Merge with Linux 2.3.99-pre3.

6 files changed, 120 insertions, 80 deletions

diff --git a/kernel/Makefile b/kernel/Makefile
index bec392fca..2a4f548af 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -7,9 +7,6 @@
 #
 # Note 2! The CFLAGS definitions are now in the main makefile...
 
-.S.s:
-	$(CPP) $(CPPFLAGS) -traditional $< -o $*.s
-
 O_TARGET := kernel.o
 O_OBJS    = sched.o dma.o fork.o exec_domain.o panic.o printk.o sys.o \
 	    module.o exit.o itimer.o info.o time.o softirq.o resource.o \
@@ -29,7 +26,7 @@ ifeq ($(CONFIG_MODULES),y)
 OX_OBJS  += ksyms.o
 endif
 
-ifdef CONFIG_PM
+ifeq ($(CONFIG_PM),y)
 OX_OBJS += pm.o
 endif
 
diff --git a/kernel/ksyms.c b/kernel/ksyms.c
index a6748e70c..ad11eafea 100644
--- a/kernel/ksyms.c
+++ b/kernel/ksyms.c
@@ -207,6 +207,7 @@ EXPORT_SYMBOL(posix_unblock_lock);
 EXPORT_SYMBOL(locks_mandatory_area);
 EXPORT_SYMBOL(dput);
 EXPORT_SYMBOL(is_root_busy);
+EXPORT_SYMBOL(have_submounts);
 EXPORT_SYMBOL(prune_dcache);
 EXPORT_SYMBOL(shrink_dcache_sb);
 EXPORT_SYMBOL(shrink_dcache_parent);
@@ -324,12 +325,20 @@ EXPORT_SYMBOL(proc_doulongvec_ms_jiffies_minmax);
 EXPORT_SYMBOL(proc_doulongvec_minmax);
 
 /* interrupt handling */
+EXPORT_SYMBOL(add_timer);
+EXPORT_SYMBOL(del_timer);
 EXPORT_SYMBOL(request_irq);
 EXPORT_SYMBOL(free_irq);
+
+/* The notion of irq probe/assignment is foreign to S/390 */
+
+#if !defined(CONFIG_ARCH_S390)
 EXPORT_SYMBOL(probe_irq_on);
 EXPORT_SYMBOL(probe_irq_off);
-EXPORT_SYMBOL(add_timer);
-EXPORT_SYMBOL(del_timer);
+EXPORT_SYMBOL(autoirq_setup);
+EXPORT_SYMBOL(autoirq_report);
+#endif
+
 #ifdef __SMP__
 EXPORT_SYMBOL(del_timer_sync);
 #endif
@@ -366,10 +375,6 @@ EXPORT_SYMBOL(lock_kiovec);
 EXPORT_SYMBOL(unlock_kiovec);
 EXPORT_SYMBOL(brw_kiovec);
 
-/* autoirq from  drivers/net/auto_irq.c */
-EXPORT_SYMBOL(autoirq_setup);
-EXPORT_SYMBOL(autoirq_report);
-
 /* dma handling */
 EXPORT_SYMBOL(request_dma);
 EXPORT_SYMBOL(free_dma);
@@ -439,8 +444,8 @@ EXPORT_SYMBOL(setup_arg_pages);
 EXPORT_SYMBOL(copy_strings_kernel);
 EXPORT_SYMBOL(do_execve);
 EXPORT_SYMBOL(flush_old_exec);
-EXPORT_SYMBOL(open_dentry);
-EXPORT_SYMBOL(read_exec);
+EXPORT_SYMBOL(kernel_read);
+EXPORT_SYMBOL(open_exec);
 
 /* Miscellaneous access points */
 EXPORT_SYMBOL(si_meminfo);
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index c3f75124f..a87858804 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -44,9 +44,9 @@ repeat:
 	mapnr = pte_pagenr(*pgtable);
 	if (write && (!pte_write(*pgtable) || !pte_dirty(*pgtable)))
 		goto fault_in_page;
-	if (mapnr >= max_mapnr)
-		return 0;
 	page = mem_map + mapnr;
+	if ((mapnr >= max_mapnr) || PageReserved(page))
+		return 0;
 	flush_cache_page(vma, addr);
 
 	if (write) {
diff --git a/kernel/sched.c b/kernel/sched.c
index e1dbc62ba..c0ac80395 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -618,6 +618,11 @@ handle_softirq:
 	goto handle_softirq_back;
 
 handle_tq_scheduler:
+	/*
+	 * do not run the task queue with disabled interrupts,
+	 * cli() wouldn't work on SMP
+	 */
+	sti();
 	run_task_queue(&tq_scheduler);
 	goto tq_scheduler_back;
 
