At init time, if the system time is "warped" forward in warp_clock()
it will differ from the hardware clock by sys_tz.tz_minuteswest. This time
difference is not taken into account when ntp updates the hardware clock,
and this causes the system time to jump forward by this offset every reboot.
The kernel must take this offset into account when writing the system time
to the hardware clock in the ntp code. This patch adds
persistent_clock_is_local which indicates that an offset has been applied
in warp_clock() and accounts for the "warp" before writing the hardware
clock.
x86 does not have this problem as rtc writes are software limited to a
+/-15 minute window relative to the current rtc time. Other arches, such
as powerpc, however do a full synchronization of the system time to the
rtc and will see this problem.
[v2]: generated against tip/timers/core
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Synchronize with 'net' in order to sort out some l2tp, wireless, and
ipv6 GRE fixes that will be built on top of in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
uprobe_perf_open/close call the costly uprobe_apply() every time,
we can avoid it if:
- "nr_systemwide != 0" is not changed.
- There is another process/thread with the same ->mm.
- copy_proccess() does inherit_event(). dup_mmap() preserves the
inserted breakpoints.
- event->attr.enable_on_exec == T, we can rely on uprobe_mmap()
called by exec/mmap paths.
- tp_target is exiting. Only _close() checks PF_EXITING, I don't
think TRACE_REG_PERF_OPEN can hit the dying task too often.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Change uprobe_trace_func() and uprobe_perf_func() to return "int". Change
uprobe_dispatcher() to return "trace_ret | perf_ret" although this is not
needed, currently TP_FLAG_TRACE/TP_FLAG_PROFILE are mutually exclusive.
The only functional change is that uprobe_perf_func() checks the filtering
too and returns UPROBE_HANDLER_REMOVE if nobody wants to trace current.
Testing:
# perf probe -x /lib/libc.so.6 syscall
# perf record -e probe_libc:syscall -i perl -e 'fork; syscall -1 for 1..10; wait'
# perf report --show-total-period
100.00% 10 perl libc-2.8.so [.] syscall
Before this patch:
# cat /sys/kernel/debug/tracing/uprobe_profile
/lib/libc.so.6 syscall 20
A child process doesn't have a counter, but still it hits this breakoint
"copied" by dup_mmap().
After the patch:
# cat /sys/kernel/debug/tracing/uprobe_profile
/lib/libc.so.6 syscall 11
The child process hits this int3 only once and does unapply_uprobe().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Finally implement uprobe_perf_filter() which checks ->nr_systemwide or
->perf_events to figure out whether we need to insert the breakpoint.
uprobe_perf_open/close are changed to do uprobe_apply(true/false) when
the new perf event comes or goes away.
Note that currently this is very suboptimal:
- uprobe_register() called by TRACE_REG_PERF_REGISTER becomes a
heavy nop, consumer->filter() always returns F at this stage.
As it was already discussed we need uprobe_register_only() to
avoid the costly register_for_each_vma() when possible.
- uprobe_apply() is oftenly overkill. Unless "nr_systemwide != 0"
changes we need uprobe_apply_mm(), unapply_uprobe() is almost
what we need.
- uprobe_apply() can be simply avoided sometimes, see the next
changes.
Testing:
# perf probe -x /lib/libc.so.6 syscall
# perl -e 'syscall -1 while 1' &
[1] 530
# perf record -e probe_libc:syscall perl -e 'syscall -1 for 1..10; sleep 1'
# perf report --show-total-period
100.00% 10 perl libc-2.8.so [.] syscall
Before this patch:
# cat /sys/kernel/debug/tracing/uprobe_profile
/lib/libc.so.6 syscall 79291
A huge ->nrhit == 79291 reflects the fact that the background process
530 constantly hits this breakpoint too, even if doesn't contribute to
the output.
After the patch:
# cat /sys/kernel/debug/tracing/uprobe_profile
/lib/libc.so.6 syscall 10
This shows that only the target process was punished by int3.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Introduce "struct trace_uprobe_filter" which records the "active"
perf_event's attached to ftrace_event_call. For the start we simply
use list_head, we can optimize this later if needed. For example, we
do not really need to record an event with ->parent != NULL, we can
rely on parent->child_list. And we can certainly do some optimizations
for the case when 2 events have the same ->tp_target or tp_target->mm.
