[GIT PULL] locking changes for v4.6

From: Ingo Molnar
Date: Mon Mar 14 2016 - 07:15:10 EST

Linus,

Please pull the latest locking-core-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git locking-core-for-linus

# HEAD: 38460a2178d225b39ade5ac66586c3733391cf86 locking/csd_lock: Use smp_cond_acquire() in csd_lock_wait()

Various updates:

- Futex scalability improvements: remove page lock use for shared futex
get_futex_key(), which speeds up 'perf bench futex hash' benchmarks by over
40% on a 60-core Westmere. This makes anon-mem shared futexes perform close to
private futexes. (Mel Gorman)

- lockdep hash collision detection and fix (Alfredo Alvarez Fernandez)

- lockdep testing enhancements (Alfredo Alvarez Fernandez)

- robustify lockdep init by using hlists (Andrew Morton, Andrey Ryabinin)

- mutex and csd_lock micro-optimizations (Davidlohr Bueso)

- small x86 barriers tweaks (Michael S. Tsirkin)

- qspinlock updates (Waiman Long)

Thanks,

Ingo

------------------>
Alfredo Alvarez Fernandez (3):
tools/lib/lockdep: Add userspace version of READ_ONCE()
tools/lib/lockdep: Add tests for AA and ABBA locking
locking/lockdep: Prevent chain_key collisions

Andrew Morton (1):
locking/lockdep: Convert hash tables to hlists

Andrey Ryabinin (1):
locking/lockdep: Eliminate lockdep_init()

Arnd Bergmann (1):
locking/static_keys: Avoid nested functions

Dan Streetman (1):
locking/qspinlock: Move __ARCH_SPIN_LOCK_UNLOCKED to qspinlock_types.h

Davidlohr Bueso (4):
futex: Rename barrier references in ordering guarantees
locking/mutex: Allow next waiter lockless wakeup
locking/csd_lock: Explicitly inline csd_lock*() helpers
locking/csd_lock: Use smp_cond_acquire() in csd_lock_wait()

Ingo Molnar (3):
tools/lib/lockdep: Fix the build on recent kernels
tools/lib/lockdep: Fix link creation warning
locking/lockdep: Detect chain_key collisions

Jianyu Zhan (1):
futex: Replace barrier() in unqueue_me() with READ_ONCE()

Konrad Rzeszutek Wilk (1):
locking/atomics: Update comment about READ_ONCE() and structures

Mel Gorman (1):
futex: Remove requirement for lock_page() in get_futex_key()

Michael S. Tsirkin (4):
locking/x86: Add cc clobber for ADDL
locking/x86: Drop a comment left over from X86_OOSTORE
locking/x86: Tweak the comment about use of wmb() for IO
locking/x86: Use mb() around clflush()

Peter Zijlstra (1):
locking/mcs: Fix mcs_spin_lock() ordering

Waiman Long (3):
locking/pvqspinlock: Move lock stealing count tracking code into pv_queued_spin_steal_lock()
locking/qspinlock: Use smp_cond_acquire() in pending code
locking/pvqspinlock: Enable slowpath locking count tracking


arch/c6x/kernel/setup.c | 2 -
arch/microblaze/kernel/setup.c | 2 -
arch/powerpc/kernel/setup_32.c | 2 -
arch/powerpc/kernel/setup_64.c | 3 -
arch/s390/kernel/early.c | 1 -
arch/sparc/kernel/head_64.S | 8 --
arch/x86/include/asm/barrier.h | 15 ++-
arch/x86/kernel/process.c | 4 +-
arch/x86/lguest/boot.c | 6 --
include/asm-generic/qspinlock.h | 5 -
include/asm-generic/qspinlock_types.h | 5 +
include/linux/compiler.h | 5 +-
include/linux/lockdep.h | 2 -
init/main.c | 5 -
kernel/futex.c | 139 +++++++++++++++++++++-----
kernel/locking/lockdep.c | 132 +++++++++++-------------
kernel/locking/mcs_spinlock.h | 8 +-
kernel/locking/mutex.c | 5 +-
kernel/locking/qspinlock.c | 7 +-
kernel/locking/qspinlock_paravirt.h | 21 ++--
kernel/locking/qspinlock_stat.h | 16 +--
kernel/smp.c | 9 +-
lib/test_static_keys.c | 62 ++++++++----
tools/lib/lockdep/Makefile | 2 +-
tools/lib/lockdep/common.c | 5 -
tools/lib/lockdep/include/liblockdep/common.h | 1 -
tools/lib/lockdep/lockdep.c | 6 ++
tools/lib/lockdep/preload.c | 2 -
tools/lib/lockdep/tests/AA.c | 8 +-
tools/lib/lockdep/tests/ABA.c | 13 +++
tools/lib/lockdep/tests/ABBA_2threads.c | 46 +++++++++
tools/lib/lockdep/uinclude/linux/compiler.h | 1 +
32 files changed, 334 insertions(+), 214 deletions(-)
create mode 100644 tools/lib/lockdep/tests/ABA.c
create mode 100644 tools/lib/lockdep/tests/ABBA_2threads.c

diff --git a/arch/c6x/kernel/setup.c b/arch/c6x/kernel/setup.c
index 72e17f7ebd6f..786e36e2f61d 100644
--- a/arch/c6x/kernel/setup.c
+++ b/arch/c6x/kernel/setup.c
@@ -281,8 +281,6 @@ notrace void __init machine_init(unsigned long dt_ptr)
*/
set_ist(_vectors_start);

- lockdep_init();
-
/*
* dtb is passed in from bootloader.
* fdt is linked in blob.
diff --git a/arch/microblaze/kernel/setup.c b/arch/microblaze/kernel/setup.c
index 89a2a9394927..f31ebb5dc26c 100644
--- a/arch/microblaze/kernel/setup.c
+++ b/arch/microblaze/kernel/setup.c
@@ -130,8 +130,6 @@ void __init machine_early_init(const char *cmdline, unsigned int ram,
memset(__bss_start, 0, __bss_stop-__bss_start);
memset(_ssbss, 0, _esbss-_ssbss);