@@ -630,7 +635,7 @@ move_rr_last:
 
 scheduling_in_interrupt:
 	printk("Scheduling in interrupt\n");
-	*(int *)0 = 0;
+	BUG();
 	return;
 }
 
diff --git a/kernel/signal.c b/kernel/signal.c
index 55017e2f6..0958af05c 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -90,7 +90,7 @@ flush_signal_handlers(struct task_struct *t)
  * Dequeue a signal and return the element to the caller, which is 
  * expected to free it.
  *
- * All callers of must be holding current->sigmask_lock.
+ * All callers must be holding current->sigmask_lock.
  */
 
 int
@@ -149,19 +149,19 @@ printk("SIG dequeue (%s:%d): %d ", current->comm, current->pid,
 			kmem_cache_free(signal_queue_cachep,q);
 			atomic_dec(&nr_queued_signals);
 
-			/* then see if this signal is still pending. */
-			q = *pp;
-			while (q) {
-				if (q->info.si_signo == sig) {
-					reset = 0;
-					break;
-				}
-				q = q->next;
-			}
+			/* Then see if this signal is still pending.
+			   (Non rt signals may not be queued twice.)
+			 */
+			if (sig >= SIGRTMIN)
+				for (q = *pp; q; q = q->next)
+					if (q->info.si_signo == sig) {
+						reset = 0;
+						break;
+					}
+					
 		} else {
-			/* Ok, it wasn't in the queue.  It must have
-			   been sent either by a non-rt mechanism and
-			   we ran out of queue space.  So zero out the
+			/* Ok, it wasn't in the queue.  We must have
+			   been out of queue space.  So zero out the
 			   info.  */
 			info->si_signo = sig;
 			info->si_errno = 0;
@@ -170,9 +170,10 @@ printk("SIG dequeue (%s:%d): %d ", current->comm, current->pid,
 			info->si_uid = 0;
 		}
 
-		if (reset)
+		if (reset) {
 			sigdelset(&current->signal, sig);
-		recalc_sigpending(current);
+			recalc_sigpending(current);
+		}
 
 		/* XXX: Once POSIX.1b timers are in, if si_code == SI_TIMER,
 		   we need to xchg out the timer overrun values.  */
@@ -196,6 +197,43 @@ printk(" %d -> %d\n", signal_pending(current), sig);
 }
 
 /*
+ * Remove signal sig from queue and from t->signal.
+ * Returns 1 if sig was found in t->signal.
+ *
+ * All callers must be holding t->sigmask_lock.
+ */
+static int rm_sig_from_queue(int sig, struct task_struct *t)
+{
+	struct signal_queue *q, **pp;
+
+	if (sig >= SIGRTMIN) {
+		printk(KERN_CRIT "SIG: rm_sig_from_queue() doesn't support rt signals\n");
+		return 0;
+	}
+
+	if (!sigismember(&t->signal, sig))
+		return 0;
+
+	sigdelset(&t->signal, sig);
+
+	pp = &t->sigqueue;
+	q = t->sigqueue;
+
+	/* Find the one we're interested in ...
+	   It may appear only once. */
+	for ( ; q ; pp = &q->next, q = q->next)
+		if (q->info.si_signo == sig)
+			break;
+	if (q) {
+		if ((*pp = q->next) == NULL)
+			t->sigqueue_tail = pp;
+		kmem_cache_free(signal_queue_cachep,q);
+		atomic_dec(&nr_queued_signals);
+	}
+	return 1;
+}
+
+/*
  * Determine whether a signal should be posted or not.
  *
  * Signals with SIG_IGN can be ignored, except for the
@@ -273,18 +311,16 @@ printk("SIG queue (%s:%d): %d ", t->comm, t->pid, sig);
 		if (t->state == TASK_STOPPED)
 			wake_up_process(t);
 		t->exit_code = 0;
-		sigdelsetmask(&t->signal, (sigmask(SIGSTOP)|sigmask(SIGTSTP)|
-					   sigmask(SIGTTOU)|sigmask(SIGTTIN)));
-		/* Inflict this corner case with recalculations, not mainline */
-		recalc_sigpending(t);
+		if (rm_sig_from_queue(SIGSTOP, t) || rm_sig_from_queue(SIGTSTP, t) ||
+		    rm_sig_from_queue(SIGTTOU, t) || rm_sig_from_queue(SIGTTIN, t))
+			recalc_sigpending(t);
 		break;
 
 	case SIGSTOP: case SIGTSTP:
 	case SIGTTIN: case SIGTTOU:
 		/* If we're stopping again, cancel SIGCONT */
-		sigdelset(&t->signal, SIGCONT);
-		/* Inflict this corner case with recalculations, not mainline */
-		recalc_sigpending(t);
+		if (rm_sig_from_queue(SIGCONT, t))
+			recalc_sigpending(t);
 		break;
 	}
 
