summaryrefslogtreecommitdiff
path: root/arch/x86/um
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/um')
-rw-r--r--arch/x86/um/Kconfig70
-rw-r--r--arch/x86/um/Makefile45
-rw-r--r--arch/x86/um/asm/arch_hweight.h6
-rw-r--r--arch/x86/um/asm/archparam.h20
-rw-r--r--arch/x86/um/asm/checksum.h10
-rw-r--r--arch/x86/um/asm/checksum_32.h201
-rw-r--r--arch/x86/um/asm/checksum_64.h144
-rw-r--r--arch/x86/um/asm/elf.h221
-rw-r--r--arch/x86/um/asm/module.h23
-rw-r--r--arch/x86/um/asm/processor.h15
-rw-r--r--arch/x86/um/asm/processor_32.h73
-rw-r--r--arch/x86/um/asm/processor_64.h51
-rw-r--r--arch/x86/um/asm/ptrace.h5
-rw-r--r--arch/x86/um/asm/ptrace_32.h51
-rw-r--r--arch/x86/um/asm/ptrace_64.h72
-rw-r--r--arch/x86/um/asm/system.h133
-rw-r--r--arch/x86/um/asm/vm-flags.h25
-rw-r--r--arch/x86/um/bug.c21
-rw-r--r--arch/x86/um/bugs_32.c74
-rw-r--r--arch/x86/um/bugs_64.c15
-rw-r--r--arch/x86/um/checksum_32.S458
-rw-r--r--arch/x86/um/delay_32.c60
-rw-r--r--arch/x86/um/delay_64.c60
-rw-r--r--arch/x86/um/elfcore.c83
-rw-r--r--arch/x86/um/fault.c28
-rw-r--r--arch/x86/um/ksyms.c13
-rw-r--r--arch/x86/um/ldt.c502
-rw-r--r--arch/x86/um/mem_32.c62
-rw-r--r--arch/x86/um/mem_64.c26
-rw-r--r--arch/x86/um/os-Linux/Makefile13
-rw-r--r--arch/x86/um/os-Linux/mcontext.c31
-rw-r--r--arch/x86/um/os-Linux/prctl.c12
-rw-r--r--arch/x86/um/os-Linux/registers.c111
-rw-r--r--arch/x86/um/os-Linux/task_size.c150
-rw-r--r--arch/x86/um/os-Linux/tls.c35
-rw-r--r--arch/x86/um/ptrace_32.c273
-rw-r--r--arch/x86/um/ptrace_64.c271
-rw-r--r--arch/x86/um/ptrace_user.c21
-rw-r--r--arch/x86/um/setjmp_32.S58
-rw-r--r--arch/x86/um/setjmp_64.S54
-rw-r--r--arch/x86/um/shared/sysdep/archsetjmp.h5
-rw-r--r--arch/x86/um/shared/sysdep/archsetjmp_32.h22
-rw-r--r--arch/x86/um/shared/sysdep/archsetjmp_64.h24
-rw-r--r--arch/x86/um/shared/sysdep/faultinfo.h5
-rw-r--r--arch/x86/um/shared/sysdep/faultinfo_32.h35
-rw-r--r--arch/x86/um/shared/sysdep/faultinfo_64.h35
-rw-r--r--arch/x86/um/shared/sysdep/host_ldt.h5
-rw-r--r--arch/x86/um/shared/sysdep/host_ldt_32.h34
-rw-r--r--arch/x86/um/shared/sysdep/host_ldt_64.h38
-rw-r--r--arch/x86/um/shared/sysdep/kernel-offsets.h21
-rw-r--r--arch/x86/um/shared/sysdep/mcontext.h31
-rw-r--r--arch/x86/um/shared/sysdep/ptrace.h5
-rw-r--r--arch/x86/um/shared/sysdep/ptrace_32.h117
-rw-r--r--arch/x86/um/shared/sysdep/ptrace_64.h160
-rw-r--r--arch/x86/um/shared/sysdep/ptrace_user.h5
-rw-r--r--arch/x86/um/shared/sysdep/ptrace_user_32.h26
-rw-r--r--arch/x86/um/shared/sysdep/ptrace_user_64.h38
-rw-r--r--arch/x86/um/shared/sysdep/skas_ptrace.h22
-rw-r--r--arch/x86/um/shared/sysdep/stub.h14
-rw-r--r--arch/x86/um/shared/sysdep/stub_32.h93
-rw-r--r--arch/x86/um/shared/sysdep/stub_64.h99
-rw-r--r--arch/x86/um/shared/sysdep/syscalls.h5
-rw-r--r--arch/x86/um/shared/sysdep/syscalls_32.h20
-rw-r--r--arch/x86/um/shared/sysdep/syscalls_64.h32
-rw-r--r--arch/x86/um/shared/sysdep/tls.h5
-rw-r--r--arch/x86/um/shared/sysdep/tls_32.h32
-rw-r--r--arch/x86/um/shared/sysdep/tls_64.h29
-rw-r--r--arch/x86/um/signal_32.c498
-rw-r--r--arch/x86/um/signal_64.c255
-rw-r--r--arch/x86/um/stub_32.S51
-rw-r--r--arch/x86/um/stub_64.S66
-rw-r--r--arch/x86/um/stub_segv.c19
-rw-r--r--arch/x86/um/sys_call_table_32.S28
-rw-r--r--arch/x86/um/sys_call_table_64.c64
-rw-r--r--arch/x86/um/syscalls_32.c66
-rw-r--r--arch/x86/um/syscalls_64.c102
-rw-r--r--arch/x86/um/sysrq_32.c101
-rw-r--r--arch/x86/um/sysrq_64.c41
-rw-r--r--arch/x86/um/tls_32.c396
-rw-r--r--arch/x86/um/tls_64.c17
-rw-r--r--arch/x86/um/user-offsets.c79
-rw-r--r--arch/x86/um/vdso/Makefile90
-rw-r--r--arch/x86/um/vdso/checkundef.sh10
-rw-r--r--arch/x86/um/vdso/um_vdso.c71
-rw-r--r--arch/x86/um/vdso/vdso-layout.lds.S64
-rw-r--r--arch/x86/um/vdso/vdso-note.S12
-rw-r--r--arch/x86/um/vdso/vdso.S10
-rw-r--r--arch/x86/um/vdso/vdso.lds.S32
-rw-r--r--arch/x86/um/vdso/vma.c74
89 files changed, 6699 insertions, 0 deletions
diff --git a/arch/x86/um/Kconfig b/arch/x86/um/Kconfig
new file mode 100644
index 000000000000..21bebe63df66
--- /dev/null
+++ b/arch/x86/um/Kconfig
@@ -0,0 +1,70 @@
+mainmenu "User Mode Linux/$SUBARCH $KERNELVERSION Kernel Configuration"
+
+source "arch/um/Kconfig.common"
+
+menu "UML-specific options"
+
+menu "Host processor type and features"
+
+config CMPXCHG_LOCAL
+ bool
+ default n
+
+config CMPXCHG_DOUBLE
+ bool
+ default n
+
+source "arch/x86/Kconfig.cpu"
+
+endmenu
+
+config UML_X86
+ def_bool y
+ select GENERIC_FIND_FIRST_BIT
+
+config 64BIT
+ bool
+ default SUBARCH = "x86_64"
+
+config X86_32
+ def_bool !64BIT
+ select HAVE_AOUT
+
+config X86_64
+ def_bool 64BIT
+
+config RWSEM_XCHGADD_ALGORITHM
+ def_bool X86_XADD && 64BIT
+
+config RWSEM_GENERIC_SPINLOCK
+ def_bool !RWSEM_XCHGADD_ALGORITHM
+
+config 3_LEVEL_PGTABLES
+ bool "Three-level pagetables (EXPERIMENTAL)" if !64BIT
+ default 64BIT
+ depends on EXPERIMENTAL
+ help
+ Three-level pagetables will let UML have more than 4G of physical
+ memory. All the memory that can't be mapped directly will be treated
+ as high memory.
+
+ However, this it experimental on 32-bit architectures, so if unsure say
+ N (on x86-64 it's automatically enabled, instead, as it's safe there).
+
+config ARCH_HAS_SC_SIGNALS
+ def_bool !64BIT
+
+config ARCH_REUSE_HOST_VSYSCALL_AREA
+ def_bool !64BIT
+
+config SMP_BROKEN
+ def_bool 64BIT
+
+config GENERIC_HWEIGHT
+ def_bool y
+
+source "arch/um/Kconfig.um"
+
+endmenu
+
+source "arch/um/Kconfig.rest"
diff --git a/arch/x86/um/Makefile b/arch/x86/um/Makefile
new file mode 100644
index 000000000000..df419896693f
--- /dev/null
+++ b/arch/x86/um/Makefile
@@ -0,0 +1,45 @@
+#
+# Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+#
+
+ifeq ($(CONFIG_X86_32),y)
+ BITS := 32
+else
+ BITS := 64
+endif
+
+obj-y = bug.o bugs_$(BITS).o delay_$(BITS).o fault.o ksyms.o ldt.o \
+ ptrace_$(BITS).o ptrace_user.o setjmp_$(BITS).o signal_$(BITS).o \
+ stub_$(BITS).o stub_segv.o syscalls_$(BITS).o \
+ sys_call_table_$(BITS).o sysrq_$(BITS).o tls_$(BITS).o \
+ mem_$(BITS).o subarch.o os-$(OS)/
+
+ifeq ($(CONFIG_X86_32),y)
+
+obj-y += checksum_32.o
+obj-$(CONFIG_BINFMT_ELF) += elfcore.o
+
+subarch-y = ../lib/string_32.o ../lib/atomic64_32.o ../lib/atomic64_cx8_32.o
+subarch-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += ../lib/rwsem.o
+subarch-$(CONFIG_HIGHMEM) += ../mm/highmem_32.o
+
+else
+
+obj-y += vdso/
+
+subarch-y = ../lib/csum-partial_64.o ../lib/memcpy_64.o ../lib/thunk_64.o \
+ ../lib/rwsem.o
+
+endif
+
+subarch-$(CONFIG_MODULES) += ../kernel/module.o
+
+USER_OBJS := bugs_$(BITS).o ptrace_user.o fault.o
+
+extra-y += user-offsets.s
+$(obj)/user-offsets.s: c_flags = -Wp,-MD,$(depfile) $(USER_CFLAGS)
+
+UNPROFILE_OBJS := stub_segv.o
+CFLAGS_stub_segv.o := $(CFLAGS_NO_HARDENING)
+
+include arch/um/scripts/Makefile.rules
diff --git a/arch/x86/um/asm/arch_hweight.h b/arch/x86/um/asm/arch_hweight.h
new file mode 100644
index 000000000000..c656cf443f4a
--- /dev/null
+++ b/arch/x86/um/asm/arch_hweight.h
@@ -0,0 +1,6 @@
+#ifndef _ASM_UM_HWEIGHT_H
+#define _ASM_UM_HWEIGHT_H
+
+#include <asm-generic/bitops/arch_hweight.h>
+
+#endif
diff --git a/arch/x86/um/asm/archparam.h b/arch/x86/um/asm/archparam.h
new file mode 100644
index 000000000000..c17cf68dda0f
--- /dev/null
+++ b/arch/x86/um/asm/archparam.h
@@ -0,0 +1,20 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Copyright 2003 PathScale, Inc.
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_ARCHPARAM_H
+#define __UM_ARCHPARAM_H
+
+#ifdef CONFIG_X86_32
+
+#ifdef CONFIG_X86_PAE
+#define LAST_PKMAP 512
+#else
+#define LAST_PKMAP 1024
+#endif
+
+#endif
+
+#endif
diff --git a/arch/x86/um/asm/checksum.h b/arch/x86/um/asm/checksum.h
new file mode 100644
index 000000000000..b6efe2381b5d
--- /dev/null
+++ b/arch/x86/um/asm/checksum.h
@@ -0,0 +1,10 @@
+#ifndef __UM_CHECKSUM_H
+#define __UM_CHECKSUM_H
+
+#ifdef CONFIG_X86_32
+# include "checksum_32.h"
+#else
+# include "checksum_64.h"
+#endif
+
+#endif
diff --git a/arch/x86/um/asm/checksum_32.h b/arch/x86/um/asm/checksum_32.h
new file mode 100644
index 000000000000..caab74252e27
--- /dev/null
+++ b/arch/x86/um/asm/checksum_32.h
@@ -0,0 +1,201 @@
+/*
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_SYSDEP_CHECKSUM_H
+#define __UM_SYSDEP_CHECKSUM_H
+
+#include "linux/in6.h"
+#include "linux/string.h"
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+__wsum csum_partial(const void *buff, int len, __wsum sum);
+
+/*
+ * Note: when you get a NULL pointer exception here this means someone
+ * passed in an incorrect kernel address to one of these functions.
+ *
+ * If you use these functions directly please don't forget the
+ * access_ok().
+ */
+
+static __inline__
+__wsum csum_partial_copy_nocheck(const void *src, void *dst,
+ int len, __wsum sum)
+{
+ memcpy(dst, src, len);
+ return csum_partial(dst, len, sum);
+}
+
+/*
+ * the same as csum_partial, but copies from src while it
+ * checksums, and handles user-space pointer exceptions correctly, when needed.
+ *
+ * here even more important to align src and dst on a 32-bit (or even
+ * better 64-bit) boundary
+ */
+
+static __inline__
+__wsum csum_partial_copy_from_user(const void __user *src, void *dst,
+ int len, __wsum sum, int *err_ptr)
+{
+ if (copy_from_user(dst, src, len)) {
+ *err_ptr = -EFAULT;
+ return (__force __wsum)-1;
+ }
+
+ return csum_partial(dst, len, sum);
+}
+
+/*
+ * This is a version of ip_compute_csum() optimized for IP headers,
+ * which always checksum on 4 octet boundaries.
+ *
+ * By Jorge Cwik <jorge@laser.satlink.net>, adapted for linux by
+ * Arnt Gulbrandsen.
+ */
+static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
+{
+ unsigned int sum;
+
+ __asm__ __volatile__(
+ "movl (%1), %0 ;\n"
+ "subl $4, %2 ;\n"
+ "jbe 2f ;\n"
+ "addl 4(%1), %0 ;\n"
+ "adcl 8(%1), %0 ;\n"
+ "adcl 12(%1), %0 ;\n"
+"1: adcl 16(%1), %0 ;\n"
+ "lea 4(%1), %1 ;\n"
+ "decl %2 ;\n"
+ "jne 1b ;\n"
+ "adcl $0, %0 ;\n"
+ "movl %0, %2 ;\n"
+ "shrl $16, %0 ;\n"
+ "addw %w2, %w0 ;\n"
+ "adcl $0, %0 ;\n"
+ "notl %0 ;\n"
+"2: ;\n"
+ /* Since the input registers which are loaded with iph and ipl
+ are modified, we must also specify them as outputs, or gcc
+ will assume they contain their original values. */
+ : "=r" (sum), "=r" (iph), "=r" (ihl)
+ : "1" (iph), "2" (ihl)
+ : "memory");
+ return (__force __sum16)sum;
+}
+
+/*
+ * Fold a partial checksum
+ */
+
+static inline __sum16 csum_fold(__wsum sum)
+{
+ __asm__(
+ "addl %1, %0 ;\n"
+ "adcl $0xffff, %0 ;\n"
+ : "=r" (sum)
+ : "r" ((__force u32)sum << 16),
+ "0" ((__force u32)sum & 0xffff0000)
+ );
+ return (__force __sum16)(~(__force u32)sum >> 16);
+}
+
+static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+ unsigned short len,
+ unsigned short proto,
+ __wsum sum)
+{
+ __asm__(
+ "addl %1, %0 ;\n"
+ "adcl %2, %0 ;\n"
+ "adcl %3, %0 ;\n"
+ "adcl $0, %0 ;\n"
+ : "=r" (sum)
+ : "g" (daddr), "g"(saddr), "g"((len + proto) << 8), "0"(sum));
+ return sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+ unsigned short len,
+ unsigned short proto,
+ __wsum sum)
+{
+ return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+
+static inline __sum16 ip_compute_csum(const void *buff, int len)
+{
+ return csum_fold (csum_partial(buff, len, 0));
+}
+
+#define _HAVE_ARCH_IPV6_CSUM
+static __inline__ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
+ const struct in6_addr *daddr,
+ __u32 len, unsigned short proto,
+ __wsum sum)
+{
+ __asm__(
+ "addl 0(%1), %0 ;\n"
+ "adcl 4(%1), %0 ;\n"
+ "adcl 8(%1), %0 ;\n"
+ "adcl 12(%1), %0 ;\n"
+ "adcl 0(%2), %0 ;\n"
+ "adcl 4(%2), %0 ;\n"
+ "adcl 8(%2), %0 ;\n"
+ "adcl 12(%2), %0 ;\n"
+ "adcl %3, %0 ;\n"
+ "adcl %4, %0 ;\n"
+ "adcl $0, %0 ;\n"
+ : "=&r" (sum)
+ : "r" (saddr), "r" (daddr),
+ "r"(htonl(len)), "r"(htonl(proto)), "0"(sum));
+
+ return csum_fold(sum);
+}
+
+/*
+ * Copy and checksum to user
+ */
+#define HAVE_CSUM_COPY_USER
+static __inline__ __wsum csum_and_copy_to_user(const void *src,
+ void __user *dst,
+ int len, __wsum sum, int *err_ptr)
+{
+ if (access_ok(VERIFY_WRITE, dst, len)) {
+ if (copy_to_user(dst, src, len)) {
+ *err_ptr = -EFAULT;
+ return (__force __wsum)-1;
+ }
+
+ return csum_partial(src, len, sum);
+ }
+
+ if (len)
+ *err_ptr = -EFAULT;
+
+ return (__force __wsum)-1; /* invalid checksum */
+}
+
+#endif
+
diff --git a/arch/x86/um/asm/checksum_64.h b/arch/x86/um/asm/checksum_64.h
new file mode 100644
index 000000000000..a5be9031ea85
--- /dev/null
+++ b/arch/x86/um/asm/checksum_64.h
@@ -0,0 +1,144 @@
+/*
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_SYSDEP_CHECKSUM_H
+#define __UM_SYSDEP_CHECKSUM_H
+
+#include "linux/string.h"
+#include "linux/in6.h"
+#include "asm/uaccess.h"
+
+extern __wsum csum_partial(const void *buff, int len, __wsum sum);
+
+/*
+ * Note: when you get a NULL pointer exception here this means someone
+ * passed in an incorrect kernel address to one of these functions.
+ *
+ * If you use these functions directly please don't forget the
+ * access_ok().
+ */
+
+static __inline__
+__wsum csum_partial_copy_nocheck(const void *src, void *dst,
+ int len, __wsum sum)
+{
+ memcpy(dst, src, len);
+ return(csum_partial(dst, len, sum));
+}
+
+static __inline__
+__wsum csum_partial_copy_from_user(const void __user *src,
+ void *dst, int len, __wsum sum,
+ int *err_ptr)
+{
+ if (copy_from_user(dst, src, len)) {
+ *err_ptr = -EFAULT;
+ return (__force __wsum)-1;
+ }
+ return csum_partial(dst, len, sum);
+}
+
+/**
+ * csum_fold - Fold and invert a 32bit checksum.
+ * sum: 32bit unfolded sum
+ *
+ * Fold a 32bit running checksum to 16bit and invert it. This is usually
+ * the last step before putting a checksum into a packet.
+ * Make sure not to mix with 64bit checksums.
+ */
+static inline __sum16 csum_fold(__wsum sum)
+{
+ __asm__(
+ " addl %1,%0\n"
+ " adcl $0xffff,%0"
+ : "=r" (sum)
+ : "r" ((__force u32)sum << 16),
+ "0" ((__force u32)sum & 0xffff0000)
+ );
+ return (__force __sum16)(~(__force u32)sum >> 16);
+}
+
+/**
+ * csum_tcpup_nofold - Compute an IPv4 pseudo header checksum.
+ * @saddr: source address
+ * @daddr: destination address
+ * @len: length of packet
+ * @proto: ip protocol of packet
+ * @sum: initial sum to be added in (32bit unfolded)
+ *
+ * Returns the pseudo header checksum the input data. Result is
+ * 32bit unfolded.
+ */
+static inline __wsum
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
+ unsigned short proto, __wsum sum)
+{
+ asm(" addl %1, %0\n"
+ " adcl %2, %0\n"
+ " adcl %3, %0\n"
+ " adcl $0, %0\n"
+ : "=r" (sum)
+ : "g" (daddr), "g" (saddr), "g" ((len + proto) << 8), "0" (sum));
+ return sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
+ unsigned short len,
+ unsigned short proto,
+ __wsum sum)
+{
+ return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+}
+
+/**
+ * ip_fast_csum - Compute the IPv4 header checksum efficiently.
+ * iph: ipv4 header
+ * ihl: length of header / 4
+ */
+static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl)
+{
+ unsigned int sum;
+
+ asm( " movl (%1), %0\n"
+ " subl $4, %2\n"
+ " jbe 2f\n"
+ " addl 4(%1), %0\n"
+ " adcl 8(%1), %0\n"
+ " adcl 12(%1), %0\n"
+ "1: adcl 16(%1), %0\n"
+ " lea 4(%1), %1\n"
+ " decl %2\n"
+ " jne 1b\n"
+ " adcl $0, %0\n"
+ " movl %0, %2\n"
+ " shrl $16, %0\n"
+ " addw %w2, %w0\n"
+ " adcl $0, %0\n"
+ " notl %0\n"
+ "2:"
+ /* Since the input registers which are loaded with iph and ipl
+ are modified, we must also specify them as outputs, or gcc
+ will assume they contain their original values. */
+ : "=r" (sum), "=r" (iph), "=r" (ihl)
+ : "1" (iph), "2" (ihl)
+ : "memory");
+ return (__force __sum16)sum;
+}
+
+static inline unsigned add32_with_carry(unsigned a, unsigned b)
+{
+ asm("addl %2,%0\n\t"
+ "adcl $0,%0"
+ : "=r" (a)
+ : "0" (a), "r" (b));
+ return a;
+}
+
+extern __sum16 ip_compute_csum(const void *buff, int len);
+
+#endif
diff --git a/arch/x86/um/asm/elf.h b/arch/x86/um/asm/elf.h
new file mode 100644
index 000000000000..f3b0633b69a1
--- /dev/null
+++ b/arch/x86/um/asm/elf.h
@@ -0,0 +1,221 @@
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+#ifndef __UM_ELF_X86_H
+#define __UM_ELF_X86_H
+
+#include <asm/user.h>
+#include "skas.h"
+
+#ifdef CONFIG_X86_32
+
+#define R_386_NONE 0
+#define R_386_32 1
+#define R_386_PC32 2
+#define R_386_GOT32 3
+#define R_386_PLT32 4
+#define R_386_COPY 5
+#define R_386_GLOB_DAT 6
+#define R_386_JMP_SLOT 7
+#define R_386_RELATIVE 8
+#define R_386_GOTOFF 9
+#define R_386_GOTPC 10
+#define R_386_NUM 11
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x) \
+ (((x)->e_machine == EM_386) || ((x)->e_machine == EM_486))
+
+#define ELF_CLASS ELFCLASS32
+#define ELF_DATA ELFDATA2LSB
+#define ELF_ARCH EM_386
+
+#define ELF_PLAT_INIT(regs, load_addr) do { \
+ PT_REGS_EBX(regs) = 0; \
+ PT_REGS_ECX(regs) = 0; \
+ PT_REGS_EDX(regs) = 0; \
+ PT_REGS_ESI(regs) = 0; \
+ PT_REGS_EDI(regs) = 0; \
+ PT_REGS_EBP(regs) = 0; \
+ PT_REGS_EAX(regs) = 0; \
+} while (0)
+
+/* Shamelessly stolen from include/asm-i386/elf.h */
+
+#define ELF_CORE_COPY_REGS(pr_reg, regs) do { \
+ pr_reg[0] = PT_REGS_EBX(regs); \
+ pr_reg[1] = PT_REGS_ECX(regs); \
+ pr_reg[2] = PT_REGS_EDX(regs); \
+ pr_reg[3] = PT_REGS_ESI(regs); \
+ pr_reg[4] = PT_REGS_EDI(regs); \
+ pr_reg[5] = PT_REGS_EBP(regs); \
+ pr_reg[6] = PT_REGS_EAX(regs); \
+ pr_reg[7] = PT_REGS_DS(regs); \
+ pr_reg[8] = PT_REGS_ES(regs); \
+ /* fake once used fs and gs selectors? */ \
+ pr_reg[9] = PT_REGS_DS(regs); \
+ pr_reg[10] = PT_REGS_DS(regs); \
+ pr_reg[11] = PT_REGS_SYSCALL_NR(regs); \
+ pr_reg[12] = PT_REGS_IP(regs); \
+ pr_reg[13] = PT_REGS_CS(regs); \
+ pr_reg[14] = PT_REGS_EFLAGS(regs); \
+ pr_reg[15] = PT_REGS_SP(regs); \
+ pr_reg[16] = PT_REGS_SS(regs); \
+} while (0);
+
+extern char * elf_aux_platform;
+#define ELF_PLATFORM (elf_aux_platform)
+
+extern unsigned long vsyscall_ehdr;
+extern unsigned long vsyscall_end;
+extern unsigned long __kernel_vsyscall;
+
+/*
+ * This is the range that is readable by user mode, and things
+ * acting like user mode such as get_user_pages.
+ */
+#define FIXADDR_USER_START vsyscall_ehdr
+#define FIXADDR_USER_END vsyscall_end
+
+
+/*
+ * Architecture-neutral AT_ values in 0-17, leave some room
+ * for more of them, start the x86-specific ones at 32.
