ {
+-	free_thread_xstate(ti->task);
+ 	free_pages((unsigned long)ti, THREAD_ORDER);
+ }
+ 
++static struct kmem_cache *task_struct_cachep;
++
+ void arch_task_cache_init(void)
+ {
+-        task_xstate_cachep =
+-        	kmem_cache_create("task_xstate", xstate_size,
++	/* create a slab on which task_structs can be allocated */
++	task_struct_cachep =
++		kmem_cache_create("task_struct", sizeof(struct task_struct),
++			ARCH_MIN_TASKALIGN, SLAB_PANIC | SLAB_NOTRACK, NULL);
++
++	task_xstate_cachep =
++		kmem_cache_create("task_xstate", xstate_size,
+ 				  __alignof__(union thread_xstate),
+-				  SLAB_PANIC | SLAB_NOTRACK, NULL);
++				  SLAB_PANIC | SLAB_NOTRACK | SLAB_USERCOPY, NULL);
++}
++
++struct task_struct *alloc_task_struct_node(int node)
++{
++	return kmem_cache_alloc_node(task_struct_cachep, GFP_KERNEL, node);
++}
++
++void free_task_struct(struct task_struct *task)
++{
++	free_thread_xstate(task);
++	kmem_cache_free(task_struct_cachep, task);
+ }
+ 
+ /*
+@@ -70,7 +87,7 @@ void exit_thread(void)
+ 	unsigned long *bp = t->io_bitmap_ptr;
+ 
+ 	if (bp) {
+-		struct tss_struct *tss = &per_cpu(init_tss, get_cpu());
++		struct tss_struct *tss = init_tss + get_cpu();
+ 
+ 		t->io_bitmap_ptr = NULL;
+ 		clear_thread_flag(TIF_IO_BITMAP);
+@@ -106,7 +123,7 @@ void show_regs_common(void)
+ 
+ 	printk(KERN_CONT "\n");
+ 	printk(KERN_DEFAULT "Pid: %d, comm: %.20s %s %s %.*s",
+-		current->pid, current->comm, print_tainted(),
++		task_pid_nr(current), current->comm, print_tainted(),
+ 		init_utsname()->release,
+ 		(int)strcspn(init_utsname()->version, " "),
+ 		init_utsname()->version);
+@@ -120,6 +137,9 @@ void flush_thread(void)
+ {
+ 	struct task_struct *tsk = current;
+ 
++#if defined(CONFIG_X86_32) && !defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_PAX_MEMORY_UDEREF)
++	loadsegment(gs, 0);
++#endif
+ 	flush_ptrace_hw_breakpoint(tsk);
+ 	memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
+ 	/*
+@@ -282,10 +302,10 @@ int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
+ 	regs.di = (unsigned long) arg;
+ 
+ #ifdef CONFIG_X86_32
+-	regs.ds = __USER_DS;
+-	regs.es = __USER_DS;
++	regs.ds = __KERNEL_DS;
++	regs.es = __KERNEL_DS;
+ 	regs.fs = __KERNEL_PERCPU;
+-	regs.gs = __KERNEL_STACK_CANARY;
++	savesegment(gs, regs.gs);
+ #else
+ 	regs.ss = __KERNEL_DS;
+ #endif
+@@ -411,7 +431,7 @@ bool set_pm_idle_to_default(void)
+ 
+ 	return ret;
+ }
+-void stop_this_cpu(void *dummy)
++__noreturn void stop_this_cpu(void *dummy)
+ {
+ 	local_irq_disable();
+ 	/*
+@@ -653,16 +673,37 @@ static int __init idle_setup(char *str)
+ }
+ early_param("idle", idle_setup);
+ 
+-unsigned long arch_align_stack(unsigned long sp)
++#ifdef CONFIG_PAX_RANDKSTACK
++void pax_randomize_kstack(struct pt_regs *regs)
+ {
+-	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+-		sp -= get_random_int() % 8192;
+-	return sp & ~0xf;
+-}
++	struct thread_struct *thread = &current->thread;
++	unsigned long time;
+ 
+-unsigned long arch_randomize_brk(struct mm_struct *mm)
+-{
+-	unsigned long range_end = mm->brk + 0x02000000;
+-	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
+-}
++	if (!randomize_va_space)
++		return;
++
++	if (v8086_mode(regs))
++		return;
+ 
++	rdtscl(time);
++
++	/* P4 seems to return a 0 LSB, ignore it */
++#ifdef CONFIG_MPENTIUM4
++	time &= 0x3EUL;
++	time <<= 2;
++#elif defined(CONFIG_X86_64)
++	time &= 0xFUL;
++	time <<= 4;
++#else
++	time &= 0x1FUL;
++	time <<= 3;
++#endif
++
++	thread->sp0 ^= time;
++	load_sp0(init_tss + smp_processor_id(), thread);
++
++#ifdef CONFIG_X86_64
++	percpu_write(kernel_stack, thread->sp0);
++#endif
++}
++#endif
+diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
+index 8598296..bfadef0 100644
+--- a/arch/x86/kernel/process_32.c
++++ b/arch/x86/kernel/process_32.c
+@@ -67,6 +67,7 @@ asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
+ unsigned long thread_saved_pc(struct task_struct *tsk)
+ {
+ 	return ((unsigned long *)tsk->thread.sp)[3];
++//XXX	return tsk->thread.eip;
+ }
+ 
+ #ifndef CONFIG_SMP
+@@ -130,15 +131,14 @@ void __show_regs(struct pt_regs *regs, int all)
+ 	unsigned long sp;
+ 	unsigned short ss, gs;
+ 
+-	if (user_mode_vm(regs)) {
++	if (user_mode(regs)) {
+ 		sp = regs->sp;
+ 		ss = regs->ss & 0xffff;
+-		gs = get_user_gs(regs);
+ 	} else {
+ 		sp = kernel_stack_pointer(regs);
+ 		savesegment(ss, ss);
+-		savesegment(gs, gs);
+ 	}
++	gs = get_user_gs(regs);
+ 
+ 	show_regs_common();
+ 
+@@ -200,13 +200,14 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
+ 	struct task_struct *tsk;
+ 	int err;
+ 
+-	childregs = task_pt_regs(p);
++	childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 8;
+ 	*childregs = *regs;
+ 	childregs->ax = 0;
+ 	childregs->sp = sp;
+ 
+ 	p->thread.sp = (unsigned long) childregs;
+ 	p->thread.sp0 = (unsigned long) (childregs+1);
++	p->tinfo.lowest_stack = (unsigned long)task_stack_page(p);
+ 
+ 	p->thread.ip = (unsigned long) ret_from_fork;
+ 
+@@ -296,7 +297,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ 	struct thread_struct *prev = &prev_p->thread,
+ 				 *next = &next_p->thread;
+ 	int cpu = smp_processor_id();
+-	struct tss_struct *tss = &per_cpu(init_tss, cpu);
++	struct tss_struct *tss = init_tss + cpu;
+ 	fpu_switch_t fpu;
+ 
+ 	/* never put a printk in __switch_to... printk() calls wake_up*() indirectly */
+@@ -320,6 +321,10 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ 	 */
+ 	lazy_save_gs(prev->gs);
+ 
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	__set_fs(task_thread_info(next_p)->addr_limit);
++#endif
++
+ 	/*
+ 	 * Load the per-thread Thread-Local Storage descriptor.