@@ -338,8 +374,12 @@ printk("SIG queue (%s:%d): %d ", t->comm, t->pid, sig);
 				q->info = *info;
 				break;
 		}
-	} else {
-		/* Queue overflow, we have to abort. */
+	} else if (sig >= SIGRTMIN && info && (unsigned long)info != 1
+		   && info->si_code < 0) {
+		/*
+		 * Queue overflow, abort.  We may abort if the signal was rt
+		 * and sent by user using something other than kill().
+		 */
 		ret = -EAGAIN;
 		goto out;
 	}
@@ -406,7 +446,7 @@ force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
 }
 
 /*
- * kill_pg() sends a signal to a process group: this is what the tty
+ * kill_pg_info() sends a signal to a process group: this is what the tty
  * control characters do (^C, ^Z etc)
  */
 
@@ -437,7 +477,7 @@ kill_pg_info(int sig, struct siginfo *info, pid_t pgrp)
 }
 
 /*
- * kill_sl() sends a signal to the session leader: this is used
+ * kill_sl_info() sends a signal to the session leader: this is used
  * to send SIGHUP to the controlling process of a terminal when
  * the connection is lost.
  */
@@ -484,7 +524,7 @@ kill_proc_info(int sig, struct siginfo *info, pid_t pid)
 }
 
 /*
- * kill_something() interprets pid in interesting ways just like kill(2).
+ * kill_something_info() interprets pid in interesting ways just like kill(2).
  *
  * POSIX specifies that kill(-1,sig) is unspecified, but what we have
  * is probably wrong.  Should make it like BSD or SYSV.
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 4e654fa3a..35883ed4c 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -14,6 +14,8 @@
  * Added proc_doulongvec_minmax, 09/08/99, Carlos H. Bauer.
  * Changed linked lists to use list.h instead of lists.h, 02/24/00, Bill
  *  Wendling.
+ * The list_for_each() macro wasn't appropriate for the sysctl loop.
+ *  Removed it and replaced it with older style, 03/23/00, Bill Wendling
  */
 
 #include <linux/config.h>
@@ -84,9 +86,9 @@ static int parse_table(int *, int, void *, size_t *, void *, size_t,
 static int proc_doutsstring(ctl_table *table, int write, struct file *filp,
 		  void *buffer, size_t *lenp);
 
-
 static ctl_table root_table[];
-static LIST_HEAD(root_table_header);
+static struct ctl_table_header root_table_header =
+	{ root_table, LIST_HEAD_INIT(root_table_header.ctl_entry) };
 
 static ctl_table kern_table[];
 static ctl_table vm_table[];
@@ -99,7 +101,6 @@ static ctl_table debug_table[];
 static ctl_table dev_table[];
 extern ctl_table random_table[];
 
-
 /* /proc declarations: */
 
 #ifdef CONFIG_PROC_FS
@@ -108,14 +109,12 @@ static ssize_t proc_readsys(struct file *, char *, size_t, loff_t *);
 static ssize_t proc_writesys(struct file *, const char *, size_t, loff_t *);
 static int proc_sys_permission(struct inode *, int);
 
-struct file_operations proc_sys_file_operations =
-{
+struct file_operations proc_sys_file_operations = {
 	read:		proc_readsys,
 	write:		proc_writesys,
 };
 
-static struct inode_operations proc_sys_inode_operations =
-{
+static struct inode_operations proc_sys_inode_operations = {
 	permission:	proc_sys_permission,
 };
 
@@ -124,6 +123,7 @@ extern struct proc_dir_entry *proc_sys_root;
 static void register_proc_table(ctl_table *, struct proc_dir_entry *);
 static void unregister_proc_table(ctl_table *, struct proc_dir_entry *);
 #endif
+
 extern int inodes_stat[];
 extern int dentry_stat[];
 
@@ -294,39 +294,33 @@ void __init sysctl_init(void)
 	register_proc_table(root_table, proc_sys_root);
 	init_irq_proc();
 #endif
-
 }
 
