]> git.saurik.com Git - apple/xnu.git/blobdiff - osfmk/kern/syscall_subr.c
xnu-3248.40.184.tar.gz
[apple/xnu.git] / osfmk / kern / syscall_subr.c
index dd4716262c6fe4cf618e6ba937d60944c1fa16df..1ebf39e4f9d4f405b1a9e9a0f41f3624b0595bb9 100644 (file)
@@ -1,16 +1,19 @@
 /*
- * Copyright (c) 2000 Apple Computer, Inc. All rights reserved.
+ * Copyright (c) 2000-2009 Apple Inc. All rights reserved.
  *
- * @APPLE_LICENSE_HEADER_START@
- * 
- * Copyright (c) 1999-2003 Apple Computer, Inc.  All Rights Reserved.
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
  * 
  * This file contains Original Code and/or Modifications of Original Code
  * as defined in and that are subject to the Apple Public Source License
  * Version 2.0 (the 'License'). You may not use this file except in
- * compliance with the License. Please obtain a copy of the License at
- * http://www.opensource.apple.com/apsl/ and read it before using this
- * file.
+ * compliance with the License. The rights granted to you under the License
+ * may not be used to create, or enable the creation or redistribution of,
+ * unlawful or unlicensed copies of an Apple operating system, or to
+ * circumvent, violate, or enable the circumvention or violation of, any
+ * terms of an Apple operating system software license agreement.
+ * 
+ * Please obtain a copy of the License at
+ * http://www.opensource.apple.com/apsl/ and read it before using this file.
  * 
  * The Original Code and all software distributed under the License are
  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
@@ -20,7 +23,7 @@
  * Please see the License for the specific language governing rights and
  * limitations under the License.
  * 
- * @APPLE_LICENSE_HEADER_END@
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  */
 /*
  * @OSF_COPYRIGHT@
@@ -58,7 +61,6 @@
 #include <ipc/ipc_port.h>
 #include <ipc/ipc_space.h>
 #include <kern/counters.h>
-#include <kern/etap_macros.h>
 #include <kern/ipc_kobject.h>
 #include <kern/processor.h>
 #include <kern/sched.h>
 #include <kern/spl.h>
 #include <kern/task.h>
 #include <kern/thread.h>
-#include <kern/ast.h>
 #include <mach/policy.h>
 
 #include <kern/syscall_subr.h>
 #include <mach/mach_host_server.h>
 #include <mach/mach_syscalls.h>
+#include <sys/kdebug.h>
+
+#ifdef MACH_BSD
+extern void workqueue_thread_yielded(void);
+extern sched_call_t workqueue_get_sched_callback(void);
+#endif /* MACH_BSD */
+
+
+/* Called from commpage to take a delayed preemption when exiting
+ * the "Preemption Free Zone" (PFZ).
+ */
+kern_return_t
+pfz_exit(
+__unused       struct pfz_exit_args *args)
+{
+       /* For now, nothing special to do.  We'll pick up the ASTs on kernel exit. */
+
+       return (KERN_SUCCESS);
+}
 
-#include <kern/mk_sp.h>
 
 /*
  *     swtch and swtch_pri both attempt to context switch (logic in
  *     thread_block no-ops the context switch if nothing would happen).
  *     A boolean is returned that indicates whether there is anything
- *     else runnable.
- *
- *     This boolean can be used by a thread waiting on a
- *     lock or condition:  If FALSE is returned, the thread is justified
- *     in becoming a resource hog by continuing to spin because there's
- *     nothing else useful that the processor could do.  If TRUE is
- *     returned, the thread should make one more check on the
- *     lock and then be a good citizen and really suspend.
+ *     else runnable.  That's no excuse to spin, though.
  */
 
-void
+static void
 swtch_continue(void)
 {
        register processor_t    myprocessor;
     boolean_t                          result;
 
-    mp_disable_preemption();
+    disable_preemption();
        myprocessor = current_processor();
-       result =                myprocessor->runq.count > 0                                     ||
-                               myprocessor->processor_set->runq.count > 0;
-       mp_enable_preemption();
+       result = !SCHED(processor_queue_empty)(myprocessor) || rt_runq.count > 0;
+       enable_preemption();
 
        thread_syscall_return(result);
        /*NOTREACHED*/
 }
 
 boolean_t
-swtch(void)
+swtch(
+       __unused struct swtch_args *args)
 {
        register processor_t    myprocessor;
        boolean_t                               result;
 
