X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=drivers%2Foprofile%2Fbuffer_sync.c;h=fd8b050596eeb207a0a343c993876c6b30da8ed8;hb=refs%2Fheads%2Fvserver;hp=9ac2dab4d70556272427d1a68a8c7f2f4a4c9206;hpb=6a77f38946aaee1cd85eeec6cf4229b204c15071;p=linux-2.6.git

diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c
index 9ac2dab4d..fd8b05059 100644
--- a/drivers/oprofile/buffer_sync.c
+++ b/drivers/oprofile/buffer_sync.c
@@ -6,6 +6,10 @@
  *
  * @author John Levon <levon@movementarian.org>
  *
+ * Modified by Aravind Menon for Xen
+ * These modifications are:
+ * Copyright (C) 2005 Hewlett-Packard Co.
+ *
  * This is the core of the buffer management. Each
  * CPU buffer is processed and entered into the
  * global event buffer. Such processing is necessary
@@ -34,22 +38,25 @@
  
 static LIST_HEAD(dying_tasks);
 static LIST_HEAD(dead_tasks);
-cpumask_t marked_cpus = CPU_MASK_NONE;
+static cpumask_t marked_cpus = CPU_MASK_NONE;
 static DEFINE_SPINLOCK(task_mortuary);
-void process_task_mortuary(void);
+static void process_task_mortuary(void);
 
 
 /* Take ownership of the task struct and place it on the
  * list for processing. Only after two full buffer syncs
  * does the task eventually get freed, because by then
  * we are sure we will not reference it again.
+ * Can be invoked from softirq via RCU callback due to
+ * call_rcu() of the task struct, hence the _irqsave.
  */
 static int task_free_notify(struct notifier_block * self, unsigned long val, void * data)
 {
+	unsigned long flags;
 	struct task_struct * task = data;
-	spin_lock(&task_mortuary);
+	spin_lock_irqsave(&task_mortuary, flags);
 	list_add(&task->tasks, &dying_tasks);
-	spin_unlock(&task_mortuary);
+	spin_unlock_irqrestore(&task_mortuary, flags);
 	return NOTIFY_OK;
 }
 
@@ -62,7 +69,7 @@ static int task_exit_notify(struct notifier_block * self, unsigned long val, voi
 	/* To avoid latency problems, we only process the current CPU,
 	 * hoping that most samples for the task are on this CPU
 	 */
-	sync_buffer(_smp_processor_id());
+	sync_buffer(raw_smp_processor_id());
   	return 0;
 }
 
@@ -86,7 +93,7 @@ static int munmap_notify(struct notifier_block * self, unsigned long val, void *
 		/* To avoid latency problems, we only process the current CPU,
 		 * hoping that most samples for the task are on this CPU
 		 */
-		sync_buffer(_smp_processor_id());
+		sync_buffer(raw_smp_processor_id());
 		return 0;
 	}
 
@@ -105,10 +112,10 @@ static int module_load_notify(struct notifier_block * self, unsigned long val, v
 		return 0;
 
 	/* FIXME: should we process all CPU buffers ? */
-	down(&buffer_sem);
+	mutex_lock(&buffer_mutex);
 	add_event_entry(ESCAPE_CODE);
 	add_event_entry(MODULE_LOADED_CODE);
-	up(&buffer_sem);
+	mutex_unlock(&buffer_mutex);
 #endif
 	return 0;
 }
@@ -206,7 +213,7 @@ static inline unsigned long fast_get_dcookie(struct dentry * dentry,
  */
 static unsigned long get_exec_dcookie(struct mm_struct * mm)
 {
-	unsigned long cookie = 0;
+	unsigned long cookie = NO_COOKIE;
 	struct vm_area_struct * vma;
  
 	if (!mm)
@@ -217,8 +224,8 @@ static unsigned long get_exec_dcookie(struct mm_struct * mm)
 			continue;
 		if (!(vma->vm_flags & VM_EXECUTABLE))
 			continue;
-		cookie = fast_get_dcookie(vma->vm_file->f_dentry,
-			vma->vm_file->f_vfsmnt);
+		cookie = fast_get_dcookie(vma->vm_file->f_path.dentry,
+			vma->vm_file->f_path.mnt);
 		break;
 	}
 
