| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
fix event_enter_##sname->event
fix event_exit_##sname->event
remove unused event_syscall_enter and event_syscall_exit
Signed-off-by: Lai Jiangshan <[email protected]>
Acked-by: Jason Baron <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Move slow_work's debugging proc file to debugfs.
Signed-off-by: David Howells <[email protected]>
Requested-and-acked-by: Ingo Molnar <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Commits 3d7a641 ("SLOW_WORK: Wait for outstanding work items belonging to a
module to clear") introduced some code to make sure that all of a module's
slow-work items were complete before that module was removed, and commit
3bde31a ("SLOW_WORK: Allow a requeueable work item to sleep till the thread is
needed") further extended that, breaking it in the process if CONFIG_MODULES=n:
CC kernel/slow-work.o
kernel/slow-work.c: In function 'slow_work_execute':
kernel/slow-work.c:313: error: 'slow_work_thread_processing' undeclared (first use in this function)
kernel/slow-work.c:313: error: (Each undeclared identifier is reported only once
kernel/slow-work.c:313: error: for each function it appears in.)
kernel/slow-work.c: In function 'slow_work_wait_for_items':
kernel/slow-work.c:950: error: 'slow_work_unreg_sync_lock' undeclared (first use in this function)
kernel/slow-work.c:951: error: 'slow_work_unreg_wq' undeclared (first use in this function)
kernel/slow-work.c:961: error: 'slow_work_unreg_work_item' undeclared (first use in this function)
kernel/slow-work.c:974: error: 'slow_work_unreg_module' undeclared (first use in this function)
kernel/slow-work.c:977: error: 'slow_work_thread_processing' undeclared (first use in this function)
make[1]: *** [kernel/slow-work.o] Error 1
Fix this by:
(1) Extracting the bits of slow_work_execute() that are contingent on
CONFIG_MODULES, and the bits that should be, into inline functions and
placing them into the #ifdef'd section that defines the relevant variables
and adding stubs for moduleless kernels. This allows the removal of some
#ifdefs.
(2) #ifdef'ing out the contents of slow_work_wait_for_items() in moduleless
kernels.
The four functions related to handling module unloading synchronisation (and
their associated variables) could be offloaded into a separate .c file, but
each function is only used once and three of them are tiny, so doing so would
prevent them from being inlined.
Reported-by: Ingo Molnar <[email protected]>
Signed-off-by: David Howells <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
In current code in perf_swevent_hrtimer(), data.period is not
initialized, The result is obvious wrong:
# ./perf record -f -e cpu-clock make
# ./perf report
# Samples: 1740
#
# Overhead Command ......
# ........ ........ ..........................................
#
1025422183050275328.00% sh libc-2.9.90.so ...
1025422183050275328.00% perl libperl.so ...
1025422168240043264.00% perl [kernel] ...
1025422030011210752.00% perl [kernel] ...
Signed-off-by: Xiao Guangrong <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Fix a memory leak case in create_trace_probe(). When an argument
is too long (> MAX_ARGSTR_LEN), it just jumps to error path. In
that case tp->args[i].name is not released.
This also fixes a bug to check kstrdup()'s return value.
Signed-off-by: Masami Hiramatsu <[email protected]>
Cc: systemtap <[email protected]>
Cc: DLE <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Jim Keniston <[email protected]>
Cc: Ananth N Mavinakayanahalli <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Frank Ch. Eigler <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: Jason Baron <[email protected]>
Cc: K.Prasad <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Srikar Dronamraju <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <20091201001919.10235.56455.stgit@harusame>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
fork() clones all thread_info flags, including
TIF_USER_RETURN_NOTIFY; if the new task is first scheduled on a cpu
which doesn't have user return notifiers set, this causes user
return notifiers to trigger without any way of clearing itself.
This is easy to trigger with a forky workload on the host in
parallel with kvm, resulting in a cpu in an endless loop on the
verge of returning to userspace.
Fix by dropping the TIF_USER_RETURN_NOTIFY immediately after fork.
Signed-off-by: Avi Kivity <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
"symbol_name+0" is not so friendly.
It makes the output longer.
Signed-off-by: Lai Jiangshan <[email protected]>
Acked-by: Masami Hiramatsu <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Sometimes the group name is not "kprobes",
It'll be better if we can read it from tracing/kprobe_events.