-       mp_disable_preemption();
+       disable_preemption();
        myprocessor = current_processor();
-       if (            myprocessor->runq.count == 0                            &&
-                       myprocessor->processor_set->runq.count == 0                     ) {
+       if (SCHED(processor_queue_empty)(myprocessor) &&        rt_runq.count == 0) {
                mp_enable_preemption();
 
                return (FALSE);
        }
-       mp_enable_preemption();
+       enable_preemption();
 
        counter(c_swtch_block++);
 
-       thread_block_reason(swtch_continue, AST_YIELD);
+       thread_block_reason((thread_continue_t)swtch_continue, NULL, AST_YIELD);
 
-       mp_disable_preemption();
+       disable_preemption();
        myprocessor = current_processor();
-       result =                myprocessor->runq.count > 0                                     ||
-                               myprocessor->processor_set->runq.count > 0;
-       mp_enable_preemption();
+       result = !SCHED(processor_queue_empty)(myprocessor) || rt_runq.count > 0;
+       enable_preemption();
 
        return (result);
 }
 
-void
+static void
 swtch_pri_continue(void)
 {
        register processor_t    myprocessor;
     boolean_t                          result;
 
-       _mk_sp_thread_depress_abort(current_thread(), FALSE);
+       thread_depress_abort_internal(current_thread());
 
-    mp_disable_preemption();
+    disable_preemption();
        myprocessor = current_processor();
-       result =                myprocessor->runq.count > 0                                     ||
-                               myprocessor->processor_set->runq.count > 0;
+       result = !SCHED(processor_queue_empty)(myprocessor) || rt_runq.count > 0;
        mp_enable_preemption();
 
        thread_syscall_return(result);
@@ -154,38 +163,82 @@ swtch_pri_continue(void)
 
 boolean_t
 swtch_pri(
-       int                             pri)
+__unused       struct swtch_pri_args *args)
 {
        register processor_t    myprocessor;
        boolean_t                               result;
 
-       mp_disable_preemption();
+       disable_preemption();
        myprocessor = current_processor();
-       if (    myprocessor->runq.count == 0                                    &&
-                       myprocessor->processor_set->runq.count == 0                     ) {
+       if (SCHED(processor_queue_empty)(myprocessor) && rt_runq.count == 0) {
                mp_enable_preemption();
 
                return (FALSE);
        }
-       mp_enable_preemption();
+       enable_preemption();
 
        counter(c_swtch_pri_block++);
 
-       _mk_sp_thread_depress_abstime(std_quantum);
+       thread_depress_abstime(thread_depress_time);
 
-       thread_block_reason(swtch_pri_continue, AST_YIELD);
+       thread_block_reason((thread_continue_t)swtch_pri_continue, NULL, AST_YIELD);
 
-       _mk_sp_thread_depress_abort(current_thread(), FALSE);
+       thread_depress_abort_internal(current_thread());
 
-       mp_disable_preemption();
+       disable_preemption();
        myprocessor = current_processor();
-       result =        myprocessor->runq.count > 0                                             ||
-                               myprocessor->processor_set->runq.count > 0;
-       mp_enable_preemption();
+       result = !SCHED(processor_queue_empty)(myprocessor) || rt_runq.count > 0;
+       enable_preemption();
 
