17 files changed, 110 insertions, 187 deletions

diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile
index 9736ca27c5f0..a8918e80df96 100644
--- a/arch/um/kernel/Makefile
+++ b/arch/um/kernel/Makefile
@@ -14,7 +14,7 @@ obj-y = config.o exec_kern.o exitcode.o \
         tlb.o trap_kern.o trap_user.o uaccess_user.o um_arch.o umid.o \
         user_util.o
 
-obj-$(CONFIG_BLK_DEV_INITRD) += initrd_kern.o initrd_user.o
+obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o
 obj-$(CONFIG_GPROF)     += gprof_syms.o
 obj-$(CONFIG_GCOV)      += gmon_syms.o
 obj-$(CONFIG_TTY_LOG)   += tty_log.o
diff --git a/arch/um/kernel/checksum.c b/arch/um/kernel/checksum.c
deleted file mode 100644
index e69de29bb2d1..000000000000
--- a/arch/um/kernel/checksum.c
+++ /dev/null
diff --git a/arch/um/kernel/exec_kern.c b/arch/um/kernel/exec_kern.c
index 49ddabe69be7..efd222ffe20e 100644
--- a/arch/um/kernel/exec_kern.c
+++ b/arch/um/kernel/exec_kern.c
@@ -16,7 +16,6 @@
 #include "kern.h"
 #include "irq_user.h"
 #include "tlb.h"
-#include "2_5compat.h"
 #include "os.h"
 #include "time_user.h"
 #include "choose-mode.h"
diff --git a/arch/um/kernel/initrd_kern.c b/arch/um/kernel/initrd.c
index fc568af468b9..82ecf904b09c 100644
--- a/arch/um/kernel/initrd_kern.c
+++ b/arch/um/kernel/initrd.c
@@ -41,12 +41,31 @@ static int __init uml_initrd_setup(char *line, int *add)
         return 0;
 }
 
-__uml_setup("initrd=", uml_initrd_setup, 
+__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
diff --git a/arch/um/kernel/initrd_user.c b/arch/um/kernel/initrd_user.c
deleted file mode 100644
index cb90681e151c..000000000000
--- a/arch/um/kernel/initrd_user.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
- * Licensed under the GPL
- */
-
-#include <unistd.h>
-#include <sys/types.h>
-#include <sys/stat.h>
-#include <errno.h>
-
-#include "user_util.h"
-#include "kern_util.h"
-#include "user.h"
-#include "initrd.h"
-#include "os.h"
-
-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:
- */
diff --git a/arch/um/kernel/irq_user.c b/arch/um/kernel/irq_user.c
index 6d6f9484b884..b3074cbaa479 100644
--- a/arch/um/kernel/irq_user.c
+++ b/arch/um/kernel/irq_user.c
@@ -236,9 +236,15 @@ static void free_irq_by_cb(int (*test)(struct irq_fd *, void *), void *arg)
                                        (*prev)->fd, pollfds[i].fd);
                                 goto out;
                         }
-                        memcpy(&pollfds[i], &pollfds[i + 1],
-                               (pollfds_num - i - 1) * sizeof(pollfds[0]));
+
                         pollfds_num--;
+
+                        /* This moves the *whole* array after pollfds[i] (though
+                         * it doesn't spot as such)! */
+
+                        memmove(&pollfds[i], &pollfds[i + 1],
+                               (pollfds_num - i) * sizeof(pollfds[0]));
+
                         if(last_irq_ptr == &old_fd->next) 
                                 last_irq_ptr = prev;
                         *prev = (*prev)->next;
diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c
index 78d69dc74b26..99439fa15ef4 100644
--- a/arch/um/kernel/ksyms.c
+++ b/arch/um/kernel/ksyms.c
@@ -57,6 +57,7 @@ EXPORT_SYMBOL(copy_to_user_tt);
 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);
 
diff --git a/arch/um/kernel/main.c b/arch/um/kernel/main.c
index a17c49703f9b..e42e6364ca13 100644
--- a/arch/um/kernel/main.c
+++ b/arch/um/kernel/main.c
@@ -71,7 +71,7 @@ static __init void do_uml_initcalls(void)
 
 static void last_ditch_exit(int sig)
 {
-        CHOOSE_MODE(kmalloc_ok = 0, (void) 0);
+        kmalloc_ok = 0;
         signal(SIGINT, SIG_DFL);
         signal(SIGTERM, SIG_DFL);
         signal(SIGHUP, SIG_DFL);
@@ -87,7 +87,7 @@ int main(int argc, char **argv, char **envp)
 {
         char **new_argv;
         sigset_t mask;
-        int ret, i;
+        int ret, i, err;
 
