Creation of Cybook 2416 (actually Gen4) repository

arch/um/kernel/Makefile

@@ -0,0 +1,47 @@
+#
+# Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+# Licensed under the GPL
+#
+
+extra-y := vmlinux.lds
+clean-files :=
+
+obj-y = config.o exec.o exitcode.o init_task.o irq.o ksyms.o mem.o \
+	physmem.o process.o ptrace.o reboot.o sigio.o \
+	signal.o smp.o syscall.o sysrq.o time.o tlb.o trap.o uaccess.o \
+	um_arch.o umid.o
+
+obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o
+obj-$(CONFIG_GPROF)	+= gprof_syms.o
+obj-$(CONFIG_GCOV)	+= gmon_syms.o
+
+obj-$(CONFIG_MODE_TT) += tt/
+obj-$(CONFIG_MODE_SKAS) += skas/
+
+USER_OBJS := config.o
+
+include arch/um/scripts/Makefile.rules
+
+targets := config.c config.tmp
+
+# Be careful with the below Sed code - sed is pitfall-rich!
+# We use sed to lower build requirements, for "embedded" builders for instance.
+
+$(obj)/config.tmp: $(objtree)/.config FORCE
+	$(call if_changed,quote1)
+
+quiet_cmd_quote1 = QUOTE   $@
+      cmd_quote1 = sed -e 's/"/\\"/g' -e 's/^/"/' -e 's/$$/\\n"/' \
+		   $< > $@
+
+$(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE
+	$(call if_changed,quote2)
+
+quiet_cmd_quote2 = QUOTE   $@
+      cmd_quote2 = sed -e '/CONFIG/{'          \
+		  -e 's/"CONFIG"\;/""/'        \
+		  -e 'r $(obj)/config.tmp'     \
+		  -e 'a \'                     \
+		  -e '""\;'                    \
+		  -e '}'                       \
+		  $< > $@

arch/um/kernel/asm-offsets.c

@@ -0,0 +1 @@
+#include "sysdep/kernel-offsets.h"

arch/um/kernel/config.c.in

@@ -0,0 +1,32 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include "init.h"
+
+static __initdata char *config = "CONFIG";
+
+static int __init print_config(char *line, int *add)
+{
+	printf("%s", config);
+	exit(0);
+}
+
+__uml_setup("--showconfig", print_config,
+"--showconfig\n"
+"    Prints the config file that this UML binary was generated from.\n\n"
+);
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/dyn.lds.S

@@ -0,0 +1,155 @@
+#include <asm-generic/vmlinux.lds.h>
+
+OUTPUT_FORMAT(ELF_FORMAT)
+OUTPUT_ARCH(ELF_ARCH)
+ENTRY(_start)
+jiffies = jiffies_64;
+
+SECTIONS
+{
+  PROVIDE (__executable_start = START);
+  . = START + SIZEOF_HEADERS;
+  .interp         : { *(.interp) }
+  /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start
+   * is remapped.*/
+  __binary_start = .;
+  . = ALIGN(4096);		/* Init code and data */
+  _text = .;
+  _stext = .;
+  __init_begin = .;
+  .init.text : {
+	_sinittext = .;
+	*(.init.text)
+	_einittext = .;
+  }
+
+  . = ALIGN(4096);
+
+  /* Read-only sections, merged into text segment: */
+  .hash           : { *(.hash) }
+  .gnu.hash       : { *(.gnu.hash) }
+  .dynsym         : { *(.dynsym) }
+  .dynstr         : { *(.dynstr) }
+  .gnu.version    : { *(.gnu.version) }
+  .gnu.version_d  : { *(.gnu.version_d) }
+  .gnu.version_r  : { *(.gnu.version_r) }
+  .rel.init       : { *(.rel.init) }
+  .rela.init      : { *(.rela.init) }
+  .rel.text       : { *(.rel.text .rel.text.* .rel.gnu.linkonce.t.*) }
+  .rela.text      : { *(.rela.text .rela.text.* .rela.gnu.linkonce.t.*) }
+  .rel.fini       : { *(.rel.fini) }
+  .rela.fini      : { *(.rela.fini) }
+  .rel.rodata     : { *(.rel.rodata .rel.rodata.* .rel.gnu.linkonce.r.*) }
+  .rela.rodata    : { *(.rela.rodata .rela.rodata.* .rela.gnu.linkonce.r.*) }
+  .rel.data       : { *(.rel.data .rel.data.* .rel.gnu.linkonce.d.*) }
+  .rela.data      : { *(.rela.data .rela.data.* .rela.gnu.linkonce.d.*) }
+  .rel.tdata	  : { *(.rel.tdata .rel.tdata.* .rel.gnu.linkonce.td.*) }
+  .rela.tdata	  : { *(.rela.tdata .rela.tdata.* .rela.gnu.linkonce.td.*) }
+  .rel.tbss	  : { *(.rel.tbss .rel.tbss.* .rel.gnu.linkonce.tb.*) }
+  .rela.tbss	  : { *(.rela.tbss .rela.tbss.* .rela.gnu.linkonce.tb.*) }
+  .rel.ctors      : { *(.rel.ctors) }
+  .rela.ctors     : { *(.rela.ctors) }
+  .rel.dtors      : { *(.rel.dtors) }
+  .rela.dtors     : { *(.rela.dtors) }
+  .rel.got        : { *(.rel.got) }
+  .rela.got       : { *(.rela.got) }
+  .rel.bss        : { *(.rel.bss .rel.bss.* .rel.gnu.linkonce.b.*) }
+  .rela.bss       : { *(.rela.bss .rela.bss.* .rela.gnu.linkonce.b.*) }
+  .rel.plt        : { *(.rel.plt) }
+  .rela.plt       : { *(.rela.plt) }
+  .init           : {
+    KEEP (*(.init))
+  } =0x90909090
+  .plt            : { *(.plt) }
+  .text           : {
+    *(.text)
+    SCHED_TEXT
+    LOCK_TEXT
+    *(.fixup)
+    *(.stub .text.* .gnu.linkonce.t.*)
+    /* .gnu.warning sections are handled specially by elf32.em.  */
+    *(.gnu.warning)
+
+    . = ALIGN(4096);
+    __syscall_stub_start = .;
+    *(.__syscall_stub*)
+    __syscall_stub_end = .;
+    . = ALIGN(4096);
+  } =0x90909090
+  .fini           : {
+    KEEP (*(.fini))
+  } =0x90909090
+
+  .kstrtab : { *(.kstrtab) }
+
+  #include "asm/common.lds.S"
+
+  init.data : { *(.init.data) }
+
+  /* Ensure the __preinit_array_start label is properly aligned.  We
+     could instead move the label definition inside the section, but
+     the linker would then create the section even if it turns out to
+     be empty, which isn't pretty.  */
+  . = ALIGN(32 / 8);
+  .preinit_array     : { *(.preinit_array) }
+  .init_array     : { *(.init_array) }
+  .fini_array     : { *(.fini_array) }
+  .data           : {
+    . = ALIGN(KERNEL_STACK_SIZE);		/* init_task */
+    *(.data.init_task)
+    *(.data .data.* .gnu.linkonce.d.*)
+    SORT(CONSTRUCTORS)
+  }
+  .data1          : { *(.data1) }
+  .tdata	  : { *(.tdata .tdata.* .gnu.linkonce.td.*) }
+  .tbss		  : { *(.tbss .tbss.* .gnu.linkonce.tb.*) *(.tcommon) }
+  .eh_frame       : { KEEP (*(.eh_frame)) }
+  .gcc_except_table   : { *(.gcc_except_table) }
+  .dynamic        : { *(.dynamic) }
+  .ctors          : {
+    /* gcc uses crtbegin.o to find the start of
+       the constructors, so we make sure it is
+       first.  Because this is a wildcard, it
+       doesn't matter if the user does not
+       actually link against crtbegin.o; the
+       linker won't look for a file to match a
+       wildcard.  The wildcard also means that it
+       doesn't matter which directory crtbegin.o
+       is in.  */
+    KEEP (*crtbegin.o(.ctors))
+    /* We don't want to include the .ctor section from
+       from the crtend.o file until after the sorted ctors.
+       The .ctor section from the crtend file contains the
+       end of ctors marker and it must be last */
+    KEEP (*(EXCLUDE_FILE (*crtend.o ) .ctors))
+    KEEP (*(SORT(.ctors.*)))
+    KEEP (*(.ctors))
+  }
+  .dtors          : {
+    KEEP (*crtbegin.o(.dtors))
+    KEEP (*(EXCLUDE_FILE (*crtend.o ) .dtors))
+    KEEP (*(SORT(.dtors.*)))
+    KEEP (*(.dtors))
+  }
+  .jcr            : { KEEP (*(.jcr)) }
+  .got            : { *(.got.plt) *(.got) }
+  _edata = .;
+  PROVIDE (edata = .);
+  __bss_start = .;
+  .bss            : {
+   *(.dynbss)
+   *(.bss .bss.* .gnu.linkonce.b.*)
+   *(COMMON)
+   /* Align here to ensure that the .bss section occupies space up to
+      _end.  Align after .bss to ensure correct alignment even if the
+      .bss section disappears because there are no input sections.  */
+   . = ALIGN(32 / 8);
+  . = ALIGN(32 / 8);
+  }
+  _end = .;
+  PROVIDE (end = .);
+
+  STABS_DEBUG
+
+  DWARF_DEBUG
+}

arch/um/kernel/exec.c

@@ -0,0 +1,89 @@
+/*
+ * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/slab.h"
+#include "linux/smp_lock.h"
+#include "linux/ptrace.h"
+#include "asm/ptrace.h"
+#include "asm/pgtable.h"
+#include "asm/tlbflush.h"
+#include "asm/uaccess.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "mem_user.h"
+#include "kern.h"
+#include "irq_user.h"
+#include "tlb.h"
+#include "os.h"
+#include "choose-mode.h"
+#include "mode_kern.h"
+
+void flush_thread(void)
+{
+	arch_flush_thread(&current->thread.arch);
+	CHOOSE_MODE(flush_thread_tt(), flush_thread_skas());
+}
+
+void start_thread(struct pt_regs *regs, unsigned long eip, unsigned long esp)
+{
+	CHOOSE_MODE_PROC(start_thread_tt, start_thread_skas, regs, eip, esp);
+}
+
+#ifdef CONFIG_TTY_LOG
+extern void log_exec(char **argv, void *tty);
+#endif
+
+static long execve1(char *file, char __user * __user *argv,
+		    char __user *__user *env)
+{
+        long error;
+#ifdef CONFIG_TTY_LOG
+	struct tty_struct *tty;
+
+	mutex_lock(&tty_mutex);
+	tty = get_current_tty();
+	if (tty)
+		log_exec(argv, tty);
+	mutex_unlock(&tty_mutex);
+#endif
+        error = do_execve(file, argv, env, &current->thread.regs);
+        if (error == 0){
+		task_lock(current);
+                current->ptrace &= ~PT_DTRACE;
+#ifdef SUBARCH_EXECVE1
+		SUBARCH_EXECVE1(&current->thread.regs.regs);
+#endif
+		task_unlock(current);
+                set_cmdline(current_cmd());
+        }
+        return(error);
+}
+
+long um_execve(char *file, char __user *__user *argv, char __user *__user *env)
+{
+	long err;
+
+	err = execve1(file, argv, env);
+	if(!err)
+		do_longjmp(current->thread.exec_buf, 1);
+	return(err);
+}
+
+long sys_execve(char __user *file, char __user *__user *argv,
+		char __user *__user *env)
+{
+	long error;
+	char *filename;
+
+	lock_kernel();
+	filename = getname(file);
+	error = PTR_ERR(filename);
+	if (IS_ERR(filename)) goto out;
+	error = execve1(filename, argv, env);
+	putname(filename);
+ out:
+	unlock_kernel();
+	return(error);
+}

arch/um/kernel/exitcode.c

@@ -0,0 +1,77 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/init.h"
+#include "linux/ctype.h"
+#include "linux/proc_fs.h"
+#include "asm/uaccess.h"
+
+/* If read and write race, the read will still atomically read a valid
+ * value.
+ */
+int uml_exitcode = 0;
+
+static int read_proc_exitcode(char *page, char **start, off_t off,
+			      int count, int *eof, void *data)
+{
+	int len, val;
+
+	/* Save uml_exitcode in a local so that we don't need to guarantee
+	 * that sprintf accesses it atomically.
+	 */
+	val = uml_exitcode;
+	len = sprintf(page, "%d\n", val);
+	len -= off;
+	if(len <= off+count) *eof = 1;
+	*start = page + off;
+	if(len > count) len = count;
+	if(len < 0) len = 0;
+	return(len);
+}
+
+static int write_proc_exitcode(struct file *file, const char __user *buffer,
+			       unsigned long count, void *data)
+{
+	char *end, buf[sizeof("nnnnn\0")];
+	int tmp;
+
+	if(copy_from_user(buf, buffer, count))
+		return(-EFAULT);
+	tmp = simple_strtol(buf, &end, 0);
+	if((*end != '\0') && !isspace(*end))
+		return(-EINVAL);
+	uml_exitcode = tmp;
+	return(count);
+}
+
+static int make_proc_exitcode(void)
+{
+	struct proc_dir_entry *ent;
+
+	ent = create_proc_entry("exitcode", 0600, &proc_root);
+	if(ent == NULL){
+		printk(KERN_WARNING "make_proc_exitcode : Failed to register "
+		       "/proc/exitcode\n");
+		return(0);
+	}
+
+	ent->read_proc = read_proc_exitcode;
+	ent->write_proc = write_proc_exitcode;
+	
+	return(0);
+}
+
+__initcall(make_proc_exitcode);
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/gmon_syms.c

@@ -0,0 +1,23 @@
+/* 
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/module.h"
+
+extern void __bb_init_func(void *)  __attribute__((weak));
+EXPORT_SYMBOL(__bb_init_func);
+
+/* This is defined (and referred to in profiling stub code) only by some GCC
+ * versions in libgcov.
+ *
+ * Since SuSE backported the fix, we cannot handle it depending on GCC version.
+ * So, unconditinally export it. But also give it a weak declaration, which will
+ * be overriden by any other one.
+ */
+
+extern void __gcov_init(void *) __attribute__((weak));
+EXPORT_SYMBOL(__gcov_init);
+
+extern void __gcov_merge_add(void *) __attribute__((weak));
+EXPORT_SYMBOL(__gcov_merge_add);

arch/um/kernel/gprof_syms.c

@@ -0,0 +1,20 @@
+/* 
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/module.h"
+
+extern void mcount(void);
+EXPORT_SYMBOL(mcount);
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/init_task.c

@@ -0,0 +1,61 @@
+/* 
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/mm.h"
+#include "linux/module.h"
+#include "linux/sched.h"
+#include "linux/init_task.h"
+#include "linux/mqueue.h"
+#include "asm/uaccess.h"
+#include "asm/pgtable.h"
+#include "user_util.h"
+#include "mem_user.h"
+#include "os.h"
+
+static struct fs_struct init_fs = INIT_FS;
+struct mm_struct init_mm = INIT_MM(init_mm);
+static struct files_struct init_files = INIT_FILES;
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+EXPORT_SYMBOL(init_mm);
+
+/*
+ * Initial task structure.
+ *
+ * All other task structs will be allocated on slabs in fork.c
+ */
+
+struct task_struct init_task = INIT_TASK(init_task);
+
+EXPORT_SYMBOL(init_task);
+
+/*
+ * Initial thread structure.
+ *
+ * We need to make sure that this is 16384-byte aligned due to the
+ * way process stacks are handled. This is done by having a special
+ * "init_task" linker map entry..
+ */
+
+union thread_union init_thread_union 
+__attribute__((__section__(".data.init_task"))) = 
+{ INIT_THREAD_INFO(init_task) };
+
+void unprotect_stack(unsigned long stack)
+{
+	os_protect_memory((void *) stack, (1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE,
+		       1, 1, 0);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/initrd.c

@@ -0,0 +1,78 @@
+/*
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/init.h"
+#include "linux/bootmem.h"
+#include "linux/initrd.h"
+#include "asm/types.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "initrd.h"
+#include "init.h"
+#include "os.h"
+
+/* Changed by uml_initrd_setup, which is a setup */
+static char *initrd __initdata = NULL;
+
+static int __init read_initrd(void)
+{
+	void *area;
+	long long size;
+	int err;
+
+	if(initrd == NULL) return 0;
+	err = os_file_size(initrd, &size);
+	if(err) return 0;
+	area = alloc_bootmem(size);
+	if(area == NULL) return 0;
+	if(load_initrd(initrd, area, size) == -1) return 0;
+	initrd_start = (unsigned long) area;
+	initrd_end = initrd_start + size;
+	return 0;
+}
+
+__uml_postsetup(read_initrd);
+
+static int __init uml_initrd_setup(char *line, int *add)
+{
+	initrd = line;
+	return 0;
+}
+
+__uml_setup("initrd=", uml_initrd_setup,
+"initrd=<initrd image>\n"
+"    This is used to boot UML from an initrd image.  The argument is the\n"
+"    name of the file containing the image.\n\n"
+);
+
+int load_initrd(char *filename, void *buf, int size)
+{
+	int fd, n;
+
+	fd = os_open_file(filename, of_read(OPENFLAGS()), 0);
+	if(fd < 0){
+		printk("Opening '%s' failed - err = %d\n", filename, -fd);
+		return(-1);
+	}
+	n = os_read_file(fd, buf, size);
+	if(n != size){
+		printk("Read of %d bytes from '%s' failed, err = %d\n", size,
+		       filename, -n);
+		return(-1);
+	}
+
+	os_close_file(fd);
+	return(0);
+}
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/irq.c

@@ -0,0 +1,456 @@
+/* 
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ * Derived (i.e. mostly copied) from arch/i386/kernel/irq.c:
+ *	Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
+ */
+
+#include "linux/kernel.h"
+#include "linux/module.h"
+#include "linux/smp.h"
+#include "linux/kernel_stat.h"
+#include "linux/interrupt.h"
+#include "linux/random.h"
+#include "linux/slab.h"
+#include "linux/file.h"
+#include "linux/proc_fs.h"
+#include "linux/init.h"
+#include "linux/seq_file.h"
+#include "linux/profile.h"
+#include "linux/hardirq.h"
+#include "asm/irq.h"
+#include "asm/hw_irq.h"
+#include "asm/atomic.h"
+#include "asm/signal.h"
+#include "asm/system.h"
+#include "asm/errno.h"
+#include "asm/uaccess.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "irq_user.h"
+#include "irq_kern.h"
+#include "os.h"
+#include "sigio.h"
+#include "um_malloc.h"
+#include "misc_constants.h"
+
+/*
+ * Generic, controller-independent functions:
+ */
+
+int show_interrupts(struct seq_file *p, void *v)
+{
+	int i = *(loff_t *) v, j;
+	struct irqaction * action;
+	unsigned long flags;
+
+	if (i == 0) {
+		seq_printf(p, "           ");
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ",j);
+		seq_putc(p, '\n');
+	}
+
+	if (i < NR_IRQS) {
+		spin_lock_irqsave(&irq_desc[i].lock, flags);
+		action = irq_desc[i].action;
+		if (!action) 
+			goto skip;
+		seq_printf(p, "%3d: ",i);
+#ifndef CONFIG_SMP
+		seq_printf(p, "%10u ", kstat_irqs(i));
+#else
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+#endif
+		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, "  %s", action->name);
+
+		for (action=action->next; action; action = action->next)
+			seq_printf(p, ", %s", action->name);
+
+		seq_putc(p, '\n');
+skip:
+		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
+	} else if (i == NR_IRQS) {
+		seq_putc(p, '\n');
+	}
+
+	return 0;
+}
+
+static struct irq_fd *active_fds = NULL;
+static struct irq_fd **last_irq_ptr = &active_fds;
+
+extern void free_irqs(void);
+
+void sigio_handler(int sig, union uml_pt_regs *regs)
+{
+	struct irq_fd *irq_fd;
+	int n;
+
+	if (smp_sigio_handler())
+		return;
+
+	while (1) {
+		n = os_waiting_for_events(active_fds);
+		if (n <= 0) {
+			if(n == -EINTR) continue;
+			else break;
+		}
+
+		for (irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next) {
+			if (irq_fd->current_events != 0) {
+				irq_fd->current_events = 0;
+				do_IRQ(irq_fd->irq, regs);
+			}
+		}
+	}
+
+	free_irqs();
+}
+
+static DEFINE_SPINLOCK(irq_lock);
+
+int activate_fd(int irq, int fd, int type, void *dev_id)
+{
+	struct pollfd *tmp_pfd;
+	struct irq_fd *new_fd, *irq_fd;
+	unsigned long flags;
+	int pid, events, err, n;
+
+	pid = os_getpid();
+	err = os_set_fd_async(fd, pid);
+	if (err < 0)
+		goto out;
+
+	err = -ENOMEM;
+	new_fd = kmalloc(sizeof(struct irq_fd), GFP_KERNEL);
+	if (new_fd == NULL)
+		goto out;
+
+	if (type == IRQ_READ)
+		events = UM_POLLIN | UM_POLLPRI;
+	else
+		events = UM_POLLOUT;
+	*new_fd = ((struct irq_fd) { .next  		= NULL,
+				     .id 		= dev_id,
+				     .fd 		= fd,
+				     .type 		= type,
+				     .irq 		= irq,
+				     .pid  		= pid,
+				     .events 		= events,
+				     .current_events 	= 0 } );
+
+	err = -EBUSY;
+	spin_lock_irqsave(&irq_lock, flags);
+	for (irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next) {
+		if ((irq_fd->fd == fd) && (irq_fd->type == type)) {
+			printk("Registering fd %d twice\n", fd);
+			printk("Irqs : %d, %d\n", irq_fd->irq, irq);
+			printk("Ids : 0x%p, 0x%p\n", irq_fd->id, dev_id);
+			goto out_unlock;
+		}
+	}
+
+	if (type == IRQ_WRITE)
+		fd = -1;
+
+	tmp_pfd = NULL;
+	n = 0;
+
+	while (1) {
+		n = os_create_pollfd(fd, events, tmp_pfd, n);
+		if (n == 0)
+			break;
+
+		/* n > 0
+		 * It means we couldn't put new pollfd to current pollfds
+		 * and tmp_fds is NULL or too small for new pollfds array.
+		 * Needed size is equal to n as minimum.
+		 *
+		 * Here we have to drop the lock in order to call
+		 * kmalloc, which might sleep.
+		 * If something else came in and changed the pollfds array
+		 * so we will not be able to put new pollfd struct to pollfds
+		 * then we free the buffer tmp_fds and try again.
+		 */
+		spin_unlock_irqrestore(&irq_lock, flags);
+		kfree(tmp_pfd);
+
+		tmp_pfd = kmalloc(n, GFP_KERNEL);
+		if (tmp_pfd == NULL)
+			goto out_kfree;
+
+		spin_lock_irqsave(&irq_lock, flags);
+	}
+
+	*last_irq_ptr = new_fd;
+	last_irq_ptr = &new_fd->next;
+
+	spin_unlock_irqrestore(&irq_lock, flags);
+
+	/* This calls activate_fd, so it has to be outside the critical
+	 * section.
+	 */
+	maybe_sigio_broken(fd, (type == IRQ_READ));
+
+	return 0;
+
+ out_unlock:
+	spin_unlock_irqrestore(&irq_lock, flags);
+ out_kfree:
+	kfree(new_fd);
+ out:
+	return err;
+}
+
+static void free_irq_by_cb(int (*test)(struct irq_fd *, void *), void *arg)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&irq_lock, flags);
+	os_free_irq_by_cb(test, arg, active_fds, &last_irq_ptr);
+	spin_unlock_irqrestore(&irq_lock, flags);
+}
+
+struct irq_and_dev {
+	int irq;
+	void *dev;
+};
+
+static int same_irq_and_dev(struct irq_fd *irq, void *d)
+{
+	struct irq_and_dev *data = d;
+
+	return ((irq->irq == data->irq) && (irq->id == data->dev));
+}
+
+void free_irq_by_irq_and_dev(unsigned int irq, void *dev)
+{
+	struct irq_and_dev data = ((struct irq_and_dev) { .irq  = irq,
+							  .dev  = dev });
+
+	free_irq_by_cb(same_irq_and_dev, &data);
+}
+
+static int same_fd(struct irq_fd *irq, void *fd)
+{
+	return (irq->fd == *((int *)fd));
+}
+
+void free_irq_by_fd(int fd)
+{
+	free_irq_by_cb(same_fd, &fd);
+}
+
+static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out)
+{
+	struct irq_fd *irq;
+	int i = 0;
+	int fdi;
+
+	for (irq = active_fds; irq != NULL; irq = irq->next) {
+		if ((irq->fd == fd) && (irq->irq == irqnum))
+			break;
+		i++;
+	}
+	if (irq == NULL) {
+		printk("find_irq_by_fd doesn't have descriptor %d\n", fd);
+		goto out;
+	}
+	fdi = os_get_pollfd(i);
+	if ((fdi != -1) && (fdi != fd)) {
+		printk("find_irq_by_fd - mismatch between active_fds and "
+		       "pollfds, fd %d vs %d, need %d\n", irq->fd,
+		       fdi, fd);
+		irq = NULL;
+		goto out;
+	}
+	*index_out = i;
+ out:
+	return irq;
+}
+
+void reactivate_fd(int fd, int irqnum)
+{
+	struct irq_fd *irq;
+	unsigned long flags;
+	int i;
+
+	spin_lock_irqsave(&irq_lock, flags);
+	irq = find_irq_by_fd(fd, irqnum, &i);
+	if (irq == NULL) {
+		spin_unlock_irqrestore(&irq_lock, flags);
+		return;
+	}
+	os_set_pollfd(i, irq->fd);
+	spin_unlock_irqrestore(&irq_lock, flags);
+
+	add_sigio_fd(fd);
+}
+
+void deactivate_fd(int fd, int irqnum)
+{
+	struct irq_fd *irq;
+	unsigned long flags;
+	int i;
+
+	spin_lock_irqsave(&irq_lock, flags);
+	irq = find_irq_by_fd(fd, irqnum, &i);
+	if(irq == NULL){
+		spin_unlock_irqrestore(&irq_lock, flags);
+		return;
+	}
+
+	os_set_pollfd(i, -1);
+	spin_unlock_irqrestore(&irq_lock, flags);
+
+	ignore_sigio_fd(fd);
+}
+
+int deactivate_all_fds(void)
+{
+	struct irq_fd *irq;
+	int err;
+
+	for (irq = active_fds; irq != NULL; irq = irq->next) {
+		err = os_clear_fd_async(irq->fd);
+		if (err)
+			return err;
+	}
+	/* If there is a signal already queued, after unblocking ignore it */
+	os_set_ioignore();
+
+	return 0;
+}
+
+#ifdef CONFIG_MODE_TT
+void forward_interrupts(int pid)
+{
+	struct irq_fd *irq;
+	unsigned long flags;
+	int err;
+
+	spin_lock_irqsave(&irq_lock, flags);
+	for (irq = active_fds; irq != NULL; irq = irq->next) {
+		err = os_set_owner(irq->fd, pid);
+		if (err < 0) {
+			/* XXX Just remove the irq rather than
+			 * print out an infinite stream of these
+			 */
+			printk("Failed to forward %d to pid %d, err = %d\n",
+			       irq->fd, pid, -err);
+		}
+
+		irq->pid = pid;
+	}
+	spin_unlock_irqrestore(&irq_lock, flags);
+}
+#endif
+
+/*
+ * do_IRQ handles all normal device IRQ's (the special
+ * SMP cross-CPU interrupts have their own specific
+ * handlers).
+ */
+unsigned int do_IRQ(int irq, union uml_pt_regs *regs)
+{
+	struct pt_regs *old_regs = set_irq_regs((struct pt_regs *)regs);
+	irq_enter();
+	__do_IRQ(irq);
+	irq_exit();
+	set_irq_regs(old_regs);
+	return 1;
+}
+
+int um_request_irq(unsigned int irq, int fd, int type,
+		   irq_handler_t handler,
+		   unsigned long irqflags, const char * devname,
+		   void *dev_id)
+{
+	int err;
+
+	err = request_irq(irq, handler, irqflags, devname, dev_id);
+	if (err)
+		return err;
+
+	if (fd != -1)
+		err = activate_fd(irq, fd, type, dev_id);
+	return err;
+}
+EXPORT_SYMBOL(um_request_irq);
+EXPORT_SYMBOL(reactivate_fd);
+
+/* hw_interrupt_type must define (startup || enable) &&
+ * (shutdown || disable) && end */
+static void dummy(unsigned int irq)
+{
+}
+
+/* This is used for everything else than the timer. */
+static struct hw_interrupt_type normal_irq_type = {
+	.typename = "SIGIO",
+	.release = free_irq_by_irq_and_dev,
+	.disable = dummy,
+	.enable = dummy,
+	.ack = dummy,
+	.end = dummy
+};
+
+static struct hw_interrupt_type SIGVTALRM_irq_type = {
+	.typename = "SIGVTALRM",
+	.release = free_irq_by_irq_and_dev,
+	.shutdown = dummy, /* never called */
+	.disable = dummy,
+	.enable = dummy,
+	.ack = dummy,
+	.end = dummy
+};
+
+void __init init_IRQ(void)
+{
+	int i;
+
+	irq_desc[TIMER_IRQ].status = IRQ_DISABLED;
+	irq_desc[TIMER_IRQ].action = NULL;
+	irq_desc[TIMER_IRQ].depth = 1;
+	irq_desc[TIMER_IRQ].chip = &SIGVTALRM_irq_type;
+	enable_irq(TIMER_IRQ);
+	for (i = 1; i < NR_IRQS; i++) {
+		irq_desc[i].status = IRQ_DISABLED;
+		irq_desc[i].action = NULL;
+		irq_desc[i].depth = 1;
+		irq_desc[i].chip = &normal_irq_type;
+		enable_irq(i);
+	}
+}
+
+int init_aio_irq(int irq, char *name, irq_handler_t handler)
+{
+	int fds[2], err;
+
+	err = os_pipe(fds, 1, 1);
+	if (err) {
+		printk("init_aio_irq - os_pipe failed, err = %d\n", -err);
+		goto out;
+	}
+
+	err = um_request_irq(irq, fds[0], IRQ_READ, handler,
+			     IRQF_DISABLED | IRQF_SAMPLE_RANDOM, name,
+			     (void *) (long) fds[0]);
+	if (err) {
+		printk("init_aio_irq - : um_request_irq failed, err = %d\n",
+		       err);
+		goto out_close;
+	}
+
+	err = fds[1];
+	goto out;
+
+ out_close:
+	os_close_file(fds[0]);
+	os_close_file(fds[1]);
+ out:
+	return err;
+}