@@ -234,46 +241,69 @@ out:
  */
 static unsigned long lookup_dcookie(struct mm_struct * mm, unsigned long addr, off_t * offset)
 {
-	unsigned long cookie = 0;
+	unsigned long cookie = NO_COOKIE;
 	struct vm_area_struct * vma;
 
 	for (vma = find_vma(mm, addr); vma; vma = vma->vm_next) {
  
-		if (!vma->vm_file)
-			continue;
-
 		if (addr < vma->vm_start || addr >= vma->vm_end)
 			continue;
 
-		cookie = fast_get_dcookie(vma->vm_file->f_dentry,
-			vma->vm_file->f_vfsmnt);
-		*offset = (vma->vm_pgoff << PAGE_SHIFT) + addr - vma->vm_start; 
+		if (vma->vm_file) {
+			cookie = fast_get_dcookie(vma->vm_file->f_path.dentry,
+				vma->vm_file->f_path.mnt);
+			*offset = (vma->vm_pgoff << PAGE_SHIFT) + addr -
+				vma->vm_start;
+		} else {
+			/* must be an anonymous map */
+			*offset = addr;
+		}
+
 		break;
 	}
 
+	if (!vma)
+		cookie = INVALID_COOKIE;
+
 	return cookie;
 }
 
 
-static unsigned long last_cookie = ~0UL;
+static unsigned long last_cookie = INVALID_COOKIE;
  
 static void add_cpu_switch(int i)
 {
 	add_event_entry(ESCAPE_CODE);
 	add_event_entry(CPU_SWITCH_CODE);
 	add_event_entry(i);
-	last_cookie = ~0UL;
+	last_cookie = INVALID_COOKIE;
 }
 
-static void add_kernel_ctx_switch(unsigned int in_kernel)
+static void add_cpu_mode_switch(unsigned int cpu_mode)
 {
 	add_event_entry(ESCAPE_CODE);
-	if (in_kernel)
-		add_event_entry(KERNEL_ENTER_SWITCH_CODE); 
-	else
-		add_event_entry(KERNEL_EXIT_SWITCH_CODE); 
+	switch (cpu_mode) {
+	case CPU_MODE_USER:
+		add_event_entry(USER_ENTER_SWITCH_CODE);
+		break;
+	case CPU_MODE_KERNEL:
+		add_event_entry(KERNEL_ENTER_SWITCH_CODE);
+		break;
+	case CPU_MODE_XEN:
+		add_event_entry(XEN_ENTER_SWITCH_CODE);
+	  	break;
+	default:
+		break;
+	}
 }
- 
+
+static void add_domain_switch(unsigned long domain_id)
+{
+	add_event_entry(ESCAPE_CODE);
+	add_event_entry(DOMAIN_SWITCH_CODE);
+	add_event_entry(domain_id);
+}
+
 static void
 add_user_ctx_switch(struct task_struct const * task, unsigned long cookie)
 {
@@ -317,7 +347,7 @@ static int add_us_sample(struct mm_struct * mm, struct op_sample * s)
  
  	cookie = lookup_dcookie(mm, s->eip, &offset);
  
-	if (!cookie) {
+	if (cookie == INVALID_COOKIE) {
 		atomic_inc(&oprofile_stats.sample_lost_no_mapping);
 		return 0;
 	}
@@ -338,9 +368,9 @@ static int add_us_sample(struct mm_struct * mm, struct op_sample * s)
  * for later lookup from userspace.
  */
 static int
-add_sample(struct mm_struct * mm, struct op_sample * s, int in_kernel)
+add_sample(struct mm_struct * mm, struct op_sample * s, int cpu_mode)
 {
-	if (in_kernel) {
+	if (cpu_mode >= CPU_MODE_KERNEL) {
 		add_sample_entry(s->eip, s->event);
 		return 1;
 	} else if (mm) {
@@ -422,27 +452,24 @@ static void increment_tail(struct oprofile_cpu_buffer * b)
  * and to have reached the list, it must have gone through
  * one full sync already.
  */
-void process_task_mortuary(void)
+static void process_task_mortuary(void)
 {
-	struct list_head * pos;
-	struct list_head * pos2;
+	unsigned long flags;
+	LIST_HEAD(local_dead_tasks);
 	struct task_struct * task;
+	struct task_struct * ttask;
 
