Skip to content

K
kernel_linux

OpenHarmony / kernel_linux
上一次同步 4 年多

通知 15
Star 8
Fork 2
K
kernel_linux
切换分支/标签
  1. 03 11月, 2015 1 次提交
  3. 26 10月, 2015 2 次提交
    • S
      tracing: Implement event pid filtering · 3fdaf80f
      由 Steven Rostedt (Red Hat) 提交于 
      Add the necessary hooks to use the pids loaded in set_event_pid to filter
all the events enabled in the tracing instance that match the pids listed.

Two probes are added to both sched_switch and sched_wakeup tracepoints to be
called before other probes are called and after the other probes are called.
The first is used to set the necessary flags to let the probes know to test
if they should be traced or not.

The sched_switch pre probe will set the "ignore_pid" flag if neither the
previous or next task has a matching pid.

The sched_switch probe will set the "ignore_pid" flag if the next task
does not match the matching pid.

The pre probe allows for probes tracing sched_switch to be traced if
necessary.

The sched_wakeup pre probe will set the "ignore_pid" flag if neither the
current task nor the wakee task has a matching pid.

The sched_wakeup post probe will set the "ignore_pid" flag if the current
task does not have a matching pid.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      3fdaf80f
    • S
      tracing: Add set_event_pid directory for future use · 49090107
      由 Steven Rostedt (Red Hat) 提交于 
      Create a tracing directory called set_event_pid, which currently has no
function, but will be used to filter all events for the tracing instance or
the pids that are added to the file.

The reason no functionality is added with this commit is that this commit
focuses on the creation and removal of the pids in a safe manner. And tests
can be made against this change to make sure things are correct before
hooking features to the list of pids.

Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      49090107
  5. 01 10月, 2015 4 次提交
    • S
      tracing: Add trace options for tracer options to instances · 37aea98b
      由 Steven Rostedt (Red Hat) 提交于 
      Add the tracer options to instances options directory as well. Only add the
options for tracers that are allowed to be enabled by an instance. But note,
that tracer options are global. That is, tracer options enabled in an
instance, also take affect at the top level and in other instances.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      37aea98b
    • S
      tracing: Add a method to pass in trace_array descriptor to option files · 9a38a885
      由 Steven Rostedt (Red Hat) 提交于 
      In preparation of having the multi buffer instances having their own trace
option flags, the trace option files needs a way to not only pass in the
flag they represent, but also the trace_array descriptor.

A new field is added to the trace_array descriptor called trace_flags_index,
which is a 32 byte character array representing a bit. This array is simply
filled with the index of the array, where

  index_array[n] = n;

Then the address of this array is passed to the file callbacks instead of
the index of the flag index. Then to retrieve both the flag index and the
trace_array descriptor:

  data is the passed in argument.

  index = *(unsigned char *)data;

  data -= index;

  /* Now data points to the address of the array in the trace_array */

  tr = container_of(data, struct trace_array, trace_flags_index);
Suggested-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      9a38a885
    • S
      tracing: Move trace_flags from global to a trace_array field · 983f938a
      由 Steven Rostedt (Red Hat) 提交于 
      In preparation to make trace options per instance, the global trace_flags
needs to be moved from being a global variable to a field within the trace
instance trace_array structure.

There's still more work to do, as there's some functions that use
trace_flags without passing in a way to get to the current_trace array. For
those, the global_trace is used directly (from trace.c). This includes
setting and clearing the trace_flags. This means that when a new instance is
created, it just gets the trace_flags of the global_trace and will not be
able to modify them. Depending on the functions that have access to the
trace_array, the flags of an instance may not affect parts of its trace,
where the global_trace is used. These will be fixed in future changes.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      983f938a
    • S
      tracing: Move sleep-time and graph-time options out of the core trace_flags · 55577204
      由 Steven Rostedt (Red Hat) 提交于 
      The sleep-time and graph-time options are only for the function graph tracer
and are not used by anything else. As tracer options are now visible when
the tracer is not activated, its better to move the function graph specific
tracer options into the function graph tracer.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      55577204
  7. 30 9月, 2015 11 次提交
    • S
      tracing: Remove access to trace_flags in trace_printk.c · b9f9108c
      由 Steven Rostedt (Red Hat) 提交于 
      In the effort to move the global trace_flags to the tracing instances, the
direct access to trace_flags must be removed from trace_printk.c

Instead, add a new trace_printk_enabled boolean that is set by a new access
function trace_printk_control(), that will enable or disable trace_printk.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      b9f9108c
    • S
      tracing: Add build bug if we have more trace_flags than bits · b5e87c05
      由 Steven Rostedt (Red Hat) 提交于 
      Add a enum that denotes the last bit of the trace_flags and have a
BUILD_BUG_ON(last_bit > 32).

