linux-toradex.git/kernel/lockdep.c, branch v2.6.22.11

lockdep: fix mismatched lockdep_depth/curr_chain_hash

2007-11-02T15:44:16+00:00

patch 3aa416b07f0adf01c090baab26fb70c35ec17623 in mainline.

lockdep: fix mismatched lockdep_depth/curr_chain_hash

It is possible for the current->curr_chain_key to become inconsistent with the
current index if the chain fails to validate.  The end result is that future
lock_acquire() operations may inadvertently fail to find a hit in the cache
resulting in a new node being added to the graph for every acquire.

[ peterz: this might explain some of the lockdep is so _slow_ complaints. ]
[ mingo: this does not impact the correctness of validation, but may slow
  down future operations significantly, if the chain gets very long. ]

Signed-off-by: Gregory Haskins 
Signed-off-by: Peter Zijlstra 
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

lockdep: removed unused ip argument in mark_lock & mark_held_locks

2007-05-08T18:15:13+00:00

It looks like a remainder from designing...

Signed-off-by: Jarek Poplawski 
Cc: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

lockdep: lookup_chain_cache comment errata

2007-05-08T18:15:11+00:00

Signed-off-by: Jarek Poplawski 
Acked-by: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Simplify kallsyms_lookup()

2007-05-08T18:15:08+00:00

Several kallsyms_lookup() pass dummy arguments but only need, say, module's
name.  Make kallsyms_lookup() accept NULLs where possible.

Also, makes picture clearer about what interfaces are needed for all symbol
resolving business.

Signed-off-by: Alexey Dobriyan 
Cc: Rusty Russell 
Acked-by: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

simplify the stacktrace code

2007-05-08T18:14:58+00:00

Simplify the stacktrace code:

 - remove the unused task argument to save_stack_trace, it's always
   current
 - remove the all_contexts flag, it's alwasy 0

Signed-off-by: Christoph Hellwig 
Cc: Paul Mundt 
Cc: Ralf Baechle 
Cc: Martin Schwidefsky 
Cc: "David S. Miller" 
Cc: Andi Kleen 
Cc: Akinobu Mita 
Acked-by: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] lockdep: debug_show_all_locks & debug_show_held_locks vs. debug_locks

2007-03-23T02:39:06+00:00

lockdep's data shouldn't be used when debug_locks == 0 because it's not
updated after this, so it's more misleading than helpful.

PS: probably lockdep's current-> fields should be reset after it turns
debug_locks off: so, after printing a bug report, but before return from
exported functions, but there are really a lot of these possibilities (e.g.
 after DEBUG_LOCKS_WARN_ON), so, something could be missed.  (Of course
direct use of this fields isn't recommended either.)

Reported-by: Folkert van Heusden 
Inspired-by: Oleg Nesterov 
Signed-off-by: Jarek Poplawski 
Acked-by: Peter Zijlstra 
Cc: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] fix section mismatch warning in lockdep

2007-03-01T22:53:37+00:00

lockdep_init() is marked __init but used in several places
outside __init code. This causes following warnings:
$ scripts/mod/modpost kernel/lockdep.o
WARNING: kernel/built-in.o - Section mismatch: reference to .init.text:lockdep_init from .text.lockdep_init_map after 'lockdep_init_map' (at offset 0x105)
WARNING: kernel/built-in.o - Section mismatch: reference to .init.text:lockdep_init from .text.lockdep_reset_lock after 'lockdep_reset_lock' (at offset 0x35)
WARNING: kernel/built-in.o - Section mismatch: reference to .init.text:lockdep_init from .text.__lock_acquire after '__lock_acquire' (at offset 0xb2)

The warnings are less obviously due to heavy inlining by gcc - this is not
altered.

Fix the section mismatch warnings by removing the __init marking, which
seems obviously wrong.

Signed-off-by: Sam Ravnborg 
Acked-by: Ingo Molnar 
Cc: 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] lockdep: debug_locks check after check_chain_key

2007-02-21T01:10:14+00:00

In __lock_acquire check_chain_key can turn off debug_locks, so check is
needed to assure proper return code.

Signed-off-by: Jarek Poplawski 
Cc: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] lockdep: add graph depth information to /proc/lockdep

2007-02-11T18:51:26+00:00

Generate locking graph information into /proc/lockdep, for lock hierarchy
documentation and visualization purposes.

sample output:

 c089fd5c OPS:     138 FD:   14 BD:    1 --..: &tty->termios_mutex
  -> [c07a3430] tty_ldisc_lock
  -> [c07a37f0] &port_lock_key
  -> [c07afdc0] &rq->rq_lock_key#2

The lock classes listed are all the first-hop lock dependencies that
lockdep has seen so far.

Signed-off-by: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] lockdep: more unlock-on-error fixes

2007-02-11T18:51:26+00:00

- returns after DEBUG_LOCKS_WARN_ON added in 3 places

- debug_locks checking after lookup_chain_cache() added in
  __lock_acquire()

- locking for testing and changing global variable max_lockdep_depth
  added in __lock_acquire()

From: Ingo Molnar 

My __acquire_lock() cleanup introduced a locking bug: on SMP systems we'd
release a non-owned graph lock.  Fix this by moving the graph unlock back,
and by leaving the max_lockdep_depth variable update possibly racy.  (we
dont care, it's just statistics)

Also add some minimal debugging code to graph_unlock()/graph_lock(),
which caught this locking bug.

Signed-off-by: Jarek Poplawski 
Signed-off-by: Ingo Molnar 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds