Merge samba.org:/stuff/paulus/kernel/linux-2.5

into samba.org:/stuff/paulus/kernel/for-linus-ppc

7 files changed, 152 insertions, 73 deletions

diff --git a/kernel/Makefile b/kernel/Makefile
index 2929aae0c2fe..1e652214037c 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -5,9 +5,10 @@
 obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
 	    exit.o itimer.o time.o softirq.o resource.o \
 	    sysctl.o capability.o ptrace.o timer.o user.o \
-	    signal.o sys.o kmod.o workqueue.o futex.o pid.o \
+	    signal.o sys.o kmod.o workqueue.o pid.o \
 	    rcupdate.o intermodule.o extable.o params.o posix-timers.o
 
+obj-$(CONFIG_FUTEX) += futex.o
 obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o
 obj-$(CONFIG_SMP) += cpu.o
 obj-$(CONFIG_UID16) += uid16.o
diff --git a/kernel/compat.c b/kernel/compat.c
index 0dcab5fc6acd..e0998f98b72b 100644
--- a/kernel/compat.c
+++ b/kernel/compat.c
@@ -18,6 +18,7 @@
 #include <linux/signal.h>
 #include <linux/sched.h>	/* for MAX_SCHEDULE_TIMEOUT */
 #include <linux/futex.h>	/* for FUTEX_WAIT */
+#include <linux/unistd.h>
 
 #include <asm/uaccess.h>
 
@@ -211,21 +212,25 @@ asmlinkage long compat_sys_sigprocmask(int how, compat_old_sigset_t *set,
 	return ret;
 }
 
-extern long do_futex(unsigned long, int, int, unsigned long);
-
+#ifdef CONFIG_FUTEX
 asmlinkage long compat_sys_futex(u32 *uaddr, int op, int val,
 		struct compat_timespec *utime)
 {
 	struct timespec t;
 	unsigned long timeout = MAX_SCHEDULE_TIMEOUT;
+	int val2 = 0;
 
 	if ((op == FUTEX_WAIT) && utime) {
 		if (get_compat_timespec(&t, utime))
 			return -EFAULT;
 		timeout = timespec_to_jiffies(&t) + 1;
 	}
-	return do_futex((unsigned long)uaddr, op, val, timeout);
+	if (op == FUTEX_REQUEUE)
+		val2 = (int) utime;
+
+	return do_futex((unsigned long)uaddr, op, val, timeout, uaddr2, val2);
 }
+#endif
 
 asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit *rlim);
 
diff --git a/kernel/exit.c b/kernel/exit.c
index c4130eb03ca1..c5b8ec241a83 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -21,6 +21,7 @@
 #include <linux/ptrace.h>
 #include <linux/profile.h>
 #include <linux/mount.h>
+#include <linux/proc_fs.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -31,10 +32,8 @@ extern struct task_struct *child_reaper;
 
 int getrusage(struct task_struct *, int, struct rusage *);
 
-static struct dentry * __unhash_process(struct task_struct *p)
+static void __unhash_process(struct task_struct *p)
 {
-	struct dentry *proc_dentry;
-
 	nr_threads--;
 	detach_pid(p, PIDTYPE_PID);
 	detach_pid(p, PIDTYPE_TGID);
@@ -46,34 +45,25 @@ static struct dentry * __unhash_process(struct task_struct *p)
 	}
 
 	REMOVE_LINKS(p);
-	proc_dentry = p->proc_dentry;
-	if (unlikely(proc_dentry != NULL)) {
-		spin_lock(&dcache_lock);
-		if (!d_unhashed(proc_dentry)) {
-			dget_locked(proc_dentry);
-			__d_drop(proc_dentry);
-		} else
-			proc_dentry = NULL;
-		spin_unlock(&dcache_lock);
-	}
-	return proc_dentry;
 }
 
 void release_task(struct task_struct * p)
 {
-	struct dentry *proc_dentry;
 	task_t *leader;
+	struct dentry *proc_dentry;
  
 	BUG_ON(p->state < TASK_ZOMBIE);
  
 	atomic_dec(&p->user->processes);
+	spin_lock(&p->proc_lock);
+	proc_dentry = proc_pid_unhash(p);
 	write_lock_irq(&tasklist_lock);
 	if (unlikely(p->ptrace))
 		__ptrace_unlink(p);
 	BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children));
 	__exit_signal(p);
 	__exit_sighand(p);
-	proc_dentry = __unhash_process(p);
+	__unhash_process(p);
 
