Merge branch 'uprobes/core' of...

Merge branch 'uprobes/core' of git://git.kernel.org/pub/scm/linux/kernel/git/oleg/misc into perf/core

Improve uprobes performance by adding 'pre-filtering' support,
by Oleg Nesterov:

	# time perl -e 'syscall -1 for 1..100_000'
	real    0m0.040s
	user    0m0.027s
	sys     0m0.010s

	# perf probe -x /lib/libc.so.6 syscall
	# perf record -e probe_libc:syscall sleep 100 &

Before this series:

	# time perl -e 'syscall -1 for 1..100_000'
	real    0m1.714s
	user    0m0.103s
	sys     0m1.607s

After:

	# time perl -e 'syscall -1 for 1..100_000'
	real    0m0.037s
	user    0m0.013s
	sys     0m0.023s
Signed-off-by: Ingo Molnar <mingo@kernel.org>

arch/x86/kernel/uprobes.c

@@ -680,8 +680,10 @@ static bool __skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs)
 		if (auprobe->insn[i] == 0x66)
 			continue;
 
-		if (auprobe->insn[i] == 0x90)
+		if (auprobe->insn[i] == 0x90) {
+			regs->ip += i + 1;
 			return true;
+		}
 
 		break;
 	}

include/linux/perf_event.h

@@ -135,16 +135,21 @@ struct hw_perf_event {
 		struct { /* software */
 			struct hrtimer	hrtimer;
 		};
+		struct { /* tracepoint */
+			struct task_struct	*tp_target;
+			/* for tp_event->class */
+			struct list_head	tp_list;
+		};
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 		struct { /* breakpoint */
-			struct arch_hw_breakpoint	info;
-			struct list_head		bp_list;
 			/*
 			 * Crufty hack to avoid the chicken and egg
 			 * problem hw_breakpoint has with context
 			 * creation and event initalization.
 			 */
 			struct task_struct		*bp_target;
+			struct arch_hw_breakpoint	info;
+			struct list_head		bp_list;
 		};
 #endif
 	};

include/linux/uprobes.h

@@ -35,13 +35,20 @@ struct inode;
 # include <asm/uprobes.h>
 #endif
 
+#define UPROBE_HANDLER_REMOVE		1
+#define UPROBE_HANDLER_MASK		1
+
+enum uprobe_filter_ctx {
+	UPROBE_FILTER_REGISTER,
+	UPROBE_FILTER_UNREGISTER,
+	UPROBE_FILTER_MMAP,
+};
+
 struct uprobe_consumer {
 	int (*handler)(struct uprobe_consumer *self, struct pt_regs *regs);
-	/*
-	 * filter is optional; If a filter exists, handler is run
-	 * if and only if filter returns true.
-	 */
-	bool (*filter)(struct uprobe_consumer *self, struct task_struct *task);
+	bool (*filter)(struct uprobe_consumer *self,
+				enum uprobe_filter_ctx ctx,
+				struct mm_struct *mm);
 
 	struct uprobe_consumer *next;
 };
@@ -94,6 +101,7 @@ extern int __weak set_swbp(struct arch_uprobe *aup, struct mm_struct *mm, unsign
 extern int __weak set_orig_insn(struct arch_uprobe *aup, struct mm_struct *mm, unsigned long vaddr);
 extern bool __weak is_swbp_insn(uprobe_opcode_t *insn);
 extern int uprobe_register(struct inode *inode, loff_t offset, struct uprobe_consumer *uc);
+extern int uprobe_apply(struct inode *inode, loff_t offset, struct uprobe_consumer *uc, bool);
 extern void uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consumer *uc);
 extern int uprobe_mmap(struct vm_area_struct *vma);
 extern void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned long end);
@@ -117,6 +125,11 @@ uprobe_register(struct inode *inode, loff_t offset, struct uprobe_consumer *uc)
 {
 	return -ENOSYS;
 }
+static inline int
+uprobe_apply(struct inode *inode, loff_t offset, struct uprobe_consumer *uc, bool add)
+{
+	return -ENOSYS;
+}
 static inline void
 uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consumer *uc)
 {

kernel/events/core.c

@@ -6162,11 +6162,14 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 
 	if (task) {
 		event->attach_state = PERF_ATTACH_TASK;
+
+		if (attr->type == PERF_TYPE_TRACEPOINT)
+			event->hw.tp_target = task;
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 		/*
 		 * hw_breakpoint is a bit difficult here..
 		 */
-		if (attr->type == PERF_TYPE_BREAKPOINT)
+		else if (attr->type == PERF_TYPE_BREAKPOINT)
 			event->hw.bp_target = task;
 #endif
 	}

kernel/events/uprobes.c

@@ -27,6 +27,7 @@
 #include <linux/pagemap.h>	/* read_mapping_page */
 #include <linux/slab.h>
 #include <linux/sched.h>
+#include <linux/export.h>
 #include <linux/rmap.h>		/* anon_vma_prepare */
 #include <linux/mmu_notifier.h>	/* set_pte_at_notify */
 #include <linux/swap.h>		/* try_to_free_swap */
@@ -41,58 +42,31 @@
 #define MAX_UPROBE_XOL_SLOTS		UINSNS_PER_PAGE
 
 static struct rb_root uprobes_tree = RB_ROOT;
-
-static DEFINE_SPINLOCK(uprobes_treelock);	/* serialize rbtree access */
-
-#define UPROBES_HASH_SZ	13
-
 /*
- * We need separate register/unregister and mmap/munmap lock hashes because
- * of mmap_sem nesting.
- *
- * uprobe_register() needs to install probes on (potentially) all processes
- * and thus needs to acquire multiple mmap_sems (consequtively, not
- * concurrently), whereas uprobe_mmap() is called while holding mmap_sem
- * for the particular process doing the mmap.
- *
- * uprobe_register()->register_for_each_vma() needs to drop/acquire mmap_sem
- * because of lock order against i_mmap_mutex. This means there's a hole in
- * the register vma iteration where a mmap() can happen.
- *
- * Thus uprobe_register() can race with uprobe_mmap() and we can try and
- * install a probe where one is already installed.
+ * allows us to skip the uprobe_mmap if there are no uprobe events active
+ * at this time.  Probably a fine grained per inode count is better?
  */
+#define no_uprobe_events()	RB_EMPTY_ROOT(&uprobes_tree)
 
