The ftrace_preempt_disable/enable functions were to address a
recursive race caused by the function tracer. The function tracer
traces all functions which makes it easily susceptible to recursion.
One area was preempt_enable(). This would call the scheduler and
the schedulre would call the function tracer and loop.
(So was it thought).

The ftrace_preempt_disable/enable was made to protect against recursion
inside the scheduler by storing the NEED_RESCHED flag. If it was
set before the ftrace_preempt_disable() it would not call schedule
on ftrace_preempt_enable(), thinking that if it was set before then
it would have already scheduled unless it was already in the scheduler.

This worked fine except in the case of SMP, where another task would set
the NEED_RESCHED flag for a task on another CPU, and then kick off an
IPI to trigger it. This could cause the NEED_RESCHED to be saved at
ftrace_preempt_disable() but the IPI to arrive in the the preempt
disabled section. The ftrace_preempt_enable() would not call the scheduler
because the flag was already set before entring the section.

This bug would cause a missed preemption check and cause lower latencies.

Investigating further, I found that the recusion caused by the function
tracer was not due to schedule(), but due to preempt_schedule(). Now
that preempt_schedule is completely annotated with notrace, the recusion
no longer is an issue.
Reported-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

When we cat <debugfs>/tracing/stack_trace, we may cause circular lock:
sys_read()
  t_start()
     arch_spin_lock(&max_stack_lock);

  t_show()
     seq_printf(), vsnprintf() .... /* they are all trace-able,
       when they are traced, max_stack_lock may be required again. */

The following script can trigger this circular dead lock very easy:
#!/bin/bash

echo 1 > /proc/sys/kernel/stack_tracer_enabled

mount -t debugfs xxx /mnt > /dev/null 2>&1

(
# make check_stack() zealous to require max_stack_lock
for ((; ;))
{
	echo 1 > /mnt/tracing/stack_max_size
}
) &

for ((; ;))
{
	cat /mnt/tracing/stack_trace > /dev/null
}

To fix this bug, we increase the percpu trace_active before
require the lock.
Reported-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NLai Jiangshan <laijs@cn.fujitsu.com>
LKML-Reference: <4B67D4F9.9080905@cn.fujitsu.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

Name space cleanup. No functional change.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: linux-arch@vger.kernel.org

Further name space cleanup. No functional change
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: linux-arch@vger.kernel.org

The raw_spin* namespace was taken by lockdep for the architecture
specific implementations. raw_spin_* would be the ideal name space for
the spinlocks which are not converted to sleeping locks in preempt-rt.

Linus suggested to convert the raw_ to arch_ locks and cleanup the
name space instead of using an artifical name like core_spin,
atomic_spin or whatever

No functional change.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NPeter Zijlstra <peterz@infradead.org>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: linux-arch@vger.kernel.org

It's unused.

It isn't needed -- read or write flag is already passed and sysctl
shouldn't care about the rest.

It _was_ used in two places at arch/frv for some reason.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Extract duplicate code in t_start() and t_next().
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A891A91.4030602@cn.fujitsu.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Every time we cat stack_trace, we leak memory allocated by seq_open().
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <4A67D8E8.3020500@cn.fujitsu.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

We can directly use %pF input format instead of sprint_symbol()
and %s input format.
Signed-off-by: NXiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Reviewed-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>

This made my machine completely frozen:

  # echo 1 > /proc/sys/kernel/stack_tracer_enabled
  # echo 2 > /proc/sys/kernel/stack_tracer_enabled

The cause is register_ftrace_function() was called twice.

Also fix ftrace_enabled sysctl, though seems nothing bad happened
as I tested it.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A448D17.9010305@cn.fujitsu.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The last entry in the stack_dump_trace is ULONG_MAX, which is not
a valid entry, but max_stack_trace.nr_entries has accounted for it.
So when printing the header, we should decrease it by one.
Before fix, print as following, for example:

	Depth    Size   Location    (53 entries)	<--- should be 52
	-----    ----   --------
  0)     3264     108   update_wall_time+0x4d5/0x9a0
  ...
 51)       80      80   syscall_call+0x7/0xb
 ^^^
   it's correct.
Signed-off-by: Nwalimis <walimisdev@gmail.com>
LKML-Reference: <1244016090-7814-1-git-send-email-walimisdev@gmail.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

Impact: cleanup

Most of the tracing files creation follow the same pattern:

ret = debugfs_create_file(...)
if (!ret)
	pr_warning("Couldn't create ... entry\n")

Unify it!
Reported-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <1238109938-11840-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

Ingo Molnar suggested, instead of:

        Depth    Size      Location    (27 entries)
        -----    ----      --------
  0)     2880      48   lock_timer_base+0x2b/0x4f
  1)     2832      80   __mod_timer+0x33/0xe0
  2)     2752      16   __ide_set_handler+0x63/0x65

To have it be:

        Depth    Size   Location    (27 entries)
        -----    ----   --------
  0)     2880      48   lock_timer_base+0x2b/0x4f
  1)     2832      80   __mod_timer+0x33/0xe0
  2)     2752      16   __ide_set_handler+0x63/0x65
Requested-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>