 	/*
 	 * If we are the last non-leader member of the thread
@@ -92,11 +82,8 @@ void release_task(struct task_struct * p)
 	p->parent->cnswap += p->nswap + p->cnswap;
 	sched_exit(p);
 	write_unlock_irq(&tasklist_lock);
-
-	if (unlikely(proc_dentry != NULL)) {
-		shrink_dcache_parent(proc_dentry);
-		dput(proc_dentry);
-	}
+	spin_unlock(&p->proc_lock);
+	proc_pid_flush(proc_dentry);
 	release_thread(p);
 	put_task_struct(p);
 }
@@ -107,14 +94,13 @@ void unhash_process(struct task_struct *p)
 {
 	struct dentry *proc_dentry;
 
+	spin_lock(&p->proc_lock);
+	proc_dentry = proc_pid_unhash(p);
 	write_lock_irq(&tasklist_lock);
-	proc_dentry = __unhash_process(p);
+	__unhash_process(p);
 	write_unlock_irq(&tasklist_lock);
-
-	if (unlikely(proc_dentry != NULL)) {
-		shrink_dcache_parent(proc_dentry);
-		dput(proc_dentry);
-	}
+	spin_unlock(&p->proc_lock);
+	proc_pid_flush(proc_dentry);
 }
 
 /*
diff --git a/kernel/fork.c b/kernel/fork.c
index a509e6da132f..23c6d34f800f 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -457,7 +457,7 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm)
 		 * not set up a proper pointer then tough luck.
 		 */
 		put_user(0, tidptr);
-		sys_futex(tidptr, FUTEX_WAKE, 1, NULL);
+		sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL);
 	}
 }
 
diff --git a/kernel/futex.c b/kernel/futex.c
index ce5c894e459e..df2dcbf557d0 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -2,6 +2,9 @@
  *  Fast Userspace Mutexes (which I call "Futexes!").
  *  (C) Rusty Russell, IBM 2002
  *
+ *  Generalized futexes, futex requeueing, misc fixes by Ingo Molnar
+ *  (C) Copyright 2003 Red Hat Inc, All Rights Reserved
+ *
  *  Thanks to Ben LaHaise for yelling "hashed waitqueues" loudly
  *  enough at me, Linus for the original (flawed) idea, Matthew
  *  Kirkwood for proof-of-concept implementation.
@@ -9,9 +12,6 @@
  *  "The futexes are also cursed."
  *  "But they come in a choice of three flavours!"
  *
- *  Generalized futexes for every mapping type, Ingo Molnar, 2002
- *
- *
  *  This program is free software; you can redistribute it and/or modify
  *  it under the terms of the GNU General Public License as published by
  *  the Free Software Foundation; either version 2 of the License, or
@@ -93,19 +93,18 @@ static inline struct list_head *hash_futex(struct page *page, int offset)
 							FUTEX_HASHBITS)];
 }
 
-/* Waiter either waiting in FUTEX_WAIT or poll(), or expecting signal */
-static inline void tell_waiter(struct futex_q *q)
-{
-	wake_up_all(&q->waiters);
-	if (q->filp)
-		send_sigio(&q->filp->f_owner, q->fd, POLL_IN);
-}
-
 /*
  * Get kernel address of the user page and pin it.
  *
  * Must be called with (and returns with) all futex-MM locks held.
  */
+static inline struct page *__pin_page_atomic (struct page *page)
+{
+	if (!PageReserved(page))
+		get_page(page);
+	return page;
+}
+
 static struct page *__pin_page(unsigned long addr)
 {
 	struct mm_struct *mm = current->mm;
@@ -116,11 +115,8 @@ static struct page *__pin_page(unsigned long addr)
 	 * Do a quick atomic lookup first - this is the fastpath.
 	 */
 	page = follow_page(mm, addr, 0);
-	if (likely(page != NULL)) {	
-		if (!PageReserved(page))
-			get_page(page);
-		return page;
-	}
+	if (likely(page != NULL))
+		return __pin_page_atomic(page);
 
 	/*
 	 * No luck - need to fault in the page:
@@ -150,16 +146,11 @@ repeat_lookup:
 	return page;
 }
 
-static inline void unpin_page(struct page *page)
-{
-	put_page(page);
-}
-
 /*
  * Wake up all waiters hashed on the physical page that is mapped
  * to this virtual address:
  */
-static int futex_wake(unsigned long uaddr, int offset, int num)
+static inline int futex_wake(unsigned long uaddr, int offset, int num)
 {
 	struct list_head *i, *next, *head;
 	struct page *page;
@@ -181,7 +172,9 @@ static int futex_wake(unsigned long uaddr, int offset, int num)
 		if (this->page == page && this->offset == offset) {
 			list_del_init(i);
 			__detach_vcache(&this->vcache);
-			tell_waiter(this);
+			wake_up_all(&this->waiters);
+			if (this->filp)
+				send_sigio(&this->filp->f_owner, this->fd, POLL_IN);
 			ret++;
 			if (ret >= num)
 				break;
@@ -189,7 +182,7 @@ static int futex_wake(unsigned long uaddr, int offset, int num)
 	}
 
