23 files changed, 270 insertions, 374 deletions

diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig
index 875d008828b8..ea7f61e8bc9e 100644
--- a/arch/tile/Kconfig
+++ b/arch/tile/Kconfig
@@ -21,6 +21,8 @@ config TILE
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	select GENERIC_CLOCKEVENTS
 	select MODULES_USE_ELF_RELA
+	select GENERIC_KERNEL_THREAD
+	select GENERIC_KERNEL_EXECVE
 
 # FIXME: investigate whether we need/want these options.
 #	select HAVE_IOREMAP_PROT
diff --git a/arch/tile/include/asm/Kbuild b/arch/tile/include/asm/Kbuild
index 6948015e08a2..b17b9b8e53cd 100644
--- a/arch/tile/include/asm/Kbuild
+++ b/arch/tile/include/asm/Kbuild
@@ -34,5 +34,6 @@ generic-y += sockios.h
 generic-y += statfs.h
 generic-y += termbits.h
 generic-y += termios.h
+generic-y += trace_clock.h
 generic-y += types.h
 generic-y += xor.h
diff --git a/arch/tile/include/asm/compat.h b/arch/tile/include/asm/compat.h
index 3063e6fc8daa..88f3c227afd9 100644
--- a/arch/tile/include/asm/compat.h
+++ b/arch/tile/include/asm/compat.h
@@ -275,18 +275,14 @@ extern int compat_setup_rt_frame(int sig, struct k_sigaction *ka,
 struct compat_sigaction;
 struct compat_siginfo;
 struct compat_sigaltstack;
-long compat_sys_execve(const char __user *path,
-		       compat_uptr_t __user *argv,
-		       compat_uptr_t __user *envp, struct pt_regs *);
 long compat_sys_rt_sigaction(int sig, struct compat_sigaction __user *act,
 			     struct compat_sigaction __user *oact,
 			     size_t sigsetsize);
 long compat_sys_rt_sigqueueinfo(int pid, int sig,
 				struct compat_siginfo __user *uinfo);
-long compat_sys_rt_sigreturn(struct pt_regs *);
+long compat_sys_rt_sigreturn(void);
 long compat_sys_sigaltstack(const struct compat_sigaltstack __user *uss_ptr,
-			    struct compat_sigaltstack __user *uoss_ptr,
-			    struct pt_regs *);
+			    struct compat_sigaltstack __user *uoss_ptr);
 long compat_sys_truncate64(char __user *filename, u32 dummy, u32 low, u32 high);
 long compat_sys_ftruncate64(unsigned int fd, u32 dummy, u32 low, u32 high);
 long compat_sys_pread64(unsigned int fd, char __user *ubuf, size_t count,
@@ -300,15 +296,8 @@ long compat_sys_sync_file_range2(int fd, unsigned int flags,
 long compat_sys_fallocate(int fd, int mode,
 			  u32 offset_lo, u32 offset_hi,
 			  u32 len_lo, u32 len_hi);
-long compat_sys_sched_rr_get_interval(compat_pid_t pid,
-				      struct compat_timespec __user *interval);
-
-/* These are the intvec_64.S trampolines. */
-long _compat_sys_execve(const char __user *path,
-			const compat_uptr_t __user *argv,
-			const compat_uptr_t __user *envp);
-long _compat_sys_sigaltstack(const struct compat_sigaltstack __user *uss_ptr,
-			    struct compat_sigaltstack __user *uoss_ptr);
+
+/* Assembly trampoline to avoid clobbering r0. */
 long _compat_sys_rt_sigreturn(void);
 
 #endif /* _ASM_TILE_COMPAT_H */
diff --git a/arch/tile/include/asm/elf.h b/arch/tile/include/asm/elf.h
index f8ccf08f6934..ff8a93408823 100644
--- a/arch/tile/include/asm/elf.h
+++ b/arch/tile/include/asm/elf.h
@@ -148,6 +148,7 @@ extern int arch_setup_additional_pages(struct linux_binprm *bprm,
 #define compat_start_thread(regs, ip, usp) do { \
 		regs->pc = ptr_to_compat_reg((void *)(ip)); \
 		regs->sp = ptr_to_compat_reg((void *)(usp)); \
+		single_step_execve();	\
 	} while (0)
 
 /*
@@ -169,4 +170,6 @@ do { \
 
 #endif /* CONFIG_COMPAT */
 
+#define CORE_DUMP_USE_REGSET
+
 #endif /* _ASM_TILE_ELF_H */
diff --git a/arch/tile/include/asm/processor.h b/arch/tile/include/asm/processor.h
index 8c4dd9ff91eb..2b70dfb1442e 100644
--- a/arch/tile/include/asm/processor.h
+++ b/arch/tile/include/asm/processor.h
@@ -211,6 +211,7 @@ static inline void start_thread(struct pt_regs *regs,
 {
 	regs->pc = pc;
 	regs->sp = usp;
+	single_step_execve();
 }
 
 /* Free all resources held by a thread. */
@@ -219,8 +220,6 @@ static inline void release_thread(struct task_struct *dead_task)
 	/* Nothing for now */
 }
 
-extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
-
 extern int do_work_pending(struct pt_regs *regs, u32 flags);
 
 
@@ -239,6 +238,9 @@ unsigned long get_wchan(struct task_struct *p);
 #define KSTK_TOP(task)	(task_ksp0(task) - STACK_TOP_DELTA)
 #define task_pt_regs(task) \
   ((struct pt_regs *)(task_ksp0(task) - KSTK_PTREGS_GAP) - 1)
+#define current_pt_regs()                                   \
+  ((struct pt_regs *)((stack_pointer | (THREAD_SIZE - 1)) - \
+                      (KSTK_PTREGS_GAP - 1)) - 1)
 #define task_sp(task)	(task_pt_regs(task)->sp)
 #define task_pc(task)	(task_pt_regs(task)->pc)
 /* Aliases for pc and sp (used in fs/proc/array.c) */
diff --git a/arch/tile/include/asm/ptrace.h b/arch/tile/include/asm/ptrace.h
index 1a4fd9ab0ee1..5ce052e16b7b 100644
--- a/arch/tile/include/asm/ptrace.h
+++ b/arch/tile/include/asm/ptrace.h
@@ -24,8 +24,7 @@ typedef unsigned long pt_reg_t;
 #include <uapi/asm/ptrace.h>
 