arch/um/kernel/ksyms.c

@@ -0,0 +1,101 @@
+/* 
+ * Copyright (C) 2001 - 2004 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/module.h"
+#include "linux/string.h"
+#include "linux/smp_lock.h"
+#include "linux/spinlock.h"
+#include "linux/highmem.h"
+#include "asm/current.h"
+#include "asm/processor.h"
+#include "asm/unistd.h"
+#include "asm/pgalloc.h"
+#include "asm/pgtable.h"
+#include "asm/page.h"
+#include "asm/tlbflush.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "mem_user.h"
+#include "os.h"
+
+EXPORT_SYMBOL(uml_physmem);
+EXPORT_SYMBOL(set_signals);
+EXPORT_SYMBOL(get_signals);
+EXPORT_SYMBOL(kernel_thread);
+EXPORT_SYMBOL(sys_waitpid);
+EXPORT_SYMBOL(task_size);
+EXPORT_SYMBOL(flush_tlb_range);
+EXPORT_SYMBOL(host_task_size);
+EXPORT_SYMBOL(arch_validate);
+EXPORT_SYMBOL(get_kmem_end);
+
+EXPORT_SYMBOL(high_physmem);
+EXPORT_SYMBOL(empty_zero_page);
+EXPORT_SYMBOL(um_virt_to_phys);
+EXPORT_SYMBOL(mode_tt);
+EXPORT_SYMBOL(handle_page_fault);
+EXPORT_SYMBOL(find_iomem);
+
+#ifdef CONFIG_MODE_TT
+EXPORT_SYMBOL(stop);
+EXPORT_SYMBOL(strncpy_from_user_tt);
+EXPORT_SYMBOL(copy_from_user_tt);
+EXPORT_SYMBOL(copy_to_user_tt);
+#endif
+
+#ifdef CONFIG_MODE_SKAS
+EXPORT_SYMBOL(strnlen_user_skas);
+EXPORT_SYMBOL(strncpy_from_user_skas);
+EXPORT_SYMBOL(copy_to_user_skas);
+EXPORT_SYMBOL(copy_from_user_skas);
+EXPORT_SYMBOL(clear_user_skas);
+#endif
+EXPORT_SYMBOL(uml_strdup);
+
+EXPORT_SYMBOL(os_stat_fd);
+EXPORT_SYMBOL(os_stat_file);
+EXPORT_SYMBOL(os_access);
+EXPORT_SYMBOL(os_print_error);
+EXPORT_SYMBOL(os_get_exec_close);
+EXPORT_SYMBOL(os_set_exec_close);
+EXPORT_SYMBOL(os_getpid);
+EXPORT_SYMBOL(os_open_file);
+EXPORT_SYMBOL(os_read_file);
+EXPORT_SYMBOL(os_write_file);
+EXPORT_SYMBOL(os_seek_file);
+EXPORT_SYMBOL(os_lock_file);
+EXPORT_SYMBOL(os_ioctl_generic);
+EXPORT_SYMBOL(os_pipe);
+EXPORT_SYMBOL(os_file_type);
+EXPORT_SYMBOL(os_file_mode);
+EXPORT_SYMBOL(os_file_size);
+EXPORT_SYMBOL(os_flush_stdout);
+EXPORT_SYMBOL(os_close_file);
+EXPORT_SYMBOL(os_set_fd_async);
+EXPORT_SYMBOL(os_set_fd_block);
+EXPORT_SYMBOL(helper_wait);
+EXPORT_SYMBOL(os_shutdown_socket);
+EXPORT_SYMBOL(os_create_unix_socket);
+EXPORT_SYMBOL(os_connect_socket);
+EXPORT_SYMBOL(os_accept_connection);
+EXPORT_SYMBOL(os_rcv_fd);
+EXPORT_SYMBOL(run_helper);
+EXPORT_SYMBOL(start_thread);
+EXPORT_SYMBOL(dump_thread);
+
+EXPORT_SYMBOL(do_gettimeofday);
+EXPORT_SYMBOL(do_settimeofday);
+
+#ifdef CONFIG_SMP
+
+/* required for SMP */
+
+extern void FASTCALL( __write_lock_failed(rwlock_t *rw));
+EXPORT_SYMBOL(__write_lock_failed);
+
+extern void FASTCALL( __read_lock_failed(rwlock_t *rw));
+EXPORT_SYMBOL(__read_lock_failed);
+
+#endif

arch/um/kernel/mem.c

@@ -0,0 +1,370 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/stddef.h"
+#include "linux/kernel.h"
+#include "linux/mm.h"
+#include "linux/bootmem.h"
+#include "linux/swap.h"
+#include "linux/highmem.h"
+#include "linux/gfp.h"
+#include "asm/page.h"
+#include "asm/fixmap.h"
+#include "asm/pgalloc.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "mem_user.h"
+#include "uml_uaccess.h"
+#include "os.h"
+#include "linux/types.h"
+#include "linux/string.h"
+#include "init.h"
+#include "kern_constants.h"
+
+/* allocated in paging_init, zeroed in mem_init, and unchanged thereafter */
+unsigned long *empty_zero_page = NULL;
+/* allocated in paging_init and unchanged thereafter */
+unsigned long *empty_bad_page = NULL;
+pgd_t swapper_pg_dir[PTRS_PER_PGD];
+unsigned long long highmem;
+int kmalloc_ok = 0;
+
+static unsigned long brk_end;
+
+void unmap_physmem(void)
+{
+	os_unmap_memory((void *) brk_end, uml_reserved - brk_end);
+}
+
+static void map_cb(void *unused)
+{
+	map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0);
+}
+
+#ifdef CONFIG_HIGHMEM
+static void setup_highmem(unsigned long highmem_start,
+			  unsigned long highmem_len)
+{
+	struct page *page;
+	unsigned long highmem_pfn;
+	int i;
+
+	highmem_pfn = __pa(highmem_start) >> PAGE_SHIFT;
+	for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){
+		page = &mem_map[highmem_pfn + i];
+		ClearPageReserved(page);
+		init_page_count(page);
+		__free_page(page);
+	}
+}
+#endif
+
+void mem_init(void)
+{
+	/* clear the zero-page */
+	memset((void *) empty_zero_page, 0, PAGE_SIZE);
+
+	/* Map in the area just after the brk now that kmalloc is about
+	 * to be turned on.
+	 */
+	brk_end = (unsigned long) UML_ROUND_UP(sbrk(0));
+	map_cb(NULL);
+	initial_thread_cb(map_cb, NULL);
+	free_bootmem(__pa(brk_end), uml_reserved - brk_end);
+	uml_reserved = brk_end;
+
+	/* this will put all low memory onto the freelists */
+	totalram_pages = free_all_bootmem();
+	max_low_pfn = totalram_pages;
+#ifdef CONFIG_HIGHMEM
+	totalhigh_pages = highmem >> PAGE_SHIFT;
+	totalram_pages += totalhigh_pages;
+#endif
+	num_physpages = totalram_pages;
+	max_pfn = totalram_pages;
+	printk(KERN_INFO "Memory: %luk available\n", 
+	       (unsigned long) nr_free_pages() << (PAGE_SHIFT-10));
+	kmalloc_ok = 1;
+
+#ifdef CONFIG_HIGHMEM
+	setup_highmem(end_iomem, highmem);
+#endif
+}
+
+/*
+ * Create a page table and place a pointer to it in a middle page
+ * directory entry.
+ */
+static void __init one_page_table_init(pmd_t *pmd)
+{
+	if (pmd_none(*pmd)) {
+		pte_t *pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+		set_pmd(pmd, __pmd(_KERNPG_TABLE +
+					   (unsigned long) __pa(pte)));
+		if (pte != pte_offset_kernel(pmd, 0))
+			BUG();
+	}
+}
+
+static void __init one_md_table_init(pud_t *pud)
+{
+#ifdef CONFIG_3_LEVEL_PGTABLES
+	pmd_t *pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+	set_pud(pud, __pud(_KERNPG_TABLE + (unsigned long) __pa(pmd_table)));
+	if (pmd_table != pmd_offset(pud, 0))
+		BUG();
+#endif
+}
+
+static void __init fixrange_init(unsigned long start, unsigned long end, 
+				 pgd_t *pgd_base)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	int i, j;
+	unsigned long vaddr;
+
+	vaddr = start;
+	i = pgd_index(vaddr);
+	j = pmd_index(vaddr);
+	pgd = pgd_base + i;
+
+	for ( ; (i < PTRS_PER_PGD) && (vaddr < end); pgd++, i++) {
+		pud = pud_offset(pgd, vaddr);
+		if (pud_none(*pud))
+			one_md_table_init(pud);
+		pmd = pmd_offset(pud, vaddr);
+		for (; (j < PTRS_PER_PMD) && (vaddr != end); pmd++, j++) {
+			one_page_table_init(pmd);
+			vaddr += PMD_SIZE;
+		}
+		j = 0;
+	}
+}
+
+#ifdef CONFIG_HIGHMEM
+pte_t *kmap_pte;
+pgprot_t kmap_prot;
+
+#define kmap_get_fixmap_pte(vaddr)					\
+	pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)),\
+ 			  (vaddr)), (vaddr))
+
+static void __init kmap_init(void)
+{
+	unsigned long kmap_vstart;
+
+	/* cache the first kmap pte */
+	kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
+	kmap_pte = kmap_get_fixmap_pte(kmap_vstart);
+
+	kmap_prot = PAGE_KERNEL;
+}
+
+static void init_highmem(void)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long vaddr;
+
+	/*
+	 * Permanent kmaps:
+	 */
+	vaddr = PKMAP_BASE;
+	fixrange_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, swapper_pg_dir);
+
+	pgd = swapper_pg_dir + pgd_index(vaddr);
+	pud = pud_offset(pgd, vaddr);
+	pmd = pmd_offset(pud, vaddr);
+	pte = pte_offset_kernel(pmd, vaddr);
+	pkmap_page_table = pte;
+
+	kmap_init();
+}
+#endif /* CONFIG_HIGHMEM */
+
+static void __init fixaddr_user_init( void)
+{
+#ifdef CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA
+	long size = FIXADDR_USER_END - FIXADDR_USER_START;
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long paddr, vaddr = FIXADDR_USER_START;
+
+	if (  ! size )
+		return;
+
+	fixrange_init( FIXADDR_USER_START, FIXADDR_USER_END, swapper_pg_dir);
+	paddr = (unsigned long)alloc_bootmem_low_pages( size);
+	memcpy( (void *)paddr, (void *)FIXADDR_USER_START, size);
+	paddr = __pa(paddr);
+	for ( ; size > 0; size-=PAGE_SIZE, vaddr+=PAGE_SIZE, paddr+=PAGE_SIZE){
+		pgd = swapper_pg_dir + pgd_index(vaddr);
+		pud = pud_offset(pgd, vaddr);
+		pmd = pmd_offset(pud, vaddr);
+		pte = pte_offset_kernel(pmd, vaddr);
+		pte_set_val( (*pte), paddr, PAGE_READONLY);
+	}
+#endif
+}
+
+void paging_init(void)
+{
+	unsigned long zones_size[MAX_NR_ZONES], vaddr;
+	int i;
+
+	empty_zero_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE);
+	empty_bad_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE);
+	for(i = 0; i < ARRAY_SIZE(zones_size); i++)
+		zones_size[i] = 0;
+
+	zones_size[ZONE_NORMAL] = (end_iomem >> PAGE_SHIFT) -
+		(uml_physmem >> PAGE_SHIFT);
+#ifdef CONFIG_HIGHMEM
+	zones_size[ZONE_HIGHMEM] = highmem >> PAGE_SHIFT;
+#endif
+	free_area_init(zones_size);
+
+	/*
+	 * Fixed mappings, only the page table structure has to be
+	 * created - mappings will be set by set_fixmap():
+	 */
+	vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
+	fixrange_init(vaddr, FIXADDR_TOP, swapper_pg_dir);
+
+	fixaddr_user_init();
+
+#ifdef CONFIG_HIGHMEM
+	init_highmem();
+#endif
+}
+
+struct page *arch_validate(struct page *page, gfp_t mask, int order)
+{
+	unsigned long addr, zero = 0;
+	int i;
+
+ again:
+	if(page == NULL)
+		return page;
+	if(PageHighMem(page))
+		return page;
+
+	addr = (unsigned long) page_address(page);
+	for(i = 0; i < (1 << order); i++){
+		current->thread.fault_addr = (void *) addr;
+		if(__do_copy_to_user((void __user *) addr, &zero,
+				     sizeof(zero),
+				     &current->thread.fault_addr,
+				     &current->thread.fault_catcher)){
+			if(!(mask & __GFP_WAIT))
+				return NULL;
+			else break;
+		}
+		addr += PAGE_SIZE;
+	}
+
+	if(i == (1 << order))
+		return page;
+	page = alloc_pages(mask, order);
+	goto again;
+}
+
+/* This can't do anything because nothing in the kernel image can be freed
+ * since it's not in kernel physical memory.
+ */
+
+void free_initmem(void)
+{
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+	if (start < end)
+		printk ("Freeing initrd memory: %ldk freed\n", 
+			(end - start) >> 10);
+	for (; start < end; start += PAGE_SIZE) {
+		ClearPageReserved(virt_to_page(start));
+		init_page_count(virt_to_page(start));
+		free_page(start);
+		totalram_pages++;
+	}
+}
+#endif
+
+void show_mem(void)
+{
+	int pfn, total = 0, reserved = 0;
+	int shared = 0, cached = 0;
+	int highmem = 0;
+	struct page *page;
+
+	printk("Mem-info:\n");
+	show_free_areas();
+	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+	pfn = max_mapnr;
+	while(pfn-- > 0) {
+		page = pfn_to_page(pfn);
+		total++;
+		if(PageHighMem(page))
+			highmem++;
+		if(PageReserved(page))
+			reserved++;
+		else if(PageSwapCache(page))
+			cached++;
+		else if(page_count(page))
+			shared += page_count(page) - 1;
+	}
+	printk("%d pages of RAM\n", total);
+	printk("%d pages of HIGHMEM\n", highmem);
+	printk("%d reserved pages\n", reserved);
+	printk("%d pages shared\n", shared);
+	printk("%d pages swap cached\n", cached);
+}
+
+/*
+ * Allocate and free page tables.
+ */
+
+pgd_t *pgd_alloc(struct mm_struct *mm)
+{
+	pgd_t *pgd = (pgd_t *)__get_free_page(GFP_KERNEL);
+
+	if (pgd) {
+		memset(pgd, 0, USER_PTRS_PER_PGD * sizeof(pgd_t));
+		memcpy(pgd + USER_PTRS_PER_PGD, 
+		       swapper_pg_dir + USER_PTRS_PER_PGD, 
+		       (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t));
+	}
+	return pgd;
+}
+
+void pgd_free(pgd_t *pgd)
+{
+	free_page((unsigned long) pgd);
+}
+
+pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+{
+	pte_t *pte;
+
+	pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
+	return pte;
+}
+
+struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
+{
+	struct page *pte;
+
+	pte = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
+	return pte;
+}

arch/um/kernel/physmem.c

@@ -0,0 +1,458 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/mm.h"
+#include "linux/rbtree.h"
+#include "linux/slab.h"
+#include "linux/vmalloc.h"
+#include "linux/bootmem.h"
+#include "linux/module.h"
+#include "linux/pfn.h"
+#include "asm/types.h"
+#include "asm/pgtable.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "mode_kern.h"
+#include "mem.h"
+#include "mem_user.h"
+#include "os.h"
+#include "kern.h"
+#include "init.h"
+
+struct phys_desc {
+	struct rb_node rb;
+	int fd;
+	__u64 offset;
+	void *virt;
+	unsigned long phys;
+	struct list_head list;
+};
+
+static struct rb_root phys_mappings = RB_ROOT;
+
+static struct rb_node **find_rb(void *virt)
+{
+	struct rb_node **n = &phys_mappings.rb_node;
+	struct phys_desc *d;
+
+	while(*n != NULL){
+		d = rb_entry(*n, struct phys_desc, rb);
+		if(d->virt == virt)
+			return n;
+
+		if(d->virt > virt)
+			n = &(*n)->rb_left;
+		else
+			n = &(*n)->rb_right;
+	}
+
+	return n;
+}
+
+static struct phys_desc *find_phys_mapping(void *virt)
+{
+	struct rb_node **n = find_rb(virt);
+
+	if(*n == NULL)
+		return NULL;
+
+	return rb_entry(*n, struct phys_desc, rb);
+}
+
+static void insert_phys_mapping(struct phys_desc *desc)
+{
+	struct rb_node **n = find_rb(desc->virt);
+
+	if(*n != NULL)
+		panic("Physical remapping for %p already present",
+		      desc->virt);
+
+	rb_link_node(&desc->rb, rb_parent(*n), n);
+	rb_insert_color(&desc->rb, &phys_mappings);
+}
+
+LIST_HEAD(descriptor_mappings);
+
+struct desc_mapping {
+	int fd;
+	struct list_head list;
+	struct list_head pages;
+};
+
+static struct desc_mapping *find_mapping(int fd)
+{
+	struct desc_mapping *desc;
+	struct list_head *ele;
+
+	list_for_each(ele, &descriptor_mappings){
+		desc = list_entry(ele, struct desc_mapping, list);
+		if(desc->fd == fd)
+			return desc;
+	}
+
+	return NULL;
+}
+
+static struct desc_mapping *descriptor_mapping(int fd)
+{
+	struct desc_mapping *desc;
+
+	desc = find_mapping(fd);
+	if(desc != NULL)
+		return desc;
+
+	desc = kmalloc(sizeof(*desc), GFP_ATOMIC);
+	if(desc == NULL)
+		return NULL;
+
+	*desc = ((struct desc_mapping)
+		{ .fd =		fd,
+		  .list =	LIST_HEAD_INIT(desc->list),
+		  .pages =	LIST_HEAD_INIT(desc->pages) });
+	list_add(&desc->list, &descriptor_mappings);
+
+	return desc;
+}
+
+int physmem_subst_mapping(void *virt, int fd, __u64 offset, int w)
+{
+	struct desc_mapping *fd_maps;
+	struct phys_desc *desc;
+	unsigned long phys;
+	int err;
+
+	fd_maps = descriptor_mapping(fd);
+	if(fd_maps == NULL)
+		return -ENOMEM;
+
+	phys = __pa(virt);
+	desc = find_phys_mapping(virt);
+	if(desc != NULL)
+		panic("Address 0x%p is already substituted\n", virt);
+
+	err = -ENOMEM;
+	desc = kmalloc(sizeof(*desc), GFP_ATOMIC);
+	if(desc == NULL)
+		goto out;
+
+	*desc = ((struct phys_desc)
+		{ .fd =			fd,
+		  .offset =		offset,
+		  .virt =		virt,
+		  .phys =		__pa(virt),
+		  .list = 		LIST_HEAD_INIT(desc->list) });
+	insert_phys_mapping(desc);
+
+	list_add(&desc->list, &fd_maps->pages);
+
+	virt = (void *) ((unsigned long) virt & PAGE_MASK);
+	err = os_map_memory(virt, fd, offset, PAGE_SIZE, 1, w, 0);
+	if(!err)
+		goto out;
+
+	rb_erase(&desc->rb, &phys_mappings);
+	kfree(desc);
+ out:
+	return err;
+}
+
+static int physmem_fd = -1;
+
+static void remove_mapping(struct phys_desc *desc)
+{
+	void *virt = desc->virt;
+	int err;
+
+	rb_erase(&desc->rb, &phys_mappings);
+	list_del(&desc->list);
+	kfree(desc);
+
+	err = os_map_memory(virt, physmem_fd, __pa(virt), PAGE_SIZE, 1, 1, 0);
+	if(err)
+		panic("Failed to unmap block device page from physical memory, "
+		      "errno = %d", -err);
+}
+
+int physmem_remove_mapping(void *virt)
+{
+	struct phys_desc *desc;
+
+	virt = (void *) ((unsigned long) virt & PAGE_MASK);
+	desc = find_phys_mapping(virt);
+	if(desc == NULL)
+		return 0;
+
+	remove_mapping(desc);
+	return 1;
+}
+
+void physmem_forget_descriptor(int fd)
+{
+	struct desc_mapping *desc;
+	struct phys_desc *page;
+	struct list_head *ele, *next;
+	__u64 offset;
+	void *addr;
+	int err;
+
+	desc = find_mapping(fd);
+	if(desc == NULL)
+		return;
+
+	list_for_each_safe(ele, next, &desc->pages){
+		page = list_entry(ele, struct phys_desc, list);
+		offset = page->offset;
+		addr = page->virt;
+		remove_mapping(page);
+		err = os_seek_file(fd, offset);
+		if(err)
+			panic("physmem_forget_descriptor - failed to seek "
+			      "to %lld in fd %d, error = %d\n",
+			      offset, fd, -err);
+		err = os_read_file(fd, addr, PAGE_SIZE);
+		if(err < 0)
+			panic("physmem_forget_descriptor - failed to read "
+			      "from fd %d to 0x%p, error = %d\n",
+			      fd, addr, -err);
+	}
+
+	list_del(&desc->list);
+	kfree(desc);
+}
+
+EXPORT_SYMBOL(physmem_forget_descriptor);
+EXPORT_SYMBOL(physmem_remove_mapping);
+EXPORT_SYMBOL(physmem_subst_mapping);
+
+void arch_free_page(struct page *page, int order)
+{
+	void *virt;
+	int i;
+
+	for(i = 0; i < (1 << order); i++){
+		virt = __va(page_to_phys(page + i));
+		physmem_remove_mapping(virt);
+	}
+}
+
+int is_remapped(void *virt)
+{
+ 	struct phys_desc *desc = find_phys_mapping(virt);
+
+	return desc != NULL;
+}
+
+/* Changed during early boot */
+unsigned long high_physmem;
+
+extern unsigned long long physmem_size;
+
+int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem)
+{
+	struct page *p, *map;
+	unsigned long phys_len, phys_pages, highmem_len, highmem_pages;
+	unsigned long iomem_len, iomem_pages, total_len, total_pages;
+	int i;
+
+	phys_pages = physmem >> PAGE_SHIFT;
+	phys_len = phys_pages * sizeof(struct page);
+
+	iomem_pages = iomem >> PAGE_SHIFT;
+	iomem_len = iomem_pages * sizeof(struct page);
+
+	highmem_pages = highmem >> PAGE_SHIFT;
+	highmem_len = highmem_pages * sizeof(struct page);
+
+	total_pages = phys_pages + iomem_pages + highmem_pages;
+	total_len = phys_len + iomem_len + highmem_len;
+
+	if(kmalloc_ok){
+		map = kmalloc(total_len, GFP_KERNEL);
+		if(map == NULL)
+			map = vmalloc(total_len);
+	}
+	else map = alloc_bootmem_low_pages(total_len);
+
+	if(map == NULL)
+		return -ENOMEM;
+
+	for(i = 0; i < total_pages; i++){
+		p = &map[i];
+		memset(p, 0, sizeof(struct page));
+		SetPageReserved(p);
+		INIT_LIST_HEAD(&p->lru);
+	}
+
+	max_mapnr = total_pages;
+	return 0;
+}
+
+/* Changed during early boot */
+static unsigned long kmem_top = 0;
+
+unsigned long get_kmem_end(void)
+{
+	if(kmem_top == 0)
+		kmem_top = CHOOSE_MODE(kmem_end_tt, kmem_end_skas);
+	return kmem_top;
+}
+
+void map_memory(unsigned long virt, unsigned long phys, unsigned long len,
+		int r, int w, int x)
+{
+	__u64 offset;
+	int fd, err;
+
+	fd = phys_mapping(phys, &offset);
+	err = os_map_memory((void *) virt, fd, offset, len, r, w, x);
+	if(err) {
+		if(err == -ENOMEM)
+			printk("try increasing the host's "
+			       "/proc/sys/vm/max_map_count to <physical "
+			       "memory size>/4096\n");
+		panic("map_memory(0x%lx, %d, 0x%llx, %ld, %d, %d, %d) failed, "
+		      "err = %d\n", virt, fd, offset, len, r, w, x, err);
+	}
+}
+
+extern int __syscall_stub_start;
+
+void setup_physmem(unsigned long start, unsigned long reserve_end,
+		   unsigned long len, unsigned long long highmem)
+{
+	unsigned long reserve = reserve_end - start;
+	int pfn = PFN_UP(__pa(reserve_end));
+	int delta = (len - reserve) >> PAGE_SHIFT;
+	int err, offset, bootmap_size;
+
+	physmem_fd = create_mem_file(len + highmem);
+
+	offset = uml_reserved - uml_physmem;
+	err = os_map_memory((void *) uml_reserved, physmem_fd, offset,
+			    len - offset, 1, 1, 0);
+	if(err < 0){
+		os_print_error(err, "Mapping memory");
+		exit(1);
+	}
+
+	/* Special kludge - This page will be mapped in to userspace processes
+	 * from physmem_fd, so it needs to be written out there.
+	 */
+	os_seek_file(physmem_fd, __pa(&__syscall_stub_start));
+	os_write_file(physmem_fd, &__syscall_stub_start, PAGE_SIZE);
+
+	bootmap_size = init_bootmem(pfn, pfn + delta);
+	free_bootmem(__pa(reserve_end) + bootmap_size,
+		     len - bootmap_size - reserve);
+}
+
+int phys_mapping(unsigned long phys, __u64 *offset_out)
+{
+	struct phys_desc *desc = find_phys_mapping(__va(phys & PAGE_MASK));
+	int fd = -1;
+
+	if(desc != NULL){
+		fd = desc->fd;
+		*offset_out = desc->offset;
+	}
+	else if(phys < physmem_size){
+		fd = physmem_fd;
+		*offset_out = phys;
+	}
+	else if(phys < __pa(end_iomem)){
+		struct iomem_region *region = iomem_regions;
+
+		while(region != NULL){
+			if((phys >= region->phys) &&
+			   (phys < region->phys + region->size)){
+				fd = region->fd;
+				*offset_out = phys - region->phys;
+				break;
+			}
+			region = region->next;
+		}
+	}
+	else if(phys < __pa(end_iomem) + highmem){
+		fd = physmem_fd;
+		*offset_out = phys - iomem_size;
+	}
+
+	return fd;
+}
+
+static int __init uml_mem_setup(char *line, int *add)
+{
+	char *retptr;
+	physmem_size = memparse(line,&retptr);
+	return 0;
+}
+__uml_setup("mem=", uml_mem_setup,
+"mem=<Amount of desired ram>\n"
+"    This controls how much \"physical\" memory the kernel allocates\n"
+"    for the system. The size is specified as a number followed by\n"
+"    one of 'k', 'K', 'm', 'M', which have the obvious meanings.\n"
+"    This is not related to the amount of memory in the host.  It can\n"
+"    be more, and the excess, if it's ever used, will just be swapped out.\n"
+"	Example: mem=64M\n\n"
+);
+
+extern int __init parse_iomem(char *str, int *add);
+
+__uml_setup("iomem=", parse_iomem,
+"iomem=<name>,<file>\n"
+"    Configure <file> as an IO memory region named <name>.\n\n"
+);
+
+/*
+ * This list is constructed in parse_iomem and addresses filled in in
+ * setup_iomem, both of which run during early boot.  Afterwards, it's
+ * unchanged.
+ */
+struct iomem_region *iomem_regions = NULL;
+
+/* Initialized in parse_iomem */
+int iomem_size = 0;
+
+unsigned long find_iomem(char *driver, unsigned long *len_out)
+{
+	struct iomem_region *region = iomem_regions;
+
+	while(region != NULL){
+		if(!strcmp(region->driver, driver)){
+			*len_out = region->size;
+			return region->virt;
+		}
+
+		region = region->next;
+	}
+
+	return 0;
+}
+
+int setup_iomem(void)
+{
+	struct iomem_region *region = iomem_regions;
+	unsigned long iomem_start = high_physmem + PAGE_SIZE;
+	int err;
+
+	while(region != NULL){
+		err = os_map_memory((void *) iomem_start, region->fd, 0,
+				    region->size, 1, 1, 0);
+		if(err)
+			printk("Mapping iomem region for driver '%s' failed, "
+			       "errno = %d\n", region->driver, -err);
+		else {
+			region->virt = iomem_start;
+			region->phys = __pa(region->virt);
+		}
+
+		iomem_start += region->size + PAGE_SIZE;
+		region = region->next;
+	}
+
+	return 0;
+}
+
+__initcall(setup_iomem);

arch/um/kernel/process.c

@@ -0,0 +1,484 @@
+/*
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright 2003 PathScale, Inc.
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/sched.h"
+#include "linux/interrupt.h"
+#include "linux/string.h"
+#include "linux/mm.h"
+#include "linux/slab.h"
+#include "linux/utsname.h"
+#include "linux/fs.h"
+#include "linux/utime.h"
+#include "linux/smp_lock.h"
+#include "linux/module.h"
+#include "linux/init.h"
+#include "linux/capability.h"
+#include "linux/vmalloc.h"
+#include "linux/spinlock.h"
+#include "linux/proc_fs.h"
+#include "linux/ptrace.h"
+#include "linux/random.h"
+#include "linux/personality.h"
+#include "asm/unistd.h"
+#include "asm/mman.h"
+#include "asm/segment.h"
+#include "asm/stat.h"
+#include "asm/pgtable.h"
+#include "asm/processor.h"
+#include "asm/tlbflush.h"
+#include "asm/uaccess.h"
+#include "asm/user.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "signal_kern.h"
+#include "init.h"
+#include "irq_user.h"
+#include "mem_user.h"
+#include "tlb.h"
+#include "frame_kern.h"
+#include "sigcontext.h"
+#include "os.h"
+#include "mode.h"
+#include "mode_kern.h"
+#include "choose-mode.h"
+#include "um_malloc.h"
+
+/* This is a per-cpu array.  A processor only modifies its entry and it only
+ * cares about its entry, so it's OK if another processor is modifying its
+ * entry.
+ */
+struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } };
+
+int external_pid(void *t)
+{
+	struct task_struct *task = t ? t : current;
+
+	return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task));
+}
+
+int pid_to_processor_id(int pid)
+{
+	int i;
+
+	for(i = 0; i < ncpus; i++){
+		if(cpu_tasks[i].pid == pid) return(i);
+	}
+	return(-1);
+}
+
+void free_stack(unsigned long stack, int order)
+{
+	free_pages(stack, order);
+}
+
+unsigned long alloc_stack(int order, int atomic)
+{
+	unsigned long page;
+	gfp_t flags = GFP_KERNEL;
+
+	if (atomic)
+		flags = GFP_ATOMIC;
+	page = __get_free_pages(flags, order);
+	if(page == 0)
+		return(0);
+	stack_protections(page);
+	return(page);
+}
+
+int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
+{
+	int pid;
+
+	current->thread.request.u.thread.proc = fn;
+	current->thread.request.u.thread.arg = arg;
+	pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0,
+		      &current->thread.regs, 0, NULL, NULL);
+	if(pid < 0)
+		panic("do_fork failed in kernel_thread, errno = %d", pid);
+	return(pid);
+}
+
+void set_current(void *t)
+{
+	struct task_struct *task = t;
+
+	cpu_tasks[task_thread_info(task)->cpu] = ((struct cpu_task)
+		{ external_pid(task), task });
+}
+
+void *_switch_to(void *prev, void *next, void *last)
+{
+	struct task_struct *from = prev;
+	struct task_struct *to= next;
+
+	to->thread.prev_sched = from;
+	set_current(to);
+
+	do {
+		current->thread.saved_task = NULL ;
+		CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next);
+		if(current->thread.saved_task)
+			show_regs(&(current->thread.regs));
+		next= current->thread.saved_task;
+		prev= current;
+	} while(current->thread.saved_task);
+
+	return(current->thread.prev_sched);
+
+}
+
+void interrupt_end(void)
+{
+	if(need_resched()) schedule();
+	if(test_tsk_thread_flag(current, TIF_SIGPENDING)) do_signal();
+}
+
+void release_thread(struct task_struct *task)
+{
+	CHOOSE_MODE(release_thread_tt(task), release_thread_skas(task));
+}
+
+void exit_thread(void)
+{
+	unprotect_stack((unsigned long) current_thread);
+}
+
+void *get_current(void)
+{
+	return(current);
+}
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
+		unsigned long stack_top, struct task_struct * p,
+		struct pt_regs *regs)
+{
+	int ret;
+
+	p->thread = (struct thread_struct) INIT_THREAD;
+	ret = CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr,
+				clone_flags, sp, stack_top, p, regs);
+
+	if (ret || !current->thread.forking)
+		goto out;
+
+	clear_flushed_tls(p);
+
+	/*
+	 * Set a new TLS for the child thread?
+	 */
+	if (clone_flags & CLONE_SETTLS)
+		ret = arch_copy_tls(p);
+
+out:
+	return ret;
+}
+
+void initial_thread_cb(void (*proc)(void *), void *arg)
+{
+	int save_kmalloc_ok = kmalloc_ok;
+
+	kmalloc_ok = 0;
+	CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc,
+			 arg);
+	kmalloc_ok = save_kmalloc_ok;
+}
+
+unsigned long stack_sp(unsigned long page)
+{
+	return(page + PAGE_SIZE - sizeof(void *));
+}
+
+int current_pid(void)
+{
+	return(current->pid);
+}
+
+void default_idle(void)
+{
+	CHOOSE_MODE(uml_idle_timer(), (void) 0);
+
+	while(1){
+		/* endless idle loop with no priority at all */
+
+		/*
+		 * although we are an idle CPU, we do not want to
+		 * get into the scheduler unnecessarily.
+		 */
+		if(need_resched())
+			schedule();
+
+		idle_sleep(10);
+	}
+}
+
+void cpu_idle(void)
+{
+	CHOOSE_MODE(init_idle_tt(), init_idle_skas());
+}
+
+int page_size(void)
+{
+	return(PAGE_SIZE);
+}
+
+void *um_virt_to_phys(struct task_struct *task, unsigned long addr,
+		      pte_t *pte_out)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	pte_t ptent;
+
+	if(task->mm == NULL)
+		return(ERR_PTR(-EINVAL));
+	pgd = pgd_offset(task->mm, addr);
+	if(!pgd_present(*pgd))
+		return(ERR_PTR(-EINVAL));
+
+	pud = pud_offset(pgd, addr);
+	if(!pud_present(*pud))
+		return(ERR_PTR(-EINVAL));
+
+	pmd = pmd_offset(pud, addr);
+	if(!pmd_present(*pmd))
+		return(ERR_PTR(-EINVAL));
+
+	pte = pte_offset_kernel(pmd, addr);
+	ptent = *pte;
+	if(!pte_present(ptent))
+		return(ERR_PTR(-EINVAL));
+
+	if(pte_out != NULL)
+		*pte_out = ptent;
+	return((void *) (pte_val(ptent) & PAGE_MASK) + (addr & ~PAGE_MASK));
+}
+
+char *current_cmd(void)
+{
+#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM)
+	return("(Unknown)");
+#else
+	void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL);
+	return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr);
+#endif
+}
+
+void force_sigbus(void)
+{
+	printk(KERN_ERR "Killing pid %d because of a lack of memory\n",
+	       current->pid);
+	lock_kernel();
+	sigaddset(&current->pending.signal, SIGBUS);
+	recalc_sigpending();
+	current->flags |= PF_SIGNALED;
+	do_exit(SIGBUS | 0x80);
+}
+
+void dump_thread(struct pt_regs *regs, struct user *u)
+{
+}
+
+void enable_hlt(void)
+{
+	panic("enable_hlt");
+}
+
+EXPORT_SYMBOL(enable_hlt);
+
+void disable_hlt(void)
+{
+	panic("disable_hlt");
+}
+
+EXPORT_SYMBOL(disable_hlt);
+
+void *um_kmalloc(int size)
+{
+	return kmalloc(size, GFP_KERNEL);
+}
+
+void *um_kmalloc_atomic(int size)
+{
+	return kmalloc(size, GFP_ATOMIC);
+}
+
+void *um_vmalloc(int size)
+{
+	return vmalloc(size);
+}
+
+void *um_vmalloc_atomic(int size)
+{
+	return __vmalloc(size, GFP_ATOMIC | __GFP_HIGHMEM, PAGE_KERNEL);
+}
+
+int __cant_sleep(void) {
+	return in_atomic() || irqs_disabled() || in_interrupt();
+	/* Is in_interrupt() really needed? */
+}
+
+unsigned long get_fault_addr(void)
+{
+	return((unsigned long) current->thread.fault_addr);
+}
+
+EXPORT_SYMBOL(get_fault_addr);
+
+void not_implemented(void)
+{
+	printk(KERN_DEBUG "Something isn't implemented in here\n");
+}
+
+EXPORT_SYMBOL(not_implemented);
+
+int user_context(unsigned long sp)
+{
+	unsigned long stack;
+
+	stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER);
+	return(stack != (unsigned long) current_thread);
+}
+
+extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end;
+
+void do_uml_exitcalls(void)
+{
+	exitcall_t *call;
+
+	call = &__uml_exitcall_end;
+	while (--call >= &__uml_exitcall_begin)
+		(*call)();
+}
+
+char *uml_strdup(char *string)
+{
+	return kstrdup(string, GFP_KERNEL);
+}
+
+int copy_to_user_proc(void __user *to, void *from, int size)
+{
+	return(copy_to_user(to, from, size));
+}
+
+int copy_from_user_proc(void *to, void __user *from, int size)
+{
+	return(copy_from_user(to, from, size));
+}
+
+int clear_user_proc(void __user *buf, int size)
+{
+	return(clear_user(buf, size));
+}
+
+int strlen_user_proc(char __user *str)
+{
+	return(strlen_user(str));
+}
+
+int smp_sigio_handler(void)
+{
+#ifdef CONFIG_SMP
+	int cpu = current_thread->cpu;
+	IPI_handler(cpu);
+	if(cpu != 0)
+		return(1);
+#endif
+	return(0);
+}
+
+int cpu(void)
+{
+	return(current_thread->cpu);
+}
+
+static atomic_t using_sysemu = ATOMIC_INIT(0);
+int sysemu_supported;
+
+void set_using_sysemu(int value)
+{
+	if (value > sysemu_supported)
+		return;
+	atomic_set(&using_sysemu, value);
+}
+
+int get_using_sysemu(void)
+{
+	return atomic_read(&using_sysemu);
+}
+
+static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data)
+{
+	if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/
+		*eof = 1;
+
+	return strlen(buf);
+}
+
+static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned long count,void *data)
+{
+	char tmp[2];
+
+	if (copy_from_user(tmp, buf, 1))
+		return -EFAULT;
+
+	if (tmp[0] >= '0' && tmp[0] <= '2')
+		set_using_sysemu(tmp[0] - '0');
+	return count; /*We use the first char, but pretend to write everything*/
+}
+
+int __init make_proc_sysemu(void)
+{
+	struct proc_dir_entry *ent;
+	if (!sysemu_supported)
+		return 0;
+
+	ent = create_proc_entry("sysemu", 0600, &proc_root);
+
+	if (ent == NULL)
+	{
+		printk(KERN_WARNING "Failed to register /proc/sysemu\n");
+		return(0);
+	}
+
+	ent->read_proc  = proc_read_sysemu;
+	ent->write_proc = proc_write_sysemu;
+
+	return 0;
+}
+
+late_initcall(make_proc_sysemu);
+
+int singlestepping(void * t)
+{
+	struct task_struct *task = t ? t : current;
+
+	if ( ! (task->ptrace & PT_DTRACE) )
+		return(0);
+
+	if (task->thread.singlestep_syscall)
+		return(1);
+
+	return 2;
+}
+
+/*
+ * Only x86 and x86_64 have an arch_align_stack().
+ * All other arches have "#define arch_align_stack(x) (x)"
+ * in their asm/system.h
+ * As this is included in UML from asm-um/system-generic.h,
+ * we can use it to behave as the subarch does.
+ */
+#ifndef arch_align_stack
+unsigned long arch_align_stack(unsigned long sp)
+{
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+		sp -= get_random_int() % 8192;
+	return sp & ~0xf;
+}
+#endif

arch/um/kernel/ptrace.c

@@ -0,0 +1,321 @@
+/* 
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/mm.h"
+#include "linux/errno.h"
+#include "linux/smp_lock.h"
+#include "linux/security.h"
+#include "linux/ptrace.h"
+#include "linux/audit.h"
+#ifdef CONFIG_PROC_MM
+#include "linux/proc_mm.h"
+#endif
+#include "asm/ptrace.h"
+#include "asm/uaccess.h"
+#include "kern_util.h"
+#include "skas_ptrace.h"
+#include "sysdep/ptrace.h"
+#include "os.h"
+
+static inline void set_singlestepping(struct task_struct *child, int on)
+{
+        if (on)
+                child->ptrace |= PT_DTRACE;
+        else
+                child->ptrace &= ~PT_DTRACE;
+        child->thread.singlestep_syscall = 0;
+
+#ifdef SUBARCH_SET_SINGLESTEPPING
+        SUBARCH_SET_SINGLESTEPPING(child, on);
+#endif
+}
+
+/*
+ * Called by kernel/ptrace.c when detaching..
+ */
+void ptrace_disable(struct task_struct *child)
+{ 
+        set_singlestepping(child,0);
+}
+
+extern int peek_user(struct task_struct * child, long addr, long data);
+extern int poke_user(struct task_struct * child, long addr, long data);
+
+long arch_ptrace(struct task_struct *child, long request, long addr, long data)
+{
+	int i, ret;
+	unsigned long __user *p = (void __user *)(unsigned long)data;
+
+	switch (request) {
+		/* when I and D space are separate, these will need to be fixed. */
+	case PTRACE_PEEKTEXT: /* read word at location addr. */ 
+	case PTRACE_PEEKDATA: {
+		unsigned long tmp;
+		int copied;
+
+		ret = -EIO;
+		copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
+		if (copied != sizeof(tmp))
+			break;
+		ret = put_user(tmp, p);
+		break;
+	}
+
+	/* read the word at location addr in the USER area. */
+        case PTRACE_PEEKUSR:
+                ret = peek_user(child, addr, data);
+                break;
+
+	/* when I and D space are separate, this will have to be fixed. */
+	case PTRACE_POKETEXT: /* write the word at location addr. */
+	case PTRACE_POKEDATA:
+		ret = -EIO;
+		if (access_process_vm(child, addr, &data, sizeof(data), 
+				      1) != sizeof(data))
+			break;
+		ret = 0;
+		break;
+
+	case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
+                ret = poke_user(child, addr, data);
+                break;
+
+	case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */
+	case PTRACE_CONT: { /* restart after signal. */
+		ret = -EIO;
+		if (!valid_signal(data))
+			break;
+
+                set_singlestepping(child, 0);
+		if (request == PTRACE_SYSCALL) {
+			set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		}
+		else {
+			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		}
+		child->exit_code = data;
+		wake_up_process(child);
+		ret = 0;
+		break;
+	}
+
+/*
+ * make the child exit.  Best I can do is send it a sigkill. 
+ * perhaps it should be put in the status that it wants to 
+ * exit.
+ */
+	case PTRACE_KILL: {
+		ret = 0;
+		if (child->exit_state == EXIT_ZOMBIE)	/* already dead */
+			break;
+
+                set_singlestepping(child, 0);
+		child->exit_code = SIGKILL;
+		wake_up_process(child);
+		break;
+	}
+
+	case PTRACE_SINGLESTEP: {  /* set the trap flag. */
+		ret = -EIO;
+		if (!valid_signal(data))
+			break;
+		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+                set_singlestepping(child, 1);
+		child->exit_code = data;
+		/* give it a chance to run. */
+		wake_up_process(child);
+		ret = 0;
+		break;
+	}
+
+	case PTRACE_DETACH:
+		/* detach a process that was attached. */
+		ret = ptrace_detach(child, data);
+ 		break;
+
+#ifdef PTRACE_GETREGS
+	case PTRACE_GETREGS: { /* Get all gp regs from the child. */
+		if (!access_ok(VERIFY_WRITE, p, MAX_REG_OFFSET)) {
+			ret = -EIO;
+			break;
+		}
+		for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) {
+			__put_user(getreg(child, i), p);
+			p++;
+		}
+		ret = 0;
+		break;
+	}
+#endif
+#ifdef PTRACE_SETREGS
+	case PTRACE_SETREGS: { /* Set all gp regs in the child. */
+		unsigned long tmp = 0;
+		if (!access_ok(VERIFY_READ, p, MAX_REG_OFFSET)) {
+			ret = -EIO;
+			break;
+		}
+		for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) {
+			__get_user(tmp, p);
+			putreg(child, i, tmp);
+			p++;
+		}
+		ret = 0;
+		break;
+	}
+#endif
+#ifdef PTRACE_GETFPREGS
+	case PTRACE_GETFPREGS: /* Get the child FPU state. */
+		ret = get_fpregs(data, child);
+		break;
+#endif
+#ifdef PTRACE_SETFPREGS
+	case PTRACE_SETFPREGS: /* Set the child FPU state. */
+	        ret = set_fpregs(data, child);
+		break;
+#endif
+#ifdef PTRACE_GETFPXREGS
+	case PTRACE_GETFPXREGS: /* Get the child FPU state. */
+		ret = get_fpxregs(data, child);
+		break;
+#endif
+#ifdef PTRACE_SETFPXREGS
+	case PTRACE_SETFPXREGS: /* Set the child FPU state. */
+		ret = set_fpxregs(data, child);
+		break;
+#endif
+	case PTRACE_GET_THREAD_AREA:
+		ret = ptrace_get_thread_area(child, addr,
+					     (struct user_desc __user *) data);
+		break;
+
+	case PTRACE_SET_THREAD_AREA:
+		ret = ptrace_set_thread_area(child, addr,
+					     (struct user_desc __user *) data);
+		break;
+
+	case PTRACE_FAULTINFO: {
+		/* Take the info from thread->arch->faultinfo,
+		 * but transfer max. sizeof(struct ptrace_faultinfo).
+		 * On i386, ptrace_faultinfo is smaller!
+		 */
+		ret = copy_to_user(p, &child->thread.arch.faultinfo,
+				   sizeof(struct ptrace_faultinfo));
+		if(ret)
+			break;
+		break;
+	}
+
+#ifdef PTRACE_LDT
+	case PTRACE_LDT: {
+		struct ptrace_ldt ldt;
+
+		if(copy_from_user(&ldt, p, sizeof(ldt))){
+			ret = -EIO;
+			break;
+		}
+
+		/* This one is confusing, so just punt and return -EIO for 
+		 * now
+		 */
+		ret = -EIO;
+		break;
+	}
+#endif
+#ifdef CONFIG_PROC_MM
+	case PTRACE_SWITCH_MM: {
+		struct mm_struct *old = child->mm;
+		struct mm_struct *new = proc_mm_get_mm(data);
+
+		if(IS_ERR(new)){
+			ret = PTR_ERR(new);
+			break;
+		}
+
+		atomic_inc(&new->mm_users);
+		child->mm = new;
+		child->active_mm = new;
+		mmput(old);
+		ret = 0;
+		break;
+	}
+#endif
+#ifdef PTRACE_ARCH_PRCTL
+        case PTRACE_ARCH_PRCTL:
+                /* XXX Calls ptrace on the host - needs some SMP thinking */
+                ret = arch_prctl_skas(child, data, (void *) addr);
+                break;
+#endif
+	default:
+		ret = ptrace_request(child, request, addr, data);
+		break;
+	}
+
+	return ret;
+}
+
+void send_sigtrap(struct task_struct *tsk, union uml_pt_regs *regs,
+		  int error_code)
+{
+	struct siginfo info;
+
+	memset(&info, 0, sizeof(info));
+	info.si_signo = SIGTRAP;
+	info.si_code = TRAP_BRKPT;
+
+	/* User-mode eip? */
+	info.si_addr = UPT_IS_USER(regs) ? (void __user *) UPT_IP(regs) : NULL;
+
+	/* Send us the fakey SIGTRAP */
+	force_sig_info(SIGTRAP, &info, tsk);
+}
+
+/* XXX Check PT_DTRACE vs TIF_SINGLESTEP for singlestepping check and
+ * PT_PTRACED vs TIF_SYSCALL_TRACE for syscall tracing check
+ */
+void syscall_trace(union uml_pt_regs *regs, int entryexit)
+{
+	int is_singlestep = (current->ptrace & PT_DTRACE) && entryexit;
+	int tracesysgood;
+
+	if (unlikely(current->audit_context)) {
+		if (!entryexit)
+			audit_syscall_entry(HOST_AUDIT_ARCH,
+					    UPT_SYSCALL_NR(regs),
+					    UPT_SYSCALL_ARG1(regs),
+					    UPT_SYSCALL_ARG2(regs),
+					    UPT_SYSCALL_ARG3(regs),
+					    UPT_SYSCALL_ARG4(regs));
+		else audit_syscall_exit(AUDITSC_RESULT(UPT_SYSCALL_RET(regs)),
+                                        UPT_SYSCALL_RET(regs));
+	}
+
+	/* Fake a debug trap */
+	if (is_singlestep)
+		send_sigtrap(current, regs, 0);
+
+	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+		return;
+
+	if (!(current->ptrace & PT_PTRACED))
+		return;
+
+	/* the 0x80 provides a way for the tracing parent to distinguish
+	   between a syscall stop and SIGTRAP delivery */
+	tracesysgood = (current->ptrace & PT_TRACESYSGOOD);
+	ptrace_notify(SIGTRAP | (tracesysgood ? 0x80 : 0));
+
+	if (entryexit) /* force do_signal() --> is_syscall() */
+		set_thread_flag(TIF_SIGPENDING);
+
+	/* this isn't the same as continuing with a signal, but it will do
+	 * for normal use.  strace only continues with a signal if the
+	 * stopping signal is not SIGTRAP.  -brl
+	 */
+	if (current->exit_code) {
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+}

arch/um/kernel/reboot.c

@@ -0,0 +1,64 @@
+/* 
+ * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/module.h"
+#include "linux/sched.h"
+#include "asm/smp.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "os.h"
+#include "mode.h"
+#include "choose-mode.h"
+
+void (*pm_power_off)(void);
+
+#ifdef CONFIG_SMP
+static void kill_idlers(int me)
+{
+#ifdef CONFIG_MODE_TT
+	struct task_struct *p;
+	int i;
+
+	for(i = 0; i < ARRAY_SIZE(idle_threads); i++){
+		p = idle_threads[i];
+		if((p != NULL) && (p->thread.mode.tt.extern_pid != me))
+			os_kill_process(p->thread.mode.tt.extern_pid, 0);
+	}
+#endif
+}
+#endif
+
+static void kill_off_processes(void)
+{
+	CHOOSE_MODE(kill_off_processes_tt(), kill_off_processes_skas());
+#ifdef CONFIG_SMP
+	kill_idlers(os_getpid());
+#endif
+}
+
+void uml_cleanup(void)
+{
+        kmalloc_ok = 0;
+	do_uml_exitcalls();
+	kill_off_processes();
+}
+
+void machine_restart(char * __unused)
+{
+        uml_cleanup();
+	CHOOSE_MODE(reboot_tt(), reboot_skas());
+}
+
+void machine_power_off(void)
+{
+        uml_cleanup();
+	CHOOSE_MODE(halt_tt(), halt_skas());
+}
+
+void machine_halt(void)
+{
+	machine_power_off();
+}

arch/um/kernel/sigio.c

@@ -0,0 +1,56 @@
+/*
+ * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/list.h"
+#include "linux/slab.h"
+#include "linux/signal.h"
+#include "linux/interrupt.h"
+#include "init.h"
+#include "sigio.h"
+#include "irq_user.h"
+#include "irq_kern.h"
+#include "os.h"
+
+/* Protected by sigio_lock() called from write_sigio_workaround */
+static int sigio_irq_fd = -1;
+
+static irqreturn_t sigio_interrupt(int irq, void *data)
+{
+	char c;
+
+	os_read_file(sigio_irq_fd, &c, sizeof(c));
+	reactivate_fd(sigio_irq_fd, SIGIO_WRITE_IRQ);
+	return IRQ_HANDLED;
+}
+
+int write_sigio_irq(int fd)
+{
+	int err;
+
+	err = um_request_irq(SIGIO_WRITE_IRQ, fd, IRQ_READ, sigio_interrupt,
+			     IRQF_DISABLED|IRQF_SAMPLE_RANDOM, "write sigio",
+			     NULL);
+	if(err){
+		printk("write_sigio_irq : um_request_irq failed, err = %d\n",
+		       err);
+		return -1;
+	}
+	sigio_irq_fd = fd;
+	return 0;
+}
+
+/* These are called from os-Linux/sigio.c to protect its pollfds arrays. */
+static DEFINE_SPINLOCK(sigio_spinlock);
+
+void sigio_lock(void)
+{
+	spin_lock(&sigio_spinlock);
+}
+
+void sigio_unlock(void)
+{
+	spin_unlock(&sigio_spinlock);
+}

arch/um/kernel/signal.c

@@ -0,0 +1,190 @@
+/*
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/stddef.h"
+#include "linux/sys.h"
+#include "linux/sched.h"
+#include "linux/wait.h"
+#include "linux/kernel.h"
+#include "linux/smp_lock.h"
+#include "linux/module.h"
+#include "linux/slab.h"
+#include "linux/tty.h"
+#include "linux/binfmts.h"
+#include "linux/ptrace.h"
+#include "asm/signal.h"
+#include "asm/uaccess.h"
+#include "asm/unistd.h"
+#include "user_util.h"
+#include "asm/ucontext.h"
+#include "kern_util.h"
+#include "signal_kern.h"
+#include "kern.h"
+#include "frame_kern.h"
+#include "sigcontext.h"
+#include "mode.h"
+
+EXPORT_SYMBOL(block_signals);
+EXPORT_SYMBOL(unblock_signals);
+
+#define _S(nr) (1<<((nr)-1))
+
+#define _BLOCKABLE (~(_S(SIGKILL) | _S(SIGSTOP)))
+
+/*
+ * OK, we're invoking a handler
+ */
+static int handle_signal(struct pt_regs *regs, unsigned long signr,
+			 struct k_sigaction *ka, siginfo_t *info,
+			 sigset_t *oldset)
+{
+	unsigned long sp;
+	int err;
+
+	/* Always make any pending restarted system calls return -EINTR */
+	current_thread_info()->restart_block.fn = do_no_restart_syscall;
+
+	/* Did we come from a system call? */
+	if(PT_REGS_SYSCALL_NR(regs) >= 0){
+		/* If so, check system call restarting.. */
+		switch(PT_REGS_SYSCALL_RET(regs)){
+		case -ERESTART_RESTARTBLOCK:
+		case -ERESTARTNOHAND:
+			PT_REGS_SYSCALL_RET(regs) = -EINTR;
+			break;
+
+		case -ERESTARTSYS:
+			if (!(ka->sa.sa_flags & SA_RESTART)) {
+				PT_REGS_SYSCALL_RET(regs) = -EINTR;
+				break;
+			}
+		/* fallthrough */
+		case -ERESTARTNOINTR:
+			PT_REGS_RESTART_SYSCALL(regs);
+			PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs);
+			break;
+		}
+	}
+
+	sp = PT_REGS_SP(regs);
+	if((ka->sa.sa_flags & SA_ONSTACK) && (sas_ss_flags(sp) == 0))
+		sp = current->sas_ss_sp + current->sas_ss_size;
+
+#ifdef CONFIG_ARCH_HAS_SC_SIGNALS
+	if(!(ka->sa.sa_flags & SA_SIGINFO))
+		err = setup_signal_stack_sc(sp, signr, ka, regs, oldset);
+	else
+#endif
+		err = setup_signal_stack_si(sp, signr, ka, regs, info, oldset);
+
+	if(err){
+		spin_lock_irq(&current->sighand->siglock);
+		current->blocked = *oldset;
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+		force_sigsegv(signr, current);
+	} else {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked, &current->blocked,
+			  &ka->sa.sa_mask);
+		 if(!(ka->sa.sa_flags & SA_NODEFER))
+			sigaddset(&current->blocked, signr);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	return err;
+}
+
+static int kern_do_signal(struct pt_regs *regs)
+{
+	struct k_sigaction ka_copy;
+	siginfo_t info;
+	sigset_t *oldset;
+	int sig, handled_sig = 0;
+
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
+		oldset = &current->blocked;
+
+	while((sig = get_signal_to_deliver(&info, &ka_copy, regs, NULL)) > 0){
+		handled_sig = 1;
+		/* Whee!  Actually deliver the signal.  */
+		if(!handle_signal(regs, sig, &ka_copy, &info, oldset)){
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+			break;
+		}
+	}
+
+	/* Did we come from a system call? */
+	if(!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)){
+		/* Restart the system call - no handlers present */
+		switch(PT_REGS_SYSCALL_RET(regs)){
+		case -ERESTARTNOHAND:
+		case -ERESTARTSYS:
+		case -ERESTARTNOINTR:
+			PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs);
+			PT_REGS_RESTART_SYSCALL(regs);
+			break;
+		case -ERESTART_RESTARTBLOCK:
+			PT_REGS_ORIG_SYSCALL(regs) = __NR_restart_syscall;
+			PT_REGS_RESTART_SYSCALL(regs);
+			break;
+ 		}
+	}
+
+	/* This closes a way to execute a system call on the host.  If
+	 * you set a breakpoint on a system call instruction and singlestep
+	 * from it, the tracing thread used to PTRACE_SINGLESTEP the process
+	 * rather than PTRACE_SYSCALL it, allowing the system call to execute
+	 * on the host.  The tracing thread will check this flag and
+	 * PTRACE_SYSCALL if necessary.
+	 */
+	if(current->ptrace & PT_DTRACE)
+		current->thread.singlestep_syscall =
+			is_syscall(PT_REGS_IP(&current->thread.regs));
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	 * back */
+	if (!handled_sig && test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+	return handled_sig;
+}
+
+int do_signal(void)
+{
+	return kern_do_signal(&current->thread.regs);
+}
+
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+long sys_sigsuspend(int history0, int history1, old_sigset_t mask)
+{
+	mask &= _BLOCKABLE;
+	spin_lock_irq(&current->sighand->siglock);
+	current->saved_sigmask = current->blocked;
+	siginitset(&current->blocked, mask);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
+}
+
+long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss)
+{
+	return do_sigaltstack(uss, uoss, PT_REGS_SP(&current->thread.regs));
+}

arch/um/kernel/skas/Makefile

@@ -0,0 +1,15 @@
+#
+# Copyright (C) 2002 - 2004 Jeff Dike (jdike@addtoit.com)
+# Licensed under the GPL
+#
+
+obj-y := clone.o exec.o mem.o mmu.o process.o syscall.o tlb.o uaccess.o
+
+# clone.o is in the stub, so it can't be built with profiling
+# GCC hardened also auto-enables -fpic, but we need %ebx so it can't work ->
+# disable it
+
+CFLAGS_clone.o := $(CFLAGS_NO_HARDENING)
+UNPROFILE_OBJS := clone.o
+
+include arch/um/scripts/Makefile.rules

arch/um/kernel/skas/clone.c

@@ -0,0 +1,54 @@
+#include <sched.h>
+#include <signal.h>
+#include <sys/mman.h>
+#include <sys/time.h>
+#include <asm/unistd.h>
+#include <asm/page.h>
+#include "ptrace_user.h"
+#include "skas.h"
+#include "stub-data.h"
+#include "uml-config.h"
+#include "sysdep/stub.h"
+#include "kern_constants.h"
+
+/* This is in a separate file because it needs to be compiled with any
+ * extraneous gcc flags (-pg, -fprofile-arcs, -ftest-coverage) disabled
+ *
+ * Use UM_KERN_PAGE_SIZE instead of PAGE_SIZE because that calls getpagesize
+ * on some systems.
+ */
+
+void __attribute__ ((__section__ (".__syscall_stub")))
+stub_clone_handler(void)
+{
+	struct stub_data *data = (struct stub_data *) UML_CONFIG_STUB_DATA;
+	long err;
+
+	err = stub_syscall2(__NR_clone, CLONE_PARENT | CLONE_FILES | SIGCHLD,
+			    UML_CONFIG_STUB_DATA + UM_KERN_PAGE_SIZE / 2 -
+			    sizeof(void *));
+	if(err != 0)
+		goto out;
+
+	err = stub_syscall4(__NR_ptrace, PTRACE_TRACEME, 0, 0, 0);
+	if(err)
+		goto out;
+
+	err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL, 
+			    (long) &data->timer, 0);
+	if(err)
+		goto out;
+
+	remap_stack(data->fd, data->offset);
+	goto done;
+
+ out:
+	/* save current result. 
+	 * Parent: pid; 
+	 * child: retcode of mmap already saved and it jumps around this 
+	 * assignment
+	 */
+	data->err = err;
+ done:
+	trap_myself();
+}

arch/um/kernel/skas/exec.c

@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "asm/current.h"
+#include "asm/page.h"
+#include "asm/signal.h"
+#include "asm/ptrace.h"
+#include "asm/uaccess.h"
+#include "asm/mmu_context.h"
+#include "tlb.h"
+#include "skas.h"
+#include "um_mmu.h"
+#include "os.h"
+
+void flush_thread_skas(void)
+{
+	force_flush_all();
+	switch_mm_skas(&current->mm->context.skas.id);
+}
+
+void start_thread_skas(struct pt_regs *regs, unsigned long eip,
+		       unsigned long esp)
+{
+	set_fs(USER_DS);
+	PT_REGS_IP(regs) = eip;
+	PT_REGS_SP(regs) = esp;
+}

arch/um/kernel/skas/mem.c

@@ -0,0 +1,22 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/mm.h"
+#include "asm/pgtable.h"
+#include "mem_user.h"
+#include "skas.h"
+
+unsigned long set_task_sizes_skas(unsigned long *task_size_out)
+{
+	/* Round up to the nearest 4M */
+	unsigned long host_task_size = ROUND_4M((unsigned long)
+						&host_task_size);
+
+	if (!skas_needs_stub)
+		*task_size_out = host_task_size;
+	else *task_size_out = CONFIG_STUB_START & PGDIR_MASK;
+
+	return host_task_size;
+}

arch/um/kernel/skas/mmu.c

@@ -0,0 +1,158 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/list.h"
+#include "linux/spinlock.h"
+#include "linux/slab.h"
+#include "linux/errno.h"
+#include "linux/mm.h"
+#include "asm/current.h"
+#include "asm/segment.h"
+#include "asm/mmu.h"
+#include "asm/pgalloc.h"
+#include "asm/pgtable.h"
+#include "asm/ldt.h"
+#include "os.h"
+#include "skas.h"
+
+extern int __syscall_stub_start;
+
+static int init_stub_pte(struct mm_struct *mm, unsigned long proc,
+			 unsigned long kernel)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+
+	pgd = pgd_offset(mm, proc);
+	pud = pud_alloc(mm, pgd, proc);
+	if (!pud)
+		goto out;
+
+	pmd = pmd_alloc(mm, pud, proc);
+	if (!pmd)
+		goto out_pmd;
+
+	pte = pte_alloc_map(mm, pmd, proc);
+	if (!pte)
+		goto out_pte;
+
+	/* There's an interaction between the skas0 stub pages, stack
+	 * randomization, and the BUG at the end of exit_mmap.  exit_mmap
+         * checks that the number of page tables freed is the same as had
+         * been allocated.  If the stack is on the last page table page,
+	 * then the stack pte page will be freed, and if not, it won't.  To
+	 * avoid having to know where the stack is, or if the process mapped
+	 * something at the top of its address space for some other reason,
+	 * we set TASK_SIZE to end at the start of the last page table.
+	 * This keeps exit_mmap off the last page, but introduces a leak
+	 * of that page.  So, we hang onto it here and free it in
+	 * destroy_context_skas.
+	 */
+
+        mm->context.skas.last_page_table = pmd_page_vaddr(*pmd);
+#ifdef CONFIG_3_LEVEL_PGTABLES
+        mm->context.skas.last_pmd = (unsigned long) __va(pud_val(*pud));
+#endif
+
+	*pte = mk_pte(virt_to_page(kernel), __pgprot(_PAGE_PRESENT));
+	*pte = pte_mkread(*pte);
+	return(0);
+
+ out_pmd:
+	pud_free(pud);
+ out_pte:
+	pmd_free(pmd);
+ out:
+	return(-ENOMEM);
+}
+
+int init_new_context_skas(struct task_struct *task, struct mm_struct *mm)
+{
+ 	struct mmu_context_skas *from_mm = NULL;
+	struct mmu_context_skas *to_mm = &mm->context.skas;
+	unsigned long stack = 0;
+	int ret = -ENOMEM;
+
+	if(skas_needs_stub){
+		stack = get_zeroed_page(GFP_KERNEL);
+		if(stack == 0)
+			goto out;
+
+		/* This zeros the entry that pgd_alloc didn't, needed since
+		 * we are about to reinitialize it, and want mm.nr_ptes to
+		 * be accurate.
+		 */
+		mm->pgd[USER_PTRS_PER_PGD] = __pgd(0);
+
+		ret = init_stub_pte(mm, CONFIG_STUB_CODE,
+				    (unsigned long) &__syscall_stub_start);
+		if(ret)
+			goto out_free;
+
+		ret = init_stub_pte(mm, CONFIG_STUB_DATA, stack);
+		if(ret)
+			goto out_free;
+
+		mm->nr_ptes--;
+	}
+
+	to_mm->id.stack = stack;
+	if(current->mm != NULL && current->mm != &init_mm)
+		from_mm = &current->mm->context.skas;
+
+	if(proc_mm){
+		ret = new_mm(stack);
+		if(ret < 0){
+			printk("init_new_context_skas - new_mm failed, "
+			       "errno = %d\n", ret);
+			goto out_free;
+		}
+		to_mm->id.u.mm_fd = ret;
+	}
+	else {
+		if(from_mm)
+			to_mm->id.u.pid = copy_context_skas0(stack,
+							     from_mm->id.u.pid);
+		else to_mm->id.u.pid = start_userspace(stack);
+	}
+
+	ret = init_new_ldt(to_mm, from_mm);
+	if(ret < 0){
+		printk("init_new_context_skas - init_ldt"
+		       " failed, errno = %d\n", ret);
+		goto out_free;
+	}
+
+	return 0;
+
+ out_free:
+	if(to_mm->id.stack != 0)
+		free_page(to_mm->id.stack);
+ out:
+	return ret;
+}
+
+void destroy_context_skas(struct mm_struct *mm)
+{
+	struct mmu_context_skas *mmu = &mm->context.skas;
+
+	if(proc_mm)
+		os_close_file(mmu->id.u.mm_fd);
+	else
+		os_kill_ptraced_process(mmu->id.u.pid, 1);
+
+	if(!proc_mm || !ptrace_faultinfo){
+		free_page(mmu->id.stack);
+		pte_lock_deinit(virt_to_page(mmu->last_page_table));
+		pte_free_kernel((pte_t *) mmu->last_page_table);
+		dec_zone_page_state(virt_to_page(mmu->last_page_table), NR_PAGETABLE);
+#ifdef CONFIG_3_LEVEL_PGTABLES
+		pmd_free((pmd_t *) mmu->last_pmd);
+#endif
+	}
+}