- lockdep_init();
-
/* initialize device tree for usage in early_printk */
early_init_devtree(_fdt_start);

diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index ad8c9db61237..d544fa311757 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -114,8 +114,6 @@ extern unsigned int memset_nocache_branch; /* Insn to be replaced by NOP */

notrace void __init machine_init(u64 dt_ptr)
{
- lockdep_init();
-
/* Enable early debugging if any specified (see udbg.h) */
udbg_early_init();

diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 5c03a6a9b054..f98be8383a39 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -255,9 +255,6 @@ void __init early_setup(unsigned long dt_ptr)
setup_paca(&boot_paca);
fixup_boot_paca();

- /* Initialize lockdep early or else spinlocks will blow */
- lockdep_init();
-
/* -------- printk is now safe to use ------- */

/* Enable early debugging if any specified (see udbg.h) */
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index c55576bbaa1f..a0684de5a93b 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -448,7 +448,6 @@ void __init startup_init(void)
rescue_initrd();
clear_bss_section();
init_kernel_storage_key();
- lockdep_init();
lockdep_off();
setup_lowcore_early();
setup_facility_list();
diff --git a/arch/sparc/kernel/head_64.S b/arch/sparc/kernel/head_64.S
index f2d30cab5b3f..cd1f592cd347 100644
--- a/arch/sparc/kernel/head_64.S
+++ b/arch/sparc/kernel/head_64.S
@@ -696,14 +696,6 @@
call __bzero
sub %o1, %o0, %o1

-#ifdef CONFIG_LOCKDEP
- /* We have this call this super early, as even prom_init can grab
- * spinlocks and thus call into the lockdep code.
- */
- call lockdep_init
- nop
-#endif
-
call prom_init
mov %l7, %o0 ! OpenPROM cif handler

diff --git a/arch/x86/include/asm/barrier.h b/arch/x86/include/asm/barrier.h
index a584e1c50918..bfb28caf97b1 100644
--- a/arch/x86/include/asm/barrier.h
+++ b/arch/x86/include/asm/barrier.h
@@ -6,18 +6,17 @@

/*
* Force strict CPU ordering.
- * And yes, this is required on UP too when we're talking
+ * And yes, this might be required on UP too when we're talking
* to devices.
*/

#ifdef CONFIG_X86_32
-/*
- * Some non-Intel clones support out of order store. wmb() ceases to be a
- * nop for these.
- */
-#define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
-#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
-#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
+#define mb() asm volatile(ALTERNATIVE("lock; addl $0,0(%%esp)", "mfence", \
+ X86_FEATURE_XMM2) ::: "memory", "cc")
+#define rmb() asm volatile(ALTERNATIVE("lock; addl $0,0(%%esp)", "lfence", \
+ X86_FEATURE_XMM2) ::: "memory", "cc")
+#define wmb() asm volatile(ALTERNATIVE("lock; addl $0,0(%%esp)", "sfence", \
+ X86_FEATURE_XMM2) ::: "memory", "cc")
#else
#define mb() asm volatile("mfence":::"memory")
#define rmb() asm volatile("lfence":::"memory")
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 9f7c21c22477..9decee2bfdbe 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -418,9 +418,9 @@ static void mwait_idle(void)
if (!current_set_polling_and_test()) {
trace_cpu_idle_rcuidle(1, smp_processor_id());
if (this_cpu_has(X86_BUG_CLFLUSH_MONITOR)) {
- smp_mb(); /* quirk */
+ mb(); /* quirk */
clflush((void *)&current_thread_info()->flags);
- smp_mb(); /* quirk */
+ mb(); /* quirk */
}

__monitor((void *)&current_thread_info()->flags, 0, 0);
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index 4ba229ac3f4f..f56cc418c87d 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -1520,12 +1520,6 @@ __init void lguest_init(void)
*/
reserve_top_address(lguest_data.reserve_mem);

- /*
- * If we don't initialize the lock dependency checker now, it crashes
- * atomic_notifier_chain_register, then paravirt_disable_iospace.
- */
- lockdep_init();
-
/* Hook in our special panic hypercall code. */
atomic_notifier_chain_register(&panic_notifier_list, &paniced);

diff --git a/include/asm-generic/qspinlock.h b/include/asm-generic/qspinlock.h
index 39e1cb201b8e..35a52a880b2f 100644
--- a/include/asm-generic/qspinlock.h
+++ b/include/asm-generic/qspinlock.h
@@ -120,11 +120,6 @@ static __always_inline bool virt_spin_lock(struct qspinlock *lock)
#endif

/*
- * Initializier
- */
-#define __ARCH_SPIN_LOCK_UNLOCKED { ATOMIC_INIT(0) }
-
-/*
* Remapping spinlock architecture specific functions to the corresponding
* queued spinlock functions.
*/
diff --git a/include/asm-generic/qspinlock_types.h b/include/asm-generic/qspinlock_types.h
index 85f888e86761..034acd0c4956 100644
--- a/include/asm-generic/qspinlock_types.h
+++ b/include/asm-generic/qspinlock_types.h
@@ -33,6 +33,11 @@ typedef struct qspinlock {
} arch_spinlock_t;

/*
+ * Initializier
+ */
+#define __ARCH_SPIN_LOCK_UNLOCKED { ATOMIC_INIT(0) }
+
+/*
* Bitfields in the atomic value:
*
* When NR_CPUS < 16K
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 48f5aab117ae..a27f4f17c382 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -263,8 +263,9 @@ static __always_inline void __write_once_size(volatile void *p, void *res, int s
* In contrast to ACCESS_ONCE these two macros will also work on aggregate
* data types like structs or unions. If the size of the accessed data
* type exceeds the word size of the machine (e.g., 32 bits or 64 bits)
- * READ_ONCE() and WRITE_ONCE() will fall back to memcpy and print a
- * compile-time warning.
+ * READ_ONCE() and WRITE_ONCE() will fall back to memcpy(). There's at
+ * least two memcpy()s: one for the __builtin_memcpy() and then one for
+ * the macro doing the copy of variable - '__u' allocated on the stack.
*
* Their two major use cases are: (1) Mediating communication between
* process-level code and irq/NMI handlers, all running on the same CPU,
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 4dca42fd32f5..d026b190c530 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -261,7 +261,6 @@ struct held_lock {
/*
* Initialization, self-test and debugging-output methods:
*/
-extern void lockdep_init(void);
extern void lockdep_info(void);
extern void lockdep_reset(void);
extern void lockdep_reset_lock(struct lockdep_map *lock);
@@ -392,7 +391,6 @@ static inline void lockdep_on(void)
# define lockdep_set_current_reclaim_state(g) do { } while (0)
# define lockdep_clear_current_reclaim_state() do { } while (0)
# define lockdep_trace_alloc(g) do { } while (0)
-# define lockdep_init() do { } while (0)
# define lockdep_info() do { } while (0)
# define lockdep_init_map(lock, name, key, sub) \
do { (void)(name); (void)(key); } while (0)
diff --git a/init/main.c b/init/main.c
index 58c9e374704b..b3008bcfb1dc 100644
--- a/init/main.c
+++ b/init/main.c
@@ -499,11 +499,6 @@ asmlinkage __visible void __init start_kernel(void)
char *command_line;
char *after_dashes;