If we add more bits than we have in trace_flags, the kernel wont build.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      b5e87c05
    • S
      tracing: Always show all tracer options in the options directory · 41d9c0be
      由 Steven Rostedt (Red Hat) 提交于 
      There are options that are unique to a specific tracer (like function and
function graph). Currently, these options are only visible in the options
directory when the tracer is enabled.

This has been a pain, especially for something like the func_stack_trace
option that if used inappropriately, could bring the system to a crawl. But
the only way to see it, is to enable the function tracer.

For example, if one had done:

 # cd /sys/kernel/tracing
 # echo __schedule > set_ftrace_filter
 # echo 1 > options/func_stack_trace
 # echo function > current_tracer

The __schedule call will be traced and a stack trace will also be recorded
there. Now when you were done, you may do...

 # echo nop > current_tracer
 # echo > set_ftrace_filter

But you forgot to disable the func_stack_trace. The only way to disable it
is to re-enable function tracing first. If you do not add a filter to
set_ftrace_filter and just do:

 # echo function > current_tracer

Now you would be performing a stack trace on *every* function! On some
systems, that causes a live lock. Others may take a few minutes to fix your
mistake.

Having the func_stack_trace option visible allows you to check it and
disable it before enabling the funtion tracer.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      41d9c0be
    • S
      tracing: Only create stacktrace option when STACKTRACE is configured · 73dddbb5
      由 Steven Rostedt (Red Hat) 提交于 
      Only create the stacktrace trace option when CONFIG_STACKTRACE is
configured.

Cleaned up the ftrace_trace_stack() function call a little to allow better
encapsulation of the stacktrace trace flag.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      73dddbb5
    • S
      tracing: Do not create function tracer options when not compiled in · 8179e8a1
      由 Steven Rostedt (Red Hat) 提交于 
      When the function tracer is not compiled in, do not create the option files
for it.

Fix up both the sched_wakeup and irqsoff tracers to handle the change.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      8179e8a1
    • S
      tracing: Only create branch tracer options when compiled in · 4ee4301c
      由 Steven Rostedt (Red Hat) 提交于 
      When the branch tracer is not compiled in, do not create the option files
associated to it.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      4ee4301c
    • S
      tracing: Only create function graph options when it is compiled in · 729358da
      由 Steven Rostedt (Red Hat) 提交于 
      Do not create fuction graph tracer options when function graph tracer is not
even compiled in.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      729358da
    • S
      tracing: Use TRACE_FLAGS macro to keep enums and strings matched · a3418a36
      由 Steven Rostedt (Red Hat) 提交于 
      Use a cute little macro trick to keep the names of the trace flags file
guaranteed to match the corresponding masks.

The macro TRACE_FLAGS is defined as a serious of enum names followed by
the string name of the file that matches it. For example:

 #define TRACE_FLAGS						\
		C(PRINT_PARENT,		"print-parent"),	\
		C(SYM_OFFSET,		"sym-offset"),		\
		C(SYM_ADDR,		"sym-addr"),		\
		C(VERBOSE,		"verbose"),

Now we can define the following:

 #undef C
 #define C(a, b) TRACE_ITER_##a##_BIT
 enum trace_iterator_bits { TRACE_FLAGS };

The above creates:

 enum trace_iterator_bits {
	TRACE_ITER_PRINT_PARENT_BIT,
	TRACE_ITER_SYM_OFFSET_BIT,
	TRACE_ITER_SYM_ADDR_BIT,
	TRACE_ITER_VERBOSE_BIT,
 };

Then we can redefine C as:

 #undef C
 #define C(a, b) TRACE_ITER_##a = (1 << TRACE_ITER_##a##_BIT)
 enum trace_iterator_flags { TRACE_FLAGS };

Which creates:

 enum trace_iterator_flags {
	TRACE_ITER_PRINT_PARENT	= (1 << TRACE_ITER_PRINT_PARENT_BIT),
	TRACE_ITER_SYM_OFFSET	= (1 << TRACE_ITER_SYM_OFFSET_BIT),
	TRACE_ITER_SYM_ADDR	= (1 << TRACE_ITER_SYM_ADDR_BIT),
	TRACE_ITER_VERBOSE	= (1 << TRACE_ITER_VERBOSE_BIT),
 };