arch/um/kernel/skas/process.c

@@ -0,0 +1,217 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/slab.h"
+#include "linux/ptrace.h"
+#include "linux/proc_fs.h"
+#include "linux/file.h"
+#include "linux/errno.h"
+#include "linux/init.h"
+#include "asm/uaccess.h"
+#include "asm/atomic.h"
+#include "kern_util.h"
+#include "skas.h"
+#include "os.h"
+#include "user_util.h"
+#include "tlb.h"
+#include "kern.h"
+#include "mode.h"
+#include "registers.h"
+
+void switch_to_skas(void *prev, void *next)
+{
+	struct task_struct *from, *to;
+
+	from = prev;
+	to = next;
+
+	/* XXX need to check runqueues[cpu].idle */
+	if(current->pid == 0)
+		switch_timers(0);
+
+	switch_threads(&from->thread.mode.skas.switch_buf,
+		       &to->thread.mode.skas.switch_buf);
+
+	arch_switch_to_skas(current->thread.prev_sched, current);
+
+	if(current->pid == 0)
+		switch_timers(1);
+}
+
+extern void schedule_tail(struct task_struct *prev);
+
+/* This is called magically, by its address being stuffed in a jmp_buf
+ * and being longjmp-d to.
+ */
+void new_thread_handler(void)
+{
+	int (*fn)(void *), n;
+	void *arg;
+
+	if(current->thread.prev_sched != NULL)
+		schedule_tail(current->thread.prev_sched);
+	current->thread.prev_sched = NULL;
+
+	fn = current->thread.request.u.thread.proc;
+	arg = current->thread.request.u.thread.arg;
+
+	/* The return value is 1 if the kernel thread execs a process,
+	 * 0 if it just exits
+	 */
+	n = run_kernel_thread(fn, arg, &current->thread.exec_buf);
+	if(n == 1){
+		/* Handle any immediate reschedules or signals */
+		interrupt_end();
+		userspace(&current->thread.regs.regs);
+	}
+	else do_exit(0);
+}
+
+void release_thread_skas(struct task_struct *task)
+{
+}
+
+/* Called magically, see new_thread_handler above */
+void fork_handler(void)
+{
+	force_flush_all();
+	if(current->thread.prev_sched == NULL)
+		panic("blech");
+
+	schedule_tail(current->thread.prev_sched);
+
+	/* XXX: if interrupt_end() calls schedule, this call to
+	 * arch_switch_to_skas isn't needed. We could want to apply this to
+	 * improve performance. -bb */
+	arch_switch_to_skas(current->thread.prev_sched, current);
+
+	current->thread.prev_sched = NULL;
+
+/* Handle any immediate reschedules or signals */
+	interrupt_end();
+
+	userspace(&current->thread.regs.regs);
+}
+
+int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
+		     unsigned long stack_top, struct task_struct * p,
+		     struct pt_regs *regs)
+{
+	void (*handler)(void);
+
+	if(current->thread.forking){
+	  	memcpy(&p->thread.regs.regs.skas, &regs->regs.skas,
+		       sizeof(p->thread.regs.regs.skas));
+		REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0);
+		if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp;
+
+		handler = fork_handler;
+
+		arch_copy_thread(&current->thread.arch, &p->thread.arch);
+	}
+	else {
+		init_thread_registers(&p->thread.regs.regs);
+		p->thread.request.u.thread = current->thread.request.u.thread;
+		handler = new_thread_handler;
+	}
+
+	new_thread(task_stack_page(p), &p->thread.mode.skas.switch_buf,
+		   handler);
+	return(0);
+}
+
+int new_mm(unsigned long stack)
+{
+	int fd;
+
+	fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0);
+	if(fd < 0)
+		return(fd);
+
+	if(skas_needs_stub)
+		map_stub_pages(fd, CONFIG_STUB_CODE, CONFIG_STUB_DATA, stack);
+
+	return(fd);
+}
+
+void init_idle_skas(void)
+{
+	cpu_tasks[current_thread->cpu].pid = os_getpid();
+	default_idle();
+}
+
+extern void start_kernel(void);
+
+static int start_kernel_proc(void *unused)
+{
+	int pid;
+
+	block_signals();
+	pid = os_getpid();
+
+	cpu_tasks[0].pid = pid;
+	cpu_tasks[0].task = current;
+#ifdef CONFIG_SMP
+	cpu_online_map = cpumask_of_cpu(0);
+#endif
+	start_kernel();
+	return(0);
+}
+
+extern int userspace_pid[];
+
+int start_uml_skas(void)
+{
+	if(proc_mm)
+		userspace_pid[0] = start_userspace(0);
+
+	init_new_thread_signals();
+
+	init_task.thread.request.u.thread.proc = start_kernel_proc;
+	init_task.thread.request.u.thread.arg = NULL;
+	return(start_idle_thread(task_stack_page(&init_task),
+				 &init_task.thread.mode.skas.switch_buf));
+}
+
+int external_pid_skas(struct task_struct *task)
+{
+#warning Need to look up userspace_pid by cpu
+	return(userspace_pid[0]);
+}
+
+int thread_pid_skas(struct task_struct *task)
+{
+#warning Need to look up userspace_pid by cpu
+	return(userspace_pid[0]);
+}
+
+void kill_off_processes_skas(void)
+{
+	if(proc_mm)
+#warning need to loop over userspace_pids in kill_off_processes_skas
+		os_kill_ptraced_process(userspace_pid[0], 1);
+	else {
+		struct task_struct *p;
+		int pid, me;
+
+		me = os_getpid();
+		for_each_process(p){
+			if(p->mm == NULL)
+				continue;
+
+			pid = p->mm->context.skas.id.u.pid;
+			os_kill_ptraced_process(pid, 1);
+		}
+	}
+}
+
+unsigned long current_stub_stack(void)
+{
+	if(current->mm == NULL)
+		return(0);
+
+	return(current->mm->context.skas.id.stack);
+}

arch/um/kernel/skas/syscall.c

@@ -0,0 +1,43 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sys.h"
+#include "linux/ptrace.h"
+#include "asm/errno.h"
+#include "asm/unistd.h"
+#include "asm/ptrace.h"
+#include "asm/current.h"
+#include "sysdep/syscalls.h"
+#include "kern_util.h"
+#include "syscall.h"
+
+void handle_syscall(union uml_pt_regs *r)
+{
+	struct pt_regs *regs = container_of(r, struct pt_regs, regs);
+	long result;
+	int syscall;
+
+	syscall_trace(r, 0);
+
+	current->thread.nsyscalls++;
+	nsyscalls++;
+
+	/* This should go in the declaration of syscall, but when I do that,
+	 * strace -f -c bash -c 'ls ; ls' breaks, sometimes not tracing
+	 * children at all, sometimes hanging when bash doesn't see the first
+	 * ls exit.
+	 * The assembly looks functionally the same to me.  This is
+	 *     gcc version 4.0.1 20050727 (Red Hat 4.0.1-5)
+	 * in case it's a compiler bug.
+	 */
+	syscall = UPT_SYSCALL_NR(r);
+	if((syscall >= NR_syscalls) || (syscall < 0))
+		result = -ENOSYS;
+	else result = EXECUTE_SYSCALL(syscall, regs);
+
+	REGS_SET_SYSCALL_RETURN(r->skas.regs, result);
+
+	syscall_trace(r, 1);
+}

arch/um/kernel/skas/tlb.c

@@ -0,0 +1,97 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright 2003 PathScale, Inc.
+ * Licensed under the GPL
+ */
+
+#include "linux/stddef.h"
+#include "linux/sched.h"
+#include "linux/mm.h"
+#include "asm/page.h"
+#include "asm/pgtable.h"
+#include "asm/mmu.h"
+#include "user_util.h"
+#include "mem_user.h"
+#include "mem.h"
+#include "skas.h"
+#include "os.h"
+#include "tlb.h"
+
+static int do_ops(union mm_context *mmu, struct host_vm_op *ops, int last,
+		  int finished, void **flush)
+{
+	struct host_vm_op *op;
+        int i, ret = 0;
+
+        for(i = 0; i <= last && !ret; i++){
+		op = &ops[i];
+		switch(op->type){
+		case MMAP:
+			ret = map(&mmu->skas.id, op->u.mmap.addr,
+				  op->u.mmap.len, op->u.mmap.r, op->u.mmap.w,
+				  op->u.mmap.x, op->u.mmap.fd,
+				  op->u.mmap.offset, finished, flush);
+			break;
+		case MUNMAP:
+			ret = unmap(&mmu->skas.id,
+				    (void *) op->u.munmap.addr,
+				    op->u.munmap.len, finished, flush);
+			break;
+		case MPROTECT:
+			ret = protect(&mmu->skas.id, op->u.mprotect.addr,
+				      op->u.mprotect.len, op->u.mprotect.r,
+				      op->u.mprotect.w, op->u.mprotect.x,
+				      finished, flush);
+			break;
+		default:
+			printk("Unknown op type %d in do_ops\n", op->type);
+			break;
+		}
+	}
+
+	return ret;
+}
+
+extern int proc_mm;
+
+static void fix_range(struct mm_struct *mm, unsigned long start_addr,
+		      unsigned long end_addr, int force)
+{
+        if(!proc_mm && (end_addr > CONFIG_STUB_START))
+                end_addr = CONFIG_STUB_START;
+
+        fix_range_common(mm, start_addr, end_addr, force, do_ops);
+}
+
+void __flush_tlb_one_skas(unsigned long addr)
+{
+        flush_tlb_kernel_range_common(addr, addr + PAGE_SIZE);
+}
+
+void flush_tlb_range_skas(struct vm_area_struct *vma, unsigned long start, 
+		     unsigned long end)
+{
+        if(vma->vm_mm == NULL)
+                flush_tlb_kernel_range_common(start, end);
+        else fix_range(vma->vm_mm, start, end, 0);
+}
+
+void flush_tlb_mm_skas(struct mm_struct *mm)
+{
+	unsigned long end;
+
+	/* Don't bother flushing if this address space is about to be
+         * destroyed.
+         */
+        if(atomic_read(&mm->mm_users) == 0)
+                return;
+
+	end = proc_mm ? task_size : CONFIG_STUB_START;
+        fix_range(mm, 0, end, 0);
+}
+
+void force_flush_all_skas(void)
+{
+	unsigned long end = proc_mm ? task_size : CONFIG_STUB_START;
+        fix_range(current->mm, 0, end, 1);
+}

arch/um/kernel/skas/uaccess.c

@@ -0,0 +1,265 @@
+/*
+ * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/compiler.h"
+#include "linux/stddef.h"
+#include "linux/kernel.h"
+#include "linux/string.h"
+#include "linux/fs.h"
+#include "linux/hardirq.h"
+#include "linux/highmem.h"
+#include "asm/page.h"
+#include "asm/pgtable.h"
+#include "asm/uaccess.h"
+#include "kern_util.h"
+#include "os.h"
+
+extern void *um_virt_to_phys(struct task_struct *task, unsigned long addr,
+			     pte_t *pte_out);
+
+static unsigned long maybe_map(unsigned long virt, int is_write)
+{
+	pte_t pte;
+	int err;
+
+	void *phys = um_virt_to_phys(current, virt, &pte);
+	int dummy_code;
+
+	if(IS_ERR(phys) || (is_write && !pte_write(pte))){
+		err = handle_page_fault(virt, 0, is_write, 1, &dummy_code);
+		if(err)
+			return(-1UL);
+		phys = um_virt_to_phys(current, virt, NULL);
+	}
+        if(IS_ERR(phys))
+                phys = (void *) -1;
+
+	return((unsigned long) phys);
+}
+
+static int do_op_one_page(unsigned long addr, int len, int is_write,
+		 int (*op)(unsigned long addr, int len, void *arg), void *arg)
+{
+	struct page *page;
+	int n;
+
+	addr = maybe_map(addr, is_write);
+	if(addr == -1UL)
+		return(-1);
+
+	page = phys_to_page(addr);
+	addr = (unsigned long) kmap_atomic(page, KM_UML_USERCOPY) + (addr & ~PAGE_MASK);
+
+	n = (*op)(addr, len, arg);
+
+	kunmap_atomic(page, KM_UML_USERCOPY);
+
+	return(n);
+}
+
+static void do_buffer_op(void *jmpbuf, void *arg_ptr)
+{
+	va_list args;
+	unsigned long addr;
+	int len, is_write, size, remain, n;
+	int (*op)(unsigned long, int, void *);
+	void *arg;
+	int *res;
+
+	va_copy(args, *(va_list *)arg_ptr);
+	addr = va_arg(args, unsigned long);
+	len = va_arg(args, int);
+	is_write = va_arg(args, int);
+	op = va_arg(args, void *);
+	arg = va_arg(args, void *);
+	res = va_arg(args, int *);
+	va_end(args);
+	size = min(PAGE_ALIGN(addr) - addr, (unsigned long) len);
+	remain = len;
+
+	current->thread.fault_catcher = jmpbuf;
+	n = do_op_one_page(addr, size, is_write, op, arg);
+	if(n != 0){
+		*res = (n < 0 ? remain : 0);
+		goto out;
+	}
+
+	addr += size;
+	remain -= size;
+	if(remain == 0){
+		*res = 0;
+		goto out;
+	}
+
+	while(addr < ((addr + remain) & PAGE_MASK)){
+		n = do_op_one_page(addr, PAGE_SIZE, is_write, op, arg);
+		if(n != 0){
+			*res = (n < 0 ? remain : 0);
+			goto out;
+		}
+
+		addr += PAGE_SIZE;
+		remain -= PAGE_SIZE;
+	}
+	if(remain == 0){
+		*res = 0;
+		goto out;
+	}
+
+	n = do_op_one_page(addr, remain, is_write, op, arg);
+	if(n != 0)
+		*res = (n < 0 ? remain : 0);
+	else *res = 0;
+ out:
+	current->thread.fault_catcher = NULL;
+}
+
+static int buffer_op(unsigned long addr, int len, int is_write,
+		     int (*op)(unsigned long addr, int len, void *arg),
+		     void *arg)
+{
+	int faulted, res;
+
+	faulted = setjmp_wrapper(do_buffer_op, addr, len, is_write, op, arg,
+				 &res);
+	if(!faulted)
+		return(res);
+
+	return(addr + len - (unsigned long) current->thread.fault_addr);
+}
+
+static int copy_chunk_from_user(unsigned long from, int len, void *arg)
+{
+	unsigned long *to_ptr = arg, to = *to_ptr;
+
+	memcpy((void *) to, (void *) from, len);
+	*to_ptr += len;
+	return(0);
+}
+
+int copy_from_user_skas(void *to, const void __user *from, int n)
+{
+	if(segment_eq(get_fs(), KERNEL_DS)){
+		memcpy(to, (__force void*)from, n);
+		return(0);
+	}
+
+	return(access_ok(VERIFY_READ, from, n) ?
+	       buffer_op((unsigned long) from, n, 0, copy_chunk_from_user, &to):
+	       n);
+}
+
+static int copy_chunk_to_user(unsigned long to, int len, void *arg)
+{
+	unsigned long *from_ptr = arg, from = *from_ptr;
+
+	memcpy((void *) to, (void *) from, len);
+	*from_ptr += len;
+	return(0);
+}
+
+int copy_to_user_skas(void __user *to, const void *from, int n)
+{
+	if(segment_eq(get_fs(), KERNEL_DS)){
+		memcpy((__force void*)to, from, n);
+		return(0);
+	}
+
+	return(access_ok(VERIFY_WRITE, to, n) ?
+	       buffer_op((unsigned long) to, n, 1, copy_chunk_to_user, &from) :
+	       n);
+}
+
+static int strncpy_chunk_from_user(unsigned long from, int len, void *arg)
+{
+	char **to_ptr = arg, *to = *to_ptr;
+	int n;
+
+	strncpy(to, (void *) from, len);
+	n = strnlen(to, len);
+	*to_ptr += n;
+
+	if(n < len)
+	        return(1);
+	return(0);
+}
+
+int strncpy_from_user_skas(char *dst, const char __user *src, int count)
+{
+	int n;
+	char *ptr = dst;
+
+	if(segment_eq(get_fs(), KERNEL_DS)){
+		strncpy(dst, (__force void*)src, count);
+		return(strnlen(dst, count));
+	}
+
+	if(!access_ok(VERIFY_READ, src, 1))
+		return(-EFAULT);
+
+	n = buffer_op((unsigned long) src, count, 0, strncpy_chunk_from_user,
+		      &ptr);
+	if(n != 0)
+		return(-EFAULT);
+	return(strnlen(dst, count));
+}
+
+static int clear_chunk(unsigned long addr, int len, void *unused)
+{
+	memset((void *) addr, 0, len);
+	return(0);
+}
+
+int __clear_user_skas(void __user *mem, int len)
+{
+	return(buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL));
+}
+
+int clear_user_skas(void __user *mem, int len)
+{
+	if(segment_eq(get_fs(), KERNEL_DS)){
+		memset((__force void*)mem, 0, len);
+		return(0);
+	}
+
+	return(access_ok(VERIFY_WRITE, mem, len) ?
+	       buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL) : len);
+}
+
+static int strnlen_chunk(unsigned long str, int len, void *arg)
+{
+	int *len_ptr = arg, n;
+
+	n = strnlen((void *) str, len);
+	*len_ptr += n;
+
+	if(n < len)
+		return(1);
+	return(0);
+}
+
+int strnlen_user_skas(const void __user *str, int len)
+{
+	int count = 0, n;
+
+	if(segment_eq(get_fs(), KERNEL_DS))
+		return(strnlen((__force char*)str, len) + 1);
+
+	n = buffer_op((unsigned long) str, len, 0, strnlen_chunk, &count);
+	if(n == 0)
+		return(count + 1);
+	return(-EFAULT);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/smp.c

@@ -0,0 +1,267 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/percpu.h"
+#include "asm/pgalloc.h"
+#include "asm/tlb.h"
+
+/* For some reason, mmu_gathers are referenced when CONFIG_SMP is off. */
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+
+#ifdef CONFIG_SMP
+
+#include "linux/sched.h"
+#include "linux/module.h"
+#include "linux/threads.h"
+#include "linux/interrupt.h"
+#include "linux/err.h"
+#include "linux/hardirq.h"
+#include "asm/smp.h"
+#include "asm/processor.h"
+#include "asm/spinlock.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "irq_user.h"
+#include "os.h"
+
+/* CPU online map, set by smp_boot_cpus */
+cpumask_t cpu_online_map = CPU_MASK_NONE;
+cpumask_t cpu_possible_map = CPU_MASK_NONE;
+
+EXPORT_SYMBOL(cpu_online_map);
+EXPORT_SYMBOL(cpu_possible_map);
+
+/* Per CPU bogomips and other parameters
+ * The only piece used here is the ipi pipe, which is set before SMP is
+ * started and never changed.
+ */
+struct cpuinfo_um cpu_data[NR_CPUS];
+
+/* A statistic, can be a little off */
+int num_reschedules_sent = 0;
+
+/* Not changed after boot */
+struct task_struct *idle_threads[NR_CPUS];
+
+void smp_send_reschedule(int cpu)
+{
+	os_write_file(cpu_data[cpu].ipi_pipe[1], "R", 1);
+	num_reschedules_sent++;
+}
+
+void smp_send_stop(void)
+{
+	int i;
+
+	printk(KERN_INFO "Stopping all CPUs...");
+	for(i = 0; i < num_online_cpus(); i++){
+		if(i == current_thread->cpu)
+			continue;
+		os_write_file(cpu_data[i].ipi_pipe[1], "S", 1);
+	}
+	printk("done\n");
+}
+
+static cpumask_t smp_commenced_mask = CPU_MASK_NONE;
+static cpumask_t cpu_callin_map = CPU_MASK_NONE;
+
+static int idle_proc(void *cpup)
+{
+	int cpu = (int) cpup, err;
+
+	err = os_pipe(cpu_data[cpu].ipi_pipe, 1, 1);
+	if(err < 0)
+		panic("CPU#%d failed to create IPI pipe, err = %d", cpu, -err);
+
+	os_set_fd_async(cpu_data[cpu].ipi_pipe[0],
+		     current->thread.mode.tt.extern_pid);
+
+	wmb();
+	if (cpu_test_and_set(cpu, cpu_callin_map)) {
+		printk("huh, CPU#%d already present??\n", cpu);
+		BUG();
+	}
+
+	while (!cpu_isset(cpu, smp_commenced_mask))
+		cpu_relax();
+
+	cpu_set(cpu, cpu_online_map);
+	default_idle();
+	return(0);
+}
+
+static struct task_struct *idle_thread(int cpu)
+{
+	struct task_struct *new_task;
+	unsigned char c;
+
+        current->thread.request.u.thread.proc = idle_proc;
+        current->thread.request.u.thread.arg = (void *) cpu;
+	new_task = fork_idle(cpu);
+	if(IS_ERR(new_task))
+		panic("copy_process failed in idle_thread, error = %ld",
+		      PTR_ERR(new_task));
+
+	cpu_tasks[cpu] = ((struct cpu_task)
+		          { .pid = 	new_task->thread.mode.tt.extern_pid,
+			    .task = 	new_task } );
+	idle_threads[cpu] = new_task;
+	CHOOSE_MODE(os_write_file(new_task->thread.mode.tt.switch_pipe[1], &c,
+			  sizeof(c)),
+		    ({ panic("skas mode doesn't support SMP"); }));
+	return(new_task);
+}
+
+void smp_prepare_cpus(unsigned int maxcpus)
+{
+	struct task_struct *idle;
+	unsigned long waittime;
+	int err, cpu, me = smp_processor_id();
+	int i;
+
+	for (i = 0; i < ncpus; ++i)
+		cpu_set(i, cpu_possible_map);
+
+	cpu_clear(me, cpu_online_map);
+	cpu_set(me, cpu_online_map);
+	cpu_set(me, cpu_callin_map);
+
+	err = os_pipe(cpu_data[me].ipi_pipe, 1, 1);
+	if(err < 0)
+		panic("CPU#0 failed to create IPI pipe, errno = %d", -err);
+
+	os_set_fd_async(cpu_data[me].ipi_pipe[0],
+		     current->thread.mode.tt.extern_pid);
+
+	for(cpu = 1; cpu < ncpus; cpu++){
+		printk("Booting processor %d...\n", cpu);
+
+		idle = idle_thread(cpu);
+
+		init_idle(idle, cpu);
+
+		waittime = 200000000;
+		while (waittime-- && !cpu_isset(cpu, cpu_callin_map))
+			cpu_relax();
+
+		if (cpu_isset(cpu, cpu_callin_map))
+			printk("done\n");
+		else printk("failed\n");
+	}
+}
+
+void smp_prepare_boot_cpu(void)
+{
+	cpu_set(smp_processor_id(), cpu_online_map);
+}
+
+int __cpu_up(unsigned int cpu)
+{
+	cpu_set(cpu, smp_commenced_mask);
+	while (!cpu_isset(cpu, cpu_online_map))
+		mb();
+	return(0);
+}
+
+int setup_profiling_timer(unsigned int multiplier)
+{
+	printk(KERN_INFO "setup_profiling_timer\n");
+	return(0);
+}
+
+void smp_call_function_slave(int cpu);
+
+void IPI_handler(int cpu)
+{
+	unsigned char c;
+	int fd;
+
+	fd = cpu_data[cpu].ipi_pipe[0];
+	while (os_read_file(fd, &c, 1) == 1) {
+		switch (c) {
+		case 'C':
+			smp_call_function_slave(cpu);
+			break;
+
+		case 'R':
+			set_tsk_need_resched(current);
+			break;
+
+		case 'S':
+			printk("CPU#%d stopping\n", cpu);
+			while(1)
+				pause();
+			break;
+
+		default:
+			printk("CPU#%d received unknown IPI [%c]!\n", cpu, c);
+			break;
+		}
+	}
+}
+
+int hard_smp_processor_id(void)
+{
+	return(pid_to_processor_id(os_getpid()));
+}
+
+static DEFINE_SPINLOCK(call_lock);
+static atomic_t scf_started;
+static atomic_t scf_finished;
+static void (*func)(void *info);
+static void *info;
+
+void smp_call_function_slave(int cpu)
+{
+	atomic_inc(&scf_started);
+	(*func)(info);
+	atomic_inc(&scf_finished);
+}
+
+int smp_call_function(void (*_func)(void *info), void *_info, int nonatomic,
+		      int wait)
+{
+	int cpus = num_online_cpus() - 1;
+	int i;
+
+	if (!cpus)
+		return 0;
+
+	/* Can deadlock when called with interrupts disabled */
+	WARN_ON(irqs_disabled());
+
+	spin_lock_bh(&call_lock);
+	atomic_set(&scf_started, 0);
+	atomic_set(&scf_finished, 0);
+	func = _func;
+	info = _info;
+
+	for_each_online_cpu(i)
+		os_write_file(cpu_data[i].ipi_pipe[1], "C", 1);
+
+	while (atomic_read(&scf_started) != cpus)
+		barrier();
+
+	if (wait)
+		while (atomic_read(&scf_finished) != cpus)
+			barrier();
+
+	spin_unlock_bh(&call_lock);
+	return 0;
+}
+
+#endif
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/syscall.c

@@ -0,0 +1,163 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/file.h"
+#include "linux/smp_lock.h"
+#include "linux/mm.h"
+#include "linux/utsname.h"
+#include "linux/msg.h"
+#include "linux/shm.h"
+#include "linux/sys.h"
+#include "linux/syscalls.h"
+#include "linux/unistd.h"
+#include "linux/slab.h"
+#include "linux/utime.h"
+#include "asm/mman.h"
+#include "asm/uaccess.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "sysdep/syscalls.h"
+#include "mode_kern.h"
+#include "choose-mode.h"
+
+/*  Unlocked, I don't care if this is a bit off */
+int nsyscalls = 0;
+
+long sys_fork(void)
+{
+	long ret;
+
+	current->thread.forking = 1;
+	ret = do_fork(SIGCHLD, UPT_SP(&current->thread.regs.regs),
+		      &current->thread.regs, 0, NULL, NULL);
+	current->thread.forking = 0;
+	return(ret);
+}
+
+long sys_vfork(void)
+{
+	long ret;
+
+	current->thread.forking = 1;
+	ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD,
+		      UPT_SP(&current->thread.regs.regs),
+		      &current->thread.regs, 0, NULL, NULL);
+	current->thread.forking = 0;
+	return(ret);
+}
+
+/* common code for old and new mmaps */
+long sys_mmap2(unsigned long addr, unsigned long len,
+	       unsigned long prot, unsigned long flags,
+	       unsigned long fd, unsigned long pgoff)
+{
+	long error = -EBADF;
+	struct file * file = NULL;
+
+	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
+	if (!(flags & MAP_ANONYMOUS)) {
+		file = fget(fd);
+		if (!file)
+			goto out;
+	}
+
+	down_write(&current->mm->mmap_sem);
+	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
+	up_write(&current->mm->mmap_sem);
+
+	if (file)
+		fput(file);
+ out:
+	return error;
+}
+
+long old_mmap(unsigned long addr, unsigned long len,
+	      unsigned long prot, unsigned long flags,
+	      unsigned long fd, unsigned long offset)
+{
+	long err = -EINVAL;
+	if (offset & ~PAGE_MASK)
+		goto out;
+
+	err = sys_mmap2(addr, len, prot, flags, fd, offset >> PAGE_SHIFT);
+ out:
+	return err;
+}
+/*
+ * sys_pipe() is the normal C calling standard for creating
+ * a pipe. It's not the way unix traditionally does this, though.
+ */
+long sys_pipe(unsigned long __user * fildes)
+{
+        int fd[2];
+        long error;
+
+        error = do_pipe(fd);
+        if (!error) {
+		if (copy_to_user(fildes, fd, sizeof(fd)))
+                        error = -EFAULT;
+        }
+        return error;
+}
+
+
+long sys_uname(struct old_utsname __user * name)
+{
+	long err;
+	if (!name)
+		return -EFAULT;
+	down_read(&uts_sem);
+	err = copy_to_user(name, utsname(), sizeof (*name));
+	up_read(&uts_sem);
+	return err?-EFAULT:0;
+}
+
+long sys_olduname(struct oldold_utsname __user * name)
+{
+	long error;
+
+	if (!name)
+		return -EFAULT;
+	if (!access_ok(VERIFY_WRITE,name,sizeof(struct oldold_utsname)))
+		return -EFAULT;
+
+  	down_read(&uts_sem);
+
+	error = __copy_to_user(&name->sysname, &utsname()->sysname,
+			       __OLD_UTS_LEN);
+	error |= __put_user(0, name->sysname + __OLD_UTS_LEN);
+	error |= __copy_to_user(&name->nodename, &utsname()->nodename,
+				__OLD_UTS_LEN);
+	error |= __put_user(0, name->nodename + __OLD_UTS_LEN);
+	error |= __copy_to_user(&name->release, &utsname()->release,
+				__OLD_UTS_LEN);
+	error |= __put_user(0, name->release + __OLD_UTS_LEN);
+	error |= __copy_to_user(&name->version, &utsname()->version,
+				__OLD_UTS_LEN);
+	error |= __put_user(0, name->version + __OLD_UTS_LEN);
+	error |= __copy_to_user(&name->machine, &utsname()->machine,
+				__OLD_UTS_LEN);
+	error |= __put_user(0, name->machine + __OLD_UTS_LEN);
+
+	up_read(&uts_sem);
+
+	error = error ? -EFAULT : 0;
+
+	return error;
+}
+
+int kernel_execve(const char *filename, char *const argv[], char *const envp[])
+{
+	mm_segment_t fs;
+	int ret;
+
+	fs = get_fs();
+	set_fs(KERNEL_DS);
+	ret = um_execve(filename, argv, envp);
+	set_fs(fs);
+
+	return ret;
+}

arch/um/kernel/sysrq.c

@@ -0,0 +1,81 @@
+/* 
+ * Copyright (C) 2001 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/kernel.h"
+#include "linux/module.h"
+#include "linux/kallsyms.h"
+#include "asm/page.h"
+#include "asm/processor.h"
+#include "sysrq.h"
+#include "user_util.h"
+
+/* Catch non-i386 SUBARCH's. */
+#if !defined(CONFIG_UML_X86) || defined(CONFIG_64BIT)
+void show_trace(struct task_struct *task, unsigned long * stack)
+{
+        unsigned long addr;
+
+        if (!stack) {
+		stack = (unsigned long*) &stack;
+		WARN_ON(1);
+	}
+
+        printk("Call Trace: \n");
+        while (((long) stack & (THREAD_SIZE-1)) != 0) {
+                addr = *stack;
+		if (__kernel_text_address(addr)) {
+			printk("%08lx:  [<%08lx>]", (unsigned long) stack, addr);
+			print_symbol(" %s", addr);
+			printk("\n");
+                }
+                stack++;
+        }
+        printk("\n");
+}
+#endif
+
+/*
+ * stack dumps generator - this is used by arch-independent code.
+ * And this is identical to i386 currently.
+ */
+void dump_stack(void)
+{
+	unsigned long stack;
+
+	show_trace(current, &stack);
+}
+EXPORT_SYMBOL(dump_stack);
+
+/*Stolen from arch/i386/kernel/traps.c */
+static const int kstack_depth_to_print = 24;
+
+/* This recently started being used in arch-independent code too, as in
+ * kernel/sched.c.*/
+void show_stack(struct task_struct *task, unsigned long *esp)
+{
+	unsigned long *stack;
+	int i;
+
+	if (esp == NULL) {
+		if (task != current && task != NULL) {
+			esp = (unsigned long *) KSTK_ESP(task);
+		} else {
+			esp = (unsigned long *) &esp;
+		}
+	}
+
+	stack = esp;
+	for(i = 0; i < kstack_depth_to_print; i++) {
+		if (kstack_end(stack))
+			break;
+		if (i && ((i % 8) == 0))
+			printk("\n       ");
+		printk("%08lx ", *stack++);
+	}
+
+	printk("Call Trace: \n");
+	show_trace(task, esp);
+}

