Merge branch 'v2.6.24-lockdep' of git://git.kernel.org/pub/scm/linux/…

…kernel/git/peterz/linux-2.6-lockdep

* 'v2.6.24-lockdep' of git://git.kernel.org/pub/scm/linux/kernel/git/peterz/linux-2.6-lockdep:
  lockdep: annotate dir vs file i_mutex
  lockdep: per filesystem inode lock class
  lockdep: annotate kprobes irq fiddling
  lockdep: annotate rcu_read_{,un}lock{,_bh}
  lockdep: annotate journal_start()
  lockdep: s390: connect the sysexit hook
  lockdep: x86_64: connect the sysexit hook
  lockdep: i386: connect the sysexit hook
  lockdep: syscall exit check
  lockdep: fixup mutex annotations
  lockdep: fix mismatched lockdep_depth/curr_chain_hash
  lockdep: Avoid /proc/lockdep & lock_stat infinite output
  lockdep: maintainers

MAINTAINERS

@@ -2404,6 +2404,15 @@ M:	khali@linux-fr.org
 L:	lm-sensors@lm-sensors.org
 S:	Maintained
 
+LOCKDEP AND LOCKSTAT
+P:	Peter Zijlstra
+M:	peterz@infradead.org
+P:	Ingo Molnar
+M:	mingo@redhat.com
+L:	linux-kernel@vger.kernel.org
+T:	git://git.kernel.org/pub/scm/linux/kernel/git/peterz/linux-2.6-lockdep.git
+S:	Maintained
+
 LOGICAL DISK MANAGER SUPPORT (LDM, Windows 2000/XP/Vista Dynamic Disks)
 P:	Richard Russon (FlatCap)
 M:	ldm@flatcap.org

arch/s390/kernel/entry.S

@@ -68,9 +68,15 @@ STACK_SIZE  = 1 << STACK_SHIFT
 	l	%r1,BASED(.Ltrace_irq_off)
 	basr	%r14,%r1
 	.endm
+
+	.macro	LOCKDEP_SYS_EXIT
+	l	%r1,BASED(.Llockdep_sys_exit)
+	basr	%r14,%r1
+	.endm
 #else
 #define TRACE_IRQS_ON
 #define TRACE_IRQS_OFF
+#define LOCKDEP_SYS_EXIT
 #endif
 
 /*
@@ -260,6 +266,7 @@ sysc_return:
 	bno	BASED(sysc_leave)
 	tm	__TI_flags+3(%r9),_TIF_WORK_SVC
 	bnz	BASED(sysc_work)  # there is work to do (signals etc.)
+	LOCKDEP_SYS_EXIT
 sysc_leave:
 	RESTORE_ALL __LC_RETURN_PSW,1
 
@@ -283,6 +290,7 @@ sysc_work:
 	bo	BASED(sysc_restart)
 	tm	__TI_flags+3(%r9),_TIF_SINGLE_STEP
 	bo	BASED(sysc_singlestep)
+	LOCKDEP_SYS_EXIT
 	b	BASED(sysc_leave)
 
 #
@@ -572,6 +580,7 @@ io_return:
 #endif
 	tm	__TI_flags+3(%r9),_TIF_WORK_INT
 	bnz	BASED(io_work)		# there is work to do (signals etc.)
+	LOCKDEP_SYS_EXIT
 io_leave:
 	RESTORE_ALL __LC_RETURN_PSW,0
 io_done:
@@ -618,6 +627,7 @@ io_work_loop:
 	bo	BASED(io_reschedule)
 	tm	__TI_flags+3(%r9),(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)
 	bnz	BASED(io_sigpending)
+	LOCKDEP_SYS_EXIT
 	b	BASED(io_leave)
 