 	unlock_futex_mm();
-	unpin_page(page);
+	put_page(page);
 
 	return ret;
 }
@@ -208,7 +201,9 @@ static void futex_vcache_callback(vcache_t *vcache, struct page *new_page)
 	spin_lock(&futex_lock);
 
 	if (!list_empty(&q->list)) {
+		put_page(q->page);
 		q->page = new_page;
+		__pin_page_atomic(new_page);
 		list_del(&q->list);
 		list_add_tail(&q->list, head);
 	}
@@ -216,6 +211,65 @@ static void futex_vcache_callback(vcache_t *vcache, struct page *new_page)
 	spin_unlock(&futex_lock);
 }
 
+/*
+ * Requeue all waiters hashed on one physical page to another
+ * physical page.
+ */
+static inline int futex_requeue(unsigned long uaddr1, int offset1,
+	unsigned long uaddr2, int offset2, int nr_wake, int nr_requeue)
+{
+	struct list_head *i, *next, *head1, *head2;
+	struct page *page1 = NULL, *page2 = NULL;
+	int ret = 0;
+
+	lock_futex_mm();
+
+	page1 = __pin_page(uaddr1 - offset1);
+	if (!page1)
+		goto out;
+	page2 = __pin_page(uaddr2 - offset2);
+	if (!page2)
+		goto out;
+
+	head1 = hash_futex(page1, offset1);
+	head2 = hash_futex(page2, offset2);
+
+	list_for_each_safe(i, next, head1) {
+		struct futex_q *this = list_entry(i, struct futex_q, list);
+
+		if (this->page == page1 && this->offset == offset1) {
+			list_del_init(i);
+			__detach_vcache(&this->vcache);
+			if (++ret <= nr_wake) {
+				wake_up_all(&this->waiters);
+				if (this->filp)
+					send_sigio(&this->filp->f_owner,
+							this->fd, POLL_IN);
+			} else {
+				put_page(this->page);
+				__pin_page_atomic (page2);
+				list_add_tail(i, head2);
+				__attach_vcache(&this->vcache, uaddr2,
+					current->mm, futex_vcache_callback);
+				this->offset = offset2;
+				this->page = page2;
+				if (ret - nr_wake >= nr_requeue)
+					break;
+			}
+		}
+	}
+
+out:
+	unlock_futex_mm();
+
+	if (page1)
+		put_page(page1);
+	if (page2)
+		put_page(page2);
+
+	return ret;
+}
+
 static inline void __queue_me(struct futex_q *q, struct page *page,
 				unsigned long uaddr, int offset,
 				int fd, struct file *filp)
@@ -252,7 +306,7 @@ static inline int unqueue_me(struct futex_q *q)
 	return ret;
 }
 
-static int futex_wait(unsigned long uaddr,
+static inline int futex_wait(unsigned long uaddr,
 		      int offset,
 		      int val,
 		      unsigned long time)
@@ -273,14 +327,17 @@ static int futex_wait(unsigned long uaddr,
 	}
 	__queue_me(&q, page, uaddr, offset, -1, NULL);
 
-	unlock_futex_mm();
-
-	/* Page is pinned, but may no longer be in this address space. */
+	/*
+	 * Page is pinned, but may no longer be in this address space.
+	 * It cannot schedule, so we access it with the spinlock held.
+	 */
 	if (get_user(curval, (int *)uaddr) != 0) {
+		unlock_futex_mm();
 		ret = -EFAULT;
 		goto out;
 	}
 	if (curval != val) {
+		unlock_futex_mm();
 		ret = -EWOULDBLOCK;
 		goto out;
 	}
@@ -288,13 +345,15 @@ static int futex_wait(unsigned long uaddr,
 	 * The get_user() above might fault and schedule so we
 	 * cannot just set TASK_INTERRUPTIBLE state when queueing
 	 * ourselves into the futex hash. This code thus has to
-	 * rely on the FUTEX_WAKE code doing a wakeup after removing
+	 * rely on the futex_wake() code doing a wakeup after removing
 	 * the waiter from the list.
 	 */
 	add_wait_queue(&q.waiters, &wait);
 	set_current_state(TASK_INTERRUPTIBLE);
-	if (!list_empty(&q.list))
+	if (!list_empty(&q.list)) {
+		unlock_futex_mm();
 		time = schedule_timeout(time);
+	}
 	set_current_state(TASK_RUNNING);
 	/*
 	 * NOTE: we don't remove ourselves from the waitqueue because
@@ -310,7 +369,7 @@ out:
 	/* Were we woken up anyway? */
 	if (!unqueue_me(&q))
 		ret = 0;
-	unpin_page(page);
+	put_page(q.page);
 
 	return ret;
 }
@@ -320,7 +379,7 @@ static int futex_close(struct inode *inode, struct file *filp)
 	struct futex_q *q = filp->private_data;
 
 	unqueue_me(q);
-	unpin_page(q->page);
+	put_page(q->page);
 	kfree(filp->private_data);
 	return 0;
 }
@@ -416,11 +475,12 @@ static int futex_fd(unsigned long uaddr, int offset, int signal)
 	page = NULL;
 out:
 	if (page)
-		unpin_page(page);
+		put_page(page);
 	return ret;
 }
 