+ */
+#define AT_SYSINFO 32
+#define AT_SYSINFO_EHDR 33
+
+#define ARCH_DLINFO \
+do { \
+ if ( vsyscall_ehdr ) { \
+ NEW_AUX_ENT(AT_SYSINFO, __kernel_vsyscall); \
+ NEW_AUX_ENT(AT_SYSINFO_EHDR, vsyscall_ehdr); \
+ } \
+} while (0)
+
+#else
+
+/* x86-64 relocation types, taken from asm-x86_64/elf.h */
+#define R_X86_64_NONE 0 /* No reloc */
+#define R_X86_64_64 1 /* Direct 64 bit */
+#define R_X86_64_PC32 2 /* PC relative 32 bit signed */
+#define R_X86_64_GOT32 3 /* 32 bit GOT entry */
+#define R_X86_64_PLT32 4 /* 32 bit PLT address */
+#define R_X86_64_COPY 5 /* Copy symbol at runtime */
+#define R_X86_64_GLOB_DAT 6 /* Create GOT entry */
+#define R_X86_64_JUMP_SLOT 7 /* Create PLT entry */
+#define R_X86_64_RELATIVE 8 /* Adjust by program base */
+#define R_X86_64_GOTPCREL 9 /* 32 bit signed pc relative
+ offset to GOT */
+#define R_X86_64_32 10 /* Direct 32 bit zero extended */
+#define R_X86_64_32S 11 /* Direct 32 bit sign extended */
+#define R_X86_64_16 12 /* Direct 16 bit zero extended */
+#define R_X86_64_PC16 13 /* 16 bit sign extended pc relative */
+#define R_X86_64_8 14 /* Direct 8 bit sign extended */
+#define R_X86_64_PC8 15 /* 8 bit sign extended pc relative */
+
+#define R_X86_64_NUM 16
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x) \
+ ((x)->e_machine == EM_X86_64)
+
+#define ELF_CLASS ELFCLASS64
+#define ELF_DATA ELFDATA2LSB
+#define ELF_ARCH EM_X86_64
+
+#define ELF_PLAT_INIT(regs, load_addr) do { \
+ PT_REGS_RBX(regs) = 0; \
+ PT_REGS_RCX(regs) = 0; \
+ PT_REGS_RDX(regs) = 0; \
+ PT_REGS_RSI(regs) = 0; \
+ PT_REGS_RDI(regs) = 0; \
+ PT_REGS_RBP(regs) = 0; \
+ PT_REGS_RAX(regs) = 0; \
+ PT_REGS_R8(regs) = 0; \
+ PT_REGS_R9(regs) = 0; \
+ PT_REGS_R10(regs) = 0; \
+ PT_REGS_R11(regs) = 0; \
+ PT_REGS_R12(regs) = 0; \
+ PT_REGS_R13(regs) = 0; \
+ PT_REGS_R14(regs) = 0; \
+ PT_REGS_R15(regs) = 0; \
+} while (0)
+
+#define ELF_CORE_COPY_REGS(pr_reg, _regs) \
+ (pr_reg)[0] = (_regs)->regs.gp[0]; \
+ (pr_reg)[1] = (_regs)->regs.gp[1]; \
+ (pr_reg)[2] = (_regs)->regs.gp[2]; \
+ (pr_reg)[3] = (_regs)->regs.gp[3]; \
+ (pr_reg)[4] = (_regs)->regs.gp[4]; \
+ (pr_reg)[5] = (_regs)->regs.gp[5]; \
+ (pr_reg)[6] = (_regs)->regs.gp[6]; \
+ (pr_reg)[7] = (_regs)->regs.gp[7]; \
+ (pr_reg)[8] = (_regs)->regs.gp[8]; \
+ (pr_reg)[9] = (_regs)->regs.gp[9]; \
+ (pr_reg)[10] = (_regs)->regs.gp[10]; \
+ (pr_reg)[11] = (_regs)->regs.gp[11]; \
+ (pr_reg)[12] = (_regs)->regs.gp[12]; \
+ (pr_reg)[13] = (_regs)->regs.gp[13]; \
+ (pr_reg)[14] = (_regs)->regs.gp[14]; \
+ (pr_reg)[15] = (_regs)->regs.gp[15]; \
+ (pr_reg)[16] = (_regs)->regs.gp[16]; \
+ (pr_reg)[17] = (_regs)->regs.gp[17]; \
+ (pr_reg)[18] = (_regs)->regs.gp[18]; \
+ (pr_reg)[19] = (_regs)->regs.gp[19]; \
+ (pr_reg)[20] = (_regs)->regs.gp[20]; \
+ (pr_reg)[21] = current->thread.arch.fs; \
+ (pr_reg)[22] = 0; \
+ (pr_reg)[23] = 0; \
+ (pr_reg)[24] = 0; \
+ (pr_reg)[25] = 0; \
+ (pr_reg)[26] = 0;
+
+#define ELF_PLATFORM "x86_64"
+
+/* No user-accessible fixmap addresses, i.e. vsyscall */
+#define FIXADDR_USER_START 0
+#define FIXADDR_USER_END 0
+
+#define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
+struct linux_binprm;
+extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+ int uses_interp);
+
+extern unsigned long um_vdso_addr;
+#define AT_SYSINFO_EHDR 33
+#define ARCH_DLINFO NEW_AUX_ENT(AT_SYSINFO_EHDR, um_vdso_addr)
+
+#endif
+
+typedef unsigned long elf_greg_t;
+
+#define ELF_NGREG (sizeof (struct user_regs_struct) / sizeof(elf_greg_t))
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+typedef struct user_i387_struct elf_fpregset_t;
+
+#define task_pt_regs(t) (&(t)->thread.regs)
+
+struct task_struct;
+
+extern int elf_core_copy_fpregs(struct task_struct *t, elf_fpregset_t *fpu);
+
+#define ELF_CORE_COPY_FPREGS(t, fpu) elf_core_copy_fpregs(t, fpu)
+
+#define ELF_EXEC_PAGESIZE 4096
+
+#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3)
+
+extern long elf_aux_hwcap;
+#define ELF_HWCAP (elf_aux_hwcap)
+
+#define SET_PERSONALITY(ex) do ; while(0)
+#define __HAVE_ARCH_GATE_AREA 1
+
+#endif
diff --git a/arch/x86/um/asm/module.h b/arch/x86/um/asm/module.h
new file mode 100644
index 000000000000..61af80e932eb
--- /dev/null
+++ b/arch/x86/um/asm/module.h
@@ -0,0 +1,23 @@
+#ifndef __UM_MODULE_H
+#define __UM_MODULE_H
+
+/* UML is simple */
+struct mod_arch_specific
+{
+};
+
+#ifdef CONFIG_X86_32
+
+#define Elf_Shdr Elf32_Shdr
+#define Elf_Sym Elf32_Sym
+#define Elf_Ehdr Elf32_Ehdr
+
+#else
+
+#define Elf_Shdr Elf64_Shdr
+#define Elf_Sym Elf64_Sym
+#define Elf_Ehdr Elf64_Ehdr
+
+#endif
+
+#endif
diff --git a/arch/x86/um/asm/processor.h b/arch/x86/um/asm/processor.h
new file mode 100644
index 000000000000..d3ac1cecf0f4
--- /dev/null
+++ b/arch/x86/um/asm/processor.h
@@ -0,0 +1,15 @@
+#ifndef __UM_PROCESSOR_H
+#define __UM_PROCESSOR_H
+
+/* include faultinfo structure */
+#include <sysdep/faultinfo.h>
+
+#ifdef CONFIG_X86_32
+# include "processor_32.h"
+#else
+# include "processor_64.h"
+#endif
+
+#include <asm/processor-generic.h>
+
+#endif
diff --git a/arch/x86/um/asm/processor_32.h b/arch/x86/um/asm/processor_32.h
new file mode 100644
index 000000000000..ae0d189aafcf
--- /dev/null
+++ b/arch/x86/um/asm/processor_32.h
@@ -0,0 +1,73 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_PROCESSOR_I386_H
+#define __UM_PROCESSOR_I386_H
+
+#include <linux/string.h>
+#include <sysdep/host_ldt.h>
+#include <asm/segment.h>
+
+extern int host_has_cmov;
+
+struct uml_tls_struct {
+ struct user_desc tls;
+ unsigned flushed:1;
+ unsigned present:1;
+};
+
+struct arch_thread {
+ struct uml_tls_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
+ unsigned long debugregs[8];
+ int debugregs_seq;
+ struct faultinfo faultinfo;
+};
+
+#define INIT_ARCH_THREAD { \
+ .tls_array = { [ 0 ... GDT_ENTRY_TLS_ENTRIES - 1 ] = \
+ { .present = 0, .flushed = 0 } }, \
+ .debugregs = { [ 0 ... 7 ] = 0 }, \
+ .debugregs_seq = 0, \
+ .faultinfo = { 0, 0, 0 } \
+}
+
+static inline void arch_flush_thread(struct arch_thread *thread)
+{
+ /* Clear any TLS still hanging */
+ memset(&thread->tls_array, 0, sizeof(thread->tls_array));
+}
+
+static inline void arch_copy_thread(struct arch_thread *from,
+ struct arch_thread *to)
+{
+ memcpy(&to->tls_array, &from->tls_array, sizeof(from->tls_array));
+}
+
+#include <asm/user.h>
+
+/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
+static inline void rep_nop(void)
+{
+ __asm__ __volatile__("rep;nop": : :"memory");
+}
+
+#define cpu_relax() rep_nop()
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter"). Stolen
+ * from asm-i386/processor.h
+ */
+#define current_text_addr() \
+ ({ void *pc; __asm__("movl $1f,%0\n1:":"=g" (pc)); pc; })
+
+#define ARCH_IS_STACKGROW(address) \
+ (address + 32 >= UPT_SP(&current->thread.regs.regs))
+
+#define KSTK_EIP(tsk) KSTK_REG(tsk, EIP)
+#define KSTK_ESP(tsk) KSTK_REG(tsk, UESP)
+#define KSTK_EBP(tsk) KSTK_REG(tsk, EBP)
+
+#endif
diff --git a/arch/x86/um/asm/processor_64.h b/arch/x86/um/asm/processor_64.h
new file mode 100644
index 000000000000..6db812b24f48
--- /dev/null
+++ b/arch/x86/um/asm/processor_64.h
@@ -0,0 +1,51 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_PROCESSOR_X86_64_H
+#define __UM_PROCESSOR_X86_64_H
+
+struct arch_thread {
+ unsigned long debugregs[8];
+ int debugregs_seq;
+ unsigned long fs;
+ struct faultinfo faultinfo;
+};
+
+/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
+static inline void rep_nop(void)
+{
+ __asm__ __volatile__("rep;nop": : :"memory");
+}
+
+#define cpu_relax() rep_nop()
+
+#define INIT_ARCH_THREAD { .debugregs = { [ 0 ... 7 ] = 0 }, \
+ .debugregs_seq = 0, \
+ .fs = 0, \
+ .faultinfo = { 0, 0, 0 } }
+
+static inline void arch_flush_thread(struct arch_thread *thread)
+{
+}
+
+static inline void arch_copy_thread(struct arch_thread *from,
+ struct arch_thread *to)
+{
+ to->fs = from->fs;
+}
+
+#include <asm/user.h>
+
+#define current_text_addr() \
+ ({ void *pc; __asm__("movq $1f,%0\n1:":"=g" (pc)); pc; })
+
+#define ARCH_IS_STACKGROW(address) \
+ (address + 128 >= UPT_SP(&current->thread.regs.regs))
+
+#define KSTK_EIP(tsk) KSTK_REG(tsk, RIP)
+#define KSTK_ESP(tsk) KSTK_REG(tsk, RSP)
+
+#endif
diff --git a/arch/x86/um/asm/ptrace.h b/arch/x86/um/asm/ptrace.h
new file mode 100644
index 000000000000..c8aca8c501b0
--- /dev/null
+++ b/arch/x86/um/asm/ptrace.h
@@ -0,0 +1,5 @@
+#ifdef CONFIG_X86_32
+# include "ptrace_32.h"
+#else
+# include "ptrace_64.h"
+#endif
diff --git a/arch/x86/um/asm/ptrace_32.h b/arch/x86/um/asm/ptrace_32.h
new file mode 100644
index 000000000000..5d2a59112537
--- /dev/null
+++ b/arch/x86/um/asm/ptrace_32.h
@@ -0,0 +1,51 @@
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_PTRACE_I386_H
+#define __UM_PTRACE_I386_H
+
+#define HOST_AUDIT_ARCH AUDIT_ARCH_I386
+
+#include "linux/compiler.h"
+#include "asm/ptrace-generic.h"
+
+#define PT_REGS_EAX(r) UPT_EAX(&(r)->regs)
+#define PT_REGS_EBX(r) UPT_EBX(&(r)->regs)
+#define PT_REGS_ECX(r) UPT_ECX(&(r)->regs)
+#define PT_REGS_EDX(r) UPT_EDX(&(r)->regs)
+#define PT_REGS_ESI(r) UPT_ESI(&(r)->regs)
+#define PT_REGS_EDI(r) UPT_EDI(&(r)->regs)
+#define PT_REGS_EBP(r) UPT_EBP(&(r)->regs)
+
+#define PT_REGS_CS(r) UPT_CS(&(r)->regs)
+#define PT_REGS_SS(r) UPT_SS(&(r)->regs)
+#define PT_REGS_DS(r) UPT_DS(&(r)->regs)
+#define PT_REGS_ES(r) UPT_ES(&(r)->regs)
+#define PT_REGS_FS(r) UPT_FS(&(r)->regs)
+#define PT_REGS_GS(r) UPT_GS(&(r)->regs)
+
+#define PT_REGS_EFLAGS(r) UPT_EFLAGS(&(r)->regs)
+
+#define PT_REGS_ORIG_SYSCALL(r) PT_REGS_EAX(r)
+#define PT_REGS_SYSCALL_RET(r) PT_REGS_EAX(r)
+#define PT_FIX_EXEC_STACK(sp) do ; while(0)
+
+#define profile_pc(regs) PT_REGS_IP(regs)
+
+#define user_mode(r) UPT_IS_USER(&(r)->regs)
+
+/*
+ * Forward declaration to avoid including sysdep/tls.h, which causes a
+ * circular include, and compilation failures.
+ */
+struct user_desc;
+
+extern int ptrace_get_thread_area(struct task_struct *child, int idx,
+ struct user_desc __user *user_desc);
+
+extern int ptrace_set_thread_area(struct task_struct *child, int idx,
+ struct user_desc __user *user_desc);
+
+#endif
diff --git a/arch/x86/um/asm/ptrace_64.h b/arch/x86/um/asm/ptrace_64.h
new file mode 100644
index 000000000000..83d8c473b905
--- /dev/null
+++ b/arch/x86/um/asm/ptrace_64.h
@@ -0,0 +1,72 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#ifndef __UM_PTRACE_X86_64_H
+#define __UM_PTRACE_X86_64_H
+
+#include "linux/compiler.h"
+#include "asm/errno.h"
+
+#define __FRAME_OFFSETS /* Needed to get the R* macros */
+#include "asm/ptrace-generic.h"
+
+#define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64
+
+#define PT_REGS_RBX(r) UPT_RBX(&(r)->regs)
+#define PT_REGS_RCX(r) UPT_RCX(&(r)->regs)
+#define PT_REGS_RDX(r) UPT_RDX(&(r)->regs)
+#define PT_REGS_RSI(r) UPT_RSI(&(r)->regs)
+#define PT_REGS_RDI(r) UPT_RDI(&(r)->regs)
+#define PT_REGS_RBP(r) UPT_RBP(&(r)->regs)
+#define PT_REGS_RAX(r) UPT_RAX(&(r)->regs)
+#define PT_REGS_R8(r) UPT_R8(&(r)->regs)
+#define PT_REGS_R9(r) UPT_R9(&(r)->regs)
+#define PT_REGS_R10(r) UPT_R10(&(r)->regs)
+#define PT_REGS_R11(r) UPT_R11(&(r)->regs)
+#define PT_REGS_R12(r) UPT_R12(&(r)->regs)
+#define PT_REGS_R13(r) UPT_R13(&(r)->regs)
+#define PT_REGS_R14(r) UPT_R14(&(r)->regs)
+#define PT_REGS_R15(r) UPT_R15(&(r)->regs)
+
+#define PT_REGS_FS(r) UPT_FS(&(r)->regs)
+#define PT_REGS_GS(r) UPT_GS(&(r)->regs)
+#define PT_REGS_DS(r) UPT_DS(&(r)->regs)
+#define PT_REGS_ES(r) UPT_ES(&(r)->regs)
+#define PT_REGS_SS(r) UPT_SS(&(r)->regs)
+#define PT_REGS_CS(r) UPT_CS(&(r)->regs)
+
+#define PT_REGS_ORIG_RAX(r) UPT_ORIG_RAX(&(r)->regs)
+#define PT_REGS_RIP(r) UPT_IP(&(r)->regs)
+#define PT_REGS_RSP(r) UPT_SP(&(r)->regs)
+
+#define PT_REGS_EFLAGS(r) UPT_EFLAGS(&(r)->regs)
+
+/* XXX */
+#define user_mode(r) UPT_IS_USER(&(r)->regs)
+#define PT_REGS_ORIG_SYSCALL(r) PT_REGS_RAX(r)
+#define PT_REGS_SYSCALL_RET(r) PT_REGS_RAX(r)
+
+#define PT_FIX_EXEC_STACK(sp) do ; while(0)
+
+#define profile_pc(regs) PT_REGS_IP(regs)
+
+struct user_desc;
+
+static inline int ptrace_get_thread_area(struct task_struct *child, int idx,
+ struct user_desc __user *user_desc)
+{
+ return -ENOSYS;
+}
+
+static inline int ptrace_set_thread_area(struct task_struct *child, int idx,
+ struct user_desc __user *user_desc)
+{
+ return -ENOSYS;
+}
+
+extern long arch_prctl(struct task_struct *task, int code,
+ unsigned long __user *addr);
+#endif
diff --git a/arch/x86/um/asm/system.h b/arch/x86/um/asm/system.h
new file mode 100644
index 000000000000..a89113bc74f2
--- /dev/null
+++ b/arch/x86/um/asm/system.h
@@ -0,0 +1,133 @@
+#ifndef _ASM_X86_SYSTEM_H_
+#define _ASM_X86_SYSTEM_H_
+
+#include <asm/asm.h>
+#include <asm/segment.h>
+#include <asm/cpufeature.h>
+#include <asm/cmpxchg.h>
+#include <asm/nops.h>
+#include <asm/system-um.h>
+
+#include <linux/kernel.h>
+#include <linux/irqflags.h>
+
+/* entries in ARCH_DLINFO: */
+#ifdef CONFIG_IA32_EMULATION
+# define AT_VECTOR_SIZE_ARCH 2
+#else
+# define AT_VECTOR_SIZE_ARCH 1
+#endif
+
+extern unsigned long arch_align_stack(unsigned long sp);
+
+void default_idle(void);
+
+/*
+ * Force strict CPU ordering.
+ * And yes, this is required on UP too when we're talking
+ * to devices.
+ */
+#ifdef CONFIG_X86_32
+/*
+ * Some non-Intel clones support out of order store. wmb() ceases to be a
+ * nop for these.
+ */
+#define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
+#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
+#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
+#else
+#define mb() asm volatile("mfence":::"memory")
+#define rmb() asm volatile("lfence":::"memory")
+#define wmb() asm volatile("sfence" ::: "memory")
+#endif
+
+/**
+ * read_barrier_depends - Flush all pending reads that subsequents reads
+ * depend on.
+ *
+ * No data-dependent reads from memory-like regions are ever reordered
+ * over this barrier. All reads preceding this primitive are guaranteed
+ * to access memory (but not necessarily other CPUs' caches) before any
+ * reads following this primitive that depend on the data return by
+ * any of the preceding reads. This primitive is much lighter weight than
+ * rmb() on most CPUs, and is never heavier weight than is
+ * rmb().
+ *
+ * These ordering constraints are respected by both the local CPU
+ * and the compiler.
+ *
+ * Ordering is not guaranteed by anything other than these primitives,
+ * not even by data dependencies. See the documentation for
+ * memory_barrier() for examples and URLs to more information.
+ *
+ * For example, the following code would force ordering (the initial
+ * value of "a" is zero, "b" is one, and "p" is "&a"):
+ *
+ * <programlisting>
+ * CPU 0 CPU 1
+ *
+ * b = 2;
+ * memory_barrier();
+ * p = &b; q = p;
+ * read_barrier_depends();
+ * d = *q;
+ * </programlisting>
+ *
+ * because the read of "*q" depends on the read of "p" and these
+ * two reads are separated by a read_barrier_depends(). However,
+ * the following code, with the same initial values for "a" and "b":
+ *
+ * <programlisting>
+ * CPU 0 CPU 1
+ *
+ * a = 2;
+ * memory_barrier();
+ * b = 3; y = b;
+ * read_barrier_depends();
+ * x = a;
+ * </programlisting>
+ *
+ * does not enforce ordering, since there is no data dependency between
+ * the read of "a" and the read of "b". Therefore, on some CPUs, such
+ * as Alpha, "y" could be set to 3 and "x" to 0. Use rmb()
+ * in cases like this where there are no data dependencies.
+ **/
+
+#define read_barrier_depends() do { } while (0)
+
+#ifdef CONFIG_SMP
+#define smp_mb() mb()
+#ifdef CONFIG_X86_PPRO_FENCE
+# define smp_rmb() rmb()
+#else
+# define smp_rmb() barrier()
+#endif
+#ifdef CONFIG_X86_OOSTORE
+# define smp_wmb() wmb()
+#else
+# define smp_wmb() barrier()
+#endif
+#define smp_read_barrier_depends() read_barrier_depends()
+#define set_mb(var, value) do { (void)xchg(&var, value); } while (0)
+#else
+#define smp_mb() barrier()
+#define smp_rmb() barrier()
+#define smp_wmb() barrier()
+#define smp_read_barrier_depends() do { } while (0)
+#define set_mb(var, value) do { var = value; barrier(); } while (0)
+#endif
+
+/*
+ * Stop RDTSC speculation. This is needed when you need to use RDTSC
+ * (or get_cycles or vread that possibly accesses the TSC) in a defined
+ * code region.
+ *
+ * (Could use an alternative three way for this if there was one.)
+ */
+static inline void rdtsc_barrier(void)
+{
+ alternative(ASM_NOP3, "mfence", X86_FEATURE_MFENCE_RDTSC);
+ alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC);
+}
+
+#endif
diff --git a/arch/x86/um/asm/vm-flags.h b/arch/x86/um/asm/vm-flags.h
new file mode 100644
index 000000000000..7c297e9e2413
--- /dev/null
+++ b/arch/x86/um/asm/vm-flags.h
@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com)
+ * Copyright 2003 PathScale, Inc.
+ * Licensed under the GPL
+ */
+
+#ifndef __VM_FLAGS_X86_H
+#define __VM_FLAGS_X86_H
+
+#ifdef CONFIG_X86_32
+
+#define VM_DATA_DEFAULT_FLAGS \
+ (VM_READ | VM_WRITE | \
+ ((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
+ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+#else
+
+#define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \
+ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+#define VM_STACK_DEFAULT_FLAGS (VM_GROWSDOWN | VM_READ | VM_WRITE | \
+ VM_EXEC | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+#endif
+#endif
diff --git a/arch/x86/um/bug.c b/arch/x86/um/bug.c
new file mode 100644
index 000000000000..e8034e363d83
--- /dev/null
+++ b/arch/x86/um/bug.c
@@ -0,0 +1,21 @@
+/*
+ * Copyright (C) 2006 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL V2
+ */
+
+#include <linux/uaccess.h>
+
+/*
+ * Mostly copied from i386/x86_86 - eliminated the eip < PAGE_OFFSET because
+ * that's not relevant in skas mode.
+ */
+
+int is_valid_bugaddr(unsigned long eip)
+{
+ unsigned short ud2;
+
+ if (probe_kernel_address((unsigned short __user *)eip, ud2))
+ return 0;
+
+ return ud2 == 0x0b0f;
+}
diff --git a/arch/x86/um/bugs_32.c b/arch/x86/um/bugs_32.c
new file mode 100644
index 000000000000..7058e1fa903b
--- /dev/null
+++ b/arch/x86/um/bugs_32.c
@@ -0,0 +1,74 @@
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include <signal.h>
+#include "kern_util.h"
+#include "longjmp.h"
+#include "task.h"
+#include "sysdep/ptrace.h"
+
+/* Set during early boot */
+static int host_has_cmov = 1;
+static jmp_buf cmov_test_return;
+
+static void cmov_sigill_test_handler(int sig)
+{
+ host_has_cmov = 0;
+ longjmp(cmov_test_return, 1);
+}
+
+void arch_check_bugs(void)
+{
+ struct sigaction old, new;
+
+ printk(UM_KERN_INFO "Checking for host processor cmov support...");
+ new.sa_handler = cmov_sigill_test_handler;
+
+ /* Make sure that SIGILL is enabled after the handler longjmps back */
+ new.sa_flags = SA_NODEFER;
+ sigemptyset(&new.sa_mask);
+ sigaction(SIGILL, &new, &old);
+
+ if (setjmp(cmov_test_return) == 0) {
+ unsigned long foo = 0;
+ __asm__ __volatile__("cmovz %0, %1" : "=r" (foo) : "0" (foo));
+ printk(UM_KERN_CONT "Yes\n");
+ } else
+ printk(UM_KERN_CONT "No\n");
+
+ sigaction(SIGILL, &old, &new);
+}
+
+void arch_examine_signal(int sig, struct uml_pt_regs *regs)
+{
+ unsigned char tmp[2];
+
+ /*
+ * This is testing for a cmov (0x0f 0x4x) instruction causing a
+ * SIGILL in init.
+ */
+ if ((sig != SIGILL) || (TASK_PID(get_current()) != 1))
+ return;
+
+ if (copy_from_user_proc(tmp, (void *) UPT_IP(regs), 2)) {
+ printk(UM_KERN_ERR "SIGILL in init, could not read "
+ "instructions!\n");
+ return;
+ }
+
+ if ((tmp[0] != 0x0f) || ((tmp[1] & 0xf0) != 0x40))
+ return;
+
+ if (host_has_cmov == 0)
+ printk(UM_KERN_ERR "SIGILL caused by cmov, which this "
+ "processor doesn't implement. Boot a filesystem "
+ "compiled for older processors");
+ else if (host_has_cmov == 1)
+ printk(UM_KERN_ERR "SIGILL caused by cmov, which this "
+ "processor claims to implement");
+ else
+ printk(UM_KERN_ERR "Bad value for host_has_cmov (%d)",
+ host_has_cmov);
+}
diff --git a/arch/x86/um/bugs_64.c b/arch/x86/um/bugs_64.c
new file mode 100644
index 000000000000..44e02ba2a265
--- /dev/null
+++ b/arch/x86/um/bugs_64.c
@@ -0,0 +1,15 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#include "sysdep/ptrace.h"
+
+void arch_check_bugs(void)
+{
+}
+
+void arch_examine_signal(int sig, struct uml_pt_regs *regs)
+{
+}
diff --git a/arch/x86/um/checksum_32.S b/arch/x86/um/checksum_32.S
new file mode 100644
index 000000000000..f058d2f82e18
--- /dev/null
+++ b/arch/x86/um/checksum_32.S
@@ -0,0 +1,458 @@
+/*
+ * INET An implementation of the TCP/IP protocol suite for the LINUX
+ * operating system. INET is implemented using the BSD Socket
+ * interface as the means of communication with the user level.
+ *
+ * IP/TCP/UDP checksumming routines
+ *
+ * Authors: Jorge Cwik, <jorge@laser.satlink.net>
+ * Arnt Gulbrandsen, <agulbra@nvg.unit.no>
+ * Tom May, <ftom@netcom.com>
+ * Pentium Pro/II routines:
+ * Alexander Kjeldaas <astor@guardian.no>
+ * Finn Arne Gangstad <finnag@guardian.no>
+ * Lots of code moved from tcp.c and ip.c; see those files
+ * for more names.
+ *
+ * Changes: Ingo Molnar, converted csum_partial_copy() to 2.1 exception
+ * handling.
+ * Andi Kleen, add zeroing on error
+ * converted to pure assembler
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <asm/errno.h>
+
+/*
+ * computes a partial checksum, e.g. for TCP/UDP fragments
+ */
+
+/*
+unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum)
+ */
+
+.text
+.align 4
+.globl csum_partial
+
+#ifndef CONFIG_X86_USE_PPRO_CHECKSUM
+
+ /*
+ * Experiments with Ethernet and SLIP connections show that buff
+ * is aligned on either a 2-byte or 4-byte boundary. We get at
+ * least a twofold speedup on 486 and Pentium if it is 4-byte aligned.
+ * Fortunately, it is easy to convert 2-byte alignment to 4-byte
+ * alignment for the unrolled loop.
+ */
+csum_partial:
+ pushl %esi
+ pushl %ebx
+ movl 20(%esp),%eax # Function arg: unsigned int sum
+ movl 16(%esp),%ecx # Function arg: int len
+ movl 12(%esp),%esi # Function arg: unsigned char *buff
+ testl $2, %esi # Check alignment.
+ jz 2f # Jump if alignment is ok.
+ subl $2, %ecx # Alignment uses up two bytes.
+ jae 1f # Jump if we had at least two bytes.
+ addl $2, %ecx # ecx was < 2. Deal with it.
+ jmp 4f
+1: movw (%esi), %bx
+ addl $2, %esi
+ addw %bx, %ax
+ adcl $0, %eax
+2:
+ movl %ecx, %edx
+ shrl $5, %ecx
+ jz 2f
+ testl %esi, %esi
+1: movl (%esi), %ebx
+ adcl %ebx, %eax
+ movl 4(%esi), %ebx
+ adcl %ebx, %eax
+ movl 8(%esi), %ebx
+ adcl %ebx, %eax
+ movl 12(%esi), %ebx
+ adcl %ebx, %eax
+ movl 16(%esi), %ebx
+ adcl %ebx, %eax
+ movl 20(%esi), %ebx
+ adcl %ebx, %eax
+ movl 24(%esi), %ebx
+ adcl %ebx, %eax
+ movl 28(%esi), %ebx
+ adcl %ebx, %eax
+ lea 32(%esi), %esi
+ dec %ecx
+ jne 1b
+ adcl $0, %eax
+2: movl %edx, %ecx
+ andl $0x1c, %edx
+ je 4f
+ shrl $2, %edx # This clears CF
+3: adcl (%esi), %eax
+ lea 4(%esi), %esi
+ dec %edx
+ jne 3b
+ adcl $0, %eax
+4: andl $3, %ecx
+ jz 7f
+ cmpl $2, %ecx
+ jb 5f
+ movw (%esi),%cx
+ leal 2(%esi),%esi
+ je 6f
+ shll $16,%ecx
+5: movb (%esi),%cl
+6: addl %ecx,%eax
+ adcl $0, %eax
+7:
+ popl %ebx
+ popl %esi
+ ret
+
+#else
+
+/* Version for PentiumII/PPro */
+
+csum_partial:
+ pushl %esi
+ pushl %ebx
+ movl 20(%esp),%eax # Function arg: unsigned int sum
+ movl 16(%esp),%ecx # Function arg: int len
+ movl 12(%esp),%esi # Function arg: const unsigned char *buf
+
+ testl $2, %esi
+ jnz 30f
+10:
+ movl %ecx, %edx
+ movl %ecx, %ebx
+ andl $0x7c, %ebx
+ shrl $7, %ecx
+ addl %ebx,%esi
+ shrl $2, %ebx
+ negl %ebx
+ lea 45f(%ebx,%ebx,2), %ebx
+ testl %esi, %esi
+ jmp *%ebx
+
+ # Handle 2-byte-aligned regions
+20: addw (%esi), %ax
+ lea 2(%esi), %esi
+ adcl $0, %eax
+ jmp 10b
+
+30: subl $2, %ecx
+ ja 20b
+ je 32f
+ movzbl (%esi),%ebx # csumming 1 byte, 2-aligned
+ addl %ebx, %eax
+ adcl $0, %eax
+ jmp 80f
+32:
+ addw (%esi), %ax # csumming 2 bytes, 2-aligned
+ adcl $0, %eax
+ jmp 80f
+
+40:
+ addl -128(%esi), %eax
+ adcl -124(%esi), %eax
+ adcl -120(%esi), %eax
+ adcl -116(%esi), %eax
+ adcl -112(%esi), %eax
+ adcl -108(%esi), %eax
+ adcl -104(%esi), %eax
+ adcl -100(%esi), %eax
+ adcl -96(%esi), %eax
+ adcl -92(%esi), %eax
+ adcl -88(%esi), %eax
+ adcl -84(%esi), %eax
+ adcl -80(%esi), %eax
+ adcl -76(%esi), %eax
+ adcl -72(%esi), %eax
+ adcl -68(%esi), %eax
+ adcl -64(%esi), %eax
+ adcl -60(%esi), %eax
+ adcl -56(%esi), %eax
+ adcl -52(%esi), %eax
+ adcl -48(%esi), %eax
+ adcl -44(%esi), %eax
+ adcl -40(%esi), %eax
+ adcl -36(%esi), %eax
+ adcl -32(%esi), %eax
+ adcl -28(%esi), %eax
+ adcl -24(%esi), %eax
+ adcl -20(%esi), %eax
+ adcl -16(%esi), %eax
+ adcl -12(%esi), %eax
+ adcl -8(%esi), %eax
+ adcl -4(%esi), %eax
+45:
+ lea 128(%esi), %esi
+ adcl $0, %eax
+ dec %ecx
+ jge 40b
+ movl %edx, %ecx
+50: andl $3, %ecx
+ jz 80f
+
+ # Handle the last 1-3 bytes without jumping
+ notl %ecx # 1->2, 2->1, 3->0, higher bits are masked
+ movl $0xffffff,%ebx # by the shll and shrl instructions
+ shll $3,%ecx
+ shrl %cl,%ebx
+ andl -128(%esi),%ebx # esi is 4-aligned so should be ok
+ addl %ebx,%eax
+ adcl $0,%eax
+80:
+ popl %ebx
+ popl %esi
+ ret
+
+#endif
+
+/*
+unsigned int csum_partial_copy_generic (const char *src, char *dst,
+ int len, int sum, int *src_err_ptr, int *dst_err_ptr)
+ */
+
+/*
+ * Copy from ds while checksumming, otherwise like csum_partial
+ *
+ * The macros SRC and DST specify the type of access for the instruction.
+ * thus we can call a custom exception handler for all access types.
+ *
+ * FIXME: could someone double-check whether I haven't mixed up some SRC and
+ * DST definitions? It's damn hard to trigger all cases. I hope I got
+ * them all but there's no guarantee.
+ */
+
+#define SRC(y...) \
+ 9999: y; \
+ .section __ex_table, "a"; \
+ .long 9999b, 6001f ; \
+ .previous
+
+#define DST(y...) \
+ 9999: y; \
+ .section __ex_table, "a"; \
+ .long 9999b, 6002f ; \
+ .previous
+
+.align 4
+
+#ifndef CONFIG_X86_USE_PPRO_CHECKSUM
+
+#define ARGBASE 16
+#define FP 12
+
+csum_partial_copy_generic_i386:
+ subl $4,%esp
+ pushl %edi
+ pushl %esi
+ pushl %ebx
+ movl ARGBASE+16(%esp),%eax # sum
+ movl ARGBASE+12(%esp),%ecx # len
+ movl ARGBASE+4(%esp),%esi # src
+ movl ARGBASE+8(%esp),%edi # dst
+
+ testl $2, %edi # Check alignment.
