linux-kernel - [PATCH v3 01/30] locking/lockdep: Rename deadlock check functions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20190628091528.17059-2-duyuyang@gmail.com>
Date:   Fri, 28 Jun 2019 17:14:59 +0800
From:   Yuyang Du <duyuyang@...il.com>
To:     peterz@...radead.org, will.deacon@....com, mingo@...nel.org
Cc:     bvanassche@....org, ming.lei@...hat.com, frederic@...nel.org,
        tglx@...utronix.de, linux-kernel@...r.kernel.org,
        longman@...hat.com, paulmck@...ux.vnet.ibm.com,
        boqun.feng@...il.com, Yuyang Du <duyuyang@...il.com>
Subject: [PATCH v3 01/30] locking/lockdep: Rename deadlock check functions

Deadlock checks are performed at two places:

 - Within current's held lock stack, check for lock recursion deadlock.
 - Within dependency graph, check for lock inversion deadlock.

Rename the two relevant functions for later use. Plus, with
recursive-read locks, only a dependency circle in graph is not a
sufficient condition for lock inversion deadlocks anymore, so
check_noncircular() is not entirely accurate.

No functional change.

Signed-off-by: Yuyang Du <duyuyang@...il.com>
---
 kernel/locking/lockdep.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 341f521..e30e9e4 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -1714,8 +1714,8 @@ unsigned long lockdep_count_backward_deps(struct lock_class *class)
  * Print an error and return 0 if it does.
  */
 static noinline int
-check_noncircular(struct held_lock *src, struct held_lock *target,
-		  struct lock_trace *trace)
+check_deadlock_graph(struct held_lock *src, struct held_lock *target,
+		     struct lock_trace *trace)
 {
 	int ret;
 	struct lock_list *uninitialized_var(target_entry);
@@ -2302,7 +2302,8 @@ static inline void inc_chains(void)
 }
 
 /*
- * Check whether we are holding such a class already.
+ * Check whether we are holding such a class already in the current
+ * held lock stack.
  *
  * (Note that this has to be done separately, because the graph cannot
  * detect such classes of deadlocks.)
@@ -2310,7 +2311,7 @@ static inline void inc_chains(void)
  * Returns: 0 on deadlock detected, 1 on OK, 2 on recursive read
  */
 static int
-check_deadlock(struct task_struct *curr, struct held_lock *next)
+check_deadlock_current(struct task_struct *curr, struct held_lock *next)
 {
 	struct held_lock *prev;
 	struct held_lock *nest = NULL;
@@ -2394,7 +2395,7 @@ static inline void inc_chains(void)
 
 	/*
 	 * Prove that the new <prev> -> <next> dependency would not
-	 * create a circular dependency in the graph. (We do this by
+	 * create a deadlock scenario in the graph. (We do this by
 	 * a breadth-first search into the graph starting at <next>,
 	 * and check whether we can reach <prev>.)
 	 *
@@ -2402,7 +2403,7 @@ static inline void inc_chains(void)
 	 * MAX_CIRCULAR_QUEUE_SIZE) which keeps track of a breadth of nodes
 	 * in the graph whose neighbours are to be checked.
 	 */
-	ret = check_noncircular(next, prev, trace);
+	ret = check_deadlock_graph(next, prev, trace);
 	if (unlikely(ret <= 0))
 		return 0;
 
@@ -2878,7 +2879,7 @@ static int validate_chain(struct task_struct *curr,
 		 * The simple case: does the current hold the same lock
 		 * already?
 		 */
-		int ret = check_deadlock(curr, hlock);
+		int ret = check_deadlock_current(curr, hlock);
 
 		if (!ret)
 			return 0;
-- 
1.8.3.1