All of lore.kernel.org
 help / color / mirror / Atom feed
From: Byungchul Park <byungchul.park@lge.com>
To: peterz@infradead.org, mingo@kernel.org
Cc: tglx@linutronix.de, walken@google.com, boqun.feng@gmail.com,
	kirill@shutemov.name, linux-kernel@vger.kernel.org,
	linux-mm@kvack.org, iamjoonsoo.kim@lge.com,
	akpm@linux-foundation.org, npiggin@gmail.com
Subject: [PATCH v4 04/15] lockdep: Add a function building a chain between two classes
Date: Fri,  9 Dec 2016 14:12:00 +0900	[thread overview]
Message-ID: <1481260331-360-5-git-send-email-byungchul.park@lge.com> (raw)
In-Reply-To: <1481260331-360-1-git-send-email-byungchul.park@lge.com>

add_chain_cache() should be used in the context where the hlock is
owned since it might be racy in another context. However crossrelease
feature needs to build a chain between two locks regardless of context.
So introduce a new function making it possible.

Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
 kernel/locking/lockdep.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 56 insertions(+)

diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 5df56aa..111839f 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -2105,6 +2105,62 @@ static int check_no_collision(struct task_struct *curr,
 	return 1;
 }
 
+/*
+ * This is for building a chain between just two different classes,
+ * instead of adding a new hlock upon current, which is done by
+ * add_chain_cache().
+ *
+ * This can be called in any context with two classes, while
+ * add_chain_cache() must be done within the lock owener's context
+ * since it uses hlock which might be racy in another context.
+ */
+static inline int add_chain_cache_classes(unsigned int prev,
+					  unsigned int next,
+					  unsigned int irq_context,
+					  u64 chain_key)
+{
+	struct hlist_head *hash_head = chainhashentry(chain_key);
+	struct lock_chain *chain;
+
+	/*
+	 * Allocate a new chain entry from the static array, and add
+	 * it to the hash:
+	 */
+
+	/*
+	 * We might need to take the graph lock, ensure we've got IRQs
+	 * disabled to make this an IRQ-safe lock.. for recursion reasons
+	 * lockdep won't complain about its own locking errors.
+	 */
+	if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
+		return 0;
+
+	if (unlikely(nr_lock_chains >= MAX_LOCKDEP_CHAINS)) {
+		if (!debug_locks_off_graph_unlock())
+			return 0;
+
+		print_lockdep_off("BUG: MAX_LOCKDEP_CHAINS too low!");
+		dump_stack();
+		return 0;
+	}
+
+	chain = lock_chains + nr_lock_chains++;
+	chain->chain_key = chain_key;
+	chain->irq_context = irq_context;
+	chain->depth = 2;
+	if (likely(nr_chain_hlocks + chain->depth <= MAX_LOCKDEP_CHAIN_HLOCKS)) {
+		chain->base = nr_chain_hlocks;
+		nr_chain_hlocks += chain->depth;
+		chain_hlocks[chain->base] = prev - 1;
+		chain_hlocks[chain->base + 1] = next -1;
+	}
+	hlist_add_head_rcu(&chain->entry, hash_head);
+	debug_atomic_inc(chain_lookup_misses);
+	inc_chains();
+
+	return 1;
+}
+
 static inline int add_chain_cache(struct task_struct *curr,
 				  struct held_lock *hlock,
 				  u64 chain_key)
-- 
1.9.1

WARNING: multiple messages have this Message-ID (diff)
From: Byungchul Park <byungchul.park@lge.com>
To: peterz@infradead.org, mingo@kernel.org
Cc: tglx@linutronix.de, walken@google.com, boqun.feng@gmail.com,
	kirill@shutemov.name, linux-kernel@vger.kernel.org,
	linux-mm@kvack.org, iamjoonsoo.kim@lge.com,
	akpm@linux-foundation.org, npiggin@gmail.com
Subject: [PATCH v4 04/15] lockdep: Add a function building a chain between two classes
Date: Fri,  9 Dec 2016 14:12:00 +0900	[thread overview]
Message-ID: <1481260331-360-5-git-send-email-byungchul.park@lge.com> (raw)
In-Reply-To: <1481260331-360-1-git-send-email-byungchul.park@lge.com>

add_chain_cache() should be used in the context where the hlock is
owned since it might be racy in another context. However crossrelease
feature needs to build a chain between two locks regardless of context.
So introduce a new function making it possible.

Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
 kernel/locking/lockdep.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 56 insertions(+)

diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 5df56aa..111839f 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -2105,6 +2105,62 @@ static int check_no_collision(struct task_struct *curr,
 	return 1;
 }
 
+/*
+ * This is for building a chain between just two different classes,
+ * instead of adding a new hlock upon current, which is done by
+ * add_chain_cache().
+ *
+ * This can be called in any context with two classes, while
+ * add_chain_cache() must be done within the lock owener's context
+ * since it uses hlock which might be racy in another context.
+ */
+static inline int add_chain_cache_classes(unsigned int prev,
+					  unsigned int next,
+					  unsigned int irq_context,
+					  u64 chain_key)
+{
+	struct hlist_head *hash_head = chainhashentry(chain_key);
+	struct lock_chain *chain;
+
+	/*
+	 * Allocate a new chain entry from the static array, and add
+	 * it to the hash:
+	 */
+
+	/*
+	 * We might need to take the graph lock, ensure we've got IRQs
+	 * disabled to make this an IRQ-safe lock.. for recursion reasons
+	 * lockdep won't complain about its own locking errors.
+	 */
+	if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
+		return 0;
+
+	if (unlikely(nr_lock_chains >= MAX_LOCKDEP_CHAINS)) {
+		if (!debug_locks_off_graph_unlock())
+			return 0;
+
+		print_lockdep_off("BUG: MAX_LOCKDEP_CHAINS too low!");
+		dump_stack();
+		return 0;
+	}
+
+	chain = lock_chains + nr_lock_chains++;
+	chain->chain_key = chain_key;
+	chain->irq_context = irq_context;
+	chain->depth = 2;
+	if (likely(nr_chain_hlocks + chain->depth <= MAX_LOCKDEP_CHAIN_HLOCKS)) {
+		chain->base = nr_chain_hlocks;
+		nr_chain_hlocks += chain->depth;
+		chain_hlocks[chain->base] = prev - 1;
+		chain_hlocks[chain->base + 1] = next -1;
+	}
+	hlist_add_head_rcu(&chain->entry, hash_head);
+	debug_atomic_inc(chain_lookup_misses);
+	inc_chains();
+
+	return 1;
+}
+
 static inline int add_chain_cache(struct task_struct *curr,
 				  struct held_lock *hlock,
 				  u64 chain_key)
-- 
1.9.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2016-12-09  5:19 UTC|newest]