+ 	 */
+@@ -350,6 +355,9 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ 	 */
+ 	arch_end_context_switch(next_p);
+ 
++	percpu_write(current_task, next_p);
++	percpu_write(current_tinfo, &next_p->tinfo);
++
+ 	/*
+ 	 * Restore %gs if needed (which is common)
+ 	 */
+@@ -358,8 +366,6 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ 
+ 	switch_fpu_finish(next_p, fpu);
+ 
+-	percpu_write(current_task, next_p);
+-
+ 	return prev_p;
+ }
+ 
+@@ -389,4 +395,3 @@ unsigned long get_wchan(struct task_struct *p)
+ 	} while (count++ < 16);
+ 	return 0;
+ }
+-
+diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
+index 6a364a6..b147d11 100644
+--- a/arch/x86/kernel/process_64.c
++++ b/arch/x86/kernel/process_64.c
+@@ -89,7 +89,7 @@ static void __exit_idle(void)
+ void exit_idle(void)
+ {
+ 	/* idle loop has pid 0 */
+-	if (current->pid)
++	if (task_pid_nr(current))
+ 		return;
+ 	__exit_idle();
+ }
+@@ -264,8 +264,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
+ 	struct pt_regs *childregs;
+ 	struct task_struct *me = current;
+ 
+-	childregs = ((struct pt_regs *)
+-			(THREAD_SIZE + task_stack_page(p))) - 1;
++	childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 16;
+ 	*childregs = *regs;
+ 
+ 	childregs->ax = 0;
+@@ -277,6 +276,7 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
+ 	p->thread.sp = (unsigned long) childregs;
+ 	p->thread.sp0 = (unsigned long) (childregs+1);
+ 	p->thread.usersp = me->thread.usersp;
++	p->tinfo.lowest_stack = (unsigned long)task_stack_page(p);
+ 
+ 	set_tsk_thread_flag(p, TIF_FORK);
+ 
+@@ -379,7 +379,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ 	struct thread_struct *prev = &prev_p->thread;
+ 	struct thread_struct *next = &next_p->thread;
+ 	int cpu = smp_processor_id();
+-	struct tss_struct *tss = &per_cpu(init_tss, cpu);
++	struct tss_struct *tss = init_tss + cpu;
+ 	unsigned fsindex, gsindex;
+ 	fpu_switch_t fpu;
+ 
+@@ -461,10 +461,9 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+ 	prev->usersp = percpu_read(old_rsp);
+ 	percpu_write(old_rsp, next->usersp);
+ 	percpu_write(current_task, next_p);
++	percpu_write(current_tinfo, &next_p->tinfo);
+ 
+-	percpu_write(kernel_stack,
+-		  (unsigned long)task_stack_page(next_p) +
+-		  THREAD_SIZE - KERNEL_STACK_OFFSET);
++	percpu_write(kernel_stack, next->sp0);
+ 
+ 	/*
+ 	 * Now maybe reload the debug registers and handle I/O bitmaps
+@@ -519,12 +518,11 @@ unsigned long get_wchan(struct task_struct *p)
+ 	if (!p || p == current || p->state == TASK_RUNNING)
+ 		return 0;
+ 	stack = (unsigned long)task_stack_page(p);
+-	if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE)
++	if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE-16-sizeof(u64))
+ 		return 0;
+ 	fp = *(u64 *)(p->thread.sp);
+ 	do {
+-		if (fp < (unsigned long)stack ||
+-		    fp >= (unsigned long)stack+THREAD_SIZE)
++		if (fp < stack || fp > stack+THREAD_SIZE-16-sizeof(u64))
+ 			return 0;
+ 		ip = *(u64 *)(fp+8);
+ 		if (!in_sched_functions(ip))
+diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
+index 8252879..f367ec9 100644
+--- a/arch/x86/kernel/ptrace.c
++++ b/arch/x86/kernel/ptrace.c
+@@ -791,6 +791,10 @@ static int ioperm_active(struct task_struct *target,
+ static int ioperm_get(struct task_struct *target,
+ 		      const struct user_regset *regset,
+ 		      unsigned int pos, unsigned int count,
++		      void *kbuf, void __user *ubuf) __size_overflow(3,4);
++static int ioperm_get(struct task_struct *target,
++		      const struct user_regset *regset,
++		      unsigned int pos, unsigned int count,
+ 		      void *kbuf, void __user *ubuf)
+ {
+ 	if (!target->thread.io_bitmap_ptr)
+@@ -822,7 +826,7 @@ long arch_ptrace(struct task_struct *child, long request,
+ 		 unsigned long addr, unsigned long data)
+ {
+ 	int ret;
+-	unsigned long __user *datap = (unsigned long __user *)data;