If the stack tracing is disabled (by default) the stack_trace file
will only contain the header:

 # cat /debug/tracing/stack_trace
        Depth    Size      Location    (0 entries)
        -----    ----      --------

This can be frustrating to a developer that does not realize that the
stack tracer is disabled. This patch adds the following text:

  # cat /debug/tracing/stack_trace
        Depth    Size      Location    (0 entries)
        -----    ----      --------
 #
 #  Stack tracer disabled
 #
 # To enable the stack tracer, either add 'stacktrace' to the
 # kernel command line
 # or 'echo 1 > /proc/sys/kernel/stack_tracer_enabled'
 #
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>

The stack tracer use to look like this:

 # cat /debug/tracing/stack_trace
         Depth  Size      Location    (57 entries)
         -----  ----      --------
  0)     5088      16   mempool_alloc_slab+0x16/0x18
  1)     5072     144   mempool_alloc+0x4d/0xfe
  2)     4928      16   scsi_sg_alloc+0x48/0x4a [scsi_mod]

Now it looks like this:

 # cat /debug/tracing/stack_trace

        Depth    Size      Location    (57 entries)
        -----    ----      --------
  0)     5088      16   mempool_alloc_slab+0x16/0x18
  1)     5072     144   mempool_alloc+0x4d/0xfe
  2)     4928      16   scsi_sg_alloc+0x48/0x4a [scsi_mod]
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>

Impact: clean up

Andrew Morton suggested to use the stack_tracer_enabled variable
to decide whether or not to start stack tracing on bootup.
This lets us remove the start_stack_trace variable.
Reported-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: enhancement to stack tracer

The stack tracer currently is either on when configured in or
off when it is not. It can not be disabled when it is configured on.
(besides disabling the function tracer that it uses)

This patch adds a way to enable or disable the stack tracer at
run time. It defaults off on bootup, but a kernel parameter 'stacktrace'
has been added to enable it on bootup.

A new sysctl has been added "kernel.stack_tracer_enabled" to let
the user enable or disable the stack tracer at run time.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: fix to output of stack trace

If a function is not found in the stack of the stack tracer, the
number printed is quite strange. This fixes the algorithm to handle
missing functions better.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: fix for lockdep and ftrace

The raw_local_irq_save/restore confuses lockdep. This patch
converts them to the local_irq_save/restore variants.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: make output of stack_trace complete if buffer overruns

When read buffer overruns, the output of stack_trace isn't complete.

When printing records with seq_printf in t_show, if the read buffer
has overruned by the current record, then this record won't be
printed to user space through read buffer, it will just be dropped in
this printing.

When next printing, t_start should return the "*pos"th record, which
is the one dropped by previous printing, but it just returns
(m->private + *pos)th record.

Here we use a more sane method to implement seq_operations which can
be found in kernel code. Thus we needn't initialize m->private.

About testing, it's not easy to overrun read buffer, but we can use
seq_printf to print more padding bytes in t_show, then it's easy to
check whether or not records are lost.

This commit has been tested on both condition of overrun and non
overrun.
Signed-off-by: NLiming Wang <liming.wang@windriver.com>
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: use new, consolidated APIs in ftrace plugins

This patch replaces the schedule safe preempt disable code with the
ftrace_preempt_disable() and ftrace_preempt_enable() safe functions.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The stack trace API does not record if the stack is not on the current
task's stack. That is, if the stack is the interrupt stack or NMI stack,
the output does not show. Also, the size of those stacks are not
consistent with the size of the thread stack, this makes the calculation
of the stack size usually bogus.

This all confuses the stack tracer. I unfortunately do not have time to
fix all these problems, but this patch does record the worst stack when
the stack pointer is on the tasks stack (instead of bogus numbers).

The patch simply returns if the stack pointer is not on the task's stack.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch adds indexes into the stack that the functions in the
stack dump were found at. As an added bonus, I also added a diff
to show which function is the most notorious consumer of the stack.

The output now looks like this:

# cat /debug/tracing/stack_trace
        Depth   Size      Location    (48 entries)
        -----   ----      --------
  0)     2476     212   blk_recount_segments+0x39/0x59
  1)     2264      12   bio_phys_segments+0x16/0x1d
  2)     2252      20   blk_rq_bio_prep+0x23/0xaf
  3)     2232      12   init_request_from_bio+0x74/0x77
  4)     2220      56   __make_request+0x294/0x331
  5)     2164     136   generic_make_request+0x34f/0x37d
  6)     2028      56   submit_bio+0xe7/0xef
  7)     1972      28   submit_bh+0xd1/0xf0
  8)     1944     112   block_read_full_page+0x299/0x2a9
  9)     1832       8   blkdev_readpage+0x14/0x16
 10)     1824      28   read_cache_page_async+0x7e/0x109
 11)     1796      16   read_cache_page+0x11/0x49
 12)     1780      32   read_dev_sector+0x3c/0x72
 13)     1748      48   read_lba+0x4d/0xaa
 14)     1700     168   efi_partition+0x85/0x61b
 15)     1532      72   rescan_partitions+0x10e/0x266
 16)     1460      40   do_open+0x1c7/0x24e
 17)     1420     292   __blkdev_get+0x79/0x84
 18)     1128      12   blkdev_get+0x12/0x14
 19)     1116      20   register_disk+0xd1/0x11e
 20)     1096      28   add_disk+0x34/0x90
 21)     1068      52   sd_probe+0x2b1/0x366
 22)     1016      20   driver_probe_device+0xa5/0x120
 23)      996       8   __device_attach+0xd/0xf
 24)      988      32   bus_for_each_drv+0x3e/0x68
 25)      956      24   device_attach+0x56/0x6c
 26)      932      16   bus_attach_device+0x26/0x4d
 27)      916      64   device_add+0x380/0x4b4
 28)      852      28   scsi_sysfs_add_sdev+0xa1/0x1c9
 29)      824     160   scsi_probe_and_add_lun+0x919/0xa2a
 30)      664      36   __scsi_add_device+0x88/0xae
 31)      628      44   ata_scsi_scan_host+0x9e/0x21c
 32)      584      28   ata_host_register+0x1cb/0x1db
 33)      556      24   ata_host_activate+0x98/0xb5
 34)      532     192   ahci_init_one+0x9bd/0x9e9
 35)      340      20   pci_device_probe+0x3e/0x5e
 36)      320      20   driver_probe_device+0xa5/0x120
 37)      300      20   __driver_attach+0x3f/0x5e
 38)      280      36   bus_for_each_dev+0x40/0x62
 39)      244      12   driver_attach+0x19/0x1b
 40)      232      28   bus_add_driver+0x9c/0x1af
 41)      204      28   driver_register+0x76/0xd2
 42)      176      20   __pci_register_driver+0x44/0x71
 43)      156       8   ahci_init+0x14/0x16
 44)      148     100   _stext+0x42/0x122
 45)       48      20   kernel_init+0x175/0x1dc
 46)       28      28   kernel_thread_helper+0x7/0x10

The first column is simply an index starting from the inner most function
and counting down to the outer most.

The next column is the location that the function was found on the stack.

The next column is the size of the stack for that function.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This is another tracer using the ftrace infrastructure, that examines
at each function call the size of the stack. If the stack use is greater
than the previous max it is recorded.

You can always see (and set) the max stack size seen. By setting it
to zero will start the recording again. The backtrace is also available.

For example:

# cat /debug/tracing/stack_max_size
1856

# cat /debug/tracing/stack_trace
[<c027764d>] stack_trace_call+0x8f/0x101
[<c021b966>] ftrace_call+0x5/0x8
[<c02553cc>] clocksource_get_next+0x12/0x48
[<c02542a5>] update_wall_time+0x538/0x6d1
[<c0245913>] do_timer+0x23/0xb0
[<c0257657>] tick_do_update_jiffies64+0xd9/0xf1
[<c02576b9>] tick_sched_timer+0x4a/0xad
[<c0250fe6>] __run_hrtimer+0x3e/0x75
[<c02518ed>] hrtimer_interrupt+0xf1/0x154
[<c022c870>] smp_apic_timer_interrupt+0x71/0x84
[<c021b7e9>] apic_timer_interrupt+0x2d/0x34
[<c0238597>] finish_task_switch+0x29/0xa0
[<c05abd13>] schedule+0x765/0x7be
[<c05abfca>] schedule_timeout+0x1b/0x90
[<c05ab4d4>] wait_for_common+0xab/0x101
[<c05ab5ac>] wait_for_completion+0x12/0x14
[<c033cfc3>] blk_execute_rq+0x84/0x99
[<c0402470>] scsi_execute+0xc2/0x105
[<c040250a>] scsi_execute_req+0x57/0x7f
[<c043afe0>] sr_test_unit_ready+0x3e/0x97
[<c043bbd6>] sr_media_change+0x43/0x205
[<c046b59f>] media_changed+0x48/0x77
[<c046b5ff>] cdrom_media_changed+0x31/0x37
[<c043b091>] sr_block_media_changed+0x16/0x18
[<c02b9e69>] check_disk_change+0x1b/0x63
[<c046f4c3>] cdrom_open+0x7a1/0x806
[<c043b148>] sr_block_open+0x78/0x8d
[<c02ba4c0>] do_open+0x90/0x257
[<c02ba869>] blkdev_open+0x2d/0x56
[<c0296a1f>] __dentry_open+0x14d/0x23c
[<c0296b32>] nameidata_to_filp+0x24/0x38
[<c02a1c68>] do_filp_open+0x347/0x626
[<c02967ef>] do_sys_open+0x47/0xbc
[<c02968b0>] sys_open+0x23/0x2b
[<c021aadd>] sysenter_do_call+0x12/0x26

I've tested this on both x86_64 and i386.
Signed-off-by: NSteven Rostedt <srostedt@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>