-/* serialize (un)register */
-static struct mutex uprobes_mutex[UPROBES_HASH_SZ];
-
-#define uprobes_hash(v)		(&uprobes_mutex[((unsigned long)(v)) % UPROBES_HASH_SZ])
+static DEFINE_SPINLOCK(uprobes_treelock);	/* serialize rbtree access */
 
+#define UPROBES_HASH_SZ	13
 /* serialize uprobe->pending_list */
 static struct mutex uprobes_mmap_mutex[UPROBES_HASH_SZ];
 #define uprobes_mmap_hash(v)	(&uprobes_mmap_mutex[((unsigned long)(v)) % UPROBES_HASH_SZ])
 
 static struct percpu_rw_semaphore dup_mmap_sem;
 
-/*
- * uprobe_events allows us to skip the uprobe_mmap if there are no uprobe
- * events active at this time.  Probably a fine grained per inode count is
- * better?
- */
-static atomic_t uprobe_events = ATOMIC_INIT(0);
-
 /* Have a copy of original instruction */
 #define UPROBE_COPY_INSN	0
-/* Dont run handlers when first register/ last unregister in progress*/
-#define UPROBE_RUN_HANDLER	1
 /* Can skip singlestep */
-#define UPROBE_SKIP_SSTEP	2
+#define UPROBE_SKIP_SSTEP	1
 
 struct uprobe {
 	struct rb_node		rb_node;	/* node in the rb tree */
 	atomic_t		ref;
+	struct rw_semaphore	register_rwsem;
 	struct rw_semaphore	consumer_rwsem;
-	struct mutex		copy_mutex;	/* TODO: kill me and UPROBE_COPY_INSN */
 	struct list_head	pending_list;
 	struct uprobe_consumer	*consumers;
 	struct inode		*inode;		/* Also hold a ref to inode */
@@ -430,9 +404,6 @@ static struct uprobe *insert_uprobe(struct uprobe *uprobe)
 	u = __insert_uprobe(uprobe);
 	spin_unlock(&uprobes_treelock);
 
-	/* For now assume that the instruction need not be single-stepped */
-	__set_bit(UPROBE_SKIP_SSTEP, &uprobe->flags);
-
 	return u;
 }
 
@@ -452,8 +423,10 @@ static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset)
 
 	uprobe->inode = igrab(inode);
 	uprobe->offset = offset;
+	init_rwsem(&uprobe->register_rwsem);
 	init_rwsem(&uprobe->consumer_rwsem);
-	mutex_init(&uprobe->copy_mutex);
+	/* For now assume that the instruction need not be single-stepped */
+	__set_bit(UPROBE_SKIP_SSTEP, &uprobe->flags);
 
 	/* add to uprobes_tree, sorted on inode:offset */
 	cur_uprobe = insert_uprobe(uprobe);
@@ -463,38 +436,17 @@ static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset)
 		kfree(uprobe);
 		uprobe = cur_uprobe;
 		iput(inode);
-	} else {
-		atomic_inc(&uprobe_events);
 	}
 
 	return uprobe;
 }
 
-static void handler_chain(struct uprobe *uprobe, struct pt_regs *regs)
-{
-	struct uprobe_consumer *uc;
-
-	if (!test_bit(UPROBE_RUN_HANDLER, &uprobe->flags))
-		return;
-
-	down_read(&uprobe->consumer_rwsem);
-	for (uc = uprobe->consumers; uc; uc = uc->next) {
-		if (!uc->filter || uc->filter(uc, current))
-			uc->handler(uc, regs);
-	}
-	up_read(&uprobe->consumer_rwsem);
-}
-
-/* Returns the previous consumer */
-static struct uprobe_consumer *
-consumer_add(struct uprobe *uprobe, struct uprobe_consumer *uc)
+static void consumer_add(struct uprobe *uprobe, struct uprobe_consumer *uc)
 {
 	down_write(&uprobe->consumer_rwsem);
 	uc->next = uprobe->consumers;
 	uprobe->consumers = uc;
 	up_write(&uprobe->consumer_rwsem);
-
-	return uc->next;
 }
 
 /*
@@ -588,7 +540,8 @@ static int prepare_uprobe(struct uprobe *uprobe, struct file *file,
 	if (test_bit(UPROBE_COPY_INSN, &uprobe->flags))
 		return ret;
 
-	mutex_lock(&uprobe->copy_mutex);
+	/* TODO: move this into _register, until then we abuse this sem. */
+	down_write(&uprobe->consumer_rwsem);
 	if (test_bit(UPROBE_COPY_INSN, &uprobe->flags))
 		goto out;
 
@@ -612,7 +565,30 @@ static int prepare_uprobe(struct uprobe *uprobe, struct file *file,
 	set_bit(UPROBE_COPY_INSN, &uprobe->flags);
 
  out:
-	mutex_unlock(&uprobe->copy_mutex);
+	up_write(&uprobe->consumer_rwsem);
+
+	return ret;
+}
+
+static inline bool consumer_filter(struct uprobe_consumer *uc,
+				   enum uprobe_filter_ctx ctx, struct mm_struct *mm)
+{
+	return !uc->filter || uc->filter(uc, ctx, mm);
+}
+
+static bool filter_chain(struct uprobe *uprobe,
+			 enum uprobe_filter_ctx ctx, struct mm_struct *mm)
+{
+	struct uprobe_consumer *uc;
+	bool ret = false;
+
+	down_read(&uprobe->consumer_rwsem);
+	for (uc = uprobe->consumers; uc; uc = uc->next) {
+		ret = consumer_filter(uc, ctx, mm);
+		if (ret)
+			break;
+	}
+	up_read(&uprobe->consumer_rwsem);
 
 	return ret;
 }
@@ -624,16 +600,6 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm,
 	bool first_uprobe;
 	int ret;
 
