aboutsummaryrefslogtreecommitdiff
path: root/arch/v850/kernel/process.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/v850/kernel/process.c')
-rw-r--r--arch/v850/kernel/process.c217
1 files changed, 0 insertions, 217 deletions
diff --git a/arch/v850/kernel/process.c b/arch/v850/kernel/process.c
deleted file mode 100644
index e4a4b8e7d5a..00000000000
--- a/arch/v850/kernel/process.c
+++ /dev/null
@@ -1,217 +0,0 @@
-/*
- * arch/v850/kernel/process.c -- Arch-dependent process handling
- *
- * Copyright (C) 2001,02,03 NEC Electronics Corporation
- * Copyright (C) 2001,02,03 Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License. See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/stddef.h>
-#include <linux/unistd.h>
-#include <linux/ptrace.h>
-#include <linux/slab.h>
-#include <linux/user.h>
-#include <linux/a.out.h>
-#include <linux/reboot.h>
-
-#include <asm/uaccess.h>
-#include <asm/system.h>
-#include <asm/pgtable.h>
-
-void (*pm_power_off)(void) = NULL;
-EXPORT_SYMBOL(pm_power_off);
-
-extern void ret_from_fork (void);
-
-
-/* The idle loop. */
-static void default_idle (void)
-{
- while (! need_resched ())
- asm ("halt; nop; nop; nop; nop; nop" ::: "cc");
-}
-
-void (*idle)(void) = default_idle;
-
-/*
- * The idle thread. There's no useful work to be
- * done, so just try to conserve power and have a
- * low exit latency (ie sit in a loop waiting for
- * somebody to say that they'd like to reschedule)
- */
-void cpu_idle (void)
-{
- /* endless idle loop with no priority at all */
- while (1) {
- while (!need_resched())
- (*idle) ();
-
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
- }
-}
-
-/*
- * This is the mechanism for creating a new kernel thread.
- *
- * NOTE! Only a kernel-only process (ie the swapper or direct descendants who
- * haven't done an "execve()") should use this: it will work within a system
- * call from a "real" process, but the process memory space will not be free'd
- * until both the parent and the child have exited.
- */
-int kernel_thread (int (*fn)(void *), void *arg, unsigned long flags)
-{
- register mm_segment_t fs = get_fs ();
- register unsigned long syscall asm (SYSCALL_NUM);
- register unsigned long arg0 asm (SYSCALL_ARG0);
- register unsigned long ret asm (SYSCALL_RET);
-
- set_fs (KERNEL_DS);
-
- /* Clone this thread. Note that we don't pass the clone syscall's
- second argument -- it's ignored for calls from kernel mode (the
- child's SP is always set to the top of the kernel stack). */
- arg0 = flags | CLONE_VM;
- syscall = __NR_clone;
- asm volatile ("trap " SYSCALL_SHORT_TRAP
- : "=r" (ret), "=r" (syscall)
- : "1" (syscall), "r" (arg0)
- : SYSCALL_SHORT_CLOBBERS);
-
- if (ret == 0) {
- /* In child thread, call FN and exit. */
- arg0 = (*fn) (arg);
- syscall = __NR_exit;
- asm volatile ("trap " SYSCALL_SHORT_TRAP
- : "=r" (ret), "=r" (syscall)
- : "1" (syscall), "r" (arg0)
- : SYSCALL_SHORT_CLOBBERS);
- }
-
- /* In parent. */
- set_fs (fs);
-
- return ret;
-}
-
-void flush_thread (void)
-{
- set_fs (USER_DS);
-}
-
-int copy_thread (int nr, unsigned long clone_flags,
- unsigned long stack_start, unsigned long stack_size,
- struct task_struct *p, struct pt_regs *regs)
-{
- /* Start pushing stuff from the top of the child's kernel stack. */
- unsigned long orig_ksp = task_tos(p);
- unsigned long ksp = orig_ksp;
- /* We push two `state save' stack fames (see entry.S) on the new
- kernel stack:
- 1) The innermost one is what switch_thread would have
- pushed, and is used when we context switch to the child
- thread for the first time. It's set up to return to
- ret_from_fork in entry.S.
- 2) The outermost one (nearest the top) is what a syscall
- trap would have pushed, and is set up to return to the
- same location as the parent thread, but with a return
- value of 0. */
- struct pt_regs *child_switch_regs, *child_trap_regs;
-
- /* Trap frame. */
- ksp -= STATE_SAVE_SIZE;
- child_trap_regs = (struct pt_regs *)(ksp + STATE_SAVE_PT_OFFSET);
- /* Switch frame. */
- ksp -= STATE_SAVE_SIZE;
- child_switch_regs = (struct pt_regs *)(ksp + STATE_SAVE_PT_OFFSET);
-
- /* First copy parent's register state to child. */
- *child_switch_regs = *regs;
- *child_trap_regs = *regs;
-
- /* switch_thread returns to the restored value of the lp
- register (r31), so we make that the place where we want to
- jump when the child thread begins running. */
- child_switch_regs->gpr[GPR_LP] = (v850_reg_t)ret_from_fork;
-
- if (regs->kernel_mode)
- /* Since we're returning to kernel-mode, make sure the child's
- stored kernel stack pointer agrees with what the actual
- stack pointer will be at that point (the trap return code
- always restores the SP, even when returning to
- kernel-mode). */
- child_trap_regs->gpr[GPR_SP] = orig_ksp;
- else
- /* Set the child's user-mode stack-pointer (the name
- `stack_start' is a misnomer, it's just the initial SP
- value). */
- child_trap_regs->gpr[GPR_SP] = stack_start;
-
- /* Thread state for the child (everything else is on the stack). */
- p->thread.ksp = ksp;
-
- return 0;
-}
-
-/*
- * sys_execve() executes a new program.
- */
-int sys_execve (char *name, char **argv, char **envp, struct pt_regs *regs)
-{
- char *filename = getname (name);
- int error = PTR_ERR (filename);
-
- if (! IS_ERR (filename)) {
- error = do_execve (filename, argv, envp, regs);
- putname (filename);
- }
-
- return error;
-}
-
-
-/*
- * These bracket the sleeping functions..
- */
-#define first_sched ((unsigned long)__sched_text_start)
-#define last_sched ((unsigned long)__sched_text_end)
-
-unsigned long get_wchan (struct task_struct *p)
-{
-#if 0 /* Barf. Figure out the stack-layout later. XXX */
- unsigned long fp, pc;
- int count = 0;
-
- if (!p || p == current || p->state == TASK_RUNNING)
- return 0;
-
- pc = thread_saved_pc (p);
-
- /* This quite disgusting function walks up the stack, following
- saved return address, until it something that's out of bounds
- (as defined by `first_sched' and `last_sched'). It then
- returns the last PC that was in-bounds. */
- do {
- if (fp < stack_page + sizeof (struct task_struct) ||
- fp >= 8184+stack_page)
- return 0;
- pc = ((unsigned long *)fp)[1];
- if (pc < first_sched || pc >= last_sched)
- return pc;
- fp = *(unsigned long *) fp;
- } while (count++ < 16);
-#endif
-
- return 0;
-}