1 files changed, 227 insertions, 247 deletions

diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c
index 2ffda012385..1d8505b1e29 100644
--- a/arch/um/kernel/irq.c
+++ b/arch/um/kernel/irq.c
@@ -1,90 +1,36 @@
-/* 
- * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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/config.h"
-#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 "misc_constants.h"
+#include <linux/cpumask.h>
+#include <linux/hardirq.h>
+#include <linux/interrupt.h>
+#include <linux/kernel_stat.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <as-layout.h>
+#include <kern_util.h>
+#include <os.h>
 
 /*
- * Generic, controller-independent functions:
+ * This list is accessed under irq_lock, except in sigio_handler,
+ * where it is safe from being modified.  IRQ handlers won't change it -
+ * if an IRQ source has vanished, it will be freed by free_irqs just
+ * before returning from sigio_handler.  That will process a separate
+ * list of irqs to free, with its own locking, coming back here to
+ * remove list elements, taking the irq_lock to do so.
  */
-
-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].handler->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;
-}
-
-struct irq_fd *active_fds = NULL;
+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)
+void sigio_handler(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs)
 {
 	struct irq_fd *irq_fd;
 	int n;
@@ -95,11 +41,13 @@ void sigio_handler(int sig, union uml_pt_regs *regs)
 	while (1) {
 		n = os_waiting_for_events(active_fds);
 		if (n <= 0) {
-			if(n == -EINTR) continue;
+			if (n == -EINTR)
+				continue;
 			else break;
 		}
 
-		for (irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next) {
+		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);
@@ -110,73 +58,47 @@ void sigio_handler(int sig, union uml_pt_regs *regs)
 	free_irqs();
 }
 
-static void maybe_sigio_broken(int fd, int type)
-{
-	if (os_isatty(fd)) {
-		if ((type == IRQ_WRITE) && !pty_output_sigio) {
-			write_sigio_workaround();
-			add_sigio_fd(fd, 0);
-		} else if ((type == IRQ_READ) && !pty_close_sigio) {
-			write_sigio_workaround();
-			add_sigio_fd(fd, 1);
-		}
-	}
-}
+static DEFINE_SPINLOCK(irq_lock);
 
-int activate_fd(int irq, int fd, int type, void *dev_id)
+static 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;
+	int events, err, n;
 
-	pid = os_getpid();
-	err = os_set_fd_async(fd, pid);
+	err = os_set_fd_async(fd);
 	if (err < 0)
 		goto out;
 
-	new_fd = um_kmalloc(sizeof(*new_fd));
 	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;
+	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 } );
 
-	/* Critical section - locked by a spinlock because this stuff can
-	 * be changed from interrupt handlers.  The stuff above is done
-	 * outside the lock because it allocates memory.
-	 */
-
-	/* Actually, it only looks like it can be called from interrupt
-	 * context.  The culprit is reactivate_fd, which calls
-	 * maybe_sigio_broken, which calls write_sigio_workaround,
-	 * which calls activate_fd.  However, write_sigio_workaround should
-	 * only be called once, at boot time.  That would make it clear that
-	 * this is called only from process context, and can be locked with
-	 * a semaphore.
-	 */
-	flags = irq_lock();
+	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);
+			printk(KERN_ERR "Registering fd %d twice\n", fd);
+			printk(KERN_ERR "Irqs : %d, %d\n", irq_fd->irq, irq);
+			printk(KERN_ERR "Ids : 0x%p, 0x%p\n", irq_fd->id,
+			       dev_id);
 			goto out_unlock;
 		}
 	}
 
-	/*-------------*/
 	if (type == IRQ_WRITE)
 		fd = -1;
 
@@ -188,7 +110,8 @@ int activate_fd(int irq, int fd, int type, void *dev_id)
 		if (n == 0)
 			break;
 
-		/* n > 0
+		/*
+		 * 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.
@@ -199,45 +122,44 @@ int activate_fd(int irq, int fd, int type, void *dev_id)
 		 * so we will not be able to put new pollfd struct to pollfds
 		 * then we free the buffer tmp_fds and try again.
 		 */
-		irq_unlock(flags);
+		spin_unlock_irqrestore(&irq_lock, flags);
 		kfree(tmp_pfd);
-		tmp_pfd = NULL;
 