 #
@@ -1040,6 +1050,8 @@ cleanup_io_leave_insn:
 .Ltrace_irq_on: .long	trace_hardirqs_on
 .Ltrace_irq_off:
 		.long	trace_hardirqs_off
+.Llockdep_sys_exit:
+		.long	lockdep_sys_exit
 #endif
 .Lcritical_start:
 		.long	__critical_start + 0x80000000

arch/s390/kernel/entry64.S

@@ -66,9 +66,14 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK | _TIF_NEED_RESCHED | \
 	.macro	TRACE_IRQS_OFF
 	 brasl	%r14,trace_hardirqs_off
 	.endm
+
+	.macro	LOCKDEP_SYS_EXIT
+	 brasl	%r14,lockdep_sys_exit
+	.endm
 #else
 #define TRACE_IRQS_ON
 #define TRACE_IRQS_OFF
+#define LOCKDEP_SYS_EXIT
 #endif
 
 	.macro	STORE_TIMER lc_offset
@@ -255,6 +260,7 @@ sysc_return:
 	jno	sysc_leave
 	tm	__TI_flags+7(%r9),_TIF_WORK_SVC
 	jnz	sysc_work	# there is work to do (signals etc.)
+	LOCKDEP_SYS_EXIT
 sysc_leave:
 	RESTORE_ALL __LC_RETURN_PSW,1
 
@@ -278,6 +284,7 @@ sysc_work:
 	jo	sysc_restart
 	tm	__TI_flags+7(%r9),_TIF_SINGLE_STEP
 	jo	sysc_singlestep
+	LOCKDEP_SYS_EXIT
 	j	sysc_leave
 
 #
@@ -558,6 +565,7 @@ io_return:
 #endif
 	tm	__TI_flags+7(%r9),_TIF_WORK_INT
 	jnz	io_work 		# there is work to do (signals etc.)
+	LOCKDEP_SYS_EXIT
 io_leave:
 	RESTORE_ALL __LC_RETURN_PSW,0
 io_done:
@@ -605,6 +613,7 @@ io_work_loop:
 	jo	io_reschedule
 	tm	__TI_flags+7(%r9),(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)
 	jnz	io_sigpending
+	LOCKDEP_SYS_EXIT
 	j	io_leave
 
 #

arch/x86/kernel/entry_32.S

@@ -251,6 +251,7 @@ check_userspace:
 	jb resume_kernel		# not returning to v8086 or userspace
 
 ENTRY(resume_userspace)
+	LOCKDEP_SYS_EXIT
  	DISABLE_INTERRUPTS(CLBR_ANY)	# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
@@ -338,6 +339,7 @@ sysenter_past_esp:
 	jae syscall_badsys
 	call *sys_call_table(,%eax,4)
 	movl %eax,PT_EAX(%esp)
+	LOCKDEP_SYS_EXIT
 	DISABLE_INTERRUPTS(CLBR_ANY)
 	TRACE_IRQS_OFF
 	movl TI_flags(%ebp), %ecx
@@ -377,6 +379,7 @@ syscall_call:
 	call *sys_call_table(,%eax,4)
 	movl %eax,PT_EAX(%esp)		# store the return value
 syscall_exit:
+	LOCKDEP_SYS_EXIT
 	DISABLE_INTERRUPTS(CLBR_ANY)	# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
@@ -467,6 +470,7 @@ work_pending:
 	jz work_notifysig
 work_resched:
 	call schedule
+	LOCKDEP_SYS_EXIT
 	DISABLE_INTERRUPTS(CLBR_ANY)	# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret

arch/x86/kernel/entry_64.S

@@ -244,6 +244,7 @@ ret_from_sys_call:
 	movl $_TIF_ALLWORK_MASK,%edi
 	/* edi:	flagmask */
 sysret_check:		
+	LOCKDEP_SYS_EXIT
 	GET_THREAD_INFO(%rcx)
 	cli
 	TRACE_IRQS_OFF
@@ -333,6 +334,7 @@ int_ret_from_sys_call:
 	movl $_TIF_ALLWORK_MASK,%edi
 	/* edi:	mask to check */
 int_with_check:
+	LOCKDEP_SYS_EXIT_IRQ
 	GET_THREAD_INFO(%rcx)
 	movl threadinfo_flags(%rcx),%edx
 	andl %edi,%edx
@@ -544,11 +546,13 @@ exit_intr:
 retint_with_reschedule:
 	movl $_TIF_WORK_MASK,%edi
 retint_check:
+	LOCKDEP_SYS_EXIT_IRQ
 	movl threadinfo_flags(%rcx),%edx
 	andl %edi,%edx
 	CFI_REMEMBER_STATE
 	jnz  retint_careful
-retint_swapgs:	 	
+
+retint_swapgs:		/* return to user-space */
 	/*
 	 * The iretq could re-enable interrupts:
 	 */
@@ -557,7 +561,7 @@ retint_swapgs:
 	swapgs 
 	jmp restore_args
 