         /* Enable all signals except SIGIO - in some environments, we can
          * enter with some signals blocked
@@ -160,27 +160,29 @@ int main(int argc, char **argv, char **envp)
          */
         change_sig(SIGPROF, 0);
 
-        /* Reboot */
-        if(ret){
-                int err;
-
-                printf("\n");
+        /* This signal stuff used to be in the reboot case.  However,
+         * sometimes a SIGVTALRM can come in when we're halting (reproducably
+         * when writing out gcov information, presumably because that takes
+         * some time) and cause a segfault.
+         */
 
-                /* stop timers and set SIG*ALRM to be ignored */
-                disable_timer();
+        /* stop timers and set SIG*ALRM to be ignored */
+        disable_timer();
 
-                /* disable SIGIO for the fds and set SIGIO to be ignored */
-                err = deactivate_all_fds();
-                if(err)
-                        printf("deactivate_all_fds failed, errno = %d\n",
-                               -err);
+        /* disable SIGIO for the fds and set SIGIO to be ignored */
+        err = deactivate_all_fds();
+        if(err)
+                printf("deactivate_all_fds failed, errno = %d\n", -err);
 
-                /* Let any pending signals fire now.  This ensures
-                 * that they won't be delivered after the exec, when
-                 * they are definitely not expected.
-                 */
-                unblock_signals();
+        /* Let any pending signals fire now.  This ensures
+         * that they won't be delivered after the exec, when
+         * they are definitely not expected.
+         */
+        unblock_signals();
 
+        /* Reboot */
+        if(ret){
+                printf("\n");
                 execvp(new_argv[0], new_argv);
                 perror("Failed to exec kernel");
                 ret = 1;
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index f156661781cb..c22825f13e40 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -100,12 +100,37 @@ void mem_init(void)
 #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;
-        pte_t *pte;
         int i, j;
         unsigned long vaddr;
 
@@ -115,15 +140,12 @@ static void __init fixrange_init(unsigned long start, unsigned long end,
         pgd = pgd_base + i;
 
         for ( ; (i < PTRS_PER_PGD) && (vaddr < end); pgd++, i++) {
-                pmd = (pmd_t *)pgd;
+                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++) {
-                        if (pmd_none(*pmd)) {
-                                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();
-                        }
+                        one_page_table_init(pmd);
                         vaddr += PMD_SIZE;
                 }
                 j = 0;
diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c
index c1adf7ba3fd1..804c6bbdf67c 100644
--- a/arch/um/kernel/process_kern.c
+++ b/arch/um/kernel/process_kern.c
@@ -43,7 +43,6 @@
 #include "tlb.h"
 #include "frame_kern.h"
 #include "sigcontext.h"
-#include "2_5compat.h"
 #include "os.h"
 #include "mode.h"
 #include "mode_kern.h"
@@ -55,18 +54,6 @@
  */
 struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } };
 
-struct task_struct *get_task(int pid, int require)
-{
-        struct task_struct *ret;
-
-        read_lock(&tasklist_lock);
-        ret = find_task_by_pid(pid);
-        read_unlock(&tasklist_lock);
-
-        if(require && (ret == NULL)) panic("get_task couldn't find a task\n");
-        return(ret);
-}
-
 int external_pid(void *t)
 {
         struct task_struct *task = t ? t : current;
@@ -189,7 +176,6 @@ void default_idle(void)
 
         while(1){
                 /* endless idle loop with no priority at all */
-                SET_PRI(current);
 
                 /*
                  * although we are an idle CPU, we do not want to
@@ -212,11 +198,6 @@ int page_size(void)
         return(PAGE_SIZE);
 }
 
-unsigned long page_mask(void)
-{
-        return(PAGE_MASK);
-}
-
 void *um_virt_to_phys(struct task_struct *task, unsigned long addr, 
                       pte_t *pte_out)
 {
@@ -349,11 +330,6 @@ char *uml_strdup(char *string)
         return(new);
 }
 