+ jz 2f # Jump if alignment is ok.
+ subl $2, %ecx # Alignment uses up two bytes.
+ jae 1f # Jump if we had at least two bytes.
+ addl $2, %ecx # ecx was < 2. Deal with it.
+ jmp 4f
+SRC(1: movw (%esi), %bx )
+ addl $2, %esi
+DST( movw %bx, (%edi) )
+ addl $2, %edi
+ addw %bx, %ax
+ adcl $0, %eax
+2:
+ movl %ecx, FP(%esp)
+ shrl $5, %ecx
+ jz 2f
+ testl %esi, %esi
+SRC(1: movl (%esi), %ebx )
+SRC( movl 4(%esi), %edx )
+ adcl %ebx, %eax
+DST( movl %ebx, (%edi) )
+ adcl %edx, %eax
+DST( movl %edx, 4(%edi) )
+
+SRC( movl 8(%esi), %ebx )
+SRC( movl 12(%esi), %edx )
+ adcl %ebx, %eax
+DST( movl %ebx, 8(%edi) )
+ adcl %edx, %eax
+DST( movl %edx, 12(%edi) )
+
+SRC( movl 16(%esi), %ebx )
+SRC( movl 20(%esi), %edx )
+ adcl %ebx, %eax
+DST( movl %ebx, 16(%edi) )
+ adcl %edx, %eax
+DST( movl %edx, 20(%edi) )
+
+SRC( movl 24(%esi), %ebx )
+SRC( movl 28(%esi), %edx )
+ adcl %ebx, %eax
+DST( movl %ebx, 24(%edi) )
+ adcl %edx, %eax
+DST( movl %edx, 28(%edi) )
+
+ lea 32(%esi), %esi
+ lea 32(%edi), %edi
+ dec %ecx
+ jne 1b
+ adcl $0, %eax
+2: movl FP(%esp), %edx
+ movl %edx, %ecx
+ andl $0x1c, %edx
+ je 4f
+ shrl $2, %edx # This clears CF
+SRC(3: movl (%esi), %ebx )
+ adcl %ebx, %eax
+DST( movl %ebx, (%edi) )
+ lea 4(%esi), %esi
+ lea 4(%edi), %edi
+ dec %edx
+ jne 3b
+ adcl $0, %eax
+4: andl $3, %ecx
+ jz 7f
+ cmpl $2, %ecx
+ jb 5f
+SRC( movw (%esi), %cx )
+ leal 2(%esi), %esi
+DST( movw %cx, (%edi) )
+ leal 2(%edi), %edi
+ je 6f
+ shll $16,%ecx
+SRC(5: movb (%esi), %cl )
+DST( movb %cl, (%edi) )
+6: addl %ecx, %eax
+ adcl $0, %eax
+7:
+5000:
+
+# Exception handler:
+.section .fixup, "ax"
+
+6001:
+ movl ARGBASE+20(%esp), %ebx # src_err_ptr
+ movl $-EFAULT, (%ebx)
+
+ # zero the complete destination - computing the rest
+ # is too much work
+ movl ARGBASE+8(%esp), %edi # dst
+ movl ARGBASE+12(%esp), %ecx # len
+ xorl %eax,%eax
+ rep ; stosb
+
+ jmp 5000b
+
+6002:
+ movl ARGBASE+24(%esp), %ebx # dst_err_ptr
+ movl $-EFAULT,(%ebx)
+ jmp 5000b
+
+.previous
+
+ popl %ebx
+ popl %esi
+ popl %edi
+ popl %ecx # equivalent to addl $4,%esp
+ ret
+
+#else
+
+/* Version for PentiumII/PPro */
+
+#define ROUND1(x) \
+ SRC(movl x(%esi), %ebx ) ; \
+ addl %ebx, %eax ; \
+ DST(movl %ebx, x(%edi) ) ;
+
+#define ROUND(x) \
+ SRC(movl x(%esi), %ebx ) ; \
+ adcl %ebx, %eax ; \
+ DST(movl %ebx, x(%edi) ) ;
+
+#define ARGBASE 12
+
+csum_partial_copy_generic_i386:
+ pushl %ebx
+ pushl %edi
+ pushl %esi
+ movl ARGBASE+4(%esp),%esi #src
+ movl ARGBASE+8(%esp),%edi #dst
+ movl ARGBASE+12(%esp),%ecx #len
+ movl ARGBASE+16(%esp),%eax #sum
+# movl %ecx, %edx
+ movl %ecx, %ebx
+ movl %esi, %edx
+ shrl $6, %ecx
+ andl $0x3c, %ebx
+ negl %ebx
+ subl %ebx, %esi
+ subl %ebx, %edi
+ lea -1(%esi),%edx
+ andl $-32,%edx
+ lea 3f(%ebx,%ebx), %ebx
+ testl %esi, %esi
+ jmp *%ebx
+1: addl $64,%esi
+ addl $64,%edi
+ SRC(movb -32(%edx),%bl) ; SRC(movb (%edx),%bl)
+ ROUND1(-64) ROUND(-60) ROUND(-56) ROUND(-52)
+ ROUND (-48) ROUND(-44) ROUND(-40) ROUND(-36)
+ ROUND (-32) ROUND(-28) ROUND(-24) ROUND(-20)
+ ROUND (-16) ROUND(-12) ROUND(-8) ROUND(-4)
+3: adcl $0,%eax
+ addl $64, %edx
+ dec %ecx
+ jge 1b
+4: movl ARGBASE+12(%esp),%edx #len
+ andl $3, %edx
+ jz 7f
+ cmpl $2, %edx
+ jb 5f
+SRC( movw (%esi), %dx )
+ leal 2(%esi), %esi
+DST( movw %dx, (%edi) )
+ leal 2(%edi), %edi
+ je 6f
+ shll $16,%edx
+5:
+SRC( movb (%esi), %dl )
+DST( movb %dl, (%edi) )
+6: addl %edx, %eax
+ adcl $0, %eax
+7:
+.section .fixup, "ax"
+6001: movl ARGBASE+20(%esp), %ebx # src_err_ptr
+ movl $-EFAULT, (%ebx)
+ # zero the complete destination (computing the rest is too much work)
+ movl ARGBASE+8(%esp),%edi # dst
+ movl ARGBASE+12(%esp),%ecx # len
+ xorl %eax,%eax
+ rep; stosb
+ jmp 7b
+6002: movl ARGBASE+24(%esp), %ebx # dst_err_ptr
+ movl $-EFAULT, (%ebx)
+ jmp 7b
+.previous
+
+ popl %esi
+ popl %edi
+ popl %ebx
+ ret
+
+#undef ROUND
+#undef ROUND1
+
+#endif
diff --git a/arch/x86/um/delay_32.c b/arch/x86/um/delay_32.c
new file mode 100644
index 000000000000..f3fe1a688f7e
--- /dev/null
+++ b/arch/x86/um/delay_32.c
@@ -0,0 +1,60 @@
+/*
+ * Copyright (C) 2011 Richard Weinberger <richrd@nod.at>
+ * Mostly copied from arch/x86/lib/delay.c
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <asm/param.h>
+
+void __delay(unsigned long loops)
+{
+ asm volatile(
+ "test %0,%0\n"
+ "jz 3f\n"
+ "jmp 1f\n"
+
+ ".align 16\n"
+ "1: jmp 2f\n"
+
+ ".align 16\n"
+ "2: dec %0\n"
+ " jnz 2b\n"
+ "3: dec %0\n"
+
+ : /* we don't need output */
+ : "a" (loops)
+ );
+}
+EXPORT_SYMBOL(__delay);
+
+inline void __const_udelay(unsigned long xloops)
+{
+ int d0;
+
+ xloops *= 4;
+ asm("mull %%edx"
+ : "=d" (xloops), "=&a" (d0)
+ : "1" (xloops), "0"
+ (loops_per_jiffy * (HZ/4)));
+
+ __delay(++xloops);
+}
+EXPORT_SYMBOL(__const_udelay);
+
+void __udelay(unsigned long usecs)
+{
+ __const_udelay(usecs * 0x000010c7); /* 2**32 / 1000000 (rounded up) */
+}
+EXPORT_SYMBOL(__udelay);
+
+void __ndelay(unsigned long nsecs)
+{
+ __const_udelay(nsecs * 0x00005); /* 2**32 / 1000000000 (rounded up) */
+}
+EXPORT_SYMBOL(__ndelay);
diff --git a/arch/x86/um/delay_64.c b/arch/x86/um/delay_64.c
new file mode 100644
index 000000000000..f3fe1a688f7e
--- /dev/null
+++ b/arch/x86/um/delay_64.c
@@ -0,0 +1,60 @@
+/*
+ * Copyright (C) 2011 Richard Weinberger <richrd@nod.at>
+ * Mostly copied from arch/x86/lib/delay.c
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <asm/param.h>
+
+void __delay(unsigned long loops)
+{
+ asm volatile(
+ "test %0,%0\n"
+ "jz 3f\n"
+ "jmp 1f\n"
+
+ ".align 16\n"
+ "1: jmp 2f\n"
+
+ ".align 16\n"
+ "2: dec %0\n"
+ " jnz 2b\n"
+ "3: dec %0\n"
+
+ : /* we don't need output */
+ : "a" (loops)
+ );
+}
+EXPORT_SYMBOL(__delay);
+
+inline void __const_udelay(unsigned long xloops)
+{
+ int d0;
+
+ xloops *= 4;
+ asm("mull %%edx"
+ : "=d" (xloops), "=&a" (d0)
+ : "1" (xloops), "0"
+ (loops_per_jiffy * (HZ/4)));
+
+ __delay(++xloops);
+}
+EXPORT_SYMBOL(__const_udelay);
+
+void __udelay(unsigned long usecs)
+{
+ __const_udelay(usecs * 0x000010c7); /* 2**32 / 1000000 (rounded up) */
+}
+EXPORT_SYMBOL(__udelay);
+
+void __ndelay(unsigned long nsecs)
+{
+ __const_udelay(nsecs * 0x00005); /* 2**32 / 1000000000 (rounded up) */
+}
+EXPORT_SYMBOL(__ndelay);
diff --git a/arch/x86/um/elfcore.c b/arch/x86/um/elfcore.c
new file mode 100644
index 000000000000..6bb49b687c97
--- /dev/null
+++ b/arch/x86/um/elfcore.c
@@ -0,0 +1,83 @@
+#include <linux/elf.h>
+#include <linux/coredump.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+
+#include <asm/elf.h>
+
+
+Elf32_Half elf_core_extra_phdrs(void)
+{
+ return vsyscall_ehdr ? (((struct elfhdr *)vsyscall_ehdr)->e_phnum) : 0;
+}
+
+int elf_core_write_extra_phdrs(struct file *file, loff_t offset, size_t *size,
+ unsigned long limit)
+{
+ if ( vsyscall_ehdr ) {
+ const struct elfhdr *const ehdrp =
+ (struct elfhdr *) vsyscall_ehdr;
+ const struct elf_phdr *const phdrp =
+ (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff);
+ int i;
+ Elf32_Off ofs = 0;
+
+ for (i = 0; i < ehdrp->e_phnum; ++i) {
+ struct elf_phdr phdr = phdrp[i];
+
+ if (phdr.p_type == PT_LOAD) {
+ ofs = phdr.p_offset = offset;
+ offset += phdr.p_filesz;
+ } else {
+ phdr.p_offset += ofs;
+ }
+ phdr.p_paddr = 0; /* match other core phdrs */
+ *size += sizeof(phdr);
+ if (*size > limit
+ || !dump_write(file, &phdr, sizeof(phdr)))
+ return 0;
+ }
+ }
+ return 1;
+}
+
+int elf_core_write_extra_data(struct file *file, size_t *size,
+ unsigned long limit)
+{
+ if ( vsyscall_ehdr ) {
+ const struct elfhdr *const ehdrp =
+ (struct elfhdr *) vsyscall_ehdr;
+ const struct elf_phdr *const phdrp =
+ (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff);
+ int i;
+
+ for (i = 0; i < ehdrp->e_phnum; ++i) {
+ if (phdrp[i].p_type == PT_LOAD) {
+ void *addr = (void *) phdrp[i].p_vaddr;
+ size_t filesz = phdrp[i].p_filesz;
+
+ *size += filesz;
+ if (*size > limit
+ || !dump_write(file, addr, filesz))
+ return 0;
+ }
+ }
+ }
+ return 1;
+}
+
+size_t elf_core_extra_data_size(void)
+{
+ if ( vsyscall_ehdr ) {
+ const struct elfhdr *const ehdrp =
+ (struct elfhdr *)vsyscall_ehdr;
+ const struct elf_phdr *const phdrp =
+ (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff);
+ int i;
+
+ for (i = 0; i < ehdrp->e_phnum; ++i)
+ if (phdrp[i].p_type == PT_LOAD)
+ return (size_t) phdrp[i].p_filesz;
+ }
+ return 0;
+}
diff --git a/arch/x86/um/fault.c b/arch/x86/um/fault.c
new file mode 100644
index 000000000000..d670f68532f4
--- /dev/null
+++ b/arch/x86/um/fault.c
@@ -0,0 +1,28 @@
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include "sysdep/ptrace.h"
+
+/* These two are from asm-um/uaccess.h and linux/module.h, check them. */
+struct exception_table_entry
+{
+ unsigned long insn;
+ unsigned long fixup;
+};
+
+const struct exception_table_entry *search_exception_tables(unsigned long add);
+
+/* Compare this to arch/i386/mm/extable.c:fixup_exception() */
+int arch_fixup(unsigned long address, struct uml_pt_regs *regs)
+{
+ const struct exception_table_entry *fixup;
+
+ fixup = search_exception_tables(address);
+ if (fixup != 0) {
+ UPT_IP(regs) = fixup->fixup;
+ return 1;
+ }
+ return 0;
+}
diff --git a/arch/x86/um/ksyms.c b/arch/x86/um/ksyms.c
new file mode 100644
index 000000000000..2e8f43ec6214
--- /dev/null
+++ b/arch/x86/um/ksyms.c
@@ -0,0 +1,13 @@
+#include <linux/module.h>
+#include <asm/string.h>
+#include <asm/checksum.h>
+
+#ifndef CONFIG_X86_32
+/*XXX: we need them because they would be exported by x86_64 */
+#if (__GNUC__ == 4 && __GNUC_MINOR__ >= 3) || __GNUC__ > 4
+EXPORT_SYMBOL(memcpy);
+#else
+EXPORT_SYMBOL(__memcpy);
+#endif
+#endif
+EXPORT_SYMBOL(csum_partial);
diff --git a/arch/x86/um/ldt.c b/arch/x86/um/ldt.c
new file mode 100644
index 000000000000..3f2bf208d884
--- /dev/null
+++ b/arch/x86/um/ldt.c
@@ -0,0 +1,502 @@
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <asm/unistd.h>
+#include "os.h"
+#include "proc_mm.h"
+#include "skas.h"
+#include "skas_ptrace.h"
+#include "sysdep/tls.h"
+
+extern int modify_ldt(int func, void *ptr, unsigned long bytecount);
+
+static long write_ldt_entry(struct mm_id *mm_idp, int func,
+ struct user_desc *desc, void **addr, int done)
+{
+ long res;
+
+ if (proc_mm) {
+ /*
+ * This is a special handling for the case, that the mm to
+ * modify isn't current->active_mm.
+ * If this is called directly by modify_ldt,
+ * (current->active_mm->context.skas.u == mm_idp)
+ * will be true. So no call to __switch_mm(mm_idp) is done.
+ * If this is called in case of init_new_ldt or PTRACE_LDT,
+ * mm_idp won't belong to current->active_mm, but child->mm.
+ * So we need to switch child's mm into our userspace, then
+ * later switch back.
+ *
+ * Note: I'm unsure: should interrupts be disabled here?
+ */
+ if (!current->active_mm || current->active_mm == &init_mm ||
+ mm_idp != &current->active_mm->context.id)
+ __switch_mm(mm_idp);
+ }
+
+ if (ptrace_ldt) {
+ struct ptrace_ldt ldt_op = (struct ptrace_ldt) {
+ .func = func,
+ .ptr = desc,
+ .bytecount = sizeof(*desc)};
+ u32 cpu;
+ int pid;
+
+ if (!proc_mm)
+ pid = mm_idp->u.pid;
+ else {
+ cpu = get_cpu();
+ pid = userspace_pid[cpu];
+ }
+
+ res = os_ptrace_ldt(pid, 0, (unsigned long) &ldt_op);
+
+ if (proc_mm)
+ put_cpu();
+ }
+ else {
+ void *stub_addr;
+ res = syscall_stub_data(mm_idp, (unsigned long *)desc,
+ (sizeof(*desc) + sizeof(long) - 1) &
+ ~(sizeof(long) - 1),
+ addr, &stub_addr);
+ if (!res) {
+ unsigned long args[] = { func,
+ (unsigned long)stub_addr,
+ sizeof(*desc),
+ 0, 0, 0 };
+ res = run_syscall_stub(mm_idp, __NR_modify_ldt, args,
+ 0, addr, done);
+ }
+ }
+
+ if (proc_mm) {
+ /*
+ * This is the second part of special handling, that makes
+ * PTRACE_LDT possible to implement.
+ */
+ if (current->active_mm && current->active_mm != &init_mm &&
+ mm_idp != &current->active_mm->context.id)
+ __switch_mm(&current->active_mm->context.id);
+ }
+
+ return res;
+}
+
+static long read_ldt_from_host(void __user * ptr, unsigned long bytecount)
+{
+ int res, n;
+ struct ptrace_ldt ptrace_ldt = (struct ptrace_ldt) {
+ .func = 0,
+ .bytecount = bytecount,
+ .ptr = kmalloc(bytecount, GFP_KERNEL)};
+ u32 cpu;
+
+ if (ptrace_ldt.ptr == NULL)
+ return -ENOMEM;
+
+ /*
+ * This is called from sys_modify_ldt only, so userspace_pid gives
+ * us the right number
+ */
+
+ cpu = get_cpu();
+ res = os_ptrace_ldt(userspace_pid[cpu], 0, (unsigned long) &ptrace_ldt);
+ put_cpu();
+ if (res < 0)
+ goto out;
+
+ n = copy_to_user(ptr, ptrace_ldt.ptr, res);
+ if (n != 0)
+ res = -EFAULT;
+
+ out:
+ kfree(ptrace_ldt.ptr);
+
+ return res;
+}
+
+/*
+ * In skas mode, we hold our own ldt data in UML.
+ * Thus, the code implementing sys_modify_ldt_skas
+ * is very similar to (and mostly stolen from) sys_modify_ldt
+ * for arch/i386/kernel/ldt.c
+ * The routines copied and modified in part are:
+ * - read_ldt
+ * - read_default_ldt
+ * - write_ldt
+ * - sys_modify_ldt_skas
+ */
+
+static int read_ldt(void __user * ptr, unsigned long bytecount)
+{
+ int i, err = 0;
+ unsigned long size;
+ uml_ldt_t * ldt = &current->mm->context.ldt;
+
+ if (!ldt->entry_count)
+ goto out;
+ if (bytecount > LDT_ENTRY_SIZE*LDT_ENTRIES)
+ bytecount = LDT_ENTRY_SIZE*LDT_ENTRIES;
+ err = bytecount;
+
+ if (ptrace_ldt)
+ return read_ldt_from_host(ptr, bytecount);
+
+ mutex_lock(&ldt->lock);
+ if (ldt->entry_count <= LDT_DIRECT_ENTRIES) {
+ size = LDT_ENTRY_SIZE*LDT_DIRECT_ENTRIES;
+ if (size > bytecount)
+ size = bytecount;
+ if (copy_to_user(ptr, ldt->u.entries, size))
+ err = -EFAULT;
+ bytecount -= size;
+ ptr += size;
+ }
+ else {
+ for (i=0; i<ldt->entry_count/LDT_ENTRIES_PER_PAGE && bytecount;
+ i++) {
+ size = PAGE_SIZE;
+ if (size > bytecount)
+ size = bytecount;
+ if (copy_to_user(ptr, ldt->u.pages[i], size)) {
+ err = -EFAULT;
+ break;
+ }
+ bytecount -= size;
+ ptr += size;
+ }
+ }
+ mutex_unlock(&ldt->lock);
+
+ if (bytecount == 0 || err == -EFAULT)
+ goto out;
+
+ if (clear_user(ptr, bytecount))
+ err = -EFAULT;
+
+out:
+ return err;
+}
+
+static int read_default_ldt(void __user * ptr, unsigned long bytecount)
+{
+ int err;
+
+ if (bytecount > 5*LDT_ENTRY_SIZE)
+ bytecount = 5*LDT_ENTRY_SIZE;
+
+ err = bytecount;
+ /*
+ * UML doesn't support lcall7 and lcall27.
+ * So, we don't really have a default ldt, but emulate
+ * an empty ldt of common host default ldt size.
+ */
+ if (clear_user(ptr, bytecount))
+ err = -EFAULT;
+
+ return err;
+}
+
+static int write_ldt(void __user * ptr, unsigned long bytecount, int func)
+{
+ uml_ldt_t * ldt = &current->mm->context.ldt;
+ struct mm_id * mm_idp = &current->mm->context.id;
+ int i, err;
+ struct user_desc ldt_info;
+ struct ldt_entry entry0, *ldt_p;
+ void *addr = NULL;
+
+ err = -EINVAL;
+ if (bytecount != sizeof(ldt_info))
+ goto out;
+ err = -EFAULT;
+ if (copy_from_user(&ldt_info, ptr, sizeof(ldt_info)))
+ goto out;
+
+ err = -EINVAL;
+ if (ldt_info.entry_number >= LDT_ENTRIES)
+ goto out;
+ if (ldt_info.contents == 3) {
+ if (func == 1)
+ goto out;
+ if (ldt_info.seg_not_present == 0)
+ goto out;
+ }
+
+ if (!ptrace_ldt)
+ mutex_lock(&ldt->lock);
+
+ err = write_ldt_entry(mm_idp, func, &ldt_info, &addr, 1);
+ if (err)
+ goto out_unlock;
+ else if (ptrace_ldt) {
+ /* With PTRACE_LDT available, this is used as a flag only */
+ ldt->entry_count = 1;
+ goto out;
+ }
+
+ if (ldt_info.entry_number >= ldt->entry_count &&
+ ldt_info.entry_number >= LDT_DIRECT_ENTRIES) {
+ for (i=ldt->entry_count/LDT_ENTRIES_PER_PAGE;
+ i*LDT_ENTRIES_PER_PAGE <= ldt_info.entry_number;
+ i++) {
+ if (i == 0)
+ memcpy(&entry0, ldt->u.entries,
+ sizeof(entry0));
+ ldt->u.pages[i] = (struct ldt_entry *)
+ __get_free_page(GFP_KERNEL|__GFP_ZERO);
+ if (!ldt->u.pages[i]) {
+ err = -ENOMEM;
+ /* Undo the change in host */
+ memset(&ldt_info, 0, sizeof(ldt_info));
+ write_ldt_entry(mm_idp, 1, &ldt_info, &addr, 1);
+ goto out_unlock;
+ }
+ if (i == 0) {
+ memcpy(ldt->u.pages[0], &entry0,
+ sizeof(entry0));
+ memcpy(ldt->u.pages[0]+1, ldt->u.entries+1,
+ sizeof(entry0)*(LDT_DIRECT_ENTRIES-1));
+ }
+ ldt->entry_count = (i + 1) * LDT_ENTRIES_PER_PAGE;
+ }
+ }
+ if (ldt->entry_count <= ldt_info.entry_number)
+ ldt->entry_count = ldt_info.entry_number + 1;
+
+ if (ldt->entry_count <= LDT_DIRECT_ENTRIES)
+ ldt_p = ldt->u.entries + ldt_info.entry_number;
+ else
+ ldt_p = ldt->u.pages[ldt_info.entry_number/LDT_ENTRIES_PER_PAGE] +
+ ldt_info.entry_number%LDT_ENTRIES_PER_PAGE;
+
+ if (ldt_info.base_addr == 0 && ldt_info.limit == 0 &&
+ (func == 1 || LDT_empty(&ldt_info))) {
+ ldt_p->a = 0;
+ ldt_p->b = 0;
+ }
+ else{
+ if (func == 1)
+ ldt_info.useable = 0;
+ ldt_p->a = LDT_entry_a(&ldt_info);
+ ldt_p->b = LDT_entry_b(&ldt_info);
+ }
+ err = 0;
+
+out_unlock:
+ mutex_unlock(&ldt->lock);
+out:
+ return err;
+}
+
+static long do_modify_ldt_skas(int func, void __user *ptr,
+ unsigned long bytecount)
+{
+ int ret = -ENOSYS;
+
+ switch (func) {
+ case 0:
+ ret = read_ldt(ptr, bytecount);
+ break;
+ case 1:
+ case 0x11:
+ ret = write_ldt(ptr, bytecount, func);
+ break;
+ case 2:
+ ret = read_default_ldt(ptr, bytecount);
+ break;
+ }
+ return ret;
+}
+
+static DEFINE_SPINLOCK(host_ldt_lock);
+static short dummy_list[9] = {0, -1};
+static short * host_ldt_entries = NULL;
+
+static void ldt_get_host_info(void)
+{
+ long ret;
+ struct ldt_entry * ldt;
+ short *tmp;
+ int i, size, k, order;
+
+ spin_lock(&host_ldt_lock);
+
+ if (host_ldt_entries != NULL) {
+ spin_unlock(&host_ldt_lock);
+ return;
+ }
+ host_ldt_entries = dummy_list+1;
+
+ spin_unlock(&host_ldt_lock);
+
+ for (i = LDT_PAGES_MAX-1, order=0; i; i>>=1, order++)
+ ;
+
+ ldt = (struct ldt_entry *)
+ __get_free_pages(GFP_KERNEL|__GFP_ZERO, order);
+ if (ldt == NULL) {
+ printk(KERN_ERR "ldt_get_host_info: couldn't allocate buffer "
+ "for host ldt\n");
+ return;
+ }
+
+ ret = modify_ldt(0, ldt, (1<<order)*PAGE_SIZE);
+ if (ret < 0) {
+ printk(KERN_ERR "ldt_get_host_info: couldn't read host ldt\n");
+ goto out_free;
+ }
+ if (ret == 0) {
+ /* default_ldt is active, simply write an empty entry 0 */
+ host_ldt_entries = dummy_list;
+ goto out_free;
+ }
+
+ for (i=0, size=0; i<ret/LDT_ENTRY_SIZE; i++) {
+ if (ldt[i].a != 0 || ldt[i].b != 0)
+ size++;
+ }
+
+ if (size < ARRAY_SIZE(dummy_list))
+ host_ldt_entries = dummy_list;
+ else {
+ size = (size + 1) * sizeof(dummy_list[0]);
+ tmp = kmalloc(size, GFP_KERNEL);
+ if (tmp == NULL) {
+ printk(KERN_ERR "ldt_get_host_info: couldn't allocate "
+ "host ldt list\n");
+ goto out_free;
+ }
+ host_ldt_entries = tmp;
+ }
+
+ for (i=0, k=0; i<ret/LDT_ENTRY_SIZE; i++) {
+ if (ldt[i].a != 0 || ldt[i].b != 0)
+ host_ldt_entries[k++] = i;
+ }
+ host_ldt_entries[k] = -1;
+
+out_free:
+ free_pages((unsigned long)ldt, order);
+}
+
+long init_new_ldt(struct mm_context *new_mm, struct mm_context *from_mm)
+{
+ struct user_desc desc;
+ short * num_p;
+ int i;
+ long page, err=0;
+ void *addr = NULL;
+ struct proc_mm_op copy;
+
+
+ if (!ptrace_ldt)
+ mutex_init(&new_mm->ldt.lock);
+
+ if (!from_mm) {
+ memset(&desc, 0, sizeof(desc));
+ /*
+ * We have to initialize a clean ldt.
+ */
+ if (proc_mm) {
+ /*
+ * If the new mm was created using proc_mm, host's
+ * default-ldt currently is assigned, which normally
+ * contains the call-gates for lcall7 and lcall27.
+ * To remove these gates, we simply write an empty
+ * entry as number 0 to the host.
+ */
+ err = write_ldt_entry(&new_mm->id, 1, &desc, &addr, 1);
+ }
+ else{
+ /*
+ * Now we try to retrieve info about the ldt, we
+ * inherited from the host. All ldt-entries found
+ * will be reset in the following loop
+ */
+ ldt_get_host_info();
+ for (num_p=host_ldt_entries; *num_p != -1; num_p++) {
+ desc.entry_number = *num_p;
+ err = write_ldt_entry(&new_mm->id, 1, &desc,
+ &addr, *(num_p + 1) == -1);
+ if (err)
+ break;
+ }
+ }
+ new_mm->ldt.entry_count = 0;
+
+ goto out;
+ }
+
+ if (proc_mm) {
+ /*
+ * We have a valid from_mm, so we now have to copy the LDT of
+ * from_mm to new_mm, because using proc_mm an new mm with
+ * an empty/default LDT was created in new_mm()
+ */
+ copy = ((struct proc_mm_op) { .op = MM_COPY_SEGMENTS,
+ .u =
+ { .copy_segments =
+ from_mm->id.u.mm_fd } } );
+ i = os_write_file(new_mm->id.u.mm_fd, &copy, sizeof(copy));
+ if (i != sizeof(copy))
+ printk(KERN_ERR "new_mm : /proc/mm copy_segments "
+ "failed, err = %d\n", -i);
+ }
+
+ if (!ptrace_ldt) {
+ /*
+ * Our local LDT is used to supply the data for
+ * modify_ldt(READLDT), if PTRACE_LDT isn't available,
+ * i.e., we have to use the stub for modify_ldt, which
+ * can't handle the big read buffer of up to 64kB.
