1 files changed, 130 insertions, 98 deletions

diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index cb346f26a..4e5e9798a 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -86,15 +86,6 @@ struct uprobe {
 	struct arch_uprobe	arch;
 };
 
-struct return_instance {
-	struct uprobe		*uprobe;
-	unsigned long		func;
-	unsigned long		orig_ret_vaddr; /* original return address */
-	bool			chained;	/* true, if instance is nested */
-
-	struct return_instance	*next;		/* keep as stack */
-};
-
 /*
  * Execute out of line area: anonymous executable mapping installed
  * by the probed task to execute the copy of the original instruction
@@ -105,17 +96,18 @@ struct return_instance {
  * allocated.
  */
 struct xol_area {
-	wait_queue_head_t 	wq;		/* if all slots are busy */
-	atomic_t 		slot_count;	/* number of in-use slots */
-	unsigned long 		*bitmap;	/* 0 = free slot */
-	struct page 		*page;
+	wait_queue_head_t 		wq;		/* if all slots are busy */
+	atomic_t 			slot_count;	/* number of in-use slots */
+	unsigned long 			*bitmap;	/* 0 = free slot */
 
+	struct vm_special_mapping	xol_mapping;
+	struct page 			*pages[2];
 	/*
 	 * We keep the vma's vm_start rather than a pointer to the vma
 	 * itself.  The probed process or a naughty kernel module could make
 	 * the vma go away, and we must handle that reasonably gracefully.
 	 */
-	unsigned long 		vaddr;		/* Page(s) of instruction slots */
+	unsigned long 			vaddr;		/* Page(s) of instruction slots */
 };
 
 /*
@@ -366,6 +358,18 @@ set_orig_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long v
 	return uprobe_write_opcode(mm, vaddr, *(uprobe_opcode_t *)&auprobe->insn);
 }
 
+static struct uprobe *get_uprobe(struct uprobe *uprobe)
+{
+	atomic_inc(&uprobe->ref);
+	return uprobe;
+}
+
+static void put_uprobe(struct uprobe *uprobe)
+{
+	if (atomic_dec_and_test(&uprobe->ref))
+		kfree(uprobe);
+}
+
 static int match_uprobe(struct uprobe *l, struct uprobe *r)
 {
 	if (l->inode < r->inode)
@@ -393,10 +397,8 @@ static struct uprobe *__find_uprobe(struct inode *inode, loff_t offset)
 	while (n) {
 		uprobe = rb_entry(n, struct uprobe, rb_node);
 		match = match_uprobe(&u, uprobe);
-		if (!match) {
-			atomic_inc(&uprobe->ref);
-			return uprobe;
-		}
+		if (!match)
+			return get_uprobe(uprobe);
 
 		if (match < 0)
 			n = n->rb_left;
@@ -432,10 +434,8 @@ static struct uprobe *__insert_uprobe(struct uprobe *uprobe)
 		parent = *p;
 		u = rb_entry(parent, struct uprobe, rb_node);
 		match = match_uprobe(uprobe, u);
-		if (!match) {
-			atomic_inc(&u->ref);
-			return u;
-		}
+		if (!match)
+			return get_uprobe(u);
 
 		if (match < 0)
 			p = &parent->rb_left;
@@ -472,12 +472,6 @@ static struct uprobe *insert_uprobe(struct uprobe *uprobe)
 	return u;
 }
 
-static void put_uprobe(struct uprobe *uprobe)
-{
-	if (atomic_dec_and_test(&uprobe->ref))
-		kfree(uprobe);
-}
-
 static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset)
 {
 	struct uprobe *uprobe, *cur_uprobe;
@@ -1039,14 +1033,14 @@ static void build_probe_list(struct inode *inode,
 			if (u->inode != inode || u->offset < min)
 				break;
 			list_add(&u->pending_list, head);
-			atomic_inc(&u->ref);
+			get_uprobe(u);
 		}
 		for (t = n; (t = rb_next(t)); ) {
 			u = rb_entry(t, struct uprobe, rb_node);
 			if (u->inode != inode || u->offset > max)
 				break;
 			list_add(&u->pending_list, head);
-			atomic_inc(&u->ref);
+			get_uprobe(u);
 		}
 	}
 	spin_unlock(&uprobes_treelock);
@@ -1132,11 +1126,14 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon
 /* Slot allocation for XOL */
 static int xol_add_vma(struct mm_struct *mm, struct xol_area *area)
 {
-	int ret = -EALREADY;
+	struct vm_area_struct *vma;
+	int ret;
 