-	/*
-	 * If probe is being deleted, unregister thread could be done with
-	 * the vma-rmap-walk through. Adding a probe now can be fatal since
-	 * nobody will be able to cleanup. Also we could be from fork or
-	 * mremap path, where the probe might have already been inserted.
-	 * Hence behave as if probe already existed.
-	 */
-	if (!uprobe->consumers)
-		return 0;
-
 	ret = prepare_uprobe(uprobe, vma->vm_file, mm, vaddr);
 	if (ret)
 		return ret;
@@ -658,14 +624,14 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm,
 static int
 remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, unsigned long vaddr)
 {
-	/* can happen if uprobe_register() fails */
-	if (!test_bit(MMF_HAS_UPROBES, &mm->flags))
-		return 0;
-
 	set_bit(MMF_RECALC_UPROBES, &mm->flags);
 	return set_orig_insn(&uprobe->arch, mm, vaddr);
 }
 
+static inline bool uprobe_is_active(struct uprobe *uprobe)
+{
+	return !RB_EMPTY_NODE(&uprobe->rb_node);
+}
 /*
  * There could be threads that have already hit the breakpoint. They
  * will recheck the current insn and restart if find_uprobe() fails.
@@ -673,12 +639,15 @@ remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, unsigned long vad
  */
 static void delete_uprobe(struct uprobe *uprobe)
 {
+	if (WARN_ON(!uprobe_is_active(uprobe)))
+		return;
+
 	spin_lock(&uprobes_treelock);
 	rb_erase(&uprobe->rb_node, &uprobes_tree);
 	spin_unlock(&uprobes_treelock);
+	RB_CLEAR_NODE(&uprobe->rb_node); /* for uprobe_is_active() */
 	iput(uprobe->inode);
 	put_uprobe(uprobe);
-	atomic_dec(&uprobe_events);
 }
 
 struct map_info {
@@ -764,8 +733,10 @@ build_map_info(struct address_space *mapping, loff_t offset, bool is_register)
 	return curr;
 }
 
-static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
+static int
+register_for_each_vma(struct uprobe *uprobe, struct uprobe_consumer *new)
 {
+	bool is_register = !!new;
 	struct map_info *info;
 	int err = 0;
 
@@ -794,10 +765,16 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
 		    vaddr_to_offset(vma, info->vaddr) != uprobe->offset)
 			goto unlock;
 
-		if (is_register)
+		if (is_register) {
+			/* consult only the "caller", new consumer. */
+			if (consumer_filter(new,
+					UPROBE_FILTER_REGISTER, mm))
 				err = install_breakpoint(uprobe, mm, vma, info->vaddr);
-		else
+		} else if (test_bit(MMF_HAS_UPROBES, &mm->flags)) {
+			if (!filter_chain(uprobe,
+					UPROBE_FILTER_UNREGISTER, mm))
 				err |= remove_breakpoint(uprobe, mm, info->vaddr);
+		}
 
  unlock:
 		up_write(&mm->mmap_sem);
@@ -810,17 +787,23 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
 	return err;
 }
 
-static int __uprobe_register(struct uprobe *uprobe)
+static int __uprobe_register(struct uprobe *uprobe, struct uprobe_consumer *uc)
 {
-	return register_for_each_vma(uprobe, true);
+	consumer_add(uprobe, uc);
+	return register_for_each_vma(uprobe, uc);
 }
 
-static void __uprobe_unregister(struct uprobe *uprobe)
+static void __uprobe_unregister(struct uprobe *uprobe, struct uprobe_consumer *uc)
 {
-	if (!register_for_each_vma(uprobe, false))
-		delete_uprobe(uprobe);
+	int err;
 
+	if (!consumer_del(uprobe, uc))	/* WARN? */
+		return;
+
+	err = register_for_each_vma(uprobe, NULL);
 	/* TODO : cant unregister? schedule a worker thread */
+	if (!uprobe->consumers && !err)
+		delete_uprobe(uprobe);
 }
 
 /*
@@ -845,30 +828,58 @@ int uprobe_register(struct inode *inode, loff_t offset, struct uprobe_consumer *
 	struct uprobe *uprobe;
 	int ret;
 
-	if (!inode || !uc || uc->next)
-		return -EINVAL;
-
+	/* Racy, just to catch the obvious mistakes */
 	if (offset > i_size_read(inode))
 		return -EINVAL;
 
-	ret = 0;
-	mutex_lock(uprobes_hash(inode));
+ retry:
 	uprobe = alloc_uprobe(inode, offset);
-
-	if (!uprobe) {
-		ret = -ENOMEM;
-	} else if (!consumer_add(uprobe, uc)) {
-		ret = __uprobe_register(uprobe);
-		if (ret) {
-			uprobe->consumers = NULL;
-			__uprobe_unregister(uprobe);
-		} else {
-			set_bit(UPROBE_RUN_HANDLER, &uprobe->flags);
-		}
+	if (!uprobe)
+		return -ENOMEM;
+	/*
+	 * We can race with uprobe_unregister()->delete_uprobe().
+	 * Check uprobe_is_active() and retry if it is false.
+	 */
+	down_write(&uprobe->register_rwsem);
+	ret = -EAGAIN;
+	if (likely(uprobe_is_active(uprobe))) {
+		ret = __uprobe_register(uprobe, uc);
+		if (ret)
+			__uprobe_unregister(uprobe, uc);
 	}
+	up_write(&uprobe->register_rwsem);
+	put_uprobe(uprobe);
+
+	if (unlikely(ret == -EAGAIN))
+		goto retry;
+	return ret;
+}
+EXPORT_SYMBOL_GPL(uprobe_register);
+
+/*
+ * uprobe_apply - unregister a already registered probe.
+ * @inode: the file in which the probe has to be removed.
+ * @offset: offset from the start of the file.
+ * @uc: consumer which wants to add more or remove some breakpoints
+ * @add: add or remove the breakpoints
+ */
+int uprobe_apply(struct inode *inode, loff_t offset,
+			struct uprobe_consumer *uc, bool add)
+{
+	struct uprobe *uprobe;
+	struct uprobe_consumer *con;
+	int ret = -ENOENT;
+
+	uprobe = find_uprobe(inode, offset);
+	if (!uprobe)
+		return ret;
 