 #define PTRACE_O_MASK_TILE	(PTRACE_O_TRACEMIGRATE)
-#define PT_TRACE_MIGRATE	0x00080000
-#define PT_TRACE_MASK_TILE	(PT_TRACE_MIGRATE)
+#define PT_TRACE_MIGRATE	PT_EVENT_FLAG(PTRACE_EVENT_MIGRATE)
 
 /* Flag bits in pt_regs.flags */
 #define PT_FLAGS_DISABLE_IRQ    1  /* on return to kernel, disable irqs */
diff --git a/arch/tile/include/asm/switch_to.h b/arch/tile/include/asm/switch_to.h
index 1d48c5fee8b7..b8f888cbe6b0 100644
--- a/arch/tile/include/asm/switch_to.h
+++ b/arch/tile/include/asm/switch_to.h
@@ -68,7 +68,10 @@ extern unsigned long get_switch_to_pc(void);
 /* Support function for forking a new task. */
 void ret_from_fork(void);
 
-/* Called from ret_from_fork() when a new process starts up. */
+/* Support function for forking a new kernel thread. */
+void ret_from_kernel_thread(void *fn, void *arg);
+
+/* Called from ret_from_xxx() when a new process starts up. */
 struct task_struct *sim_notify_fork(struct task_struct *prev);
 
 #endif /* !__ASSEMBLY__ */
diff --git a/arch/tile/include/asm/syscalls.h b/arch/tile/include/asm/syscalls.h
index 06f0464cfed9..4c8462a62cb6 100644
--- a/arch/tile/include/asm/syscalls.h
+++ b/arch/tile/include/asm/syscalls.h
@@ -51,8 +51,7 @@ long sys_cacheflush(unsigned long addr, unsigned long len,
 
 #ifndef __tilegx__
 /* mm/fault.c */
-long sys_cmpxchg_badaddr(unsigned long address, struct pt_regs *);
-long _sys_cmpxchg_badaddr(unsigned long address);
+long sys_cmpxchg_badaddr(unsigned long address);
 #endif
 
 #ifdef CONFIG_COMPAT
@@ -63,14 +62,16 @@ long sys_truncate64(const char __user *path, loff_t length);
 long sys_ftruncate64(unsigned int fd, loff_t length);
 #endif
 
+/* Provide versions of standard syscalls that use current_pt_regs(). */
+long sys_rt_sigreturn(void);
+long sys_sigaltstack(const stack_t __user *, stack_t __user *);
+#define sys_rt_sigreturn sys_rt_sigreturn
+#define sys_sigaltstack sys_sigaltstack
+
 /* These are the intvec*.S trampolines. */
-long _sys_sigaltstack(const stack_t __user *, stack_t __user *);
 long _sys_rt_sigreturn(void);
 long _sys_clone(unsigned long clone_flags, unsigned long newsp,
 		void __user *parent_tid, void __user *child_tid);
-long _sys_execve(const char __user *filename,
-		 const char __user *const __user *argv,
-		 const char __user *const __user *envp);
 
 #include <asm-generic/syscalls.h>
 
diff --git a/arch/tile/include/asm/unistd.h b/arch/tile/include/asm/unistd.h
index 6e032a0a268e..fe841e7d4963 100644
--- a/arch/tile/include/asm/unistd.h
+++ b/arch/tile/include/asm/unistd.h
@@ -14,6 +14,9 @@
 /* In compat mode, we use sys_llseek() for compat_sys_llseek(). */
 #ifdef CONFIG_COMPAT
 #define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_COMPAT_SYS_SCHED_RR_GET_INTERVAL
 #endif
 #define __ARCH_WANT_SYS_NEWFSTATAT
+#define __ARCH_WANT_SYS_EXECVE
+#define __ARCH_WANT_SYS_CLONE
 #include <uapi/asm/unistd.h>
diff --git a/arch/tile/include/uapi/asm/ptrace.h b/arch/tile/include/uapi/asm/ptrace.h
index c717d0fec72e..7757e1985fb6 100644
--- a/arch/tile/include/uapi/asm/ptrace.h
+++ b/arch/tile/include/uapi/asm/ptrace.h
@@ -81,8 +81,14 @@ struct pt_regs {
 #define PTRACE_SETFPREGS	15
 
 /* Support TILE-specific ptrace options, with events starting at 16. */
-#define PTRACE_O_TRACEMIGRATE	0x00010000
 #define PTRACE_EVENT_MIGRATE	16
+#define PTRACE_O_TRACEMIGRATE	(1 << PTRACE_EVENT_MIGRATE)
 
+/*
+ * Flag bits in pt_regs.flags that are part of the ptrace API.
+ * We start our numbering higher up to avoid confusion with the
+ * non-ABI kernel-internal values that use the low 16 bits.
+ */
+#define PT_FLAGS_COMPAT		0x10000  /* process is an -m32 compat process */
 
 #endif /* _UAPI_ASM_TILE_PTRACE_H */
diff --git a/arch/tile/kernel/compat.c b/arch/tile/kernel/compat.c
index d67459b9ac2a..7f72401b4f45 100644
--- a/arch/tile/kernel/compat.c
+++ b/arch/tile/kernel/compat.c
@@ -76,24 +76,6 @@ long compat_sys_fallocate(int fd, int mode,
 			     ((loff_t)len_hi << 32) | len_lo);
 }
 
-
-
-long compat_sys_sched_rr_get_interval(compat_pid_t pid,
-				      struct compat_timespec __user *interval)
-{
-	struct timespec t;
-	int ret;
-	mm_segment_t old_fs = get_fs();
-
-	set_fs(KERNEL_DS);
-	ret = sys_sched_rr_get_interval(pid,
-					(struct timespec __force __user *)&t);
-	set_fs(old_fs);
-	if (put_compat_timespec(&t, interval))
-		return -EFAULT;
-	return ret;
-}
-
 /* Provide the compat syscall number to call mapping. */
 #undef __SYSCALL
 #define __SYSCALL(nr, call) [nr] = (call),
@@ -102,9 +84,7 @@ long compat_sys_sched_rr_get_interval(compat_pid_t pid,
 #define compat_sys_fadvise64_64 sys32_fadvise64_64
 #define compat_sys_readahead sys32_readahead
 