 	down_write(&mm->mmap_sem);
-	if (mm->uprobes_state.xol_area)
+	if (mm->uprobes_state.xol_area) {
+		ret = -EALREADY;
 		goto fail;
+	}
 
 	if (!area->vaddr) {
 		/* Try to map as high as possible, this is only a hint. */
@@ -1148,11 +1145,15 @@ static int xol_add_vma(struct mm_struct *mm, struct xol_area *area)
 		}
 	}
 
-	ret = install_special_mapping(mm, area->vaddr, PAGE_SIZE,
-				VM_EXEC|VM_MAYEXEC|VM_DONTCOPY|VM_IO, &area->page);
-	if (ret)
+	vma = _install_special_mapping(mm, area->vaddr, PAGE_SIZE,
+				VM_EXEC|VM_MAYEXEC|VM_DONTCOPY|VM_IO,
+				&area->xol_mapping);
+	if (IS_ERR(vma)) {
+		ret = PTR_ERR(vma);
 		goto fail;
+	}
 
+	ret = 0;
 	smp_wmb();	/* pairs with get_xol_area() */
 	mm->uprobes_state.xol_area = area;
  fail:
@@ -1175,21 +1176,24 @@ static struct xol_area *__create_xol_area(unsigned long vaddr)
 	if (!area->bitmap)
 		goto free_area;
 
-	area->page = alloc_page(GFP_HIGHUSER);
-	if (!area->page)
+	area->xol_mapping.name = "[uprobes]";
+	area->xol_mapping.pages = area->pages;
+	area->pages[0] = alloc_page(GFP_HIGHUSER);
+	if (!area->pages[0])
 		goto free_bitmap;
+	area->pages[1] = NULL;
 
 	area->vaddr = vaddr;
 	init_waitqueue_head(&area->wq);
 	/* Reserve the 1st slot for get_trampoline_vaddr() */
 	set_bit(0, area->bitmap);
 	atomic_set(&area->slot_count, 1);
-	copy_to_page(area->page, 0, &insn, UPROBE_SWBP_INSN_SIZE);
+	copy_to_page(area->pages[0], 0, &insn, UPROBE_SWBP_INSN_SIZE);
 
 	if (!xol_add_vma(mm, area))
 		return area;
 
-	__free_page(area->page);
+	__free_page(area->pages[0]);
  free_bitmap:
 	kfree(area->bitmap);
  free_area:
@@ -1227,7 +1231,7 @@ void uprobe_clear_state(struct mm_struct *mm)
 	if (!area)
 		return;
 
-	put_page(area->page);
+	put_page(area->pages[0]);
 	kfree(area->bitmap);
 	kfree(area);
 }
@@ -1296,7 +1300,7 @@ static unsigned long xol_get_insn_slot(struct uprobe *uprobe)
 	if (unlikely(!xol_vaddr))
 		return 0;
 
-	arch_uprobe_copy_ixol(area->page, xol_vaddr,
+	arch_uprobe_copy_ixol(area->pages[0], xol_vaddr,
 			      &uprobe->arch.ixol, sizeof(uprobe->arch.ixol));
 
 	return xol_vaddr;
@@ -1333,6 +1337,7 @@ static void xol_free_insn_slot(struct task_struct *tsk)
 
 		clear_bit(slot_nr, area->bitmap);
 		atomic_dec(&area->slot_count);
+		smp_mb__after_atomic(); /* pairs with prepare_to_wait() */
 		if (waitqueue_active(&area->wq))
 			wake_up(&area->wq);
 
@@ -1376,6 +1381,14 @@ unsigned long uprobe_get_trap_addr(struct pt_regs *regs)
 	return instruction_pointer(regs);
 }
 