-void *get_init_task(void)
-{
-        return(&init_thread_union.thread_info.task);
-}
-
 int copy_to_user_proc(void __user *to, void *from, int size)
 {
         return(copy_to_user(to, from, size));
@@ -480,15 +456,3 @@ unsigned long arch_align_stack(unsigned long sp)
         return sp & ~0xf;
 }
 #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:
- */
diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c
index 2b75d8d9ba73..71af4d503899 100644
--- a/arch/um/kernel/ptrace.c
+++ b/arch/um/kernel/ptrace.c
@@ -28,9 +28,9 @@ static inline void set_singlestepping(struct task_struct *child, int on)
         child->thread.singlestep_syscall = 0;
 
 #ifdef SUBARCH_SET_SINGLESTEPPING
-        SUBARCH_SET_SINGLESTEPPING(child, on)
+        SUBARCH_SET_SINGLESTEPPING(child, on);
 #endif
-                }
+}
 
 /*
  * Called by kernel/ptrace.c when detaching..
@@ -83,7 +83,7 @@ long sys_ptrace(long request, long pid, long addr, long data)
         }
 
 #ifdef SUBACH_PTRACE_SPECIAL
-        SUBARCH_PTRACE_SPECIAL(child,request,addr,data)
+        SUBARCH_PTRACE_SPECIAL(child,request,addr,data);
 #endif
 
         ret = ptrace_check_attach(child, request == PTRACE_KILL);
@@ -322,11 +322,9 @@ void syscall_trace(union uml_pt_regs *regs, int entryexit)
                                             UPT_SYSCALL_ARG2(regs),
                                             UPT_SYSCALL_ARG3(regs),
                                             UPT_SYSCALL_ARG4(regs));
-                else {
-                        int res = UPT_SYSCALL_RET(regs);
-                        audit_syscall_exit(current, AUDITSC_RESULT(res),
-                                           res);
-                }
+                else audit_syscall_exit(current,
+                                        AUDITSC_RESULT(UPT_SYSCALL_RET(regs)),
+                                        UPT_SYSCALL_RET(regs));
         }
 
         /* Fake a debug trap */
@@ -356,14 +354,3 @@ void syscall_trace(union uml_pt_regs *regs, int entryexit)
                 current->exit_code = 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:
- */
diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c
index e630438f9e73..f80850091e79 100644
--- a/arch/um/kernel/sysrq.c
+++ b/arch/um/kernel/sysrq.c
@@ -3,6 +3,7 @@
  * Licensed under the GPL
  */
 
+#include "linux/config.h"
 #include "linux/sched.h"
 #include "linux/kernel.h"
 #include "linux/module.h"
@@ -12,14 +13,14 @@
 #include "sysrq.h"
 #include "user_util.h"
 
-void show_trace(unsigned long * stack)
+/* Catch non-i386 SUBARCH's. */
+#if !defined(CONFIG_UML_X86) || defined(CONFIG_64BIT)
+void show_trace(struct task_struct *task, unsigned long * stack)
 {
-        /* XXX: Copy the CONFIG_FRAME_POINTER stack-walking backtrace from
-         * arch/i386/kernel/traps.c, and then move this to sys-i386/sysrq.c.*/
         unsigned long addr;
 
         if (!stack) {
-                stack = (unsigned long*) &stack;
+                stack = (unsigned long*) &stack;
                 WARN_ON(1);
         }
 
@@ -35,6 +36,7 @@ void show_trace(unsigned long * stack)
         }
         printk("\n");
 }
+#endif
 
 /*
  * stack dumps generator - this is used by arch-independent code.
@@ -44,7 +46,7 @@ void dump_stack(void)
 {
         unsigned long stack;
 
-        show_trace(&stack);
+        show_trace(current, &stack);
 }
 EXPORT_SYMBOL(dump_stack);
 
@@ -59,7 +61,11 @@ void show_stack(struct task_struct *task, unsigned long *esp)
         int i;
 
         if (esp == NULL) {
-                if (task != current) {
+                if (task != current && task != NULL) {
+                        /* XXX: Isn't this bogus? I.e. isn't this the
+                         * *userspace* stack of this task? If not so, use this
+                         * even when task == current (as in i386).
+                         */
                         esp = (unsigned long *) KSTK_ESP(task);
                         /* Which one? No actual difference - just coding style.*/
                         //esp = (unsigned long *) PT_REGS_IP(&task->thread.regs);
@@ -77,5 +83,6 @@ void show_stack(struct task_struct *task, unsigned long *esp)
                 printk("%08lx ", *stack++);
         }
 