-/* Call the trampolines to manage pt_regs where necessary. */
-#define compat_sys_execve _compat_sys_execve
-#define compat_sys_sigaltstack _compat_sys_sigaltstack
+/* Call the assembly trampolines where necessary. */
 #define compat_sys_rt_sigreturn _compat_sys_rt_sigreturn
 #define sys_clone _sys_clone
 
diff --git a/arch/tile/kernel/compat_signal.c b/arch/tile/kernel/compat_signal.c
index 08b4fe1717bb..2e4cc69224a6 100644
--- a/arch/tile/kernel/compat_signal.c
+++ b/arch/tile/kernel/compat_signal.c
@@ -197,8 +197,7 @@ int copy_siginfo_from_user32(siginfo_t *to, struct compat_siginfo __user *from)
 }
 
 long compat_sys_sigaltstack(const struct compat_sigaltstack __user *uss_ptr,
-			    struct compat_sigaltstack __user *uoss_ptr,
-			    struct pt_regs *regs)
+			    struct compat_sigaltstack __user *uoss_ptr)
 {
 	stack_t uss, uoss;
 	int ret;
@@ -219,7 +218,7 @@ long compat_sys_sigaltstack(const struct compat_sigaltstack __user *uss_ptr,
 	set_fs(KERNEL_DS);
 	ret = do_sigaltstack(uss_ptr ? (stack_t __user __force *)&uss : NULL,
 			     (stack_t __user __force *)&uoss,
-			     (unsigned long)compat_ptr(regs->sp));
+			     (unsigned long)compat_ptr(current_pt_regs()->sp));
 	set_fs(seg);
 	if (ret >= 0 && uoss_ptr)  {
 		if (!access_ok(VERIFY_WRITE, uoss_ptr, sizeof(*uoss_ptr)) ||
@@ -232,8 +231,9 @@ long compat_sys_sigaltstack(const struct compat_sigaltstack __user *uss_ptr,
 }
 
 /* The assembly shim for this function arranges to ignore the return value. */
-long compat_sys_rt_sigreturn(struct pt_regs *regs)
+long compat_sys_rt_sigreturn(void)
 {
+	struct pt_regs *regs = current_pt_regs();
 	struct compat_rt_sigframe __user *frame =
 		(struct compat_rt_sigframe __user *) compat_ptr(regs->sp);
 	sigset_t set;
@@ -248,7 +248,7 @@ long compat_sys_rt_sigreturn(struct pt_regs *regs)
 	if (restore_sigcontext(regs, &frame->uc.uc_mcontext))
 		goto badframe;
 
-	if (compat_sys_sigaltstack(&frame->uc.uc_stack, NULL, regs) != 0)
+	if (compat_sys_sigaltstack(&frame->uc.uc_stack, NULL) == -EFAULT)
 		goto badframe;
 
 	return 0;
diff --git a/arch/tile/kernel/entry.S b/arch/tile/kernel/entry.S
index c31637baff28..f116cb0bce20 100644
--- a/arch/tile/kernel/entry.S
+++ b/arch/tile/kernel/entry.S
@@ -28,17 +28,6 @@ STD_ENTRY(current_text_addr)
 	STD_ENDPROC(current_text_addr)
 
 /*
- * Implement execve().  The i386 code has a note that forking from kernel
- * space results in no copy on write until the execve, so we should be
- * careful not to write to the stack here.
- */
-STD_ENTRY(kernel_execve)
-	moveli TREG_SYSCALL_NR_NAME, __NR_execve
-	swint1
-	jrp lr
-	STD_ENDPROC(kernel_execve)
-
-/*
  * We don't run this function directly, but instead copy it to a page
  * we map into every user process.  See vdso_setup().
  *
diff --git a/arch/tile/kernel/intvec_32.S b/arch/tile/kernel/intvec_32.S
index 6943515100f8..f212bf7cea86 100644
--- a/arch/tile/kernel/intvec_32.S
+++ b/arch/tile/kernel/intvec_32.S
@@ -1291,6 +1291,21 @@ STD_ENTRY(ret_from_fork)
 	}
 	STD_ENDPROC(ret_from_fork)
 
+STD_ENTRY(ret_from_kernel_thread)
+	jal     sim_notify_fork
+	jal     schedule_tail
+	FEEDBACK_REENTER(ret_from_fork)
+	{
+	 move   r0, r31
+	 jalr   r30
+	}
+	FEEDBACK_REENTER(ret_from_kernel_thread)
+	{
+	 movei  r30, 0               /* not an NMI */
+	 j      .Lresume_userspace   /* jump into middle of interrupt_return */
+	}
+	STD_ENDPROC(ret_from_kernel_thread)
+
 	/*
 	 * Code for ill interrupt.
 	 */
@@ -1437,15 +1452,6 @@ STD_ENTRY_LOCAL(bad_intr)
 	panic   "Unhandled interrupt %#x: PC %#lx"
 	STD_ENDPROC(bad_intr)
 
-/* Put address of pt_regs in reg and jump. */
-#define PTREGS_SYSCALL(x, reg)                          \
-	STD_ENTRY(_##x);                                \
-	{                                               \
-	 PTREGS_PTR(reg, PTREGS_OFFSET_BASE);           \
-	 j      x                                       \
-	};                                              \
-	STD_ENDPROC(_##x)
-
 /*
  * Special-case sigreturn to not write r0 to the stack on return.
  * This is technically more efficient, but it also avoids difficulties
@@ -1461,12 +1467,9 @@ STD_ENTRY_LOCAL(bad_intr)
 	};                                              \
 	STD_ENDPROC(_##x)
 
-PTREGS_SYSCALL(sys_execve, r3)
-PTREGS_SYSCALL(sys_sigaltstack, r2)
 PTREGS_SYSCALL_SIGRETURN(sys_rt_sigreturn, r0)
-PTREGS_SYSCALL(sys_cmpxchg_badaddr, r1)
 
-/* Save additional callee-saves to pt_regs, put address in r4 and jump. */
+/* Save additional callee-saves to pt_regs and jump to standard function. */
 STD_ENTRY(_sys_clone)
 	push_extra_callee_saves r4
 	j       sys_clone
diff --git a/arch/tile/kernel/intvec_64.S b/arch/tile/kernel/intvec_64.S
index 7c06d597ffd0..54bc9a6678e8 100644
--- a/arch/tile/kernel/intvec_64.S
+++ b/arch/tile/kernel/intvec_64.S
@@ -1150,6 +1150,21 @@ STD_ENTRY(ret_from_fork)
 	}
 	STD_ENDPROC(ret_from_fork)
 
+STD_ENTRY(ret_from_kernel_thread)
+	jal     sim_notify_fork
+	jal     schedule_tail
+	FEEDBACK_REENTER(ret_from_fork)
+	{
+	 move   r0, r31
+	 jalr   r30
+	}
+	FEEDBACK_REENTER(ret_from_kernel_thread)
+	{
+	 movei  r30, 0               /* not an NMI */
+	 j      .Lresume_userspace   /* jump into middle of interrupt_return */
+	}
+	STD_ENDPROC(ret_from_kernel_thread)
+
 /* Various stub interrupt handlers and syscall handlers */
 
 STD_ENTRY_LOCAL(_kernel_double_fault)
@@ -1166,15 +1181,6 @@ STD_ENTRY_LOCAL(bad_intr)
 	panic   "Unhandled interrupt %#x: PC %#lx"
 	STD_ENDPROC(bad_intr)
 
-/* Put address of pt_regs in reg and jump. */
-#define PTREGS_SYSCALL(x, reg)                          \
-	STD_ENTRY(_##x);                                \
-	{                                               \
-	 PTREGS_PTR(reg, PTREGS_OFFSET_BASE);           \
-	 j      x                                       \
-	};                                              \
-	STD_ENDPROC(_##x)
-
 /*
  * Special-case sigreturn to not write r0 to the stack on return.
  * This is technically more efficient, but it also avoids difficulties
@@ -1190,16 +1196,12 @@ STD_ENTRY_LOCAL(bad_intr)
 	};                                              \
 	STD_ENDPROC(_##x)
 
-PTREGS_SYSCALL(sys_execve, r3)
-PTREGS_SYSCALL(sys_sigaltstack, r2)
 PTREGS_SYSCALL_SIGRETURN(sys_rt_sigreturn, r0)
 #ifdef CONFIG_COMPAT
-PTREGS_SYSCALL(compat_sys_execve, r3)
-PTREGS_SYSCALL(compat_sys_sigaltstack, r2)
 PTREGS_SYSCALL_SIGRETURN(compat_sys_rt_sigreturn, r0)
 #endif
 
-/* Save additional callee-saves to pt_regs, put address in r4 and jump. */
+/* Save additional callee-saves to pt_regs and jump to standard function. */
 STD_ENTRY(_sys_clone)
 	push_extra_callee_saves r4
 	j       sys_clone
diff --git a/arch/tile/kernel/pci.c b/arch/tile/kernel/pci.c
index 759822687e8f..aac1cd586966 100644
--- a/arch/tile/kernel/pci.c
+++ b/arch/tile/kernel/pci.c
@@ -245,7 +245,7 @@ static void __devinit fixup_read_and_payload_sizes(void)
 	u16 new_values;
 
 	/* Scan for the smallest maximum payload size. */
-	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+	for_each_pci_dev(dev) {
 		u32 devcap;
 		int max_payload;
 
@@ -260,7 +260,7 @@ static void __devinit fixup_read_and_payload_sizes(void)
 
 	/* Now, set the max_payload_size for all devices to that value. */
 	new_values = (max_read_size << 12) | (smallest_max_payload << 5);
-	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL)
+	for_each_pci_dev(dev)
 		pcie_capability_clear_and_set_word(dev, PCI_EXP_DEVCTL,
 				PCI_EXP_DEVCTL_PAYLOAD | PCI_EXP_DEVCTL_READRQ,
 				new_values);
diff --git a/arch/tile/kernel/pci_gx.c b/arch/tile/kernel/pci_gx.c
index 2ba6d052f85d..94810d4a6332 100644
--- a/arch/tile/kernel/pci_gx.c
+++ b/arch/tile/kernel/pci_gx.c
@@ -1047,8 +1047,7 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
 }
 
 /* Called for each device after PCI setup is done. */
-static void __init
-pcibios_fixup_final(struct pci_dev *pdev)
+static void pcibios_fixup_final(struct pci_dev *pdev)
 {
 	set_dma_ops(&pdev->dev, gx_pci_dma_map_ops);
 	set_dma_offset(&pdev->dev, TILE_PCI_MEM_MAP_BASE_OFFSET);
diff --git a/arch/tile/kernel/process.c b/arch/tile/kernel/process.c
index 307d010696c9..0e5661e7d00d 100644
--- a/arch/tile/kernel/process.c
+++ b/arch/tile/kernel/process.c
@@ -157,24 +157,43 @@ void arch_release_thread_info(struct thread_info *info)
 static void save_arch_state(struct thread_struct *t);
 
 int copy_thread(unsigned long clone_flags, unsigned long sp,
-		unsigned long stack_size,
-		struct task_struct *p, struct pt_regs *regs)
+		unsigned long arg, struct task_struct *p)
 {
-	struct pt_regs *childregs;
+	struct pt_regs *childregs = task_pt_regs(p), *regs = current_pt_regs();
 	unsigned long ksp;
+	unsigned long *callee_regs;
 