-retint_restore_args:				
+retint_restore_args:	/* return to kernel space */
 	cli
 	/*
 	 * The iretq could re-enable interrupts:
@@ -866,26 +870,21 @@ error_sti:
 	movq ORIG_RAX(%rsp),%rsi	/* get error code */ 
 	movq $-1,ORIG_RAX(%rsp)
 	call *%rax
-	/* ebx:	no swapgs flag (1: don't need swapgs, 0: need it) */	 
-error_exit:		
-	movl %ebx,%eax		
+	/* ebx:	no swapgs flag (1: don't need swapgs, 0: need it) */
+error_exit:
+	movl %ebx,%eax
 	RESTORE_REST
 	cli
 	TRACE_IRQS_OFF
 	GET_THREAD_INFO(%rcx)	
 	testl %eax,%eax
 	jne  retint_kernel
+	LOCKDEP_SYS_EXIT_IRQ
 	movl  threadinfo_flags(%rcx),%edx
 	movl  $_TIF_WORK_MASK,%edi
 	andl  %edi,%edx
 	jnz  retint_careful
-	/*
-	 * The iret might restore flags:
-	 */
-	TRACE_IRQS_IRETQ
-	swapgs 
-	RESTORE_ARGS 0,8,0						
-	jmp iret_label
+	jmp retint_swapgs
 	CFI_ENDPROC
 
 error_kernelspace:

arch/x86/kernel/kprobes_32.c

@@ -557,6 +557,12 @@ static int __kprobes post_kprobe_handler(struct pt_regs *regs)
 
 	resume_execution(cur, regs, kcb);
 	regs->eflags |= kcb->kprobe_saved_eflags;
+#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
+	if (raw_irqs_disabled_flags(regs->eflags))
+		trace_hardirqs_off();
+	else
+		trace_hardirqs_on();
+#endif
 
 	/*Restore back the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
@@ -694,6 +700,7 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
 			MIN_STACK_SIZE(addr));
 	regs->eflags &= ~IF_MASK;
+	trace_hardirqs_off();
 	regs->eip = (unsigned long)(jp->entry);
 	return 1;
 }

arch/x86/kernel/kprobes_64.c

@@ -544,6 +544,12 @@ int __kprobes post_kprobe_handler(struct pt_regs *regs)
 
 	resume_execution(cur, regs, kcb);
 	regs->eflags |= kcb->kprobe_saved_rflags;
+#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
+	if (raw_irqs_disabled_flags(regs->eflags))
+		trace_hardirqs_off();
+	else
+		trace_hardirqs_on();
+#endif
 
 	/* Restore the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
@@ -684,6 +690,7 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr,
 			MIN_STACK_SIZE(addr));
 	regs->eflags &= ~IF_MASK;
+	trace_hardirqs_off();
 	regs->rip = (unsigned long)(jp->entry);
 	return 1;
 }

arch/x86/lib/thunk_64.S

@@ -50,6 +50,10 @@
 	thunk trace_hardirqs_on_thunk,trace_hardirqs_on
 	thunk trace_hardirqs_off_thunk,trace_hardirqs_off
 #endif
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	thunk lockdep_sys_exit_thunk,lockdep_sys_exit
+#endif
 
 	/* SAVE_ARGS below is used only for the .cfi directives it contains. */
 	CFI_STARTPROC

fs/inode.c

@@ -142,6 +142,15 @@ static struct inode *alloc_inode(struct super_block *sb)
 			return NULL;
 		}
 