        return (result);
 }
 
+static int
+thread_switch_disable_workqueue_sched_callback(void)
+{
+       sched_call_t callback = workqueue_get_sched_callback();
+       thread_t self = current_thread();
+       if (!callback || self->sched_call != callback) {
+               return FALSE;
+       }
+       spl_t s = splsched();
+       thread_lock(self);
+       thread_sched_call(self, NULL);
+       thread_unlock(self);
+       splx(s);
+       return TRUE;
+}
+
+static void
+thread_switch_enable_workqueue_sched_callback(void)
+{
+       sched_call_t callback = workqueue_get_sched_callback();
+       thread_t self = current_thread();
+       spl_t s = splsched();
+       thread_lock(self);
+       thread_sched_call(self, callback);
+       thread_unlock(self);
+       splx(s);
+}
+
+static void
+thread_switch_continue(void)
+{
+       register thread_t       self = current_thread();
+       int                                     option = self->saved.swtch.option;
+       boolean_t                       reenable_workq_callback = self->saved.swtch.reenable_workq_callback;
+
+
+       if (option == SWITCH_OPTION_DEPRESS || option == SWITCH_OPTION_OSLOCK_DEPRESS)
+               thread_depress_abort_internal(self);
+
+       if (reenable_workq_callback)
+               thread_switch_enable_workqueue_sched_callback();
+
+       thread_syscall_return(KERN_SUCCESS);
+       /*NOTREACHED*/
+}
+
 /*
  *     thread_switch:
  *
@@ -193,39 +246,325 @@ swtch_pri(
  */
 kern_return_t
 thread_switch(
-       mach_port_name_t                thread_name,
-       int                                             option,
-       mach_msg_timeout_t              option_time)
+       struct thread_switch_args *args)
 {
-    register thread_t          self = current_thread();
-    register thread_act_t      hint_act = THR_ACT_NULL;
+       thread_t                        thread = THREAD_NULL;
+       thread_t                        self = current_thread();
+       mach_port_name_t                thread_name = args->thread_name;
+       int                                             option = args->option;
+       mach_msg_timeout_t              option_time = args->option_time;
+       uint32_t                                scale_factor = NSEC_PER_MSEC;
+       boolean_t                               reenable_workq_callback = FALSE;
+       boolean_t                               depress_option = FALSE;
+       boolean_t                               wait_option = FALSE;
 
     /*
-     * Process option.
+     * Validate and process option.
      */
     switch (option) {
 
        case SWITCH_OPTION_NONE:
-       case SWITCH_OPTION_DEPRESS:
+               workqueue_thread_yielded();
+               break;
        case SWITCH_OPTION_WAIT:
-           break;
-
+               wait_option = TRUE;
+               workqueue_thread_yielded();
+               break;
+       case SWITCH_OPTION_DEPRESS:
+               depress_option = TRUE;
+               workqueue_thread_yielded();
+               break;
+       case SWITCH_OPTION_DISPATCH_CONTENTION:
+               scale_factor = NSEC_PER_USEC;
+               wait_option = TRUE;
+               if (thread_switch_disable_workqueue_sched_callback())
+                       reenable_workq_callback = TRUE;
+               break;
+       case SWITCH_OPTION_OSLOCK_DEPRESS:
+               depress_option = TRUE;
+               if (thread_switch_disable_workqueue_sched_callback())
+                       reenable_workq_callback = TRUE;
+               break;
+       case SWITCH_OPTION_OSLOCK_WAIT:
+               wait_option = TRUE;
+               if (thread_switch_disable_workqueue_sched_callback())
+                       reenable_workq_callback = TRUE;
+               break;
        default:
            return (KERN_INVALID_ARGUMENT);
     }
 
-    if (thread_name != MACH_PORT_NULL) {
-               ipc_port_t                      port;
+       /*
+        * Translate the port name if supplied.
+        */
+       if (thread_name != MACH_PORT_NULL) {
+               ipc_port_t port;
 
-               if (ipc_port_translate_send(self->top_act->task->itk_space,
-                                                                       thread_name, &port) == KERN_SUCCESS) {
+               if (ipc_port_translate_send(self->task->itk_space,
+                                           thread_name, &port) == KERN_SUCCESS) {
                        ip_reference(port);
                        ip_unlock(port);
 