-		tmp_pfd = um_kmalloc(n);
+		tmp_pfd = kmalloc(n, GFP_KERNEL);
 		if (tmp_pfd == NULL)
 			goto out_kfree;
 
-		flags = irq_lock();
+		spin_lock_irqsave(&irq_lock, flags);
 	}
-	/*-------------*/
 
 	*last_irq_ptr = new_fd;
 	last_irq_ptr = &new_fd->next;
 
-	irq_unlock(flags);
+	spin_unlock_irqrestore(&irq_lock, flags);
 
-	/* This calls activate_fd, so it has to be outside the critical
+	/*
+	 * This calls activate_fd, so it has to be outside the critical
 	 * section.
 	 */
-	maybe_sigio_broken(fd, type);
+	maybe_sigio_broken(fd, (type == IRQ_READ));
 
-	return(0);
+	return 0;
 
  out_unlock:
-	irq_unlock(flags);
+	spin_unlock_irqrestore(&irq_lock, flags);
  out_kfree:
 	kfree(new_fd);
  out:
-	return(err);
+	return err;
 }
 
 static void free_irq_by_cb(int (*test)(struct irq_fd *, void *), void *arg)
 {
 	unsigned long flags;
 
-	flags = irq_lock();
+	spin_lock_irqsave(&irq_lock, flags);
 	os_free_irq_by_cb(test, arg, active_fds, &last_irq_ptr);
-	irq_unlock(flags);
+	spin_unlock_irqrestore(&irq_lock, flags);
 }
 
 struct irq_and_dev {
@@ -252,7 +174,7 @@ static int same_irq_and_dev(struct irq_fd *irq, void *d)
 	return ((irq->irq == data->irq) && (irq->id == data->dev));
 }
 
-void free_irq_by_irq_and_dev(unsigned int irq, void *dev)
+static 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 });
@@ -270,6 +192,7 @@ void free_irq_by_fd(int fd)
 	free_irq_by_cb(same_fd, &fd);
 }
 
+/* Must be called with irq_lock held */
 static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out)
 {
 	struct irq_fd *irq;
@@ -282,13 +205,14 @@ static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out)
 		i++;
 	}
 	if (irq == NULL) {
-		printk("find_irq_by_fd doesn't have descriptor %d\n", fd);
+		printk(KERN_ERR "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,
+		printk(KERN_ERR "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;
@@ -304,19 +228,16 @@ void reactivate_fd(int fd, int irqnum)
 	unsigned long flags;
 	int i;
 
-	flags = irq_lock();
+	spin_lock_irqsave(&irq_lock, flags);
 	irq = find_irq_by_fd(fd, irqnum, &i);
 	if (irq == NULL) {
-		irq_unlock(flags);
+		spin_unlock_irqrestore(&irq_lock, flags);
 		return;
 	}
 	os_set_pollfd(i, irq->fd);
-	irq_unlock(flags);
+	spin_unlock_irqrestore(&irq_lock, flags);
 
-	/* This calls activate_fd, so it has to be outside the critical
-	 * section.
-	 */
-	maybe_sigio_broken(fd, irq->type);
+	add_sigio_fd(fd);
 }
 
 void deactivate_fd(int fd, int irqnum)
@@ -325,15 +246,26 @@ void deactivate_fd(int fd, int irqnum)
 	unsigned long flags;
 	int i;
 
-	flags = irq_lock();
+	spin_lock_irqsave(&irq_lock, flags);
 	irq = find_irq_by_fd(fd, irqnum, &i);
-	if (irq == NULL)
-		goto out;
+	if (irq == NULL) {
+		spin_unlock_irqrestore(&irq_lock, flags);
+		return;
+	}
+
 	os_set_pollfd(i, -1);
- out:
-	irq_unlock(flags);
+	spin_unlock_irqrestore(&irq_lock, flags);
+
+	ignore_sigio_fd(fd);
 }
+EXPORT_SYMBOL(deactivate_fd);
 
+/*
+ * Called just before shutdown in order to provide a clean exec
+ * environment in case the system is rebooting.  No locking because
+ * that would cause a pointless shutdown hang if something hadn't
+ * released the lock.
+ */
 int deactivate_all_fds(void)
 {
 	struct irq_fd *irq;
@@ -350,144 +282,192 @@ int deactivate_all_fds(void)
 	return 0;
 }
 
-void forward_interrupts(int pid)
-{
-	struct irq_fd *irq;
-	unsigned long flags;
-	int err;
-
-	flags = irq_lock();
-	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;
-	}
-	irq_unlock(flags);
-}
-
 /*
- * do_IRQ handles all normal device IRQ's (the special
+ * do_IRQ handles all normal device IRQs (the special
  * SMP cross-CPU interrupts have their own specific
  * handlers).
  */
-unsigned int do_IRQ(int irq, union uml_pt_regs *regs)
+unsigned int do_IRQ(int irq, struct uml_pt_regs *regs)
 {
-       irq_enter();
-       __do_IRQ(irq, (struct pt_regs *)regs);
-       irq_exit();
-       return 1;
+	struct pt_regs *old_regs = set_irq_regs((struct pt_regs *)regs);
+	irq_enter();
+	generic_handle_irq(irq);
+	irq_exit();
+	set_irq_regs(old_regs);
+	return 1;
 }
 
+void um_free_irq(unsigned int irq, void *dev)
+{
+	free_irq_by_irq_and_dev(irq, dev);
+	free_irq(irq, dev);
+}
+EXPORT_SYMBOL(um_free_irq);
+
 int um_request_irq(unsigned int irq, int fd, int type,
-		   irqreturn_t (*handler)(int, void *, struct pt_regs *),
+		   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)
+	if (fd != -1) {
 		err = activate_fd(irq, fd, type, dev_id);
-	return err;
-}
-EXPORT_SYMBOL(um_request_irq);
-EXPORT_SYMBOL(reactivate_fd);
-
-static DEFINE_SPINLOCK(irq_spinlock);
-
-unsigned long irq_lock(void)
-{
-	unsigned long flags;
+		if (err)
+			return err;
+	}
 
-	spin_lock_irqsave(&irq_spinlock, flags);
-	return flags;
+	return request_irq(irq, handler, irqflags, devname, dev_id);
 }
 
-void irq_unlock(unsigned long flags)
-{
-	spin_unlock_irqrestore(&irq_spinlock, flags);
-}
+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)
+/*
+ * irq_chip must define at least enable/disable and ack when
+ * the edge handler is used.
+ */
+static void dummy(struct irq_data *d)
 {
 }
 
 /* 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 irq_chip normal_irq_type = {
+	.name = "SIGIO",
+	.irq_disable = dummy,
+	.irq_enable = dummy,
+	.irq_ack = dummy,
+	.irq_mask = dummy,
+	.irq_unmask = 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
+static struct irq_chip SIGVTALRM_irq_type = {
+	.name = "SIGVTALRM",
+	.irq_disable = dummy,
+	.irq_enable = dummy,
+	.irq_ack = dummy,
+	.irq_mask = dummy,
+	.irq_unmask = 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].handler = &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].handler = &normal_irq_type;
-		enable_irq(i);
-	}
+	irq_set_chip_and_handler(TIMER_IRQ, &SIGVTALRM_irq_type, handle_edge_irq);
+
+	for (i = 1; i < NR_IRQS; i++)
+		irq_set_chip_and_handler(i, &normal_irq_type, handle_edge_irq);
 }
 