-	mutex_unlock(uprobes_hash(inode));
-	if (uprobe)
+	down_write(&uprobe->register_rwsem);
+	for (con = uprobe->consumers; con && con != uc ; con = con->next)
+		;
+	if (con)
+		ret = register_for_each_vma(uprobe, add ? uc : NULL);
+	up_write(&uprobe->register_rwsem);
 	put_uprobe(uprobe);
 
 	return ret;
@@ -884,24 +895,42 @@ void uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consume
 {
 	struct uprobe *uprobe;
 
-	if (!inode || !uc)
-		return;
-
 	uprobe = find_uprobe(inode, offset);
 	if (!uprobe)
 		return;
 
-	mutex_lock(uprobes_hash(inode));
+	down_write(&uprobe->register_rwsem);
+	__uprobe_unregister(uprobe, uc);
+	up_write(&uprobe->register_rwsem);
+	put_uprobe(uprobe);
+}
+EXPORT_SYMBOL_GPL(uprobe_unregister);
 
-	if (consumer_del(uprobe, uc)) {
-		if (!uprobe->consumers) {
-			__uprobe_unregister(uprobe);
-			clear_bit(UPROBE_RUN_HANDLER, &uprobe->flags);
-		}
+static int unapply_uprobe(struct uprobe *uprobe, struct mm_struct *mm)
+{
+	struct vm_area_struct *vma;
+	int err = 0;
+
+	down_read(&mm->mmap_sem);
+	for (vma = mm->mmap; vma; vma = vma->vm_next) {
+		unsigned long vaddr;
+		loff_t offset;
+
+		if (!valid_vma(vma, false) ||
+		    vma->vm_file->f_mapping->host != uprobe->inode)
+			continue;
+
+		offset = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
+		if (uprobe->offset <  offset ||
+		    uprobe->offset >= offset + vma->vm_end - vma->vm_start)
+			continue;
+
+		vaddr = offset_to_vaddr(vma, uprobe->offset);
+		err |= remove_breakpoint(uprobe, mm, vaddr);
 	}
+	up_read(&mm->mmap_sem);
 
-	mutex_unlock(uprobes_hash(inode));
-	put_uprobe(uprobe);
+	return err;
 }
 
 static struct rb_node *
@@ -978,7 +1007,7 @@ int uprobe_mmap(struct vm_area_struct *vma)
 	struct uprobe *uprobe, *u;
 	struct inode *inode;
 
-	if (!atomic_read(&uprobe_events) || !valid_vma(vma, true))
+	if (no_uprobe_events() || !valid_vma(vma, true))
 		return 0;
 
 	inode = vma->vm_file->f_mapping->host;
@@ -987,9 +1016,14 @@ int uprobe_mmap(struct vm_area_struct *vma)
 
 	mutex_lock(uprobes_mmap_hash(inode));
 	build_probe_list(inode, vma, vma->vm_start, vma->vm_end, &tmp_list);
-
+	/*
+	 * We can race with uprobe_unregister(), this uprobe can be already
+	 * removed. But in this case filter_chain() must return false, all
+	 * consumers have gone away.
+	 */
 	list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) {
-		if (!fatal_signal_pending(current)) {
+		if (!fatal_signal_pending(current) &&
+		    filter_chain(uprobe, UPROBE_FILTER_MMAP, vma->vm_mm)) {
 			unsigned long vaddr = offset_to_vaddr(vma, uprobe->offset);
 			install_breakpoint(uprobe, vma->vm_mm, vma, vaddr);
 		}
@@ -1024,7 +1058,7 @@ vma_has_uprobes(struct vm_area_struct *vma, unsigned long start, unsigned long e
  */
 void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned long end)
 {
-	if (!atomic_read(&uprobe_events) || !valid_vma(vma, false))
+	if (no_uprobe_events() || !valid_vma(vma, false))
 		return;
 
 	if (!atomic_read(&vma->vm_mm->mm_users)) /* called by mmput() ? */
@@ -1041,22 +1075,14 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon
 /* Slot allocation for XOL */
 static int xol_add_vma(struct xol_area *area)
 {
-	struct mm_struct *mm;
-	int ret;
-
-	area->page = alloc_page(GFP_HIGHUSER);
-	if (!area->page)
-		return -ENOMEM;
-
-	ret = -EALREADY;
-	mm = current->mm;
+	struct mm_struct *mm = current->mm;
+	int ret = -EALREADY;
 
 	down_write(&mm->mmap_sem);
 	if (mm->uprobes_state.xol_area)
 		goto fail;
 
 	ret = -ENOMEM;
-
 	/* Try to map as high as possible, this is only a hint. */
 	area->vaddr = get_unmapped_area(NULL, TASK_SIZE - PAGE_SIZE, PAGE_SIZE, 0, 0);
 	if (area->vaddr & ~PAGE_MASK) {
@@ -1072,54 +1098,53 @@ static int xol_add_vma(struct xol_area *area)
 	smp_wmb();	/* pairs with get_xol_area() */
 	mm->uprobes_state.xol_area = area;
 	ret = 0;
-
-fail:
+ fail:
 	up_write(&mm->mmap_sem);
-	if (ret)
-		__free_page(area->page);
 
 	return ret;
 }
 
-static struct xol_area *get_xol_area(struct mm_struct *mm)
-{
-	struct xol_area *area;
-
-	area = mm->uprobes_state.xol_area;
-	smp_read_barrier_depends();	/* pairs with wmb in xol_add_vma() */
-
-	return area;
-}
-
 /*
- * xol_alloc_area - Allocate process's xol_area.
- * This area will be used for storing instructions for execution out of
- * line.
+ * get_xol_area - Allocate process's xol_area if necessary.
+ * This area will be used for storing instructions for execution out of line.
  *
  * Returns the allocated area or NULL.
  */
-static struct xol_area *xol_alloc_area(void)
+static struct xol_area *get_xol_area(void)
 {
+	struct mm_struct *mm = current->mm;
 	struct xol_area *area;
 