++	unsigned long __user *datap = (__force unsigned long __user *)data;
+ 
+ 	switch (request) {
+ 	/* read the word at location addr in the USER area. */
+@@ -907,14 +911,14 @@ long arch_ptrace(struct task_struct *child, long request,
+ 		if ((int) addr < 0)
+ 			return -EIO;
+ 		ret = do_get_thread_area(child, addr,
+-					(struct user_desc __user *)data);
++					(__force struct user_desc __user *) data);
+ 		break;
+ 
+ 	case PTRACE_SET_THREAD_AREA:
+ 		if ((int) addr < 0)
+ 			return -EIO;
+ 		ret = do_set_thread_area(child, addr,
+-					(struct user_desc __user *)data, 0);
++					(__force struct user_desc __user *) data, 0);
+ 		break;
+ #endif
+ 
+@@ -1331,7 +1335,7 @@ static void fill_sigtrap_info(struct task_struct *tsk,
+ 	memset(info, 0, sizeof(*info));
+ 	info->si_signo = SIGTRAP;
+ 	info->si_code = si_code;
+-	info->si_addr = user_mode_vm(regs) ? (void __user *)regs->ip : NULL;
++	info->si_addr = user_mode(regs) ? (__force void __user *)regs->ip : NULL;
+ }
+ 
+ void user_single_step_siginfo(struct task_struct *tsk,
+diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c
+index 42eb330..139955c 100644
+--- a/arch/x86/kernel/pvclock.c
++++ b/arch/x86/kernel/pvclock.c
+@@ -81,11 +81,11 @@ unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src)
+ 	return pv_tsc_khz;
+ }
+ 
+-static atomic64_t last_value = ATOMIC64_INIT(0);
++static atomic64_unchecked_t last_value = ATOMIC64_INIT(0);
+ 
+ void pvclock_resume(void)
+ {
+-	atomic64_set(&last_value, 0);
++	atomic64_set_unchecked(&last_value, 0);
+ }
+ 
+ cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
+@@ -121,11 +121,11 @@ cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
+ 	 * updating at the same time, and one of them could be slightly behind,
+ 	 * making the assumption that last_value always go forward fail to hold.
+ 	 */
+-	last = atomic64_read(&last_value);
++	last = atomic64_read_unchecked(&last_value);
+ 	do {
+ 		if (ret < last)
+ 			return last;
+-		last = atomic64_cmpxchg(&last_value, last, ret);
++		last = atomic64_cmpxchg_unchecked(&last_value, last, ret);
+ 	} while (unlikely(last != ret));
+ 
+ 	return ret;
+diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
+index 37a458b..e63d183 100644
+--- a/arch/x86/kernel/reboot.c
++++ b/arch/x86/kernel/reboot.c
+@@ -35,7 +35,7 @@ void (*pm_power_off)(void);
+ EXPORT_SYMBOL(pm_power_off);
+ 
+ static const struct desc_ptr no_idt = {};
+-static int reboot_mode;
++static unsigned short reboot_mode;
+ enum reboot_type reboot_type = BOOT_ACPI;
+ int reboot_force;
+ 
+@@ -324,13 +324,17 @@ core_initcall(reboot_init);
+ extern const unsigned char machine_real_restart_asm[];
+ extern const u64 machine_real_restart_gdt[3];
+ 
+-void machine_real_restart(unsigned int type)
++__noreturn void machine_real_restart(unsigned int type)
+ {
+ 	void *restart_va;
+ 	unsigned long restart_pa;
+-	void (*restart_lowmem)(unsigned int);
++	void (* __noreturn restart_lowmem)(unsigned int);
+ 	u64 *lowmem_gdt;
+ 
++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF))
++	struct desc_struct *gdt;
++#endif
++
+ 	local_irq_disable();
+ 
+ 	/* Write zero to CMOS register number 0x0f, which the BIOS POST
+@@ -356,14 +360,14 @@ void machine_real_restart(unsigned int type)
+ 	   boot)".  This seems like a fairly standard thing that gets set by
+ 	   REBOOT.COM programs, and the previous reset routine did this
+ 	   too. */
+-	*((unsigned short *)0x472) = reboot_mode;
++	*(unsigned short *)(__va(0x472)) = reboot_mode;
+ 
+ 	/* Patch the GDT in the low memory trampoline */
+ 	lowmem_gdt = TRAMPOLINE_SYM(machine_real_restart_gdt);
+ 
+ 	restart_va = TRAMPOLINE_SYM(machine_real_restart_asm);
+ 	restart_pa = virt_to_phys(restart_va);
+-	restart_lowmem = (void (*)(unsigned int))restart_pa;
++	restart_lowmem = (void *)restart_pa;
+ 
+ 	/* GDT[0]: GDT self-pointer */
+ 	lowmem_gdt[0] =
+@@ -374,7 +378,33 @@ void machine_real_restart(unsigned int type)
+ 		GDT_ENTRY(0x009b, restart_pa, 0xffff);
+ 
+ 	/* Jump to the identity-mapped low memory code */
++
++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF))
++	gdt = get_cpu_gdt_table(smp_processor_id());
++	pax_open_kernel();
++#ifdef CONFIG_PAX_MEMORY_UDEREF
++	gdt[GDT_ENTRY_KERNEL_DS].type = 3;
++	gdt[GDT_ENTRY_KERNEL_DS].limit = 0xf;
++	asm("mov %0, %%ds; mov %0, %%es; mov %0, %%ss" : : "r" (__KERNEL_DS) : "memory");
++#endif
++#ifdef CONFIG_PAX_KERNEXEC
++	gdt[GDT_ENTRY_KERNEL_CS].base0 = 0;
++	gdt[GDT_ENTRY_KERNEL_CS].base1 = 0;
++	gdt[GDT_ENTRY_KERNEL_CS].base2 = 0;
++	gdt[GDT_ENTRY_KERNEL_CS].limit0 = 0xffff;
++	gdt[GDT_ENTRY_KERNEL_CS].limit = 0xf;
++	gdt[GDT_ENTRY_KERNEL_CS].g = 1;
++#endif
++	pax_close_kernel();
++#endif
++
++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC)
++	asm volatile("push %0; push %1; lret\n" : : "i" (__KERNEL_CS), "rm" (restart_lowmem), "a" (type));
++	unreachable();
++#else
+ 	restart_lowmem(type);
++#endif
++
+ }
+ #ifdef CONFIG_APM_MODULE
+ EXPORT_SYMBOL(machine_real_restart);
+@@ -540,7 +570,7 @@ void __attribute__((weak)) mach_reboot_fixups(void)
+  * try to force a triple fault and then cycle between hitting the keyboard
+  * controller and doing that
+  */
+-static void native_machine_emergency_restart(void)
++__noreturn static void native_machine_emergency_restart(void)
+ {
+ 	int i;
+ 	int attempt = 0;
+@@ -664,13 +694,13 @@ void native_machine_shutdown(void)
+ #endif
+ }
+ 
+-static void __machine_emergency_restart(int emergency)
++static __noreturn void __machine_emergency_restart(int emergency)
+ {
+ 	reboot_emergency = emergency;
+ 	machine_ops.emergency_restart();
+ }
+ 
+-static void native_machine_restart(char *__unused)
++static __noreturn void native_machine_restart(char *__unused)
+ {
+ 	printk("machine restart\n");
+ 
+@@ -679,7 +709,7 @@ static void native_machine_restart(char *__unused)
+ 	__machine_emergency_restart(0);
+ }
+ 
+-static void native_machine_halt(void)
++static __noreturn void native_machine_halt(void)
+ {
+ 	/* stop other cpus and apics */
+ 	machine_shutdown();
+@@ -690,7 +720,7 @@ static void native_machine_halt(void)
+ 	stop_this_cpu(NULL);
+ }
+ 
+-static void native_machine_power_off(void)
++__noreturn static void native_machine_power_off(void)
+ {
+ 	if (pm_power_off) {
+ 		if (!reboot_force)
+@@ -699,6 +729,7 @@ static void native_machine_power_off(void)
+ 	}
+ 	/* a fallback in case there is no PM info available */
+ 	tboot_shutdown(TB_SHUTDOWN_HALT);
++	unreachable();
+ }
+ 
+ struct machine_ops machine_ops = {
+diff --git a/arch/x86/kernel/relocate_kernel_64.S b/arch/x86/kernel/relocate_kernel_64.S
+index 7a6f3b3..bed145d7 100644
+--- a/arch/x86/kernel/relocate_kernel_64.S
++++ b/arch/x86/kernel/relocate_kernel_64.S
+@@ -11,6 +11,7 @@
+ #include <asm/kexec.h>
+ #include <asm/processor-flags.h>
+ #include <asm/pgtable_types.h>
++#include <asm/alternative-asm.h>
+ 
+ /*
+  * Must be relocatable PIC code callable as a C function
+@@ -160,13 +161,14 @@ identity_mapped:
+ 	xorq    %rbp, %rbp
+ 	xorq	%r8,  %r8
+ 	xorq	%r9,  %r9
+-	xorq	%r10, %r9
++	xorq	%r10, %r10
+ 	xorq	%r11, %r11
+ 	xorq	%r12, %r12
+ 	xorq	%r13, %r13
+ 	xorq	%r14, %r14
+ 	xorq	%r15, %r15
+ 
++	pax_force_retaddr 0, 1
+ 	ret
+ 
+ 1:
+diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
+index cf0ef98..e3f780b 100644
+--- a/arch/x86/kernel/setup.c
++++ b/arch/x86/kernel/setup.c
+@@ -447,7 +447,7 @@ static void __init parse_setup_data(void)
+ 
+ 		switch (data->type) {
+ 		case SETUP_E820_EXT:
+-			parse_e820_ext(data);
++			parse_e820_ext((struct setup_data __force_kernel *)data);
+ 			break;
+ 		case SETUP_DTB:
+ 			add_dtb(pa_data);
+@@ -650,7 +650,7 @@ static void __init trim_bios_range(void)
+ 	 * area (640->1Mb) as ram even though it is not.
+ 	 * take them out.
+ 	 */
+-	e820_remove_range(BIOS_BEGIN, BIOS_END - BIOS_BEGIN, E820_RAM, 1);
++	e820_remove_range(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS, E820_RAM, 1);
+ 	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+ }
+ 
+@@ -773,14 +773,14 @@ void __init setup_arch(char **cmdline_p)
+ 
+ 	if (!boot_params.hdr.root_flags)
+ 		root_mountflags &= ~MS_RDONLY;
+-	init_mm.start_code = (unsigned long) _text;
+-	init_mm.end_code = (unsigned long) _etext;
++	init_mm.start_code = ktla_ktva((unsigned long) _text);
++	init_mm.end_code = ktla_ktva((unsigned long) _etext);
+ 	init_mm.end_data = (unsigned long) _edata;
+ 	init_mm.brk = _brk_end;
+ 
+-	code_resource.start = virt_to_phys(_text);
+-	code_resource.end = virt_to_phys(_etext)-1;
+-	data_resource.start = virt_to_phys(_etext);
++	code_resource.start = virt_to_phys(ktla_ktva(_text));
++	code_resource.end = virt_to_phys(ktla_ktva(_etext))-1;
++	data_resource.start = virt_to_phys(_sdata);
+ 	data_resource.end = virt_to_phys(_edata)-1;
+ 	bss_resource.start = virt_to_phys(&__bss_start);
+ 	bss_resource.end = virt_to_phys(&__bss_stop)-1;
+diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
+index 71f4727..217419b 100644
+--- a/arch/x86/kernel/setup_percpu.c
++++ b/arch/x86/kernel/setup_percpu.c
+@@ -21,19 +21,17 @@
+ #include <asm/cpu.h>
+ #include <asm/stackprotector.h>
+ 
+-DEFINE_PER_CPU(int, cpu_number);
++#ifdef CONFIG_SMP
++DEFINE_PER_CPU(unsigned int, cpu_number);
+ EXPORT_PER_CPU_SYMBOL(cpu_number);
++#endif
+ 
+-#ifdef CONFIG_X86_64
+ #define BOOT_PERCPU_OFFSET ((unsigned long)__per_cpu_load)
+-#else
+-#define BOOT_PERCPU_OFFSET 0
+-#endif
+ 
+ DEFINE_PER_CPU(unsigned long, this_cpu_off) = BOOT_PERCPU_OFFSET;
+ EXPORT_PER_CPU_SYMBOL(this_cpu_off);
+ 
+-unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = {
++unsigned long __per_cpu_offset[NR_CPUS] __read_only = {
+ 	[0 ... NR_CPUS-1] = BOOT_PERCPU_OFFSET,
+ };
+ EXPORT_SYMBOL(__per_cpu_offset);
+@@ -96,6 +94,8 @@ static bool __init pcpu_need_numa(void)
+  * 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) __size_overflow(2);
++static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size,
+ 					unsigned long align)
+ {
+ 	const unsigned long goal = __pa(MAX_DMA_ADDRESS);
+@@ -124,6 +124,8 @@ static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size,
+ /*
+  * Helpers for first chunk memory allocation
+  */
++static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, size_t align) __size_overflow(2);
++
+ static void * __init pcpu_fc_alloc(unsigned int cpu, size_t size, size_t align)
+ {
+ 	return pcpu_alloc_bootmem(cpu, size, align);
+@@ -155,10 +157,10 @@ static inline void setup_percpu_segment(int cpu)
+ {
+ #ifdef CONFIG_X86_32
+ 	struct desc_struct gdt;
++	unsigned long base = per_cpu_offset(cpu);
+ 
+-	pack_descriptor(&gdt, per_cpu_offset(cpu), 0xFFFFF,
+-			0x2 | DESCTYPE_S, 0x8);
+-	gdt.s = 1;
++	pack_descriptor(&gdt, base, (VMALLOC_END - base - 1) >> PAGE_SHIFT,
++			0x83 | DESCTYPE_S, 0xC);
+ 	write_gdt_entry(get_cpu_gdt_table(cpu),
+ 			GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S);
+ #endif
+@@ -207,6 +209,11 @@ void __init setup_per_cpu_areas(void)
+ 	/* alrighty, percpu areas up and running */
+ 	delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
+ 	for_each_possible_cpu(cpu) {
++#ifdef CONFIG_CC_STACKPROTECTOR
++#ifdef CONFIG_X86_32
++		unsigned long canary = per_cpu(stack_canary.canary, cpu);
++#endif
++#endif
+ 		per_cpu_offset(cpu) = delta + pcpu_unit_offsets[cpu];
+ 		per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu);
+ 		per_cpu(cpu_number, cpu) = cpu;
+@@ -247,6 +254,12 @@ void __init setup_per_cpu_areas(void)
+ 		 */
+ 		set_cpu_numa_node(cpu, early_cpu_to_node(cpu));
+ #endif
++#ifdef CONFIG_CC_STACKPROTECTOR
++#ifdef CONFIG_X86_32
++		if (!cpu)
++			per_cpu(stack_canary.canary, cpu) = canary;
++#endif
++#endif
+ 		/*
+ 		 * Up to this point, the boot CPU has been using .init.data
+ 		 * area.  Reload any changed state for the boot CPU.
+diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
+index 54ddaeb2..22c3bdc 100644
+--- a/arch/x86/kernel/signal.c
++++ b/arch/x86/kernel/signal.c
+@@ -198,7 +198,7 @@ static unsigned long align_sigframe(unsigned long sp)
+ 	 * 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;
++	sp = ((sp - 12) & -16ul) - 4;
+ #else /* !CONFIG_X86_32 */
+ 	sp = round_down(sp, 16) - 8;
+ #endif
+@@ -249,11 +249,11 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
+ 	 * Return an always-bogus address instead so we will die with SIGSEGV.
+ 	 */
+ 	if (onsigstack && !likely(on_sig_stack(sp)))
+-		return (void __user *)-1L;
++		return (__force void __user *)-1L;
+ 
+ 	/* save i387 state */
+ 	if (used_math() && save_i387_xstate(*fpstate) < 0)
+-		return (void __user *)-1L;
++		return (__force void __user *)-1L;
+ 
+ 	return (void __user *)sp;
+ }
+@@ -308,9 +308,9 @@ __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
+ 	}
+ 
+ 	if (current->mm->context.vdso)
+-		restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn);
++		restorer = (__force void __user *)VDSO32_SYMBOL(current->mm->context.vdso, sigreturn);
+ 	else
+-		restorer = &frame->retcode;
++		restorer = (void __user *)&frame->retcode;
+ 	if (ka->sa.sa_flags & SA_RESTORER)
+ 		restorer = ka->sa.sa_restorer;
+ 
+@@ -324,7 +324,7 @@ __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
+ 	 * reasons and because gdb uses it as a signature to notice
+ 	 * signal handler stack frames.
+ 	 */
+-	err |= __put_user(*((u64 *)&retcode), (u64 *)frame->retcode);
++	err |= __put_user(*((u64 *)&retcode), (u64 __user *)frame->retcode);
+ 
+ 	if (err)
+ 		return -EFAULT;
+@@ -378,7 +378,10 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+ 		err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+ 
+ 		/* Set up to return from userspace.  */
+-		restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
++		if (current->mm->context.vdso)
++			restorer = (__force void __user *)VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
++		else
++			restorer = (void __user *)&frame->retcode;
+ 		if (ka->sa.sa_flags & SA_RESTORER)
+ 			restorer = ka->sa.sa_restorer;
+ 		put_user_ex(restorer, &frame->pretcode);
+@@ -390,7 +393,7 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+ 		 * reasons and because gdb uses it as a signature to notice
+ 		 * signal handler stack frames.
