Add down_timeout and change ACPI to use it

ACPI currently emulates a timeout for semaphores with calls to
down_trylock and sleep.  This produces horrible behaviour in terms of
fairness and excessive wakeups.  Now that we have a unified semaphore
implementation, adding a real down_trylock is almost trivial.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>

3 files changed, 62 insertions, 75 deletions

diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index a697fb6cf050..a498a6cc68fe 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -4,6 +4,8 @@
  *  Copyright (C) 2000       Andrew Henroid
  *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
  *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *  Copyright (c) 2008 Intel Corporation
+ *   Author: Matthew Wilcox <willy@linux.intel.com>
  *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
@@ -37,15 +39,18 @@
 #include <linux/workqueue.h>
 #include <linux/nmi.h>
 #include <linux/acpi.h>
-#include <acpi/acpi.h>
-#include <asm/io.h>
-#include <acpi/acpi_bus.h>
-#include <acpi/processor.h>
-#include <asm/uaccess.h>
-
 #include <linux/efi.h>
 #include <linux/ioport.h>
 #include <linux/list.h>
+#include <linux/jiffies.h>
+#include <linux/semaphore.h>
+
+#include <asm/io.h>
+#include <asm/uaccess.h>
+
+#include <acpi/acpi.h>
+#include <acpi/acpi_bus.h>
+#include <acpi/processor.h>
 
 #define _COMPONENT              ACPI_OS_SERVICES
 ACPI_MODULE_NAME("osl");
@@ -764,7 +769,6 @@ acpi_os_create_semaphore(u32 max_units, u32 initial_units, acpi_handle * handle)
 {
         struct semaphore *sem = NULL;
 
-
         sem = acpi_os_allocate(sizeof(struct semaphore));
         if (!sem)
                 return AE_NO_MEMORY;
@@ -791,12 +795,12 @@ acpi_status acpi_os_delete_semaphore(acpi_handle handle)
 {
         struct semaphore *sem = (struct semaphore *)handle;
 
-
         if (!sem)
                 return AE_BAD_PARAMETER;
 
         ACPI_DEBUG_PRINT((ACPI_DB_MUTEX, "Deleting semaphore[%p].\n", handle));
 
+        BUG_ON(!list_empty(&sem->wait_list));
         kfree(sem);
         sem = NULL;
 
@@ -804,21 +808,15 @@ acpi_status acpi_os_delete_semaphore(acpi_handle handle)
 }
 
 /*
- * TODO: The kernel doesn't have a 'down_timeout' function -- had to
- * improvise.  The process is to sleep for one scheduler quantum
- * until the semaphore becomes available.  Downside is that this
- * may result in starvation for timeout-based waits when there's
- * lots of semaphore activity.
- *
  * TODO: Support for units > 1?
  */
 acpi_status acpi_os_wait_semaphore(acpi_handle handle, u32 units, u16 timeout)
 {
         acpi_status status = AE_OK;
         struct semaphore *sem = (struct semaphore *)handle;
+        long jiffies;
         int ret = 0;
 
-
         if (!sem || (units < 1))
                 return AE_BAD_PARAMETER;
 
@@ -828,58 +826,14 @@ acpi_status acpi_os_wait_semaphore(acpi_handle handle, u32 units, u16 timeout)
         ACPI_DEBUG_PRINT((ACPI_DB_MUTEX, "Waiting for semaphore[%p|%d|%d]\n",
                           handle, units, timeout));
 
-        /*
-         * This can be called during resume with interrupts off.
-         * Like boot-time, we should be single threaded and will
-         * always get the lock if we try -- timeout or not.
-         * If this doesn't succeed, then we will oops courtesy of
-         * might_sleep() in down().
-         */
-        if (!down_trylock(sem))
-                return AE_OK;
-
-        switch (timeout) {
-                /*
-                 * No Wait:
-                 * --------
-                 * A zero timeout value indicates that we shouldn't wait - just
-                 * acquire the semaphore if available otherwise return AE_TIME
-                 * (a.k.a. 'would block').
-                 */
-        case 0:
-                if (down_trylock(sem))
-                        status = AE_TIME;
-                break;
-
-                /*
-                 * Wait Indefinitely:
-                 * ------------------
-                 */
-        case ACPI_WAIT_FOREVER:
-                down(sem);
-                break;
-
-                /*
-                 * Wait w/ Timeout:
-                 * ----------------
-                 */
-        default:
-                // TODO: A better timeout algorithm?
-                {
-                        int i = 0;
-                        static const int quantum_ms = 1000 / HZ;
-
-                        ret = down_trylock(sem);
-                        for (i = timeout; (i > 0 && ret != 0); i -= quantum_ms) {
-                                schedule_timeout_interruptible(1);
-                                ret = down_trylock(sem);
-                        }
-
-                        if (ret != 0)
-                                status = AE_TIME;
-                }
-                break;
-        }
+        if (timeout == ACPI_WAIT_FOREVER)
+                jiffies = MAX_SCHEDULE_TIMEOUT;
+        else
+                jiffies = msecs_to_jiffies(timeout);
+        
+        ret = down_timeout(sem, jiffies);
+        if (ret)
+                status = AE_TIME;
 