-	spin_lock(&task_mortuary);
+	spin_lock_irqsave(&task_mortuary, flags);
 
-	list_for_each_safe(pos, pos2, &dead_tasks) {
-		task = list_entry(pos, struct task_struct, tasks);
-		list_del(&task->tasks);
-		free_task(task);
-	}
+	list_splice_init(&dead_tasks, &local_dead_tasks);
+	list_splice_init(&dying_tasks, &dead_tasks);
+
+	spin_unlock_irqrestore(&task_mortuary, flags);
 
-	list_for_each_safe(pos, pos2, &dying_tasks) {
-		task = list_entry(pos, struct task_struct, tasks);
+	list_for_each_entry_safe(task, ttask, &local_dead_tasks, tasks) {
 		list_del(&task->tasks);
-		list_add_tail(&task->tasks, &dead_tasks);
+		free_task(task);
 	}
-
-	spin_unlock(&task_mortuary);
 }
 
 
@@ -489,12 +516,13 @@ void sync_buffer(int cpu)
 	struct mm_struct *mm = NULL;
 	struct task_struct * new;
 	unsigned long cookie = 0;
-	int in_kernel = 1;
+	int cpu_mode = 1;
 	unsigned int i;
 	sync_buffer_state state = sb_buffer_start;
 	unsigned long available;
+	int domain_switch = 0;
 
-	down(&buffer_sem);
+	mutex_lock(&buffer_mutex);
  
 	add_cpu_switch(cpu);
 
@@ -505,16 +533,18 @@ void sync_buffer(int cpu)
 	for (i = 0; i < available; ++i) {
 		struct op_sample * s = &cpu_buf->buffer[cpu_buf->tail_pos];
  
-		if (is_code(s->eip)) {
-			if (s->event <= CPU_IS_KERNEL) {
-				/* kernel/userspace switch */
-				in_kernel = s->event;
+		if (is_code(s->eip) && !domain_switch) {
+			if (s->event <= CPU_MODE_XEN) {
+				/* xen/kernel/userspace switch */
+				cpu_mode = s->event;
 				if (state == sb_buffer_start)
 					state = sb_sample_start;
-				add_kernel_ctx_switch(s->event);
+				add_cpu_mode_switch(s->event);
 			} else if (s->event == CPU_TRACE_BEGIN) {
 				state = sb_bt_start;
 				add_trace_begin();
+			} else if (s->event == CPU_DOMAIN_SWITCH) {
+					domain_switch = 1;				
 			} else {
 				struct mm_struct * oldmm = mm;
 
@@ -528,11 +558,16 @@ void sync_buffer(int cpu)
 				add_user_ctx_switch(new, cookie);
 			}
 		} else {
-			if (state >= sb_bt_start &&
-			    !add_sample(mm, s, in_kernel)) {
-				if (state == sb_bt_start) {
-					state = sb_bt_ignore;
-					atomic_inc(&oprofile_stats.bt_lost_no_mapping);
+			if (domain_switch) {
+				add_domain_switch(s->eip);
+				domain_switch = 0;
+			} else {
+				if (state >= sb_bt_start &&
+				    !add_sample(mm, s, cpu_mode)) {
+					if (state == sb_bt_start) {
+						state = sb_bt_ignore;
+						atomic_inc(&oprofile_stats.bt_lost_no_mapping);
+					}
 				}
 			}
 		}
@@ -543,5 +578,5 @@ void sync_buffer(int cpu)
 
 	mark_done(cpu);
 
-	up(&buffer_sem);
+	mutex_unlock(&buffer_mutex);
 }