- /*
- * Need to run as early as possible, to initialize the
- * lockdep hash:
- */
- lockdep_init();
set_task_stack_end_magic(&init_task);
smp_setup_processor_id();
debug_objects_early_init();
diff --git a/kernel/futex.c b/kernel/futex.c
index 5d6ce6413ef1..a5d2e74c89e0 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -124,16 +124,16 @@
* futex_wait(futex, val);
*
* waiters++; (a)
- * mb(); (A) <-- paired with -.
- * |
- * lock(hash_bucket(futex)); |
- * |
- * uval = *futex; |
- * | *futex = newval;
- * | sys_futex(WAKE, futex);
- * | futex_wake(futex);
- * |
- * `-------> mb(); (B)
+ * smp_mb(); (A) <-- paired with -.
+ * |
+ * lock(hash_bucket(futex)); |
+ * |
+ * uval = *futex; |
+ * | *futex = newval;
+ * | sys_futex(WAKE, futex);
+ * | futex_wake(futex);
+ * |
+ * `--------> smp_mb(); (B)
* if (uval == val)
* queue();
* unlock(hash_bucket(futex));
@@ -334,7 +334,7 @@ static inline void futex_get_mm(union futex_key *key)
/*
* Ensure futex_get_mm() implies a full barrier such that
* get_futex_key() implies a full barrier. This is relied upon
- * as full barrier (B), see the ordering comment above.
+ * as smp_mb(); (B), see the ordering comment above.
*/
smp_mb__after_atomic();
}
@@ -407,10 +407,10 @@ static void get_futex_key_refs(union futex_key *key)

switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) {
case FUT_OFF_INODE:
- ihold(key->shared.inode); /* implies MB (B) */
+ ihold(key->shared.inode); /* implies smp_mb(); (B) */
break;
case FUT_OFF_MMSHARED:
- futex_get_mm(key); /* implies MB (B) */
+ futex_get_mm(key); /* implies smp_mb(); (B) */
break;
default:
/*
@@ -418,7 +418,7 @@ static void get_futex_key_refs(union futex_key *key)
* mm, therefore the only purpose of calling get_futex_key_refs
* is because we need the barrier for the lockless waiter check.
*/
- smp_mb(); /* explicit MB (B) */
+ smp_mb(); /* explicit smp_mb(); (B) */
}
}

@@ -497,7 +497,7 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw)
if (!fshared) {
key->private.mm = mm;
key->private.address = address;
- get_futex_key_refs(key); /* implies MB (B) */
+ get_futex_key_refs(key); /* implies smp_mb(); (B) */
return 0;
}

@@ -520,7 +520,20 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw)
else
err = 0;

- lock_page(page);
+ /*
+ * The treatment of mapping from this point on is critical. The page
+ * lock protects many things but in this context the page lock
+ * stabilizes mapping, prevents inode freeing in the shared
+ * file-backed region case and guards against movement to swap cache.
+ *
+ * Strictly speaking the page lock is not needed in all cases being
+ * considered here and page lock forces unnecessarily serialization
+ * From this point on, mapping will be re-verified if necessary and
+ * page lock will be acquired only if it is unavoidable
+ */
+ page = compound_head(page);
+ mapping = READ_ONCE(page->mapping);
+
/*
* If page->mapping is NULL, then it cannot be a PageAnon
* page; but it might be the ZERO_PAGE or in the gate area or
@@ -536,19 +549,31 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw)
* shmem_writepage move it from filecache to swapcache beneath us:
* an unlikely race, but we do need to retry for page->mapping.
*/
- mapping = compound_head(page)->mapping;
- if (!mapping) {
- int shmem_swizzled = PageSwapCache(page);
+ if (unlikely(!mapping)) {
+ int shmem_swizzled;
+
+ /*
+ * Page lock is required to identify which special case above
+ * applies. If this is really a shmem page then the page lock
+ * will prevent unexpected transitions.
+ */
+ lock_page(page);
+ shmem_swizzled = PageSwapCache(page) || page->mapping;
unlock_page(page);
put_page(page);
+
if (shmem_swizzled)
goto again;
+
return -EFAULT;
}

/*
* Private mappings are handled in a simple way.
*
+ * If the futex key is stored on an anonymous page, then the associated
+ * object is the mm which is implicitly pinned by the calling process.
+ *
* NOTE: When userspace waits on a MAP_SHARED mapping, even if
* it's a read-only handle, it's expected that futexes attach to
* the object not the particular process.
@@ -566,16 +591,74 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw)
key->both.offset |= FUT_OFF_MMSHARED; /* ref taken on mm */
key->private.mm = mm;
key->private.address = address;
+
+ get_futex_key_refs(key); /* implies smp_mb(); (B) */
+
} else {
+ struct inode *inode;
+
+ /*
+ * The associated futex object in this case is the inode and
+ * the page->mapping must be traversed. Ordinarily this should
+ * be stabilised under page lock but it's not strictly
+ * necessary in this case as we just want to pin the inode, not
+ * update the radix tree or anything like that.
+ *
+ * The RCU read lock is taken as the inode is finally freed
+ * under RCU. If the mapping still matches expectations then the
+ * mapping->host can be safely accessed as being a valid inode.
+ */
+ rcu_read_lock();
+
+ if (READ_ONCE(page->mapping) != mapping) {
+ rcu_read_unlock();
+ put_page(page);
+
+ goto again;
+ }
+
+ inode = READ_ONCE(mapping->host);
+ if (!inode) {
+ rcu_read_unlock();
+ put_page(page);
+
+ goto again;
+ }
+
+ /*
+ * Take a reference unless it is about to be freed. Previously
+ * this reference was taken by ihold under the page lock
+ * pinning the inode in place so i_lock was unnecessary. The
+ * only way for this check to fail is if the inode was
+ * truncated in parallel so warn for now if this happens.
+ *
+ * We are not calling into get_futex_key_refs() in file-backed
+ * cases, therefore a successful atomic_inc return below will
+ * guarantee that get_futex_key() will still imply smp_mb(); (B).
+ */
+ if (WARN_ON_ONCE(!atomic_inc_not_zero(&inode->i_count))) {
+ rcu_read_unlock();
+ put_page(page);
+
+ goto again;
+ }
+
+ /* Should be impossible but lets be paranoid for now */
+ if (WARN_ON_ONCE(inode->i_mapping != mapping)) {
+ err = -EFAULT;
+ rcu_read_unlock();
+ iput(inode);
+
+ goto out;
+ }
+
key->both.offset |= FUT_OFF_INODE; /* inode-based key */
- key->shared.inode = mapping->host;
+ key->shared.inode = inode;
key->shared.pgoff = basepage_index(page);
+ rcu_read_unlock();
}