# echo 'r:laijs/vfs_read vfs_read %ax' > kprobe_events
# cat kprobe_events
r:laijs/vfs_read vfs_read %ax=%ax
Signed-off-by: Lai Jiangshan <[email protected]>
Acked-by: Masami Hiramatsu <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
tp->nr_args is not set before we "goto error",
it causes memory leak for free_trace_probe() use tp->nr_args
to free memory of args.
Signed-off-by: Lai Jiangshan <[email protected]>
Acked-by: Masami Hiramatsu <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Field syscall number is missed in syscall_enter_define_fields()/
syscall_exit_define_fields().
Syscall number is also needed for event filter or other users.
Signed-off-by: Lai Jiangshan <[email protected]>
Acked-by: Frederic Weisbecker <[email protected]>
Cc: Jason Baron <[email protected]>
Cc: Steven Rostedt <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Kernel breakpoints are created using functions in which we pass
breakpoint parameters as individual variables: address, length
and type.
Although it fits well for x86, this just does not scale across
architectures that may support this api later as these may have
more or different needs. Pass in a perf_event_attr structure
instead because it is meant to evolve as much as possible into
a generic hardware breakpoint parameter structure.
Reported-by: K.Prasad <[email protected]>
Signed-off-by: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
In-kernel user breakpoints are created using functions in which
we pass breakpoint parameters as individual variables: address,
length and type.
Although it fits well for x86, this just does not scale across
archictectures that may support this api later as these may have
more or different needs. Pass in a perf_event_attr structure
instead because it is meant to evolve as much as possible into
a generic hardware breakpoint parameter structure.
Reported-by: K.Prasad <[email protected]>
Signed-off-by: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
I'm seeing spikes of up to 0.5ms in khungtaskd on a large
machine. To reduce this source of jitter I tried setting
hung_task_check_count to 0:
# echo 0 > /proc/sys/kernel/hung_task_check_count
which didn't have the intended response. Change to a post
increment of max_count, so a value of 0 means check 0 tasks.
Signed-off-by: Anton Blanchard <[email protected]>
Acked-by: Frederic Weisbecker <[email protected]>
Cc: [email protected]
LKML-Reference: <20091127022820.GU32182@kryten>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
When a pinned group cannot be scheduled it goes into error state.
Normally a group cannot go out of error state without being
explicitly re-enabled or disabled. There was a bug in per-thread
mode, whereby upon termination of the thread, the group would
transition from error to off leading to bogus counts and timing
information returned by read().
Fix it by clearing the error state.
Signed-off-by: Stephane Eranian <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Use of msecs_to_jiffies() for nsecs_to_cputime() have some
problems:
- The type of msecs_to_jiffies()'s argument is unsigned int, so
it cannot convert msecs greater than UINT_MAX = about 49.7 days.
- msecs_to_jiffies() returns MAX_JIFFY_OFFSET if MSB of argument
is set, assuming that input was negative value. So it cannot
convert msecs greater than INT_MAX = about 24.8 days too.
This patch defines a new function nsecs_to_jiffies() that can
deal greater values, and that can deal all incoming values as
unsigned.
Signed-off-by: Hidetoshi Seto <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: Stanislaw Gruszka <[email protected]>
Cc: Spencer Candland <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Amrico Wang <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: John Stultz <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Now all task_{u,s}time() pairs are replaced by task_times().
And task_gtime() is too simple to be an inline function.
Cleanup them all.
Signed-off-by: Hidetoshi Seto <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: Stanislaw Gruszka <[email protected]>
Cc: Spencer Candland <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Americo Wang <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Functions task_{u,s}time() are called in pair in almost all
cases. However task_stime() is implemented to call task_utime()
from its inside, so such paired calls run task_utime() twice.
It means we do heavy divisions (div_u64 + do_div) twice to get
utime and stime which can be obtained at same time by one set
of divisions.
This patch introduces a function task_times(*tsk, *utime,
*stime) to retrieve utime and stime at once in better, optimized
way.
Signed-off-by: Hidetoshi Seto <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: Stanislaw Gruszka <[email protected]>
Cc: Spencer Candland <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Americo Wang <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Add signal_overflow_fail and signal_lose_info tracepoints
for signal-lost events.