+ */
+ mutex_lock(&from_mm->ldt.lock);
+ if (from_mm->ldt.entry_count <= LDT_DIRECT_ENTRIES)
+ memcpy(new_mm->ldt.u.entries, from_mm->ldt.u.entries,
+ sizeof(new_mm->ldt.u.entries));
+ else {
+ i = from_mm->ldt.entry_count / LDT_ENTRIES_PER_PAGE;
+ while (i-->0) {
+ page = __get_free_page(GFP_KERNEL|__GFP_ZERO);
+ if (!page) {
+ err = -ENOMEM;
+ break;
+ }
+ new_mm->ldt.u.pages[i] =
+ (struct ldt_entry *) page;
+ memcpy(new_mm->ldt.u.pages[i],
+ from_mm->ldt.u.pages[i], PAGE_SIZE);
+ }
+ }
+ new_mm->ldt.entry_count = from_mm->ldt.entry_count;
+ mutex_unlock(&from_mm->ldt.lock);
+ }
+
+ out:
+ return err;
+}
+
+
+void free_ldt(struct mm_context *mm)
+{
+ int i;
+
+ if (!ptrace_ldt && mm->ldt.entry_count > LDT_DIRECT_ENTRIES) {
+ i = mm->ldt.entry_count / LDT_ENTRIES_PER_PAGE;
+ while (i-- > 0)
+ free_page((long) mm->ldt.u.pages[i]);
+ }
+ mm->ldt.entry_count = 0;
+}
+
+int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount)
+{
+ return do_modify_ldt_skas(func, ptr, bytecount);
+}
diff --git a/arch/x86/um/mem_32.c b/arch/x86/um/mem_32.c
new file mode 100644
index 000000000000..639900a6fde9
--- /dev/null
+++ b/arch/x86/um/mem_32.c
@@ -0,0 +1,62 @@
+/*
+ * Copyright (C) 2011 Richard Weinberger <richrd@nod.at>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <asm/page.h>
+#include <asm/mman.h>
+
+static struct vm_area_struct gate_vma;
+
+static int __init gate_vma_init(void)
+{
+ if (!FIXADDR_USER_START)
+ return 0;
+
+ gate_vma.vm_mm = NULL;
+ gate_vma.vm_start = FIXADDR_USER_START;
+ gate_vma.vm_end = FIXADDR_USER_END;
+ gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC;
+ gate_vma.vm_page_prot = __P101;
+
+ /*
+ * Make sure the vDSO gets into every core dump.
+ * Dumping its contents makes post-mortem fully interpretable later
+ * without matching up the same kernel and hardware config to see
+ * what PC values meant.
+ */
+ gate_vma.vm_flags |= VM_ALWAYSDUMP;
+
+ return 0;
+}
+__initcall(gate_vma_init);
+
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
+{
+ return FIXADDR_USER_START ? &gate_vma : NULL;
+}
+
+int in_gate_area_no_mm(unsigned long addr)
+{
+ if (!FIXADDR_USER_START)
+ return 0;
+
+ if ((addr >= FIXADDR_USER_START) && (addr < FIXADDR_USER_END))
+ return 1;
+
+ return 0;
+}
+
+int in_gate_area(struct mm_struct *mm, unsigned long addr)
+{
+ struct vm_area_struct *vma = get_gate_vma(mm);
+
+ if (!vma)
+ return 0;
+
+ return (addr >= vma->vm_start) && (addr < vma->vm_end);
+}
diff --git a/arch/x86/um/mem_64.c b/arch/x86/um/mem_64.c
new file mode 100644
index 000000000000..546518727a73
--- /dev/null
+++ b/arch/x86/um/mem_64.c
@@ -0,0 +1,26 @@
+#include "linux/mm.h"
+#include "asm/page.h"
+#include "asm/mman.h"
+
+const char *arch_vma_name(struct vm_area_struct *vma)
+{
+ if (vma->vm_mm && vma->vm_start == um_vdso_addr)
+ return "[vdso]";
+
+ return NULL;
+}
+
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
+{
+ return NULL;
+}
+
+int in_gate_area(struct mm_struct *mm, unsigned long addr)
+{
+ return 0;
+}
+
+int in_gate_area_no_mm(unsigned long addr)
+{
+ return 0;
+}
diff --git a/arch/x86/um/os-Linux/Makefile b/arch/x86/um/os-Linux/Makefile
new file mode 100644
index 000000000000..253bfb8cb702
--- /dev/null
+++ b/arch/x86/um/os-Linux/Makefile
@@ -0,0 +1,13 @@
+#
+# Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+# Licensed under the GPL
+#
+
+obj-y = registers.o task_size.o mcontext.o
+
+obj-$(CONFIG_X86_32) += tls.o
+obj-$(CONFIG_64BIT) += prctl.o
+
+USER_OBJS := $(obj-y)
+
+include arch/um/scripts/Makefile.rules
diff --git a/arch/x86/um/os-Linux/mcontext.c b/arch/x86/um/os-Linux/mcontext.c
new file mode 100644
index 000000000000..1d33d72c6284
--- /dev/null
+++ b/arch/x86/um/os-Linux/mcontext.c
@@ -0,0 +1,31 @@
+#include <sys/ucontext.h>
+#define __FRAME_OFFSETS
+#include <asm/ptrace.h>
+#include <sysdep/ptrace.h>
+
+void get_regs_from_mc(struct uml_pt_regs *regs, mcontext_t *mc)
+{
+#ifdef __i386__
+#define COPY2(X,Y) regs->gp[X] = mc->gregs[REG_##Y]
+#define COPY(X) regs->gp[X] = mc->gregs[REG_##X]
+#define COPY_SEG(X) regs->gp[X] = mc->gregs[REG_##X] & 0xffff;
+#define COPY_SEG_CPL3(X) regs->gp[X] = (mc->gregs[REG_##X] & 0xffff) | 3;
+ COPY_SEG(GS); COPY_SEG(FS); COPY_SEG(ES); COPY_SEG(DS);
+ COPY(EDI); COPY(ESI); COPY(EBP);
+ COPY2(UESP, ESP); /* sic */
+ COPY(EBX); COPY(EDX); COPY(ECX); COPY(EAX);
+ COPY(EIP); COPY_SEG_CPL3(CS); COPY(EFL); COPY_SEG_CPL3(SS);
+#else
+#define COPY2(X,Y) regs->gp[X/sizeof(unsigned long)] = mc->gregs[REG_##Y]
+#define COPY(X) regs->gp[X/sizeof(unsigned long)] = mc->gregs[REG_##X]
+ COPY(R8); COPY(R9); COPY(R10); COPY(R11);
+ COPY(R12); COPY(R13); COPY(R14); COPY(R15);
+ COPY(RDI); COPY(RSI); COPY(RBP); COPY(RBX);
+ COPY(RDX); COPY(RAX); COPY(RCX); COPY(RSP);
+ COPY(RIP);
+ COPY2(EFLAGS, EFL);
+ COPY2(CS, CSGSFS);
+ regs->gp[CS / sizeof(unsigned long)] &= 0xffff;
+ regs->gp[CS / sizeof(unsigned long)] |= 3;
+#endif
+}
diff --git a/arch/x86/um/os-Linux/prctl.c b/arch/x86/um/os-Linux/prctl.c
new file mode 100644
index 000000000000..9d34eddb517f
--- /dev/null
+++ b/arch/x86/um/os-Linux/prctl.c
@@ -0,0 +1,12 @@
+/*
+ * Copyright (C) 2007 Jeff Dike (jdike@{addtoit.com,linux.intel.com})
+ * Licensed under the GPL
+ */
+
+#include <sys/ptrace.h>
+#include <linux/ptrace.h>
+
+int os_arch_prctl(int pid, int code, unsigned long *addr)
+{
+ return ptrace(PTRACE_ARCH_PRCTL, pid, (unsigned long) addr, code);
+}
diff --git a/arch/x86/um/os-Linux/registers.c b/arch/x86/um/os-Linux/registers.c
new file mode 100644
index 000000000000..3a9b6247bbbc
--- /dev/null
+++ b/arch/x86/um/os-Linux/registers.c
@@ -0,0 +1,111 @@
+/*
+ * Copyright (C) 2004 PathScale, Inc
+ * Copyright (C) 2004 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include <errno.h>
+#include <sys/ptrace.h>
+#include <sys/user.h>
+#include "longjmp.h"
+#include "sysdep/ptrace_user.h"
+
+int save_fp_registers(int pid, unsigned long *fp_regs)
+{
+ if (ptrace(PTRACE_GETFPREGS, pid, 0, fp_regs) < 0)
+ return -errno;
+ return 0;
+}
+
+int restore_fp_registers(int pid, unsigned long *fp_regs)
+{
+ if (ptrace(PTRACE_SETFPREGS, pid, 0, fp_regs) < 0)
+ return -errno;
+ return 0;
+}
+
+#ifdef __i386__
+int have_fpx_regs = 1;
+int save_fpx_registers(int pid, unsigned long *fp_regs)
+{
+ if (ptrace(PTRACE_GETFPXREGS, pid, 0, fp_regs) < 0)
+ return -errno;
+ return 0;
+}
+
+int restore_fpx_registers(int pid, unsigned long *fp_regs)
+{
+ if (ptrace(PTRACE_SETFPXREGS, pid, 0, fp_regs) < 0)
+ return -errno;
+ return 0;
+}
+
+int get_fp_registers(int pid, unsigned long *regs)
+{
+ if (have_fpx_regs)
+ return save_fpx_registers(pid, regs);
+ else
+ return save_fp_registers(pid, regs);
+}
+
+int put_fp_registers(int pid, unsigned long *regs)
+{
+ if (have_fpx_regs)
+ return restore_fpx_registers(pid, regs);
+ else
+ return restore_fp_registers(pid, regs);
+}
+
+void arch_init_registers(int pid)
+{
+ struct user_fpxregs_struct fpx_regs;
+ int err;
+
+ err = ptrace(PTRACE_GETFPXREGS, pid, 0, &fpx_regs);
+ if (!err)
+ return;
+
+ if (errno != EIO)
+ panic("check_ptrace : PTRACE_GETFPXREGS failed, errno = %d",
+ errno);
+
+ have_fpx_regs = 0;
+}
+#else
+
+int get_fp_registers(int pid, unsigned long *regs)
+{
+ return save_fp_registers(pid, regs);
+}
+
+int put_fp_registers(int pid, unsigned long *regs)
+{
+ return restore_fp_registers(pid, regs);
+}
+
+#endif
+
+unsigned long get_thread_reg(int reg, jmp_buf *buf)
+{
+ switch (reg) {
+#ifdef __i386__
+ case EIP:
+ return buf[0]->__eip;
+ case UESP:
+ return buf[0]->__esp;
+ case EBP:
+ return buf[0]->__ebp;
+#else
+ case RIP:
+ return buf[0]->__rip;
+ case RSP:
+ return buf[0]->__rsp;
+ case RBP:
+ return buf[0]->__rbp;
+#endif
+ default:
+ printk(UM_KERN_ERR "get_thread_regs - unknown register %d\n",
+ reg);
+ return 0;
+ }
+}
diff --git a/arch/x86/um/os-Linux/task_size.c b/arch/x86/um/os-Linux/task_size.c
new file mode 100644
index 000000000000..efb16c5c9bcf
--- /dev/null
+++ b/arch/x86/um/os-Linux/task_size.c
@@ -0,0 +1,150 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <sys/mman.h>
+#include "longjmp.h"
+
+#ifdef __i386__
+
+static jmp_buf buf;
+
+static void segfault(int sig)
+{
+ longjmp(buf, 1);
+}
+
+static int page_ok(unsigned long page)
+{
+ unsigned long *address = (unsigned long *) (page << UM_KERN_PAGE_SHIFT);
+ unsigned long n = ~0UL;
+ void *mapped = NULL;
+ int ok = 0;
+
+ /*
+ * First see if the page is readable. If it is, it may still
+ * be a VDSO, so we go on to see if it's writable. If not
+ * then try mapping memory there. If that fails, then we're
+ * still in the kernel area. As a sanity check, we'll fail if
+ * the mmap succeeds, but gives us an address different from
+ * what we wanted.
+ */
+ if (setjmp(buf) == 0)
+ n = *address;
+ else {
+ mapped = mmap(address, UM_KERN_PAGE_SIZE,
+ PROT_READ | PROT_WRITE,
+ MAP_FIXED | MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+ if (mapped == MAP_FAILED)
+ return 0;
+ if (mapped != address)
+ goto out;
+ }
+
+ /*
+ * Now, is it writeable? If so, then we're in user address
+ * space. If not, then try mprotecting it and try the write
+ * again.
+ */
+ if (setjmp(buf) == 0) {
+ *address = n;
+ ok = 1;
+ goto out;
+ } else if (mprotect(address, UM_KERN_PAGE_SIZE,
+ PROT_READ | PROT_WRITE) != 0)
+ goto out;
+
+ if (setjmp(buf) == 0) {
+ *address = n;
+ ok = 1;
+ }
+
+ out:
+ if (mapped != NULL)
+ munmap(mapped, UM_KERN_PAGE_SIZE);
+ return ok;
+}
+
+unsigned long os_get_top_address(void)
+{
+ struct sigaction sa, old;
+ unsigned long bottom = 0;
+ /*
+ * A 32-bit UML on a 64-bit host gets confused about the VDSO at
+ * 0xffffe000. It is mapped, is readable, can be reprotected writeable
+ * and written. However, exec discovers later that it can't be
+ * unmapped. So, just set the highest address to be checked to just
+ * below it. This might waste some address space on 4G/4G 32-bit
+ * hosts, but shouldn't hurt otherwise.
+ */
+ unsigned long top = 0xffffd000 >> UM_KERN_PAGE_SHIFT;
+ unsigned long test, original;
+
+ printf("Locating the bottom of the address space ... ");
+ fflush(stdout);
+
+ /*
+ * We're going to be longjmping out of the signal handler, so
+ * SA_DEFER needs to be set.
+ */
+ sa.sa_handler = segfault;
+ sigemptyset(&sa.sa_mask);
+ sa.sa_flags = SA_NODEFER;
+ if (sigaction(SIGSEGV, &sa, &old)) {
+ perror("os_get_top_address");
+ exit(1);
+ }
+
+ /* Manually scan the address space, bottom-up, until we find
+ * the first valid page (or run out of them).
+ */
+ for (bottom = 0; bottom < top; bottom++) {
+ if (page_ok(bottom))
+ break;
+ }
+
+ /* If we've got this far, we ran out of pages. */
+ if (bottom == top) {
+ fprintf(stderr, "Unable to determine bottom of address "
+ "space.\n");
+ exit(1);
+ }
+
+ printf("0x%x\n", bottom << UM_KERN_PAGE_SHIFT);
+ printf("Locating the top of the address space ... ");
+ fflush(stdout);
+
+ original = bottom;
+
+ /* This could happen with a 4G/4G split */
+ if (page_ok(top))
+ goto out;
+
+ do {
+ test = bottom + (top - bottom) / 2;
+ if (page_ok(test))
+ bottom = test;
+ else
+ top = test;
+ } while (top - bottom > 1);
+
+out:
+ /* Restore the old SIGSEGV handling */
+ if (sigaction(SIGSEGV, &old, NULL)) {
+ perror("os_get_top_address");
+ exit(1);
+ }
+ top <<= UM_KERN_PAGE_SHIFT;
+ printf("0x%x\n", top);
+
+ return top;
+}
+
+#else
+
+unsigned long os_get_top_address(void)
+{
+ /* The old value of CONFIG_TOP_ADDR */
+ return 0x7fc0000000;
+}
+
+#endif
diff --git a/arch/x86/um/os-Linux/tls.c b/arch/x86/um/os-Linux/tls.c
new file mode 100644
index 000000000000..281e83ecce3d
--- /dev/null
+++ b/arch/x86/um/os-Linux/tls.c
@@ -0,0 +1,35 @@
+#include <errno.h>
+#include <linux/unistd.h>
+
+#include <sys/syscall.h>
+#include <unistd.h>
+
+#include "sysdep/tls.h"
+
+/* Checks whether host supports TLS, and sets *tls_min according to the value
+ * valid on the host.
+ * i386 host have it == 6; x86_64 host have it == 12, for i386 emulation. */
+void check_host_supports_tls(int *supports_tls, int *tls_min) {
+ /* Values for x86 and x86_64.*/
+ int val[] = {GDT_ENTRY_TLS_MIN_I386, GDT_ENTRY_TLS_MIN_X86_64};
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(val); i++) {
+ user_desc_t info;
+ info.entry_number = val[i];
+
+ if (syscall(__NR_get_thread_area, &info) == 0) {
+ *tls_min = val[i];
+ *supports_tls = 1;
+ return;
+ } else {
+ if (errno == EINVAL)
+ continue;
+ else if (errno == ENOSYS)
+ *supports_tls = 0;
+ return;
+ }
+ }
+
+ *supports_tls = 0;
+}
diff --git a/arch/x86/um/ptrace_32.c b/arch/x86/um/ptrace_32.c
new file mode 100644
index 000000000000..a174fde2531c
--- /dev/null
+++ b/arch/x86/um/ptrace_32.c
@@ -0,0 +1,273 @@
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/mm.h"
+#include "linux/sched.h"
+#include "asm/uaccess.h"
+#include "skas.h"
+
+extern int arch_switch_tls(struct task_struct *to);
+
+void arch_switch_to(struct task_struct *to)
+{
+ int err = arch_switch_tls(to);
+ if (!err)
+ return;
+
+ if (err != -EINVAL)
+ printk(KERN_WARNING "arch_switch_tls failed, errno %d, "
+ "not EINVAL\n", -err);
+ else
+ printk(KERN_WARNING "arch_switch_tls failed, errno = EINVAL\n");
+}
+
+int is_syscall(unsigned long addr)
+{
+ unsigned short instr;
+ int n;
+
+ n = copy_from_user(&instr, (void __user *) addr, sizeof(instr));
+ if (n) {
+ /* access_process_vm() grants access to vsyscall and stub,
+ * while copy_from_user doesn't. Maybe access_process_vm is
+ * slow, but that doesn't matter, since it will be called only
+ * in case of singlestepping, if copy_from_user failed.
+ */
+ n = access_process_vm(current, addr, &instr, sizeof(instr), 0);
+ if (n != sizeof(instr)) {
+ printk(KERN_ERR "is_syscall : failed to read "
+ "instruction from 0x%lx\n", addr);
+ return 1;
+ }
+ }
+ /* int 0x80 or sysenter */
+ return (instr == 0x80cd) || (instr == 0x340f);
+}
+
+/* determines which flags the user has access to. */
+/* 1 = access 0 = no access */
+#define FLAG_MASK 0x00044dd5
+
+static const int reg_offsets[] = {
+ [EBX] = HOST_EBX,
+ [ECX] = HOST_ECX,
+ [EDX] = HOST_EDX,
+ [ESI] = HOST_ESI,
+ [EDI] = HOST_EDI,
+ [EBP] = HOST_EBP,
+ [EAX] = HOST_EAX,
+ [DS] = HOST_DS,
+ [ES] = HOST_ES,
+ [FS] = HOST_FS,
+ [GS] = HOST_GS,
+ [EIP] = HOST_IP,
+ [CS] = HOST_CS,
+ [EFL] = HOST_EFLAGS,
+ [UESP] = HOST_SP,
+ [SS] = HOST_SS,
+};
+
+int putreg(struct task_struct *child, int regno, unsigned long value)
+{
+ regno >>= 2;
+ switch (regno) {
+ case EBX:
+ case ECX:
+ case EDX:
+ case ESI:
+ case EDI:
+ case EBP:
+ case EAX:
+ case EIP:
+ case UESP:
+ break;
+ case FS:
+ if (value && (value & 3) != 3)
+ return -EIO;
+ break;
+ case GS:
+ if (value && (value & 3) != 3)
+ return -EIO;
+ break;
+ case DS:
+ case ES:
+ if (value && (value & 3) != 3)
+ return -EIO;
+ value &= 0xffff;
+ break;
+ case SS:
+ case CS:
+ if ((value & 3) != 3)
+ return -EIO;
+ value &= 0xffff;
+ break;
+ case EFL:
+ value &= FLAG_MASK;
+ child->thread.regs.regs.gp[HOST_EFLAGS] |= value;
+ return 0;
+ case ORIG_EAX:
+ child->thread.regs.regs.syscall = value;
+ return 0;
+ default :
+ panic("Bad register in putreg() : %d\n", regno);
+ }
+ child->thread.regs.regs.gp[reg_offsets[regno]] = value;
+ return 0;
+}
+
+int poke_user(struct task_struct *child, long addr, long data)
+{
+ if ((addr & 3) || addr < 0)
+ return -EIO;
+
+ if (addr < MAX_REG_OFFSET)
+ return putreg(child, addr, data);
+ else if ((addr >= offsetof(struct user, u_debugreg[0])) &&
+ (addr <= offsetof(struct user, u_debugreg[7]))) {
+ addr -= offsetof(struct user, u_debugreg[0]);
+ addr = addr >> 2;
+ if ((addr == 4) || (addr == 5))
+ return -EIO;
+ child->thread.arch.debugregs[addr] = data;
+ return 0;
+ }
+ return -EIO;
+}
+
+unsigned long getreg(struct task_struct *child, int regno)
+{
+ unsigned long mask = ~0UL;
+
+ regno >>= 2;
+ switch (regno) {
+ case ORIG_EAX:
+ return child->thread.regs.regs.syscall;
+ case FS:
+ case GS:
+ case DS:
+ case ES:
+ case SS:
+ case CS:
+ mask = 0xffff;
+ break;
+ case EIP:
+ case UESP:
+ case EAX:
+ case EBX:
+ case ECX:
+ case EDX:
+ case ESI:
+ case EDI:
+ case EBP:
+ case EFL:
+ break;
+ default:
+ panic("Bad register in getreg() : %d\n", regno);
+ }
+ return mask & child->thread.regs.regs.gp[reg_offsets[regno]];
+}
+
+/* read the word at location addr in the USER area. */
+int peek_user(struct task_struct *child, long addr, long data)
+{
+ unsigned long tmp;
+
+ if ((addr & 3) || addr < 0)
+ return -EIO;
+
+ tmp = 0; /* Default return condition */
+ if (addr < MAX_REG_OFFSET) {
+ tmp = getreg(child, addr);
+ }
+ else if ((addr >= offsetof(struct user, u_debugreg[0])) &&
+ (addr <= offsetof(struct user, u_debugreg[7]))) {
+ addr -= offsetof(struct user, u_debugreg[0]);
+ addr = addr >> 2;
+ tmp = child->thread.arch.debugregs[addr];
+ }
+ return put_user(tmp, (unsigned long __user *) data);
+}
+
+static int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child)
+{
+ int err, n, cpu = ((struct thread_info *) child->stack)->cpu;
+ struct user_i387_struct fpregs;
+
+ err = save_fp_registers(userspace_pid[cpu], (unsigned long *) &fpregs);
+ if (err)
+ return err;
+
+ n = copy_to_user(buf, &fpregs, sizeof(fpregs));
+ if(n > 0)
+ return -EFAULT;
+
+ return n;
+}
+
+static int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child)
+{
+ int n, cpu = ((struct thread_info *) child->stack)->cpu;
+ struct user_i387_struct fpregs;
+
+ n = copy_from_user(&fpregs, buf, sizeof(fpregs));
+ if (n > 0)
+ return -EFAULT;
+
+ return restore_fp_registers(userspace_pid[cpu],
+ (unsigned long *) &fpregs);
+}
+
+static int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child)
+{
+ int err, n, cpu = ((struct thread_info *) child->stack)->cpu;
+ struct user_fxsr_struct fpregs;
+
+ err = save_fpx_registers(userspace_pid[cpu], (unsigned long *) &fpregs);
+ if (err)
+ return err;
+
+ n = copy_to_user(buf, &fpregs, sizeof(fpregs));
+ if(n > 0)
+ return -EFAULT;
+
+ return n;
+}
+
+static int set_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *child)
+{
+ int n, cpu = ((struct thread_info *) child->stack)->cpu;
+ struct user_fxsr_struct fpregs;
+
+ n = copy_from_user(&fpregs, buf, sizeof(fpregs));
+ if (n > 0)
+ return -EFAULT;
+
+ return restore_fpx_registers(userspace_pid[cpu],
+ (unsigned long *) &fpregs);
+}
+
+long subarch_ptrace(struct task_struct *child, long request,
+ unsigned long addr, unsigned long data)
+{
+ int ret = -EIO;
+ void __user *datap = (void __user *) data;
+ switch (request) {
+ case PTRACE_GETFPREGS: /* Get the child FPU state. */
+ ret = get_fpregs(datap, child);
+ break;
+ case PTRACE_SETFPREGS: /* Set the child FPU state. */
+ ret = set_fpregs(datap, child);
+ break;
+ case PTRACE_GETFPXREGS: /* Get the child FPU state. */
+ ret = get_fpxregs(datap, child);
+ break;
+ case PTRACE_SETFPXREGS: /* Set the child FPU state. */
+ ret = set_fpxregs(datap, child);
+ break;
+ default:
+ ret = -EIO;
+ }
+ return ret;
+}
diff --git a/arch/x86/um/ptrace_64.c b/arch/x86/um/ptrace_64.c
new file mode 100644
index 000000000000..44e68e0c0d10
--- /dev/null
+++ b/arch/x86/um/ptrace_64.c
@@ -0,0 +1,271 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ * Copyright (C) 2003 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ *
+ * Licensed under the GPL
+ */
+
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#define __FRAME_OFFSETS
+#include <asm/ptrace.h>
+#include <asm/uaccess.h>
+
+/*
+ * determines which flags the user has access to.
+ * 1 = access 0 = no access
+ */
+#define FLAG_MASK 0x44dd5UL
+
+static const int reg_offsets[] =
+{
+ [R8 >> 3] = HOST_R8,
+ [R9 >> 3] = HOST_R9,
+ [R10 >> 3] = HOST_R10,
+ [R11 >> 3] = HOST_R11,
+ [R12 >> 3] = HOST_R12,
+ [R13 >> 3] = HOST_R13,
+ [R14 >> 3] = HOST_R14,
+ [R15 >> 3] = HOST_R15,
+ [RIP >> 3] = HOST_IP,
+ [RSP >> 3] = HOST_SP,
+ [RAX >> 3] = HOST_RAX,
+ [RBX >> 3] = HOST_RBX,
+ [RCX >> 3] = HOST_RCX,
+ [RDX >> 3] = HOST_RDX,
+ [RSI >> 3] = HOST_RSI,
+ [RDI >> 3] = HOST_RDI,
+ [RBP >> 3] = HOST_RBP,
+ [CS >> 3] = HOST_CS,
+ [SS >> 3] = HOST_SS,
+ [FS_BASE >> 3] = HOST_FS_BASE,
+ [GS_BASE >> 3] = HOST_GS_BASE,
+ [DS >> 3] = HOST_DS,
+ [ES >> 3] = HOST_ES,
+ [FS >> 3] = HOST_FS,
+ [GS >> 3] = HOST_GS,
+ [EFLAGS >> 3] = HOST_EFLAGS,
+ [ORIG_RAX >> 3] = HOST_ORIG_RAX,
+};
+
+int putreg(struct task_struct *child, int regno, unsigned long value)
+{
+#ifdef TIF_IA32
+ /*
+ * Some code in the 64bit emulation may not be 64bit clean.
+ * Don't take any chances.
+ */
+ if (test_tsk_thread_flag(child, TIF_IA32))
+ value &= 0xffffffff;
+#endif
+ switch (regno) {
+ case R8:
+ case R9:
+ case R10:
+ case R11:
+ case R12:
+ case R13:
+ case R14:
+ case R15:
+ case RIP:
+ case RSP:
+ case RAX:
+ case RBX:
+ case RCX:
+ case RDX:
+ case RSI:
+ case RDI:
+ case RBP:
+ case ORIG_RAX:
+ break;
+
+ case FS:
+ case GS:
+ case DS:
+ case ES:
+ case SS:
+ case CS:
+ if (value && (value & 3) != 3)
+ return -EIO;
+ value &= 0xffff;
+ break;
+
+ case FS_BASE:
+ case GS_BASE:
+ if (!((value >> 48) == 0 || (value >> 48) == 0xffff))
+ return -EIO;
+ break;
+
+ case EFLAGS:
+ value &= FLAG_MASK;
+ child->thread.regs.regs.gp[HOST_EFLAGS] |= value;
+ return 0;
+
+ default:
+ panic("Bad register in putreg(): %d\n", regno);
+ }
+
+ child->thread.regs.regs.gp[reg_offsets[regno >> 3]] = value;
+ return 0;
+}
+
+int poke_user(struct task_struct *child, long addr, long data)
+{
+ if ((addr & 3) || addr < 0)
+ return -EIO;
+
+ if (addr < MAX_REG_OFFSET)
+ return putreg(child, addr, data);
+ else if ((addr >= offsetof(struct user, u_debugreg[0])) &&
+ (addr <= offsetof(struct user, u_debugreg[7]))) {
+ addr -= offsetof(struct user, u_debugreg[0]);
+ addr = addr >> 2;
+ if ((addr == 4) || (addr == 5))
+ return -EIO;
+ child->thread.arch.debugregs[addr] = data;
+ return 0;
+ }
+ return -EIO;
+}
+
+unsigned long getreg(struct task_struct *child, int regno)
+{
+ unsigned long mask = ~0UL;
+#ifdef TIF_IA32
+ if (test_tsk_thread_flag(child, TIF_IA32))
+ mask = 0xffffffff;
+#endif
+ switch (regno) {
+ case R8:
+ case R9:
+ case R10:
+ case R11:
+ case R12:
+ case R13:
+ case R14:
+ case R15:
+ case RIP:
+ case RSP:
+ case RAX:
+ case RBX:
+ case RCX:
+ case RDX:
+ case RSI:
+ case RDI:
+ case RBP:
+ case ORIG_RAX:
+ case EFLAGS:
+ case FS_BASE:
+ case GS_BASE:
+ break;
+ case FS:
+ case GS:
+ case DS:
+ case ES:
+ case SS:
+ case CS:
+ mask = 0xffff;
+ break;
+ default:
+ panic("Bad register in getreg: %d\n", regno);
+ }
+ return mask & child->thread.regs.regs.gp[reg_offsets[regno >> 3]];
+}
+
+int peek_user(struct task_struct *child, long addr, long data)
+{
+ /* read the word at location addr in the USER area. */
+ unsigned long tmp;
+
+ if ((addr & 3) || addr < 0)
+ return -EIO;
+
+ tmp = 0; /* Default return condition */
+ if (addr < MAX_REG_OFFSET)
+ tmp = getreg(child, addr);
+ else if ((addr >= offsetof(struct user, u_debugreg[0])) &&
+ (addr <= offsetof(struct user, u_debugreg[7]))) {
+ addr -= offsetof(struct user, u_debugreg[0]);
+ addr = addr >> 2;
+ tmp = child->thread.arch.debugregs[addr];
+ }
+ return put_user(tmp, (unsigned long *) data);
+}
+
+/* XXX Mostly copied from sys-i386 */
+int is_syscall(unsigned long addr)
+{
+ unsigned short instr;
+ int n;
+
+ n = copy_from_user(&instr, (void __user *) addr, sizeof(instr));
+ if (n) {
+ /*
+ * access_process_vm() grants access to vsyscall and stub,
+ * while copy_from_user doesn't. Maybe access_process_vm is
+ * slow, but that doesn't matter, since it will be called only
+ * in case of singlestepping, if copy_from_user failed.