- get_futex_key_refs(key); /* implies MB (B) */
-
out:
- unlock_page(page);
put_page(page);
return err;
}
@@ -1864,7 +1947,7 @@ static inline struct futex_hash_bucket *queue_lock(struct futex_q *q)

q->lock_ptr = &hb->lock;

- spin_lock(&hb->lock); /* implies MB (A) */
+ spin_lock(&hb->lock); /* implies smp_mb(); (A) */
return hb;
}

@@ -1927,8 +2010,12 @@ static int unqueue_me(struct futex_q *q)

/* In the common case we don't take the spinlock, which is nice. */
retry:
- lock_ptr = q->lock_ptr;
- barrier();
+ /*
+ * q->lock_ptr can change between this read and the following spin_lock.
+ * Use READ_ONCE to forbid the compiler from reloading q->lock_ptr and
+ * optimizing lock_ptr out of the logic below.
+ */
+ lock_ptr = READ_ONCE(q->lock_ptr);
if (lock_ptr != NULL) {
spin_lock(lock_ptr);
/*
diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 716547fdb873..f894a2cd9b2a 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -123,8 +123,6 @@ static inline int debug_locks_off_graph_unlock(void)
return ret;
}

-static int lockdep_initialized;
-
unsigned long nr_list_entries;
static struct lock_list list_entries[MAX_LOCKDEP_ENTRIES];

@@ -434,19 +432,6 @@ unsigned int max_lockdep_depth;

#ifdef CONFIG_DEBUG_LOCKDEP
/*
- * We cannot printk in early bootup code. Not even early_printk()
- * might work. So we mark any initialization errors and printk
- * about it later on, in lockdep_info().
- */
-static int lockdep_init_error;
-static const char *lock_init_error;
-static unsigned long lockdep_init_trace_data[20];
-static struct stack_trace lockdep_init_trace = {
- .max_entries = ARRAY_SIZE(lockdep_init_trace_data),
- .entries = lockdep_init_trace_data,
-};
-
-/*
* Various lockdep statistics:
*/
DEFINE_PER_CPU(struct lockdep_stats, lockdep_stats);
@@ -669,20 +654,6 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
struct hlist_head *hash_head;
struct lock_class *class;

-#ifdef CONFIG_DEBUG_LOCKDEP
- /*
- * If the architecture calls into lockdep before initializing
- * the hashes then we'll warn about it later. (we cannot printk
- * right now)
- */
- if (unlikely(!lockdep_initialized)) {
- lockdep_init();
- lockdep_init_error = 1;
- lock_init_error = lock->name;
- save_stack_trace(&lockdep_init_trace);
- }
-#endif
-
if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) {
debug_locks_off();
printk(KERN_ERR
@@ -2011,6 +1982,53 @@ struct lock_class *lock_chain_get_class(struct lock_chain *chain, int i)
}

/*
+ * Returns the index of the first held_lock of the current chain
+ */
+static inline int get_first_held_lock(struct task_struct *curr,
+ struct held_lock *hlock)
+{
+ int i;
+ struct held_lock *hlock_curr;
+
+ for (i = curr->lockdep_depth - 1; i >= 0; i--) {
+ hlock_curr = curr->held_locks + i;
+ if (hlock_curr->irq_context != hlock->irq_context)
+ break;
+
+ }
+
+ return ++i;
+}
+
+/*
+ * Checks whether the chain and the current held locks are consistent
+ * in depth and also in content. If they are not it most likely means
+ * that there was a collision during the calculation of the chain_key.
+ * Returns: 0 not passed, 1 passed
+ */
+static int check_no_collision(struct task_struct *curr,
+ struct held_lock *hlock,
+ struct lock_chain *chain)
+{
+#ifdef CONFIG_DEBUG_LOCKDEP
+ int i, j, id;
+
+ i = get_first_held_lock(curr, hlock);
+
+ if (DEBUG_LOCKS_WARN_ON(chain->depth != curr->lockdep_depth - (i - 1)))
+ return 0;
+
+ for (j = 0; j < chain->depth - 1; j++, i++) {
+ id = curr->held_locks[i].class_idx - 1;
+
+ if (DEBUG_LOCKS_WARN_ON(chain_hlocks[chain->base + j] != id))
+ return 0;
+ }
+#endif
+ return 1;
+}
+
+/*
* Look up a dependency chain. If the key is not present yet then
* add it and return 1 - in this case the new dependency chain is
* validated. If the key is already hashed, return 0.
@@ -2023,7 +2041,6 @@ static inline int lookup_chain_cache(struct task_struct *curr,
struct lock_class *class = hlock_class(hlock);
struct hlist_head *hash_head = chainhashentry(chain_key);
struct lock_chain *chain;
- struct held_lock *hlock_curr;
int i, j;