+ 		 */
+-		put_user_ex(*((u64 *)&rt_retcode), (u64 *)frame->retcode);
++		put_user_ex(*((u64 *)&rt_retcode), (u64 __user *)frame->retcode);
+ 	} put_user_catch(err);
+ 
+ 	if (err)
+@@ -769,7 +772,7 @@ static void do_signal(struct pt_regs *regs)
+ 	 * X86_32: vm86 regs switched out by assembly code before reaching
+ 	 * here, so testing against kernel CS suffices.
+ 	 */
+-	if (!user_mode(regs))
++	if (!user_mode_novm(regs))
+ 		return;
+ 
+ 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
+index 9f548cb..caf76f7 100644
+--- a/arch/x86/kernel/smpboot.c
++++ b/arch/x86/kernel/smpboot.c
+@@ -709,17 +709,20 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
+ 	set_idle_for_cpu(cpu, c_idle.idle);
+ do_rest:
+ 	per_cpu(current_task, cpu) = c_idle.idle;
++	per_cpu(current_tinfo, cpu) = &c_idle.idle->tinfo;
+ #ifdef CONFIG_X86_32
+ 	/* Stack for startup_32 can be just as for start_secondary onwards */
+ 	irq_ctx_init(cpu);
+ #else
+ 	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
+ 	initial_gs = per_cpu_offset(cpu);
+-	per_cpu(kernel_stack, cpu) =
+-		(unsigned long)task_stack_page(c_idle.idle) -
+-		KERNEL_STACK_OFFSET + THREAD_SIZE;
++	per_cpu(kernel_stack, cpu) = (unsigned long)task_stack_page(c_idle.idle) - 16 + THREAD_SIZE;
+ #endif
++
++	pax_open_kernel();
+ 	early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
++	pax_close_kernel();
++
+ 	initial_code = (unsigned long)start_secondary;
+ 	stack_start  = c_idle.idle->thread.sp;
+ 
+@@ -861,6 +864,12 @@ int __cpuinit native_cpu_up(unsigned int cpu)
+ 
+ 	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
+ 
++#ifdef CONFIG_PAX_PER_CPU_PGD
++	clone_pgd_range(get_cpu_pgd(cpu) + KERNEL_PGD_BOUNDARY,
++			swapper_pg_dir + KERNEL_PGD_BOUNDARY,
++			KERNEL_PGD_PTRS);
++#endif
++
+ 	err = do_boot_cpu(apicid, cpu);
+ 	if (err) {
+ 		pr_debug("do_boot_cpu failed %d\n", err);
+diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
+index c346d11..d43b163 100644
+--- a/arch/x86/kernel/step.c
++++ b/arch/x86/kernel/step.c
+@@ -27,10 +27,10 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re
+ 		struct desc_struct *desc;
+ 		unsigned long base;
+ 
+-		seg &= ~7UL;
++		seg >>= 3;
+ 
+ 		mutex_lock(&child->mm->context.lock);
+-		if (unlikely((seg >> 3) >= child->mm->context.size))
++		if (unlikely(seg >= child->mm->context.size))
+ 			addr = -1L; /* bogus selector, access would fault */
+ 		else {
+ 			desc = child->mm->context.ldt + seg;
+@@ -42,7 +42,8 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re
+ 			addr += base;
+ 		}
+ 		mutex_unlock(&child->mm->context.lock);
+-	}
++	} else if (seg == __KERNEL_CS || seg == __KERNEXEC_KERNEL_CS)
++		addr = ktla_ktva(addr);
+ 
+ 	return addr;
+ }
+@@ -53,6 +54,9 @@ static int is_setting_trap_flag(struct task_struct *child, struct pt_regs *regs)
+ 	unsigned char opcode[15];
+ 	unsigned long addr = convert_ip_to_linear(child, regs);
+ 
++	if (addr == -EINVAL)
++		return 0;
++
+ 	copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0);
+ 	for (i = 0; i < copied; i++) {
+ 		switch (opcode[i]) {
+diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c
+index 0b0cb5f..db6b9ed 100644
+--- a/arch/x86/kernel/sys_i386_32.c
++++ b/arch/x86/kernel/sys_i386_32.c
+@@ -24,17 +24,224 @@
+ 
+ #include <asm/syscalls.h>
+ 
+-/*
+- * Do a system call from kernel instead of calling sys_execve so we
+- * end up with proper pt_regs.