Changes in v3:
- Add docbook style comments
Changes in v2:
- Use siginfo string macro
Suggested-by: Roland McGrath <[email protected]>
Reviewed-by: Jason Baron <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Acked-by: Roland McGrath <[email protected]>
Cc: systemtap <[email protected]>
Cc: DLE <[email protected]>
Cc: Oleg Nesterov <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Add a tracepoint where a process gets a signal. This tracepoint
shows signal-number, sa-handler and sa-flag.
Changes in v3:
- Add docbook style comments
Changes in v2:
- Add siginfo argument
- Fix comment
Signed-off-by: Masami Hiramatsu <[email protected]>
Reviewed-by: Jason Baron <[email protected]>
Acked-by: Roland McGrath <[email protected]>
Cc: systemtap <[email protected]>
Cc: DLE <[email protected]>
Cc: Oleg Nesterov <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Move signal sending event to events/signal.h. This patch also
renames sched_signal_send event to signal_generate.
Changes in v4:
- Fix a typo of task_struct pointer.
Changes in v3:
- Add docbook style comments
Changes in v2:
- Add siginfo argument
- Add siginfo storing macro
Signed-off-by: Masami Hiramatsu <[email protected]>
Reviewed-by: Jason Baron <[email protected]>
Acked-by: Roland McGrath <[email protected]>
Cc: systemtap <[email protected]>
Cc: DLE <[email protected]>
Cc: Oleg Nesterov <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Merge reason: Pick up fixes that did not make it into .32.0
Signed-off-by: Ingo Molnar <[email protected]>
|
|
bp_perf_event_destroy() is unused in its off-case version, let's
remove it to fix the following warning reported by Stephen
Rothwell in linux-next:
kernel/perf_event.c:4306: warning: 'bp_perf_event_destroy' defined but not used
Reported-by: Stephen Rothwell <[email protected]>
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Limit the number of per cpu calibration messages by only
printing out results for the first cpu to boot.
Also, don't print "CPUx is down" as this is expected, and we
don't need 4096 reminders... ;-)
Signed-off-by: Mike Travis <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Roland Dreier <[email protected]>
Cc: Randy Dunlap <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Yinghai Lu <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Rusty Russell <[email protected]>
Cc: Hidetoshi Seto <[email protected]>
Cc: Jack Steiner <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Remove the verbose scheduler debug messages unless kernel
parameter "sched_debug" set. /proc/sched_debug unchanged.
Signed-off-by: Mike Travis <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Roland Dreier <[email protected]>
Cc: Randy Dunlap <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Yinghai Lu <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Rusty Russell <[email protected]>
Cc: Hidetoshi Seto <[email protected]>
Cc: Jack Steiner <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
If the new percpu tree is combined with the perf events tree
the following new warning triggers:
kernel/hw_breakpoint.c: In function 'toggle_bp_task_slot':
kernel/hw_breakpoint.c:151: warning: 'task_bp_pinned' is used uninitialized in this function
Because it's not valid anymore to define a local variable
and a percpu variable (even if it's file scope local) with
the same name.
Rename the local variable to resolve this.
Signed-off-by: Andrew Morton <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: K.Prasad <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Linus Torvalds <[email protected]>
LKML-Reference: <[email protected]>
[ v2: added changelog ]
Signed-off-by: Ingo Molnar <[email protected]>
|
|
This simplifies the error handling when we create a breakpoint.
We don't need to check the NULL return value corner case anymore
since we have improved perf_event_create_kernel_counter() to
always return an error code in the failure case.
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Prasad <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
In fail case, perf_event_create_kernel_counter() returns NULL
instead of an error, which doesn't help us to inform the user
about the origin of the problem from the outer most callers.
Often we can just return -EINVAL, which doesn't help anyone when
it's eventually about a memory allocation failure.
Then, this patch makes perf_event_create_kernel_counter() always
return a detailed error code.
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Prasad <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
The error path of a breakpoint modification is broken in
the ksym tracer. A modified breakpoint hlist node is immediately
released after its removal. Also we leak a breakpoint in this
case.
Fix the path.
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Prasad <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Running the ring-buffer-benchmark's threads at the lowest priority may
work well for keeping it in the background, but it is not appropriate
for the benchmarks.