arch/um/kernel/time.c

@@ -0,0 +1,180 @@
+/*
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/module.h"
+#include "linux/unistd.h"
+#include "linux/stddef.h"
+#include "linux/spinlock.h"
+#include "linux/time.h"
+#include "linux/sched.h"
+#include "linux/interrupt.h"
+#include "linux/init.h"
+#include "linux/delay.h"
+#include "linux/hrtimer.h"
+#include "asm/irq.h"
+#include "asm/param.h"
+#include "asm/current.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "mode.h"
+#include "os.h"
+
+int hz(void)
+{
+	return(HZ);
+}
+
+/*
+ * Scheduler clock - returns current time in nanosec units.
+ */
+unsigned long long sched_clock(void)
+{
+	return (unsigned long long)jiffies_64 * (1000000000 / HZ);
+}
+
+static unsigned long long prev_nsecs[NR_CPUS];
+#ifdef CONFIG_UML_REAL_TIME_CLOCK
+static long long delta[NR_CPUS];		/* Deviation per interval */
+#endif
+
+void timer_irq(union uml_pt_regs *regs)
+{
+	unsigned long long ticks = 0;
+#ifdef CONFIG_UML_REAL_TIME_CLOCK
+	int c = cpu();
+	if(prev_nsecs[c]){
+		/* We've had 1 tick */
+		unsigned long long nsecs = os_nsecs();
+
+		delta[c] += nsecs - prev_nsecs[c];
+		prev_nsecs[c] = nsecs;
+
+		/* Protect against the host clock being set backwards */
+		if(delta[c] < 0)
+			delta[c] = 0;
+
+		ticks += (delta[c] * HZ) / BILLION;
+		delta[c] -= (ticks * BILLION) / HZ;
+	}
+	else prev_nsecs[c] = os_nsecs();
+#else
+	ticks = 1;
+#endif
+	while(ticks > 0){
+		do_IRQ(TIMER_IRQ, regs);
+		ticks--;
+	}
+}
+
+/* Protects local_offset */
+static DEFINE_SPINLOCK(timer_spinlock);
+static unsigned long long local_offset = 0;
+
+static inline unsigned long long get_time(void)
+{
+	unsigned long long nsecs;
+	unsigned long flags;
+
+	spin_lock_irqsave(&timer_spinlock, flags);
+	nsecs = os_nsecs();
+	nsecs += local_offset;
+	spin_unlock_irqrestore(&timer_spinlock, flags);
+
+	return nsecs;
+}
+
+irqreturn_t um_timer(int irq, void *dev)
+{
+	unsigned long long nsecs;
+	unsigned long flags;
+
+	write_seqlock_irqsave(&xtime_lock, flags);
+
+	do_timer(1);
+
+	nsecs = get_time();
+	xtime.tv_sec = nsecs / NSEC_PER_SEC;
+	xtime.tv_nsec = nsecs - xtime.tv_sec * NSEC_PER_SEC;
+
+	write_sequnlock_irqrestore(&xtime_lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+static void register_timer(void)
+{
+	int err;
+
+	err = request_irq(TIMER_IRQ, um_timer, IRQF_DISABLED, "timer", NULL);
+	if(err != 0)
+		printk(KERN_ERR "register_timer : request_irq failed - "
+		       "errno = %d\n", -err);
+
+	err = set_interval(1);
+	if(err != 0)
+		printk(KERN_ERR "register_timer : set_interval failed - "
+		       "errno = %d\n", -err);
+}
+
+extern void (*late_time_init)(void);
+
+void time_init(void)
+{
+	long long nsecs;
+
+	nsecs = os_nsecs();
+	set_normalized_timespec(&wall_to_monotonic, -nsecs / BILLION,
+				-nsecs % BILLION);
+	late_time_init = register_timer;
+}
+
+void do_gettimeofday(struct timeval *tv)
+{
+	unsigned long long nsecs = get_time();
+
+	tv->tv_sec = nsecs / NSEC_PER_SEC;
+	/* Careful about calculations here - this was originally done as
+	 * (nsecs - tv->tv_sec * NSEC_PER_SEC) / NSEC_PER_USEC
+	 * which gave bogus (> 1000000) values.  Dunno why, suspect gcc
+	 * (4.0.0) miscompiled it, or there's a subtle 64/32-bit conversion
+	 * problem that I missed.
+	 */
+	nsecs -= tv->tv_sec * NSEC_PER_SEC;
+	tv->tv_usec = (unsigned long) nsecs / NSEC_PER_USEC;
+}
+
+static inline void set_time(unsigned long long nsecs)
+{
+	unsigned long long now;
+	unsigned long flags;
+
+	spin_lock_irqsave(&timer_spinlock, flags);
+	now = os_nsecs();
+	local_offset = nsecs - now;
+	spin_unlock_irqrestore(&timer_spinlock, flags);
+
+	clock_was_set();
+}
+
+int do_settimeofday(struct timespec *tv)
+{
+	set_time((unsigned long long) tv->tv_sec * NSEC_PER_SEC + tv->tv_nsec);
+
+	return 0;
+}
+
+void timer_handler(int sig, union uml_pt_regs *regs)
+{
+	local_irq_disable();
+	irq_enter();
+	update_process_times(CHOOSE_MODE(
+	                     (UPT_SC(regs) && user_context(UPT_SP(regs))),
+			     (regs)->skas.is_user));
+	irq_exit();
+	local_irq_enable();
+	if(current_thread->cpu == 0)
+		timer_irq(regs);
+}

arch/um/kernel/tlb.c

@@ -0,0 +1,390 @@
+/*
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/mm.h"
+#include "asm/page.h"
+#include "asm/pgalloc.h"
+#include "asm/tlbflush.h"
+#include "choose-mode.h"
+#include "mode_kern.h"
+#include "user_util.h"
+#include "tlb.h"
+#include "mem.h"
+#include "mem_user.h"
+#include "os.h"
+
+static int add_mmap(unsigned long virt, unsigned long phys, unsigned long len,
+		    int r, int w, int x, struct host_vm_op *ops, int *index,
+		    int last_filled, union mm_context *mmu, void **flush,
+		    int (*do_ops)(union mm_context *, struct host_vm_op *,
+				  int, int, void **))
+{
+	__u64 offset;
+	struct host_vm_op *last;
+	int fd, ret = 0;
+
+	fd = phys_mapping(phys, &offset);
+	if(*index != -1){
+		last = &ops[*index];
+		if((last->type == MMAP) &&
+		   (last->u.mmap.addr + last->u.mmap.len == virt) &&
+		   (last->u.mmap.r == r) && (last->u.mmap.w == w) &&
+		   (last->u.mmap.x == x) && (last->u.mmap.fd == fd) &&
+		   (last->u.mmap.offset + last->u.mmap.len == offset)){
+			last->u.mmap.len += len;
+			return 0;
+		}
+	}
+
+	if(*index == last_filled){
+		ret = (*do_ops)(mmu, ops, last_filled, 0, flush);
+		*index = -1;
+	}
+
+	ops[++*index] = ((struct host_vm_op) { .type	= MMAP,
+			     			.u = { .mmap = {
+						       .addr	= virt,
+						       .len	= len,
+						       .r	= r,
+						       .w	= w,
+						       .x	= x,
+						       .fd	= fd,
+						       .offset	= offset }
+			   } });
+	return ret;
+}
+
+static int add_munmap(unsigned long addr, unsigned long len,
+		      struct host_vm_op *ops, int *index, int last_filled,
+		      union mm_context *mmu, void **flush,
+		      int (*do_ops)(union mm_context *, struct host_vm_op *,
+				    int, int, void **))
+{
+	struct host_vm_op *last;
+	int ret = 0;
+
+	if(*index != -1){
+		last = &ops[*index];
+		if((last->type == MUNMAP) &&
+		   (last->u.munmap.addr + last->u.mmap.len == addr)){
+			last->u.munmap.len += len;
+			return 0;
+		}
+	}
+
+	if(*index == last_filled){
+		ret = (*do_ops)(mmu, ops, last_filled, 0, flush);
+		*index = -1;
+	}
+
+	ops[++*index] = ((struct host_vm_op) { .type	= MUNMAP,
+			     		       .u = { .munmap = {
+						        .addr	= addr,
+							.len	= len } } });
+	return ret;
+}
+
+static int add_mprotect(unsigned long addr, unsigned long len, int r, int w,
+			int x, struct host_vm_op *ops, int *index,
+			int last_filled, union mm_context *mmu, void **flush,
+			int (*do_ops)(union mm_context *, struct host_vm_op *,
+				      int, int, void **))
+{
+	struct host_vm_op *last;
+	int ret = 0;
+
+	if(*index != -1){
+		last = &ops[*index];
+		if((last->type == MPROTECT) &&
+		   (last->u.mprotect.addr + last->u.mprotect.len == addr) &&
+		   (last->u.mprotect.r == r) && (last->u.mprotect.w == w) &&
+		   (last->u.mprotect.x == x)){
+			last->u.mprotect.len += len;
+			return 0;
+		}
+	}
+
+	if(*index == last_filled){
+		ret = (*do_ops)(mmu, ops, last_filled, 0, flush);
+		*index = -1;
+	}
+
+	ops[++*index] = ((struct host_vm_op) { .type	= MPROTECT,
+			     		       .u = { .mprotect = {
+						       .addr	= addr,
+						       .len	= len,
+						       .r	= r,
+						       .w	= w,
+						       .x	= x } } });
+	return ret;
+}
+
+#define ADD_ROUND(n, inc) (((n) + (inc)) & ~((inc) - 1))
+
+void fix_range_common(struct mm_struct *mm, unsigned long start_addr,
+		      unsigned long end_addr, int force,
+		      int (*do_ops)(union mm_context *, struct host_vm_op *,
+				    int, int, void **))
+{
+	pgd_t *npgd;
+	pud_t *npud;
+	pmd_t *npmd;
+	pte_t *npte;
+	union mm_context *mmu = &mm->context;
+	unsigned long addr, end;
+	int r, w, x;
+	struct host_vm_op ops[1];
+	void *flush = NULL;
+	int op_index = -1, last_op = ARRAY_SIZE(ops) - 1;
+	int ret = 0;
+
+	if(mm == NULL)
+		return;
+
+	ops[0].type = NONE;
+	for(addr = start_addr; addr < end_addr && !ret;){
+		npgd = pgd_offset(mm, addr);
+		if(!pgd_present(*npgd)){
+			end = ADD_ROUND(addr, PGDIR_SIZE);
+			if(end > end_addr)
+				end = end_addr;
+			if(force || pgd_newpage(*npgd)){
+				ret = add_munmap(addr, end - addr, ops,
+						 &op_index, last_op, mmu,
+						 &flush, do_ops);
+				pgd_mkuptodate(*npgd);
+			}
+			addr = end;
+			continue;
+		}
+
+		npud = pud_offset(npgd, addr);
+		if(!pud_present(*npud)){
+			end = ADD_ROUND(addr, PUD_SIZE);
+			if(end > end_addr)
+				end = end_addr;
+			if(force || pud_newpage(*npud)){
+				ret = add_munmap(addr, end - addr, ops,
+						 &op_index, last_op, mmu,
+						 &flush, do_ops);
+				pud_mkuptodate(*npud);
+			}
+			addr = end;
+			continue;
+		}
+
+		npmd = pmd_offset(npud, addr);
+		if(!pmd_present(*npmd)){
+			end = ADD_ROUND(addr, PMD_SIZE);
+			if(end > end_addr)
+				end = end_addr;
+			if(force || pmd_newpage(*npmd)){
+				ret = add_munmap(addr, end - addr, ops,
+						 &op_index, last_op, mmu,
+						 &flush, do_ops);
+				pmd_mkuptodate(*npmd);
+			}
+			addr = end;
+			continue;
+		}
+
+		npte = pte_offset_kernel(npmd, addr);
+		r = pte_read(*npte);
+		w = pte_write(*npte);
+		x = pte_exec(*npte);
+		if (!pte_young(*npte)) {
+			r = 0;
+			w = 0;
+		} else if (!pte_dirty(*npte)) {
+			w = 0;
+		}
+		if(force || pte_newpage(*npte)){
+			if(pte_present(*npte))
+				ret = add_mmap(addr,
+					       pte_val(*npte) & PAGE_MASK,
+					       PAGE_SIZE, r, w, x, ops,
+					       &op_index, last_op, mmu,
+					       &flush, do_ops);
+			else ret = add_munmap(addr, PAGE_SIZE, ops,
+					      &op_index, last_op, mmu,
+					      &flush, do_ops);
+		}
+		else if(pte_newprot(*npte))
+			ret = add_mprotect(addr, PAGE_SIZE, r, w, x, ops,
+					   &op_index, last_op, mmu,
+					   &flush, do_ops);
+
+		*npte = pte_mkuptodate(*npte);
+		addr += PAGE_SIZE;
+	}
+	if(!ret)
+		ret = (*do_ops)(mmu, ops, op_index, 1, &flush);
+
+/* This is not an else because ret is modified above */
+	if(ret) {
+		printk("fix_range_common: failed, killing current process\n");
+		force_sig(SIGKILL, current);
+	}
+}
+
+int flush_tlb_kernel_range_common(unsigned long start, unsigned long end)
+{
+	struct mm_struct *mm;
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long addr, last;
+	int updated = 0, err;
+
+	mm = &init_mm;
+	for(addr = start; addr < end;){
+		pgd = pgd_offset(mm, addr);
+		if(!pgd_present(*pgd)){
+			last = ADD_ROUND(addr, PGDIR_SIZE);
+			if(last > end)
+				last = end;
+			if(pgd_newpage(*pgd)){
+				updated = 1;
+				err = os_unmap_memory((void *) addr,
+						      last - addr);
+				if(err < 0)
+					panic("munmap failed, errno = %d\n",
+					      -err);
+			}
+			addr = last;
+			continue;
+		}
+
+		pud = pud_offset(pgd, addr);
+		if(!pud_present(*pud)){
+			last = ADD_ROUND(addr, PUD_SIZE);
+			if(last > end)
+				last = end;
+			if(pud_newpage(*pud)){
+				updated = 1;
+				err = os_unmap_memory((void *) addr,
+						      last - addr);
+				if(err < 0)
+					panic("munmap failed, errno = %d\n",
+					      -err);
+			}
+			addr = last;
+			continue;
+		}
+
+		pmd = pmd_offset(pud, addr);
+		if(!pmd_present(*pmd)){
+			last = ADD_ROUND(addr, PMD_SIZE);
+			if(last > end)
+				last = end;
+			if(pmd_newpage(*pmd)){
+				updated = 1;
+				err = os_unmap_memory((void *) addr,
+						      last - addr);
+				if(err < 0)
+					panic("munmap failed, errno = %d\n",
+					      -err);
+			}
+			addr = last;
+			continue;
+		}
+
+		pte = pte_offset_kernel(pmd, addr);
+		if(!pte_present(*pte) || pte_newpage(*pte)){
+			updated = 1;
+			err = os_unmap_memory((void *) addr,
+					      PAGE_SIZE);
+			if(err < 0)
+				panic("munmap failed, errno = %d\n",
+				      -err);
+			if(pte_present(*pte))
+				map_memory(addr,
+					   pte_val(*pte) & PAGE_MASK,
+					   PAGE_SIZE, 1, 1, 1);
+		}
+		else if(pte_newprot(*pte)){
+			updated = 1;
+			os_protect_memory((void *) addr, PAGE_SIZE, 1, 1, 1);
+		}
+		addr += PAGE_SIZE;
+	}
+	return(updated);
+}
+
+pgd_t *pgd_offset_proc(struct mm_struct *mm, unsigned long address)
+{
+	return(pgd_offset(mm, address));
+}
+
+pud_t *pud_offset_proc(pgd_t *pgd, unsigned long address)
+{
+	return(pud_offset(pgd, address));
+}
+
+pmd_t *pmd_offset_proc(pud_t *pud, unsigned long address)
+{
+	return(pmd_offset(pud, address));
+}
+
+pte_t *pte_offset_proc(pmd_t *pmd, unsigned long address)
+{
+	return(pte_offset_kernel(pmd, address));
+}
+
+pte_t *addr_pte(struct task_struct *task, unsigned long addr)
+{
+	pgd_t *pgd = pgd_offset(task->mm, addr);
+	pud_t *pud = pud_offset(pgd, addr);
+	pmd_t *pmd = pmd_offset(pud, addr);
+
+	return(pte_offset_map(pmd, addr));
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long address)
+{
+	address &= PAGE_MASK;
+	flush_tlb_range(vma, address, address + PAGE_SIZE);
+}
+
+void flush_tlb_all(void)
+{
+	flush_tlb_mm(current->mm);
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt,
+			 flush_tlb_kernel_range_common, start, end);
+}
+
+void flush_tlb_kernel_vm(void)
+{
+	CHOOSE_MODE(flush_tlb_kernel_vm_tt(),
+		    flush_tlb_kernel_range_common(start_vm, end_vm));
+}
+
+void __flush_tlb_one(unsigned long addr)
+{
+	CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr);
+}
+
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+		     unsigned long end)
+{
+	CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start,
+			 end);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm);
+}
+
+void force_flush_all(void)
+{
+	CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas());
+}
+

arch/um/kernel/trap.c

@@ -0,0 +1,258 @@
+/*
+ * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "asm/errno.h"
+#include "linux/sched.h"
+#include "linux/mm.h"
+#include "linux/spinlock.h"
+#include "linux/init.h"
+#include "linux/ptrace.h"
+#include "asm/semaphore.h"
+#include "asm/pgtable.h"
+#include "asm/pgalloc.h"
+#include "asm/tlbflush.h"
+#include "asm/a.out.h"
+#include "asm/current.h"
+#include "asm/irq.h"
+#include "sysdep/sigcontext.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "chan_kern.h"
+#include "mconsole_kern.h"
+#include "mem.h"
+#include "mem_kern.h"
+#include "sysdep/sigcontext.h"
+#include "sysdep/ptrace.h"
+#include "os.h"
+#ifdef CONFIG_MODE_SKAS
+#include "skas.h"
+#endif
+#include "os.h"
+
+/* Note this is constrained to return 0, -EFAULT, -EACCESS, -ENOMEM by segv(). */
+int handle_page_fault(unsigned long address, unsigned long ip,
+		      int is_write, int is_user, int *code_out)
+{
+	struct mm_struct *mm = current->mm;
+	struct vm_area_struct *vma;
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	int err = -EFAULT;
+
+	*code_out = SEGV_MAPERR;
+
+	/* If the fault was during atomic operation, don't take the fault, just
+	 * fail. */
+	if (in_atomic())
+		goto out_nosemaphore;
+
+	down_read(&mm->mmap_sem);
+	vma = find_vma(mm, address);
+	if(!vma)
+		goto out;
+	else if(vma->vm_start <= address)
+		goto good_area;
+	else if(!(vma->vm_flags & VM_GROWSDOWN))
+		goto out;
+	else if(is_user && !ARCH_IS_STACKGROW(address))
+		goto out;
+	else if(expand_stack(vma, address))
+		goto out;
+
+good_area:
+	*code_out = SEGV_ACCERR;
+	if(is_write && !(vma->vm_flags & VM_WRITE))
+		goto out;
+
+	/* Don't require VM_READ|VM_EXEC for write faults! */
+        if(!is_write && !(vma->vm_flags & (VM_READ | VM_EXEC)))
+                goto out;
+
+	do {
+survive:
+		switch (handle_mm_fault(mm, vma, address, is_write)){
+		case VM_FAULT_MINOR:
+			current->min_flt++;
+			break;
+		case VM_FAULT_MAJOR:
+			current->maj_flt++;
+			break;
+		case VM_FAULT_SIGBUS:
+			err = -EACCES;
+			goto out;
+		case VM_FAULT_OOM:
+			err = -ENOMEM;
+			goto out_of_memory;
+		default:
+			BUG();
+		}
+		pgd = pgd_offset(mm, address);
+		pud = pud_offset(pgd, address);
+		pmd = pmd_offset(pud, address);
+		pte = pte_offset_kernel(pmd, address);
+	} while(!pte_present(*pte));
+	err = 0;
+	/* The below warning was added in place of
+	 *	pte_mkyoung(); if (is_write) pte_mkdirty();
+	 * If it's triggered, we'd see normally a hang here (a clean pte is
+	 * marked read-only to emulate the dirty bit).
+	 * However, the generic code can mark a PTE writable but clean on a
+	 * concurrent read fault, triggering this harmlessly. So comment it out.
+	 */
+#if 0
+	WARN_ON(!pte_young(*pte) || (is_write && !pte_dirty(*pte)));
+#endif
+	flush_tlb_page(vma, address);
+out:
+	up_read(&mm->mmap_sem);
+out_nosemaphore:
+	return(err);
+
+/*
+ * We ran out of memory, or some other thing happened to us that made
+ * us unable to handle the page fault gracefully.
+ */
+out_of_memory:
+	if (is_init(current)) {
+		up_read(&mm->mmap_sem);
+		yield();
+		down_read(&mm->mmap_sem);
+		goto survive;
+	}
+	goto out;
+}
+
+static void bad_segv(struct faultinfo fi, unsigned long ip)
+{
+	struct siginfo si;
+
+	si.si_signo = SIGSEGV;
+	si.si_code = SEGV_ACCERR;
+	si.si_addr = (void __user *) FAULT_ADDRESS(fi);
+	current->thread.arch.faultinfo = fi;
+	force_sig_info(SIGSEGV, &si, current);
+}
+
+static void segv_handler(int sig, union uml_pt_regs *regs)
+{
+	struct faultinfo * fi = UPT_FAULTINFO(regs);
+
+	if(UPT_IS_USER(regs) && !SEGV_IS_FIXABLE(fi)){
+		bad_segv(*fi, UPT_IP(regs));
+		return;
+	}
+	segv(*fi, UPT_IP(regs), UPT_IS_USER(regs), regs);
+}
+
+/*
+ * We give a *copy* of the faultinfo in the regs to segv.
+ * This must be done, since nesting SEGVs could overwrite
+ * the info in the regs. A pointer to the info then would
+ * give us bad data!
+ */
+unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc)
+{
+	struct siginfo si;
+	void *catcher;
+	int err;
+        int is_write = FAULT_WRITE(fi);
+        unsigned long address = FAULT_ADDRESS(fi);
+
+        if(!is_user && (address >= start_vm) && (address < end_vm)){
+                flush_tlb_kernel_vm();
+                return(0);
+        }
+	else if(current->mm == NULL)
+		panic("Segfault with no mm");
+
+	if (SEGV_IS_FIXABLE(&fi) || SEGV_MAYBE_FIXABLE(&fi))
+		err = handle_page_fault(address, ip, is_write, is_user, &si.si_code);
+	else {
+		err = -EFAULT;
+		/* A thread accessed NULL, we get a fault, but CR2 is invalid.
+		 * This code is used in __do_copy_from_user() of TT mode. */
+		address = 0;
+	}
+
+	catcher = current->thread.fault_catcher;
+	if(!err)
+		return(0);
+	else if(catcher != NULL){
+		current->thread.fault_addr = (void *) address;
+		do_longjmp(catcher, 1);
+	}
+	else if(current->thread.fault_addr != NULL)
+		panic("fault_addr set but no fault catcher");
+        else if(!is_user && arch_fixup(ip, sc))
+		return(0);
+
+ 	if(!is_user)
+		panic("Kernel mode fault at addr 0x%lx, ip 0x%lx",
+		      address, ip);
+
+	if (err == -EACCES) {
+		si.si_signo = SIGBUS;
+		si.si_errno = 0;
+		si.si_code = BUS_ADRERR;
+		si.si_addr = (void __user *)address;
+                current->thread.arch.faultinfo = fi;
+		force_sig_info(SIGBUS, &si, current);
+	} else if (err == -ENOMEM) {
+		printk("VM: killing process %s\n", current->comm);
+		do_exit(SIGKILL);
+	} else {
+		BUG_ON(err != -EFAULT);
+		si.si_signo = SIGSEGV;
+		si.si_addr = (void __user *) address;
+                current->thread.arch.faultinfo = fi;
+		force_sig_info(SIGSEGV, &si, current);
+	}
+	return(0);
+}
+
+void relay_signal(int sig, union uml_pt_regs *regs)
+{
+	if(arch_handle_signal(sig, regs))
+		return;
+
+	if(!UPT_IS_USER(regs)){
+		if(sig == SIGBUS)
+			printk("Bus error - the /dev/shm or /tmp mount likely "
+			       "just ran out of space\n");
+		panic("Kernel mode signal %d", sig);
+	}
+
+        current->thread.arch.faultinfo = *UPT_FAULTINFO(regs);
+	force_sig(sig, current);
+}
+
+static void bus_handler(int sig, union uml_pt_regs *regs)
+{
+	if(current->thread.fault_catcher != NULL)
+		do_longjmp(current->thread.fault_catcher, 1);
+	else relay_signal(sig, regs);
+}
+
+static void winch(int sig, union uml_pt_regs *regs)
+{
+	do_IRQ(WINCH_IRQ, regs);
+}
+
+const struct kern_handlers handlinfo_kern = {
+	.relay_signal = relay_signal,
+	.winch = winch,
+	.bus_handler = bus_handler,
+	.page_fault = segv_handler,
+	.sigio_handler = sigio_handler,
+	.timer_handler = timer_handler
+};
+
+void trap_init(void)
+{
+}

arch/um/kernel/tt/Makefile

@@ -0,0 +1,14 @@
+# 
+# Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com)
+# Licensed under the GPL
+#
+
+obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \
+	syscall_kern.o syscall_user.o tlb.o tracer.o trap_user.o \
+	uaccess.o uaccess_user.o
+
+obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/
+
+USER_OBJS := gdb.o tracer.o
+
+include arch/um/scripts/Makefile.rules

arch/um/kernel/tt/exec_kern.c

@@ -0,0 +1,85 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/mm.h"
+#include "asm/signal.h"
+#include "asm/ptrace.h"
+#include "asm/uaccess.h"
+#include "asm/pgalloc.h"
+#include "asm/tlbflush.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "irq_user.h"
+#include "mem_user.h"
+#include "os.h"
+#include "tlb.h"
+#include "mode.h"
+
+static int exec_tramp(void *sig_stack)
+{
+	init_new_thread_stack(sig_stack, NULL);
+	init_new_thread_signals();
+	os_stop_process(os_getpid());
+	return(0);
+}
+
+void flush_thread_tt(void)
+{
+	unsigned long stack;
+	int new_pid;
+
+	stack = alloc_stack(0, 0);
+	if(stack == 0){
+		printk(KERN_ERR 
+		       "flush_thread : failed to allocate temporary stack\n");
+		do_exit(SIGKILL);
+	}
+		
+	new_pid = start_fork_tramp(task_stack_page(current), stack, 0, exec_tramp);
+	if(new_pid < 0){
+		printk(KERN_ERR 
+		       "flush_thread : new thread failed, errno = %d\n",
+		       -new_pid);
+		do_exit(SIGKILL);
+	}
+
+	if(current_thread->cpu == 0)
+		forward_interrupts(new_pid);
+	current->thread.request.op = OP_EXEC;
+	current->thread.request.u.exec.pid = new_pid;
+	unprotect_stack((unsigned long) current_thread);
+	os_usr1_process(os_getpid());
+	change_sig(SIGUSR1, 1);
+
+	change_sig(SIGUSR1, 0);
+	enable_timer();
+	free_page(stack);
+	protect_memory(uml_reserved, high_physmem - uml_reserved, 1, 1, 0, 1);
+	task_protections((unsigned long) current_thread);
+	force_flush_all();
+	unblock_signals();
+}
+
+void start_thread_tt(struct pt_regs *regs, unsigned long eip, 
+		     unsigned long esp)
+{
+	set_fs(USER_DS);
+	flush_tlb_mm(current->mm);
+	PT_REGS_IP(regs) = eip;
+	PT_REGS_SP(regs) = esp;
+	PT_FIX_EXEC_STACK(esp);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/exec_user.c

@@ -0,0 +1,57 @@
+/* 
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <sched.h>
+#include <errno.h>
+#include <sys/wait.h>
+#include <signal.h>
+#include "user_util.h"
+#include "kern_util.h"
+#include "user.h"
+#include "ptrace_user.h"
+#include "os.h"
+
+void do_exec(int old_pid, int new_pid)
+{
+	unsigned long regs[FRAME_SIZE];
+	int err;
+
+	if((ptrace(PTRACE_ATTACH, new_pid, 0, 0) < 0) ||
+	   (ptrace(PTRACE_CONT, new_pid, 0, 0) < 0))
+		tracer_panic("do_exec failed to attach proc - errno = %d",
+			     errno);
+
+	CATCH_EINTR(err = waitpid(new_pid, 0, WUNTRACED));
+	if (err < 0)
+		tracer_panic("do_exec failed to attach proc in waitpid - errno = %d",
+			     errno);
+
+	if(ptrace_getregs(old_pid, regs) < 0)
+		tracer_panic("do_exec failed to get registers - errno = %d",
+			     errno);
+
+	os_kill_ptraced_process(old_pid, 0);
+
+	if (ptrace(PTRACE_OLDSETOPTIONS, new_pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0)
+		tracer_panic("do_exec: PTRACE_SETOPTIONS failed, errno = %d", errno);
+
+	if(ptrace_setregs(new_pid, regs) < 0)
+		tracer_panic("do_exec failed to start new proc - errno = %d",
+			     errno);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/gdb.c

@@ -0,0 +1,279 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <string.h>
+#include <signal.h>
+#include <sys/types.h>
+#include "ptrace_user.h"
+#include "uml-config.h"
+#include "kern_constants.h"
+#include "chan_user.h"
+#include "init.h"
+#include "user.h"
+#include "debug.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "tt.h"
+#include "sysdep/thread.h"
+#include "os.h"
+
+extern int debugger_pid;
+extern int debugger_fd;
+extern int debugger_parent;
+
+int detach(int pid, int sig)
+{
+	return(ptrace(PTRACE_DETACH, pid, 0, sig));
+}
+
+int attach(int pid)
+{
+	int err;
+
+	err = ptrace(PTRACE_ATTACH, pid, 0, 0);
+	if(err < 0) return(-errno);
+	else return(err);
+}
+
+int cont(int pid)
+{
+	return(ptrace(PTRACE_CONT, pid, 0, 0));
+}
+
+#ifdef UML_CONFIG_PT_PROXY
+
+int debugger_signal(int status, pid_t pid)
+{
+	return(debugger_proxy(status, pid));
+}
+
+void child_signal(pid_t pid, int status)
+{
+	child_proxy(pid, status);
+}
+
+static void gdb_announce(char *dev_name, int dev)
+{
+	printf("gdb assigned device '%s'\n", dev_name);
+}
+
+static struct chan_opts opts = {
+	.announce  	= gdb_announce,
+	.xterm_title 	= "UML kernel debugger",
+	.raw 		= 0,
+	.tramp_stack 	= 0,
+	.in_kernel  	= 0,
+};
+
+/* Accessed by the tracing thread, which automatically serializes access */
+static void *xterm_data;
+static int xterm_fd;
+
+extern void *xterm_init(char *, int, struct chan_opts *);
+extern int xterm_open(int, int, int, void *, char **);
+extern void xterm_close(int, void *);
+
+int open_gdb_chan(void)
+{
+	char stack[UM_KERN_PAGE_SIZE], *dummy;
+
+	opts.tramp_stack = (unsigned long) stack;
+	xterm_data = xterm_init("", 0, &opts);
+	xterm_fd = xterm_open(1, 1, 1, xterm_data, &dummy);
+	return(xterm_fd);
+}
+
+static void exit_debugger_cb(void *unused)
+{
+	if(debugger_pid != -1){
+		if(gdb_pid != -1){
+			fake_child_exit();
+			gdb_pid = -1;
+		}
+		else kill_child_dead(debugger_pid);
+		debugger_pid = -1;
+		if(debugger_parent != -1)
+			detach(debugger_parent, SIGINT);
+	}
+	if(xterm_data != NULL) xterm_close(xterm_fd, xterm_data);
+}
+
+static void exit_debugger(void)
+{
+	initial_thread_cb(exit_debugger_cb, NULL);
+}
+
+__uml_exitcall(exit_debugger);
+
+struct gdb_data {
+	char *str;
+	int err;
+};
+
+static void config_gdb_cb(void *arg)
+{
+	struct gdb_data *data = arg;
+	void *task;
+	int pid;
+
+	data->err = -1;
+	if(debugger_pid != -1) exit_debugger_cb(NULL);
+	if(!strncmp(data->str, "pid,", strlen("pid,"))){
+		data->str += strlen("pid,");
+		pid = strtoul(data->str, NULL, 0);
+		task = cpu_tasks[0].task;
+		debugger_pid = attach_debugger(TASK_EXTERN_PID(task), pid, 0);
+		if(debugger_pid != -1){
+			data->err = 0;
+			gdb_pid = pid;
+		}
+		return;
+	}
+	data->err = 0;
+	debugger_pid = start_debugger(linux_prog, 0, 0, &debugger_fd);
+	init_proxy(debugger_pid, 0, 0);
+}
+
+int gdb_config(char *str, char **error_out)
+{
+	struct gdb_data data;
+
+	if(*str++ != '=') return(-1);
+	data.str = str;
+	initial_thread_cb(config_gdb_cb, &data);
+	return(data.err);
+}
+
+void remove_gdb_cb(void *unused)
+{
+	exit_debugger_cb(NULL);
+}
+
+int gdb_remove(int unused, char **error_out)
+{
+	initial_thread_cb(remove_gdb_cb, NULL);
+        return 0;
+}
+
+void signal_usr1(int sig)
+{
+	if(debugger_pid != -1){
+		printf("The debugger is already running\n");
+		return;
+	}
+	debugger_pid = start_debugger(linux_prog, 0, 0, &debugger_fd);
+	init_proxy(debugger_pid, 0, 0);
+}
+
+int init_ptrace_proxy(int idle_pid, int startup, int stop)
+{
+	int pid, status;
+
+	pid = start_debugger(linux_prog, startup, stop, &debugger_fd);
+	status = wait_for_stop(idle_pid, SIGSTOP, PTRACE_CONT, NULL);
+ 	if(pid < 0){
+		cont(idle_pid);
+		return(-1);
+	}
+	init_proxy(pid, 1, status);
+	return(pid);
+}
+
+int attach_debugger(int idle_pid, int pid, int stop)
+{
+	int status = 0, err;
+
+	err = attach(pid);
+	if(err < 0){
+		printf("Failed to attach pid %d, errno = %d\n", pid, -err);
+		return(-1);
+	}
+	if(stop) status = wait_for_stop(idle_pid, SIGSTOP, PTRACE_CONT, NULL);
+	init_proxy(pid, 1, status);
+	return(pid);
+}
+
+#ifdef notdef /* Put this back in when it does something useful */
+static int __init uml_gdb_init_setup(char *line, int *add)
+{
+	gdb_init = uml_strdup(line);
+	return 0;
+}
+
+__uml_setup("gdb=", uml_gdb_init_setup, 
+"gdb=<channel description>\n\n"
+);
+#endif
+
+static int __init uml_gdb_pid_setup(char *line, int *add)
+{
+	gdb_pid = strtoul(line, NULL, 0);
+	*add = 0;
+	return 0;
+}
+
+__uml_setup("gdb-pid=", uml_gdb_pid_setup, 
+"gdb-pid=<pid>\n"
+"    gdb-pid is used to attach an external debugger to UML.  This may be\n"
+"    an already-running gdb or a debugger-like process like strace.\n\n"
+);
+
+#else
+
+int debugger_signal(int status, pid_t pid){ return(0); }
+void child_signal(pid_t pid, int status){ }
+int init_ptrace_proxy(int idle_pid, int startup, int stop)
+{
+	printf("debug requested when CONFIG_PT_PROXY is off\n");
+	kill_child_dead(idle_pid);
+	exit(1);
+}
+
+void signal_usr1(int sig)
+{
+	printf("debug requested when CONFIG_PT_PROXY is off\n");
+}
+
+int attach_debugger(int idle_pid, int pid, int stop)
+{
+	printf("attach_debugger called when CONFIG_PT_PROXY "
+	       "is off\n");
+	return(-1);
+}
+
+int config_gdb(char *str)
+{
+	return(-1);
+}
+
+int remove_gdb(void)
+{
+	return(-1);
+}
+
+int init_parent_proxy(int pid)
+{
+	return(-1);
+}
+
+void debugger_parent_signal(int status, int pid)
+{
+}
+
+#endif
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/gdb_kern.c