/*
@@ -2041,6 +2058,9 @@ static inline int lookup_chain_cache(struct task_struct *curr,
if (chain->chain_key == chain_key) {
cache_hit:
debug_atomic_inc(chain_lookup_hits);
+ if (!check_no_collision(curr, hlock, chain))
+ return 0;
+
if (very_verbose(class))
printk("\nhash chain already cached, key: "
"%016Lx tail class: [%p] %s\n",
@@ -2078,13 +2098,7 @@ static inline int lookup_chain_cache(struct task_struct *curr,
chain = lock_chains + nr_lock_chains++;
chain->chain_key = chain_key;
chain->irq_context = hlock->irq_context;
- /* Find the first held_lock of current chain */
- for (i = curr->lockdep_depth - 1; i >= 0; i--) {
- hlock_curr = curr->held_locks + i;
- if (hlock_curr->irq_context != hlock->irq_context)
- break;
- }
- i++;
+ i = get_first_held_lock(curr, hlock);
chain->depth = curr->lockdep_depth + 1 - i;
if (likely(nr_chain_hlocks + chain->depth <= MAX_LOCKDEP_CHAIN_HLOCKS)) {
chain->base = nr_chain_hlocks;
@@ -2172,7 +2186,7 @@ static void check_chain_key(struct task_struct *curr)
{
#ifdef CONFIG_DEBUG_LOCKDEP
struct held_lock *hlock, *prev_hlock = NULL;
- unsigned int i, id;
+ unsigned int i;
u64 chain_key = 0;

for (i = 0; i < curr->lockdep_depth; i++) {
@@ -2189,17 +2203,16 @@ static void check_chain_key(struct task_struct *curr)
(unsigned long long)hlock->prev_chain_key);
return;
}
- id = hlock->class_idx - 1;
/*
* Whoops ran out of static storage again?
*/
- if (DEBUG_LOCKS_WARN_ON(id >= MAX_LOCKDEP_KEYS))
+ if (DEBUG_LOCKS_WARN_ON(hlock->class_idx > MAX_LOCKDEP_KEYS))
return;

if (prev_hlock && (prev_hlock->irq_context !=
hlock->irq_context))
chain_key = 0;
- chain_key = iterate_chain_key(chain_key, id);
+ chain_key = iterate_chain_key(chain_key, hlock->class_idx);
prev_hlock = hlock;
}
if (chain_key != curr->curr_chain_key) {
@@ -3077,7 +3090,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass,
struct task_struct *curr = current;
struct lock_class *class = NULL;
struct held_lock *hlock;
- unsigned int depth, id;
+ unsigned int depth;
int chain_head = 0;
int class_idx;
u64 chain_key;
@@ -3180,11 +3193,10 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass,
* The 'key ID' is what is the most compact key value to drive
* the hash, not class->key.
*/
- id = class - lock_classes;
/*
* Whoops, we did it again.. ran straight out of our static allocation.
*/
- if (DEBUG_LOCKS_WARN_ON(id >= MAX_LOCKDEP_KEYS))
+ if (DEBUG_LOCKS_WARN_ON(class_idx > MAX_LOCKDEP_KEYS))
return 0;

chain_key = curr->curr_chain_key;
@@ -3202,7 +3214,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass,
chain_key = 0;
chain_head = 1;
}
- chain_key = iterate_chain_key(chain_key, id);
+ chain_key = iterate_chain_key(chain_key, class_idx);

if (nest_lock && !__lock_is_held(nest_lock))
return print_lock_nested_lock_not_held(curr, hlock, ip);
@@ -4013,28 +4025,6 @@ void lockdep_reset_lock(struct lockdep_map *lock)
raw_local_irq_restore(flags);
}

-void lockdep_init(void)
-{
- int i;
-
- /*
- * Some architectures have their own start_kernel()
- * code which calls lockdep_init(), while we also
- * call lockdep_init() from the start_kernel() itself,
- * and we want to initialize the hashes only once:
- */
- if (lockdep_initialized)
- return;
-
- for (i = 0; i < CLASSHASH_SIZE; i++)
- INIT_HLIST_HEAD(classhash_table + i);
-
- for (i = 0; i < CHAINHASH_SIZE; i++)
- INIT_HLIST_HEAD(chainhash_table + i);
-
- lockdep_initialized = 1;
-}
-
void __init lockdep_info(void)
{
printk("Lock dependency validator: Copyright (c) 2006 Red Hat, Inc., Ingo Molnar\n");
@@ -4061,14 +4051,6 @@ void __init lockdep_info(void)

printk(" per task-struct memory footprint: %lu bytes\n",
sizeof(struct held_lock) * MAX_LOCK_DEPTH);
-
-#ifdef CONFIG_DEBUG_LOCKDEP
- if (lockdep_init_error) {
- printk("WARNING: lockdep init error: lock '%s' was acquired before lockdep_init().\n", lock_init_error);
- printk("Call stack leading to lockdep invocation was:\n");
- print_stack_trace(&lockdep_init_trace, 0);
- }
-#endif
}

static void
diff --git a/kernel/locking/mcs_spinlock.h b/kernel/locking/mcs_spinlock.h
index 5b9102a47ea5..c835270f0c2f 100644
--- a/kernel/locking/mcs_spinlock.h
+++ b/kernel/locking/mcs_spinlock.h
@@ -67,7 +67,13 @@ void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node)
node->locked = 0;
node->next = NULL;

- prev = xchg_acquire(lock, node);
+ /*
+ * We rely on the full barrier with global transitivity implied by the
+ * below xchg() to order the initialization stores above against any
+ * observation of @node. And to provide the ACQUIRE ordering associated
+ * with a LOCK primitive.
+ */
+ prev = xchg(lock, node);
if (likely(prev == NULL)) {
/*
* Lock acquired, don't need to set node->locked to 1. Threads
diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c
index 0551c219c40e..e364b424b019 100644
--- a/kernel/locking/mutex.c
+++ b/kernel/locking/mutex.c
@@ -716,6 +716,7 @@ static inline void
__mutex_unlock_common_slowpath(struct mutex *lock, int nested)
{
unsigned long flags;
+ WAKE_Q(wake_q);

/*
* As a performance measurement, release the lock before doing other
@@ -743,11 +744,11 @@ __mutex_unlock_common_slowpath(struct mutex *lock, int nested)
struct mutex_waiter, list);

debug_mutex_wake_waiter(lock, waiter);
-
- wake_up_process(waiter->task);
+ wake_q_add(&wake_q, waiter->task);
}

spin_unlock_mutex(&lock->wait_lock, flags);
+ wake_up_q(&wake_q);
}

/*
diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
index 393d1874b9e0..ce2f75e32ae1 100644
--- a/kernel/locking/qspinlock.c
+++ b/kernel/locking/qspinlock.c
@@ -358,8 +358,7 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
* sequentiality; this is because not all clear_pending_set_locked()
* implementations imply full barriers.
*/
- while ((val = smp_load_acquire(&lock->val.counter)) & _Q_LOCKED_MASK)
- cpu_relax();
+ smp_cond_acquire(!(atomic_read(&lock->val) & _Q_LOCKED_MASK));