+- */
+-int kernel_execve(const char *filename,
+-		  const char *const argv[],
+-		  const char *const envp[])
++int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags)
+ {
+-	long __res;
+-	asm volatile ("int $0x80"
+-	: "=a" (__res)
+-	: "0" (__NR_execve), "b" (filename), "c" (argv), "d" (envp) : "memory");
+-	return __res;
++	unsigned long pax_task_size = TASK_SIZE;
++
++#ifdef CONFIG_PAX_SEGMEXEC
++	if (current->mm->pax_flags & MF_PAX_SEGMEXEC)
++		pax_task_size = SEGMEXEC_TASK_SIZE;
++#endif
++
++	if (len > pax_task_size || addr > pax_task_size - len)
++		return -EINVAL;
++
++	return 0;
++}
++
++unsigned long
++arch_get_unmapped_area(struct file *filp, unsigned long addr,
++		unsigned long len, unsigned long pgoff, unsigned long flags)
++{
++	struct mm_struct *mm = current->mm;
++	struct vm_area_struct *vma;
++	unsigned long start_addr, pax_task_size = TASK_SIZE;
++
++#ifdef CONFIG_PAX_SEGMEXEC
++	if (mm->pax_flags & MF_PAX_SEGMEXEC)
++		pax_task_size = SEGMEXEC_TASK_SIZE;
++#endif
++
++	pax_task_size -= PAGE_SIZE;
++
++	if (len > pax_task_size)
++		return -ENOMEM;
++
++	if (flags & MAP_FIXED)
++		return addr;
++
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
++	if (addr) {
++		addr = PAGE_ALIGN(addr);
++		if (pax_task_size - len >= addr) {
++			vma = find_vma(mm, addr);
++			if (check_heap_stack_gap(vma, addr, len))
++				return addr;
++		}
++	}
++	if (len > mm->cached_hole_size) {
++		start_addr = addr = mm->free_area_cache;
++	} else {
++		start_addr = addr = mm->mmap_base;
++		mm->cached_hole_size = 0;
++	}
++
++#ifdef CONFIG_PAX_PAGEEXEC
++	if (!(__supported_pte_mask & _PAGE_NX) && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE) && start_addr >= mm->mmap_base) {
++		start_addr = 0x00110000UL;
++
++#ifdef CONFIG_PAX_RANDMMAP
++		if (mm->pax_flags & MF_PAX_RANDMMAP)
++			start_addr += mm->delta_mmap & 0x03FFF000UL;
++#endif
++
++		if (mm->start_brk <= start_addr && start_addr < mm->mmap_base)
++			start_addr = addr = mm->mmap_base;
++		else
++			addr = start_addr;
++	}
++#endif
++
++full_search:
++	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
++		/* At this point:  (!vma || addr < vma->vm_end). */
++		if (pax_task_size - len < addr) {
++			/*
++			 * Start a new search - just in case we missed
++			 * some holes.
++			 */
++			if (start_addr != mm->mmap_base) {
++				start_addr = addr = mm->mmap_base;
++				mm->cached_hole_size = 0;
++				goto full_search;
++			}
++			return -ENOMEM;
++		}
++		if (check_heap_stack_gap(vma, addr, len))
++			break;
++		if (addr + mm->cached_hole_size < vma->vm_start)
++			mm->cached_hole_size = vma->vm_start - addr;
++		addr = vma->vm_end;
++		if (mm->start_brk <= addr && addr < mm->mmap_base) {
++			start_addr = addr = mm->mmap_base;
++			mm->cached_hole_size = 0;
++			goto full_search;
++		}
++	}
++
++	/*
++	 * Remember the place where we stopped the search:
++	 */
++	mm->free_area_cache = addr + len;
++	return addr;
++}
++
++unsigned long
++arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
++			  const unsigned long len, const unsigned long pgoff,
++			  const unsigned long flags)
++{
++	struct vm_area_struct *vma;
++	struct mm_struct *mm = current->mm;
++	unsigned long base = mm->mmap_base, addr = addr0, pax_task_size = TASK_SIZE;
++
++#ifdef CONFIG_PAX_SEGMEXEC
++	if (mm->pax_flags & MF_PAX_SEGMEXEC)
++		pax_task_size = SEGMEXEC_TASK_SIZE;
++#endif
++
++	pax_task_size -= PAGE_SIZE;
++
++	/* requested length too big for entire address space */
++	if (len > pax_task_size)
++		return -ENOMEM;
++
++	if (flags & MAP_FIXED)
++		return addr;
++
++#ifdef CONFIG_PAX_PAGEEXEC
++	if (!(__supported_pte_mask & _PAGE_NX) && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE))
++		goto bottomup;
++#endif
++
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
++	/* requesting a specific address */
++	if (addr) {
++		addr = PAGE_ALIGN(addr);
++		if (pax_task_size - len >= addr) {
++			vma = find_vma(mm, addr);
++			if (check_heap_stack_gap(vma, addr, len))
++				return addr;
++		}
++	}
++
++	/* check if free_area_cache is useful for us */
++	if (len <= mm->cached_hole_size) {
++		mm->cached_hole_size = 0;
++		mm->free_area_cache = mm->mmap_base;
++	}
++
++	/* either no address requested or can't fit in requested address hole */
++	addr = mm->free_area_cache;
++
++	/* make sure it can fit in the remaining address space */
++	if (addr > len) {
++		vma = find_vma(mm, addr-len);
++		if (check_heap_stack_gap(vma, addr - len, len))
++			/* remember the address as a hint for next time */
++			return (mm->free_area_cache = addr-len);
++	}
++
++	if (mm->mmap_base < len)
++		goto bottomup;
++
++	addr = mm->mmap_base-len;
++
++	do {
++		/*
++		 * Lookup failure means no vma is above this address,
++		 * else if new region fits below vma->vm_start,
++		 * return with success:
++		 */
++		vma = find_vma(mm, addr);
++		if (check_heap_stack_gap(vma, addr, len))
++			/* remember the address as a hint for next time */
++			return (mm->free_area_cache = addr);
++
++		/* remember the largest hole we saw so far */
++		if (addr + mm->cached_hole_size < vma->vm_start)
++			mm->cached_hole_size = vma->vm_start - addr;
++
++		/* try just below the current vma->vm_start */
++		addr = skip_heap_stack_gap(vma, len);
++	} while (!IS_ERR_VALUE(addr));
++
++bottomup:
++	/*
++	 * A failed mmap() very likely causes application failure,
++	 * so fall back to the bottom-up function here. This scenario
++	 * can happen with large stack limits and large mmap()
++	 * allocations.