-long do_futex(unsigned long uaddr, int op, int val, unsigned long timeout)
+long do_futex(unsigned long uaddr, int op, int val, unsigned long timeout,
+		unsigned long uaddr2, int val2)
 {
 	unsigned long pos_in_page;
 	int ret;
@@ -442,23 +502,45 @@ long do_futex(unsigned long uaddr, int op, int val, unsigned long timeout)
 		/* non-zero val means F_SETOWN(getpid()) & F_SETSIG(val) */
 		ret = futex_fd(uaddr, pos_in_page, val);
 		break;
+	case FUTEX_REQUEUE:
+	{
+		unsigned long pos_in_page2 = uaddr2 % PAGE_SIZE;
+
+		/* Must be "naturally" aligned */
+		if (pos_in_page2 % sizeof(u32))
+			return -EINVAL;
+
+		ret = futex_requeue(uaddr, pos_in_page, uaddr2, pos_in_page2,
+				    val, val2);
+		break;
+	}
 	default:
-		ret = -EINVAL;
+		ret = -ENOSYS;
 	}
 	return ret;
 }
 
-asmlinkage long sys_futex(u32 __user *uaddr, int op, int val, struct timespec __user *utime)
+
+asmlinkage long sys_futex(u32 __user *uaddr, int op, int val,
+			  struct timespec __user *utime, u32 __user *uaddr2)
 {
 	struct timespec t;
 	unsigned long timeout = MAX_SCHEDULE_TIMEOUT;
+	int val2 = 0;
 
 	if ((op == FUTEX_WAIT) && utime) {
 		if (copy_from_user(&t, utime, sizeof(t)) != 0)
 			return -EFAULT;
 		timeout = timespec_to_jiffies(&t) + 1;
 	}
-	return do_futex((unsigned long)uaddr, op, val, timeout);
+	/*
+	 * requeue parameter in 'utime' if op == FUTEX_REQUEUE.
+	 */
+	if (op == FUTEX_REQUEUE)
+		val2 = (int) utime;
+
+	return do_futex((unsigned long)uaddr, op, val, timeout,
+			(unsigned long)uaddr2, val2);
 }
 
 static struct super_block *
diff --git a/kernel/signal.c b/kernel/signal.c
index d15a55ec9e5e..d1f02ea4b7b1 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -336,7 +336,7 @@ void __exit_signal(struct task_struct *tsk)
 		 * If there is any task waiting for the group exit
 		 * then notify it:
 		 */
-		if (sig->group_exit_task && atomic_read(&sig->count) <= 2) {
+		if (sig->group_exit_task && atomic_read(&sig->count) == sig->notify_count) {
 			wake_up_process(sig->group_exit_task);
 			sig->group_exit_task = NULL;
 		}
@@ -1346,6 +1346,9 @@ do_notify_parent_cldstop(struct task_struct *tsk, struct task_struct *parent)
 	spin_unlock_irqrestore(&sighand->siglock, flags);
 }
 
+
+#ifndef HAVE_ARCH_GET_SIGNAL_TO_DELIVER
+
 static void
 finish_stop(int stop_count)
 {
@@ -1460,9 +1463,6 @@ do_signal_stop(int signr)
 	finish_stop(stop_count);
 }
 
-
-#ifndef HAVE_ARCH_GET_SIGNAL_TO_DELIVER
-
 /*
  * Do appropriate magic when group_stop_count > 0.
  * We return nonzero if we stopped, after releasing the siglock.
diff --git a/kernel/sys.c b/kernel/sys.c
index 8e7de84e1df5..5c2c439ae6bc 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -226,6 +226,11 @@ cond_syscall(sys_shutdown)
 cond_syscall(sys_sendmsg)
 cond_syscall(sys_recvmsg)
 cond_syscall(sys_socketcall)
+cond_syscall(sys_futex)
+cond_syscall(compat_sys_futex)
+cond_syscall(sys_epoll_create)
+cond_syscall(sys_epoll_ctl)
+cond_syscall(sys_epoll_wait)
 
 static int set_one_prio(struct task_struct *p, int niceval, int error)
 {