-                       hint_act = convert_port_to_act(port);
-                       ipc_port_release(port);
+                       thread = convert_port_to_thread(port);
+                       ip_release(port);
+
+                       if (thread == self) {
+                               thread_deallocate(thread);
+                               thread = THREAD_NULL;
+                       }
+               }
+       }
+
+       if (option == SWITCH_OPTION_OSLOCK_DEPRESS || option == SWITCH_OPTION_OSLOCK_WAIT) {
+               if (thread != THREAD_NULL) {
+
+                       if (thread->task != self->task) {
+                               /*
+                                * OSLock boosting only applies to other threads
+                                * in your same task (even if you have a port for
+                                * a thread in another task)
+                                */
+
+                               thread_deallocate(thread);
+                               thread = THREAD_NULL;
+                       } else {
+                               /*
+                                * Attempt to kick the lock owner up to our same IO throttling tier.
+                                * If the thread is currently blocked in throttle_lowpri_io(),
+                                * it will immediately break out.
+                                *
+                                * TODO: SFI break out?
+                                */
+                               int new_policy = proc_get_effective_thread_policy(self, TASK_POLICY_IO);
+
+                               set_thread_iotier_override(thread, new_policy);
+                       }
+               }
+       }
+
+       /*
+        * Try to handoff if supplied.
+        */
+       if (thread != THREAD_NULL) {
+               spl_t s = splsched();
+
+               /* This may return a different thread if the target is pushing on something */
+               thread_t pulled_thread = thread_run_queue_remove_for_handoff(thread);
+
+               KERNEL_DEBUG_CONSTANT(MACHDBG_CODE(DBG_MACH_SCHED,MACH_SCHED_THREAD_SWITCH)|DBG_FUNC_NONE,
+                                     thread_tid(thread), thread->state,
+                                     pulled_thread ? TRUE : FALSE, 0, 0);
+
+               if (pulled_thread != THREAD_NULL) {
+                       /* We can't be dropping the last ref here */
+                       thread_deallocate_safe(thread);
+
+                       if (wait_option)
+                               assert_wait_timeout((event_t)assert_wait_timeout, THREAD_ABORTSAFE,
+                                                   option_time, scale_factor);
+                       else if (depress_option)
+                               thread_depress_ms(option_time);
+
+                       self->saved.swtch.option = option;
+                       self->saved.swtch.reenable_workq_callback = reenable_workq_callback;
+
+                       thread_run(self, (thread_continue_t)thread_switch_continue, NULL, pulled_thread);
+                       /* NOTREACHED */
+                       panic("returned from thread_run!");
+               }
+
+               splx(s);
+
+               thread_deallocate(thread);
+       }
+
+       if (wait_option)
+               assert_wait_timeout((event_t)assert_wait_timeout, THREAD_ABORTSAFE, option_time, scale_factor);
+       else if (depress_option)
+               thread_depress_ms(option_time);
+
+       self->saved.swtch.option = option;
+       self->saved.swtch.reenable_workq_callback = reenable_workq_callback;
+
+       thread_block_reason((thread_continue_t)thread_switch_continue, NULL, AST_YIELD);
+
+       if (depress_option)
+               thread_depress_abort_internal(self);
+
+       if (reenable_workq_callback)
+               thread_switch_enable_workqueue_sched_callback();
+
+    return (KERN_SUCCESS);
+}
+
+/*
+ * Depress thread's priority to lowest possible for the specified interval,
+ * with a value of zero resulting in no timeout being scheduled.
+ */
+void
+thread_depress_abstime(
+       uint64_t                                interval)
+{
+       register thread_t               self = current_thread();
+       uint64_t                                deadline;
+    spl_t                                      s;
+
+    s = splsched();
+    thread_lock(self);
+       if (!(self->sched_flags & TH_SFLAG_DEPRESSED_MASK)) {
+               processor_t             myprocessor = self->last_processor;
+
+               self->sched_pri = DEPRESSPRI;
+
+               KERNEL_DEBUG_CONSTANT(MACHDBG_CODE(DBG_MACH_SCHED, MACH_SCHED_CHANGE_PRIORITY),
+                                     (uintptr_t)thread_tid(self),
+                                     self->base_pri,
+                                     self->sched_pri,
+                                     0, /* eventually, 'reason' */
+                                     0);
+
+               myprocessor->current_pri = self->sched_pri;
+               self->sched_flags |= TH_SFLAG_DEPRESS;
+
+               if (interval != 0) {
+                       clock_absolutetime_interval_to_deadline(interval, &deadline);
+                       if (!timer_call_enter(&self->depress_timer, deadline, TIMER_CALL_USER_CRITICAL))
+                               self->depress_timer_active++;
                }
        }
+       thread_unlock(self);
+    splx(s);
+}
 