/*
* take ownership and clear the pending bit.
@@ -435,7 +434,7 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
*
* The PV pv_wait_head_or_lock function, if active, will acquire
* the lock and return a non-zero value. So we have to skip the
- * smp_load_acquire() call. As the next PV queue head hasn't been
+ * smp_cond_acquire() call. As the next PV queue head hasn't been
* designated yet, there is no way for the locked value to become
* _Q_SLOW_VAL. So both the set_locked() and the
* atomic_cmpxchg_relaxed() calls will be safe.
@@ -466,7 +465,7 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
break;
}
/*
- * The smp_load_acquire() call above has provided the necessary
+ * The smp_cond_acquire() call above has provided the necessary
* acquire semantics required for locking. At most two
* iterations of this loop may be ran.
*/
diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h
index 87bb235c3448..21ede57f68b3 100644
--- a/kernel/locking/qspinlock_paravirt.h
+++ b/kernel/locking/qspinlock_paravirt.h
@@ -55,6 +55,11 @@ struct pv_node {
};

/*
+ * Include queued spinlock statistics code
+ */
+#include "qspinlock_stat.h"
+
+/*
* By replacing the regular queued_spin_trylock() with the function below,
* it will be called once when a lock waiter enter the PV slowpath before
* being queued. By allowing one lock stealing attempt here when the pending
@@ -65,9 +70,11 @@ struct pv_node {
static inline bool pv_queued_spin_steal_lock(struct qspinlock *lock)
{
struct __qspinlock *l = (void *)lock;
+ int ret = !(atomic_read(&lock->val) & _Q_LOCKED_PENDING_MASK) &&
+ (cmpxchg(&l->locked, 0, _Q_LOCKED_VAL) == 0);

- return !(atomic_read(&lock->val) & _Q_LOCKED_PENDING_MASK) &&
- (cmpxchg(&l->locked, 0, _Q_LOCKED_VAL) == 0);
+ qstat_inc(qstat_pv_lock_stealing, ret);
+ return ret;
}

/*
@@ -138,11 +145,6 @@ static __always_inline int trylock_clear_pending(struct qspinlock *lock)
#endif /* _Q_PENDING_BITS == 8 */

/*
- * Include queued spinlock statistics code
- */
-#include "qspinlock_stat.h"
-
-/*
* Lock and MCS node addresses hash table for fast lookup
*
* Hashing is done on a per-cacheline basis to minimize the need to access
@@ -398,6 +400,11 @@ pv_wait_head_or_lock(struct qspinlock *lock, struct mcs_spinlock *node)
if (READ_ONCE(pn->state) == vcpu_hashed)
lp = (struct qspinlock **)1;

+ /*
+ * Tracking # of slowpath locking operations
+ */
+ qstat_inc(qstat_pv_lock_slowpath, true);
+
for (;; waitcnt++) {
/*
* Set correct vCPU state to be used by queue node wait-early
diff --git a/kernel/locking/qspinlock_stat.h b/kernel/locking/qspinlock_stat.h
index 640dcecdd1df..eb2a2c9bc3fc 100644
--- a/kernel/locking/qspinlock_stat.h
+++ b/kernel/locking/qspinlock_stat.h
@@ -22,6 +22,7 @@
* pv_kick_wake - # of vCPU kicks used for computing pv_latency_wake
* pv_latency_kick - average latency (ns) of vCPU kick operation
* pv_latency_wake - average latency (ns) from vCPU kick to wakeup
+ * pv_lock_slowpath - # of locking operations via the slowpath
* pv_lock_stealing - # of lock stealing operations
* pv_spurious_wakeup - # of spurious wakeups
* pv_wait_again - # of vCPU wait's that happened after a vCPU kick
@@ -45,6 +46,7 @@ enum qlock_stats {
qstat_pv_kick_wake,
qstat_pv_latency_kick,
qstat_pv_latency_wake,
+ qstat_pv_lock_slowpath,
qstat_pv_lock_stealing,
qstat_pv_spurious_wakeup,
qstat_pv_wait_again,
@@ -70,6 +72,7 @@ static const char * const qstat_names[qstat_num + 1] = {
[qstat_pv_spurious_wakeup] = "pv_spurious_wakeup",
[qstat_pv_latency_kick] = "pv_latency_kick",
[qstat_pv_latency_wake] = "pv_latency_wake",
+ [qstat_pv_lock_slowpath] = "pv_lock_slowpath",
[qstat_pv_lock_stealing] = "pv_lock_stealing",
[qstat_pv_wait_again] = "pv_wait_again",
[qstat_pv_wait_early] = "pv_wait_early",
@@ -279,19 +282,6 @@ static inline void __pv_wait(u8 *ptr, u8 val)
#define pv_kick(c) __pv_kick(c)
#define pv_wait(p, v) __pv_wait(p, v)

-/*
- * PV unfair trylock count tracking function
- */
-static inline int qstat_spin_steal_lock(struct qspinlock *lock)
-{
- int ret = pv_queued_spin_steal_lock(lock);
-
- qstat_inc(qstat_pv_lock_stealing, ret);
- return ret;
-}
-#undef queued_spin_trylock
-#define queued_spin_trylock(l) qstat_spin_steal_lock(l)
-
#else /* CONFIG_QUEUED_LOCK_STAT */

static inline void qstat_inc(enum qlock_stats stat, bool cond) { }
diff --git a/kernel/smp.c b/kernel/smp.c
index d903c02223af..300d29391e07 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -105,13 +105,12 @@ void __init call_function_init(void)
* previous function call. For multi-cpu calls its even more interesting
* as we'll have to ensure no other cpu is observing our csd.
*/
-static void csd_lock_wait(struct call_single_data *csd)
+static __always_inline void csd_lock_wait(struct call_single_data *csd)
{
- while (smp_load_acquire(&csd->flags) & CSD_FLAG_LOCK)
- cpu_relax();
+ smp_cond_acquire(!(csd->flags & CSD_FLAG_LOCK));
}

-static void csd_lock(struct call_single_data *csd)
+static __always_inline void csd_lock(struct call_single_data *csd)
{
csd_lock_wait(csd);
csd->flags |= CSD_FLAG_LOCK;
@@ -124,7 +123,7 @@ static void csd_lock(struct call_single_data *csd)
smp_wmb();
}