-int init_aio_irq(int irq, char *name, irqreturn_t (*handler)(int, void *,
-							     struct pt_regs *))
-{
-	int fds[2], err;
+/*
+ * IRQ stack entry and exit:
+ *
+ * Unlike i386, UML doesn't receive IRQs on the normal kernel stack
+ * and switch over to the IRQ stack after some preparation.  We use
+ * sigaltstack to receive signals on a separate stack from the start.
+ * These two functions make sure the rest of the kernel won't be too
+ * upset by being on a different stack.  The IRQ stack has a
+ * thread_info structure at the bottom so that current et al continue
+ * to work.
+ *
+ * to_irq_stack copies the current task's thread_info to the IRQ stack
+ * thread_info and sets the tasks's stack to point to the IRQ stack.
+ *
+ * from_irq_stack copies the thread_info struct back (flags may have
+ * been modified) and resets the task's stack pointer.
+ *
+ * Tricky bits -
+ *
+ * What happens when two signals race each other?  UML doesn't block
+ * signals with sigprocmask, SA_DEFER, or sa_mask, so a second signal
+ * could arrive while a previous one is still setting up the
+ * thread_info.
+ *
+ * There are three cases -
+ *     The first interrupt on the stack - sets up the thread_info and
+ * handles the interrupt
+ *     A nested interrupt interrupting the copying of the thread_info -
+ * can't handle the interrupt, as the stack is in an unknown state
+ *     A nested interrupt not interrupting the copying of the
+ * thread_info - doesn't do any setup, just handles the interrupt
+ *
+ * The first job is to figure out whether we interrupted stack setup.
+ * This is done by xchging the signal mask with thread_info->pending.
+ * If the value that comes back is zero, then there is no setup in
+ * progress, and the interrupt can be handled.  If the value is
+ * non-zero, then there is stack setup in progress.  In order to have
+ * the interrupt handled, we leave our signal in the mask, and it will
+ * be handled by the upper handler after it has set up the stack.
+ *
+ * Next is to figure out whether we are the outer handler or a nested
+ * one.  As part of setting up the stack, thread_info->real_thread is
+ * set to non-NULL (and is reset to NULL on exit).  This is the
+ * nesting indicator.  If it is non-NULL, then the stack is already
+ * set up and the handler can run.
+ */
 
-	err = os_pipe(fds, 1, 1);
-	if (err) {
-		printk("init_aio_irq - os_pipe failed, err = %d\n", -err);
-		goto out;
+static unsigned long pending_mask;
+
+unsigned long to_irq_stack(unsigned long *mask_out)
+{
+	struct thread_info *ti;
+	unsigned long mask, old;
+	int nested;
+
+	mask = xchg(&pending_mask, *mask_out);
+	if (mask != 0) {
+		/*
+		 * If any interrupts come in at this point, we want to
+		 * make sure that their bits aren't lost by our
+		 * putting our bit in.  So, this loop accumulates bits
+		 * until xchg returns the same value that we put in.
+		 * When that happens, there were no new interrupts,
+		 * and pending_mask contains a bit for each interrupt
+		 * that came in.
+		 */
+		old = *mask_out;
+		do {
+			old |= mask;
+			mask = xchg(&pending_mask, old);
+		} while (mask != old);
+		return 1;
 	}
 
-	err = um_request_irq(irq, fds[0], IRQ_READ, handler,
-			     SA_INTERRUPT | SA_SAMPLE_RANDOM, name,
-			     (void *) (long) fds[0]);
-	if (err) {
-		printk("init_aio_irq - : um_request_irq failed, err = %d\n",
-		       err);
-		goto out_close;
+	ti = current_thread_info();
+	nested = (ti->real_thread != NULL);
+	if (!nested) {
+		struct task_struct *task;
+		struct thread_info *tti;
+
+		task = cpu_tasks[ti->cpu].task;
+		tti = task_thread_info(task);
+
+		*ti = *tti;
+		ti->real_thread = tti;
+		task->stack = ti;
 	}
 
-	err = fds[1];
-	goto out;
+	mask = xchg(&pending_mask, 0);
+	*mask_out |= mask | nested;
+	return 0;
+}
 
- out_close:
-	os_close_file(fds[0]);
-	os_close_file(fds[1]);
- out:
-	return err;
+unsigned long from_irq_stack(int nested)
+{
+	struct thread_info *ti, *to;
+	unsigned long mask;
+
+	ti = current_thread_info();
+
+	pending_mask = 1;
+
+	to = ti->real_thread;
+	current->stack = to;
+	ti->real_thread = NULL;
+	*to = *ti;
+
+	mask = xchg(&pending_mask, 0);
+	return mask & ~1;
 }
+