Thread overview: 104+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-09  5:11 [PATCH v4 00/15] lockdep: Implement crossrelease feature Byungchul Park
2016-12-09  5:11 ` Byungchul Park
2016-12-09  5:11 ` [PATCH v4 01/15] x86/dumpstack: Optimize save_stack_trace Byungchul Park
2016-12-09  5:11   ` Byungchul Park
2016-12-09  5:11 ` [PATCH v4 02/15] x86/dumpstack: Add save_stack_trace()_fast() Byungchul Park
2016-12-09  5:11   ` Byungchul Park
2016-12-09  5:11 ` [PATCH v4 03/15] lockdep: Refactor lookup_chain_cache() Byungchul Park
2016-12-09  5:11   ` Byungchul Park
2016-12-09  5:12 ` Byungchul Park [this message]
2016-12-09  5:12   ` [PATCH v4 04/15] lockdep: Add a function building a chain between two classes Byungchul Park
2017-01-10 21:00   ` Peter Zijlstra
2017-01-10 21:00     ` Peter Zijlstra
2017-01-12  1:41     ` Byungchul Park
2017-01-12  1:41       ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 05/15] lockdep: Make check_prev_add can use a separate stack_trace Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2017-01-12 16:16   ` Peter Zijlstra
2017-01-12 16:16     ` Peter Zijlstra
2017-01-13  2:45     ` Byungchul Park
2017-01-13  2:45       ` Byungchul Park
2017-01-13  4:09     ` [PATCH] lockdep: Make a stack_trace instance passed to check_prev_add as an arg Byungchul Park
2017-01-13  4:38       ` Byungchul Park
2017-01-13 10:11     ` [PATCH v4 05/15] lockdep: Make check_prev_add can use a separate stack_trace Byungchul Park
2017-01-13 10:11       ` Byungchul Park
2017-01-13 10:17       ` [PATCH 1/2] lockdep: Refactor save_trace() Byungchul Park
2017-01-13 10:17       ` [PATCH 2/2] lockdep: Pass a callback arg to check_prev_add() to handle stack_trace Byungchul Park
2017-01-17 15:54       ` [PATCH v4 05/15] lockdep: Make check_prev_add can use a separate stack_trace Peter Zijlstra
2017-01-17 15:54         ` Peter Zijlstra
2017-01-18  2:04         ` Byungchul Park
2017-01-18  2:04           ` Byungchul Park
2017-01-18 15:10           ` Peter Zijlstra
2017-01-18 15:10             ` Peter Zijlstra
2017-01-19  2:47             ` Byungchul Park
2017-01-19  2:47               ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 06/15] lockdep: Make save_trace can skip stack tracing of the current Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2017-01-12 16:37   ` Peter Zijlstra
2017-01-12 16:37     ` Peter Zijlstra
2017-01-13  0:18     ` Byungchul Park
2017-01-13  0:18       ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 07/15] lockdep: Implement crossrelease feature Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2017-01-13  4:39   ` Lai Jiangshan
2017-01-13  4:39     ` Lai Jiangshan
2017-01-13  5:02     ` Byungchul Park
2017-01-13  5:02       ` Byungchul Park
2017-01-16 15:10   ` Peter Zijlstra
2017-01-16 15:10     ` Peter Zijlstra
2017-01-17  2:05     ` Byungchul Park
2017-01-17  2:05       ` Byungchul Park
2017-01-17  7:12       ` Peter Zijlstra
2017-01-17  7:12         ` Peter Zijlstra
2017-01-17  7:49         ` Byungchul Park
2017-01-17  7:49           ` Byungchul Park
2017-01-17  7:14       ` Peter Zijlstra
2017-01-17  7:14         ` Peter Zijlstra
2017-01-17  7:45         ` Byungchul Park
2017-01-17  7:45           ` Byungchul Park
2017-01-16 15:13   ` Peter Zijlstra
2017-01-16 15:13     ` Peter Zijlstra
2017-01-17  2:33     ` Byungchul Park
2017-01-17  2:33       ` Byungchul Park
2017-01-17  6:24       ` Boqun Feng
2017-01-17  7:43         ` Byungchul Park
2017-01-17  7:43           ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 08/15] lockdep: Make crossrelease use save_stack_trace_fast() Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 09/15] lockdep: Make print_circular_bug() crosslock-aware Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 10/15] lockdep: Apply crossrelease to completion operation Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 11/15] pagemap.h: Remove trailing white space Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 12/15] lockdep: Apply crossrelease to PG_locked lock Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 13/15] lockdep: Apply lock_acquire(release) on __Set(__Clear)PageLocked Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 14/15] lockdep: Move data used in CONFIG_LOCKDEP_PAGELOCK from page to page_ext Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2016-12-09  5:12 ` [PATCH v4 15/15] lockdep: Crossrelease feature documentation Byungchul Park
2016-12-09  5:12   ` Byungchul Park
2017-01-10 20:08   ` Peter Zijlstra
2017-01-10 20:08     ` Peter Zijlstra
2017-01-11  1:29     ` Byungchul Park
2017-01-11  1:29       ` Byungchul Park
2017-01-18  6:42   ` Boqun Feng
2017-01-18 10:53     ` Byungchul Park
2017-01-18 10:53       ` Byungchul Park
2017-01-18 11:03       ` Peter Zijlstra
2017-01-18 11:03         ` Peter Zijlstra
2017-01-18 11:54         ` Byungchul Park
2017-01-18 11:54           ` Byungchul Park
2017-01-18 12:07           ` Peter Zijlstra
2017-01-18 12:07             ` Peter Zijlstra
2017-01-18 12:14             ` byungchul.park
2017-01-18 12:14               ` byungchul.park
2017-01-18 14:12               ` Peter Zijlstra
2017-01-18 14:12                 ` Peter Zijlstra
2017-01-19  1:54                 ` Byungchul Park
2017-01-19  1:54                   ` Byungchul Park
2017-01-18 12:49             ` byungchul.park
2017-01-18 12:49               ` byungchul.park
2016-12-09  5:21 ` [FYI] Output of 'cat /proc/lockdep' after applying crossrelease Byungchul Park
2016-12-09  5:21   ` Byungchul Park

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1481260331-360-5-git-send-email-byungchul.park@lge.com \
    --to=byungchul.park@lge.com \
    --cc=akpm@linux-foundation.org \
    --cc=boqun.feng@gmail.com \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=kirill@shutemov.name \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mingo@kernel.org \
    --cc=npiggin@gmail.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=walken@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.