-static void csd_unlock(struct call_single_data *csd)
+static __always_inline void csd_unlock(struct call_single_data *csd)
{
WARN_ON(!(csd->flags & CSD_FLAG_LOCK));

diff --git a/lib/test_static_keys.c b/lib/test_static_keys.c
index c61b299e367f..915d75df2086 100644
--- a/lib/test_static_keys.c
+++ b/lib/test_static_keys.c
@@ -46,8 +46,11 @@ struct test_key {
bool (*test_key)(void);
};

-#define test_key_func(key, branch) \
- ({bool func(void) { return branch(key); } func; })
+#define test_key_func(key, branch) \
+static bool key ## _ ## branch(void) \
+{ \
+ return branch(&key); \
+}

static void invert_key(struct static_key *key)
{
@@ -92,6 +95,25 @@ static int verify_keys(struct test_key *keys, int size, bool invert)
return 0;
}

+test_key_func(old_true_key, static_key_true)
+test_key_func(old_false_key, static_key_false)
+test_key_func(true_key, static_branch_likely)
+test_key_func(true_key, static_branch_unlikely)
+test_key_func(false_key, static_branch_likely)
+test_key_func(false_key, static_branch_unlikely)
+test_key_func(base_old_true_key, static_key_true)
+test_key_func(base_inv_old_true_key, static_key_true)
+test_key_func(base_old_false_key, static_key_false)
+test_key_func(base_inv_old_false_key, static_key_false)
+test_key_func(base_true_key, static_branch_likely)
+test_key_func(base_true_key, static_branch_unlikely)
+test_key_func(base_inv_true_key, static_branch_likely)
+test_key_func(base_inv_true_key, static_branch_unlikely)
+test_key_func(base_false_key, static_branch_likely)
+test_key_func(base_false_key, static_branch_unlikely)
+test_key_func(base_inv_false_key, static_branch_likely)
+test_key_func(base_inv_false_key, static_branch_unlikely)
+
static int __init test_static_key_init(void)
{
int ret;
@@ -102,95 +124,95 @@ static int __init test_static_key_init(void)
{
.init_state = true,
.key = &old_true_key,
- .test_key = test_key_func(&old_true_key, static_key_true),
+ .test_key = &old_true_key_static_key_true,
},
{
.init_state = false,
.key = &old_false_key,
- .test_key = test_key_func(&old_false_key, static_key_false),
+ .test_key = &old_false_key_static_key_false,
},
/* internal keys - new keys */
{
.init_state = true,
.key = &true_key.key,
- .test_key = test_key_func(&true_key, static_branch_likely),
+ .test_key = &true_key_static_branch_likely,
},
{
.init_state = true,
.key = &true_key.key,
- .test_key = test_key_func(&true_key, static_branch_unlikely),
+ .test_key = &true_key_static_branch_unlikely,
},
{
.init_state = false,
.key = &false_key.key,
- .test_key = test_key_func(&false_key, static_branch_likely),
+ .test_key = &false_key_static_branch_likely,
},
{
.init_state = false,
.key = &false_key.key,
- .test_key = test_key_func(&false_key, static_branch_unlikely),
+ .test_key = &false_key_static_branch_unlikely,
},
/* external keys - old keys */
{
.init_state = true,
.key = &base_old_true_key,
- .test_key = test_key_func(&base_old_true_key, static_key_true),
+ .test_key = &base_old_true_key_static_key_true,
},
{
.init_state = false,
.key = &base_inv_old_true_key,
- .test_key = test_key_func(&base_inv_old_true_key, static_key_true),
+ .test_key = &base_inv_old_true_key_static_key_true,
},
{
.init_state = false,
.key = &base_old_false_key,
- .test_key = test_key_func(&base_old_false_key, static_key_false),
+ .test_key = &base_old_false_key_static_key_false,
},
{
.init_state = true,
.key = &base_inv_old_false_key,
- .test_key = test_key_func(&base_inv_old_false_key, static_key_false),
+ .test_key = &base_inv_old_false_key_static_key_false,
},
/* external keys - new keys */
{
.init_state = true,
.key = &base_true_key.key,
- .test_key = test_key_func(&base_true_key, static_branch_likely),
+ .test_key = &base_true_key_static_branch_likely,
},
{
.init_state = true,
.key = &base_true_key.key,
- .test_key = test_key_func(&base_true_key, static_branch_unlikely),
+ .test_key = &base_true_key_static_branch_unlikely,
},
{
.init_state = false,
.key = &base_inv_true_key.key,
- .test_key = test_key_func(&base_inv_true_key, static_branch_likely),
+ .test_key = &base_inv_true_key_static_branch_likely,
},
{
.init_state = false,
.key = &base_inv_true_key.key,
- .test_key = test_key_func(&base_inv_true_key, static_branch_unlikely),
+ .test_key = &base_inv_true_key_static_branch_unlikely,
},
{
.init_state = false,
.key = &base_false_key.key,
- .test_key = test_key_func(&base_false_key, static_branch_likely),
+ .test_key = &base_false_key_static_branch_likely,
},
{
.init_state = false,
.key = &base_false_key.key,
- .test_key = test_key_func(&base_false_key, static_branch_unlikely),
+ .test_key = &base_false_key_static_branch_unlikely,
},
{
.init_state = true,
.key = &base_inv_false_key.key,
- .test_key = test_key_func(&base_inv_false_key, static_branch_likely),
+ .test_key = &base_inv_false_key_static_branch_likely,
},
{
.init_state = true,
.key = &base_inv_false_key.key,
- .test_key = test_key_func(&base_inv_false_key, static_branch_unlikely),
+ .test_key = &base_inv_false_key_static_branch_unlikely,
},
};

diff --git a/tools/lib/lockdep/Makefile b/tools/lib/lockdep/Makefile
index 90d2baeb621a..1d57af56814b 100644
--- a/tools/lib/lockdep/Makefile
+++ b/tools/lib/lockdep/Makefile
@@ -100,7 +100,7 @@ include $(srctree)/tools/build/Makefile.include

do_compile_shared_library = \
($(print_shared_lib_compile) \
- $(CC) --shared $^ -o $@ -lpthread -ldl -Wl,-soname='"$@"';$(shell ln -s $@ liblockdep.so))
+ $(CC) --shared $^ -o $@ -lpthread -ldl -Wl,-soname='"$@"';$(shell ln -sf $@ liblockdep.so))

do_build_static_lib = \
($(print_static_lib_build) \
diff --git a/tools/lib/lockdep/common.c b/tools/lib/lockdep/common.c
index 9be663340f0a..d1c89cc06f5f 100644
--- a/tools/lib/lockdep/common.c
+++ b/tools/lib/lockdep/common.c
@@ -11,11 +11,6 @@ static __thread struct task_struct current_obj;
bool debug_locks = true;
bool debug_locks_silent;