Then finally we can create the list of file names:

 #undef C
 #define C(a, b) b
 static const char *trace_options[] = {
	TRACE_FLAGS
	NULL
 };

Which creates:
 static const char *trace_options[] = {
	"print-parent",
	"sym-offset",
	"sym-addr",
	"verbose",
	NULL
 };

The importance of this is that the strings match the bit index.

	trace_options[TRACE_ITER_SYM_ADDR_BIT] == "sym-addr"
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      a3418a36
    • S
      tracing: Use enums instead of hard coded bitmasks for TRACE_ITER flags · ce3fed62
      由 Steven Rostedt (Red Hat) 提交于 
      Using enums with FLAG_BIT and then defining a FLAG = (1 << FLAG_BIT), is a
bit more robust as we require that there are no bits out of order or skipped
to match the file names that represent the bits.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      ce3fed62
    • S
      tracing: Remove unused tracing option "ftrace_preempt" · 938db5f5
      由 Steven Rostedt (Red Hat) 提交于 
      There was a time where the function tracing would disable interrupts unless
specifically told not to, where it would only disable preemption. With the
new lockless code, the function tracing never disalbes interrupts and just
uses disabling of preemption. Remove the option "ftrace_preempt" as it does
nothing anyway.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      938db5f5
    • S
      tracing: Move "display-graph" option to main options · 03905582
      由 Steven Rostedt (Red Hat) 提交于 
      In order to facilitate making all tracer options visible even when the
tracer is not active, we need to get rid of duplicate options. Any option
that is shared between multiple tracers really should be a main option.

As the wakeup and irqsoff tracers both use the "display-graph" option, and
use it exactly the same way, move that option from the tracer options to the
main options and consolidate them.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      03905582
  9. 28 9月, 2015 1 次提交
  11. 26 9月, 2015 1 次提交
  13. 08 7月, 2015 1 次提交
    • S
      tracing: Have branch tracer use recursive field of task struct · 6224beb1
      由 Steven Rostedt (Red Hat) 提交于 
      Fengguang Wu's tests triggered a bug in the branch tracer's start up
test when CONFIG_DEBUG_PREEMPT set. This was because that config
adds some debug logic in the per cpu field, which calls back into
the branch tracer.

The branch tracer has its own recursive checks, but uses a per cpu
variable to implement it. If retrieving the per cpu variable calls
back into the branch tracer, you can see how things will break.

Instead of using a per cpu variable, use the trace_recursion field
of the current task struct. Simply set a bit when entering the
branch tracing and clear it when leaving. If the bit is set on
entry, just don't do the tracing.

There's also the case with lockdep, as the local_irq_save() called
before the recursion can also trigger code that can call back into
the function. Changing that to a raw_local_irq_save() will protect
that as well.

This prevents the recursion and the inevitable crash that follows.

Link: http://lkml.kernel.org/r/20150630141803.GA28071@wfg-t540p.sh.intel.com

Cc: stable@vger.kernel.org # 3.10+
Reported-by: NFengguang Wu <fengguang.wu@intel.com>
Tested-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      6224beb1
  15. 26 6月, 2015 1 次提交
  17. 14 5月, 2015 4 次提交
  19. 08 4月, 2015 1 次提交
    • S
      tracing: Add TRACE_DEFINE_ENUM() macro to map enums to their values · 0c564a53
      由 Steven Rostedt (Red Hat) 提交于 
      Several tracepoints use the helper functions __print_symbolic() or
__print_flags() and pass in enums that do the mapping between the
binary data stored and the value to print. This works well for reading
the ASCII trace files, but when the data is read via userspace tools
such as perf and trace-cmd, the conversion of the binary value to a
human string format is lost if an enum is used, as userspace does not
have access to what the ENUM is.

For example, the tracepoint trace_tlb_flush() has:

 __print_symbolic(REC->reason,
    { TLB_FLUSH_ON_TASK_SWITCH, "flush on task switch" },
    { TLB_REMOTE_SHOOTDOWN, "remote shootdown" },
    { TLB_LOCAL_SHOOTDOWN, "local shootdown" },
    { TLB_LOCAL_MM_SHOOTDOWN, "local mm shootdown" })

Which maps the enum values to the strings they represent. But perf and
trace-cmd do no know what value TLB_LOCAL_MM_SHOOTDOWN is, and would
not be able to map it.