@@ -0,0 +1,40 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/init.h"
+#include "mconsole_kern.h"
+
+#ifdef CONFIG_MCONSOLE
+
+extern int gdb_config(char *str, char **error_out);
+extern int gdb_remove(int n, char **error_out);
+
+static struct mc_device gdb_mc = {
+	.list		= INIT_LIST_HEAD(gdb_mc.list),
+	.name		= "gdb",
+	.config		= gdb_config,
+	.remove		= gdb_remove,
+};
+
+int gdb_mc_init(void)
+{
+	mconsole_register_dev(&gdb_mc);
+	return(0);
+}
+
+__initcall(gdb_mc_init);
+
+#endif
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/include/mode-tt.h

@@ -0,0 +1,34 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __MODE_TT_H__
+#define __MODE_TT_H__
+
+#include "sysdep/ptrace.h"
+
+enum { OP_NONE, OP_EXEC, OP_FORK, OP_TRACE_ON, OP_REBOOT, OP_HALT, OP_CB };
+
+extern int tracing_pid;
+
+extern int tracer(int (*init_proc)(void *), void *sp);
+extern void sig_handler_common_tt(int sig, void *sc);
+extern void syscall_handler_tt(int sig, union uml_pt_regs *regs);
+extern void reboot_tt(void);
+extern void halt_tt(void);
+extern int is_tracer_winch(int pid, int fd, void *data);
+extern void kill_off_processes_tt(void);
+
+#endif
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/include/mode_kern-tt.h

@@ -0,0 +1,52 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#ifndef __TT_MODE_KERN_H__
+#define __TT_MODE_KERN_H__
+
+#include "linux/sched.h"
+#include "asm/page.h"
+#include "asm/ptrace.h"
+#include "asm/uaccess.h"
+
+extern void switch_to_tt(void *prev, void *next);
+extern void flush_thread_tt(void);
+extern void start_thread_tt(struct pt_regs *regs, unsigned long eip,
+			   unsigned long esp);
+extern int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp,
+			  unsigned long stack_top, struct task_struct *p,
+			  struct pt_regs *regs);
+extern void release_thread_tt(struct task_struct *task);
+extern void initial_thread_cb_tt(void (*proc)(void *), void *arg);
+extern void init_idle_tt(void);
+extern void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end);
+extern void flush_tlb_kernel_vm_tt(void);
+extern void __flush_tlb_one_tt(unsigned long addr);
+extern void flush_tlb_range_tt(struct vm_area_struct *vma,
+			       unsigned long start, unsigned long end);
+extern void flush_tlb_mm_tt(struct mm_struct *mm);
+extern void force_flush_all_tt(void);
+extern long execute_syscall_tt(void *r);
+extern void before_mem_tt(unsigned long brk_start);
+extern unsigned long set_task_sizes_tt(int arg, unsigned long *host_size_out,
+				       unsigned long *task_size_out);
+extern int start_uml_tt(void);
+extern int external_pid_tt(struct task_struct *task);
+extern int thread_pid_tt(struct task_struct *task);
+
+#define kmem_end_tt (host_task_size - ABOVE_KMEM)
+
+#endif
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ksyms.c

@@ -0,0 +1,29 @@
+/* 
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/module.h"
+#include "asm/uaccess.h"
+#include "mode.h"
+
+EXPORT_SYMBOL(__do_copy_from_user);
+EXPORT_SYMBOL(__do_copy_to_user);
+EXPORT_SYMBOL(__do_strncpy_from_user);
+EXPORT_SYMBOL(__do_strnlen_user); 
+EXPORT_SYMBOL(__do_clear_user);
+EXPORT_SYMBOL(clear_user_tt);
+
+EXPORT_SYMBOL(tracing_pid);
+EXPORT_SYMBOL(honeypot);
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/mem.c

@@ -0,0 +1,35 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/stddef.h"
+#include "linux/mm.h"
+#include "asm/uaccess.h"
+#include "mem_user.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "kern.h"
+#include "tt.h"
+
+void before_mem_tt(unsigned long brk_start)
+{
+	if(debug)
+		remap_data(UML_ROUND_DOWN(&_stext), UML_ROUND_UP(&_etext), 1);
+	remap_data(UML_ROUND_DOWN(&_sdata), UML_ROUND_UP(&_edata), 1);
+	remap_data(UML_ROUND_DOWN(&__bss_start), UML_ROUND_UP(&_end), 1);
+}
+
+#define SIZE ((CONFIG_NEST_LEVEL + CONFIG_KERNEL_HALF_GIGS) * 0x20000000)
+#define START (CONFIG_TOP_ADDR - SIZE)
+
+unsigned long set_task_sizes_tt(unsigned long *task_size_out)
+{
+	unsigned long host_task_size;
+
+	/* Round up to the nearest 4M */
+	host_task_size = ROUND_4M((unsigned long) &host_task_size);
+	*task_size_out = START;
+
+	return host_task_size;
+}

arch/um/kernel/tt/mem_user.c

@@ -0,0 +1,50 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <string.h>
+#include <errno.h>
+#include <sys/mman.h>
+#include "tt.h"
+#include "mem_user.h"
+#include "user_util.h"
+#include "os.h"
+
+void remap_data(void *segment_start, void *segment_end, int w)
+{
+	void *addr;
+	unsigned long size;
+	int data, prot;
+
+	if(w) prot = PROT_WRITE;
+	else prot = 0;
+	prot |= PROT_READ | PROT_EXEC;
+	size = (unsigned long) segment_end - 
+		(unsigned long) segment_start;
+	data = create_mem_file(size);
+	addr = mmap(NULL, size, PROT_WRITE | PROT_READ, MAP_SHARED, data, 0);
+	if(addr == MAP_FAILED){
+		perror("mapping new data segment");
+		exit(1);
+	}
+	memcpy(addr, segment_start, size);
+	if(switcheroo(data, prot, addr, segment_start, size) < 0){
+		printf("switcheroo failed\n");
+		exit(1);
+	}
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/process_kern.c

@@ -0,0 +1,461 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/signal.h"
+#include "linux/kernel.h"
+#include "linux/interrupt.h"
+#include "linux/ptrace.h"
+#include "asm/system.h"
+#include "asm/pgalloc.h"
+#include "asm/ptrace.h"
+#include "asm/tlbflush.h"
+#include "irq_user.h"
+#include "kern_util.h"
+#include "user_util.h"
+#include "os.h"
+#include "kern.h"
+#include "sigcontext.h"
+#include "mem_user.h"
+#include "tlb.h"
+#include "mode.h"
+#include "mode_kern.h"
+#include "init.h"
+#include "tt.h"
+
+void switch_to_tt(void *prev, void *next)
+{
+	struct task_struct *from, *to, *prev_sched;
+	unsigned long flags;
+	int err, vtalrm, alrm, prof, cpu;
+	char c;
+
+	from = prev;
+	to = next;
+
+	cpu = task_thread_info(from)->cpu;
+	if(cpu == 0)
+		forward_interrupts(to->thread.mode.tt.extern_pid);
+#ifdef CONFIG_SMP
+	forward_ipi(cpu_data[cpu].ipi_pipe[0], to->thread.mode.tt.extern_pid);
+#endif
+	local_irq_save(flags);
+
+	vtalrm = change_sig(SIGVTALRM, 0);
+	alrm = change_sig(SIGALRM, 0);
+	prof = change_sig(SIGPROF, 0);
+
+	forward_pending_sigio(to->thread.mode.tt.extern_pid);
+
+	c = 0;
+
+	/* Notice that here we "up" the semaphore on which "to" is waiting, and
+	 * below (the read) we wait on this semaphore (which is implemented by
+	 * switch_pipe) and go sleeping. Thus, after that, we have resumed in
+	 * "to", and can't use any more the value of "from" (which is outdated),
+	 * nor the value in "to" (since it was the task which stole us the CPU,
+	 * which we don't care about). */
+
+	err = os_write_file(to->thread.mode.tt.switch_pipe[1], &c, sizeof(c));
+	if(err != sizeof(c))
+		panic("write of switch_pipe failed, err = %d", -err);
+
+	if(from->thread.mode.tt.switch_pipe[0] == -1)
+		os_kill_process(os_getpid(), 0);
+
+	err = os_read_file(from->thread.mode.tt.switch_pipe[0], &c, sizeof(c));
+	if(err != sizeof(c))
+		panic("read of switch_pipe failed, errno = %d", -err);
+
+	/* If the process that we have just scheduled away from has exited,
+	 * then it needs to be killed here.  The reason is that, even though
+	 * it will kill itself when it next runs, that may be too late.  Its
+	 * stack will be freed, possibly before then, and if that happens,
+	 * we have a use-after-free situation.  So, it gets killed here
+	 * in case it has not already killed itself.
+	 */
+	prev_sched = current->thread.prev_sched;
+        if(prev_sched->thread.mode.tt.switch_pipe[0] == -1)
+		os_kill_process(prev_sched->thread.mode.tt.extern_pid, 1);
+
+	change_sig(SIGVTALRM, vtalrm);
+	change_sig(SIGALRM, alrm);
+	change_sig(SIGPROF, prof);
+
+	arch_switch_to_tt(prev_sched, current);
+
+	flush_tlb_all();
+	local_irq_restore(flags);
+}
+
+void release_thread_tt(struct task_struct *task)
+{
+	int pid = task->thread.mode.tt.extern_pid;
+
+	/*
+         * We first have to kill the other process, before
+         * closing its switch_pipe. Else it might wake up
+         * and receive "EOF" before we could kill it.
+         */
+	if(os_getpid() != pid)
+		os_kill_process(pid, 0);
+
+        os_close_file(task->thread.mode.tt.switch_pipe[0]);
+        os_close_file(task->thread.mode.tt.switch_pipe[1]);
+	/* use switch_pipe as flag: thread is released */
+        task->thread.mode.tt.switch_pipe[0] = -1;
+}
+
+void suspend_new_thread(int fd)
+{
+	int err;
+	char c;
+
+	os_stop_process(os_getpid());
+	err = os_read_file(fd, &c, sizeof(c));
+	if(err != sizeof(c))
+		panic("read failed in suspend_new_thread, err = %d", -err);
+}
+
+void schedule_tail(struct task_struct *prev);
+
+static void new_thread_handler(int sig)
+{
+	unsigned long disable;
+	int (*fn)(void *);
+	void *arg;
+
+	fn = current->thread.request.u.thread.proc;
+	arg = current->thread.request.u.thread.arg;
+
+	UPT_SC(&current->thread.regs.regs) = (void *) (&sig + 1);
+	disable = (1 << (SIGVTALRM - 1)) | (1 << (SIGALRM - 1)) |
+		(1 << (SIGIO - 1)) | (1 << (SIGPROF - 1));
+	SC_SIGMASK(UPT_SC(&current->thread.regs.regs)) &= ~disable;
+
+	suspend_new_thread(current->thread.mode.tt.switch_pipe[0]);
+
+	force_flush_all();
+	if(current->thread.prev_sched != NULL)
+		schedule_tail(current->thread.prev_sched);
+	current->thread.prev_sched = NULL;
+
+	init_new_thread_signals();
+	enable_timer();
+	free_page(current->thread.temp_stack);
+	set_cmdline("(kernel thread)");
+
+	change_sig(SIGUSR1, 1);
+	change_sig(SIGPROF, 1);
+	local_irq_enable();
+	if(!run_kernel_thread(fn, arg, &current->thread.exec_buf))
+		do_exit(0);
+
+	/* XXX No set_user_mode here because a newly execed process will
+	 * immediately segfault on its non-existent IP, coming straight back
+	 * to the signal handler, which will call set_user_mode on its way
+	 * out.  This should probably change since it's confusing.
+	 */
+}
+
+static int new_thread_proc(void *stack)
+{
+	/* local_irq_disable is needed to block out signals until this thread is
+	 * properly scheduled.  Otherwise, the tracing thread will get mighty
+	 * upset about any signals that arrive before that.
+	 * This has the complication that it sets the saved signal mask in
+	 * the sigcontext to block signals.  This gets restored when this
+	 * thread (or a descendant, since they get a copy of this sigcontext)
+	 * returns to userspace.
+	 * So, this is compensated for elsewhere.
+	 * XXX There is still a small window until local_irq_disable() actually
+	 * finishes where signals are possible - shouldn't be a problem in
+	 * practice since SIGIO hasn't been forwarded here yet, and the
+	 * local_irq_disable should finish before a SIGVTALRM has time to be
+	 * delivered.
+	 */
+
+	local_irq_disable();
+	init_new_thread_stack(stack, new_thread_handler);
+	os_usr1_process(os_getpid());
+	change_sig(SIGUSR1, 1);
+	return(0);
+}
+
+/* Signal masking - signals are blocked at the start of fork_tramp.  They
+ * are re-enabled when finish_fork_handler is entered by fork_tramp hitting
+ * itself with a SIGUSR1.  set_user_mode has to be run with SIGUSR1 off,
+ * so it is blocked before it's called.  They are re-enabled on sigreturn
+ * despite the fact that they were blocked when the SIGUSR1 was issued because
+ * copy_thread copies the parent's sigcontext, including the signal mask
+ * onto the signal frame.
+ */
+
+void finish_fork_handler(int sig)
+{
+ 	UPT_SC(&current->thread.regs.regs) = (void *) (&sig + 1);
+	suspend_new_thread(current->thread.mode.tt.switch_pipe[0]);
+
+	force_flush_all();
+	if(current->thread.prev_sched != NULL)
+		schedule_tail(current->thread.prev_sched);
+	current->thread.prev_sched = NULL;
+
+	enable_timer();
+	change_sig(SIGVTALRM, 1);
+	local_irq_enable();
+	if(current->mm != current->parent->mm)
+		protect_memory(uml_reserved, high_physmem - uml_reserved, 1, 
+			       1, 0, 1);
+	task_protections((unsigned long) current_thread);
+
+	free_page(current->thread.temp_stack);
+	local_irq_disable();
+	change_sig(SIGUSR1, 0);
+	set_user_mode(current);
+}
+
+int fork_tramp(void *stack)
+{
+	local_irq_disable();
+	arch_init_thread();
+	init_new_thread_stack(stack, finish_fork_handler);
+
+	os_usr1_process(os_getpid());
+	change_sig(SIGUSR1, 1);
+	return(0);
+}
+
+int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp,
+		   unsigned long stack_top, struct task_struct * p, 
+		   struct pt_regs *regs)
+{
+	int (*tramp)(void *);
+	int new_pid, err;
+	unsigned long stack;
+	
+	if(current->thread.forking)
+		tramp = fork_tramp;
+	else {
+		tramp = new_thread_proc;
+		p->thread.request.u.thread = current->thread.request.u.thread;
+	}
+
+	err = os_pipe(p->thread.mode.tt.switch_pipe, 1, 1);
+	if(err < 0){
+		printk("copy_thread : pipe failed, err = %d\n", -err);
+		return(err);
+	}
+
+	stack = alloc_stack(0, 0);
+	if(stack == 0){
+		printk(KERN_ERR "copy_thread : failed to allocate "
+		       "temporary stack\n");
+		return(-ENOMEM);
+	}
+
+	clone_flags &= CLONE_VM;
+	p->thread.temp_stack = stack;
+	new_pid = start_fork_tramp(task_stack_page(p), stack, clone_flags, tramp);
+	if(new_pid < 0){
+		printk(KERN_ERR "copy_thread : clone failed - errno = %d\n", 
+		       -new_pid);
+		return(new_pid);
+	}
+
+	if(current->thread.forking){
+		sc_to_sc(UPT_SC(&p->thread.regs.regs), UPT_SC(&regs->regs));
+		SC_SET_SYSCALL_RETURN(UPT_SC(&p->thread.regs.regs), 0);
+		if(sp != 0)
+			SC_SP(UPT_SC(&p->thread.regs.regs)) = sp;
+	}
+	p->thread.mode.tt.extern_pid = new_pid;
+
+	current->thread.request.op = OP_FORK;
+	current->thread.request.u.fork.pid = new_pid;
+	os_usr1_process(os_getpid());
+
+	/* Enable the signal and then disable it to ensure that it is handled
+	 * here, and nowhere else.
+	 */
+	change_sig(SIGUSR1, 1);
+
+	change_sig(SIGUSR1, 0);
+	err = 0;
+	return(err);
+}
+
+void reboot_tt(void)
+{
+	current->thread.request.op = OP_REBOOT;
+	os_usr1_process(os_getpid());
+	change_sig(SIGUSR1, 1);
+}
+
+void halt_tt(void)
+{
+	current->thread.request.op = OP_HALT;
+	os_usr1_process(os_getpid());
+	change_sig(SIGUSR1, 1);
+}
+
+void kill_off_processes_tt(void)
+{
+	struct task_struct *p;
+	int me;
+
+	me = os_getpid();
+        for_each_process(p){
+		if(p->thread.mode.tt.extern_pid != me) 
+			os_kill_process(p->thread.mode.tt.extern_pid, 0);
+	}
+	if(init_task.thread.mode.tt.extern_pid != me) 
+		os_kill_process(init_task.thread.mode.tt.extern_pid, 0);
+}
+
+void initial_thread_cb_tt(void (*proc)(void *), void *arg)
+{
+	if(os_getpid() == tracing_pid){
+		(*proc)(arg);
+	}
+	else {
+		current->thread.request.op = OP_CB;
+		current->thread.request.u.cb.proc = proc;
+		current->thread.request.u.cb.arg = arg;
+		os_usr1_process(os_getpid());
+		change_sig(SIGUSR1, 1);
+
+		change_sig(SIGUSR1, 0);
+	}
+}
+
+int do_proc_op(void *t, int proc_id)
+{
+	struct task_struct *task;
+	struct thread_struct *thread;
+	int op, pid;
+
+	task = t;
+	thread = &task->thread;
+	op = thread->request.op;
+	switch(op){
+	case OP_NONE:
+	case OP_TRACE_ON:
+		break;
+	case OP_EXEC:
+		pid = thread->request.u.exec.pid;
+		do_exec(thread->mode.tt.extern_pid, pid);
+		thread->mode.tt.extern_pid = pid;
+		cpu_tasks[task_thread_info(task)->cpu].pid = pid;
+		break;
+	case OP_FORK:
+		attach_process(thread->request.u.fork.pid);
+		break;
+	case OP_CB:
+		(*thread->request.u.cb.proc)(thread->request.u.cb.arg);
+		break;
+	case OP_REBOOT:
+	case OP_HALT:
+		break;
+	default:
+		tracer_panic("Bad op in do_proc_op");
+		break;
+	}
+	thread->request.op = OP_NONE;
+	return(op);
+}
+
+void init_idle_tt(void)
+{
+	default_idle();
+}
+
+extern void start_kernel(void);
+
+static int start_kernel_proc(void *unused)
+{
+	int pid;
+
+	block_signals();
+	pid = os_getpid();
+
+	cpu_tasks[0].pid = pid;
+	cpu_tasks[0].task = current;
+#ifdef CONFIG_SMP
+ 	cpu_online_map = cpumask_of_cpu(0);
+#endif
+	if(debug) os_stop_process(pid);
+	start_kernel();
+	return(0);
+}
+
+void set_tracing(void *task, int tracing)
+{
+	((struct task_struct *) task)->thread.mode.tt.tracing = tracing;
+}
+
+int is_tracing(void *t)
+{
+	return (((struct task_struct *) t)->thread.mode.tt.tracing);
+}
+
+int set_user_mode(void *t)
+{
+	struct task_struct *task;
+
+	task = t ? t : current;
+	if(task->thread.mode.tt.tracing) 
+		return(1);
+	task->thread.request.op = OP_TRACE_ON;
+	os_usr1_process(os_getpid());
+	return(0);
+}
+
+void set_init_pid(int pid)
+{
+	int err;
+
+	init_task.thread.mode.tt.extern_pid = pid;
+	err = os_pipe(init_task.thread.mode.tt.switch_pipe, 1, 1);
+	if(err)
+		panic("Can't create switch pipe for init_task, errno = %d",
+		      -err);
+}
+
+int start_uml_tt(void)
+{
+	void *sp;
+	int pages;
+
+	pages = (1 << CONFIG_KERNEL_STACK_ORDER);
+	sp = task_stack_page(&init_task) +
+		pages * PAGE_SIZE - sizeof(unsigned long);
+	return(tracer(start_kernel_proc, sp));
+}
+
+int external_pid_tt(struct task_struct *task)
+{
+	return(task->thread.mode.tt.extern_pid);
+}
+
+int thread_pid_tt(struct task_struct *task)
+{
+	return(task->thread.mode.tt.extern_pid);
+}
+
+int is_valid_pid(int pid)
+{
+	struct task_struct *task;
+
+        read_lock(&tasklist_lock);
+        for_each_process(task){
+                if(task->thread.mode.tt.extern_pid == pid){
+			read_unlock(&tasklist_lock);
+			return(1);
+                }
+        }
+	read_unlock(&tasklist_lock);
+	return(0);
+}

arch/um/kernel/tt/ptproxy/Makefile

@@ -0,0 +1,10 @@
+# 
+# Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+# Licensed under the GPL
+#
+
+obj-y = proxy.o ptrace.o sysdep.o wait.o
+
+USER_OBJS := $(obj-y)
+
+include arch/um/scripts/Makefile.rules

arch/um/kernel/tt/ptproxy/proxy.c

@@ -0,0 +1,377 @@
+/**********************************************************************
+proxy.c
+
+Copyright (C) 1999 Lars Brinkhoff.  See the file COPYING for licensing
+terms and conditions.
+
+Jeff Dike (jdike@karaya.com) : Modified for integration into uml
+**********************************************************************/
+
+/* XXX This file shouldn't refer to CONFIG_* */
+
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <signal.h>
+#include <string.h>
+#include <termios.h>
+#include <sys/wait.h>
+#include <sys/types.h>
+#include <sys/ioctl.h>
+#include <asm/unistd.h>
+#include "ptrace_user.h"
+
+#include "ptproxy.h"
+#include "sysdep.h"
+#include "wait.h"
+
+#include "user_util.h"
+#include "user.h"
+#include "os.h"
+#include "tempfile.h"
+
+static int debugger_wait(debugger_state *debugger, int *status, int options,
+			 int (*syscall)(debugger_state *debugger, pid_t child),
+			 int (*normal_return)(debugger_state *debugger, 
+					      pid_t unused),
+			 int (*wait_return)(debugger_state *debugger, 
+					    pid_t unused))
+{
+	if(debugger->real_wait){
+		debugger->handle_trace = normal_return;
+		syscall_continue(debugger->pid);
+		debugger->real_wait = 0;
+		return(1);
+	}
+	debugger->wait_status_ptr = status;
+	debugger->wait_options = options;
+	if((debugger->debugee != NULL) && debugger->debugee->event){
+		syscall_continue(debugger->pid);
+		wait_for_stop(debugger->pid, SIGTRAP, PTRACE_SYSCALL,
+			      NULL);
+		(*wait_return)(debugger, -1);
+		return(0);
+	}
+	else if(debugger->wait_options & WNOHANG){
+		syscall_cancel(debugger->pid, 0);
+		debugger->handle_trace = syscall;
+		return(0);
+	}
+	else {
+		syscall_pause(debugger->pid);
+		debugger->handle_trace = wait_return;
+		debugger->waiting = 1;
+	}
+	return(1);
+}
+
+/*
+ * Handle debugger trap, i.e. syscall.
+ */
+
+int debugger_syscall(debugger_state *debugger, pid_t child)
+{
+	long arg1, arg2, arg3, arg4, arg5, result;
+	int syscall, ret = 0;
+
+	syscall = get_syscall(debugger->pid, &arg1, &arg2, &arg3, &arg4, 
+			      &arg5);
+
+	switch(syscall){
+	case __NR_execve:
+		/* execve never returns */
+		debugger->handle_trace = debugger_syscall; 
+		break;
+
+	case __NR_ptrace:
+		if(debugger->debugee->pid != 0) arg2 = debugger->debugee->pid;
+		if(!debugger->debugee->in_context) 
+			child = debugger->debugee->pid;
+		result = proxy_ptrace(debugger, arg1, arg2, arg3, arg4, child,
+				      &ret);
+		syscall_cancel(debugger->pid, result);
+		debugger->handle_trace = debugger_syscall;
+		return(ret);
+
+#ifdef __NR_waitpid
+	case __NR_waitpid:
+#endif
+	case __NR_wait4:
+		if(!debugger_wait(debugger, (int *) arg2, arg3, 
+				  debugger_syscall, debugger_normal_return, 
+				  proxy_wait_return))
+			return(0);
+		break;
+
+	case __NR_kill:
+		if(!debugger->debugee->in_context) 
+			child = debugger->debugee->pid;
+		if(arg1 == debugger->debugee->pid){
+			result = kill(child, arg2);
+			syscall_cancel(debugger->pid, result);
+			debugger->handle_trace = debugger_syscall;
+			return(0);
+		}
+		else debugger->handle_trace = debugger_normal_return;
+		break;
+
+	default:
+		debugger->handle_trace = debugger_normal_return;
+	}
+
+	syscall_continue(debugger->pid);
+	return(0);
+}
+
+/* Used by the tracing thread */
+static debugger_state parent;
+static int parent_syscall(debugger_state *debugger, int pid);
+
+int init_parent_proxy(int pid)
+{
+	parent = ((debugger_state) { .pid 		= pid,
+				     .wait_options 	= 0,
+				     .wait_status_ptr 	= NULL,
+				     .waiting 		= 0,
+				     .real_wait 	= 0,
+				     .expecting_child 	= 0,
+				     .handle_trace  	= parent_syscall,
+				     .debugee 		= NULL } );
+	return(0);
+}
+
+int parent_normal_return(debugger_state *debugger, pid_t unused)
+{
+	debugger->handle_trace = parent_syscall;
+	syscall_continue(debugger->pid);
+	return(0);
+}
+
+static int parent_syscall(debugger_state *debugger, int pid)
+{
+	long arg1, arg2, arg3, arg4, arg5;
+	int syscall;
+
+	syscall = get_syscall(pid, &arg1, &arg2, &arg3, &arg4, &arg5);
+		
+	if((syscall == __NR_wait4)
+#ifdef __NR_waitpid
+	   || (syscall == __NR_waitpid)
+#endif
+	){
+		debugger_wait(&parent, (int *) arg2, arg3, parent_syscall,
+			      parent_normal_return, parent_wait_return);
+	}
+	else ptrace(PTRACE_SYSCALL, pid, 0, 0);
+	return(0);
+}
+
+int debugger_normal_return(debugger_state *debugger, pid_t unused)
+{
+	debugger->handle_trace = debugger_syscall;
+	syscall_continue(debugger->pid);
+	return(0);
+}
+
+void debugger_cancelled_return(debugger_state *debugger, int result)
+{
+	debugger->handle_trace = debugger_syscall;
+	syscall_set_result(debugger->pid, result);
+	syscall_continue(debugger->pid);
+}
+
+/* Used by the tracing thread */
+static debugger_state debugger;
+static debugee_state debugee;
+
+void init_proxy (pid_t debugger_pid, int stopped, int status)
+{
+	debugger.pid = debugger_pid;
+	debugger.handle_trace = debugger_syscall;
+	debugger.debugee = &debugee;
+	debugger.waiting = 0;
+	debugger.real_wait = 0;
+	debugger.expecting_child = 0;
+
+	debugee.pid = 0;
+	debugee.traced = 0;
+	debugee.stopped = stopped;
+	debugee.event = 0;
+	debugee.zombie = 0;
+	debugee.died = 0;
+	debugee.wait_status = status;
+	debugee.in_context = 1;
+}
+
+int debugger_proxy(int status, int pid)
+{
+	int ret = 0, sig;
+
+	if(WIFSTOPPED(status)){
+		sig = WSTOPSIG(status);
+		if (sig == SIGTRAP)
+			ret = (*debugger.handle_trace)(&debugger, pid);
+						       
+		else if(sig == SIGCHLD){
+			if(debugger.expecting_child){
+				ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig);
+				debugger.expecting_child = 0;
+			}
+			else if(debugger.waiting)
+				real_wait_return(&debugger);
+			else {
+				ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig);
+				debugger.real_wait = 1;
+			}
+		}
+		else ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig);
+	}
+	else if(WIFEXITED(status)){
+		tracer_panic("debugger (pid %d) exited with status %d", 
+			     debugger.pid, WEXITSTATUS(status));
+	}
+	else if(WIFSIGNALED(status)){
+		tracer_panic("debugger (pid %d) exited with signal %d", 
+			     debugger.pid, WTERMSIG(status));
+	}
+	else {
+		tracer_panic("proxy got unknown status (0x%x) on debugger "
+			     "(pid %d)", status, debugger.pid);
+	}
+	return(ret);
+}
+
+void child_proxy(pid_t pid, int status)
+{
+	debugee.event = 1;
+	debugee.wait_status = status;
+
+	if(WIFSTOPPED(status)){
+		debugee.stopped = 1;
+		debugger.expecting_child = 1;
+		kill(debugger.pid, SIGCHLD);
+	}
+	else if(WIFEXITED(status) || WIFSIGNALED(status)){
+		debugee.zombie = 1;
+		debugger.expecting_child = 1;
+		kill(debugger.pid, SIGCHLD);
+	}
+	else panic("proxy got unknown status (0x%x) on child (pid %d)", 
+		   status, pid);
+}
+
+void debugger_parent_signal(int status, int pid)
+{
+	int sig;
+
+	if(WIFSTOPPED(status)){
+		sig = WSTOPSIG(status);
+		if(sig == SIGTRAP) (*parent.handle_trace)(&parent, pid);
+		else ptrace(PTRACE_SYSCALL, pid, 0, sig);
+	}
+}
+
+void fake_child_exit(void)
+{
+	int status, pid;
+
+	child_proxy(1, W_EXITCODE(0, 0));
+	while(debugger.waiting == 1){
+		CATCH_EINTR(pid = waitpid(debugger.pid, &status, WUNTRACED));
+		if(pid != debugger.pid){
+			printk("fake_child_exit - waitpid failed, "
+			       "errno = %d\n", errno);
+			return;
+		}
+		debugger_proxy(status, debugger.pid);
+	}
+	CATCH_EINTR(pid = waitpid(debugger.pid, &status, WUNTRACED));
+	if(pid != debugger.pid){
+		printk("fake_child_exit - waitpid failed, "
+		       "errno = %d\n", errno);
+		return;
+	}
+	if(ptrace(PTRACE_DETACH, debugger.pid, 0, SIGCONT) < 0)
+		printk("fake_child_exit - PTRACE_DETACH failed, errno = %d\n",
+		       errno);
+}
+
+char gdb_init_string[] = 
+"att 1 \n\
+b panic \n\
+b stop \n\
+handle SIGWINCH nostop noprint pass \n\
+";
+
+int start_debugger(char *prog, int startup, int stop, int *fd_out)
+{
+	int slave, child;
+
+	slave = open_gdb_chan();
+	child = fork();
+	if(child == 0){
+		char *tempname = NULL;
+		int fd;
+
+	        if(setsid() < 0) perror("setsid");
+		if((dup2(slave, 0) < 0) || (dup2(slave, 1) < 0) || 
+		   (dup2(slave, 2) < 0)){
+			printk("start_debugger : dup2 failed, errno = %d\n",
+			       errno);
+			exit(1);
+		}
+		if(ioctl(0, TIOCSCTTY, 0) < 0){
+			printk("start_debugger : TIOCSCTTY failed, "
+			       "errno = %d\n", errno);
+			exit(1);
+		}
+		if(tcsetpgrp (1, os_getpid()) < 0){
+			printk("start_debugger : tcsetpgrp failed, "
+			       "errno = %d\n", errno);
+#ifdef notdef
+			exit(1);
+#endif
+		}
+		fd = make_tempfile("/tmp/gdb_init-XXXXXX", &tempname, 0);
+		if(fd < 0){
+			printk("start_debugger : make_tempfile failed,"
+			       "err = %d\n", -fd);
+			exit(1);
+		}
+		os_write_file(fd, gdb_init_string, sizeof(gdb_init_string) - 1);
+		if(startup){
+			if(stop){
+				os_write_file(fd, "b start_kernel\n",
+				      strlen("b start_kernel\n"));
+			}
+			os_write_file(fd, "c\n", strlen("c\n"));
+		}
+		if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0){
+			printk("start_debugger :  PTRACE_TRACEME failed, "
+			       "errno = %d\n", errno);
+			exit(1);
+		}
+		execlp("gdb", "gdb", "--command", tempname, prog, NULL);
+		printk("start_debugger : exec of gdb failed, errno = %d\n",
+		       errno);
+	}
+	if(child < 0){
+		printk("start_debugger : fork for gdb failed, errno = %d\n",
+		       errno);
+		return(-1);
+	}
+	*fd_out = slave;
+	return(child);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ptproxy/ptproxy.h

@@ -0,0 +1,61 @@
+/**********************************************************************
+ptproxy.h
+
+Copyright (C) 1999 Lars Brinkhoff.  See the file COPYING for licensing
+terms and conditions.
+**********************************************************************/
+
+#ifndef __PTPROXY_H
+#define __PTPROXY_H
+
+#include <sys/types.h>
+
+typedef struct debugger debugger_state;
+typedef struct debugee debugee_state;
+
+struct debugger
+{
+	pid_t pid;
+	int wait_options;
+	int *wait_status_ptr;
+	unsigned int waiting : 1;
+	unsigned int real_wait : 1;
+	unsigned int expecting_child : 1;
+	int (*handle_trace) (debugger_state *, pid_t);
+
+	debugee_state *debugee;
+};
+
+struct debugee
+{
+	pid_t pid;
+	int wait_status;
+	unsigned int died : 1;
+	unsigned int event : 1;
+	unsigned int stopped : 1;
+	unsigned int trace_singlestep : 1;
+	unsigned int trace_syscall : 1;
+	unsigned int traced : 1;
+	unsigned int zombie : 1;
+	unsigned int in_context : 1;
+};
+
+extern int debugger_syscall(debugger_state *debugger, pid_t pid);
+extern int debugger_normal_return (debugger_state *debugger, pid_t unused);
+
+extern long proxy_ptrace (struct debugger *, int, pid_t, long, long, pid_t,
+			  int *strace_out);
+extern void debugger_cancelled_return(debugger_state *debugger, int result);
+
+#endif
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ptproxy/ptrace.c

@@ -0,0 +1,238 @@
+/**********************************************************************
+ptrace.c
+
+Copyright (C) 1999 Lars Brinkhoff.  See the file COPYING for licensing
+terms and conditions.
+
+Jeff Dike (jdike@karaya.com) : Modified for integration into uml
+**********************************************************************/
+
+#include <errno.h>
+#include <unistd.h>
+#include <signal.h>
+#include <sys/types.h>
+#include <sys/time.h>
+#include <sys/wait.h>
+
+#include "ptproxy.h"
+#include "debug.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "ptrace_user.h"
+#include "tt.h"
+#include "os.h"
+
+long proxy_ptrace(struct debugger *debugger, int arg1, pid_t arg2,
+		  long arg3, long arg4, pid_t child, int *ret)
+{
+	sigset_t relay;
+	long result;
+	int status;
+
+	*ret = 0;
+	if(debugger->debugee->died) return(-ESRCH);
+
+	switch(arg1){
+	case PTRACE_ATTACH:
+		if(debugger->debugee->traced) return(-EPERM);
+
+		debugger->debugee->pid = arg2;
+		debugger->debugee->traced = 1;
+
+		if(is_valid_pid(arg2) && (arg2 != child)){
+			debugger->debugee->in_context = 0;
+			kill(arg2, SIGSTOP);
+			debugger->debugee->event = 1;
+			debugger->debugee->wait_status = W_STOPCODE(SIGSTOP);
+		}
+		else {
+			debugger->debugee->in_context = 1;
+			if(debugger->debugee->stopped) 
+				child_proxy(child, W_STOPCODE(SIGSTOP));
+			else kill(child, SIGSTOP);
+		}
+
+		return(0);
+
+	case PTRACE_DETACH:
+		if(!debugger->debugee->traced) return(-EPERM);
+		
+		debugger->debugee->traced = 0;
+		debugger->debugee->pid = 0;
+		if(!debugger->debugee->in_context)
+			kill(child, SIGCONT);
+
+		return(0);
+
+	case PTRACE_CONT:
+		if(!debugger->debugee->in_context) return(-EPERM);
+		*ret = PTRACE_CONT;
+		return(ptrace(PTRACE_CONT, child, arg3, arg4));
+
+#ifdef UM_HAVE_GETFPREGS
+	case PTRACE_GETFPREGS:
+	{
+		long regs[FP_FRAME_SIZE];
+		int i, result;
+
+		result = ptrace(PTRACE_GETFPREGS, child, 0, regs);
+		if(result == -1) return(-errno);
+		
+		for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++)
+			ptrace(PTRACE_POKEDATA, debugger->pid, arg4 + 4 * i,
+			       regs[i]);
+		return(result);
+	}
+#endif
+
+#ifdef UM_HAVE_GETFPXREGS
+	case PTRACE_GETFPXREGS:
+	{
+		long regs[FPX_FRAME_SIZE];
+		int i, result;
+
+		result = ptrace(PTRACE_GETFPXREGS, child, 0, regs);
+		if(result == -1) return(-errno);
+		
+		for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++)
+			ptrace(PTRACE_POKEDATA, debugger->pid, arg4 + 4 * i,
+			       regs[i]);
+		return(result);
+	}
+#endif
+
+#ifdef UM_HAVE_GETREGS
+	case PTRACE_GETREGS:
+	{
+		long regs[FRAME_SIZE];
+		int i, result;
+
+		result = ptrace(PTRACE_GETREGS, child, 0, regs);
+		if(result == -1) return(-errno);
+
+		for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++)
+			ptrace (PTRACE_POKEDATA, debugger->pid,
+				arg4 + 4 * i, regs[i]);
+		return(result);
+	}
+	break;
+#endif
+
+	case PTRACE_KILL:
+		result = ptrace(PTRACE_KILL, child, arg3, arg4);
+		if(result == -1) return(-errno);
+
+		return(result);
+
+	case PTRACE_PEEKDATA:
+	case PTRACE_PEEKTEXT:
+	case PTRACE_PEEKUSR:
+		/* The value being read out could be -1, so we have to 
+		 * check errno to see if there's an error, and zero it
+		 * beforehand so we're not faked out by an old error
+		 */
+
+		errno = 0;
+		result = ptrace(arg1, child, arg3, 0);
+		if((result == -1) && (errno != 0)) return(-errno);
+
+		result = ptrace(PTRACE_POKEDATA, debugger->pid, arg4, result);
+		if(result == -1) return(-errno);
+			
+		return(result);
+
+	case PTRACE_POKEDATA:
+	case PTRACE_POKETEXT:
+	case PTRACE_POKEUSR:
+		result = ptrace(arg1, child, arg3, arg4);
+		if(result == -1) return(-errno);
+
+		if(arg1 == PTRACE_POKEUSR) ptrace_pokeuser(arg3, arg4);
+		return(result);
+
+#ifdef UM_HAVE_SETFPREGS
+	case PTRACE_SETFPREGS:
+	{
+		long regs[FP_FRAME_SIZE];
+		int i;
+
+		for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++)
+			regs[i] = ptrace (PTRACE_PEEKDATA, debugger->pid,
+					  arg4 + 4 * i, 0);
+		result = ptrace(PTRACE_SETFPREGS, child, 0, regs);
+		if(result == -1) return(-errno);
+
+		return(result);
+	}
+#endif
+
+#ifdef UM_HAVE_SETFPXREGS
+	case PTRACE_SETFPXREGS:
+	{
+		long regs[FPX_FRAME_SIZE];
+		int i;
+
+		for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++)
+			regs[i] = ptrace (PTRACE_PEEKDATA, debugger->pid,
+					  arg4 + 4 * i, 0);
+		result = ptrace(PTRACE_SETFPXREGS, child, 0, regs);
+		if(result == -1) return(-errno);
+
+		return(result);
+	}
+#endif
+
+#ifdef UM_HAVE_SETREGS
+	case PTRACE_SETREGS:
+	{
+		long regs[FRAME_SIZE];
+		int i;
+
+		for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++)
+			regs[i] = ptrace(PTRACE_PEEKDATA, debugger->pid,
+					 arg4 + 4 * i, 0);
+		result = ptrace(PTRACE_SETREGS, child, 0, regs);
+		if(result == -1) return(-errno);
+
+		return(result);
+	}
+#endif
+
+	case PTRACE_SINGLESTEP:
+		if(!debugger->debugee->in_context) return(-EPERM);
+		sigemptyset(&relay);
+		sigaddset(&relay, SIGSEGV);
+		sigaddset(&relay, SIGILL);
+		sigaddset(&relay, SIGBUS);
+		result = ptrace(PTRACE_SINGLESTEP, child, arg3, arg4);
+		if(result == -1) return(-errno);
+		
+		status = wait_for_stop(child, SIGTRAP, PTRACE_SINGLESTEP,
+				       &relay);
+		child_proxy(child, status);
+		return(result);
+
+	case PTRACE_SYSCALL:
+		if(!debugger->debugee->in_context) return(-EPERM);
+		result = ptrace(PTRACE_SYSCALL, child, arg3, arg4);
+		if(result == -1) return(-errno);
+
+		*ret = PTRACE_SYSCALL;
+		return(result);
+
+	case PTRACE_TRACEME:
+	default:
+		return(-EINVAL);
+	}
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ptproxy/sysdep.c

@@ -0,0 +1,71 @@
+/**********************************************************************
+sysdep.c
+
+Copyright (C) 1999 Lars Brinkhoff.  See the file COPYING for licensing
+terms and conditions.
+**********************************************************************/
+
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <errno.h>
+#include <sys/types.h>
+#include <linux/unistd.h>
+#include "ptrace_user.h"
+#include "user_util.h"
+#include "user.h"
+#include "os.h"
+
+int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, long *arg4, 
+		long *arg5)
+{
+	*arg1 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG1_OFFSET, 0);
+	*arg2 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG2_OFFSET, 0);
+	*arg3 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG3_OFFSET, 0);
+	*arg4 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG4_OFFSET, 0);
+	*arg5 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG5_OFFSET, 0);
+	return(ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, 0));
+}
+
+void syscall_cancel(pid_t pid, int result)
+{
+	if((ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET,
+		   __NR_getpid) < 0) ||
+	   (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) ||
+	   (wait_for_stop(pid, SIGTRAP, PTRACE_SYSCALL, NULL) < 0) ||
+	   (ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, result) < 0) ||
+	   (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0))
+		printk("ptproxy: couldn't cancel syscall: errno = %d\n", 
+		       errno);
+}
+
+void syscall_set_result(pid_t pid, long result)
+{
+	ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, result);
+}
+
+void syscall_continue(pid_t pid)
+{
+	ptrace(PTRACE_SYSCALL, pid, 0, 0);
+}
+
+int syscall_pause(pid_t pid) 
+{
+	if(ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, __NR_pause) < 0){
+		printk("syscall_change - ptrace failed, errno = %d\n", errno);
+		return(-1);
+	}
+	return(0);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ptproxy/sysdep.h

@@ -0,0 +1,25 @@
+/**********************************************************************
+sysdep.h
+
+Copyright (C) 1999 Lars Brinkhoff.
+Copyright (C) 2001 Jeff Dike (jdike@karaya.com)
+See the file COPYING for licensing terms and conditions.
+**********************************************************************/
+
+extern int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, 
+		       long *arg4, long *arg5);
+extern void syscall_cancel (pid_t pid, long result);
+extern void syscall_set_result (pid_t pid, long result);
+extern void syscall_continue (pid_t pid);
+extern int syscall_pause(pid_t pid);
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ptproxy/wait.c

@@ -0,0 +1,86 @@
+/**********************************************************************
+wait.c
+
+Copyright (C) 1999 Lars Brinkhoff.  See the file COPYING for licensing
+terms and conditions.
+
+**********************************************************************/
+
+#include <errno.h>
+#include <signal.h>
+#include <sys/wait.h>
+
+#include "ptproxy.h"
+#include "sysdep.h"
+#include "wait.h"
+#include "user_util.h"
+#include "ptrace_user.h"
+#include "sysdep/ptrace.h"
+#include "sysdep/sigcontext.h"
+
+int proxy_wait_return(struct debugger *debugger, pid_t unused)
+{
+	debugger->waiting = 0;
+
+	if(debugger->debugee->died || (debugger->wait_options & __WCLONE)){
+		debugger_cancelled_return(debugger, -ECHILD);
+		return(0);
+	}
+
+	if(debugger->debugee->zombie && debugger->debugee->event)
+		debugger->debugee->died = 1;
+
+	if(debugger->debugee->event){
+		debugger->debugee->event = 0;
+		ptrace(PTRACE_POKEDATA, debugger->pid,
+		       debugger->wait_status_ptr, 
+		       debugger->debugee->wait_status);
+		/* if (wait4)
+		   ptrace (PTRACE_POKEDATA, pid, rusage_ptr, ...); */
+		debugger_cancelled_return(debugger, debugger->debugee->pid);
+		return(0);
+	}
+
+	/* pause will return -EINTR, which happens to be right for wait */
+	debugger_normal_return(debugger, -1);
+	return(0);
+}
+
+int parent_wait_return(struct debugger *debugger, pid_t unused)
+{
+	return(debugger_normal_return(debugger, -1));
+}
+
+int real_wait_return(struct debugger *debugger)
+{
+	unsigned long ip;
+	int pid;
+
+	pid = debugger->pid;
+
+	ip = ptrace(PTRACE_PEEKUSR, pid, PT_IP_OFFSET, 0);
+	IP_RESTART_SYSCALL(ip);
+
+	if(ptrace(PTRACE_POKEUSR, pid, PT_IP_OFFSET, ip) < 0)
+		tracer_panic("real_wait_return : Failed to restart system "
+			     "call, errno = %d\n", errno);
+
+	if((ptrace(PTRACE_SYSCALL, debugger->pid, 0, SIGCHLD) < 0) ||
+	   (ptrace(PTRACE_SYSCALL, debugger->pid, 0, 0) < 0) ||
+	   (ptrace(PTRACE_SYSCALL, debugger->pid, 0, 0) < 0) ||
+	   debugger_normal_return(debugger, -1))
+		tracer_panic("real_wait_return : gdb failed to wait, "
+			     "errno = %d\n", errno);
+	return(0);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/ptproxy/wait.h

@@ -0,0 +1,15 @@
+/**********************************************************************
+wait.h
+
+Copyright (C) 1999 Lars Brinkhoff.  See the file COPYING for licensing
+terms and conditions.
+**********************************************************************/
+
+#ifndef __PTPROXY_WAIT_H
+#define __PTPROXY_WAIT_H
+
+extern int proxy_wait_return(struct debugger *debugger, pid_t unused);
+extern int real_wait_return(struct debugger *debugger);
+extern int parent_wait_return(struct debugger *debugger, pid_t unused);
+
+#endif

arch/um/kernel/tt/syscall_kern.c

@@ -0,0 +1,46 @@
+/* 
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/types.h"
+#include "linux/utime.h"
+#include "linux/sys.h"
+#include "linux/ptrace.h"
+#include "asm/unistd.h"
+#include "asm/ptrace.h"
+#include "asm/uaccess.h"
+#include "asm/stat.h"
+#include "sysdep/syscalls.h"
+#include "sysdep/sigcontext.h"
+#include "kern_util.h"
+#include "syscall.h"
+
+void syscall_handler_tt(int sig, struct pt_regs *regs)
+{
+	void *sc;
+	long result;
+	int syscall;
+
+	sc = UPT_SC(&regs->regs);
+	SC_START_SYSCALL(sc);
+
+	syscall = UPT_SYSCALL_NR(&regs->regs);
+	syscall_trace(&regs->regs, 0);
+
+	current->thread.nsyscalls++;
+	nsyscalls++;
+
+	if((syscall >= NR_syscalls) || (syscall < 0))
+		result = -ENOSYS;
+	else result = EXECUTE_SYSCALL(syscall, regs);
+
+	/* regs->sc may have changed while the system call ran (there may
+	 * have been an interrupt or segfault), so it needs to be refreshed.
+	 */
+	UPT_SC(&regs->regs) = sc;
+
+	SC_SET_SYSCALL_RETURN(sc, result);
+
+	syscall_trace(&regs->regs, 1);
+}

arch/um/kernel/tt/syscall_user.c

@@ -0,0 +1,61 @@
+/* 
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <unistd.h>
+#include <signal.h>
+#include <errno.h>
+#include <asm/unistd.h>
+#include "sysdep/ptrace.h"
+#include "sigcontext.h"
+#include "ptrace_user.h"
+#include "task.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "syscall.h"
+#include "tt.h"
+
+void do_sigtrap(void *task)
+{
+	UPT_SYSCALL_NR(TASK_REGS(task)) = -1;
+}
+
+void do_syscall(void *task, int pid, int local_using_sysemu)
+{
+	unsigned long proc_regs[FRAME_SIZE];
+
+	if(ptrace_getregs(pid, proc_regs) < 0)
+		tracer_panic("Couldn't read registers");
+
+	UPT_SYSCALL_NR(TASK_REGS(task)) = PT_SYSCALL_NR(proc_regs);
+
+#ifdef UPT_ORIGGPR2
+        UPT_ORIGGPR2(TASK_REGS(task)) = REGS_ORIGGPR2(proc_regs);
+#endif
+
+	if(((unsigned long *) PT_IP(proc_regs) >= &_stext) &&
+	   ((unsigned long *) PT_IP(proc_regs) <= &_etext))
+		tracer_panic("I'm tracing myself and I can't get out");
+
+	/* advanced sysemu mode set syscall number to -1 automatically */
+	if (local_using_sysemu==2)
+		return;
+
+	/* syscall number -1 in sysemu skips syscall restarting in host */
+	if(ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET,
+		  local_using_sysemu ? -1 : __NR_getpid) < 0)
+		tracer_panic("do_syscall : Nullifying syscall failed, "
+			     "errno = %d", errno);
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/tlb.c

@@ -0,0 +1,121 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright 2003 PathScale, Inc.
+ * Licensed under the GPL
+ */
+
+#include "linux/stddef.h"
+#include "linux/kernel.h"
+#include "linux/sched.h"
+#include "linux/mm.h"
+#include "asm/page.h"
+#include "asm/pgtable.h"
+#include "asm/uaccess.h"
+#include "asm/tlbflush.h"
+#include "user_util.h"
+#include "mem_user.h"
+#include "os.h"
+#include "tlb.h"
+
+static int do_ops(union mm_context *mmu, struct host_vm_op *ops, int last,
+		    int finished, void **flush)
+{
+	struct host_vm_op *op;
+        int i, ret=0;
+
+        for(i = 0; i <= last && !ret; i++){
+		op = &ops[i];
+		switch(op->type){
+		case MMAP:
+                        ret = os_map_memory((void *) op->u.mmap.addr,
+                                            op->u.mmap.fd, op->u.mmap.offset,
+                                            op->u.mmap.len, op->u.mmap.r,
+                                            op->u.mmap.w, op->u.mmap.x);
+			break;
+		case MUNMAP:
+                        ret = os_unmap_memory((void *) op->u.munmap.addr,
+                                              op->u.munmap.len);
+			break;
+		case MPROTECT:
+                        ret = protect_memory(op->u.mprotect.addr,
+                                             op->u.munmap.len,
+                                             op->u.mprotect.r,
+                                             op->u.mprotect.w,
+                                             op->u.mprotect.x, 1);
+			protect_memory(op->u.mprotect.addr, op->u.munmap.len,
+				       op->u.mprotect.r, op->u.mprotect.w,
+				       op->u.mprotect.x, 1);
+			break;
+		default:
+			printk("Unknown op type %d in do_ops\n", op->type);
+			break;
+		}
+	}
+
+	return ret;
+}
+
+static void fix_range(struct mm_struct *mm, unsigned long start_addr, 
+		      unsigned long end_addr, int force)
+{
+        if((current->thread.mode.tt.extern_pid != -1) &&
+           (current->thread.mode.tt.extern_pid != os_getpid()))
+                panic("fix_range fixing wrong address space, current = 0x%p",
+                      current);
+
+        fix_range_common(mm, start_addr, end_addr, force, do_ops);
+}
+
+atomic_t vmchange_seq = ATOMIC_INIT(1);
+
+void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end)
+{
+        if(flush_tlb_kernel_range_common(start, end))
+                atomic_inc(&vmchange_seq);
+}
+
+void flush_tlb_kernel_vm_tt(void)
+{
+        flush_tlb_kernel_range(start_vm, end_vm);
+}
+
+void __flush_tlb_one_tt(unsigned long addr)
+{
+        flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
+}
+  
+void flush_tlb_range_tt(struct vm_area_struct *vma, unsigned long start, 
+		     unsigned long end)
+{
+	if(vma->vm_mm != current->mm) return;
+
+	/* Assumes that the range start ... end is entirely within
+	 * either process memory or kernel vm
+	 */
+	if((start >= start_vm) && (start < end_vm)){
+		if(flush_tlb_kernel_range_common(start, end))
+			atomic_inc(&vmchange_seq);
+	}
+	else fix_range(vma->vm_mm, start, end, 0);
+}
+
+void flush_tlb_mm_tt(struct mm_struct *mm)
+{
+	unsigned long seq;
+
+	if(mm != current->mm) return;
+
+	fix_range(mm, 0, STACK_TOP, 0);
+
+	seq = atomic_read(&vmchange_seq);
+	if(current->thread.mode.tt.vm_seq == seq)
+		return;
+	current->thread.mode.tt.vm_seq = seq;
+	flush_tlb_kernel_range_common(start_vm, end_vm);
+}
+
+void force_flush_all_tt(void)
+{
+	fix_range(current->mm, 0, STACK_TOP, 1);
+	flush_tlb_kernel_range_common(start_vm, end_vm);
+}

arch/um/kernel/tt/tracer.c

@@ -0,0 +1,462 @@
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <unistd.h>
+#include <signal.h>
+#include <errno.h>
+#include <sched.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/time.h>
+#include <sys/wait.h>
+#include "user.h"
+#include "sysdep/ptrace.h"
+#include "sigcontext.h"
+#include "sysdep/sigcontext.h"
+#include "os.h"
+#include "user_util.h"
+#include "mem_user.h"
+#include "process.h"
+#include "kern_util.h"
+#include "chan_user.h"
+#include "ptrace_user.h"
+#include "irq_user.h"
+#include "mode.h"
+#include "tt.h"
+
+static int tracer_winch[2];
+
+int is_tracer_winch(int pid, int fd, void *data)
+{
+	if(pid != os_getpgrp())
+		return(0);
+
+	register_winch_irq(tracer_winch[0], fd, -1, data);
+	return(1);
+}
+
+static void tracer_winch_handler(int sig)
+{
+	int n;
+	char c = 1;
+
+	n = os_write_file(tracer_winch[1], &c, sizeof(c));
+	if(n != sizeof(c))
+		printk("tracer_winch_handler - write failed, err = %d\n", -n);
+}
+
+/* Called only by the tracing thread during initialization */
+
+static void setup_tracer_winch(void)
+{
+	int err;
+
+	err = os_pipe(tracer_winch, 1, 1);
+	if(err < 0){
+		printk("setup_tracer_winch : os_pipe failed, err = %d\n", -err);
+		return;
+	}
+	signal(SIGWINCH, tracer_winch_handler);
+}
+
+void attach_process(int pid)
+{
+	if((ptrace(PTRACE_ATTACH, pid, 0, 0) < 0) ||
+	   (ptrace(PTRACE_CONT, pid, 0, 0) < 0))
+		tracer_panic("OP_FORK failed to attach pid");
+	wait_for_stop(pid, SIGSTOP, PTRACE_CONT, NULL);
+	if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0)
+		tracer_panic("OP_FORK: PTRACE_SETOPTIONS failed, errno = %d", errno);
+	if(ptrace(PTRACE_CONT, pid, 0, 0) < 0)
+		tracer_panic("OP_FORK failed to continue process");
+}
+
+void tracer_panic(char *format, ...)
+{
+	va_list ap;
+
+	va_start(ap, format);
+	vprintf(format, ap);
+	va_end(ap);
+	printf("\n");
+	while(1) pause();
+}
+
+static void tracer_segv(int sig, struct sigcontext sc)
+{
+        struct faultinfo fi;
+        GET_FAULTINFO_FROM_SC(fi, &sc);
+	printf("Tracing thread segfault at address 0x%lx, ip 0x%lx\n",
+               FAULT_ADDRESS(fi), SC_IP(&sc));
+	while(1)
+		pause();
+}
+
+/* Changed early in boot, and then only read */
+int debug = 0;
+int debug_stop = 1;
+int debug_parent = 0;
+int honeypot = 0;
+
+static int signal_tramp(void *arg)
+{
+	int (*proc)(void *);
+
+	if(honeypot && munmap((void *) (host_task_size - 0x10000000),
+			      0x10000000)) 
+		panic("Unmapping stack failed");
+	if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0)
+		panic("ptrace PTRACE_TRACEME failed");
+	os_stop_process(os_getpid());
+	change_sig(SIGWINCH, 0);
+	signal(SIGUSR1, SIG_IGN);
+	change_sig(SIGCHLD, 0);
+	signal(SIGSEGV, (__sighandler_t) sig_handler);
+	set_cmdline("(idle thread)");
+	set_init_pid(os_getpid());
+	init_irq_signals(0);
+	proc = arg;
+	return((*proc)(NULL));
+}
+
+static void sleeping_process_signal(int pid, int sig)
+{
+	switch(sig){
+	/* These two result from UML being ^Z-ed and bg-ed.  PTRACE_CONT is
+	 * right because the process must be in the kernel already.
+	 */
+	case SIGCONT:
+	case SIGTSTP:
+		if(ptrace(PTRACE_CONT, pid, 0, sig) < 0)
+			tracer_panic("sleeping_process_signal : Failed to "
+				     "continue pid %d, signal = %d, "
+				     "errno = %d\n", pid, sig, errno);
+		break;
+
+	/* This happens when the debugger (e.g. strace) is doing system call 
+	 * tracing on the kernel.  During a context switch, the current task
+	 * will be set to the incoming process and the outgoing process will
+	 * hop into write and then read.  Since it's not the current process
+	 * any more, the trace of those will land here.  So, we need to just 
+	 * PTRACE_SYSCALL it.
+	 */
+	case (SIGTRAP + 0x80):
+		if(ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0)
+			tracer_panic("sleeping_process_signal : Failed to "
+				     "PTRACE_SYSCALL pid %d, errno = %d\n",
+				     pid, errno);
+		break;
+	case SIGSTOP:
+		break;
+	default:
+		tracer_panic("sleeping process %d got unexpected "
+			     "signal : %d\n", pid, sig);
+		break;
+	}
+}
+
+/* Accessed only by the tracing thread */
+int debugger_pid = -1;
+int debugger_parent = -1;
+int debugger_fd = -1;
+int gdb_pid = -1;
+
+struct {
+	int pid;
+	int signal;
+	unsigned long addr;
+	struct timeval time;
+} signal_record[1024][32];
+
+int signal_index[32];
+int nsignals = 0;
+int debug_trace = 0;
+
+extern void signal_usr1(int sig);
+
+int tracing_pid = -1;
+
+int tracer(int (*init_proc)(void *), void *sp)
+{
+	void *task = NULL;
+	int status, pid = 0, sig = 0, cont_type, tracing = 0, op = 0;
+	int proc_id = 0, n, err, old_tracing = 0, strace = 0;
+	int local_using_sysemu = 0;
+
+	signal(SIGPIPE, SIG_IGN);
+	setup_tracer_winch();
+	tracing_pid = os_getpid();
+	printf("tracing thread pid = %d\n", tracing_pid);
+
+	pid = clone(signal_tramp, sp, CLONE_FILES | SIGCHLD, init_proc);
+	CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED));
+	if(n < 0){
+		printf("waitpid on idle thread failed, errno = %d\n", errno);
+		exit(1);
+	}
+	if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) {
+		printf("Failed to PTRACE_SETOPTIONS for idle thread, errno = %d\n", errno);
+		exit(1);
+	}
+	if((ptrace(PTRACE_CONT, pid, 0, 0) < 0)){
+		printf("Failed to continue idle thread, errno = %d\n", errno);
+		exit(1);
+	}
+
+	signal(SIGSEGV, (sighandler_t) tracer_segv);
+	signal(SIGUSR1, signal_usr1);
+	if(debug_trace){
+		printf("Tracing thread pausing to be attached\n");
+		stop();
+	}
+	if(debug){
+		if(gdb_pid != -1) 
+			debugger_pid = attach_debugger(pid, gdb_pid, 1);
+		else debugger_pid = init_ptrace_proxy(pid, 1, debug_stop);
+		if(debug_parent){
+			debugger_parent = os_process_parent(debugger_pid);
+			init_parent_proxy(debugger_parent);
+			err = attach(debugger_parent);
+			if(err){
+				printf("Failed to attach debugger parent %d, "
+				       "errno = %d\n", debugger_parent, -err);
+				debugger_parent = -1;
+			}
+			else {
+				if(ptrace(PTRACE_SYSCALL, debugger_parent, 
+					  0, 0) < 0){
+					printf("Failed to continue debugger "
+					       "parent, errno = %d\n", errno);
+					debugger_parent = -1;
+				}
+			}
+		}
+	}
+	set_cmdline("(tracing thread)");
+	while(1){
+		CATCH_EINTR(pid = waitpid(-1, &status, WUNTRACED));
+		if(pid <= 0){
+			if(errno != ECHILD){
+				printf("wait failed - errno = %d\n", errno);
+			}
+			continue;
+		}
+		if(pid == debugger_pid){
+			int cont = 0;
+
+			if(WIFEXITED(status) || WIFSIGNALED(status))
+				debugger_pid = -1;
+			/* XXX Figure out how to deal with gdb and SMP */
+			else cont = debugger_signal(status, cpu_tasks[0].pid);
+			if(cont == PTRACE_SYSCALL) strace = 1;
+			continue;
+		}
+		else if(pid == debugger_parent){
+			debugger_parent_signal(status, pid);
+			continue;
+		}
+		nsignals++;
+		if(WIFEXITED(status)) ;
+#ifdef notdef
+		{
+			printf("Child %d exited with status %d\n", pid, 
+			       WEXITSTATUS(status));
+		}
+#endif
+		else if(WIFSIGNALED(status)){
+			sig = WTERMSIG(status);
+			if(sig != 9){
+				printf("Child %d exited with signal %d\n", pid,
+				       sig);
+			}
+		}
+		else if(WIFSTOPPED(status)){
+			proc_id = pid_to_processor_id(pid);
+			sig = WSTOPSIG(status);
+			if(proc_id == -1){
+				sleeping_process_signal(pid, sig);
+				continue;
+			}
+
+			task = cpu_tasks[proc_id].task;
+			tracing = is_tracing(task);
+			old_tracing = tracing;
+
+			/* Assume: no syscall, when coming from user */
+			if ( tracing )
+				do_sigtrap(task);
+
+			switch(sig){
+			case SIGUSR1:
+				sig = 0;
+				op = do_proc_op(task, proc_id);
+				switch(op){
+				/*
+				 * This is called when entering user mode; after
+				 * this, we start intercepting syscalls.
+				 *
+				 * In fact, a process is started in kernel mode,
+				 * so with is_tracing() == 0 (and that is reset
+				 * when executing syscalls, since UML kernel has
+				 * the right to do syscalls);
+				 */
+				case OP_TRACE_ON:
+					arch_leave_kernel(task, pid);
+					tracing = 1;
+					break;
+				case OP_REBOOT:
+				case OP_HALT:
+					unmap_physmem();
+					kmalloc_ok = 0;
+					os_kill_ptraced_process(pid, 0);
+					/* Now let's reap remaining zombies */
+					errno = 0;
+					do {
+						waitpid(-1, &status,
+							WUNTRACED);
+					} while (errno != ECHILD);
+					return(op == OP_REBOOT);
+				case OP_NONE:
+					printf("Detaching pid %d\n", pid);
+					detach(pid, SIGSTOP);
+					continue;
+				default:
+					break;
+				}
+				/* OP_EXEC switches host processes on us,
+				 * we want to continue the new one.
+				 */
+				pid = cpu_tasks[proc_id].pid;
+				break;
+			case (SIGTRAP + 0x80):
+				if(!tracing && (debugger_pid != -1)){
+					child_signal(pid, status & 0x7fff);
+					continue;
+				}
+				tracing = 0;
+				/* local_using_sysemu has been already set
+				 * below, since if we are here, is_tracing() on
+				 * the traced task was 1, i.e. the process had
+				 * already run through one iteration of the
+				 * loop which executed a OP_TRACE_ON request.*/
+				do_syscall(task, pid, local_using_sysemu);
+				sig = SIGUSR2;
+				break;
+			case SIGTRAP:
+				if(!tracing && (debugger_pid != -1)){
+					child_signal(pid, status);
+					continue;
+				}
+				tracing = 0;
+				break;
+			case SIGPROF:
+				if(tracing) sig = 0;
+				break;
+			case SIGCHLD:
+			case SIGHUP:
+				sig = 0;
+				break;
+			case SIGSEGV:
+			case SIGIO:
+			case SIGALRM:
+			case SIGVTALRM:
+			case SIGFPE:
+			case SIGBUS:
+			case SIGILL:
+			case SIGWINCH:
+
+			default:
+				tracing = 0;
+				break;
+			}
+			set_tracing(task, tracing);
+
+			if(!tracing && old_tracing)
+				arch_enter_kernel(task, pid);
+
+			if(!tracing && (debugger_pid != -1) && (sig != 0) &&
+				(sig != SIGALRM) && (sig != SIGVTALRM) &&
+				(sig != SIGSEGV) && (sig != SIGTRAP) &&
+				(sig != SIGUSR2) && (sig != SIGIO) &&
+				(sig != SIGFPE)){
+				child_signal(pid, status);
+				continue;
+			}
+
+			local_using_sysemu = get_using_sysemu();
+
+			if(tracing)
+				cont_type = SELECT_PTRACE_OPERATION(local_using_sysemu,
+				                                    singlestepping(task));
+			else if((debugger_pid != -1) && strace)
+				cont_type = PTRACE_SYSCALL;
+			else
+				cont_type = PTRACE_CONT;
+
+			if(ptrace(cont_type, pid, 0, sig) != 0){
+				tracer_panic("ptrace failed to continue "
+					     "process - errno = %d\n", 
+					     errno);
+			}
+		}
+	}
+	return(0);
+}
+
+static int __init uml_debug_setup(char *line, int *add)
+{
+	char *next;
+
+	debug = 1;
+	*add = 0;
+	if(*line != '=') return(0);
+	line++;
+
+	while(line != NULL){
+		next = strchr(line, ',');
+		if(next) *next++ = '\0';
+		
+		if(!strcmp(line, "go"))	debug_stop = 0;
+		else if(!strcmp(line, "parent")) debug_parent = 1;
+		else printf("Unknown debug option : '%s'\n", line);
+
+		line = next;
+	}
+	return(0);
+}
+
+__uml_setup("debug", uml_debug_setup,
+"debug\n"
+"    Starts up the kernel under the control of gdb. See the \n"
+"    kernel debugging tutorial and the debugging session pages\n"
+"    at http://user-mode-linux.sourceforge.net/ for more information.\n\n"
+);
+
+static int __init uml_debugtrace_setup(char *line, int *add)
+{
+	debug_trace = 1;
+	return 0;
+}
+__uml_setup("debugtrace", uml_debugtrace_setup,
+"debugtrace\n"
+"    Causes the tracing thread to pause until it is attached by a\n"
+"    debugger and continued.  This is mostly for debugging crashes\n"
+"    early during boot, and should be pretty much obsoleted by\n"
+"    the debug switch.\n\n"
+);
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/trap_user.c