This patch adds 4 parameters to the module:
consumer_fifo
consumer_nice
producer_fifo
producer_nice
By default the consumer and producer still run at nice +19.
If the *_fifo options are set, they will override the *_nice values.
modprobe ring_buffer_benchmark consumer_nice=0 producer_fifo=10
The above will set the consumer thread to a nice value of 0, and
the producer thread to a RT SCHED_FIFO priority of 10.
Note, this patch also fixes a bug where calling set_user_nice on the
consumer thread would oops the kernel when the parameter "disable_reader"
is set.
Signed-off-by: Steven Rostedt <[email protected]>
|
|
In commit v2.6.21-691-g39bc89f ("make SysRq-T show all tasks
again") the interface of show_state_filter() was changed: zero
valued 'state_filter' specifies "dump all tasks" (instead of -1).
However, the condition for calling debug_show_all_locks() ("show
locks if all tasks are dumped") was not updated accordingly.
Signed-off-by: Shmulik Ladkani <[email protected]>
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Commit ee949a86b3aef15845ea677aa60231008de62672 ("tracing/syscalls:
Use long for syscall ret format and field definitions") changed the
syscall exit return type to long, but forgot to change it in the
struct.
Signed-off-by: Tom Zanussi <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Commit 4ed7c92d68a5387ba5f7030dc76eab03558e27f5
(perf_events: Undo some recursion damage) has introduced a bad
reference counting of the recursion context. putting the context
behaves like getting it, dropping every software/trace events
after the first one in a context.
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arjan van de Ven <[email protected]>
Cc: Li Zefan <[email protected]>
Cc: Steven Rostedt <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Branch hint profiling on my nehalem machine showed over 90%
incorrect branch hints:
10420275 170645395 94 context_switch sched.c
3043
10408421 171098521 94 context_switch sched.c
3050
Signed-off-by: Tim Blechmann <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Branch hint profiling on my nehalem machine showed 90%
incorrect branch hints:
15728471 158903754 90 pick_next_task_fair
sched_fair.c 1555
Signed-off-by: Tim Blechmann <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
When using an event group, the value and id for non leaders events
were wrong due to invalid offset into the outgoing buffer.
Signed-off-by: Stephane Eranian <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: [email protected]
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
As far as I know, all distros currently ship kernels with default
CONFIG_SECURITY_FILE_CAPABILITIES=y. Since having the option on
leaves a 'no_file_caps' option to boot without file capabilities,
the main reason to keep the option is that turning it off saves
you (on my s390x partition) 5k. In particular, vmlinux sizes
came to:
without patch fscaps=n: 53598392
without patch fscaps=y: 53603406
with this patch applied: 53603342
with the security-next tree.
Against this we must weigh the fact that there is no simple way for
userspace to figure out whether file capabilities are supported,
while things like per-process securebits, capability bounding
sets, and adding bits to pI if CAP_SETPCAP is in pE are not supported
with SECURITY_FILE_CAPABILITIES=n, leaving a bit of a problem for
applications wanting to know whether they can use them and/or why
something failed.
It also adds another subtly different set of semantics which we must
maintain at the risk of severe security regressions.
So this patch removes the SECURITY_FILE_CAPABILITIES compile
option. It drops the kernel size by about 50k over the stock
SECURITY_FILE_CAPABILITIES=y kernel, by removing the
cap_limit_ptraced_target() function.
Changelog:
Nov 20: remove cap_limit_ptraced_target() as it's logic
was ifndef'ed.
Signed-off-by: Serge E. Hallyn <[email protected]>
Acked-by: Andrew G. Morgan" <[email protected]>
Signed-off-by: James Morris <[email protected]>
|
|
When libcap, or other libraries attempt to confirm/determine the supported
capability version magic, they generally supply a NULL dataptr to capget().
In this case, while returning the supported/preferred magic (via a
modified header content), the return code of this system call may be 0,
-EINVAL, or -EFAULT.
No libcap code depends on the previous -EINVAL etc. return code, and
all of the above three return codes can accompany a valid (successful)
attempt to determine the requested magic value.
This patch cleans up the system call to return 0, if the call is
successfully being used to determine the supported/preferred capability
magic value.