+ */
+ n = access_process_vm(current, addr, &instr, sizeof(instr), 0);
+ if (n != sizeof(instr)) {
+ printk("is_syscall : failed to read instruction from "
+ "0x%lx\n", addr);
+ return 1;
+ }
+ }
+ /* sysenter */
+ return instr == 0x050f;
+}
+
+static int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *child)
+{
+ int err, n, cpu = ((struct thread_info *) child->stack)->cpu;
+ long fpregs[HOST_FP_SIZE];
+
+ BUG_ON(sizeof(*buf) != sizeof(fpregs));
+ err = save_fp_registers(userspace_pid[cpu], fpregs);
+ if (err)
+ return err;
+
+ n = copy_to_user(buf, fpregs, sizeof(fpregs));
+ if (n > 0)
+ return -EFAULT;
+
+ return n;
+}
+
+static int set_fpregs(struct user_i387_struct __user *buf, struct task_struct *child)
+{
+ int n, cpu = ((struct thread_info *) child->stack)->cpu;
+ long fpregs[HOST_FP_SIZE];
+
+ BUG_ON(sizeof(*buf) != sizeof(fpregs));
+ n = copy_from_user(fpregs, buf, sizeof(fpregs));
+ if (n > 0)
+ return -EFAULT;
+
+ return restore_fp_registers(userspace_pid[cpu], fpregs);
+}
+
+long subarch_ptrace(struct task_struct *child, long request,
+ unsigned long addr, unsigned long data)
+{
+ int ret = -EIO;
+ void __user *datap = (void __user *) data;
+
+ switch (request) {
+ case PTRACE_GETFPREGS: /* Get the child FPU state. */
+ ret = get_fpregs(datap, child);
+ break;
+ case PTRACE_SETFPREGS: /* Set the child FPU state. */
+ ret = set_fpregs(datap, child);
+ break;
+ case PTRACE_ARCH_PRCTL:
+ /* XXX Calls ptrace on the host - needs some SMP thinking */
+ ret = arch_prctl(child, data, (void __user *) addr);
+ break;
+ }
+
+ return ret;
+}
diff --git a/arch/x86/um/ptrace_user.c b/arch/x86/um/ptrace_user.c
new file mode 100644
index 000000000000..3960ca1dd35a
--- /dev/null
+++ b/arch/x86/um/ptrace_user.c
@@ -0,0 +1,21 @@
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include <errno.h>
+#include "ptrace_user.h"
+
+int ptrace_getregs(long pid, unsigned long *regs_out)
+{
+ if (ptrace(PTRACE_GETREGS, pid, 0, regs_out) < 0)
+ return -errno;
+ return 0;
+}
+
+int ptrace_setregs(long pid, unsigned long *regs)
+{
+ if (ptrace(PTRACE_SETREGS, pid, 0, regs) < 0)
+ return -errno;
+ return 0;
+}
diff --git a/arch/x86/um/setjmp_32.S b/arch/x86/um/setjmp_32.S
new file mode 100644
index 000000000000..b766792c9933
--- /dev/null
+++ b/arch/x86/um/setjmp_32.S
@@ -0,0 +1,58 @@
+#
+# arch/i386/setjmp.S
+#
+# setjmp/longjmp for the i386 architecture
+#
+
+#
+# The jmp_buf is assumed to contain the following, in order:
+# %ebx
+# %esp
+# %ebp
+# %esi
+# %edi
+# <return address>
+#
+
+ .text
+ .align 4
+ .globl setjmp
+ .type setjmp, @function
+setjmp:
+#ifdef _REGPARM
+ movl %eax,%edx
+#else
+ movl 4(%esp),%edx
+#endif
+ popl %ecx # Return address, and adjust the stack
+ xorl %eax,%eax # Return value
+ movl %ebx,(%edx)
+ movl %esp,4(%edx) # Post-return %esp!
+ pushl %ecx # Make the call/return stack happy
+ movl %ebp,8(%edx)
+ movl %esi,12(%edx)
+ movl %edi,16(%edx)
+ movl %ecx,20(%edx) # Return address
+ ret
+
+ .size setjmp,.-setjmp
+
+ .text
+ .align 4
+ .globl longjmp
+ .type longjmp, @function
+longjmp:
+#ifdef _REGPARM
+ xchgl %eax,%edx
+#else
+ movl 4(%esp),%edx # jmp_ptr address
+ movl 8(%esp),%eax # Return value
+#endif
+ movl (%edx),%ebx
+ movl 4(%edx),%esp
+ movl 8(%edx),%ebp
+ movl 12(%edx),%esi
+ movl 16(%edx),%edi
+ jmp *20(%edx)
+
+ .size longjmp,.-longjmp
diff --git a/arch/x86/um/setjmp_64.S b/arch/x86/um/setjmp_64.S
new file mode 100644
index 000000000000..45f547b4043e
--- /dev/null
+++ b/arch/x86/um/setjmp_64.S
@@ -0,0 +1,54 @@
+#
+# arch/x86_64/setjmp.S
+#
+# setjmp/longjmp for the x86-64 architecture
+#
+
+#
+# The jmp_buf is assumed to contain the following, in order:
+# %rbx
+# %rsp (post-return)
+# %rbp
+# %r12
+# %r13
+# %r14
+# %r15
+# <return address>
+#
+
+ .text
+ .align 4
+ .globl setjmp
+ .type setjmp, @function
+setjmp:
+ pop %rsi # Return address, and adjust the stack
+ xorl %eax,%eax # Return value
+ movq %rbx,(%rdi)
+ movq %rsp,8(%rdi) # Post-return %rsp!
+ push %rsi # Make the call/return stack happy
+ movq %rbp,16(%rdi)
+ movq %r12,24(%rdi)
+ movq %r13,32(%rdi)
+ movq %r14,40(%rdi)
+ movq %r15,48(%rdi)
+ movq %rsi,56(%rdi) # Return address
+ ret
+
+ .size setjmp,.-setjmp
+
+ .text
+ .align 4
+ .globl longjmp
+ .type longjmp, @function
+longjmp:
+ movl %esi,%eax # Return value (int)
+ movq (%rdi),%rbx
+ movq 8(%rdi),%rsp
+ movq 16(%rdi),%rbp
+ movq 24(%rdi),%r12
+ movq 32(%rdi),%r13
+ movq 40(%rdi),%r14
+ movq 48(%rdi),%r15
+ jmp *56(%rdi)
+
+ .size longjmp,.-longjmp
diff --git a/arch/x86/um/shared/sysdep/archsetjmp.h b/arch/x86/um/shared/sysdep/archsetjmp.h
new file mode 100644
index 000000000000..ff7766d28226
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/archsetjmp.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "archsetjmp_32.h"
+#else
+#include "archsetjmp_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/archsetjmp_32.h b/arch/x86/um/shared/sysdep/archsetjmp_32.h
new file mode 100644
index 000000000000..0f312085ce1d
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/archsetjmp_32.h
@@ -0,0 +1,22 @@
+/*
+ * arch/um/include/sysdep-i386/archsetjmp.h
+ */
+
+#ifndef _KLIBC_ARCHSETJMP_H
+#define _KLIBC_ARCHSETJMP_H
+
+struct __jmp_buf {
+ unsigned int __ebx;
+ unsigned int __esp;
+ unsigned int __ebp;
+ unsigned int __esi;
+ unsigned int __edi;
+ unsigned int __eip;
+};
+
+typedef struct __jmp_buf jmp_buf[1];
+
+#define JB_IP __eip
+#define JB_SP __esp
+
+#endif /* _SETJMP_H */
diff --git a/arch/x86/um/shared/sysdep/archsetjmp_64.h b/arch/x86/um/shared/sysdep/archsetjmp_64.h
new file mode 100644
index 000000000000..2af8f12ca161
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/archsetjmp_64.h
@@ -0,0 +1,24 @@
+/*
+ * arch/um/include/sysdep-x86_64/archsetjmp.h
+ */
+
+#ifndef _KLIBC_ARCHSETJMP_H
+#define _KLIBC_ARCHSETJMP_H
+
+struct __jmp_buf {
+ unsigned long __rbx;
+ unsigned long __rsp;
+ unsigned long __rbp;
+ unsigned long __r12;
+ unsigned long __r13;
+ unsigned long __r14;
+ unsigned long __r15;
+ unsigned long __rip;
+};
+
+typedef struct __jmp_buf jmp_buf[1];
+
+#define JB_IP __rip
+#define JB_SP __rsp
+
+#endif /* _SETJMP_H */
diff --git a/arch/x86/um/shared/sysdep/faultinfo.h b/arch/x86/um/shared/sysdep/faultinfo.h
new file mode 100644
index 000000000000..862ecb1c7781
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/faultinfo.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "faultinfo_32.h"
+#else
+#include "faultinfo_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/faultinfo_32.h b/arch/x86/um/shared/sysdep/faultinfo_32.h
new file mode 100644
index 000000000000..a26086b8a800
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/faultinfo_32.h
@@ -0,0 +1,35 @@
+/*
+ * Copyright (C) 2004 Fujitsu Siemens Computers GmbH
+ * Author: Bodo Stroesser <bstroesser@fujitsu-siemens.com>
+ * Licensed under the GPL
+ */
+
+#ifndef __FAULTINFO_I386_H
+#define __FAULTINFO_I386_H
+
+/* this structure contains the full arch-specific faultinfo
+ * from the traps.
+ * On i386, ptrace_faultinfo unfortunately doesn't provide
+ * all the info, since trap_no is missing.
+ * All common elements are defined at the same position in
+ * both structures, thus making it easy to copy the
+ * contents without knowledge about the structure elements.
+ */
+struct faultinfo {
+ int error_code; /* in ptrace_faultinfo misleadingly called is_write */
+ unsigned long cr2; /* in ptrace_faultinfo called addr */
+ int trap_no; /* missing in ptrace_faultinfo */
+};
+
+#define FAULT_WRITE(fi) ((fi).error_code & 2)
+#define FAULT_ADDRESS(fi) ((fi).cr2)
+
+/* This is Page Fault */
+#define SEGV_IS_FIXABLE(fi) ((fi)->trap_no == 14)
+
+/* SKAS3 has no trap_no on i386, but get_skas_faultinfo() sets it to 0. */
+#define SEGV_MAYBE_FIXABLE(fi) ((fi)->trap_no == 0 && ptrace_faultinfo)
+
+#define PTRACE_FULL_FAULTINFO 0
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/faultinfo_64.h b/arch/x86/um/shared/sysdep/faultinfo_64.h
new file mode 100644
index 000000000000..f811cbe15d62
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/faultinfo_64.h
@@ -0,0 +1,35 @@
+/*
+ * Copyright (C) 2004 Fujitsu Siemens Computers GmbH
+ * Author: Bodo Stroesser <bstroesser@fujitsu-siemens.com>
+ * Licensed under the GPL
+ */
+
+#ifndef __FAULTINFO_X86_64_H
+#define __FAULTINFO_X86_64_H
+
+/* this structure contains the full arch-specific faultinfo
+ * from the traps.
+ * On i386, ptrace_faultinfo unfortunately doesn't provide
+ * all the info, since trap_no is missing.
+ * All common elements are defined at the same position in
+ * both structures, thus making it easy to copy the
+ * contents without knowledge about the structure elements.
+ */
+struct faultinfo {
+ int error_code; /* in ptrace_faultinfo misleadingly called is_write */
+ unsigned long cr2; /* in ptrace_faultinfo called addr */
+ int trap_no; /* missing in ptrace_faultinfo */
+};
+
+#define FAULT_WRITE(fi) ((fi).error_code & 2)
+#define FAULT_ADDRESS(fi) ((fi).cr2)
+
+/* This is Page Fault */
+#define SEGV_IS_FIXABLE(fi) ((fi)->trap_no == 14)
+
+/* No broken SKAS API, which doesn't pass trap_no, here. */
+#define SEGV_MAYBE_FIXABLE(fi) 0
+
+#define PTRACE_FULL_FAULTINFO 1
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/host_ldt.h b/arch/x86/um/shared/sysdep/host_ldt.h
new file mode 100644
index 000000000000..94518b3e0da5
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/host_ldt.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "host_ldt_32.h"
+#else
+#include "host_ldt_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/host_ldt_32.h b/arch/x86/um/shared/sysdep/host_ldt_32.h
new file mode 100644
index 000000000000..0953cc4df652
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/host_ldt_32.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_HOST_LDT_I386_H
+#define __ASM_HOST_LDT_I386_H
+
+#include <asm/ldt.h>
+
+/*
+ * macros stolen from include/asm-i386/desc.h
+ */
+#define LDT_entry_a(info) \
+ ((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
+
+#define LDT_entry_b(info) \
+ (((info)->base_addr & 0xff000000) | \
+ (((info)->base_addr & 0x00ff0000) >> 16) | \
+ ((info)->limit & 0xf0000) | \
+ (((info)->read_exec_only ^ 1) << 9) | \
+ ((info)->contents << 10) | \
+ (((info)->seg_not_present ^ 1) << 15) | \
+ ((info)->seg_32bit << 22) | \
+ ((info)->limit_in_pages << 23) | \
+ ((info)->useable << 20) | \
+ 0x7000)
+
+#define LDT_empty(info) (\
+ (info)->base_addr == 0 && \
+ (info)->limit == 0 && \
+ (info)->contents == 0 && \
+ (info)->read_exec_only == 1 && \
+ (info)->seg_32bit == 0 && \
+ (info)->limit_in_pages == 0 && \
+ (info)->seg_not_present == 1 && \
+ (info)->useable == 0 )
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/host_ldt_64.h b/arch/x86/um/shared/sysdep/host_ldt_64.h
new file mode 100644
index 000000000000..e8b1be1e154f
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/host_ldt_64.h
@@ -0,0 +1,38 @@
+#ifndef __ASM_HOST_LDT_X86_64_H
+#define __ASM_HOST_LDT_X86_64_H
+
+#include <asm/ldt.h>
+
+/*
+ * macros stolen from include/asm-x86_64/desc.h
+ */
+#define LDT_entry_a(info) \
+ ((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
+
+/* Don't allow setting of the lm bit. It is useless anyways because
+ * 64bit system calls require __USER_CS. */
+#define LDT_entry_b(info) \
+ (((info)->base_addr & 0xff000000) | \
+ (((info)->base_addr & 0x00ff0000) >> 16) | \
+ ((info)->limit & 0xf0000) | \
+ (((info)->read_exec_only ^ 1) << 9) | \
+ ((info)->contents << 10) | \
+ (((info)->seg_not_present ^ 1) << 15) | \
+ ((info)->seg_32bit << 22) | \
+ ((info)->limit_in_pages << 23) | \
+ ((info)->useable << 20) | \
+ /* ((info)->lm << 21) | */ \
+ 0x7000)
+
+#define LDT_empty(info) (\
+ (info)->base_addr == 0 && \
+ (info)->limit == 0 && \
+ (info)->contents == 0 && \
+ (info)->read_exec_only == 1 && \
+ (info)->seg_32bit == 0 && \
+ (info)->limit_in_pages == 0 && \
+ (info)->seg_not_present == 1 && \
+ (info)->useable == 0 && \
+ (info)->lm == 0)
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/kernel-offsets.h b/arch/x86/um/shared/sysdep/kernel-offsets.h
new file mode 100644
index 000000000000..5868526b5eef
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/kernel-offsets.h
@@ -0,0 +1,21 @@
+#include <linux/stddef.h>
+#include <linux/sched.h>
+#include <linux/elf.h>
+#include <linux/crypto.h>
+#include <asm/mman.h>
+
+#define DEFINE(sym, val) \
+ asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+
+#define STR(x) #x
+#define DEFINE_STR(sym, val) asm volatile("\n->" #sym " " STR(val) " " #val: : )
+
+#define BLANK() asm volatile("\n->" : : )
+
+#define OFFSET(sym, str, mem) \
+ DEFINE(sym, offsetof(struct str, mem));
+
+void foo(void)
+{
+#include <common-offsets.h>
+}
diff --git a/arch/x86/um/shared/sysdep/mcontext.h b/arch/x86/um/shared/sysdep/mcontext.h
new file mode 100644
index 000000000000..b724c54da316
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/mcontext.h
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __SYS_SIGCONTEXT_X86_H
+#define __SYS_SIGCONTEXT_X86_H
+
+extern void get_regs_from_mc(struct uml_pt_regs *, mcontext_t *);
+
+#ifdef __i386__
+
+#define GET_FAULTINFO_FROM_MC(fi, mc) \
+ { \
+ (fi).cr2 = (mc)->cr2; \
+ (fi).error_code = (mc)->gregs[REG_ERR]; \
+ (fi).trap_no = (mc)->gregs[REG_TRAPNO]; \
+ }
+
+#else
+
+#define GET_FAULTINFO_FROM_MC(fi, mc) \
+ { \
+ (fi).cr2 = (mc)->gregs[REG_CR2]; \
+ (fi).error_code = (mc)->gregs[REG_ERR]; \
+ (fi).trap_no = (mc)->gregs[REG_TRAPNO]; \
+ }
+
+#endif
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/ptrace.h b/arch/x86/um/shared/sysdep/ptrace.h
new file mode 100644
index 000000000000..711b1621747f
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/ptrace.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "ptrace_32.h"
+#else
+#include "ptrace_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/ptrace_32.h b/arch/x86/um/shared/sysdep/ptrace_32.h
new file mode 100644
index 000000000000..ce77fa1e2a15
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/ptrace_32.h
@@ -0,0 +1,117 @@
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_I386_PTRACE_H
+#define __SYSDEP_I386_PTRACE_H
+
+#include <generated/user_constants.h>
+#include "sysdep/faultinfo.h"
+
+#define MAX_REG_NR (UM_FRAME_SIZE / sizeof(unsigned long))
+#define MAX_REG_OFFSET (UM_FRAME_SIZE)
+
+static inline void update_debugregs(int seq) {}
+
+/* syscall emulation path in ptrace */
+
+#ifndef PTRACE_SYSEMU
+#define PTRACE_SYSEMU 31
+#endif
+
+void set_using_sysemu(int value);
+int get_using_sysemu(void);
+extern int sysemu_supported;
+
+#define REGS_IP(r) ((r)[HOST_IP])
+#define REGS_SP(r) ((r)[HOST_SP])
+#define REGS_EFLAGS(r) ((r)[HOST_EFLAGS])
+#define REGS_EAX(r) ((r)[HOST_EAX])
+#define REGS_EBX(r) ((r)[HOST_EBX])
+#define REGS_ECX(r) ((r)[HOST_ECX])
+#define REGS_EDX(r) ((r)[HOST_EDX])
+#define REGS_ESI(r) ((r)[HOST_ESI])
+#define REGS_EDI(r) ((r)[HOST_EDI])
+#define REGS_EBP(r) ((r)[HOST_EBP])
+#define REGS_CS(r) ((r)[HOST_CS])
+#define REGS_SS(r) ((r)[HOST_SS])
+#define REGS_DS(r) ((r)[HOST_DS])
+#define REGS_ES(r) ((r)[HOST_ES])
+#define REGS_FS(r) ((r)[HOST_FS])
+#define REGS_GS(r) ((r)[HOST_GS])
+
+#define REGS_SET_SYSCALL_RETURN(r, res) REGS_EAX(r) = (res)
+
+#define IP_RESTART_SYSCALL(ip) ((ip) -= 2)
+#define REGS_RESTART_SYSCALL(r) IP_RESTART_SYSCALL(REGS_IP(r))
+
+#ifndef PTRACE_SYSEMU_SINGLESTEP
+#define PTRACE_SYSEMU_SINGLESTEP 32
+#endif
+
+struct uml_pt_regs {
+ unsigned long gp[MAX_REG_NR];
+ unsigned long fp[HOST_FPX_SIZE];
+ struct faultinfo faultinfo;
+ long syscall;
+ int is_user;
+};
+
+#define EMPTY_UML_PT_REGS { }
+
+#define UPT_IP(r) REGS_IP((r)->gp)
+#define UPT_SP(r) REGS_SP((r)->gp)
+#define UPT_EFLAGS(r) REGS_EFLAGS((r)->gp)
+#define UPT_EAX(r) REGS_EAX((r)->gp)
+#define UPT_EBX(r) REGS_EBX((r)->gp)
+#define UPT_ECX(r) REGS_ECX((r)->gp)
+#define UPT_EDX(r) REGS_EDX((r)->gp)
+#define UPT_ESI(r) REGS_ESI((r)->gp)
+#define UPT_EDI(r) REGS_EDI((r)->gp)
+#define UPT_EBP(r) REGS_EBP((r)->gp)
+#define UPT_ORIG_EAX(r) ((r)->syscall)
+#define UPT_CS(r) REGS_CS((r)->gp)
+#define UPT_SS(r) REGS_SS((r)->gp)
+#define UPT_DS(r) REGS_DS((r)->gp)
+#define UPT_ES(r) REGS_ES((r)->gp)
+#define UPT_FS(r) REGS_FS((r)->gp)
+#define UPT_GS(r) REGS_GS((r)->gp)
+
+#define UPT_SYSCALL_ARG1(r) UPT_EBX(r)
+#define UPT_SYSCALL_ARG2(r) UPT_ECX(r)
+#define UPT_SYSCALL_ARG3(r) UPT_EDX(r)
+#define UPT_SYSCALL_ARG4(r) UPT_ESI(r)
+#define UPT_SYSCALL_ARG5(r) UPT_EDI(r)
+#define UPT_SYSCALL_ARG6(r) UPT_EBP(r)
+
+extern int user_context(unsigned long sp);
+
+#define UPT_IS_USER(r) ((r)->is_user)
+
+struct syscall_args {
+ unsigned long args[6];
+};
+
+#define SYSCALL_ARGS(r) ((struct syscall_args) \
+ { .args = { UPT_SYSCALL_ARG1(r), \
+ UPT_SYSCALL_ARG2(r), \
+ UPT_SYSCALL_ARG3(r), \
+ UPT_SYSCALL_ARG4(r), \
+ UPT_SYSCALL_ARG5(r), \
+ UPT_SYSCALL_ARG6(r) } } )
+
+#define UPT_SET_SYSCALL_RETURN(r, res) \
+ REGS_SET_SYSCALL_RETURN((r)->regs, (res))
+
+#define UPT_RESTART_SYSCALL(r) REGS_RESTART_SYSCALL((r)->gp)
+
+#define UPT_ORIG_SYSCALL(r) UPT_EAX(r)
+#define UPT_SYSCALL_NR(r) UPT_ORIG_EAX(r)
+#define UPT_SYSCALL_RET(r) UPT_EAX(r)
+
+#define UPT_FAULTINFO(r) (&(r)->faultinfo)
+
+extern void arch_init_registers(int pid);
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/ptrace_64.h b/arch/x86/um/shared/sysdep/ptrace_64.h
new file mode 100644
index 000000000000..866fe7e47369
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/ptrace_64.h
@@ -0,0 +1,160 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ * Copyright (C) 2003 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ *
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_X86_64_PTRACE_H
+#define __SYSDEP_X86_64_PTRACE_H
+
+#include <generated/user_constants.h>
+#include "sysdep/faultinfo.h"
+
+#define MAX_REG_OFFSET (UM_FRAME_SIZE)
+#define MAX_REG_NR ((MAX_REG_OFFSET) / sizeof(unsigned long))
+
+#define REGS_IP(r) ((r)[HOST_IP])
+#define REGS_SP(r) ((r)[HOST_SP])
+
+#define REGS_RBX(r) ((r)[HOST_RBX])
+#define REGS_RCX(r) ((r)[HOST_RCX])
+#define REGS_RDX(r) ((r)[HOST_RDX])
+#define REGS_RSI(r) ((r)[HOST_RSI])
+#define REGS_RDI(r) ((r)[HOST_RDI])
+#define REGS_RBP(r) ((r)[HOST_RBP])
+#define REGS_RAX(r) ((r)[HOST_RAX])
+#define REGS_R8(r) ((r)[HOST_R8])
+#define REGS_R9(r) ((r)[HOST_R9])
+#define REGS_R10(r) ((r)[HOST_R10])
+#define REGS_R11(r) ((r)[HOST_R11])
+#define REGS_R12(r) ((r)[HOST_R12])
+#define REGS_R13(r) ((r)[HOST_R13])
+#define REGS_R14(r) ((r)[HOST_R14])
+#define REGS_R15(r) ((r)[HOST_R15])
+#define REGS_CS(r) ((r)[HOST_CS])
+#define REGS_EFLAGS(r) ((r)[HOST_EFLAGS])
+#define REGS_SS(r) ((r)[HOST_SS])
+
+#define HOST_FS_BASE 21
+#define HOST_GS_BASE 22
+#define HOST_DS 23
+#define HOST_ES 24
+#define HOST_FS 25
+#define HOST_GS 26
+
+/* Also defined in asm/ptrace-x86_64.h, but not in libc headers. So, these
+ * are already defined for kernel code, but not for userspace code.
+ */
+#ifndef FS_BASE
+/* These aren't defined in ptrace.h, but exist in struct user_regs_struct,
+ * which is what x86_64 ptrace actually uses.
+ */
+#define FS_BASE (HOST_FS_BASE * sizeof(long))
+#define GS_BASE (HOST_GS_BASE * sizeof(long))
+#define DS (HOST_DS * sizeof(long))
+#define ES (HOST_ES * sizeof(long))
+#define FS (HOST_FS * sizeof(long))
+#define GS (HOST_GS * sizeof(long))
+#endif
+
+#define REGS_FS_BASE(r) ((r)[HOST_FS_BASE])
+#define REGS_GS_BASE(r) ((r)[HOST_GS_BASE])
+#define REGS_DS(r) ((r)[HOST_DS])
+#define REGS_ES(r) ((r)[HOST_ES])
+#define REGS_FS(r) ((r)[HOST_FS])
+#define REGS_GS(r) ((r)[HOST_GS])
+
+#define REGS_ORIG_RAX(r) ((r)[HOST_ORIG_RAX])
+
+#define REGS_SET_SYSCALL_RETURN(r, res) REGS_RAX(r) = (res)
+
+#define IP_RESTART_SYSCALL(ip) ((ip) -= 2)
+#define REGS_RESTART_SYSCALL(r) IP_RESTART_SYSCALL(REGS_IP(r))
+
+#define REGS_FAULT_ADDR(r) ((r)->fault_addr)
+
+#define REGS_FAULT_WRITE(r) FAULT_WRITE((r)->fault_type)
+
+#define REGS_TRAP(r) ((r)->trap_type)
+
+#define REGS_ERR(r) ((r)->fault_type)
+
+struct uml_pt_regs {
+ unsigned long gp[MAX_REG_NR];
+ unsigned long fp[HOST_FP_SIZE];
+ struct faultinfo faultinfo;
+ long syscall;
+ int is_user;
+};
+
+#define EMPTY_UML_PT_REGS { }
+
+#define UPT_RBX(r) REGS_RBX((r)->gp)
+#define UPT_RCX(r) REGS_RCX((r)->gp)
+#define UPT_RDX(r) REGS_RDX((r)->gp)
+#define UPT_RSI(r) REGS_RSI((r)->gp)
+#define UPT_RDI(r) REGS_RDI((r)->gp)
+#define UPT_RBP(r) REGS_RBP((r)->gp)
+#define UPT_RAX(r) REGS_RAX((r)->gp)
+#define UPT_R8(r) REGS_R8((r)->gp)
+#define UPT_R9(r) REGS_R9((r)->gp)
+#define UPT_R10(r) REGS_R10((r)->gp)
+#define UPT_R11(r) REGS_R11((r)->gp)
+#define UPT_R12(r) REGS_R12((r)->gp)
+#define UPT_R13(r) REGS_R13((r)->gp)
+#define UPT_R14(r) REGS_R14((r)->gp)
+#define UPT_R15(r) REGS_R15((r)->gp)
+#define UPT_CS(r) REGS_CS((r)->gp)
+#define UPT_FS_BASE(r) REGS_FS_BASE((r)->gp)
+#define UPT_FS(r) REGS_FS((r)->gp)
+#define UPT_GS_BASE(r) REGS_GS_BASE((r)->gp)
+#define UPT_GS(r) REGS_GS((r)->gp)
+#define UPT_DS(r) REGS_DS((r)->gp)
+#define UPT_ES(r) REGS_ES((r)->gp)
+#define UPT_CS(r) REGS_CS((r)->gp)
+#define UPT_SS(r) REGS_SS((r)->gp)
+#define UPT_ORIG_RAX(r) REGS_ORIG_RAX((r)->gp)
+
+#define UPT_IP(r) REGS_IP((r)->gp)
+#define UPT_SP(r) REGS_SP((r)->gp)
+
+#define UPT_EFLAGS(r) REGS_EFLAGS((r)->gp)
+#define UPT_SYSCALL_NR(r) ((r)->syscall)
+#define UPT_SYSCALL_RET(r) UPT_RAX(r)
+
+extern int user_context(unsigned long sp);
+
+#define UPT_IS_USER(r) ((r)->is_user)
+
+#define UPT_SYSCALL_ARG1(r) UPT_RDI(r)
+#define UPT_SYSCALL_ARG2(r) UPT_RSI(r)
+#define UPT_SYSCALL_ARG3(r) UPT_RDX(r)
+#define UPT_SYSCALL_ARG4(r) UPT_R10(r)
+#define UPT_SYSCALL_ARG5(r) UPT_R8(r)
+#define UPT_SYSCALL_ARG6(r) UPT_R9(r)
+
+struct syscall_args {
+ unsigned long args[6];
+};
+
+#define SYSCALL_ARGS(r) ((struct syscall_args) \
+ { .args = { UPT_SYSCALL_ARG1(r), \
+ UPT_SYSCALL_ARG2(r), \
+ UPT_SYSCALL_ARG3(r), \
+ UPT_SYSCALL_ARG4(r), \
+ UPT_SYSCALL_ARG5(r), \
+ UPT_SYSCALL_ARG6(r) } } )
+
+#define UPT_SET_SYSCALL_RETURN(r, res) \
+ REGS_SET_SYSCALL_RETURN((r)->regs, (res))
+
+#define UPT_RESTART_SYSCALL(r) REGS_RESTART_SYSCALL((r)->gp)
+
+#define UPT_FAULTINFO(r) (&(r)->faultinfo)
+
+static inline void arch_init_registers(int pid)
+{
+}
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/ptrace_user.h b/arch/x86/um/shared/sysdep/ptrace_user.h
new file mode 100644
index 000000000000..a92f883264ed
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/ptrace_user.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "ptrace_user_32.h"
+#else
+#include "ptrace_user_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/ptrace_user_32.h b/arch/x86/um/shared/sysdep/ptrace_user_32.h
new file mode 100644
index 000000000000..9d88a79a138b
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/ptrace_user_32.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_I386_PTRACE_USER_H__
+#define __SYSDEP_I386_PTRACE_USER_H__
+
+#include <sys/ptrace.h>
+#include <linux/ptrace.h>
+#include <asm/ptrace.h>
+#include <generated/user_constants.h>
+
+#define PT_OFFSET(r) ((r) * sizeof(long))
+
+#define PT_SYSCALL_NR(regs) ((regs)[ORIG_EAX])
+#define PT_SYSCALL_NR_OFFSET PT_OFFSET(ORIG_EAX)
+
+#define PT_SYSCALL_RET_OFFSET PT_OFFSET(EAX)
+
+#define REGS_IP_INDEX EIP
+#define REGS_SP_INDEX UESP
+
+#define FP_SIZE ((HOST_FPX_SIZE > HOST_FP_SIZE) ? HOST_FPX_SIZE : HOST_FP_SIZE)
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/ptrace_user_64.h b/arch/x86/um/shared/sysdep/ptrace_user_64.h
new file mode 100644
index 000000000000..2f1b6e33d590
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/ptrace_user_64.h
@@ -0,0 +1,38 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_X86_64_PTRACE_USER_H__
+#define __SYSDEP_X86_64_PTRACE_USER_H__
+
+#define __FRAME_OFFSETS
+#include <sys/ptrace.h>
+#include <linux/ptrace.h>
+#include <asm/ptrace.h>
+#undef __FRAME_OFFSETS
+#include <generated/user_constants.h>
+
+#define PT_INDEX(off) ((off) / sizeof(unsigned long))
+
+#define PT_SYSCALL_NR(regs) ((regs)[PT_INDEX(ORIG_RAX)])
+#define PT_SYSCALL_NR_OFFSET (ORIG_RAX)
+
+#define PT_SYSCALL_RET_OFFSET (RAX)
+
+/*
+ * x86_64 FC3 doesn't define this in /usr/include/linux/ptrace.h even though
+ * it's defined in the kernel's include/linux/ptrace.h. Additionally, use the
+ * 2.4 name and value for 2.4 host compatibility.
+ */
+#ifndef PTRACE_OLDSETOPTIONS
+#define PTRACE_OLDSETOPTIONS 21
+#endif
+
+#define REGS_IP_INDEX PT_INDEX(RIP)
+#define REGS_SP_INDEX PT_INDEX(RSP)
+
+#define FP_SIZE (HOST_FP_SIZE)
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/skas_ptrace.h b/arch/x86/um/shared/sysdep/skas_ptrace.h
new file mode 100644
index 000000000000..453febe98993
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/skas_ptrace.h
@@ -0,0 +1,22 @@
+/*
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_X86_SKAS_PTRACE_H
+#define __SYSDEP_X86_SKAS_PTRACE_H
+
+struct ptrace_faultinfo {
+ int is_write;
+ unsigned long addr;
+};
+
+struct ptrace_ldt {
+ int func;
+ void *ptr;
+ unsigned long bytecount;
+};
+
+#define PTRACE_LDT 54
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/stub.h b/arch/x86/um/shared/sysdep/stub.h
new file mode 100644
index 000000000000..bd161e300102
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/stub.h
@@ -0,0 +1,14 @@
+#include <asm/unistd.h>
+#include <sys/mman.h>
+#include <signal.h>
+#include "as-layout.h"
+#include "stub-data.h"
+
+#ifdef __i386__
+#include "stub_32.h"
+#else
+#include "stub_64.h"
+#endif
+
+extern void stub_segv_handler(int, siginfo_t *, void *);
+extern void stub_clone_handler(void);
diff --git a/arch/x86/um/shared/sysdep/stub_32.h b/arch/x86/um/shared/sysdep/stub_32.h
new file mode 100644
index 000000000000..51fd256c75f0
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/stub_32.h
@@ -0,0 +1,93 @@
+/*
+ * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_STUB_H
+#define __SYSDEP_STUB_H
+
+#include <asm/ptrace.h>
+
+#define STUB_SYSCALL_RET EAX
+#define STUB_MMAP_NR __NR_mmap2
+#define MMAP_OFFSET(o) ((o) >> UM_KERN_PAGE_SHIFT)
+
+static inline long stub_syscall0(long syscall)
+{
+ long ret;
+
+ __asm__ volatile ("int $0x80" : "=a" (ret) : "0" (syscall));
+
+ return ret;
+}
+
+static inline long stub_syscall1(long syscall, long arg1)
+{
+ long ret;
+
+ __asm__ volatile ("int $0x80" : "=a" (ret) : "0" (syscall), "b" (arg1));
+
+ return ret;
+}
+
+static inline long stub_syscall2(long syscall, long arg1, long arg2)
+{
+ long ret;
+
+ __asm__ volatile ("int $0x80" : "=a" (ret) : "0" (syscall), "b" (arg1),
+ "c" (arg2));
+
+ return ret;
+}
+
+static inline long stub_syscall3(long syscall, long arg1, long arg2, long arg3)
+{
+ long ret;
+
+ __asm__ volatile ("int $0x80" : "=a" (ret) : "0" (syscall), "b" (arg1),
+ "c" (arg2), "d" (arg3));
+
+ return ret;
+}
+
+static inline long stub_syscall4(long syscall, long arg1, long arg2, long arg3,
+ long arg4)
+{
+ long ret;
+
+ __asm__ volatile ("int $0x80" : "=a" (ret) : "0" (syscall), "b" (arg1),
+ "c" (arg2), "d" (arg3), "S" (arg4));
+
+ return ret;
+}
+
+static inline long stub_syscall5(long syscall, long arg1, long arg2, long arg3,
+ long arg4, long arg5)
+{
+ long ret;
+
+ __asm__ volatile ("int $0x80" : "=a" (ret) : "0" (syscall), "b" (arg1),
+ "c" (arg2), "d" (arg3), "S" (arg4), "D" (arg5));
+
+ return ret;
+}
+
+static inline void trap_myself(void)
+{
+ __asm("int3");
+}
+
+static inline void remap_stack(int fd, unsigned long offset)
+{
+ __asm__ volatile ("movl %%eax,%%ebp ; movl %0,%%eax ; int $0x80 ;"
+ "movl %7, %%ebx ; movl %%eax, (%%ebx)"
+ : : "g" (STUB_MMAP_NR), "b" (STUB_DATA),
+ "c" (UM_KERN_PAGE_SIZE),
+ "d" (PROT_READ | PROT_WRITE),
+ "S" (MAP_FIXED | MAP_SHARED), "D" (fd),
+ "a" (offset),
+ "i" (&((struct stub_data *) STUB_DATA)->err)
+ : "memory");
+}
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/stub_64.h b/arch/x86/um/shared/sysdep/stub_64.h
new file mode 100644
index 000000000000..994df93c5ed3
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/stub_64.h
@@ -0,0 +1,99 @@
+/*
+ * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_STUB_H
+#define __SYSDEP_STUB_H
+
+#include <sysdep/ptrace_user.h>
+
+#define STUB_SYSCALL_RET PT_INDEX(RAX)
+#define STUB_MMAP_NR __NR_mmap
+#define MMAP_OFFSET(o) (o)
+
+#define __syscall_clobber "r11","rcx","memory"
+#define __syscall "syscall"
+
+static inline long stub_syscall0(long syscall)
+{
+ long ret;
+
+ __asm__ volatile (__syscall
+ : "=a" (ret)
+ : "0" (syscall) : __syscall_clobber );
+
+ return ret;
+}
+
+static inline long stub_syscall2(long syscall, long arg1, long arg2)
+{
+ long ret;
+
+ __asm__ volatile (__syscall
+ : "=a" (ret)
+ : "0" (syscall), "D" (arg1), "S" (arg2) : __syscall_clobber );
+
+ return ret;
+}
+
+static inline long stub_syscall3(long syscall, long arg1, long arg2, long arg3)
+{
+ long ret;
+
+ __asm__ volatile (__syscall
+ : "=a" (ret)
+ : "0" (syscall), "D" (arg1), "S" (arg2), "d" (arg3)
+ : __syscall_clobber );
+
+ return ret;
+}
+
+static inline long stub_syscall4(long syscall, long arg1, long arg2, long arg3,
+ long arg4)
+{
+ long ret;
+
+ __asm__ volatile ("movq %5,%%r10 ; " __syscall
+ : "=a" (ret)
+ : "0" (syscall), "D" (arg1), "S" (arg2), "d" (arg3),
+ "g" (arg4)
+ : __syscall_clobber, "r10" );
+
+ return ret;
+}
+
+static inline long stub_syscall5(long syscall, long arg1, long arg2, long arg3,
+ long arg4, long arg5)
+{
+ long ret;
+
+ __asm__ volatile ("movq %5,%%r10 ; movq %6,%%r8 ; " __syscall
+ : "=a" (ret)
+ : "0" (syscall), "D" (arg1), "S" (arg2), "d" (arg3),
+ "g" (arg4), "g" (arg5)
+ : __syscall_clobber, "r10", "r8" );
+
+ return ret;
+}
+
+static inline void trap_myself(void)
+{
+ __asm("int3");
+}
+
+static inline void remap_stack(long fd, unsigned long offset)
+{
+ __asm__ volatile ("movq %4,%%r10 ; movq %5,%%r8 ; "
+ "movq %6, %%r9; " __syscall "; movq %7, %%rbx ; "
+ "movq %%rax, (%%rbx)":
+ : "a" (STUB_MMAP_NR), "D" (STUB_DATA),
+ "S" (UM_KERN_PAGE_SIZE),
+ "d" (PROT_READ | PROT_WRITE),
+ "g" (MAP_FIXED | MAP_SHARED), "g" (fd),
+ "g" (offset),
+ "i" (&((struct stub_data *) STUB_DATA)->err)
+ : __syscall_clobber, "r10", "r8", "r9" );
+}
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/syscalls.h b/arch/x86/um/shared/sysdep/syscalls.h
new file mode 100644
index 000000000000..bd9a89b67e41
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/syscalls.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "syscalls_32.h"
+#else
+#include "syscalls_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/syscalls_32.h b/arch/x86/um/shared/sysdep/syscalls_32.h
new file mode 100644
index 000000000000..05cb796aecb5
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/syscalls_32.h
@@ -0,0 +1,20 @@
+/*
+ * Copyright (C) 2000 - 2008 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include "asm/unistd.h"
+#include "sysdep/ptrace.h"
+
+typedef long syscall_handler_t(struct pt_regs);
+
+/* Not declared on x86, incompatible declarations on x86_64, so these have
+ * to go here rather than in sys_call_table.c
+ */
+extern syscall_handler_t sys_rt_sigaction;
+
+extern syscall_handler_t *sys_call_table[];
+
+#define EXECUTE_SYSCALL(syscall, regs) \
+ ((long (*)(struct syscall_args)) \
+ (*sys_call_table[syscall]))(SYSCALL_ARGS(&regs->regs))
diff --git a/arch/x86/um/shared/sysdep/syscalls_64.h b/arch/x86/um/shared/sysdep/syscalls_64.h
new file mode 100644
index 000000000000..8a7d5e1da98e
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/syscalls_64.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#ifndef __SYSDEP_X86_64_SYSCALLS_H__
+#define __SYSDEP_X86_64_SYSCALLS_H__
+
+#include <linux/msg.h>
+#include <linux/shm.h>
+
+typedef long syscall_handler_t(void);
+
+extern syscall_handler_t *sys_call_table[];
+
+#define EXECUTE_SYSCALL(syscall, regs) \
+ (((long (*)(long, long, long, long, long, long)) \
+ (*sys_call_table[syscall]))(UPT_SYSCALL_ARG1(&regs->regs), \
+ UPT_SYSCALL_ARG2(&regs->regs), \
+ UPT_SYSCALL_ARG3(&regs->regs), \
+ UPT_SYSCALL_ARG4(&regs->regs), \
+ UPT_SYSCALL_ARG5(&regs->regs), \
+ UPT_SYSCALL_ARG6(&regs->regs)))
+
+extern long old_mmap(unsigned long addr, unsigned long len,
+ unsigned long prot, unsigned long flags,
+ unsigned long fd, unsigned long pgoff);
+extern syscall_handler_t sys_modify_ldt;
+extern syscall_handler_t sys_arch_prctl;
+
+#endif
diff --git a/arch/x86/um/shared/sysdep/tls.h b/arch/x86/um/shared/sysdep/tls.h
new file mode 100644
index 000000000000..4d8f75262370
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/tls.h
@@ -0,0 +1,5 @@
+#ifdef __i386__
+#include "tls_32.h"
+#else
+#include "tls_64.h"
+#endif
diff --git a/arch/x86/um/shared/sysdep/tls_32.h b/arch/x86/um/shared/sysdep/tls_32.h
new file mode 100644
index 000000000000..34550755b2a1
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/tls_32.h
@@ -0,0 +1,32 @@
+#ifndef _SYSDEP_TLS_H
+#define _SYSDEP_TLS_H
+
+# ifndef __KERNEL__
+
+/* Change name to avoid conflicts with the original one from <asm/ldt.h>, which
+ * may be named user_desc (but in 2.4 and in header matching its API was named
+ * modify_ldt_ldt_s). */
+
+typedef struct um_dup_user_desc {
+ unsigned int entry_number;
+ unsigned int base_addr;
+ unsigned int limit;
+ unsigned int seg_32bit:1;
+ unsigned int contents:2;
+ unsigned int read_exec_only:1;
+ unsigned int limit_in_pages:1;
+ unsigned int seg_not_present:1;
+ unsigned int useable:1;
+} user_desc_t;
+
+# else /* __KERNEL__ */
+
+# include <ldt.h>
+typedef struct user_desc user_desc_t;
+
+# endif /* __KERNEL__ */
+
+#define GDT_ENTRY_TLS_MIN_I386 6
+#define GDT_ENTRY_TLS_MIN_X86_64 12
+
+#endif /* _SYSDEP_TLS_H */
diff --git a/arch/x86/um/shared/sysdep/tls_64.h b/arch/x86/um/shared/sysdep/tls_64.h
new file mode 100644
index 000000000000..18c000d0357a
--- /dev/null
+++ b/arch/x86/um/shared/sysdep/tls_64.h
@@ -0,0 +1,29 @@
+#ifndef _SYSDEP_TLS_H
+#define _SYSDEP_TLS_H
+
+# ifndef __KERNEL__
+
+/* Change name to avoid conflicts with the original one from <asm/ldt.h>, which
+ * may be named user_desc (but in 2.4 and in header matching its API was named
+ * modify_ldt_ldt_s). */
+
+typedef struct um_dup_user_desc {
+ unsigned int entry_number;
+ unsigned int base_addr;
+ unsigned int limit;
+ unsigned int seg_32bit:1;
+ unsigned int contents:2;
+ unsigned int read_exec_only:1;
+ unsigned int limit_in_pages:1;
+ unsigned int seg_not_present:1;
+ unsigned int useable:1;
+ unsigned int lm:1;
+} user_desc_t;
+
+# else /* __KERNEL__ */
+
+# include <ldt.h>
+typedef struct user_desc user_desc_t;
+
+# endif /* __KERNEL__ */
+#endif /* _SYSDEP_TLS_H */
diff --git a/arch/x86/um/signal_32.c b/arch/x86/um/signal_32.c
new file mode 100644
index 000000000000..bcbfb0d64813
--- /dev/null
+++ b/arch/x86/um/signal_32.c
@@ -0,0 +1,498 @@
+/*
+ * Copyright (C) 2004 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include <linux/ptrace.h>
+#include <asm/unistd.h>
+#include <asm/uaccess.h>
+#include <asm/ucontext.h>
+#include "frame_kern.h"
+#include "skas.h"
+
+/*
+ * FPU tag word conversions.
+ */
+
+static inline unsigned short twd_i387_to_fxsr(unsigned short twd)
+{
+ unsigned int tmp; /* to avoid 16 bit prefixes in the code */
+
+ /* Transform each pair of bits into 01 (valid) or 00 (empty) */
+ tmp = ~twd;
+ tmp = (tmp | (tmp>>1)) & 0x5555; /* 0V0V0V0V0V0V0V0V */
+ /* and move the valid bits to the lower byte. */
+ tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */
+ tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */
+ tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */
+ return tmp;
+}
+
+static inline unsigned long twd_fxsr_to_i387(struct user_fxsr_struct *fxsave)
+{
+ struct _fpxreg *st = NULL;
+ unsigned long twd = (unsigned long) fxsave->twd;
+ unsigned long tag;
+ unsigned long ret = 0xffff0000;
+ int i;
+
+#define FPREG_ADDR(f, n) ((char *)&(f)->st_space + (n) * 16)
+
+ for (i = 0; i < 8; i++) {
+ if (twd & 0x1) {
+ st = (struct _fpxreg *) FPREG_ADDR(fxsave, i);
+
+ switch (st->exponent & 0x7fff) {
+ case 0x7fff:
+ tag = 2; /* Special */
+ break;
+ case 0x0000:
+ if ( !st->significand[0] &&
+ !st->significand[1] &&
+ !st->significand[2] &&
+ !st->significand[3] ) {
+ tag = 1; /* Zero */
+ } else {
+ tag = 2; /* Special */
+ }
+ break;
+ default:
+ if (st->significand[3] & 0x8000) {
+ tag = 0; /* Valid */
+ } else {
+ tag = 2; /* Special */
+ }
+ break;
+ }
+ } else {
+ tag = 3; /* Empty */
+ }
+ ret |= (tag << (2 * i));
+ twd = twd >> 1;
+ }
+ return ret;
+}
+
+static int convert_fxsr_to_user(struct _fpstate __user *buf,
+ struct user_fxsr_struct *fxsave)
+{
+ unsigned long env[7];
+ struct _fpreg __user *to;
+ struct _fpxreg *from;
+ int i;
+
+ env[0] = (unsigned long)fxsave->cwd | 0xffff0000ul;
+ env[1] = (unsigned long)fxsave->swd | 0xffff0000ul;
+ env[2] = twd_fxsr_to_i387(fxsave);
+ env[3] = fxsave->fip;
+ env[4] = fxsave->fcs | ((unsigned long)fxsave->fop << 16);
+ env[5] = fxsave->foo;
+ env[6] = fxsave->fos;
+
+ if (__copy_to_user(buf, env, 7 * sizeof(unsigned long)))
+ return 1;
+
+ to = &buf->_st[0];
+ from = (struct _fpxreg *) &fxsave->st_space[0];
+ for (i = 0; i < 8; i++, to++, from++) {
+ unsigned long __user *t = (unsigned long __user *)to;
+ unsigned long *f = (unsigned long *)from;
+
+ if (__put_user(*f, t) ||
+ __put_user(*(f + 1), t + 1) ||
+ __put_user(from->exponent, &to->exponent))
+ return 1;
+ }
+ return 0;
+}
+
+static int convert_fxsr_from_user(struct user_fxsr_struct *fxsave,
+ struct _fpstate __user *buf)
+{
+ unsigned long env[7];
+ struct _fpxreg *to;
+ struct _fpreg __user *from;
+ int i;
+
+ if (copy_from_user( env, buf, 7 * sizeof(long)))
+ return 1;
+
+ fxsave->cwd = (unsigned short)(env[0] & 0xffff);
+ fxsave->swd = (unsigned short)(env[1] & 0xffff);
+ fxsave->twd = twd_i387_to_fxsr((unsigned short)(env[2] & 0xffff));
+ fxsave->fip = env[3];
+ fxsave->fop = (unsigned short)((env[4] & 0xffff0000ul) >> 16);
+ fxsave->fcs = (env[4] & 0xffff);
+ fxsave->foo = env[5];
+ fxsave->fos = env[6];
+
+ to = (struct _fpxreg *) &fxsave->st_space[0];
+ from = &buf->_st[0];
+ for (i = 0; i < 8; i++, to++, from++) {
+ unsigned long *t = (unsigned long *)to;
+ unsigned long __user *f = (unsigned long __user *)from;
+
+ if (__get_user(*t, f) ||
+ __get_user(*(t + 1), f + 1) ||
+ __get_user(to->exponent, &from->exponent))
+ return 1;
+ }
+ return 0;
+}
+
+extern int have_fpx_regs;
+
+static int copy_sc_from_user(struct pt_regs *regs,
+ struct sigcontext __user *from)
+{
+ struct sigcontext sc;
+ int err, pid;
+
+ err = copy_from_user(&sc, from, sizeof(sc));
+ if (err)
+ return err;
+
+ pid = userspace_pid[current_thread_info()->cpu];
+
+#define GETREG(regno, regname) regs->regs.gp[HOST_##regno] = sc.regname
+
+ GETREG(GS, gs);
+ GETREG(FS, fs);
+ GETREG(ES, es);
+ GETREG(DS, ds);
+ GETREG(EDI, di);
+ GETREG(ESI, si);
+ GETREG(EBP, bp);
+ GETREG(SP, sp);
+ GETREG(EBX, bx);
+ GETREG(EDX, dx);
+ GETREG(ECX, cx);
+ GETREG(EAX, ax);
+ GETREG(IP, ip);
+ GETREG(CS, cs);
+ GETREG(EFLAGS, flags);
+ GETREG(SS, ss);
+
+#undef GETREG
+ if (have_fpx_regs) {
+ struct user_fxsr_struct fpx;
+
+ err = copy_from_user(&fpx,
+ &((struct _fpstate __user *)sc.fpstate)->_fxsr_env[0],
+ sizeof(struct user_fxsr_struct));
+ if (err)
+ return 1;
+
+ err = convert_fxsr_from_user(&fpx, sc.fpstate);
+ if (err)
+ return 1;
+
+ err = restore_fpx_registers(pid, (unsigned long *) &fpx);
+ if (err < 0) {
+ printk(KERN_ERR "copy_sc_from_user - "
+ "restore_fpx_registers failed, errno = %d\n",
+ -err);
+ return 1;
+ }
+ } else {
+ struct user_i387_struct fp;
+
+ err = copy_from_user(&fp, sc.fpstate,
+ sizeof(struct user_i387_struct));
+ if (err)
+ return 1;
+
+ err = restore_fp_registers(pid, (unsigned long *) &fp);
+ if (err < 0) {
+ printk(KERN_ERR "copy_sc_from_user - "
+ "restore_fp_registers failed, errno = %d\n",
+ -err);
+ return 1;
+ }
+ }
+
+ return 0;
+}
+
+static int copy_sc_to_user(struct sigcontext __user *to,
+ struct _fpstate __user *to_fp, struct pt_regs *regs,
+ unsigned long sp)
+{
+ struct sigcontext sc;
+ struct faultinfo * fi = &current->thread.arch.faultinfo;
+ int err, pid;
+ memset(&sc, 0, sizeof(struct sigcontext));
+
+ sc.gs = REGS_GS(regs->regs.gp);
+ sc.fs = REGS_FS(regs->regs.gp);
+ sc.es = REGS_ES(regs->regs.gp);
+ sc.ds = REGS_DS(regs->regs.gp);
+ sc.di = REGS_EDI(regs->regs.gp);
+ sc.si = REGS_ESI(regs->regs.gp);
+ sc.bp = REGS_EBP(regs->regs.gp);
+ sc.sp = sp;
+ sc.bx = REGS_EBX(regs->regs.gp);
+ sc.dx = REGS_EDX(regs->regs.gp);
+ sc.cx = REGS_ECX(regs->regs.gp);
+ sc.ax = REGS_EAX(regs->regs.gp);
+ sc.ip = REGS_IP(regs->regs.gp);
+ sc.cs = REGS_CS(regs->regs.gp);
+ sc.flags = REGS_EFLAGS(regs->regs.gp);
+ sc.sp_at_signal = regs->regs.gp[UESP];
+ sc.ss = regs->regs.gp[SS];
+ sc.cr2 = fi->cr2;
+ sc.err = fi->error_code;
+ sc.trapno = fi->trap_no;
+
+ to_fp = (to_fp ? to_fp : (struct _fpstate __user *) (to + 1));
+ sc.fpstate = to_fp;
+
+ pid = userspace_pid[current_thread_info()->cpu];
+ if (have_fpx_regs) {
+ struct user_fxsr_struct fpx;
+
+ err = save_fpx_registers(pid, (unsigned long *) &fpx);
+ if (err < 0){
+ printk(KERN_ERR "copy_sc_to_user - save_fpx_registers "
+ "failed, errno = %d\n", err);
+ return 1;
+ }
+
+ err = convert_fxsr_to_user(to_fp, &fpx);
+ if (err)
+ return 1;
+
+ err |= __put_user(fpx.swd, &to_fp->status);
+ err |= __put_user(X86_FXSR_MAGIC, &to_fp->magic);
+ if (err)
+ return 1;
+
+ if (copy_to_user(&to_fp->_fxsr_env[0], &fpx,
+ sizeof(struct user_fxsr_struct)))
+ return 1;
+ }
+ else {
+ struct user_i387_struct fp;
+
+ err = save_fp_registers(pid, (unsigned long *) &fp);
+ if (copy_to_user(to_fp, &fp, sizeof(struct user_i387_struct)))
+ return 1;
+ }
+
+ return copy_to_user(to, &sc, sizeof(sc));
+}
+
+static int copy_ucontext_to_user(struct ucontext __user *uc,
+ struct _fpstate __user *fp, sigset_t *set,
+ unsigned long sp)
+{
+ int err = 0;
+
+ err |= put_user(current->sas_ss_sp, &uc->uc_stack.ss_sp);
+ err |= put_user(sas_ss_flags(sp), &uc->uc_stack.ss_flags);
+ err |= put_user(current->sas_ss_size, &uc->uc_stack.ss_size);
+ err |= copy_sc_to_user(&uc->uc_mcontext, fp, &current->thread.regs, sp);
+ err |= copy_to_user(&uc->uc_sigmask, set, sizeof(*set));
+ return err;
+}
+
+struct sigframe
+{
+ char __user *pretcode;
+ int sig;
+ struct sigcontext sc;
+ struct _fpstate fpstate;
+ unsigned long extramask[_NSIG_WORDS-1];
+ char retcode[8];
+};
+
+struct rt_sigframe
+{
+ char __user *pretcode;
+ int sig;
+ struct siginfo __user *pinfo;
+ void __user *puc;
+ struct siginfo info;
+ struct ucontext uc;
+ struct _fpstate fpstate;
+ char retcode[8];
+};
+
+int setup_signal_stack_sc(unsigned long stack_top, int sig,
+ struct k_sigaction *ka, struct pt_regs *regs,
+ sigset_t *mask)
+{
+ struct sigframe __user *frame;
+ void __user *restorer;
+ unsigned long save_sp = PT_REGS_SP(regs);
+ int err = 0;
+
+ /* This is the same calculation as i386 - ((sp + 4) & 15) == 0 */
+ stack_top = ((stack_top + 4) & -16UL) - 4;
+ frame = (struct sigframe __user *) stack_top - 1;
+ if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+ return 1;
+
+ restorer = frame->retcode;
+ if (ka->sa.sa_flags & SA_RESTORER)
+ restorer = ka->sa.sa_restorer;
+
+ /* Update SP now because the page fault handler refuses to extend
+ * the stack if the faulting address is too far below the current
+ * SP, which frame now certainly is. If there's an error, the original
+ * value is restored on the way out.
+ * When writing the sigcontext to the stack, we have to write the
+ * original value, so that's passed to copy_sc_to_user, which does
+ * the right thing with it.
+ */
+ PT_REGS_SP(regs) = (unsigned long) frame;
+
+ err |= __put_user(restorer, &frame->pretcode);
+ err |= __put_user(sig, &frame->sig);
+ err |= copy_sc_to_user(&frame->sc, NULL, regs, save_sp);
+ err |= __put_user(mask->sig[0], &frame->sc.oldmask);
+ if (_NSIG_WORDS > 1)
+ err |= __copy_to_user(&frame->extramask, &mask->sig[1],
+ sizeof(frame->extramask));
+
+ /*
+ * This is popl %eax ; movl $,%eax ; int $0x80
+ *
+ * WE DO NOT USE IT ANY MORE! It's only left here for historical
+ * reasons and because gdb uses it as a signature to notice
+ * signal handler stack frames.
+ */
+ err |= __put_user(0xb858, (short __user *)(frame->retcode+0));
+ err |= __put_user(__NR_sigreturn, (int __user *)(frame->retcode+2));
+ err |= __put_user(0x80cd, (short __user *)(frame->retcode+6));
+
+ if (err)
+ goto err;
+
+ PT_REGS_SP(regs) = (unsigned long) frame;
+ PT_REGS_IP(regs) = (unsigned long) ka->sa.sa_handler;
+ PT_REGS_EAX(regs) = (unsigned long) sig;
+ PT_REGS_EDX(regs) = (unsigned long) 0;
+ PT_REGS_ECX(regs) = (unsigned long) 0;
+
+ if ((current->ptrace & PT_DTRACE) && (current->ptrace & PT_PTRACED))
+ ptrace_notify(SIGTRAP);
+ return 0;
+
+err:
+ PT_REGS_SP(regs) = save_sp;
+ return err;
+}
+
+int setup_signal_stack_si(unsigned long stack_top, int sig,
+ struct k_sigaction *ka, struct pt_regs *regs,
+ siginfo_t *info, sigset_t *mask)
+{
+ struct rt_sigframe __user *frame;
+ void __user *restorer;
+ unsigned long save_sp = PT_REGS_SP(regs);
+ int err = 0;
+
+ stack_top &= -8UL;
+ frame = (struct rt_sigframe __user *) stack_top - 1;
+ if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+ return 1;
+
+ restorer = frame->retcode;
+ if (ka->sa.sa_flags & SA_RESTORER)
+ restorer = ka->sa.sa_restorer;
+
+ /* See comment above about why this is here */
+ PT_REGS_SP(regs) = (unsigned long) frame;
+
+ err |= __put_user(restorer, &frame->pretcode);
+ err |= __put_user(sig, &frame->sig);
+ err |= __put_user(&frame->info, &frame->pinfo);
+ err |= __put_user(&frame->uc, &frame->puc);
+ err |= copy_siginfo_to_user(&frame->info, info);
+ err |= copy_ucontext_to_user(&frame->uc, &frame->fpstate, mask,
+ save_sp);
+
+ /*
+ * This is movl $,%eax ; int $0x80
+ *
+ * WE DO NOT USE IT ANY MORE! It's only left here for historical
+ * reasons and because gdb uses it as a signature to notice
+ * signal handler stack frames.
+ */
+ err |= __put_user(0xb8, (char __user *)(frame->retcode+0));
+ err |= __put_user(__NR_rt_sigreturn, (int __user *)(frame->retcode+1));
+ err |= __put_user(0x80cd, (short __user *)(frame->retcode+5));
+
+ if (err)
+ goto err;
+
+ PT_REGS_IP(regs) = (unsigned long) ka->sa.sa_handler;
+ PT_REGS_EAX(regs) = (unsigned long) sig;
+ PT_REGS_EDX(regs) = (unsigned long) &frame->info;
+ PT_REGS_ECX(regs) = (unsigned long) &frame->uc;
+
+ if ((current->ptrace & PT_DTRACE) && (current->ptrace & PT_PTRACED))
+ ptrace_notify(SIGTRAP);
+ return 0;
+
+err:
+ PT_REGS_SP(regs) = save_sp;
+ return err;
+}
+
+long sys_sigreturn(struct pt_regs regs)
+{
+ unsigned long sp = PT_REGS_SP(&current->thread.regs);
+ struct sigframe __user *frame = (struct sigframe __user *)(sp - 8);
+ sigset_t set;
+ struct sigcontext __user *sc = &frame->sc;
+ unsigned long __user *oldmask = &sc->oldmask;
+ unsigned long __user *extramask = frame->extramask;
+ int sig_size = (_NSIG_WORDS - 1) * sizeof(unsigned long);
+
+ if (copy_from_user(&set.sig[0], oldmask, sizeof(set.sig[0])) ||
+ copy_from_user(&set.sig[1], extramask, sig_size))
+ goto segfault;
+
+ sigdelsetmask(&set, ~_BLOCKABLE);
+ set_current_blocked(&set);
+
+ if (copy_sc_from_user(&current->thread.regs, sc))
+ goto segfault;
+
+ /* Avoid ERESTART handling */
+ PT_REGS_SYSCALL_NR(&current->thread.regs) = -1;
+ return PT_REGS_SYSCALL_RET(&current->thread.regs);
+
+ segfault:
+ force_sig(SIGSEGV, current);
+ return 0;
+}
+
+long sys_rt_sigreturn(struct pt_regs regs)
+{
+ unsigned long sp = PT_REGS_SP(&current->thread.regs);
+ struct rt_sigframe __user *frame =
+ (struct rt_sigframe __user *) (sp - 4);
+ sigset_t set;
+ struct ucontext __user *uc = &frame->uc;
+ int sig_size = _NSIG_WORDS * sizeof(unsigned long);
+
+ if (copy_from_user(&set, &uc->uc_sigmask, sig_size))
+ goto segfault;
+
+ sigdelsetmask(&set, ~_BLOCKABLE);
+ set_current_blocked(&set);
+
+ if (copy_sc_from_user(&current->thread.regs, &uc->uc_mcontext))
+ goto segfault;
+
+ /* Avoid ERESTART handling */
+ PT_REGS_SYSCALL_NR(&current->thread.regs) = -1;
+ return PT_REGS_SYSCALL_RET(&current->thread.regs);
+
+ segfault:
+ force_sig(SIGSEGV, current);
+ return 0;
+}
diff --git a/arch/x86/um/signal_64.c b/arch/x86/um/signal_64.c
new file mode 100644
index 000000000000..255b2ca0ce67
--- /dev/null
+++ b/arch/x86/um/signal_64.c
@@ -0,0 +1,255 @@
+/*
+ * Copyright (C) 2003 PathScale, Inc.
+ * Copyright (C) 2003 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include <linux/personality.h>
+#include <linux/ptrace.h>
+#include <linux/kernel.h>
+#include <asm/unistd.h>
+#include <asm/uaccess.h>
+#include <asm/ucontext.h>
+#include "frame_kern.h"
+#include "skas.h"
+
+static int copy_sc_from_user(struct pt_regs *regs,
+ struct sigcontext __user *from)
+{
+ struct sigcontext sc;
+ struct user_i387_struct fp;
+ void __user *buf;
+ int err;
+
+ err = copy_from_user(&sc, from, sizeof(sc));
+ if (err)
+ return err;
+
+#define GETREG(regno, regname) regs->regs.gp[HOST_##regno] = sc.regname
+
+ GETREG(R8, r8);
+ GETREG(R9, r9);
+ GETREG(R10, r10);
+ GETREG(R11, r11);
+ GETREG(R12, r12);
+ GETREG(R13, r13);
+ GETREG(R14, r14);
+ GETREG(R15, r15);
+ GETREG(RDI, di);
+ GETREG(RSI, si);
+ GETREG(RBP, bp);
+ GETREG(RBX, bx);
+ GETREG(RDX, dx);
+ GETREG(RAX, ax);
+ GETREG(RCX, cx);
+ GETREG(SP, sp);
+ GETREG(IP, ip);
+ GETREG(EFLAGS, flags);
+ GETREG(CS, cs);
+#undef GETREG
+
+ buf = sc.fpstate;
+
+ err = copy_from_user(&fp, buf, sizeof(struct user_i387_struct));
+ if (err)
+ return 1;
+
+ err = restore_fp_registers(userspace_pid[current_thread_info()->cpu],
+ (unsigned long *) &fp);
+ if (err < 0) {
+ printk(KERN_ERR "copy_sc_from_user - "
+ "restore_fp_registers failed, errno = %d\n",
+ -err);
+ return 1;
+ }
+
+ return 0;
+}
+
+static int copy_sc_to_user(struct sigcontext __user *to,
+ struct _fpstate __user *to_fp, struct pt_regs *regs,
+ unsigned long mask, unsigned long sp)
+{
+ struct faultinfo * fi = &current->thread.arch.faultinfo;
+ struct sigcontext sc;
+ struct user_i387_struct fp;
+ int err = 0;
+ memset(&sc, 0, sizeof(struct sigcontext));
+
+#define PUTREG(regno, regname) sc.regname = regs->regs.gp[HOST_##regno]
+
+ PUTREG(RDI, di);
+ PUTREG(RSI, si);
+ PUTREG(RBP, bp);
+ /*
+ * Must use original RSP, which is passed in, rather than what's in
+ * signal frame.
+ */
+ sc.sp = sp;
+ PUTREG(RBX, bx);
+ PUTREG(RDX, dx);
+ PUTREG(RCX, cx);
+ PUTREG(RAX, ax);
+ PUTREG(R8, r8);
+ PUTREG(R9, r9);
+ PUTREG(R10, r10);
+ PUTREG(R11, r11);
+ PUTREG(R12, r12);
+ PUTREG(R13, r13);
+ PUTREG(R14, r14);
+ PUTREG(R15, r15);
+ PUTREG(CS, cs); /* XXX x86_64 doesn't do this */
+
+ sc.cr2 = fi->cr2;
+ sc.err = fi->error_code;
+ sc.trapno = fi->trap_no;
+
+ PUTREG(IP, ip);
+ PUTREG(EFLAGS, flags);
+#undef PUTREG
+
+ sc.oldmask = mask;
+
+ err = copy_to_user(to, &sc, sizeof(struct sigcontext));
+ if (err)
+ return 1;
+
+ err = save_fp_registers(userspace_pid[current_thread_info()->cpu],
+ (unsigned long *) &fp);
+ if (err < 0) {
+ printk(KERN_ERR "copy_sc_from_user - restore_fp_registers "
+ "failed, errno = %d\n", -err);
+ return 1;
+ }
+
+ if (copy_to_user(to_fp, &fp, sizeof(struct user_i387_struct)))
+ return 1;
+
+ return err;
+}
+
+struct rt_sigframe
+{
+ char __user *pretcode;
+ struct ucontext uc;
+ struct siginfo info;
+ struct _fpstate fpstate;
+};
+
+int setup_signal_stack_si(unsigned long stack_top, int sig,
+ struct k_sigaction *ka, struct pt_regs * regs,
+ siginfo_t *info, sigset_t *set)
+{
+ struct rt_sigframe __user *frame;
+ unsigned long save_sp = PT_REGS_RSP(regs);
+ int err = 0;
+ struct task_struct *me = current;
+
+ frame = (struct rt_sigframe __user *)
+ round_down(stack_top - sizeof(struct rt_sigframe), 16);
+ /* Subtract 128 for a red zone and 8 for proper alignment */
+ frame = (struct rt_sigframe __user *) ((unsigned long) frame - 128 - 8);
+
+ if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+ goto out;
+
+ if (ka->sa.sa_flags & SA_SIGINFO) {
+ err |= copy_siginfo_to_user(&frame->info, info);
+ if (err)
+ goto out;
+ }
+
+ /*
+ * Update SP now because the page fault handler refuses to extend
+ * the stack if the faulting address is too far below the current
+ * SP, which frame now certainly is. If there's an error, the original
+ * value is restored on the way out.
+ * When writing the sigcontext to the stack, we have to write the
+ * original value, so that's passed to copy_sc_to_user, which does
+ * the right thing with it.
+ */
+ PT_REGS_RSP(regs) = (unsigned long) frame;
+
+ /* Create the ucontext. */
+ err |= __put_user(0, &frame->uc.uc_flags);
+ err |= __put_user(0, &frame->uc.uc_link);
+ err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
+ err |= __put_user(sas_ss_flags(save_sp),
+ &frame->uc.uc_stack.ss_flags);
+ err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size);
+ err |= copy_sc_to_user(&frame->uc.uc_mcontext, &frame->fpstate, regs,
+ set->sig[0], save_sp);
+ err |= __put_user(&frame->fpstate, &frame->uc.uc_mcontext.fpstate);
+ if (sizeof(*set) == 16) {
+ __put_user(set->sig[0], &frame->uc.uc_sigmask.sig[0]);
+ __put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]);
+ }
+ else
+ err |= __copy_to_user(&frame->uc.uc_sigmask, set,
+ sizeof(*set));
+
+ /*
+ * Set up to return from userspace. If provided, use a stub
+ * already in userspace.
+ */
+ /* x86-64 should always use SA_RESTORER. */
+ if (ka->sa.sa_flags & SA_RESTORER)
+ err |= __put_user(ka->sa.sa_restorer, &frame->pretcode);
+ else
+ /* could use a vstub here */
+ goto restore_sp;
+
+ if (err)
+ goto restore_sp;
+
+ /* Set up registers for signal handler */
+ {
+ struct exec_domain *ed = current_thread_info()->exec_domain;
+ if (unlikely(ed && ed->signal_invmap && sig < 32))
+ sig = ed->signal_invmap[sig];
+ }
+
+ PT_REGS_RDI(regs) = sig;
+ /* In case the signal handler was declared without prototypes */
+ PT_REGS_RAX(regs) = 0;
+
+ /*
+ * This also works for non SA_SIGINFO handlers because they expect the
+ * next argument after the signal number on the stack.
+ */
+ PT_REGS_RSI(regs) = (unsigned long) &frame->info;
+ PT_REGS_RDX(regs) = (unsigned long) &frame->uc;
+ PT_REGS_RIP(regs) = (unsigned long) ka->sa.sa_handler;
+ out:
+ return err;
+
+restore_sp:
+ PT_REGS_RSP(regs) = save_sp;
+ return err;
+}
+
+long sys_rt_sigreturn(struct pt_regs *regs)
+{
+ unsigned long sp = PT_REGS_SP(&current->thread.regs);
+ struct rt_sigframe __user *frame =
+ (struct rt_sigframe __user *)(sp - 8);
+ struct ucontext __user *uc = &frame->uc;
+ sigset_t set;
+
+ if (copy_from_user(&set, &uc->uc_sigmask, sizeof(set)))
+ goto segfault;
+
+ sigdelsetmask(&set, ~_BLOCKABLE);
+ set_current_blocked(&set);
+
+ if (copy_sc_from_user(&current->thread.regs, &uc->uc_mcontext))
+ goto segfault;
+
+ /* Avoid ERESTART handling */
+ PT_REGS_SYSCALL_NR(&current->thread.regs) = -1;
+ return PT_REGS_SYSCALL_RET(&current->thread.regs);
+
+ segfault:
+ force_sig(SIGSEGV, current);
+ return 0;
+}
diff --git a/arch/x86/um/stub_32.S b/arch/x86/um/stub_32.S
new file mode 100644
index 000000000000..54a36ec20cb7
--- /dev/null
+++ b/arch/x86/um/stub_32.S
@@ -0,0 +1,51 @@
+#include "as-layout.h"
+
+ .globl syscall_stub
+.section .__syscall_stub, "ax"
+
+ .globl batch_syscall_stub
+batch_syscall_stub:
+ /* load pointer to first operation */
+ mov $(STUB_DATA+8), %esp
+
+again:
+ /* load length of additional data */
+ mov 0x0(%esp), %eax
+
+ /* if(length == 0) : end of list */
+ /* write possible 0 to header */
+ mov %eax, STUB_DATA+4
+ cmpl $0, %eax
+ jz done
+
+ /* save current pointer */
+ mov %esp, STUB_DATA+4
+
+ /* skip additional data */
+ add %eax, %esp
+
+ /* load syscall-# */
+ pop %eax
+
+ /* load syscall params */
+ pop %ebx
+ pop %ecx
+ pop %edx
+ pop %esi
+ pop %edi
+ pop %ebp
+
+ /* execute syscall */
+ int $0x80
+
+ /* check return value */
+ pop %ebx
+ cmp %ebx, %eax
+ je again
+
+done:
+ /* save return value */
+ mov %eax, STUB_DATA
+
+ /* stop */
+ int3
diff --git a/arch/x86/um/stub_64.S b/arch/x86/um/stub_64.S
new file mode 100644
index 000000000000..20e4a96a6dcb
--- /dev/null
+++ b/arch/x86/um/stub_64.S
@@ -0,0 +1,66 @@
+#include "as-layout.h"
+
+ .globl syscall_stub
+.section .__syscall_stub, "ax"
+syscall_stub:
+ syscall
+ /* We don't have 64-bit constants, so this constructs the address
+ * we need.
+ */
+ movq $(STUB_DATA >> 32), %rbx
+ salq $32, %rbx
+ movq $(STUB_DATA & 0xffffffff), %rcx
+ or %rcx, %rbx
+ movq %rax, (%rbx)
+ int3
+
+ .globl batch_syscall_stub
+batch_syscall_stub:
+ mov $(STUB_DATA >> 32), %rbx
+ sal $32, %rbx
+ mov $(STUB_DATA & 0xffffffff), %rax
+ or %rax, %rbx
+ /* load pointer to first operation */
+ mov %rbx, %rsp
+ add $0x10, %rsp
+again:
+ /* load length of additional data */
+ mov 0x0(%rsp), %rax
+
+ /* if(length == 0) : end of list */
+ /* write possible 0 to header */
+ mov %rax, 8(%rbx)
+ cmp $0, %rax
+ jz done
+
+ /* save current pointer */
+ mov %rsp, 8(%rbx)
+
+ /* skip additional data */
+ add %rax, %rsp
+
+ /* load syscall-# */
+ pop %rax
+
+ /* load syscall params */
+ pop %rdi
+ pop %rsi
+ pop %rdx
+ pop %r10
+ pop %r8
+ pop %r9
+
+ /* execute syscall */
+ syscall
+
+ /* check return value */
+ pop %rcx
+ cmp %rcx, %rax
+ je again
+
+done:
+ /* save return value */
+ mov %rax, (%rbx)
+
+ /* stop */
+ int3
diff --git a/arch/x86/um/stub_segv.c b/arch/x86/um/stub_segv.c
new file mode 100644
index 000000000000..b7450bd22e7d
--- /dev/null
+++ b/arch/x86/um/stub_segv.c
@@ -0,0 +1,19 @@
+/*
+ * Copyright (C) 2004 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Licensed under the GPL
+ */
+
+#include "sysdep/stub.h"
+#include "sysdep/faultinfo.h"
+#include "sysdep/mcontext.h"
+
+void __attribute__ ((__section__ (".__syscall_stub")))
+stub_segv_handler(int sig, siginfo_t *info, void *p)
+{
+ struct ucontext *uc = p;
+
+ GET_FAULTINFO_FROM_MC(*((struct faultinfo *) STUB_DATA),
+ &uc->uc_mcontext);
+ trap_myself();
+}
+
diff --git a/arch/x86/um/sys_call_table_32.S b/arch/x86/um/sys_call_table_32.S
new file mode 100644
index 000000000000..de274071455d
--- /dev/null
+++ b/arch/x86/um/sys_call_table_32.S
@@ -0,0 +1,28 @@
+#include <linux/linkage.h>
+/* Steal i386 syscall table for our purposes, but with some slight changes.*/
+
+#define sys_iopl sys_ni_syscall
+#define sys_ioperm sys_ni_syscall
+
+#define sys_vm86old sys_ni_syscall
+#define sys_vm86 sys_ni_syscall
+
+#define old_mmap sys_old_mmap
+
+#define ptregs_fork sys_fork
+#define ptregs_execve sys_execve
+#define ptregs_iopl sys_iopl
+#define ptregs_vm86old sys_vm86old
+#define ptregs_sigreturn sys_sigreturn
+#define ptregs_clone sys_clone
+#define ptregs_vm86 sys_vm86
+#define ptregs_rt_sigreturn sys_rt_sigreturn
+#define ptregs_sigaltstack sys_sigaltstack
+#define ptregs_vfork sys_vfork
+
+.section .rodata,"a"
+
+#include "../../x86/kernel/syscall_table_32.S"
+
+ENTRY(syscall_table_size)
+.long .-sys_call_table
diff --git a/arch/x86/um/sys_call_table_64.c b/arch/x86/um/sys_call_table_64.c
new file mode 100644
index 000000000000..f46de82d675c
--- /dev/null
+++ b/arch/x86/um/sys_call_table_64.c
@@ -0,0 +1,64 @@
+/*
+ * System call table for UML/x86-64, copied from arch/x86_64/kernel/syscall.c
+ * with some changes for UML.
+ */
+
+#include <linux/linkage.h>
+#include <linux/sys.h>
+#include <linux/cache.h>
+
+#define __NO_STUBS
+
+/*
+ * Below you can see, in terms of #define's, the differences between the x86-64
+ * and the UML syscall table.
+ */
+
+/* Not going to be implemented by UML, since we have no hardware. */
+#define stub_iopl sys_ni_syscall
+#define sys_ioperm sys_ni_syscall
+
+/*
+ * The UML TLS problem. Note that x86_64 does not implement this, so the below
+ * is needed only for the ia32 compatibility.
+ */
+
+/* On UML we call it this way ("old" means it's not mmap2) */
+#define sys_mmap old_mmap
+
+#define stub_clone sys_clone
+#define stub_fork sys_fork
+#define stub_vfork sys_vfork
+#define stub_execve sys_execve
+#define stub_rt_sigsuspend sys_rt_sigsuspend
+#define stub_sigaltstack sys_sigaltstack
+#define stub_rt_sigreturn sys_rt_sigreturn
+
+#define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ;
+#undef _ASM_X86_UNISTD_64_H
+#include "../../x86/include/asm/unistd_64.h"
+
+#undef __SYSCALL
+#define __SYSCALL(nr, sym) [ nr ] = sym,
+#undef _ASM_X86_UNISTD_64_H
+
+typedef void (*sys_call_ptr_t)(void);
+
+extern void sys_ni_syscall(void);
+
+/*
+ * We used to have a trick here which made sure that holes in the
+ * x86_64 table were filled in with sys_ni_syscall, but a comment in
+ * unistd_64.h says that holes aren't allowed, so the trick was
+ * removed.
+ * The trick looked like this
+ * [0 ... UM_NR_syscall_max] = &sys_ni_syscall
+ * before including unistd_64.h - the later initializations overwrote
+ * the sys_ni_syscall filler.
+ */
+
+sys_call_ptr_t sys_call_table[] __cacheline_aligned = {
+#include "../../x86/include/asm/unistd_64.h"
+};
+
+int syscall_table_size = sizeof(sys_call_table);
diff --git a/arch/x86/um/syscalls_32.c b/arch/x86/um/syscalls_32.c
new file mode 100644
index 000000000000..70ca357393b8
--- /dev/null
+++ b/arch/x86/um/syscalls_32.c
@@ -0,0 +1,66 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/shm.h"
+#include "linux/ipc.h"
+#include "linux/syscalls.h"
+#include "asm/mman.h"
+#include "asm/uaccess.h"
+#include "asm/unistd.h"
+
+/*
+ * The prototype on i386 is:
+ *
+ * int clone(int flags, void * child_stack, int * parent_tidptr, struct user_desc * newtls, int * child_tidptr)
+ *
+ * and the "newtls" arg. on i386 is read by copy_thread directly from the
+ * register saved on the stack.
+ */
+long sys_clone(unsigned long clone_flags, unsigned long newsp,
+ int __user *parent_tid, void *newtls, int __user *child_tid)
+{
+ long ret;
+
+ if (!newsp)
+ newsp = UPT_SP(&current->thread.regs.regs);
+
+ current->thread.forking = 1;
+ ret = do_fork(clone_flags, newsp, &current->thread.regs, 0, parent_tid,
+ child_tid);
+ current->thread.forking = 0;
+ return ret;
+}
+
+long sys_sigaction(int sig, const struct old_sigaction __user *act,
+ struct old_sigaction __user *oact)
+{
+ struct k_sigaction new_ka, old_ka;
+ int ret;
+
+ if (act) {
+ old_sigset_t mask;
+ if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
+ __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
+ __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
+ return -EFAULT;
+ __get_user(new_ka.sa.sa_flags, &act->sa_flags);
+ __get_user(mask, &act->sa_mask);
+ siginitset(&new_ka.sa.sa_mask, mask);
+ }
+
+ ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+ if (!ret && oact) {
+ if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
+ __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
+ __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
+ return -EFAULT;
+ __put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+ __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
+ }
+
+ return ret;
+}
diff --git a/arch/x86/um/syscalls_64.c b/arch/x86/um/syscalls_64.c
new file mode 100644
index 000000000000..f3d82bb6e15a
--- /dev/null
+++ b/arch/x86/um/syscalls_64.c
@@ -0,0 +1,102 @@
+/*
+ * Copyright (C) 2003 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#include "linux/linkage.h"
+#include "linux/personality.h"
+#include "linux/utsname.h"
+#include "asm/prctl.h" /* XXX This should get the constants from libc */
+#include "asm/uaccess.h"
+#include "os.h"
+
+long arch_prctl(struct task_struct *task, int code, unsigned long __user *addr)
+{
+ unsigned long *ptr = addr, tmp;
+ long ret;
+ int pid = task->mm->context.id.u.pid;
+
+ /*
+ * With ARCH_SET_FS (and ARCH_SET_GS is treated similarly to
+ * be safe), we need to call arch_prctl on the host because
+ * setting %fs may result in something else happening (like a
+ * GDT or thread.fs being set instead). So, we let the host
+ * fiddle the registers and thread struct and restore the
+ * registers afterwards.
+ *
+ * So, the saved registers are stored to the process (this
+ * needed because a stub may have been the last thing to run),
+ * arch_prctl is run on the host, then the registers are read
+ * back.
+ */
+ switch (code) {
+ case ARCH_SET_FS:
+ case ARCH_SET_GS:
+ ret = restore_registers(pid, &current->thread.regs.regs);
+ if (ret)
+ return ret;
+ break;
+ case ARCH_GET_FS:
+ case ARCH_GET_GS:
+ /*
+ * With these two, we read to a local pointer and
+ * put_user it to the userspace pointer that we were
+ * given. If addr isn't valid (because it hasn't been
+ * faulted in or is just bogus), we want put_user to
+ * fault it in (or return -EFAULT) instead of having
+ * the host return -EFAULT.
+ */
+ ptr = &tmp;
+ }
+
+ ret = os_arch_prctl(pid, code, ptr);
+ if (ret)
+ return ret;
+
+ switch (code) {
+ case ARCH_SET_FS:
+ current->thread.arch.fs = (unsigned long) ptr;
+ ret = save_registers(pid, &current->thread.regs.regs);
+ break;
+ case ARCH_SET_GS:
+ ret = save_registers(pid, &current->thread.regs.regs);
+ break;
+ case ARCH_GET_FS:
+ ret = put_user(tmp, addr);
+ break;
+ case ARCH_GET_GS:
+ ret = put_user(tmp, addr);
+ break;
+ }
+
+ return ret;
+}
+
+long sys_arch_prctl(int code, unsigned long addr)
+{
+ return arch_prctl(current, code, (unsigned long __user *) addr);
+}
+
+long sys_clone(unsigned long clone_flags, unsigned long newsp,
+ void __user *parent_tid, void __user *child_tid)
+{
+ long ret;
+
+ if (!newsp)
+ newsp = UPT_SP(&current->thread.regs.regs);
+ current->thread.forking = 1;
+ ret = do_fork(clone_flags, newsp, &current->thread.regs, 0, parent_tid,
+ child_tid);
+ current->thread.forking = 0;
+ return ret;
+}
+
+void arch_switch_to(struct task_struct *to)
+{
+ if ((to->thread.arch.fs == 0) || (to->mm == NULL))
+ return;
+
+ arch_prctl(to, ARCH_SET_FS, (void __user *) to->thread.arch.fs);
+}
diff --git a/arch/x86/um/sysrq_32.c b/arch/x86/um/sysrq_32.c
new file mode 100644
index 000000000000..171b3e9dc867
--- /dev/null
+++ b/arch/x86/um/sysrq_32.c
@@ -0,0 +1,101 @@
+/*
+ * Copyright (C) 2001 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/smp.h"
+#include "linux/sched.h"
+#include "linux/kallsyms.h"
+#include "asm/ptrace.h"
+#include "sysrq.h"
+
+/* This is declared by <linux/sched.h> */
+void show_regs(struct pt_regs *regs)
+{
+ printk("\n");
+ printk("EIP: %04lx:[<%08lx>] CPU: %d %s",
+ 0xffff & PT_REGS_CS(regs), PT_REGS_IP(regs),
+ smp_processor_id(), print_tainted());
+ if (PT_REGS_CS(regs) & 3)
+ printk(" ESP: %04lx:%08lx", 0xffff & PT_REGS_SS(regs),
+ PT_REGS_SP(regs));
+ printk(" EFLAGS: %08lx\n %s\n", PT_REGS_EFLAGS(regs),
+ print_tainted());
+ printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
+ PT_REGS_EAX(regs), PT_REGS_EBX(regs),
+ PT_REGS_ECX(regs),
+ PT_REGS_EDX(regs));
+ printk("ESI: %08lx EDI: %08lx EBP: %08lx",
+ PT_REGS_ESI(regs), PT_REGS_EDI(regs),
+ PT_REGS_EBP(regs));
+ printk(" DS: %04lx ES: %04lx\n",
+ 0xffff & PT_REGS_DS(regs),
+ 0xffff & PT_REGS_ES(regs));
+
+ show_trace(NULL, (unsigned long *) &regs);
+}
+
+/* Copied from i386. */
+static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
+{
+ return p > (void *)tinfo &&
+ p < (void *)tinfo + THREAD_SIZE - 3;
+}
+
+/* Adapted from i386 (we also print the address we read from). */
+static inline unsigned long print_context_stack(struct thread_info *tinfo,
+ unsigned long *stack, unsigned long ebp)
+{
+ unsigned long addr;
+
+#ifdef CONFIG_FRAME_POINTER
+ while (valid_stack_ptr(tinfo, (void *)ebp)) {
+ addr = *(unsigned long *)(ebp + 4);
+ printk("%08lx: [<%08lx>]", ebp + 4, addr);
+ print_symbol(" %s", addr);
+ printk("\n");
+ ebp = *(unsigned long *)ebp;
+ }
+#else
+ while (valid_stack_ptr(tinfo, stack)) {
+ addr = *stack;
+ if (__kernel_text_address(addr)) {
+ printk("%08lx: [<%08lx>]", (unsigned long) stack, addr);
+ print_symbol(" %s", addr);
+ printk("\n");
+ }
+ stack++;
+ }
+#endif
+ return ebp;
+}
+
+void show_trace(struct task_struct* task, unsigned long * stack)
+{
+ unsigned long ebp;
+ struct thread_info *context;
+
+ /* Turn this into BUG_ON if possible. */
+ if (!stack) {
+ stack = (unsigned long*) &stack;
+ printk("show_trace: got NULL stack, implicit assumption task == current");
+ WARN_ON(1);
+ }
+
+ if (!task)
+ task = current;
+
+ if (task != current) {
+ ebp = (unsigned long) KSTK_EBP(task);
+ } else {
+ asm ("movl %%ebp, %0" : "=r" (ebp) : );
+ }
+
+ context = (struct thread_info *)
+ ((unsigned long)stack & (~(THREAD_SIZE - 1)));
+ print_context_stack(context, stack, ebp);
+
+ printk("\n");
+}
+
diff --git a/arch/x86/um/sysrq_64.c b/arch/x86/um/sysrq_64.c
new file mode 100644
index 000000000000..f4f82beb3508
--- /dev/null
+++ b/arch/x86/um/sysrq_64.c
@@ -0,0 +1,41 @@
+/*
+ * Copyright 2003 PathScale, Inc.
+ *
+ * Licensed under the GPL
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/utsname.h>
+#include <asm/current.h>
+#include <asm/ptrace.h>
+#include "sysrq.h"
+
+void __show_regs(struct pt_regs *regs)
+{
+ printk("\n");
+ print_modules();
+ printk(KERN_INFO "Pid: %d, comm: %.20s %s %s\n", task_pid_nr(current),
+ current->comm, print_tainted(), init_utsname()->release);
+ printk(KERN_INFO "RIP: %04lx:[<%016lx>]\n", PT_REGS_CS(regs) & 0xffff,
+ PT_REGS_RIP(regs));
+ printk(KERN_INFO "RSP: %016lx EFLAGS: %08lx\n", PT_REGS_RSP(regs),
+ PT_REGS_EFLAGS(regs));
+ printk(KERN_INFO "RAX: %016lx RBX: %016lx RCX: %016lx\n",
+ PT_REGS_RAX(regs), PT_REGS_RBX(regs), PT_REGS_RCX(regs));
+ printk(KERN_INFO "RDX: %016lx RSI: %016lx RDI: %016lx\n",
+ PT_REGS_RDX(regs), PT_REGS_RSI(regs), PT_REGS_RDI(regs));
+ printk(KERN_INFO "RBP: %016lx R08: %016lx R09: %016lx\n",
+ PT_REGS_RBP(regs), PT_REGS_R8(regs), PT_REGS_R9(regs));
+ printk(KERN_INFO "R10: %016lx R11: %016lx R12: %016lx\n",
+ PT_REGS_R10(regs), PT_REGS_R11(regs), PT_REGS_R12(regs));
+ printk(KERN_INFO "R13: %016lx R14: %016lx R15: %016lx\n",
+ PT_REGS_R13(regs), PT_REGS_R14(regs), PT_REGS_R15(regs));
+}
+
+void show_regs(struct pt_regs *regs)
+{
+ __show_regs(regs);
+ show_trace(current, (unsigned long *) &regs);
+}
diff --git a/arch/x86/um/tls_32.c b/arch/x86/um/tls_32.c
new file mode 100644
index 000000000000..c6c7131e563b
--- /dev/null
+++ b/arch/x86/um/tls_32.c
@@ -0,0 +1,396 @@
+/*
+ * Copyright (C) 2005 Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
+ * Licensed under the GPL
+ */
+
+#include "linux/percpu.h"
+#include "linux/sched.h"
+#include "asm/uaccess.h"
+#include "os.h"
+#include "skas.h"
+#include "sysdep/tls.h"
+
+/*
+ * If needed we can detect when it's uninitialized.
+ *
+ * These are initialized in an initcall and unchanged thereafter.
+ */
+static int host_supports_tls = -1;
+int host_gdt_entry_tls_min;
+
+int do_set_thread_area(struct user_desc *info)
+{
+ int ret;
+ u32 cpu;
+
+ cpu = get_cpu();
+ ret = os_set_thread_area(info, userspace_pid[cpu]);
+ put_cpu();
+
+ if (ret)
+ printk(KERN_ERR "PTRACE_SET_THREAD_AREA failed, err = %d, "
+ "index = %d\n", ret, info->entry_number);
+
+ return ret;
+}
+
+int do_get_thread_area(struct user_desc *info)
+{
+ int ret;
+ u32 cpu;
+
+ cpu = get_cpu();
+ ret = os_get_thread_area(info, userspace_pid[cpu]);
+ put_cpu();
+
+ if (ret)
+ printk(KERN_ERR "PTRACE_GET_THREAD_AREA failed, err = %d, "
+ "index = %d\n", ret, info->entry_number);
+
+ return ret;
+}
+
+/*
+ * sys_get_thread_area: get a yet unused TLS descriptor index.
+ * XXX: Consider leaving one free slot for glibc usage at first place. This must
+ * be done here (and by changing GDT_ENTRY_TLS_* macros) and nowhere else.
+ *
+ * Also, this must be tested when compiling in SKAS mode with dynamic linking
+ * and running against NPTL.
+ */
+static int get_free_idx(struct task_struct* task)
+{
+ struct thread_struct *t = &task->thread;
+ int idx;
+
+ if (!t->arch.tls_array)
+ return GDT_ENTRY_TLS_MIN;
+
+ for (idx = 0; idx < GDT_ENTRY_TLS_ENTRIES; idx++)
+ if (!t->arch.tls_array[idx].present)
+ return idx + GDT_ENTRY_TLS_MIN;
+ return -ESRCH;
+}
+
+static inline void clear_user_desc(struct user_desc* info)
+{
+ /* Postcondition: LDT_empty(info) returns true. */
+ memset(info, 0, sizeof(*info));
+
+ /*
+ * Check the LDT_empty or the i386 sys_get_thread_area code - we obtain
+ * indeed an empty user_desc.
+ */
+ info->read_exec_only = 1;
+ info->seg_not_present = 1;
+}
+
+#define O_FORCE 1
+
+static int load_TLS(int flags, struct task_struct *to)
+{
+ int ret = 0;
+ int idx;
+
+ for (idx = GDT_ENTRY_TLS_MIN; idx < GDT_ENTRY_TLS_MAX; idx++) {
+ struct uml_tls_struct* curr =
+ &to->thread.arch.tls_array[idx - GDT_ENTRY_TLS_MIN];
+
+ /*
+ * Actually, now if it wasn't flushed it gets cleared and
+ * flushed to the host, which will clear it.
+ */
+ if (!curr->present) {
+ if (!curr->flushed) {
+ clear_user_desc(&curr->tls);
+ curr->tls.entry_number = idx;
+ } else {
+ WARN_ON(!LDT_empty(&curr->tls));
+ continue;
+ }
+ }
+
+ if (!(flags & O_FORCE) && curr->flushed)
+ continue;
+
+ ret = do_set_thread_area(&curr->tls);
+ if (ret)
+ goto out;
+
+ curr->flushed = 1;
+ }
+out:
+ return ret;
+}
+
+/*
+ * Verify if we need to do a flush for the new process, i.e. if there are any
+ * present desc's, only if they haven't been flushed.
+ */
+static inline int needs_TLS_update(struct task_struct *task)
+{
+ int i;
+ int ret = 0;
+
+ for (i = GDT_ENTRY_TLS_MIN; i < GDT_ENTRY_TLS_MAX; i++) {
+ struct uml_tls_struct* curr =
+ &task->thread.arch.tls_array[i - GDT_ENTRY_TLS_MIN];
+
+ /*
+ * Can't test curr->present, we may need to clear a descriptor
+ * which had a value.
+ */
+ if (curr->flushed)
+ continue;
+ ret = 1;
+ break;
+ }
+ return ret;
+}
+
+/*
+ * On a newly forked process, the TLS descriptors haven't yet been flushed. So
+ * we mark them as such and the first switch_to will do the job.
+ */
+void clear_flushed_tls(struct task_struct *task)
+{
+ int i;
+
+ for (i = GDT_ENTRY_TLS_MIN; i < GDT_ENTRY_TLS_MAX; i++) {
+ struct uml_tls_struct* curr =
+ &task->thread.arch.tls_array[i - GDT_ENTRY_TLS_MIN];
+
+ /*
+ * Still correct to do this, if it wasn't present on the host it
+ * will remain as flushed as it was.
+ */
+ if (!curr->present)
+ continue;
+
+ curr->flushed = 0;
+ }
+}
+
+/*
+ * In SKAS0 mode, currently, multiple guest threads sharing the same ->mm have a
+ * common host process. So this is needed in SKAS0 too.
+ *
+ * However, if each thread had a different host process (and this was discussed
+ * for SMP support) this won't be needed.
+ *
+ * And this will not need be used when (and if) we'll add support to the host
+ * SKAS patch.
+ */
+
+int arch_switch_tls(struct task_struct *to)
+{
+ if (!host_supports_tls)
+ return 0;
+
+ /*
+ * We have no need whatsoever to switch TLS for kernel threads; beyond
+ * that, that would also result in us calling os_set_thread_area with
+ * userspace_pid[cpu] == 0, which gives an error.
+ */
+ if (likely(to->mm))
+ return load_TLS(O_FORCE, to);
+
+ return 0;
+}
+
+static int set_tls_entry(struct task_struct* task, struct user_desc *info,
+ int idx, int flushed)
+{
+ struct thread_struct *t = &task->thread;
+
+ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
+ return -EINVAL;
+
+ t->arch.tls_array[idx - GDT_ENTRY_TLS_MIN].tls = *info;
+ t->arch.tls_array[idx - GDT_ENTRY_TLS_MIN].present = 1;
+ t->arch.tls_array[idx - GDT_ENTRY_TLS_MIN].flushed = flushed;
+
+ return 0;
+}
+
+int arch_copy_tls(struct task_struct *new)
+{
+ struct user_desc info;
+ int idx, ret = -EFAULT;
+
+ if (copy_from_user(&info,
+ (void __user *) UPT_ESI(&new->thread.regs.regs),
+ sizeof(info)))
+ goto out;
+
+ ret = -EINVAL;
+ if (LDT_empty(&info))
+ goto out;
+
+ idx = info.entry_number;
+
+ ret = set_tls_entry(new, &info, idx, 0);
+out:
+ return ret;
+}
+
+/* XXX: use do_get_thread_area to read the host value? I'm not at all sure! */
+static int get_tls_entry(struct task_struct *task, struct user_desc *info,
+ int idx)
+{
+ struct thread_struct *t = &task->thread;
+
+ if (!t->arch.tls_array)
+ goto clear;
+
+ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
+ return -EINVAL;
+
+ if (!t->arch.tls_array[idx - GDT_ENTRY_TLS_MIN].present)
+ goto clear;
+
+ *info = t->arch.tls_array[idx - GDT_ENTRY_TLS_MIN].tls;
+
+out:
+ /*
+ * Temporary debugging check, to make sure that things have been
+ * flushed. This could be triggered if load_TLS() failed.
+ */
+ if (unlikely(task == current &&
+ !t->arch.tls_array[idx - GDT_ENTRY_TLS_MIN].flushed)) {
+ printk(KERN_ERR "get_tls_entry: task with pid %d got here "
+ "without flushed TLS.", current->pid);
+ }
+
+ return 0;
+clear:
+ /*
+ * When the TLS entry has not been set, the values read to user in the
+ * tls_array are 0 (because it's cleared at boot, see
+ * arch/i386/kernel/head.S:cpu_gdt_table). Emulate that.
+ */
+ clear_user_desc(info);
+ info->entry_number = idx;
+ goto out;
+}
+
+int sys_set_thread_area(struct user_desc __user *user_desc)
+{
+ struct user_desc info;
+ int idx, ret;
+
+ if (!host_supports_tls)
+ return -ENOSYS;
+
+ if (copy_from_user(&info, user_desc, sizeof(info)))
+ return -EFAULT;
+
+ idx = info.entry_number;
+
+ if (idx == -1) {
+ idx = get_free_idx(current);
+ if (idx < 0)
+ return idx;
+ info.entry_number = idx;
+ /* Tell the user which slot we chose for him.*/
+ if (put_user(idx, &user_desc->entry_number))
+ return -EFAULT;
+ }
+
+ ret = do_set_thread_area(&info);
+ if (ret)
+ return ret;
+ return set_tls_entry(current, &info, idx, 1);
+}
+
+/*
+ * Perform set_thread_area on behalf of the traced child.
+ * Note: error handling is not done on the deferred load, and this differ from
+ * i386. However the only possible error are caused by bugs.
+ */
+int ptrace_set_thread_area(struct task_struct *child, int idx,
+ struct user_desc __user *user_desc)
+{
+ struct user_desc info;
+
+ if (!host_supports_tls)
+ return -EIO;
+
+ if (copy_from_user(&info, user_desc, sizeof(info)))
+ return -EFAULT;
+
+ return set_tls_entry(child, &info, idx, 0);
+}
+
+int sys_get_thread_area(struct user_desc __user *user_desc)
+{
+ struct user_desc info;
+ int idx, ret;
+
+ if (!host_supports_tls)
+ return -ENOSYS;
+
+ if (get_user(idx, &user_desc->entry_number))
+ return -EFAULT;
+
+ ret = get_tls_entry(current, &info, idx);
+ if (ret < 0)
+ goto out;
+
+ if (copy_to_user(user_desc, &info, sizeof(info)))
+ ret = -EFAULT;
+
+out:
+ return ret;
+}
+
+/*
+ * Perform get_thread_area on behalf of the traced child.
+ */
+int ptrace_get_thread_area(struct task_struct *child, int idx,
+ struct user_desc __user *user_desc)
+{
+ struct user_desc info;
+ int ret;
+
+ if (!host_supports_tls)
+ return -EIO;
+
+ ret = get_tls_entry(child, &info, idx);
+ if (ret < 0)
+ goto out;
+
+ if (copy_to_user(user_desc, &info, sizeof(info)))
+ ret = -EFAULT;
+out:
+ return ret;
+}
+
+/*
+ * This code is really i386-only, but it detects and logs x86_64 GDT indexes
+ * if a 32-bit UML is running on a 64-bit host.
+ */
+static int __init __setup_host_supports_tls(void)
+{
+ check_host_supports_tls(&host_supports_tls, &host_gdt_entry_tls_min);
+ if (host_supports_tls) {
+ printk(KERN_INFO "Host TLS support detected\n");
+ printk(KERN_INFO "Detected host type: ");
+ switch (host_gdt_entry_tls_min) {
+ case GDT_ENTRY_TLS_MIN_I386:
+ printk(KERN_CONT "i386");
+ break;
+ case GDT_ENTRY_TLS_MIN_X86_64:
+ printk(KERN_CONT "x86_64");
+ break;
+ }
+ printk(KERN_CONT " (GDT indexes %d to %d)\n",
+ host_gdt_entry_tls_min,
+ host_gdt_entry_tls_min + GDT_ENTRY_TLS_ENTRIES);
+ } else
+ printk(KERN_ERR " Host TLS support NOT detected! "
+ "TLS support inside UML will not work\n");
+ return 0;
+}
+
+__initcall(__setup_host_supports_tls);
diff --git a/arch/x86/um/tls_64.c b/arch/x86/um/tls_64.c
new file mode 100644
index 000000000000..f7ba46200ecd
--- /dev/null
+++ b/arch/x86/um/tls_64.c
@@ -0,0 +1,17 @@
+#include "linux/sched.h"
+
+void clear_flushed_tls(struct task_struct *task)
+{
+}
+
+int arch_copy_tls(struct task_struct *t)
+{
+ /*
+ * If CLONE_SETTLS is set, we need to save the thread id
+ * (which is argument 5, child_tid, of clone) so it can be set
+ * during context switches.
+ */
+ t->thread.arch.fs = t->thread.regs.regs.gp[R8 / sizeof(long)];
+
+ return 0;
+}
diff --git a/arch/x86/um/user-offsets.c b/arch/x86/um/user-offsets.c
new file mode 100644
index 000000000000..3c19c48a1d48
--- /dev/null
+++ b/arch/x86/um/user-offsets.c
@@ -0,0 +1,79 @@
+#include <stdio.h>
+#include <stddef.h>
+#include <signal.h>
+#include <sys/poll.h>
+#include <sys/mman.h>
+#include <sys/user.h>
+#define __FRAME_OFFSETS
+#include <asm/ptrace.h>
+#include <asm/types.h>
+
+#define DEFINE(sym, val) \
+ asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+
+#define DEFINE_LONGS(sym, val) \
+ asm volatile("\n->" #sym " %0 " #val : : "i" (val/sizeof(unsigned long)))
+
+void foo(void)
+{
+#ifdef __i386__
+ DEFINE_LONGS(HOST_FP_SIZE, sizeof(struct user_fpregs_struct));
+ DEFINE_LONGS(HOST_FPX_SIZE, sizeof(struct user_fpxregs_struct));
+
+ DEFINE(HOST_IP, EIP);
+ DEFINE(HOST_SP, UESP);
+ DEFINE(HOST_EFLAGS, EFL);
+ DEFINE(HOST_EAX, EAX);
+ DEFINE(HOST_EBX, EBX);
+ DEFINE(HOST_ECX, ECX);
+ DEFINE(HOST_EDX, EDX);
+ DEFINE(HOST_ESI, ESI);
+ DEFINE(HOST_EDI, EDI);
+ DEFINE(HOST_EBP, EBP);
+ DEFINE(HOST_CS, CS);
+ DEFINE(HOST_SS, SS);
+ DEFINE(HOST_DS, DS);
+ DEFINE(HOST_FS, FS);
+ DEFINE(HOST_ES, ES);
+ DEFINE(HOST_GS, GS);
+#else
+ DEFINE(HOST_FP_SIZE, sizeof(struct _fpstate) / sizeof(unsigned long));
+ DEFINE_LONGS(HOST_RBX, RBX);
+ DEFINE_LONGS(HOST_RCX, RCX);
+ DEFINE_LONGS(HOST_RDI, RDI);
+ DEFINE_LONGS(HOST_RSI, RSI);
+ DEFINE_LONGS(HOST_RDX, RDX);
+ DEFINE_LONGS(HOST_RBP, RBP);
+ DEFINE_LONGS(HOST_RAX, RAX);
+ DEFINE_LONGS(HOST_R8, R8);
+ DEFINE_LONGS(HOST_R9, R9);
+ DEFINE_LONGS(HOST_R10, R10);
+ DEFINE_LONGS(HOST_R11, R11);
+ DEFINE_LONGS(HOST_R12, R12);
+ DEFINE_LONGS(HOST_R13, R13);
+ DEFINE_LONGS(HOST_R14, R14);
+ DEFINE_LONGS(HOST_R15, R15);
+ DEFINE_LONGS(HOST_ORIG_RAX, ORIG_RAX);
+ DEFINE_LONGS(HOST_CS, CS);
+ DEFINE_LONGS(HOST_SS, SS);
+ DEFINE_LONGS(HOST_EFLAGS, EFLAGS);
+#if 0
+ DEFINE_LONGS(HOST_FS, FS);
+ DEFINE_LONGS(HOST_GS, GS);
+ DEFINE_LONGS(HOST_DS, DS);
+ DEFINE_LONGS(HOST_ES, ES);
+#endif
+
+ DEFINE_LONGS(HOST_IP, RIP);
+ DEFINE_LONGS(HOST_SP, RSP);
+#endif
+
+ DEFINE(UM_FRAME_SIZE, sizeof(struct user_regs_struct));
+ DEFINE(UM_POLLIN, POLLIN);
+ DEFINE(UM_POLLPRI, POLLPRI);
+ DEFINE(UM_POLLOUT, POLLOUT);
+
+ DEFINE(UM_PROT_READ, PROT_READ);
+ DEFINE(UM_PROT_WRITE, PROT_WRITE);
+ DEFINE(UM_PROT_EXEC, PROT_EXEC);
+}
diff --git a/arch/x86/um/vdso/Makefile b/arch/x86/um/vdso/Makefile
new file mode 100644
index 000000000000..5dffe6d46686
--- /dev/null
+++ b/arch/x86/um/vdso/Makefile
@@ -0,0 +1,90 @@
+#
+# Building vDSO images for x86.
+#
+
+VDSO64-y := y
+
+vdso-install-$(VDSO64-y) += vdso.so
+
+
+# files to link into the vdso
+vobjs-y := vdso-note.o um_vdso.o
+
+# files to link into kernel
+obj-$(VDSO64-y) += vdso.o vma.o
+
+vobjs := $(foreach F,$(vobjs-y),$(obj)/$F)
+
+$(obj)/vdso.o: $(obj)/vdso.so
+
+targets += vdso.so vdso.so.dbg vdso.lds $(vobjs-y)
+
+export CPPFLAGS_vdso.lds += -P -C
+
+VDSO_LDFLAGS_vdso.lds = -m64 -Wl,-soname=linux-vdso.so.1 \
+ -Wl,-z,max-page-size=4096 -Wl,-z,common-page-size=4096
+
+$(obj)/vdso.o: $(src)/vdso.S $(obj)/vdso.so
+
+$(obj)/vdso.so.dbg: $(src)/vdso.lds $(vobjs) FORCE
+ $(call if_changed,vdso)
+
+$(obj)/%.so: OBJCOPYFLAGS := -S
+$(obj)/%.so: $(obj)/%.so.dbg FORCE
+ $(call if_changed,objcopy)
+
+#
+# Don't omit frame pointers for ease of userspace debugging, but do
+# optimize sibling calls.
+#
+CFL := $(PROFILING) -mcmodel=small -fPIC -O2 -fasynchronous-unwind-tables -m64 \
+ $(filter -g%,$(KBUILD_CFLAGS)) $(call cc-option, -fno-stack-protector) \
+ -fno-omit-frame-pointer -foptimize-sibling-calls
+
+$(vobjs): KBUILD_CFLAGS += $(CFL)
+
+#
+# vDSO code runs in userspace and -pg doesn't help with profiling anyway.
+#
+CFLAGS_REMOVE_vdso-note.o = -pg
+CFLAGS_REMOVE_um_vdso.o = -pg
+
+targets += vdso-syms.lds
+obj-$(VDSO64-y) += vdso-syms.lds
+
+#
+# Match symbols in the DSO that look like VDSO*; produce a file of constants.
+#
+sed-vdsosym := -e 's/^00*/0/' \
+ -e 's/^\([0-9a-fA-F]*\) . \(VDSO[a-zA-Z0-9_]*\)$$/\2 = 0x\1;/p'
+quiet_cmd_vdsosym = VDSOSYM $@
+define cmd_vdsosym
+ $(NM) $< | LC_ALL=C sed -n $(sed-vdsosym) | LC_ALL=C sort > $@
+endef
+
+$(obj)/%-syms.lds: $(obj)/%.so.dbg FORCE
+ $(call if_changed,vdsosym)
+
+#
+# The DSO images are built using a special linker script.
+#
+quiet_cmd_vdso = VDSO $@
+ cmd_vdso = $(CC) -nostdlib -o $@ \
+ $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \
+ -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) && \
+ sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@'
+
+VDSO_LDFLAGS = -fPIC -shared $(call cc-ldoption, -Wl$(comma)--hash-style=sysv)
+GCOV_PROFILE := n
+
+#
+# Install the unstripped copy of vdso*.so listed in $(vdso-install-y).
+#
+quiet_cmd_vdso_install = INSTALL $@
+ cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@
+$(vdso-install-y): %.so: $(obj)/%.so.dbg FORCE
+ @mkdir -p $(MODLIB)/vdso
+ $(call cmd,vdso_install)
+
+PHONY += vdso_install $(vdso-install-y)
+vdso_install: $(vdso-install-y)
diff --git a/arch/x86/um/vdso/checkundef.sh b/arch/x86/um/vdso/checkundef.sh
new file mode 100644
index 000000000000..7ee90a9b549d
--- /dev/null
+++ b/arch/x86/um/vdso/checkundef.sh
@@ -0,0 +1,10 @@
+#!/bin/sh
+nm="$1"
+file="$2"
+$nm "$file" | grep '^ *U' > /dev/null 2>&1
+if [ $? -eq 1 ]; then
+ exit 0
+else
+ echo "$file: undefined symbols found" >&2
+ exit 1
+fi
diff --git a/arch/x86/um/vdso/um_vdso.c b/arch/x86/um/vdso/um_vdso.c
new file mode 100644
index 000000000000..7c441b59d375
--- /dev/null
+++ b/arch/x86/um/vdso/um_vdso.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright (C) 2011 Richard Weinberger <richrd@nod.at>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This vDSO turns all calls into a syscall so that UML can trap them.
+ */
+
+
+/* Disable profiling for userspace code */
+#define DISABLE_BRANCH_PROFILING
+
+#include <linux/time.h>
+#include <linux/getcpu.h>
+#include <asm/unistd.h>
+
+int __vdso_clock_gettime(clockid_t clock, struct timespec *ts)
+{
+ long ret;
+
+ asm("syscall" : "=a" (ret) :
+ "0" (__NR_clock_gettime), "D" (clock), "S" (ts) : "memory");
+
+ return ret;
+}
+int clock_gettime(clockid_t, struct timespec *)
+ __attribute__((weak, alias("__vdso_clock_gettime")));
+
+int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz)
+{
+ long ret;
+
+ asm("syscall" : "=a" (ret) :
+ "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "memory");
+
+ return ret;
+}
+int gettimeofday(struct timeval *, struct timezone *)
+ __attribute__((weak, alias("__vdso_gettimeofday")));
+
+time_t __vdso_time(time_t *t)
+{
+ long secs;
+
+ asm volatile("syscall"
+ : "=a" (secs)
+ : "0" (__NR_time), "D" (t) : "cc", "r11", "cx", "memory");
+
+ return secs;
+}
+int time(time_t *t) __attribute__((weak, alias("__vdso_time")));
+
+long
+__vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused)
+{
+ /*
+ * UML does not support SMP, we can cheat here. :)
+ */
+
+ if (cpu)
+ *cpu = 0;
+ if (node)
+ *node = 0;
+
+ return 0;
+}
+
+long getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
+ __attribute__((weak, alias("__vdso_getcpu")));
diff --git a/arch/x86/um/vdso/vdso-layout.lds.S b/arch/x86/um/vdso/vdso-layout.lds.S
new file mode 100644
index 000000000000..634a2cf62046
--- /dev/null
+++ b/arch/x86/um/vdso/vdso-layout.lds.S
@@ -0,0 +1,64 @@
+/*
+ * Linker script for vDSO. This is an ELF shared object prelinked to
+ * its virtual address, and with only one read-only segment.
+ * This script controls its layout.
+ */
+
+SECTIONS
+{
+ . = VDSO_PRELINK + SIZEOF_HEADERS;
+
+ .hash : { *(.hash) } :text
+ .gnu.hash : { *(.gnu.hash) }
+ .dynsym : { *(.dynsym) }
+ .dynstr : { *(.dynstr) }
+ .gnu.version : { *(.gnu.version) }
+ .gnu.version_d : { *(.gnu.version_d) }
+ .gnu.version_r : { *(.gnu.version_r) }
+
+ .note : { *(.note.*) } :text :note
+
+ .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr
+ .eh_frame : { KEEP (*(.eh_frame)) } :text
+
+ .dynamic : { *(.dynamic) } :text :dynamic
+
+ .rodata : { *(.rodata*) } :text
+ .data : {
+ *(.data*)
+ *(.sdata*)
+ *(.got.plt) *(.got)
+ *(.gnu.linkonce.d.*)
+ *(.bss*)
+ *(.dynbss*)
+ *(.gnu.linkonce.b.*)
+ }
+
+ .altinstructions : { *(.altinstructions) }
+ .altinstr_replacement : { *(.altinstr_replacement) }
+
+ /*
+ * Align the actual code well away from the non-instruction data.
+ * This is the best thing for the I-cache.
+ */
+ . = ALIGN(0x100);
+
+ .text : { *(.text*) } :text =0x90909090
+}
+
+/*
+ * Very old versions of ld do not recognize this name token; use the constant.
+ */
+#define PT_GNU_EH_FRAME 0x6474e550
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+ text PT_LOAD FLAGS(5) FILEHDR PHDRS; /* PF_R|PF_X */
+ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
+ note PT_NOTE FLAGS(4); /* PF_R */
+ eh_frame_hdr PT_GNU_EH_FRAME;
+}
diff --git a/arch/x86/um/vdso/vdso-note.S b/arch/x86/um/vdso/vdso-note.S
new file mode 100644
index 000000000000..79a071e4357e
--- /dev/null
+++ b/arch/x86/um/vdso/vdso-note.S
@@ -0,0 +1,12 @@
+/*
+ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
+ * Here we can supply some information useful to userland.
+ */
+
+#include <linux/uts.h>
+#include <linux/version.h>
+#include <linux/elfnote.h>
+
+ELFNOTE_START(Linux, 0, "a")
+ .long LINUX_VERSION_CODE
+ELFNOTE_END
diff --git a/arch/x86/um/vdso/vdso.S b/arch/x86/um/vdso/vdso.S
new file mode 100644
index 000000000000..1cb468adacbb
--- /dev/null
+++ b/arch/x86/um/vdso/vdso.S
@@ -0,0 +1,10 @@
+#include <linux/init.h>
+
+__INITDATA
+
+ .globl vdso_start, vdso_end
+vdso_start:
+ .incbin "arch/x86/um/vdso/vdso.so"
+vdso_end:
+
+__FINIT
diff --git a/arch/x86/um/vdso/vdso.lds.S b/arch/x86/um/vdso/vdso.lds.S
new file mode 100644
index 000000000000..b96b2677cad8
--- /dev/null
+++ b/arch/x86/um/vdso/vdso.lds.S
@@ -0,0 +1,32 @@
+/*
+ * Linker script for 64-bit vDSO.
+ * We #include the file to define the layout details.
+ * Here we only choose the prelinked virtual address.
+ *
+ * This file defines the version script giving the user-exported symbols in
+ * the DSO. We can define local symbols here called VDSO* to make their
+ * values visible using the asm-x86/vdso.h macros from the kernel proper.
+ */
+
+#define VDSO_PRELINK 0xffffffffff700000
+#include "vdso-layout.lds.S"
+
+/*
+ * This controls what userland symbols we export from the vDSO.
+ */
+VERSION {
+ LINUX_2.6 {
+ global:
+ clock_gettime;
+ __vdso_clock_gettime;
+ gettimeofday;
+ __vdso_gettimeofday;
+ getcpu;
+ __vdso_getcpu;
+ time;
+ __vdso_time;
+ local: *;
+ };
+}
+
+VDSO64_PRELINK = VDSO_PRELINK;
diff --git a/arch/x86/um/vdso/vma.c b/arch/x86/um/vdso/vma.c
new file mode 100644
index 000000000000..9495c8d0ce37
--- /dev/null
+++ b/arch/x86/um/vdso/vma.c
@@ -0,0 +1,74 @@
+/*
+ * Copyright (C) 2011 Richard Weinberger <richrd@nod.at>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <asm/page.h>
+#include <linux/init.h>
+
+unsigned int __read_mostly vdso_enabled = 1;
+unsigned long um_vdso_addr;
+
+extern unsigned long task_size;
+extern char vdso_start[], vdso_end[];
+
+static struct page **vdsop;
+
+static int __init init_vdso(void)
+{
+ struct page *um_vdso;
+
+ BUG_ON(vdso_end - vdso_start > PAGE_SIZE);
+
+ um_vdso_addr = task_size - PAGE_SIZE;
+
+ vdsop = kmalloc(GFP_KERNEL, sizeof(struct page *));
+ if (!vdsop)
+ goto oom;
+
+ um_vdso = alloc_page(GFP_KERNEL);
+ if (!um_vdso) {
+ kfree(vdsop);
+
+ goto oom;
+ }
+
+ copy_page(page_address(um_vdso), vdso_start);
+ *vdsop = um_vdso;
+
+ return 0;
+
+oom:
+ printk(KERN_ERR "Cannot allocate vdso\n");
+ vdso_enabled = 0;
+
+ return -ENOMEM;
+}
+subsys_initcall(init_vdso);
+
+int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
+{
+ int err;
+ struct mm_struct *mm = current->mm;
+
+ if (!vdso_enabled)
+ return 0;
+
+ down_write(&mm->mmap_sem);
+
+ err = install_special_mapping(mm, um_vdso_addr, PAGE_SIZE,
+ VM_READ|VM_EXEC|
+ VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC|
+ VM_ALWAYSDUMP,
+ vdsop);
+
+ up_write(&mm->mmap_sem);
+
+ return err;
+}