         if (ACPI_FAILURE(status)) {
                 ACPI_DEBUG_PRINT((ACPI_DB_MUTEX,
@@ -902,7 +856,6 @@ acpi_status acpi_os_signal_semaphore(acpi_handle handle, u32 units)
 {
         struct semaphore *sem = (struct semaphore *)handle;
 
-
         if (!sem || (units < 1))
                 return AE_BAD_PARAMETER;
 
diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index 88f2a28cc0f1..a107aebd9148 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -75,6 +75,12 @@ extern int __must_check down_killable(struct semaphore *sem);
 extern int __must_check down_trylock(struct semaphore *sem);
 
 /*
+ * As down(), except this function will return -ETIME if it fails to
+ * acquire the semaphore within the specified number of jiffies.
+ */
+extern int __must_check down_timeout(struct semaphore *sem, long jiffies);
+
+/*
  * Release the semaphore.  Unlike mutexes, up() may be called from any
  * context and even by tasks which have never called down().
  */
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index 2da2aed950f3..5a12a8558982 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -35,6 +35,7 @@
 static noinline void __down(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
 static noinline int __down_killable(struct semaphore *sem);
+static noinline int __down_timeout(struct semaphore *sem, long jiffies);
 static noinline void __up(struct semaphore *sem);
 
 void down(struct semaphore *sem)
@@ -104,6 +105,20 @@ int down_trylock(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down_trylock);
 
+int down_timeout(struct semaphore *sem, long jiffies)
+{
+        unsigned long flags;
+        int result = 0;
+
+        spin_lock_irqsave(&sem->lock, flags);
+        if (unlikely(sem->count-- <= 0))
+                result = __down_timeout(sem, jiffies);
+        spin_unlock_irqrestore(&sem->lock, flags);
+
+        return result;
+}
+EXPORT_SYMBOL(down_timeout);
+
 void up(struct semaphore *sem)
 {
         unsigned long flags;
@@ -142,10 +157,12 @@ static noinline void __sched __up_down_common(struct semaphore *sem)
 }
 
 /*
- * Because this function is inlined, the 'state' parameter will be constant,
- * and thus optimised away by the compiler.
+ * Because this function is inlined, the 'state' parameter will be
+ * constant, and thus optimised away by the compiler.  Likewise the
+ * 'timeout' parameter for the cases without timeouts.
  */
-static inline int __sched __down_common(struct semaphore *sem, long state)
+static inline int __sched __down_common(struct semaphore *sem, long state,
+                                                                long timeout)
 {
         int result = 0;
         struct task_struct *task = current;
@@ -160,14 +177,20 @@ static inline int __sched __down_common(struct semaphore *sem, long state)
                         goto interrupted;
                 if (state == TASK_KILLABLE && fatal_signal_pending(task))
                         goto interrupted;
+                if (timeout <= 0)
+                        goto timed_out;
                 __set_task_state(task, state);
                 spin_unlock_irq(&sem->lock);
-                schedule();
+                timeout = schedule_timeout(timeout);
                 spin_lock_irq(&sem->lock);
                 if (waiter.up)
                         goto woken;
         }
 
+ timed_out:
+        list_del(&waiter.list);
+        result = -ETIME;
+        goto woken;
  interrupted:
         list_del(&waiter.list);
         result = -EINTR;
@@ -187,17 +210,22 @@ static inline int __sched __down_common(struct semaphore *sem, long state)
 
 static noinline void __sched __down(struct semaphore *sem)
 {
-        __down_common(sem, TASK_UNINTERRUPTIBLE);
+        __down_common(sem, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
 }
 
 static noinline int __sched __down_interruptible(struct semaphore *sem)
 {
-        return __down_common(sem, TASK_INTERRUPTIBLE);
+        return __down_common(sem, TASK_INTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
 }
 
 static noinline int __sched __down_killable(struct semaphore *sem)
 {
-        return __down_common(sem, TASK_KILLABLE);
+        return __down_common(sem, TASK_KILLABLE, MAX_SCHEDULE_TIMEOUT);
+}
+
+static noinline int __sched __down_timeout(struct semaphore *sem, long jiffies)
+{
+        return __down_common(sem, TASK_UNINTERRUPTIBLE, jiffies);
 }
 
 static noinline void __sched __up(struct semaphore *sem)