-__attribute__((constructor)) static void liblockdep_init(void)
-{
- lockdep_init();
-}
-
__attribute__((destructor)) static void liblockdep_exit(void)
{
debug_check_no_locks_held();
diff --git a/tools/lib/lockdep/include/liblockdep/common.h b/tools/lib/lockdep/include/liblockdep/common.h
index a60c14b9662a..6e66277ec437 100644
--- a/tools/lib/lockdep/include/liblockdep/common.h
+++ b/tools/lib/lockdep/include/liblockdep/common.h
@@ -44,7 +44,6 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass,
void lock_release(struct lockdep_map *lock, int nested,
unsigned long ip);
extern void debug_check_no_locks_freed(const void *from, unsigned long len);
-extern void lockdep_init(void);

#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \
{ .name = (_name), .key = (void *)(_key), }
diff --git a/tools/lib/lockdep/lockdep.c b/tools/lib/lockdep/lockdep.c
index f42b7e9aa48f..a0a2e3a266af 100644
--- a/tools/lib/lockdep/lockdep.c
+++ b/tools/lib/lockdep/lockdep.c
@@ -1,2 +1,8 @@
#include <linux/lockdep.h>
+
+/* Trivial API wrappers, we don't (yet) have RCU in user-space: */
+#define hlist_for_each_entry_rcu hlist_for_each_entry
+#define hlist_add_head_rcu hlist_add_head
+#define hlist_del_rcu hlist_del
+
#include "../../../kernel/locking/lockdep.c"
diff --git a/tools/lib/lockdep/preload.c b/tools/lib/lockdep/preload.c
index 21cdf869a01b..52844847569c 100644
--- a/tools/lib/lockdep/preload.c
+++ b/tools/lib/lockdep/preload.c
@@ -439,7 +439,5 @@ __attribute__((constructor)) static void init_preload(void)
ll_pthread_rwlock_unlock = dlsym(RTLD_NEXT, "pthread_rwlock_unlock");
#endif

- lockdep_init();
-
__init_state = done;
}
diff --git a/tools/lib/lockdep/tests/AA.c b/tools/lib/lockdep/tests/AA.c
index 0f782ff404ac..18211a5f354f 100644
--- a/tools/lib/lockdep/tests/AA.c
+++ b/tools/lib/lockdep/tests/AA.c
@@ -1,13 +1,13 @@
#include <liblockdep/mutex.h>

-void main(void)
+int main(void)
{
- pthread_mutex_t a, b;
+ pthread_mutex_t a;

pthread_mutex_init(&a, NULL);
- pthread_mutex_init(&b, NULL);

pthread_mutex_lock(&a);
- pthread_mutex_lock(&b);
pthread_mutex_lock(&a);
+
+ return 0;
}
diff --git a/tools/lib/lockdep/tests/ABA.c b/tools/lib/lockdep/tests/ABA.c
new file mode 100644
index 000000000000..0f782ff404ac
--- /dev/null
+++ b/tools/lib/lockdep/tests/ABA.c
@@ -0,0 +1,13 @@
+#include <liblockdep/mutex.h>
+
+void main(void)
+{
+ pthread_mutex_t a, b;
+
+ pthread_mutex_init(&a, NULL);
+ pthread_mutex_init(&b, NULL);
+
+ pthread_mutex_lock(&a);
+ pthread_mutex_lock(&b);
+ pthread_mutex_lock(&a);
+}
diff --git a/tools/lib/lockdep/tests/ABBA_2threads.c b/tools/lib/lockdep/tests/ABBA_2threads.c
new file mode 100644
index 000000000000..cd807d736361
--- /dev/null
+++ b/tools/lib/lockdep/tests/ABBA_2threads.c
@@ -0,0 +1,46 @@
+#include <stdio.h>
+#include <pthread.h>
+
+pthread_mutex_t a = PTHREAD_MUTEX_INITIALIZER;
+pthread_mutex_t b = PTHREAD_MUTEX_INITIALIZER;
+pthread_barrier_t bar;
+
+void *ba_lock(void *arg)
+{
+ int ret, i;
+
+ pthread_mutex_lock(&b);
+
+ if (pthread_barrier_wait(&bar) == PTHREAD_BARRIER_SERIAL_THREAD)
+ pthread_barrier_destroy(&bar);
+
+ pthread_mutex_lock(&a);
+
+ pthread_mutex_unlock(&a);
+ pthread_mutex_unlock(&b);
+}
+
+int main(void)
+{
+ pthread_t t;
+
+ pthread_barrier_init(&bar, NULL, 2);
+
+ if (pthread_create(&t, NULL, ba_lock, NULL)) {
+ fprintf(stderr, "pthread_create() failed\n");
+ return 1;
+ }
+ pthread_mutex_lock(&a);
+
+ if (pthread_barrier_wait(&bar) == PTHREAD_BARRIER_SERIAL_THREAD)
+ pthread_barrier_destroy(&bar);
+
+ pthread_mutex_lock(&b);
+
+ pthread_mutex_unlock(&b);
+ pthread_mutex_unlock(&a);
+
+ pthread_join(t, NULL);
+
+ return 0;
+}
diff --git a/tools/lib/lockdep/uinclude/linux/compiler.h b/tools/lib/lockdep/uinclude/linux/compiler.h
index 6386dc3182a0..fd3e56a83fc2 100644
--- a/tools/lib/lockdep/uinclude/linux/compiler.h
+++ b/tools/lib/lockdep/uinclude/linux/compiler.h
@@ -3,6 +3,7 @@

#define __used __attribute__((__unused__))
#define unlikely
+#define READ_ONCE(x) (x)
#define WRITE_ONCE(x, val) x=(val)
#define RCU_INIT_POINTER(p, v) p=(v)