+static struct return_instance *free_ret_instance(struct return_instance *ri)
+{
+	struct return_instance *next = ri->next;
+	put_uprobe(ri->uprobe);
+	kfree(ri);
+	return next;
+}
+
 /*
  * Called with no locks held.
  * Called in context of a exiting or a exec-ing thread.
@@ -1383,7 +1396,7 @@ unsigned long uprobe_get_trap_addr(struct pt_regs *regs)
 void uprobe_free_utask(struct task_struct *t)
 {
 	struct uprobe_task *utask = t->utask;
-	struct return_instance *ri, *tmp;
+	struct return_instance *ri;
 
 	if (!utask)
 		return;
@@ -1392,13 +1405,8 @@ void uprobe_free_utask(struct task_struct *t)
 		put_uprobe(utask->active_uprobe);
 
 	ri = utask->return_instances;
-	while (ri) {
-		tmp = ri;
-		ri = ri->next;
-
-		put_uprobe(tmp->uprobe);
-		kfree(tmp);
-	}
+	while (ri)
+		ri = free_ret_instance(ri);
 
 	xol_free_insn_slot(t);
 	kfree(utask);
@@ -1437,7 +1445,7 @@ static int dup_utask(struct task_struct *t, struct uprobe_task *o_utask)
 			return -ENOMEM;
 
 		*n = *o;
-		atomic_inc(&n->uprobe->ref);
+		get_uprobe(n->uprobe);
 		n->next = NULL;
 
 		*p = n;
@@ -1515,12 +1523,25 @@ static unsigned long get_trampoline_vaddr(void)
 	return trampoline_vaddr;
 }
 
+static void cleanup_return_instances(struct uprobe_task *utask, bool chained,
+					struct pt_regs *regs)
+{
+	struct return_instance *ri = utask->return_instances;
+	enum rp_check ctx = chained ? RP_CHECK_CHAIN_CALL : RP_CHECK_CALL;
+
+	while (ri && !arch_uretprobe_is_alive(ri, ctx, regs)) {
+		ri = free_ret_instance(ri);
+		utask->depth--;
+	}
+	utask->return_instances = ri;
+}
+
 static void prepare_uretprobe(struct uprobe *uprobe, struct pt_regs *regs)
 {
 	struct return_instance *ri;
 	struct uprobe_task *utask;
 	unsigned long orig_ret_vaddr, trampoline_vaddr;
-	bool chained = false;
+	bool chained;
 
 	if (!get_xol_area())
 		return;
@@ -1536,49 +1557,47 @@ static void prepare_uretprobe(struct uprobe *uprobe, struct pt_regs *regs)
 		return;
 	}
 
-	ri = kzalloc(sizeof(struct return_instance), GFP_KERNEL);
+	ri = kmalloc(sizeof(struct return_instance), GFP_KERNEL);
 	if (!ri)
-		goto fail;
+		return;
 
 	trampoline_vaddr = get_trampoline_vaddr();
 	orig_ret_vaddr = arch_uretprobe_hijack_return_addr(trampoline_vaddr, regs);
 	if (orig_ret_vaddr == -1)
 		goto fail;
 
+	/* drop the entries invalidated by longjmp() */
+	chained = (orig_ret_vaddr == trampoline_vaddr);
+	cleanup_return_instances(utask, chained, regs);
+
 	/*
 	 * We don't want to keep trampoline address in stack, rather keep the
 	 * original return address of first caller thru all the consequent
 	 * instances. This also makes breakpoint unwrapping easier.
 	 */
-	if (orig_ret_vaddr == trampoline_vaddr) {
+	if (chained) {
 		if (!utask->return_instances) {
 			/*
 			 * This situation is not possible. Likely we have an
 			 * attack from user-space.
 			 */
-			pr_warn("uprobe: unable to set uretprobe pid/tgid=%d/%d\n",
-						current->pid, current->tgid);
+			uprobe_warn(current, "handle tail call");
 			goto fail;
 		}
-
-		chained = true;
 		orig_ret_vaddr = utask->return_instances->orig_ret_vaddr;
 	}
 