+		spin_lock_init(&inode->i_lock);
+		lockdep_set_class(&inode->i_lock, &sb->s_type->i_lock_key);
+
+		mutex_init(&inode->i_mutex);
+		lockdep_set_class(&inode->i_mutex, &sb->s_type->i_mutex_key);
+
+		init_rwsem(&inode->i_alloc_sem);
+		lockdep_set_class(&inode->i_alloc_sem, &sb->s_type->i_alloc_sem_key);
+
 		mapping->a_ops = &empty_aops;
  		mapping->host = inode;
 		mapping->flags = 0;
@@ -190,16 +199,13 @@ void inode_init_once(struct inode *inode)
 	INIT_HLIST_NODE(&inode->i_hash);
 	INIT_LIST_HEAD(&inode->i_dentry);
 	INIT_LIST_HEAD(&inode->i_devices);
-	mutex_init(&inode->i_mutex);
-	init_rwsem(&inode->i_alloc_sem);
 	INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
 	rwlock_init(&inode->i_data.tree_lock);
 	spin_lock_init(&inode->i_data.i_mmap_lock);
 	INIT_LIST_HEAD(&inode->i_data.private_list);
 	spin_lock_init(&inode->i_data.private_lock);
 	INIT_RAW_PRIO_TREE_ROOT(&inode->i_data.i_mmap);
 	INIT_LIST_HEAD(&inode->i_data.i_mmap_nonlinear);
-	spin_lock_init(&inode->i_lock);
 	i_size_ordered_init(inode);
 #ifdef CONFIG_INOTIFY
 	INIT_LIST_HEAD(&inode->inotify_watches);
@@ -561,6 +567,18 @@ EXPORT_SYMBOL(new_inode);
 
 void unlock_new_inode(struct inode *inode)
 {
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct file_system_type *type = inode->i_sb->s_type;
+	/*
+	 * ensure nobody is actually holding i_mutex
+	 */
+	mutex_destroy(&inode->i_mutex);
+	mutex_init(&inode->i_mutex);
+	if (inode->i_mode & S_IFDIR)
+		lockdep_set_class(&inode->i_mutex, &type->i_mutex_dir_key);
+	else
+		lockdep_set_class(&inode->i_mutex, &type->i_mutex_key);
+#endif
 	/*
 	 * This is special!  We do not need the spinlock
 	 * when clearing I_LOCK, because we're guaranteed

fs/jbd/transaction.c

@@ -233,6 +233,8 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
 	return ret;
 }
 
+static struct lock_class_key jbd_handle_key;
+
 /* Allocate a new handle.  This should probably be in a slab... */
 static handle_t *new_handle(int nblocks)
 {
@@ -243,6 +245,8 @@ static handle_t *new_handle(int nblocks)
 	handle->h_buffer_credits = nblocks;
 	handle->h_ref = 1;
 
+	lockdep_init_map(&handle->h_lockdep_map, "jbd_handle", &jbd_handle_key, 0);
+
 	return handle;
 }
 
@@ -286,6 +290,9 @@ handle_t *journal_start(journal_t *journal, int nblocks)
 		current->journal_info = NULL;
 		handle = ERR_PTR(err);
 	}
+
+	lock_acquire(&handle->h_lockdep_map, 0, 0, 0, 2, _THIS_IP_);
+
 	return handle;
 }
 
@@ -1411,6 +1418,8 @@ int journal_stop(handle_t *handle)
 		spin_unlock(&journal->j_state_lock);
 	}
 
+	lock_release(&handle->h_lockdep_map, 1, _THIS_IP_);
+
 	jbd_free_handle(handle);
 	return err;
 }

include/asm-x86/irqflags_32.h

@@ -160,4 +160,17 @@ static inline int raw_irqs_disabled(void)
 # define TRACE_IRQS_OFF
 #endif
 
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+# define LOCKDEP_SYS_EXIT			\
+	pushl %eax;				\
+	pushl %ecx;				\
+	pushl %edx;				\
+	call lockdep_sys_exit;			\
+	popl %edx;				\
+	popl %ecx;				\
+	popl %eax;
+#else
+# define LOCKDEP_SYS_EXIT
+#endif
+
 #endif