 	/*
-	 * When creating a new kernel thread we pass sp as zero.
-	 * Assign it to a reasonable value now that we have the stack.
+	 * Set up the stack and stack pointer appropriately for the
+	 * new child to find itself woken up in __switch_to().
+	 * The callee-saved registers must be on the stack to be read;
+	 * the new task will then jump to assembly support to handle
+	 * calling schedule_tail(), etc., and (for userspace tasks)
+	 * returning to the context set up in the pt_regs.
 	 */
-	if (sp == 0 && regs->ex1 == PL_ICS_EX1(KERNEL_PL, 0))
-		sp = KSTK_TOP(p);
+	ksp = (unsigned long) childregs;
+	ksp -= C_ABI_SAVE_AREA_SIZE;   /* interrupt-entry save area */
+	((long *)ksp)[0] = ((long *)ksp)[1] = 0;
+	ksp -= CALLEE_SAVED_REGS_COUNT * sizeof(unsigned long);
+	callee_regs = (unsigned long *)ksp;
+	ksp -= C_ABI_SAVE_AREA_SIZE;   /* __switch_to() save area */
+	((long *)ksp)[0] = ((long *)ksp)[1] = 0;
+	p->thread.ksp = ksp;
 
-	/*
-	 * Do not clone step state from the parent; each thread
-	 * must make its own lazily.
-	 */
-	task_thread_info(p)->step_state = NULL;
+	/* Record the pid of the task that created this one. */
+	p->thread.creator_pid = current->pid;
+
+	if (unlikely(p->flags & PF_KTHREAD)) {
+		/* kernel thread */
+		memset(childregs, 0, sizeof(struct pt_regs));
+		memset(&callee_regs[2], 0,
+		       (CALLEE_SAVED_REGS_COUNT - 2) * sizeof(unsigned long));
+		callee_regs[0] = sp;   /* r30 = function */
+		callee_regs[1] = arg;  /* r31 = arg */
+		childregs->ex1 = PL_ICS_EX1(KERNEL_PL, 0);
+		p->thread.pc = (unsigned long) ret_from_kernel_thread;
+		return 0;
+	}
 
 	/*
 	 * Start new thread in ret_from_fork so it schedules properly
@@ -182,46 +201,33 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
 	 */
 	p->thread.pc = (unsigned long) ret_from_fork;
 
-	/* Save user stack top pointer so we can ID the stack vm area later. */
-	p->thread.usp0 = sp;
-
-	/* Record the pid of the process that created this one. */
-	p->thread.creator_pid = current->pid;
+	/*
+	 * Do not clone step state from the parent; each thread
+	 * must make its own lazily.
+	 */
+	task_thread_info(p)->step_state = NULL;
 
 	/*
 	 * Copy the registers onto the kernel stack so the
 	 * return-from-interrupt code will reload it into registers.
 	 */
-	childregs = task_pt_regs(p);
-	*childregs = *regs;
+	*childregs = *current_pt_regs();
 	childregs->regs[0] = 0;         /* return value is zero */
-	childregs->sp = sp;  /* override with new user stack pointer */
+	if (sp)
+		childregs->sp = sp;  /* override with new user stack pointer */
+	memcpy(callee_regs, &childregs->regs[CALLEE_SAVED_FIRST_REG],
+	       CALLEE_SAVED_REGS_COUNT * sizeof(unsigned long));
+
+	/* Save user stack top pointer so we can ID the stack vm area later. */
+	p->thread.usp0 = childregs->sp;
 
 	/*
 	 * If CLONE_SETTLS is set, set "tp" in the new task to "r4",
 	 * which is passed in as arg #5 to sys_clone().
 	 */
 	if (clone_flags & CLONE_SETTLS)
-		childregs->tp = regs->regs[4];
+		childregs->tp = childregs->regs[4];
 
-	/*
-	 * Copy the callee-saved registers from the passed pt_regs struct
-	 * into the context-switch callee-saved registers area.
-	 * This way when we start the interrupt-return sequence, the
-	 * callee-save registers will be correctly in registers, which
-	 * is how we assume the compiler leaves them as we start doing
-	 * the normal return-from-interrupt path after calling C code.
-	 * Zero out the C ABI save area to mark the top of the stack.
-	 */
-	ksp = (unsigned long) childregs;
-	ksp -= C_ABI_SAVE_AREA_SIZE;   /* interrupt-entry save area */
-	((long *)ksp)[0] = ((long *)ksp)[1] = 0;
-	ksp -= CALLEE_SAVED_REGS_COUNT * sizeof(unsigned long);
-	memcpy((void *)ksp, &regs->regs[CALLEE_SAVED_FIRST_REG],
-	       CALLEE_SAVED_REGS_COUNT * sizeof(unsigned long));
-	ksp -= C_ABI_SAVE_AREA_SIZE;   /* __switch_to() save area */
-	((long *)ksp)[0] = ((long *)ksp)[1] = 0;
-	p->thread.ksp = ksp;
 
 #if CHIP_HAS_TILE_DMA()
 	/*
@@ -577,62 +583,6 @@ int do_work_pending(struct pt_regs *regs, u32 thread_info_flags)
 	panic("work_pending: bad flags %#x\n", thread_info_flags);
 }
 
-/* Note there is an implicit fifth argument if (clone_flags & CLONE_SETTLS). */
-SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp,
-		void __user *, parent_tidptr, void __user *, child_tidptr,
-		struct pt_regs *, regs)
-{
-	if (!newsp)
-		newsp = regs->sp;
-	return do_fork(clone_flags, newsp, regs, 0,
-		       parent_tidptr, child_tidptr);
-}
-
-/*
- * sys_execve() executes a new program.
- */
-SYSCALL_DEFINE4(execve, const char __user *, path,
-		const char __user *const __user *, argv,
-		const char __user *const __user *, envp,
-		struct pt_regs *, regs)
-{
-	long error;
-	struct filename *filename;
-
-	filename = getname(path);
-	error = PTR_ERR(filename);
-	if (IS_ERR(filename))
-		goto out;
-	error = do_execve(filename->name, argv, envp, regs);
-	putname(filename);
-	if (error == 0)
-		single_step_execve();
-out:
-	return error;
-}
-
-#ifdef CONFIG_COMPAT
-long compat_sys_execve(const char __user *path,
-		       compat_uptr_t __user *argv,
-		       compat_uptr_t __user *envp,
-		       struct pt_regs *regs)
-{
-	long error;
-	struct filename *filename;
-
-	filename = getname(path);
-	error = PTR_ERR(filename);
-	if (IS_ERR(filename))
-		goto out;
-	error = compat_do_execve(filename->name, argv, envp, regs);
-	putname(filename);
-	if (error == 0)
-		single_step_execve();
-out:
-	return error;
-}
-#endif
-
 unsigned long get_wchan(struct task_struct *p)
 {
 	struct KBacktraceIterator kbt;
@@ -650,37 +600,6 @@ unsigned long get_wchan(struct task_struct *p)
 	return 0;
 }
 