-	atomic_inc(&uprobe->ref);
-	ri->uprobe = uprobe;
+	ri->uprobe = get_uprobe(uprobe);
 	ri->func = instruction_pointer(regs);
+	ri->stack = user_stack_pointer(regs);
 	ri->orig_ret_vaddr = orig_ret_vaddr;
 	ri->chained = chained;
 
 	utask->depth++;
-
-	/* add instance to the stack */
 	ri->next = utask->return_instances;
 	utask->return_instances = ri;
 
 	return;
-
  fail:
 	kfree(ri);
 }
@@ -1766,46 +1785,58 @@ handle_uretprobe_chain(struct return_instance *ri, struct pt_regs *regs)
 	up_read(&uprobe->register_rwsem);
 }
 
-static bool handle_trampoline(struct pt_regs *regs)
+static struct return_instance *find_next_ret_chain(struct return_instance *ri)
 {
-	struct uprobe_task *utask;
-	struct return_instance *ri, *tmp;
 	bool chained;
 
+	do {
+		chained = ri->chained;
+		ri = ri->next;	/* can't be NULL if chained */
+	} while (chained);
+
+	return ri;
+}
+
+static void handle_trampoline(struct pt_regs *regs)
+{
+	struct uprobe_task *utask;
+	struct return_instance *ri, *next;
+	bool valid;
+
 	utask = current->utask;
 	if (!utask)
-		return false;
+		goto sigill;
 
 	ri = utask->return_instances;
 	if (!ri)
-		return false;
-
-	/*
-	 * TODO: we should throw out return_instance's invalidated by
-	 * longjmp(), currently we assume that the probed function always
-	 * returns.
-	 */
-	instruction_pointer_set(regs, ri->orig_ret_vaddr);
-
-	for (;;) {
-		handle_uretprobe_chain(ri, regs);
-
-		chained = ri->chained;
-		put_uprobe(ri->uprobe);
-
-		tmp = ri;
-		ri = ri->next;
-		kfree(tmp);
-		utask->depth--;
+		goto sigill;
 
-		if (!chained)
-			break;
-		BUG_ON(!ri);
-	}
+	do {
+		/*
+		 * We should throw out the frames invalidated by longjmp().
+		 * If this chain is valid, then the next one should be alive
+		 * or NULL; the latter case means that nobody but ri->func
+		 * could hit this trampoline on return. TODO: sigaltstack().
+		 */
+		next = find_next_ret_chain(ri);
+		valid = !next || arch_uretprobe_is_alive(next, RP_CHECK_RET, regs);
+
+		instruction_pointer_set(regs, ri->orig_ret_vaddr);
+		do {
+			if (valid)
+				handle_uretprobe_chain(ri, regs);
+			ri = free_ret_instance(ri);
+			utask->depth--;
+		} while (ri != next);
+	} while (!valid);
 
 	utask->return_instances = ri;
+	return;
+
+ sigill:
+	uprobe_warn(current, "handle uretprobe, sending SIGILL.");
+	force_sig_info(SIGILL, SEND_SIG_FORCED, current);
 
-	return true;
 }
 
 bool __weak arch_uprobe_ignore(struct arch_uprobe *aup, struct pt_regs *regs)
@@ -1813,6 +1844,12 @@ bool __weak arch_uprobe_ignore(struct arch_uprobe *aup, struct pt_regs *regs)
 	return false;
 }
 
+bool __weak arch_uretprobe_is_alive(struct return_instance *ret, enum rp_check ctx,
+					struct pt_regs *regs)
+{
+	return true;
+}
+
 /*
  * Run handler and ask thread to singlestep.
  * Ensure all non-fatal signals cannot interrupt thread while it singlesteps.
@@ -1824,13 +1861,8 @@ static void handle_swbp(struct pt_regs *regs)
 	int uninitialized_var(is_swbp);
 
 	bp_vaddr = uprobe_get_swbp_addr(regs);
-	if (bp_vaddr == get_trampoline_vaddr()) {
-		if (handle_trampoline(regs))
-			return;
-
-		pr_warn("uprobe: unable to handle uretprobe pid/tgid=%d/%d\n",
-						current->pid, current->tgid);
-	}
+	if (bp_vaddr == get_trampoline_vaddr())
+		return handle_trampoline(regs);
 
 	uprobe = find_active_uprobe(bp_vaddr, &is_swbp);
 	if (!uprobe) {