+	area = mm->uprobes_state.xol_area;
+	if (area)
+		goto ret;
+
 	area = kzalloc(sizeof(*area), GFP_KERNEL);
 	if (unlikely(!area))
-		return NULL;
+		goto out;
 
 	area->bitmap = kzalloc(BITS_TO_LONGS(UINSNS_PER_PAGE) * sizeof(long), GFP_KERNEL);
-
 	if (!area->bitmap)
-		goto fail;
+		goto free_area;
+
+	area->page = alloc_page(GFP_HIGHUSER);
+	if (!area->page)
+		goto free_bitmap;
 
 	init_waitqueue_head(&area->wq);
 	if (!xol_add_vma(area))
 		return area;
 
-fail:
+	__free_page(area->page);
+ free_bitmap:
 	kfree(area->bitmap);
+ free_area:
 	kfree(area);
-
-	return get_xol_area(current->mm);
+ out:
+	area = mm->uprobes_state.xol_area;
+ ret:
+	smp_read_barrier_depends();     /* pairs with wmb in xol_add_vma() */
+	return area;
 }
 
 /*
@@ -1185,33 +1210,26 @@ static unsigned long xol_take_insn_slot(struct xol_area *area)
 }
 
 /*
- * xol_get_insn_slot - If was not allocated a slot, then
- * allocate a slot.
+ * xol_get_insn_slot - allocate a slot for xol.
  * Returns the allocated slot address or 0.
  */
-static unsigned long xol_get_insn_slot(struct uprobe *uprobe, unsigned long slot_addr)
+static unsigned long xol_get_insn_slot(struct uprobe *uprobe)
 {
 	struct xol_area *area;
 	unsigned long offset;
+	unsigned long xol_vaddr;
 	void *vaddr;
 
-	area = get_xol_area(current->mm);
-	if (!area) {
-		area = xol_alloc_area();
+	area = get_xol_area();
 	if (!area)
 		return 0;
-	}
-	current->utask->xol_vaddr = xol_take_insn_slot(area);
 
-	/*
-	 * Initialize the slot if xol_vaddr points to valid
-	 * instruction slot.
-	 */
-	if (unlikely(!current->utask->xol_vaddr))
+	xol_vaddr = xol_take_insn_slot(area);
+	if (unlikely(!xol_vaddr))
 		return 0;
 
-	current->utask->vaddr = slot_addr;
-	offset = current->utask->xol_vaddr & ~PAGE_MASK;
+	/* Initialize the slot */
+	offset = xol_vaddr & ~PAGE_MASK;
 	vaddr = kmap_atomic(area->page);
 	memcpy(vaddr + offset, uprobe->arch.insn, MAX_UINSN_BYTES);
 	kunmap_atomic(vaddr);
@@ -1221,7 +1239,7 @@ static unsigned long xol_get_insn_slot(struct uprobe *uprobe, unsigned long slot
 	 */
 	flush_dcache_page(area->page);
 
-	return current->utask->xol_vaddr;
+	return xol_vaddr;
 }
 
 /*
@@ -1239,8 +1257,7 @@ static void xol_free_insn_slot(struct task_struct *tsk)
 		return;
 
 	slot_addr = tsk->utask->xol_vaddr;
-
-	if (unlikely(!slot_addr || IS_ERR_VALUE(slot_addr)))
+	if (unlikely(!slot_addr))
 		return;
 
 	area = tsk->mm->uprobes_state.xol_area;
@@ -1302,33 +1319,48 @@ void uprobe_copy_process(struct task_struct *t)
 }
 
 /*
- * Allocate a uprobe_task object for the task.
- * Called when the thread hits a breakpoint for the first time.
+ * Allocate a uprobe_task object for the task if if necessary.
+ * Called when the thread hits a breakpoint.
  *
  * Returns:
  * - pointer to new uprobe_task on success
  * - NULL otherwise
  */
-static struct uprobe_task *add_utask(void)
+static struct uprobe_task *get_utask(void)
 {
-	struct uprobe_task *utask;
-
-	utask = kzalloc(sizeof *utask, GFP_KERNEL);
-	if (unlikely(!utask))
-		return NULL;
-
-	current->utask = utask;
-	return utask;
+	if (!current->utask)
+		current->utask = kzalloc(sizeof(struct uprobe_task), GFP_KERNEL);
+	return current->utask;
 }
 
 /* Prepare to single-step probed instruction out of line. */
 static int
-pre_ssout(struct uprobe *uprobe, struct pt_regs *regs, unsigned long vaddr)
+pre_ssout(struct uprobe *uprobe, struct pt_regs *regs, unsigned long bp_vaddr)
 {
-	if (xol_get_insn_slot(uprobe, vaddr) && !arch_uprobe_pre_xol(&uprobe->arch, regs))
-		return 0;
+	struct uprobe_task *utask;
+	unsigned long xol_vaddr;
+	int err;
 
-	return -EFAULT;
+	utask = get_utask();
+	if (!utask)
+		return -ENOMEM;
+
+	xol_vaddr = xol_get_insn_slot(uprobe);
+	if (!xol_vaddr)
+		return -ENOMEM;
+
+	utask->xol_vaddr = xol_vaddr;
+	utask->vaddr = bp_vaddr;
+
+	err = arch_uprobe_pre_xol(&uprobe->arch, regs);
+	if (unlikely(err)) {
+		xol_free_insn_slot(current);
+		return err;
+	}
+
+	utask->active_uprobe = uprobe;
+	utask->state = UTASK_SSTEP;
+	return 0;
 }
 