-        show_trace(esp);
+        printk("Call Trace: \n");
+        show_trace(current, esp);
 }
diff --git a/arch/um/kernel/trap_kern.c b/arch/um/kernel/trap_kern.c
index 5fca2c61eb98..c20aef120598 100644
--- a/arch/um/kernel/trap_kern.c
+++ b/arch/um/kernel/trap_kern.c
@@ -23,7 +23,6 @@
 #include "kern.h"
 #include "chan_kern.h"
 #include "mconsole_kern.h"
-#include "2_5compat.h"
 #include "mem.h"
 #include "mem_kern.h"
 
@@ -57,10 +56,11 @@ int handle_page_fault(unsigned long address, unsigned long ip,
         *code_out = SEGV_ACCERR;
         if(is_write && !(vma->vm_flags & VM_WRITE)) 
                 goto out;
+
+        if(!(vma->vm_flags & (VM_READ | VM_EXEC)))
+                goto out;
+
         page = address & PAGE_MASK;
-        pgd = pgd_offset(mm, page);
-        pud = pud_offset(pgd, page);
-        pmd = pmd_offset(pud, page);
         do {
  survive:
                 switch (handle_mm_fault(mm, vma, address, is_write)){
@@ -106,33 +106,6 @@ out_of_memory:
         goto out;
 }
 
-LIST_HEAD(physmem_remappers);
-
-void register_remapper(struct remapper *info)
-{
-        list_add(&info->list, &physmem_remappers);
-}
-
-static int check_remapped_addr(unsigned long address, int is_write)
-{
-        struct remapper *remapper;
-        struct list_head *ele;
-        __u64 offset;
-        int fd;
-
-        fd = phys_mapping(__pa(address), &offset);
-        if(fd == -1)
-                return(0);
-
-        list_for_each(ele, &physmem_remappers){
-                remapper = list_entry(ele, struct remapper, list);
-                if((*remapper->proc)(fd, address, is_write, offset))
-                        return(1);
-        }
-
-        return(0);
-}
-
 /*
  * We give a *copy* of the faultinfo in the regs to segv.
  * This must be done, since nesting SEGVs could overwrite
@@ -151,8 +124,6 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc)
                 flush_tlb_kernel_vm();
                 return(0);
         }
-        else if(check_remapped_addr(address & PAGE_MASK, is_write))
-                return(0);
         else if(current->mm == NULL)
                 panic("Segfault with no mm");
         err = handle_page_fault(address, ip, is_write, is_user, &si.si_code);
diff --git a/arch/um/kernel/tt/ksyms.c b/arch/um/kernel/tt/ksyms.c
index 92ec85d67c7c..84a9385a8fef 100644
--- a/arch/um/kernel/tt/ksyms.c
+++ b/arch/um/kernel/tt/ksyms.c
@@ -12,6 +12,7 @@ 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);
diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c
index df810ca8fc12..776310fd5b8b 100644
--- a/arch/um/kernel/tt/process_kern.c
+++ b/arch/um/kernel/tt/process_kern.c
@@ -32,10 +32,6 @@ void *switch_to_tt(void *prev, void *next, void *last)
         unsigned long flags;
         int err, vtalrm, alrm, prof, cpu;
         char c;
-        /* jailing and SMP are incompatible, so this doesn't need to be 
-         * made per-cpu 
-         */
-        static int reading;
 
         from = prev;
         to = next;
@@ -59,13 +55,11 @@ void *switch_to_tt(void *prev, void *next, void *last)
         c = 0;
         set_current(to);
 
-        reading = 0;
         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);
 
-        reading = 1;
-        if(from->thread.mode.tt.switch_pipe[0] == -1)
+        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));
diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c
index 4d10ec372a67..418427107b29 100644
--- a/arch/um/kernel/um_arch.c
+++ b/arch/um/kernel/um_arch.c
@@ -111,12 +111,6 @@ struct seq_operations cpuinfo_op = {
         .show   = show_cpuinfo,
 };
 
-pte_t * __bad_pagetable(void)
-{
-        panic("Someone should implement __bad_pagetable");
-        return(NULL);
-}
-
 /* Set in linux_main */
 unsigned long host_task_size;
 unsigned long task_size;
diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S
index 76eadb309189..dd5355500bdc 100644
--- a/arch/um/kernel/uml.lds.S
+++ b/arch/um/kernel/uml.lds.S
@@ -73,6 +73,8 @@ SECTIONS
 
   .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.  */