-    return _mk_sp_thread_switch(hint_act, option, option_time);
+void
+thread_depress_ms(
+       mach_msg_timeout_t              interval)
+{
+       uint64_t                abstime;
+
+       clock_interval_to_absolutetime_interval(
+                                                       interval, NSEC_PER_MSEC, &abstime);
+       thread_depress_abstime(abstime);
 }
+
+/*
+ *     Priority depression expiration.
+ */
+void
+thread_depress_expire(
+       void                    *p0,
+       __unused void   *p1)
+{
+       thread_t                thread = p0;
+    spl_t                      s;
+
+    s = splsched();
+    thread_lock(thread);
+       if (--thread->depress_timer_active == 0) {
+               thread->sched_flags &= ~TH_SFLAG_DEPRESSED_MASK;
+               thread_recompute_sched_pri(thread, FALSE);
+       }
+    thread_unlock(thread);
+    splx(s);
+}
+
+/*
+ *     Prematurely abort priority depression if there is one.
+ */
+kern_return_t
+thread_depress_abort_internal(
+       thread_t                                thread)
+{
+    kern_return_t                      result = KERN_NOT_DEPRESSED;
+    spl_t                                      s;
+
+    s = splsched();
+    thread_lock(thread);
+       if (!(thread->sched_flags & TH_SFLAG_POLLDEPRESS)) {
+               if (thread->sched_flags & TH_SFLAG_DEPRESSED_MASK) {
+                       thread->sched_flags &= ~TH_SFLAG_DEPRESSED_MASK;
+                       thread_recompute_sched_pri(thread, FALSE);
+                       result = KERN_SUCCESS;
+               }
+
+               if (timer_call_cancel(&thread->depress_timer))
+                       thread->depress_timer_active--;
+       }
+       thread_unlock(thread);
+    splx(s);
+
+    return (result);
+}
+
+void
+thread_poll_yield(
+       thread_t                self)
+{
+       spl_t                   s;
+
+       assert(self == current_thread());
+
+       s = splsched();
+       if (self->sched_mode == TH_MODE_FIXED) {
+               uint64_t                        total_computation, abstime;
+
+               abstime = mach_absolute_time();
+               total_computation = abstime - self->computation_epoch;
+               total_computation += self->computation_metered;
+               if (total_computation >= max_poll_computation) {
+                       processor_t             myprocessor = current_processor();
+                       ast_t                   preempt;
+
+                       thread_lock(self);
+                       if (!(self->sched_flags & TH_SFLAG_DEPRESSED_MASK)) {
+                               self->sched_pri = DEPRESSPRI;
+
+                               KERNEL_DEBUG_CONSTANT(MACHDBG_CODE(DBG_MACH_SCHED, MACH_SCHED_CHANGE_PRIORITY),
+                                                     (uintptr_t)thread_tid(self),
+                                                     self->base_pri,
+                                                     self->sched_pri,
+                                                     0, /* eventually, 'reason' */
+                                                     0);
+
+                               myprocessor->current_pri = self->sched_pri;
+                       }
+                       self->computation_epoch = abstime;
+                       self->computation_metered = 0;
+                       self->sched_flags |= TH_SFLAG_POLLDEPRESS;
+
+                       abstime += (total_computation >> sched_poll_yield_shift);
+                       if (!timer_call_enter(&self->depress_timer, abstime, TIMER_CALL_USER_CRITICAL))
+                               self->depress_timer_active++;
+
+                       if ((preempt = csw_check(myprocessor, AST_NONE)) != AST_NONE)
+                               ast_on(preempt);
+
+                       thread_unlock(self);
+               }
+       }
+       splx(s);
+}
+
+
+void
+thread_yield_internal(
+       mach_msg_timeout_t      ms)
+{
+       processor_t     myprocessor;
+
+       disable_preemption();
+       myprocessor = current_processor();
+       if (SCHED(processor_queue_empty)(myprocessor) && rt_runq.count == 0) {
+               mp_enable_preemption();
+
+               return;
+       }
+       enable_preemption();
+
+       thread_depress_ms(ms);
+
+       thread_block_reason(THREAD_CONTINUE_NULL, NULL, AST_YIELD);
+
+       thread_depress_abort_internal(current_thread());
+}
+