 /*
@@ -1390,6 +1422,7 @@ static void mmf_recalc_uprobes(struct mm_struct *mm)
 		 * This is not strictly accurate, we can race with
 		 * uprobe_unregister() and see the already removed
 		 * uprobe if delete_uprobe() was not yet called.
+		 * Or this uprobe can be filtered out.
 		 */
 		if (vma_has_uprobes(vma, vma->vm_start, vma->vm_end))
 			return;
@@ -1451,13 +1484,33 @@ static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp)
 	return uprobe;
 }
 
+static void handler_chain(struct uprobe *uprobe, struct pt_regs *regs)
+{
+	struct uprobe_consumer *uc;
+	int remove = UPROBE_HANDLER_REMOVE;
+
+	down_read(&uprobe->register_rwsem);
+	for (uc = uprobe->consumers; uc; uc = uc->next) {
+		int rc = uc->handler(uc, regs);
+
+		WARN(rc & ~UPROBE_HANDLER_MASK,
+			"bad rc=0x%x from %pf()\n", rc, uc->handler);
+		remove &= rc;
+	}
+
+	if (remove && uprobe->consumers) {
+		WARN_ON(!uprobe_is_active(uprobe));
+		unapply_uprobe(uprobe, current->mm);
+	}
+	up_read(&uprobe->register_rwsem);
+}
+
 /*
  * Run handler and ask thread to singlestep.
  * Ensure all non-fatal signals cannot interrupt thread while it singlesteps.
  */
 static void handle_swbp(struct pt_regs *regs)
 {
-	struct uprobe_task *utask;
 	struct uprobe *uprobe;
 	unsigned long bp_vaddr;
 	int uninitialized_var(is_swbp);
@@ -1482,6 +1535,10 @@ static void handle_swbp(struct pt_regs *regs)
 		}
 		return;
 	}
+
+	/* change it in advance for ->handler() and restart */
+	instruction_pointer_set(regs, bp_vaddr);
+
 	/*
 	 * TODO: move copy_insn/etc into _register and remove this hack.
 	 * After we hit the bp, _unregister + _register can install the
@@ -1489,32 +1546,16 @@ static void handle_swbp(struct pt_regs *regs)
 	 */
 	smp_rmb(); /* pairs with wmb() in install_breakpoint() */
 	if (unlikely(!test_bit(UPROBE_COPY_INSN, &uprobe->flags)))
-		goto restart;
-
-	utask = current->utask;
-	if (!utask) {
-		utask = add_utask();
-		/* Cannot allocate; re-execute the instruction. */
-		if (!utask)
-			goto restart;
-	}
+		goto out;
 
 	handler_chain(uprobe, regs);
 	if (can_skip_sstep(uprobe, regs))
 		goto out;
 
-	if (!pre_ssout(uprobe, regs, bp_vaddr)) {
-		utask->active_uprobe = uprobe;
-		utask->state = UTASK_SSTEP;
+	if (!pre_ssout(uprobe, regs, bp_vaddr))
 		return;
-	}
 
-restart:
-	/*
-	 * cannot singlestep; cannot skip instruction;
-	 * re-execute the instruction.
-	 */
-	instruction_pointer_set(regs, bp_vaddr);
+	/* can_skip_sstep() succeeded, or restart if can't singlestep */
 out:
 	put_uprobe(uprobe);
 }
@@ -1608,10 +1649,8 @@ static int __init init_uprobes(void)
 {
 	int i;
 
-	for (i = 0; i < UPROBES_HASH_SZ; i++) {
-		mutex_init(&uprobes_mutex[i]);
+	for (i = 0; i < UPROBES_HASH_SZ; i++)
 		mutex_init(&uprobes_mmap_mutex[i]);
-	}
 
 	if (percpu_init_rwsem(&dup_mmap_sem))
 		return -ENOMEM;

kernel/ptrace.c

@@ -712,6 +712,12 @@ static int ptrace_regset(struct task_struct *task, int req, unsigned int type,
 					     kiov->iov_len, kiov->iov_base);
 }
 
+/*
+ * This is declared in linux/regset.h and defined in machine-dependent
+ * code.  We put the export here, near the primary machine-neutral use,
+ * to ensure no machine forgets it.
+ */
+EXPORT_SYMBOL_GPL(task_user_regset_view);
 #endif
 
 int ptrace_request(struct task_struct *child, long request,

kernel/trace/trace_probe.h

@@ -66,7 +66,6 @@
 #define TP_FLAG_TRACE		1
 #define TP_FLAG_PROFILE		2
 #define TP_FLAG_REGISTERED	4
-#define TP_FLAG_UPROBE		8
 
 
 /* data_rloc: data relative location, compatible with u32 */

kernel/trace/trace_uprobe.c

@@ -28,20 +28,21 @@
 
 #define UPROBE_EVENT_SYSTEM	"uprobes"
 
+struct trace_uprobe_filter {
+	rwlock_t		rwlock;
+	int			nr_systemwide;
+	struct list_head	perf_events;
+};
+
 /*
  * uprobe event core functions
  */
-struct trace_uprobe;
-struct uprobe_trace_consumer {
-	struct uprobe_consumer		cons;
-	struct trace_uprobe		*tu;
-};
-
 struct trace_uprobe {
 	struct list_head		list;
 	struct ftrace_event_class	class;
 	struct ftrace_event_call	call;
-	struct uprobe_trace_consumer	*consumer;
+	struct trace_uprobe_filter	filter;
+	struct uprobe_consumer		consumer;
 	struct inode			*inode;
 	char				*filename;
 	unsigned long			offset;
@@ -64,6 +65,18 @@ static LIST_HEAD(uprobe_list);
 
 static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs);
 
+static inline void init_trace_uprobe_filter(struct trace_uprobe_filter *filter)
+{
+	rwlock_init(&filter->rwlock);
+	filter->nr_systemwide = 0;
+	INIT_LIST_HEAD(&filter->perf_events);
+}
+
+static inline bool uprobe_filter_is_empty(struct trace_uprobe_filter *filter)
+{
+	return !filter->nr_systemwide && list_empty(&filter->perf_events);
+}
+
 /*
  * Allocate new trace_uprobe and initialize it (including uprobes).
  */
@@ -92,6 +105,8 @@ alloc_trace_uprobe(const char *group, const char *event, int nargs)
 		goto error;
 
 	INIT_LIST_HEAD(&tu->list);
+	tu->consumer.handler = uprobe_dispatcher;
+	init_trace_uprobe_filter(&tu->filter);
 	return tu;
 
 error:
@@ -253,16 +268,18 @@ static int create_trace_uprobe(int argc, char **argv)
 	if (ret)
 		goto fail_address_parse;
 
-	ret = kstrtoul(arg, 0, &offset);
-	if (ret)
-		goto fail_address_parse;
-
 	inode = igrab(path.dentry->d_inode);
-	if (!S_ISREG(inode->i_mode)) {
+	path_put(&path);
+
+	if (!inode || !S_ISREG(inode->i_mode)) {
 		ret = -EINVAL;
 		goto fail_address_parse;
 	}
 
+	ret = kstrtoul(arg, 0, &offset);
+	if (ret)
+		goto fail_address_parse;
+
 	argc -= 2;
 	argv += 2;
 
@@ -469,7 +486,7 @@ static const struct file_operations uprobe_profile_ops = {
 };
 
 /* uprobe handler */
-static void uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
+static int uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
 {
 	struct uprobe_trace_entry_head *entry;
 	struct ring_buffer_event *event;
@@ -479,8 +496,6 @@ static void uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
 	unsigned long irq_flags;
 	struct ftrace_event_call *call = &tu->call;
 
-	tu->nhit++;
-
 	local_save_flags(irq_flags);
 	pc = preempt_count();
 
@@ -489,16 +504,18 @@ static void uprobe_trace_func(struct trace_uprobe *tu, struct pt_regs *regs)
 	event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
 						  size, irq_flags, pc);
 	if (!event)
-		return;
+		return 0;
 
 	entry = ring_buffer_event_data(event);
-	entry->ip = uprobe_get_swbp_addr(task_pt_regs(current));
+	entry->ip = instruction_pointer(task_pt_regs(current));
 	data = (u8 *)&entry[1];
 	for (i = 0; i < tu->nr_args; i++)
 		call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_buffer_unlock_commit(buffer, event, irq_flags, pc);
+
+	return 0;
 }
 
 /* Event entry printers */
@@ -537,42 +554,43 @@ print_uprobe_event(struct trace_iterator *iter, int flags, struct trace_event *e
 	return TRACE_TYPE_PARTIAL_LINE;
 }
 
-static int probe_event_enable(struct trace_uprobe *tu, int flag)
+static inline bool is_trace_uprobe_enabled(struct trace_uprobe *tu)
 {
-	struct uprobe_trace_consumer *utc;
-	int ret = 0;
+	return tu->flags & (TP_FLAG_TRACE | TP_FLAG_PROFILE);
+}
 
-	if (!tu->inode || tu->consumer)
-		return -EINTR;
+typedef bool (*filter_func_t)(struct uprobe_consumer *self,
+				enum uprobe_filter_ctx ctx,
+				struct mm_struct *mm);
 
-	utc = kzalloc(sizeof(struct uprobe_trace_consumer), GFP_KERNEL);
-	if (!utc)
+static int
+probe_event_enable(struct trace_uprobe *tu, int flag, filter_func_t filter)
+{
+	int ret = 0;
+
+	if (is_trace_uprobe_enabled(tu))
 		return -EINTR;
 
-	utc->cons.handler = uprobe_dispatcher;
-	utc->cons.filter = NULL;
-	ret = uprobe_register(tu->inode, tu->offset, &utc->cons);
-	if (ret) {
-		kfree(utc);
-		return ret;
-	}
+	WARN_ON(!uprobe_filter_is_empty(&tu->filter));
 
 	tu->flags |= flag;
-	utc->tu = tu;
-	tu->consumer = utc;
+	tu->consumer.filter = filter;
+	ret = uprobe_register(tu->inode, tu->offset, &tu->consumer);
+	if (ret)
+		tu->flags &= ~flag;
 
-	return 0;
+	return ret;
 }
 
 static void probe_event_disable(struct trace_uprobe *tu, int flag)
 {
-	if (!tu->inode || !tu->consumer)
+	if (!is_trace_uprobe_enabled(tu))
 		return;
 
-	uprobe_unregister(tu->inode, tu->offset, &tu->consumer->cons);
+	WARN_ON(!uprobe_filter_is_empty(&tu->filter));
+
+	uprobe_unregister(tu->inode, tu->offset, &tu->consumer);
 	tu->flags &= ~flag;
-	kfree(tu->consumer);
-	tu->consumer = NULL;
 }
 
 static int uprobe_event_define_fields(struct ftrace_event_call *event_call)
@@ -646,8 +664,96 @@ static int set_print_fmt(struct trace_uprobe *tu)
 }
 
