[PATCH 07/15] uprobes: document uprobe_register() vs uprobe_mmap() race

[Oleg Nesterov <oleg@xxxxxxxxxx>]

From: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>

Because the mind is treacherous and makes us forget we need to write
stuff down.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Signed-off-by: Oleg Nesterov <oleg@xxxxxxxxxx>
---
 kernel/events/uprobes.c |   31 ++++++++++++++++++++++++++++---
 1 files changed, 28 insertions(+), 3 deletions(-)

diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 897417d..2671d9a 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -44,6 +44,23 @@ 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.
+ */
+
 /* serialize (un)register */
 static struct mutex uprobes_mutex[UPROBES_HASH_SZ];
 
@@ -339,7 +356,9 @@ out:
 int __weak set_swbp(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long vaddr)
 {
 	int result;
-
+	/*
+	 * See the comment near uprobes_hash().
+	 */
 	result = is_swbp_at_addr(mm, vaddr);
 	if (result == 1)
 		return -EEXIST;
@@ -845,6 +864,10 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
 
 		if (is_register) {
 			err = install_breakpoint(uprobe, mm, vma, info->vaddr);
+			/*
+			 * We can race against uprobe_mmap(), see the
+			 * comment near uprobe_hash().
+			 */
 			if (err == -EEXIST)
 				err = 0;
 		} else {
@@ -1054,8 +1077,10 @@ int uprobe_mmap(struct vm_area_struct *vma)
 			}
 
 			ret = install_breakpoint(uprobe, vma->vm_mm, vma, vaddr);
-
-			/* Ignore double add: */
+			/*
+			 * We can race against uprobe_register(), see the
+			 * comment near uprobe_hash().
+			 */
 			if (ret == -EEXIST) {
 				ret = 0;
 
-- 
1.5.5.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/