Change trace_uprobe_register() to process TRACE_REG_PERF_OPEN/CLOSE
and add/del this perf_event to the list.
We can probably avoid any locking, but lets start with the "obvioulsy
correct" trace_uprobe_filter->rwlock which protects everything.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
sys_perf_event_open()->perf_init_event(event) is called before
find_get_context(event), this means that event->ctx == NULL when
class->reg(TRACE_REG_PERF_REGISTER/OPEN) is called and thus it
can't know if this event is per-task or system-wide.
This patch adds hw_perf_event->tp_target for PERF_TYPE_TRACEPOINT,
this is analogous to PERF_TYPE_BREAKPOINT/bp_target we already have.
The patch also moves ->bp_target up so that it can overlap with the
new member, this can help the compiler to generate the better code.
trace_uprobe_register() will use it for prefiltering to avoid the
unnecessary breakpoints in mm's we do not want to trace.
->tp_target doesn't have its own reference, but we can rely on the
fact that either sys_perf_event_open() holds a reference, or it is
equal to event->ctx->task. So this pointer is always valid until
free_event().
Also add the "struct list_head tp_list" into this union. It is not
strictly necessary, but it can simplify the next changes and we can
add it for free.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Move tu->nhit++ from uprobe_trace_func() to uprobe_dispatcher().
->nhit counts how many time we hit the breakpoint inserted by this
uprobe, we do not want to loose this info if uprobe was enabled by
sys_perf_event_open().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
trace_uprobe->consumer and "struct uprobe_trace_consumer" add the
unnecessary indirection and complicate the code for no reason.
This patch simply embeds uprobe_consumer into "struct trace_uprobe",
all other changes only fix the compilation errors.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
probe_event_enable/disable() check tu->consumer != NULL to avoid the
wrong uprobe_register/unregister().
We are going to kill this pointer and "struct uprobe_trace_consumer",
so we add the new helper, is_trace_uprobe_enabled(), which can rely
on TP_FLAG_TRACE/TP_FLAG_PROFILE instead.
Note: the current logic doesn't look optimal, it is not clear why
TP_FLAG_TRACE/TP_FLAG_PROFILE are mutually exclusive, we will probably
change this later.
Also kill the unused TP_FLAG_UPROBE.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
probe_event_enable/disable() check tu->inode != NULL at the start.
This is ugly, if igrab() can fail create_trace_uprobe() should not
succeed and "postpone" the failure.
And S_ISREG(inode->i_mode) check added by d24d7dbf is not safe.
Note: alloc_uprobe() should probably check igrab() != NULL as well.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
probe_event_enable() does uprobe_register() and only after that sets
utc->tu and tu->consumer/flags. This can race with uprobe_dispatcher()
which can miss these assignments or see them out of order. Nothing
really bad can happen, but this doesn't look clean/safe.
And this does not allow to use uprobe_consumer->filter() we are going
to add, it is called by uprobe_register() and it needs utc->tu.
Change this code to initialize everything before uprobe_register(), and
reset tu->consumer/flags if it fails. We can't race with event_disable(),
the caller holds event_mutex, and if we could the code would be wrong
anyway.
In fact I think uprobe_trace_consumer should die, it buys nothing but
complicates the code. We can simply add uprobe_consumer into trace_uprobe.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
create_trace_uprobe() does kern_path() to find ->d_inode, but forgets
to do path_put(). We can do this right after igrab().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
The original pull message for uprobes (commit 654443e2) noted:
This tree includes uprobes support in 'perf probe' - but SystemTap
(and other tools) can take advantage of user probe points as well.
In order to actually be usable in module-based tools like SystemTap, the
interface needs to be exported. This patch first adds the obvious
exports for uprobe_register and uprobe_unregister. Then it also adds
one for task_user_regset_view, which is necessary to get the correct
state of userspace registers.
Signed-off-by: Josh Stone <jistone@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
utask->xol_vaddr is either zero or valid, remove the bogus
IS_ERR_VALUE() check in xol_free_insn_slot().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
handle_swbp() does get_utask() before can_skip_sstep() for no reason,
we do not need ->utask if can_skip_sstep() succeeds.