++	 */
++
++#ifdef CONFIG_PAX_SEGMEXEC
++	if (mm->pax_flags & MF_PAX_SEGMEXEC)
++		mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE;
++	else
++#endif
++
++	mm->mmap_base = TASK_UNMAPPED_BASE;
++
++#ifdef CONFIG_PAX_RANDMMAP
++	if (mm->pax_flags & MF_PAX_RANDMMAP)
++		mm->mmap_base += mm->delta_mmap;
++#endif
++
++	mm->free_area_cache = mm->mmap_base;
++	mm->cached_hole_size = ~0UL;
++	addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
++	/*
++	 * Restore the topdown base:
++	 */
++	mm->mmap_base = base;
++	mm->free_area_cache = base;
++	mm->cached_hole_size = ~0UL;
++
++	return addr;
+ }
+diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c
+index 0514890..3dbebce 100644
+--- a/arch/x86/kernel/sys_x86_64.c
++++ b/arch/x86/kernel/sys_x86_64.c
+@@ -95,8 +95,8 @@ out:
+ 	return error;
+ }
+ 
+-static void find_start_end(unsigned long flags, unsigned long *begin,
+-			   unsigned long *end)
++static void find_start_end(struct mm_struct *mm, unsigned long flags,
++			   unsigned long *begin, unsigned long *end)
+ {
+ 	if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) {
+ 		unsigned long new_begin;
+@@ -115,7 +115,7 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
+ 				*begin = new_begin;
+ 		}
+ 	} else {
+-		*begin = TASK_UNMAPPED_BASE;
++		*begin = mm->mmap_base;
+ 		*end = TASK_SIZE;
+ 	}
+ }
+@@ -132,16 +132,19 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
+ 	if (flags & MAP_FIXED)
+ 		return addr;
+ 
+-	find_start_end(flags, &begin, &end);
++	find_start_end(mm, flags, &begin, &end);
+ 
+ 	if (len > end)
+ 		return -ENOMEM;
+ 
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
+ 	if (addr) {
+ 		addr = PAGE_ALIGN(addr);
+ 		vma = find_vma(mm, addr);
+-		if (end - len >= addr &&
+-		    (!vma || addr + len <= vma->vm_start))
++		if (end - len >= addr && check_heap_stack_gap(vma, addr, len))
+ 			return addr;
+ 	}
+ 	if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32))
+@@ -172,7 +175,7 @@ full_search:
+ 			}
+ 			return -ENOMEM;
+ 		}
+-		if (!vma || addr + len <= vma->vm_start) {
++		if (check_heap_stack_gap(vma, addr, len)) {
+ 			/*
+ 			 * Remember the place where we stopped the search:
+ 			 */
+@@ -195,7 +198,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ {
+ 	struct vm_area_struct *vma;
+ 	struct mm_struct *mm = current->mm;
+-	unsigned long addr = addr0;
++	unsigned long base = mm->mmap_base, addr = addr0;
+ 
+ 	/* requested length too big for entire address space */
+ 	if (len > TASK_SIZE)
+@@ -208,13 +211,18 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 	if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT))
+ 		goto bottomup;
+ 
++#ifdef CONFIG_PAX_RANDMMAP
++	if (!(mm->pax_flags & MF_PAX_RANDMMAP))
++#endif
++
+ 	/* requesting a specific address */
+ 	if (addr) {
+ 		addr = PAGE_ALIGN(addr);
+-		vma = find_vma(mm, addr);
+-		if (TASK_SIZE - len >= addr &&
+-				(!vma || addr + len <= vma->vm_start))
+-			return addr;
++		if (TASK_SIZE - len >= addr) {
++			vma = find_vma(mm, addr);
++			if (check_heap_stack_gap(vma, addr, len))
++				return addr;
++		}
+ 	}
+ 
+ 	/* check if free_area_cache is useful for us */
+@@ -232,7 +240,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 						    ALIGN_TOPDOWN);
+ 
+ 		vma = find_vma(mm, tmp_addr);
+-		if (!vma || tmp_addr + len <= vma->vm_start)
++		if (check_heap_stack_gap(vma, tmp_addr, len))
+ 			/* remember the address as a hint for next time */
+ 			return mm->free_area_cache = tmp_addr;
+ 	}
+@@ -251,7 +259,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 		 * return with success:
+ 		 */
+ 		vma = find_vma(mm, addr);
+-		if (!vma || addr+len <= vma->vm_start)
++		if (check_heap_stack_gap(vma, addr, len))
+ 			/* remember the address as a hint for next time */
+ 			return mm->free_area_cache = addr;
+ 
+@@ -260,8 +268,8 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
+ 			mm->cached_hole_size = vma->vm_start - addr;
+ 
+ 		/* try just below the current vma->vm_start */
+-		addr = vma->vm_start-len;
+-	} while (len < vma->vm_start);
++		addr = skip_heap_stack_gap(vma, len);
++	} while (!IS_ERR_VALUE(addr));
+ 
+ bottomup:
+ 	/*
+@@ -270,13 +278,21 @@ bottomup:
+ 	 * can happen with large stack limits and large mmap()
+ 	 * allocations.
+ 	 */
++	mm->mmap_base = TASK_UNMAPPED_BASE;
++
++#ifdef CONFIG_PAX_RANDMMAP
++	if (mm->pax_flags & MF_PAX_RANDMMAP)
++		mm->mmap_base += mm->delta_mmap;
++#endif
++
++	mm->free_area_cache = mm->mmap_base;
+ 	mm->cached_hole_size = ~0UL;
+-	mm->free_area_cache = TASK_UNMAPPED_BASE;
+ 	addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags);
+ 	/*
+ 	 * Restore the topdown base:
+ 	 */
+-	mm->free_area_cache = mm->mmap_base;
++	mm->mmap_base = base;
++	mm->free_area_cache = base;
+ 	mm->cached_hole_size = ~0UL;
+ 
+ 	return addr;
+diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
+index 9a0e312..e6f66f2 100644
+--- a/arch/x86/kernel/syscall_table_32.S
++++ b/arch/x86/kernel/syscall_table_32.S
+@@ -1,3 +1,4 @@
++.section .rodata,"a",@progbits
+ ENTRY(sys_call_table)
+ 	.long sys_restart_syscall	/* 0 - old "setup()" system call, used for restarting */
+ 	.long sys_exit
+diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c
+index e2410e2..4fe3fbc 100644