Signed-off-by: Andrew G. Morgan <[email protected]>
Acked-by: Steve Grubb <[email protected]>
Acked-by: Serge Hallyn <[email protected]>
Signed-off-by: James Morris <[email protected]>
|
|
sched_feat_write() should update ppos instead of file->f_pos.
(This reduces some BKL dependencies of this code.)
Signed-off-by: Jan Blunck <[email protected]>
Cc: [email protected]
Cc: Arnd Bergmann <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: Jamie Lokier <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Alan Cox <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Add the remaining necessary bits to support breakpoints created
through perf syscall.
We don't use the software counter interface as:
- We don't need to check against recursion, this is already done
in hardware breakpoints arch level.
- We already know the perf event we are dealing with when the
event is to be committed.
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Prasad <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Perf tools create perf events as disabled in the beginning.
Breakpoints are then considered like ptrace temporary
breakpoints, only meant to reserve a breakpoint slot until we
get all the necessary informations from the user.
In this case, we don't check the address that is breakpointed as
it is NULL in the ptrace case.
But perf tools don't have the same purpose, events are created
disabled to wait for all events to be created before enabling
all of them. We want to check the breakpoint parameters in this
case.
Signed-off-by: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Prasad <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Attribute authorship to developers of hw-breakpoint related
files.
Signed-off-by: K.Prasad <[email protected]>
Cc: Alan Stern <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
[ v2: moved it to latest -tip ]
Signed-off-by: Ingo Molnar <[email protected]>
|
|
It is quite possible to call update_event_times() on a context
that isn't actually running and thereby confuse the thing.
perf stat was reporting !100% scale values for software counters
(2e2af50b perf_events: Disable events when we detach them,
solved the worst of that, but there was still some left).
The thing that happens is that because we are not self-reaping
(we have a caring parent) there is a time between the last
schedule (out) and having do_exit() called which will detach the
events.
This period would be accounted as enabled,!running because the
event->state==INACTIVE, even though !event->ctx->is_active.
Similar issues could have been observed by calling read() on a
event while the attached task was not scheduled in.
Solve this by teaching update_event_times() about
ctx->is_active.
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <1258984836.4531.480.camel@laptop>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Make perf_swevent_get_recursion_context return a context number
and disable preemption.
This could be used to remove the IRQ disable from the trace bit
and index the per-cpu buffer with.
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: Paul Mackerras <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Move the update_event_times() call in __perf_event_exit_task()
into list_del_event() because that holds the proper lock
(ctx->lock) and seems a more natural place to do the last time
update.
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
It appeared we did call update_event_times() on exit, but we
failed to update the context time, which renders the former
moot.
Locking is a bit iffy, we call update_event_times under
ctx->mutex instead of ctx->lock - the next patch fixes this.
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
If we leave the event in STATE_INACTIVE, any read of the event
after the detach will increase the running count but not the
enabled count and cause funny scaling artefacts.
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
We had two almost identical functions, avoid the duplication.
Signed-off-by: Peter Zijlstra <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
The structure init creates a bit memcpy, which shows
up big time in perf annotate output:
: ffffffff810a859d <__perf_sw_event>:
1.68 : ffffffff810a859d: 55 push %rbp
1.69 : ffffffff810a859e: 41 89 fa mov %edi,%r10d
0.01 : ffffffff810a85a1: 49 89 c9 mov %rcx,%r9
0.00 : ffffffff810a85a4: 31 c0 xor %eax,%eax
1.71 : ffffffff810a85a6: b9 16 00 00 00 mov $0x16,%ecx
0.00 : ffffffff810a85ab: 48 89 e5 mov %rsp,%rbp
0.00 : ffffffff810a85ae: 48 83 ec 60 sub $0x60,%rsp
1.52 : ffffffff810a85b2: 48 8d 7d a0 lea -0x60(%rbp),%rdi
85.20 : ffffffff810a85b6: f3 ab rep stos %eax,%es:(%rdi)
None of the callees depends on the structure being pre-initialized,
so only initialize ->addr. This gets rid of the memcpy overhead.
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <[email protected]>
|
|
Clean up strstrip() usage - which also addresses this build warning:
kernel/trace/ftrace.c: In function 'ftrace_pid_write':
kernel/trace/ftrace.c:3004: warning: ignoring return value of 'strstrip', declared with attribute warn_unused_result
Cc: Steven Rostedt <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <[email protected]>
|