With TRACE_DEFINE_ENUM(), developers can place these in the event header
files and ftrace will convert the enums to their values:

By adding:

 TRACE_DEFINE_ENUM(TLB_FLUSH_ON_TASK_SWITCH);
 TRACE_DEFINE_ENUM(TLB_REMOTE_SHOOTDOWN);
 TRACE_DEFINE_ENUM(TLB_LOCAL_SHOOTDOWN);
 TRACE_DEFINE_ENUM(TLB_LOCAL_MM_SHOOTDOWN);

 $ cat /sys/kernel/debug/tracing/events/tlb/tlb_flush/format
[...]
 __print_symbolic(REC->reason,
    { 0, "flush on task switch" },
    { 1, "remote shootdown" },
    { 2, "local shootdown" },
    { 3, "local mm shootdown" })

The above is what userspace expects to see, and tools do not need to
be modified to parse them.

Link: http://lkml.kernel.org/r/20150403013802.220157513@goodmis.org

Cc: Guilherme Cox <cox@computer.org>
Cc: Tony Luck <tony.luck@gmail.com>
Cc: Xie XiuQi <xiexiuqi@huawei.com>
Acked-by: NNamhyung Kim <namhyung@kernel.org>
Reviewed-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Tested-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      0c564a53
  21. 04 2月, 2015 1 次提交
    • S
      tracing: Convert the tracing facility over to use tracefs · 8434dc93
      由 Steven Rostedt (Red Hat) 提交于 
      debugfs was fine for the tracing facility as a quick way to get
an interface. Now that tracing has matured, it should separate itself
from debugfs such that it can be mounted separately without needing
to mount all of debugfs with it. That is, users resist using tracing
because it requires mounting debugfs. Having tracing have its own file
system lets users get the features of tracing without needing to bring
in the rest of the kernel's debug infrastructure.

Another reason for tracefs is that debubfs does not support mkdir.
Currently, to create instances, one does a mkdir in the tracing/instance
directory. This is implemented via a hack that forces debugfs to do
something it is not intended on doing. By converting over to tracefs, this
hack can be removed and mkdir can be properly implemented. This patch does
not address this yet, but it lays the ground work for that to be done.
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      8434dc93
  23. 02 2月, 2015 1 次提交
  25. 23 12月, 2014 1 次提交
  27. 15 12月, 2014 2 次提交
  29. 04 12月, 2014 1 次提交
    • B
      tracing: Add additional marks to signal very large time deltas · 8e1e1df2
      由 Byungchul Park 提交于 
      Currently, function graph tracer prints "!" or "+" just before
function execution time to signal a function overhead, depending
on the time. And some tracers tracing latency also print "!" or
"+" just after time to signal overhead, depending on the interval
between events. Even it is usually enough to do that, we sometimes
need to signal for bigger execution time than 100 micro seconds.

For example, I used function graph tracer to detect if there is
any case that exit_mm() takes too much time. I did following steps
in /sys/kernel/debug/tracing. It was easier to detect very large
excution time with patched kernel than with original kernel.

$ echo exit_mm > set_graph_function
$ echo function_graph > current_tracer
$ echo > trace
$ cat trace_pipe > $LOGFILE
 ... (do something and terminate logging)
$ grep "\\$" $LOGFILE
 3) $ 22082032 us |                      } /* kernel_map_pages */
 3) $ 22082040 us |                    } /* free_pages_prepare */
 3) $ 22082113 us |                  } /* free_hot_cold_page */
 3) $ 22083455 us |                } /* free_hot_cold_page_list */
 3) $ 22083895 us |              } /* release_pages */
 3) $ 22177873 us |            } /* free_pages_and_swap_cache */
 3) $ 22178929 us |          } /* unmap_single_vma */
 3) $ 22198885 us |        } /* unmap_vmas */
 3) $ 22206949 us |      } /* exit_mmap */
 3) $ 22207659 us |    } /* mmput */
 3) $ 22207793 us |  } /* exit_mm */

And then, it was easy to find out that a schedule-out occured by
sub_preempt_count() within kernel_map_pages().

To detect very large function exection time caused by either problematic
function implementation or scheduling issues, this patch can be useful.

Link: http://lkml.kernel.org/r/1416789259-24038-1-git-send-email-byungchul.park@lge.comSigned-off-by: NByungchul Park <byungchul.park@lge.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>
      8e1e1df2
  31. 20 11月, 2014 2 次提交
  33. 12 11月, 2014 2 次提交
  35. 19 7月, 2014 1 次提交
  37. 11 6月, 2014 1 次提交
  39. 06 6月, 2014 1 次提交