-/*
- * We pass in lr as zero (cleared in kernel_thread) and the caller
- * part of the backtrace ABI on the stack also zeroed (in copy_thread)
- * so that backtraces will stop with this function.
- * Note that we don't use r0, since copy_thread() clears it.
- */
-static void start_kernel_thread(int dummy, int (*fn)(int), int arg)
-{
-	do_exit(fn(arg));
-}
-
-/*
- * Create a kernel thread
- */
-int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
-{
-	struct pt_regs regs;
-
-	memset(&regs, 0, sizeof(regs));
-	regs.ex1 = PL_ICS_EX1(KERNEL_PL, 0);  /* run at kernel PL, no ICS */
-	regs.pc = (long) start_kernel_thread;
-	regs.flags = PT_FLAGS_CALLER_SAVES;   /* need to restore r1 and r2 */
-	regs.regs[1] = (long) fn;             /* function pointer */
-	regs.regs[2] = (long) arg;            /* parameter register */
-
-	/* Ok, create the new process.. */
-	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs,
-		       0, NULL, NULL);
-}
-EXPORT_SYMBOL(kernel_thread);
-
 /* Flush thread state. */
 void flush_thread(void)
 {
diff --git a/arch/tile/kernel/ptrace.c b/arch/tile/kernel/ptrace.c
index e92e40527d6d..9835312d5a91 100644
--- a/arch/tile/kernel/ptrace.c
+++ b/arch/tile/kernel/ptrace.c
@@ -19,7 +19,10 @@
 #include <linux/kprobes.h>
 #include <linux/compat.h>
 #include <linux/uaccess.h>
+#include <linux/regset.h>
+#include <linux/elf.h>
 #include <asm/traps.h>
+#include <arch/chip.h>
 
 void user_enable_single_step(struct task_struct *child)
 {
@@ -45,6 +48,100 @@ void ptrace_disable(struct task_struct *child)
 	clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
 }
 
+/*
+ * Get registers from task and ready the result for userspace.
+ * Note that we localize the API issues to getregs() and putregs() at
+ * some cost in performance, e.g. we need a full pt_regs copy for
+ * PEEKUSR, and two copies for POKEUSR.  But in general we expect
+ * GETREGS/PUTREGS to be the API of choice anyway.
+ */
+static char *getregs(struct task_struct *child, struct pt_regs *uregs)
+{
+	*uregs = *task_pt_regs(child);
+
+	/* Set up flags ABI bits. */
+	uregs->flags = 0;
+#ifdef CONFIG_COMPAT
+	if (task_thread_info(child)->status & TS_COMPAT)
+		uregs->flags |= PT_FLAGS_COMPAT;
+#endif
+
+	return (char *)uregs;
+}
+
+/* Put registers back to task. */
+static void putregs(struct task_struct *child, struct pt_regs *uregs)
+{
+	struct pt_regs *regs = task_pt_regs(child);
+
+	/* Don't allow overwriting the kernel-internal flags word. */
+	uregs->flags = regs->flags;
+
+	/* Only allow setting the ICS bit in the ex1 word. */
+	uregs->ex1 = PL_ICS_EX1(USER_PL, EX1_ICS(uregs->ex1));
+
+	*regs = *uregs;
+}
+
+enum tile_regset {
+	REGSET_GPR,
+};
+
+static int tile_gpr_get(struct task_struct *target,
+			  const struct user_regset *regset,
+			  unsigned int pos, unsigned int count,
+			  void *kbuf, void __user *ubuf)
+{
+	struct pt_regs regs;
+
+	getregs(target, &regs);
+
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf, &regs, 0,
+				   sizeof(regs));
+}
+
+static int tile_gpr_set(struct task_struct *target,
+			  const struct user_regset *regset,
+			  unsigned int pos, unsigned int count,
+			  const void *kbuf, const void __user *ubuf)
+{
+	int ret;
+	struct pt_regs regs;
+
+	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &regs, 0,
+				 sizeof(regs));
+	if (ret)
+		return ret;
+
+	putregs(target, &regs);
+
+	return 0;
+}
+
+static const struct user_regset tile_user_regset[] = {
+	[REGSET_GPR] = {
+		.core_note_type = NT_PRSTATUS,
+		.n = ELF_NGREG,
+		.size = sizeof(elf_greg_t),
+		.align = sizeof(elf_greg_t),
+		.get = tile_gpr_get,
+		.set = tile_gpr_set,
+	},
+};
+
+static const struct user_regset_view tile_user_regset_view = {
+	.name = CHIP_ARCH_NAME,
+	.e_machine = ELF_ARCH,
+	.ei_osabi = ELF_OSABI,
+	.regsets = tile_user_regset,
+	.n = ARRAY_SIZE(tile_user_regset),
+};
+
+const struct user_regset_view *task_user_regset_view(struct task_struct *task)
+{
+	return &tile_user_regset_view;
+}
+
 long arch_ptrace(struct task_struct *child, long request,
 		 unsigned long addr, unsigned long data)
 {
@@ -53,14 +150,13 @@ long arch_ptrace(struct task_struct *child, long request,
 	long ret = -EIO;
 	char *childreg;
 	struct pt_regs copyregs;
-	int ex1_offset;
 
 	switch (request) {
 
 	case PTRACE_PEEKUSR:  /* Read register from pt_regs. */
 		if (addr >= PTREGS_SIZE)
 			break;
-		childreg = (char *)task_pt_regs(child) + addr;
+		childreg = getregs(child, &copyregs) + addr;
 #ifdef CONFIG_COMPAT
 		if (is_compat_task()) {
 			if (addr & (sizeof(compat_long_t)-1))
@@ -79,17 +175,7 @@ long arch_ptrace(struct task_struct *child, long request,
 	case PTRACE_POKEUSR:  /* Write register in pt_regs. */
 		if (addr >= PTREGS_SIZE)
 			break;
-		childreg = (char *)task_pt_regs(child) + addr;
-
-		/* Guard against overwrites of the privilege level. */
-		ex1_offset = PTREGS_OFFSET_EX1;
-#if defined(CONFIG_COMPAT) && defined(__BIG_ENDIAN)
-		if (is_compat_task())   /* point at low word */
-			ex1_offset += sizeof(compat_long_t);
-#endif
-		if (addr == ex1_offset)
-			data = PL_ICS_EX1(USER_PL, EX1_ICS(data));
-
+		childreg = getregs(child, &copyregs) + addr;
 #ifdef CONFIG_COMPAT
 		if (is_compat_task()) {
 			if (addr & (sizeof(compat_long_t)-1))
@@ -102,24 +188,20 @@ long arch_ptrace(struct task_struct *child, long request,
 				break;
 			*(long *)childreg = data;
 		}
+		putregs(child, &copyregs);
 		ret = 0;
 		break;
 
 	case PTRACE_GETREGS:  /* Get all registers from the child. */
-		if (copy_to_user(datap, task_pt_regs(child),
-				 sizeof(struct pt_regs)) == 0) {
-			ret = 0;
-		}
+		ret = copy_regset_to_user(child, &tile_user_regset_view,
+					  REGSET_GPR, 0,
+					  sizeof(struct pt_regs), datap);
 		break;
 
 	case PTRACE_SETREGS:  /* Set all registers in the child. */
-		if (copy_from_user(&copyregs, datap,
-				   sizeof(struct pt_regs)) == 0) {
-			copyregs.ex1 =
-				PL_ICS_EX1(USER_PL, EX1_ICS(copyregs.ex1));
-			*task_pt_regs(child) = copyregs;
-			ret = 0;
-		}
+		ret = copy_regset_from_user(child, &tile_user_regset_view,
+					    REGSET_GPR, 0,
+					    sizeof(struct pt_regs), datap);
 		break;
 
 	case PTRACE_GETFPREGS:  /* Get the child FPU state. */
@@ -128,12 +210,16 @@ long arch_ptrace(struct task_struct *child, long request,
 
 	case PTRACE_SETOPTIONS:
 		/* Support TILE-specific ptrace options. */
-		child->ptrace &= ~PT_TRACE_MASK_TILE;
+		BUILD_BUG_ON(PTRACE_O_MASK_TILE & PTRACE_O_MASK);
 		tmp = data & PTRACE_O_MASK_TILE;
 		data &= ~PTRACE_O_MASK_TILE;
 		ret = ptrace_request(child, request, addr, data);
-		if (tmp & PTRACE_O_TRACEMIGRATE)
-			child->ptrace |= PT_TRACE_MIGRATE;
+		if (ret == 0) {
+			unsigned int flags = child->ptrace;
+			flags &= ~(PTRACE_O_MASK_TILE << PT_OPT_FLAG_SHIFT);
+			flags |= (tmp << PT_OPT_FLAG_SHIFT);
+			child->ptrace = flags;
+		}
 		break;
 
 	default:
diff --git a/arch/tile/kernel/signal.c b/arch/tile/kernel/signal.c
index 67efb656d104..657a7ace4ab4 100644
--- a/arch/tile/kernel/signal.c
+++ b/arch/tile/kernel/signal.c
@@ -37,10 +37,10 @@
 
 #define DEBUG_SIG 0
 
-SYSCALL_DEFINE3(sigaltstack, const stack_t __user *, uss,
-		stack_t __user *, uoss, struct pt_regs *, regs)
+SYSCALL_DEFINE2(sigaltstack, const stack_t __user *, uss,
+		stack_t __user *, uoss)
 {
-	return do_sigaltstack(uss, uoss, regs->sp);
+	return do_sigaltstack(uss, uoss, current_pt_regs()->sp);
 }
 
 
@@ -83,8 +83,9 @@ void signal_fault(const char *type, struct pt_regs *regs,
 }
 
 /* The assembly shim for this function arranges to ignore the return value. */
-SYSCALL_DEFINE1(rt_sigreturn, struct pt_regs *, regs)
+SYSCALL_DEFINE0(rt_sigreturn)
 {
+	struct pt_regs *regs = current_pt_regs();
 	struct rt_sigframe __user *frame =
 		(struct rt_sigframe __user *)(regs->sp);
 	sigset_t set;
diff --git a/arch/tile/kernel/sys.c b/arch/tile/kernel/sys.c
index b08095b402d6..b881a7be24bd 100644
--- a/arch/tile/kernel/sys.c
+++ b/arch/tile/kernel/sys.c
@@ -106,14 +106,10 @@ SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len,
 #define sys_readahead sys32_readahead
 #endif
 
-/* Call the trampolines to manage pt_regs where necessary. */
-#define sys_execve _sys_execve
-#define sys_sigaltstack _sys_sigaltstack
+/* Call the assembly trampolines where necessary. */
+#undef sys_rt_sigreturn
 #define sys_rt_sigreturn _sys_rt_sigreturn
 #define sys_clone _sys_clone
-#ifndef __tilegx__
-#define sys_cmpxchg_badaddr _sys_cmpxchg_badaddr
-#endif
 