@@ -0,0 +1,71 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <stdlib.h>
+#include <errno.h>
+#include <signal.h>
+#include "sysdep/ptrace.h"
+#include "sysdep/sigcontext.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "task.h"
+#include "tt.h"
+#include "os.h"
+
+void sig_handler_common_tt(int sig, void *sc_ptr)
+{
+	struct sigcontext *sc = sc_ptr;
+	struct tt_regs save_regs, *r;
+	int save_errno = errno, is_user = 0;
+	void (*handler)(int, union uml_pt_regs *);
+
+	/* This is done because to allow SIGSEGV to be delivered inside a SEGV
+	 * handler.  This can happen in copy_user, and if SEGV is disabled,
+	 * the process will die.
+	 */
+	if(sig == SIGSEGV)
+		change_sig(SIGSEGV, 1);
+
+	r = &TASK_REGS(get_current())->tt;
+        if ( sig == SIGFPE || sig == SIGSEGV ||
+             sig == SIGBUS || sig == SIGILL ||
+             sig == SIGTRAP ) {
+                GET_FAULTINFO_FROM_SC(r->faultinfo, sc);
+        }
+	save_regs = *r;
+	if (sc)
+		is_user = user_context(SC_SP(sc));
+	r->sc = sc;
+	if(sig != SIGUSR2) 
+		r->syscall = -1;
+
+	handler = sig_info[sig];
+
+	/* unblock SIGALRM, SIGVTALRM, SIGIO if sig isn't IRQ signal */
+	if (sig != SIGIO && sig != SIGWINCH &&
+	    sig != SIGVTALRM && sig != SIGALRM)
+		unblock_signals();
+
+	handler(sig, (union uml_pt_regs *) r);
+
+	if(is_user){
+		interrupt_end();
+		block_signals();
+		set_user_mode(NULL);
+	}
+	*r = save_regs;
+	errno = save_errno;
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/uaccess.c

@@ -0,0 +1,73 @@
+/*
+ * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "asm/uaccess.h"
+
+int copy_from_user_tt(void *to, const void __user *from, int n)
+{
+	if(!access_ok(VERIFY_READ, from, n))
+		return(n);
+
+	return(__do_copy_from_user(to, from, n, &current->thread.fault_addr,
+				   &current->thread.fault_catcher));
+}
+
+int copy_to_user_tt(void __user *to, const void *from, int n)
+{
+	if(!access_ok(VERIFY_WRITE, to, n))
+		return(n);
+
+	return(__do_copy_to_user(to, from, n, &current->thread.fault_addr,
+				 &current->thread.fault_catcher));
+}
+
+int strncpy_from_user_tt(char *dst, const char __user *src, int count)
+{
+	int n;
+
+	if(!access_ok(VERIFY_READ, src, 1))
+		return(-EFAULT);
+
+	n = __do_strncpy_from_user(dst, src, count,
+				   &current->thread.fault_addr,
+				   &current->thread.fault_catcher);
+	if(n < 0) return(-EFAULT);
+	return(n);
+}
+
+int __clear_user_tt(void __user *mem, int len)
+{
+	return(__do_clear_user(mem, len,
+			       &current->thread.fault_addr,
+			       &current->thread.fault_catcher));
+}
+
+int clear_user_tt(void __user *mem, int len)
+{
+	if(!access_ok(VERIFY_WRITE, mem, len))
+		return(len);
+
+	return(__do_clear_user(mem, len, &current->thread.fault_addr,
+			       &current->thread.fault_catcher));
+}
+
+int strnlen_user_tt(const void __user *str, int len)
+{
+	return(__do_strnlen_user(str, len,
+				 &current->thread.fault_addr,
+				 &current->thread.fault_catcher));
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/tt/uaccess_user.c

@@ -0,0 +1,106 @@
+/* 
+ * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk)
+ * Copyright (C) 2001 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include <string.h>
+#include "user_util.h"
+#include "uml_uaccess.h"
+#include "task.h"
+#include "kern_util.h"
+#include "os.h"
+#include "longjmp.h"
+
+int __do_copy_from_user(void *to, const void *from, int n,
+			void **fault_addr, void **fault_catcher)
+{
+	struct tt_regs save = TASK_REGS(get_current())->tt;
+	unsigned long fault;
+	int faulted;
+
+	fault = __do_user_copy(to, from, n, fault_addr, fault_catcher,
+			       __do_copy, &faulted);
+	TASK_REGS(get_current())->tt = save;
+
+	if(!faulted)
+		return 0;
+	else if (fault)
+		return n - (fault - (unsigned long) from);
+	else
+		/* In case of a general protection fault, we don't have the
+		 * fault address, so NULL is used instead. Pretend we didn't
+		 * copy anything. */
+		return n;
+}
+
+static void __do_strncpy(void *dst, const void *src, int count)
+{
+	strncpy(dst, src, count);
+}	
+
+int __do_strncpy_from_user(char *dst, const char *src, unsigned long count,
+			   void **fault_addr, void **fault_catcher)
+{
+	struct tt_regs save = TASK_REGS(get_current())->tt;
+	unsigned long fault;
+	int faulted;
+
+	fault = __do_user_copy(dst, src, count, fault_addr, fault_catcher,
+			       __do_strncpy, &faulted);
+	TASK_REGS(get_current())->tt = save;
+
+	if(!faulted) return(strlen(dst));
+	else return(-1);
+}
+
+static void __do_clear(void *to, const void *from, int n)
+{
+	memset(to, 0, n);
+}	
+
+int __do_clear_user(void *mem, unsigned long len,
+		    void **fault_addr, void **fault_catcher)
+{
+	struct tt_regs save = TASK_REGS(get_current())->tt;
+	unsigned long fault;
+	int faulted;
+
+	fault = __do_user_copy(mem, NULL, len, fault_addr, fault_catcher,
+			       __do_clear, &faulted);
+	TASK_REGS(get_current())->tt = save;
+
+	if(!faulted) return(0);
+	else return(len - (fault - (unsigned long) mem));
+}
+
+int __do_strnlen_user(const char *str, unsigned long n,
+		      void **fault_addr, void **fault_catcher)
+{
+	struct tt_regs save = TASK_REGS(get_current())->tt;
+	int ret;
+	unsigned long *faddrp = (unsigned long *)fault_addr;
+	jmp_buf jbuf;
+
+	*fault_catcher = &jbuf;
+	if(UML_SETJMP(&jbuf) == 0)
+		ret = strlen(str) + 1;
+	else ret = *faddrp - (unsigned long) str;
+
+	*fault_addr = NULL;
+	*fault_catcher = NULL;
+
+	TASK_REGS(get_current())->tt = save;
+	return ret;
+}
+
+/*
+ * Overrides for Emacs so that we follow Linus's tabbing style.
+ * Emacs will notice this stuff at the end of the file and automatically
+ * adjust the settings for this buffer only.  This must remain at the end
+ * of the file.
+ * ---------------------------------------------------------------------------
+ * Local variables:
+ * c-file-style: "linux"
+ * End:
+ */

arch/um/kernel/uaccess.c

@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk)
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+/* These are here rather than tt/uaccess.c because skas mode needs them in
+ * order to do SIGBUS recovery when a tmpfs mount runs out of room.
+ */
+
+#include <linux/string.h>
+#include "os.h"
+
+void __do_copy(void *to, const void *from, int n)
+{
+	memcpy(to, from, n);
+}
+
+
+int __do_copy_to_user(void *to, const void *from, int n,
+		      void **fault_addr, void **fault_catcher)
+{
+	unsigned long fault;
+	int faulted;
+
+	fault = __do_user_copy(to, from, n, fault_addr, fault_catcher,
+			       __do_copy, &faulted);
+	if(!faulted) return(0);
+	else return(n - (fault - (unsigned long) to));
+}

arch/um/kernel/um_arch.c

@@ -0,0 +1,509 @@
+/*
+ * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/sched.h"
+#include "linux/notifier.h"
+#include "linux/mm.h"
+#include "linux/types.h"
+#include "linux/tty.h"
+#include "linux/init.h"
+#include "linux/bootmem.h"
+#include "linux/spinlock.h"
+#include "linux/utsname.h"
+#include "linux/sysrq.h"
+#include "linux/seq_file.h"
+#include "linux/delay.h"
+#include "linux/module.h"
+#include "asm/page.h"
+#include "asm/pgtable.h"
+#include "asm/ptrace.h"
+#include "asm/elf.h"
+#include "asm/user.h"
+#include "asm/setup.h"
+#include "ubd_user.h"
+#include "asm/current.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "mem_user.h"
+#include "mem.h"
+#include "initrd.h"
+#include "init.h"
+#include "os.h"
+#include "choose-mode.h"
+#include "mode_kern.h"
+#include "mode.h"
+#ifdef UML_CONFIG_MODE_SKAS
+#include "skas.h"
+#endif
+
+#define DEFAULT_COMMAND_LINE "root=98:0"
+
+/* Changed in linux_main and setup_arch, which run before SMP is started */
+static char __initdata command_line[COMMAND_LINE_SIZE] = { 0 };
+
+static void __init add_arg(char *arg)
+{
+	if (strlen(command_line) + strlen(arg) + 1 > COMMAND_LINE_SIZE) {
+		printf("add_arg: Too many command line arguments!\n");
+		exit(1);
+	}
+	if(strlen(command_line) > 0)
+		strcat(command_line, " ");
+	strcat(command_line, arg);
+}
+
+struct cpuinfo_um boot_cpu_data = { 
+	.loops_per_jiffy	= 0,
+	.ipi_pipe		= { -1, -1 }
+};
+
+unsigned long thread_saved_pc(struct task_struct *task)
+{
+	return(os_process_pc(CHOOSE_MODE_PROC(thread_pid_tt, thread_pid_skas,
+					      task)));
+}
+
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+	int index = 0;
+
+#ifdef CONFIG_SMP
+	index = (struct cpuinfo_um *) v - cpu_data;
+	if (!cpu_online(index))
+		return 0;
+#endif
+
+	seq_printf(m, "processor\t: %d\n", index);
+	seq_printf(m, "vendor_id\t: User Mode Linux\n");
+	seq_printf(m, "model name\t: UML\n");
+	seq_printf(m, "mode\t\t: %s\n", CHOOSE_MODE("tt", "skas"));
+	seq_printf(m, "host\t\t: %s\n", host_info);
+	seq_printf(m, "bogomips\t: %lu.%02lu\n\n",
+		   loops_per_jiffy/(500000/HZ),
+		   (loops_per_jiffy/(5000/HZ)) % 100);
+
+	return(0);
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	return *pos < NR_CPUS ? cpu_data + *pos : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return c_start(m, pos);
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+const struct seq_operations cpuinfo_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= show_cpuinfo,
+};
+
+/* Set in linux_main */
+unsigned long host_task_size;
+unsigned long task_size;
+
+unsigned long uml_start;
+
+/* Set in early boot */
+unsigned long uml_physmem;
+unsigned long uml_reserved;
+unsigned long start_vm;
+unsigned long end_vm;
+int ncpus = 1;
+
+#ifdef CONFIG_CMDLINE_ON_HOST
+/* Pointer set in linux_main, the array itself is private to each thread,
+ * and changed at address space creation time so this poses no concurrency
+ * problems.
+ */
+static char *argv1_begin = NULL;
+static char *argv1_end = NULL;
+#endif
+
+/* Set in early boot */
+static int have_root __initdata = 0;
+long long physmem_size = 32 * 1024 * 1024;
+
+void set_cmdline(char *cmd)
+{
+#ifdef CONFIG_CMDLINE_ON_HOST
+	char *umid, *ptr;
+
+	if(CHOOSE_MODE(honeypot, 0)) return;
+
+	umid = get_umid();
+	if(*umid != '\0'){
+		snprintf(argv1_begin, 
+			 (argv1_end - argv1_begin) * sizeof(*ptr), 
+			 "(%s) ", umid);
+		ptr = &argv1_begin[strlen(argv1_begin)];
+	}
+	else ptr = argv1_begin;
+
+	snprintf(ptr, (argv1_end - ptr) * sizeof(*ptr), "[%s]", cmd);
+	memset(argv1_begin + strlen(argv1_begin), '\0', 
+	       argv1_end - argv1_begin - strlen(argv1_begin));
+#endif
+}
+
+static char *usage_string = 
+"User Mode Linux v%s\n"
+"	available at http://user-mode-linux.sourceforge.net/\n\n";
+
+static int __init uml_version_setup(char *line, int *add)
+{
+	printf("%s\n", init_utsname()->release);
+	exit(0);
+
+	return 0;
+}
+
+__uml_setup("--version", uml_version_setup,
+"--version\n"
+"    Prints the version number of the kernel.\n\n"
+);
+
+static int __init uml_root_setup(char *line, int *add)
+{
+	have_root = 1;
+	return 0;
+}
+
+__uml_setup("root=", uml_root_setup,
+"root=<file containing the root fs>\n"
+"    This is actually used by the generic kernel in exactly the same\n"
+"    way as in any other kernel. If you configure a number of block\n"
+"    devices and want to boot off something other than ubd0, you \n"
+"    would use something like:\n"
+"        root=/dev/ubd5\n\n"
+);
+
+#ifndef CONFIG_MODE_TT
+
+static int __init no_skas_debug_setup(char *line, int *add)
+{
+	printf("'debug' is not necessary to gdb UML in skas mode - run \n");
+	printf("'gdb linux' and disable CONFIG_CMDLINE_ON_HOST if gdb \n");
+	printf("doesn't work as expected\n");
+
+	return 0;
+}
+
+__uml_setup("debug", no_skas_debug_setup,
+"debug\n"
+"    this flag is not needed to run gdb on UML in skas mode\n\n"
+);
+
+#endif
+
+#ifdef CONFIG_SMP
+static int __init uml_ncpus_setup(char *line, int *add)
+{
+       if (!sscanf(line, "%d", &ncpus)) {
+               printf("Couldn't parse [%s]\n", line);
+               return -1;
+       }
+
+       return 0;
+}
+
+__uml_setup("ncpus=", uml_ncpus_setup,
+"ncpus=<# of desired CPUs>\n"
+"    This tells an SMP kernel how many virtual processors to start.\n\n" 
+);
+#endif
+
+static int force_tt = 0;
+
+#if defined(CONFIG_MODE_TT) && defined(CONFIG_MODE_SKAS)
+#define DEFAULT_TT 0
+
+static int __init mode_tt_setup(char *line, int *add)
+{
+	force_tt = 1;
+	return(0);
+}
+
+#else
+#ifdef CONFIG_MODE_SKAS
+
+#define DEFAULT_TT 0
+
+static int __init mode_tt_setup(char *line, int *add)
+{
+	printf("CONFIG_MODE_TT disabled - 'mode=tt' ignored\n");
+	return(0);
+}
+
+#else
+#ifdef CONFIG_MODE_TT
+
+#define DEFAULT_TT 1
+
+static int __init mode_tt_setup(char *line, int *add)
+{
+	printf("CONFIG_MODE_SKAS disabled - 'mode=tt' redundant\n");
+	return(0);
+}
+
+#endif
+#endif
+#endif
+
+__uml_setup("mode=tt", mode_tt_setup,
+"mode=tt\n"
+"    When both CONFIG_MODE_TT and CONFIG_MODE_SKAS are enabled, this option\n"
+"    forces UML to run in tt (tracing thread) mode.  It is not the default\n"
+"    because it's slower and less secure than skas mode.\n\n"
+);
+
+int mode_tt = DEFAULT_TT;
+
+static int __init Usage(char *line, int *add)
+{
+ 	const char **p;
+
+	printf(usage_string, init_utsname()->release);
+ 	p = &__uml_help_start;
+ 	while (p < &__uml_help_end) {
+ 		printf("%s", *p);
+ 		p++;
+ 	}
+	exit(0);
+
+	return 0;
+}
+
+__uml_setup("--help", Usage,
+"--help\n"
+"    Prints this message.\n\n"
+);
+
+static int __init uml_checksetup(char *line, int *add)
+{
+	struct uml_param *p;
+
+	p = &__uml_setup_start;
+	while(p < &__uml_setup_end) {
+		int n;
+
+		n = strlen(p->str);
+		if(!strncmp(line, p->str, n)){
+			if (p->setup_func(line + n, add)) return 1;
+		}
+		p++;
+	}
+	return 0;
+}
+
+static void __init uml_postsetup(void)
+{
+	initcall_t *p;
+
+	p = &__uml_postsetup_start;
+	while(p < &__uml_postsetup_end){
+		(*p)();
+		p++;
+	}
+	return;
+}
+
+/* Set during early boot */
+unsigned long brk_start;
+unsigned long end_iomem;
+EXPORT_SYMBOL(end_iomem);
+
+#define MIN_VMALLOC (32 * 1024 * 1024)
+
+extern char __binary_start;
+
+int __init linux_main(int argc, char **argv)
+{
+	unsigned long avail, diff;
+	unsigned long virtmem_size, max_physmem;
+	unsigned int i, add;
+	char * mode;
+
+	for (i = 1; i < argc; i++){
+		if((i == 1) && (argv[i][0] == ' ')) continue;
+		add = 1;
+		uml_checksetup(argv[i], &add);
+		if (add)
+			add_arg(argv[i]);
+	}
+	if(have_root == 0)
+		add_arg(DEFAULT_COMMAND_LINE);
+
+	os_early_checks();
+	if (force_tt)
+		clear_can_do_skas();
+	mode_tt = force_tt ? 1 : !can_do_skas();
+#ifndef CONFIG_MODE_TT
+	if (mode_tt) {
+		/*Since CONFIG_MODE_TT is #undef'ed, force_tt cannot be 1. So,
+		 * can_do_skas() returned 0, and the message is correct. */
+		printf("Support for TT mode is disabled, and no SKAS support is present on the host.\n");
+		exit(1);
+	}
+#endif
+
+#ifndef CONFIG_MODE_SKAS
+	mode = "TT";
+#else
+	/* Show to the user the result of selection */
+	if (mode_tt)
+		mode = "TT";
+	else if (proc_mm && ptrace_faultinfo)
+		mode = "SKAS3";
+	else
+		mode = "SKAS0";
+#endif
+
+	printf("UML running in %s mode\n", mode);
+
+	uml_start = (unsigned long) &__binary_start;
+	host_task_size = CHOOSE_MODE_PROC(set_task_sizes_tt,
+					  set_task_sizes_skas, &task_size);
+
+	/*
+ 	 * Setting up handlers to 'sig_info' struct
+ 	 */
+	os_fill_handlinfo(handlinfo_kern);
+
+	brk_start = (unsigned long) sbrk(0);
+	CHOOSE_MODE_PROC(before_mem_tt, before_mem_skas, brk_start);
+	/* Increase physical memory size for exec-shield users
+	so they actually get what they asked for. This should
+	add zero for non-exec shield users */
+
+	diff = UML_ROUND_UP(brk_start) - UML_ROUND_UP(&_end);
+	if(diff > 1024 * 1024){
+		printf("Adding %ld bytes to physical memory to account for "
+		       "exec-shield gap\n", diff);
+		physmem_size += UML_ROUND_UP(brk_start) - UML_ROUND_UP(&_end);
+	}
+
+	uml_physmem = uml_start & PAGE_MASK;
+
+	/* Reserve up to 4M after the current brk */
+	uml_reserved = ROUND_4M(brk_start) + (1 << 22);
+
+	setup_machinename(init_utsname()->machine);
+
+#ifdef CONFIG_CMDLINE_ON_HOST
+	argv1_begin = argv[1];
+	argv1_end = &argv[1][strlen(argv[1])];
+#endif
+  
+	highmem = 0;
+	iomem_size = (iomem_size + PAGE_SIZE - 1) & PAGE_MASK;
+	max_physmem = get_kmem_end() - uml_physmem - iomem_size - MIN_VMALLOC;
+
+	/* Zones have to begin on a 1 << MAX_ORDER page boundary,
+	 * so this makes sure that's true for highmem
+	 */
+	max_physmem &= ~((1 << (PAGE_SHIFT + MAX_ORDER)) - 1);
+	if(physmem_size + iomem_size > max_physmem){
+		highmem = physmem_size + iomem_size - max_physmem;
+		physmem_size -= highmem;
+#ifndef CONFIG_HIGHMEM
+		highmem = 0;
+		printf("CONFIG_HIGHMEM not enabled - physical memory shrunk "
+		       "to %Lu bytes\n", physmem_size);
+#endif
+	}
+
+	high_physmem = uml_physmem + physmem_size;
+	end_iomem = high_physmem + iomem_size;
+	high_memory = (void *) end_iomem;
+
+	start_vm = VMALLOC_START;
+
+	setup_physmem(uml_physmem, uml_reserved, physmem_size, highmem);
+	if(init_maps(physmem_size, iomem_size, highmem)){
+		printf("Failed to allocate mem_map for %Lu bytes of physical "
+		       "memory and %Lu bytes of highmem\n", physmem_size,
+		       highmem);
+		exit(1);
+	}
+
+	virtmem_size = physmem_size;
+	avail = get_kmem_end() - start_vm;
+	if(physmem_size > avail) virtmem_size = avail;
+	end_vm = start_vm + virtmem_size;
+
+	if(virtmem_size < physmem_size)
+		printf("Kernel virtual memory size shrunk to %lu bytes\n",
+		       virtmem_size);
+
+  	uml_postsetup();
+
+	task_protections((unsigned long) &init_thread_info);
+	os_flush_stdout();
+
+	return(CHOOSE_MODE(start_uml_tt(), start_uml_skas()));
+}
+
+extern int uml_exitcode;
+
+static int panic_exit(struct notifier_block *self, unsigned long unused1,
+		      void *unused2)
+{
+	bust_spinlocks(1);
+	show_regs(&(current->thread.regs));
+	bust_spinlocks(0);
+	uml_exitcode = 1;
+	machine_halt();
+	return(0);
+}
+
+static struct notifier_block panic_exit_notifier = {
+	.notifier_call 		= panic_exit,
+	.next 			= NULL,
+	.priority 		= 0
+};
+
+void __init setup_arch(char **cmdline_p)
+{
+	atomic_notifier_chain_register(&panic_notifier_list,
+			&panic_exit_notifier);
+	paging_init();
+	strlcpy(boot_command_line, command_line, COMMAND_LINE_SIZE);
+ 	*cmdline_p = command_line;
+	setup_hostinfo();
+}
+
+void __init check_bugs(void)
+{
+	arch_check_bugs();
+ 	os_check_bugs();
+}
+
+void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+{
+}
+
+#ifdef CONFIG_SMP
+void alternatives_smp_module_add(struct module *mod, char *name,
+				 void *locks, void *locks_end,
+				 void *text,  void *text_end)
+{
+}
+
+void alternatives_smp_module_del(struct module *mod)
+{
+}
+#endif

arch/um/kernel/umid.c

@@ -0,0 +1,39 @@
+/* 
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "asm/errno.h"
+#include "init.h"
+#include "os.h"
+#include "kern.h"
+#include "linux/kernel.h"
+
+/* Changed by set_umid_arg */
+static int umid_inited = 0;
+
+static int __init set_umid_arg(char *name, int *add)
+{
+	int err;
+
+	if(umid_inited){
+		printf("umid already set\n");
+		return 0;
+	}
+
+	*add = 0;
+	err = set_umid(name);
+	if(err == -EEXIST)
+		printf("umid '%s' already in use\n", name);
+	else if(!err)
+		umid_inited = 1;
+
+	return 0;
+}
+
+__uml_setup("umid=", set_umid_arg,
+"umid=<name>\n"
+"    This is used to assign a unique identity to this UML machine and\n"
+"    is used for naming the pid file and management console socket.\n\n"
+);
+

arch/um/kernel/uml.lds.S

@@ -0,0 +1,106 @@
+#include <asm-generic/vmlinux.lds.h>
+
+OUTPUT_FORMAT(ELF_FORMAT)
+OUTPUT_ARCH(ELF_ARCH)
+ENTRY(_start)
+jiffies = jiffies_64;
+
+SECTIONS
+{
+  /* This must contain the right address - not quite the default ELF one.*/
+  PROVIDE (__executable_start = START);
+  /* Static binaries stick stuff here, like the sigreturn trampoline,
+   * invisibly to objdump.  So, just make __binary_start equal to the very
+   * beginning of the executable, and if there are unmapped pages after this,
+   * they are forever unusable.
+   */
+  __binary_start = START;
+
+  . = START + SIZEOF_HEADERS;
+
+#ifdef MODE_TT
+  .remap_data : { UNMAP_PATH (.data .bss) }
+  .remap : { UNMAP_PATH (.text) }
+
+  . = ALIGN(4096);		/* Init code and data */
+#endif
+
+  _text = .;
+  _stext = .;
+  __init_begin = .;
+  .init.text : {
+	_sinittext = .;
+	*(.init.text)
+	_einittext = .;
+  }
+  . = ALIGN(4096);
+
+  .text      :
+  {
+    *(.text)
+    SCHED_TEXT
+    LOCK_TEXT
+    *(.fixup)
+    /* .gnu.warning sections are handled specially by elf32.em.  */
+    *(.gnu.warning)
+    *(.gnu.linkonce.t*)
+
+    . = ALIGN(4096);
+    __syscall_stub_start = .;
+    *(.__syscall_stub*)
+    __syscall_stub_end = .;
+    . = ALIGN(4096);
+  }
+
+  #include "asm/common.lds.S"
+
+  init.data : { *(init.data) }
+  .data    :
+  {
+    . = ALIGN(KERNEL_STACK_SIZE);		/* init_task */
+    *(.data.init_task)
+    *(.data)
+    *(.gnu.linkonce.d*)
+    CONSTRUCTORS
+  }
+  .data1   : { *(.data1) }
+  .ctors         :
+  {
+    *(.ctors)
+  }
+  .dtors         :
+  {
+    *(.dtors)
+  }
+
+  .got           : { *(.got.plt) *(.got) }
+  .dynamic       : { *(.dynamic) }
+  .tdata	  : { *(.tdata .tdata.* .gnu.linkonce.td.*) }
+  .tbss		  : { *(.tbss .tbss.* .gnu.linkonce.tb.*) *(.tcommon) }
+  /* We want the small data sections together, so single-instruction offsets
+     can access them all, and initialized data all before uninitialized, so
+     we can shorten the on-disk segment size.  */
+  .sdata     : { *(.sdata) }
+  _edata  =  .;
+  PROVIDE (edata = .);
+  . = ALIGN(0x1000);
+  .sbss      :
+  {
+   __bss_start = .;
+   PROVIDE(_bss_start = .);
+   *(.sbss)
+   *(.scommon)
+  }
+  .bss       :
+  {
+   *(.dynbss)
+   *(.bss)
+   *(COMMON)
+  }
+  _end = .;
+  PROVIDE (end = .);
+
+  STABS_DEBUG
+
+  DWARF_DEBUG
+}

arch/um/kernel/vmlinux.lds.S

@@ -0,0 +1,5 @@
+#ifdef CONFIG_LD_SCRIPT_STATIC
+#include "uml.lds.S"
+#else
+#include "dyn.lds.S"
+#endif