-int do_sysctl (int *name, int nlen,
-	       void *oldval, size_t *oldlenp,
+int do_sysctl(int *name, int nlen, void *oldval, size_t *oldlenp,
 	       void *newval, size_t newlen)
 {
-	int error;
 	struct list_head *tmp;
-	void *context;
-	
+
 	if (nlen == 0 || nlen >= CTL_MAXNAME)
 		return -ENOTDIR;
-	
-	if (oldval) 
-	{
+	if (oldval) {
 		int old_len;
-		if (!oldlenp)
-			return -EFAULT;
-		if(get_user(old_len, oldlenp))
+		if (!oldlenp || get_user(old_len, oldlenp))
 			return -EFAULT;
 	}
-	list_for_each(tmp, &root_table_header) {
+	tmp = &root_table_header.ctl_entry;
+	do {
 		struct ctl_table_header *head =
 			list_entry(tmp, struct ctl_table_header, ctl_entry);
-		context = NULL;
-		error = parse_table(name, nlen, oldval, oldlenp, 
-				    newval, newlen, head->ctl_table, &context);
+		void *context = NULL;
+		int error = parse_table(name, nlen, oldval, oldlenp, 
+					newval, newlen, head->ctl_table,
+					&context);
 		if (context)
 			kfree(context);
 		if (error != -ENOTDIR)
 			return error;
-	}
+	} while (tmp != &root_table_header.ctl_entry);
 	return -ENOTDIR;
 }
 
@@ -335,7 +329,7 @@ extern asmlinkage long sys_sysctl(struct __sysctl_args *args)
 	struct __sysctl_args tmp;
 	int error;
 
-	if(copy_from_user(&tmp, args, sizeof(tmp)))
+	if (copy_from_user(&tmp, args, sizeof(tmp)))
 		return -EFAULT;
 		
 	lock_kernel();
@@ -345,9 +339,10 @@ extern asmlinkage long sys_sysctl(struct __sysctl_args *args)
 	return error;
 }
 
-
-/* ctl_perm does NOT grant the superuser all rights automatically, because
-   some sysctl variables are readonly even to root. */
+/*
+ * ctl_perm does NOT grant the superuser all rights automatically, because
+ * some sysctl variables are readonly even to root.
+ */
 
 static int test_perm(int mode, int op)
 {
@@ -370,17 +365,16 @@ static int parse_table(int *name, int nlen,
 		       void *newval, size_t newlen,
 		       ctl_table *table, void **context)
 {
-	int error;
 repeat:
 	if (!nlen)
 		return -ENOTDIR;
 
 	for ( ; table->ctl_name; table++) {
 		int n;
-		if(get_user(n,name))
+		if (get_user(n, name))
 			return -EFAULT;
-		if (n == table->ctl_name ||
-		    table->ctl_name == CTL_ANY) {
+		if (n == table->ctl_name || table->ctl_name == CTL_ANY) {
+			int error;
 			if (table->child) {
 				if (ctl_perm(table, 001))
 					return -EPERM;
@@ -389,8 +383,8 @@ repeat:
 						table, name, nlen,
 						oldval, oldlenp,
 						newval, newlen, context);
-				if (error)
-					return error;
+					if (error)
+						return error;
 				}
 				name++;
 				nlen--;
@@ -465,9 +459,9 @@ struct ctl_table_header *register_sysctl_table(ctl_table * table,
 	tmp->ctl_table = table;
 	INIT_LIST_HEAD(&tmp->ctl_entry);
 	if (insert_at_head)
-		list_add(&tmp->ctl_entry, &root_table_header);
+		list_add(&tmp->ctl_entry, &root_table_header.ctl_entry);
 	else
-		list_add_tail(&tmp->ctl_entry, &root_table_header);
+		list_add_tail(&tmp->ctl_entry, &root_table_header.ctl_entry);
 #ifdef CONFIG_PROC_FS
 	register_proc_table(table, proc_sys_root);
 #endif
@@ -694,6 +688,7 @@ static int proc_doutsstring(ctl_table *table, int write, struct file *filp,
 #define OP_OR	2
 #define OP_MAX	3
 #define OP_MIN	4
+
 static int do_proc_dointvec(ctl_table *table, int write, struct file *filp,
 		  void *buffer, size_t *lenp, int conv, int op)
 {
@@ -812,7 +807,6 @@ int proc_dointvec_bset(ctl_table *table, int write, struct file *filp,
 		(current->pid == 1) ? OP_SET : OP_AND);
 }
 
-
 int proc_dointvec_minmax(ctl_table *table, int write, struct file *filp,
 		  void *buffer, size_t *lenp)
 {
@@ -911,7 +905,6 @@ int proc_dointvec_minmax(ctl_table *table, int write, struct file *filp,
 	return 0;
 }
 
-
 /*
  * an unsigned long function version
  */