 /*
  * Note that we can't include <linux/unistd.h> here since the header
diff --git a/arch/tile/mm/fault.c b/arch/tile/mm/fault.c
index fe811fa5f1b9..3d2b81c163a6 100644
--- a/arch/tile/mm/fault.c
+++ b/arch/tile/mm/fault.c
@@ -70,9 +70,10 @@ static noinline void force_sig_info_fault(const char *type, int si_signo,
  * Synthesize the fault a PL0 process would get by doing a word-load of
  * an unaligned address or a high kernel address.
  */
-SYSCALL_DEFINE2(cmpxchg_badaddr, unsigned long, address,
-		struct pt_regs *, regs)
+SYSCALL_DEFINE1(cmpxchg_badaddr, unsigned long, address)
 {
+	struct pt_regs *regs = current_pt_regs();
+
 	if (address >= PAGE_OFFSET)
 		force_sig_info_fault("atomic segfault", SIGSEGV, SEGV_MAPERR,
 				     address, INT_DTLB_MISS, current, regs);
diff --git a/arch/tile/mm/hugetlbpage.c b/arch/tile/mm/hugetlbpage.c
index 812e2d037972..650ccff8378c 100644
--- a/arch/tile/mm/hugetlbpage.c
+++ b/arch/tile/mm/hugetlbpage.c
@@ -231,42 +231,15 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file,
 		unsigned long pgoff, unsigned long flags)
 {
 	struct hstate *h = hstate_file(file);
-	struct mm_struct *mm = current->mm;
-	struct vm_area_struct *vma;
-	unsigned long start_addr;
-
-	if (len > mm->cached_hole_size) {
-		start_addr = mm->free_area_cache;
-	} else {
-		start_addr = TASK_UNMAPPED_BASE;
-		mm->cached_hole_size = 0;
-	}
-
-full_search:
-	addr = ALIGN(start_addr, huge_page_size(h));
-
-	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
-		/* At this point:  (!vma || addr < vma->vm_end). */
-		if (TASK_SIZE - len < addr) {
-			/*
-			 * Start a new search - just in case we missed
-			 * some holes.
-			 */
-			if (start_addr != TASK_UNMAPPED_BASE) {
-				start_addr = TASK_UNMAPPED_BASE;
-				mm->cached_hole_size = 0;
-				goto full_search;
-			}
-			return -ENOMEM;
-		}
-		if (!vma || addr + len <= vma->vm_start) {
-			mm->free_area_cache = addr + len;
-			return addr;
-		}
-		if (addr + mm->cached_hole_size < vma->vm_start)
-			mm->cached_hole_size = vma->vm_start - addr;
-		addr = ALIGN(vma->vm_end, huge_page_size(h));
-	}
+	struct vm_unmapped_area_info info;
+
+	info.flags = 0;
+	info.length = len;
+	info.low_limit = TASK_UNMAPPED_BASE;
+	info.high_limit = TASK_SIZE;
+	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
+	info.align_offset = 0;
+	return vm_unmapped_area(&info);
 }
 
 static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
@@ -274,92 +247,30 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
 		unsigned long pgoff, unsigned long flags)
 {
 	struct hstate *h = hstate_file(file);
-	struct mm_struct *mm = current->mm;
-	struct vm_area_struct *vma, *prev_vma;
-	unsigned long base = mm->mmap_base, addr = addr0;
-	unsigned long largest_hole = mm->cached_hole_size;
-	int first_time = 1;
-
-	/* don't allow allocations above current base */
-	if (mm->free_area_cache > base)
-		mm->free_area_cache = base;
-
-	if (len <= largest_hole) {
-		largest_hole = 0;
-		mm->free_area_cache  = base;
-	}
-try_again:
-	/* make sure it can fit in the remaining address space */
-	if (mm->free_area_cache < len)
-		goto fail;
-
-	/* either no address requested or can't fit in requested address hole */
-	addr = (mm->free_area_cache - len) & huge_page_mask(h);
-	do {
-		/*
-		 * Lookup failure means no vma is above this address,
-		 * i.e. return with success:
-		 */
-		vma = find_vma_prev(mm, addr, &prev_vma);
-		if (!vma) {
-			return addr;
-			break;
-		}
-
-		/*
-		 * new region fits between prev_vma->vm_end and
-		 * vma->vm_start, use it:
-		 */
-		if (addr + len <= vma->vm_start &&
-			    (!prev_vma || (addr >= prev_vma->vm_end))) {
-			/* remember the address as a hint for next time */
-			mm->cached_hole_size = largest_hole;
-			mm->free_area_cache = addr;
-			return addr;
-		} else {
-			/* pull free_area_cache down to the first hole */
-			if (mm->free_area_cache == vma->vm_end) {
-				mm->free_area_cache = vma->vm_start;
-				mm->cached_hole_size = largest_hole;
-			}
-		}
+	struct vm_unmapped_area_info info;
+	unsigned long addr;
 
-		/* remember the largest hole we saw so far */
-		if (addr + largest_hole < vma->vm_start)
-			largest_hole = vma->vm_start - addr;
+	info.flags = VM_UNMAPPED_AREA_TOPDOWN;
+	info.length = len;
+	info.low_limit = PAGE_SIZE;
+	info.high_limit = current->mm->mmap_base;
+	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
+	info.align_offset = 0;
+	addr = vm_unmapped_area(&info);
 
-		/* try just below the current vma->vm_start */
-		addr = (vma->vm_start - len) & huge_page_mask(h);
-
-	} while (len <= vma->vm_start);
-
-fail:
-	/*
-	 * if hint left us with no space for the requested
-	 * mapping then try again:
-	 */
-	if (first_time) {
-		mm->free_area_cache = base;
-		largest_hole = 0;
-		first_time = 0;
-		goto try_again;
-	}
 	/*
 	 * A failed mmap() very likely causes application failure,
 	 * so fall back to the bottom-up function here. This scenario
 	 * can happen with large stack limits and large mmap()
 	 * allocations.
 	 */
-	mm->free_area_cache = TASK_UNMAPPED_BASE;
-	mm->cached_hole_size = ~0UL;
-	addr = hugetlb_get_unmapped_area_bottomup(file, addr0,
-			len, pgoff, flags);
-
-	/*
-	 * Restore the topdown base:
-	 */
-	mm->free_area_cache = base;
-	mm->cached_hole_size = ~0UL;
+	if (addr & ~PAGE_MASK) {
+		VM_BUG_ON(addr != -ENOMEM);
+		info.flags = 0;
+		info.low_limit = TASK_UNMAPPED_BASE;
+		info.high_limit = TASK_SIZE;
+		addr = vm_unmapped_area(&info);
+	}
 
 	return addr;
 }