 #ifdef CONFIG_PERF_EVENTS
+static bool
+__uprobe_perf_filter(struct trace_uprobe_filter *filter, struct mm_struct *mm)
+{
+	struct perf_event *event;
+
+	if (filter->nr_systemwide)
+		return true;
+
+	list_for_each_entry(event, &filter->perf_events, hw.tp_list) {
+		if (event->hw.tp_target->mm == mm)
+			return true;
+	}
+
+	return false;
+}
+
+static inline bool
+uprobe_filter_event(struct trace_uprobe *tu, struct perf_event *event)
+{
+	return __uprobe_perf_filter(&tu->filter, event->hw.tp_target->mm);
+}
+
+static int uprobe_perf_open(struct trace_uprobe *tu, struct perf_event *event)
+{
+	bool done;
+
+	write_lock(&tu->filter.rwlock);
+	if (event->hw.tp_target) {
+		/*
+		 * event->parent != NULL means copy_process(), we can avoid
+		 * uprobe_apply(). current->mm must be probed and we can rely
+		 * on dup_mmap() which preserves the already installed bp's.
+		 *
+		 * attr.enable_on_exec means that exec/mmap will install the
+		 * breakpoints we need.
+		 */
+		done = tu->filter.nr_systemwide ||
+			event->parent || event->attr.enable_on_exec ||
+			uprobe_filter_event(tu, event);
+		list_add(&event->hw.tp_list, &tu->filter.perf_events);
+	} else {
+		done = tu->filter.nr_systemwide;
+		tu->filter.nr_systemwide++;
+	}
+	write_unlock(&tu->filter.rwlock);
+
+	if (!done)
+		uprobe_apply(tu->inode, tu->offset, &tu->consumer, true);
+
+	return 0;
+}
+
+static int uprobe_perf_close(struct trace_uprobe *tu, struct perf_event *event)
+{
+	bool done;
+
+	write_lock(&tu->filter.rwlock);
+	if (event->hw.tp_target) {
+		list_del(&event->hw.tp_list);
+		done = tu->filter.nr_systemwide ||
+			(event->hw.tp_target->flags & PF_EXITING) ||
+			uprobe_filter_event(tu, event);
+	} else {
+		tu->filter.nr_systemwide--;
+		done = tu->filter.nr_systemwide;
+	}
+	write_unlock(&tu->filter.rwlock);
+
+	if (!done)
+		uprobe_apply(tu->inode, tu->offset, &tu->consumer, false);
+
+	return 0;
+}
+
+static bool uprobe_perf_filter(struct uprobe_consumer *uc,
+				enum uprobe_filter_ctx ctx, struct mm_struct *mm)
+{
+	struct trace_uprobe *tu;
+	int ret;
+
+	tu = container_of(uc, struct trace_uprobe, consumer);
+	read_lock(&tu->filter.rwlock);
+	ret = __uprobe_perf_filter(&tu->filter, mm);
+	read_unlock(&tu->filter.rwlock);
+
+	return ret;
+}
+
 /* uprobe profile handler */
-static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
+static int uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
 {
 	struct ftrace_event_call *call = &tu->call;
 	struct uprobe_trace_entry_head *entry;
@@ -656,11 +762,14 @@ static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
 	int size, __size, i;
 	int rctx;
 
+	if (!uprobe_perf_filter(&tu->consumer, 0, current->mm))
+		return UPROBE_HANDLER_REMOVE;
+
 	__size = sizeof(*entry) + tu->size;
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough"))
-		return;
+		return 0;
 
 	preempt_disable();
 
@@ -668,7 +777,7 @@ static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
 	if (!entry)
 		goto out;
 
-	entry->ip = uprobe_get_swbp_addr(task_pt_regs(current));
+	entry->ip = instruction_pointer(task_pt_regs(current));
 	data = (u8 *)&entry[1];
 	for (i = 0; i < tu->nr_args; i++)
 		call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset);
@@ -678,6 +787,7 @@ static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs)
 
  out:
 	preempt_enable();
+	return 0;
 }
 #endif	/* CONFIG_PERF_EVENTS */
 
@@ -688,7 +798,7 @@ int trace_uprobe_register(struct ftrace_event_call *event, enum trace_reg type,
 
 	switch (type) {
 	case TRACE_REG_REGISTER:
-		return probe_event_enable(tu, TP_FLAG_TRACE);
+		return probe_event_enable(tu, TP_FLAG_TRACE, NULL);
 
 	case TRACE_REG_UNREGISTER:
 		probe_event_disable(tu, TP_FLAG_TRACE);
@@ -696,11 +806,18 @@ int trace_uprobe_register(struct ftrace_event_call *event, enum trace_reg type,
 
 #ifdef CONFIG_PERF_EVENTS
 	case TRACE_REG_PERF_REGISTER:
-		return probe_event_enable(tu, TP_FLAG_PROFILE);
+		return probe_event_enable(tu, TP_FLAG_PROFILE, uprobe_perf_filter);
 
 	case TRACE_REG_PERF_UNREGISTER:
 		probe_event_disable(tu, TP_FLAG_PROFILE);
 		return 0;
+
+	case TRACE_REG_PERF_OPEN:
+		return uprobe_perf_open(tu, data);
+
+	case TRACE_REG_PERF_CLOSE:
+		return uprobe_perf_close(tu, data);
+
 #endif
 	default:
 		return 0;
@@ -710,22 +827,20 @@ int trace_uprobe_register(struct ftrace_event_call *event, enum trace_reg type,
 
 static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs)
 {
-	struct uprobe_trace_consumer *utc;
 	struct trace_uprobe *tu;
+	int ret = 0;
 
-	utc = container_of(con, struct uprobe_trace_consumer, cons);
-	tu = utc->tu;
-	if (!tu || tu->consumer != utc)
-		return 0;
+	tu = container_of(con, struct trace_uprobe, consumer);
+	tu->nhit++;
 
 	if (tu->flags & TP_FLAG_TRACE)
-		uprobe_trace_func(tu, regs);
+		ret |= uprobe_trace_func(tu, regs);
 
 #ifdef CONFIG_PERF_EVENTS
 	if (tu->flags & TP_FLAG_PROFILE)
-		uprobe_perf_func(tu, regs);
+		ret |= uprobe_perf_func(tu, regs);
 #endif
-	return 0;
+	return ret;
 }
 
 static struct trace_event_functions uprobe_funcs = {