Move get_utask() to pre_ssout() who actually starts to use it. Move
the initialization of utask->active_uprobe/state as well. This way
the whole initialization is consolidated in pre_ssout().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
pre_ssout() should do xol_free_insn_slot() if arch_uprobe_pre_xol()
fails, otherwise nobody will free the allocated slot.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
pre_ssout()->xol_get_insn_slot() path is confusing and buggy. This patch
cleanups the code, the next one fixes the bug.
Change xol_get_insn_slot() to only allocate the slot and do nothing more,
move the initialization of utask->xol_vaddr/vaddr into pre_ssout().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Rename add_utask() into get_utask() and change it to allocate on
demand to simplify the caller. Like get_xol_area() it will have
more users.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Currently only xol_get_insn_slot() does get_xol_area() + xol_alloc_area(),
but this will have more users and we do not want to copy-and-paste this
code. This patch simply moves xol_alloc_area() into get_xol_area() to
simplify the current and future code.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Move alloc_page() from xol_add_vma() to xol_alloc_area() to cleanup
the code. This separates the memory allocations and consolidates the
-EALREADY cleanups and the error handling.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Change handle_swbp() to set regs->ip = bp_vaddr in advance, this is
what consumer->handler() needs but uprobe_get_swbp_addr() is not
exported.
This also simplifies the code and makes it more consistent across
the supported architectures. handle_swbp() becomes the only caller
of uprobe_get_swbp_addr().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Currrently the are 2 problems with pre-filtering:
1. It is not possible to add/remove a task (mm) after uprobe_register()
2. A forked child inherits all breakpoints and uprobe_consumer can not
control this.
This patch does the first step to improve the filtering. handler_chain()
removes the breakpoints installed by this uprobe from current->mm if all
handlers return UPROBE_HANDLER_REMOVE.
Note that handler_chain() relies on ->register_rwsem to avoid the race
with uprobe_register/unregister which can add/del a consumer, or even
remove and then insert the new uprobe at the same address.
Perhaps we will add uprobe_apply_mm(uprobe, mm, is_register) and teach
copy_mm() to do filter(UPROBE_FILTER_FORK), but I think this change makes
sense anyway.
Note: instead of checking the retcode from uc->handler, we could add
uc->filter(UPROBE_FILTER_BPHIT). But I think this is not optimal to
call 2 hooks in a row. This buys nothing, and if handler/filter do
something nontrivial they will probably do the same work twice.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Finally add uprobe_consumer->filter() and change consumer_filter()
to actually call this method.
Note that ->filter() accepts mm_struct, not task_struct. Because:
1. We do not have for_each_mm_user(mm, task).
2. Even if we implement for_each_mm_user(), ->filter() can
use it itself.
3. It is not clear who will actually need this interface to
do the "nontrivial" filtering.
Another argument is "enum uprobe_filter_ctx", consumer->filter() can
use it to figure out why/where it was called. For example, perhaps
we can add UPROBE_FILTER_PRE_REGISTER used by build_map_info() to
quickly "nack" the unwanted mm's. In this case consumer should know
that it is called under ->i_mmap_mutex.
See the previous discussion at http://marc.info/?t=135214229700002
Perhaps we should pass more arguments, vma/vaddr?
Note: this patch obviously can't help to filter out the child created
by fork(), this will be addressed later.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
filter_chain() was added into install_breakpoint/remove_breakpoint to
simplify the initial changes but this is sub-optimal.
This patch shifts the callsite to the callers, register_for_each_vma()
and uprobe_mmap(). This way:
- It will be easier to add the new arguments. This is the main reason,
we can do more optimizations later.
- register_for_each_vma(is_register => true) can be optimized, we only
need to consult the new consumer. The previous consumers were already
asked when they called uprobe_register().
This patch also moves the MMF_HAS_UPROBES check from remove_breakpoint(),
this allows to avoid the potentionally costly filter_chain(). Note that
register_for_each_vma(is_register => false) doesn't really need to take
->consumer_rwsem, but I don't think it makes sense to optimize this and
introduce filter_chain_lockless().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
uprobe_register() and uprobe_unregister() are the only users of
mutex_lock(uprobes_hash(inode)), and the only reason why we can't
simply remove it is that we need to ensure that delete_uprobe() is
not possible after alloc_uprobe() and before consumer_add().
IOW, we need to ensure that when we take uprobe->register_rwsem
this uprobe is still valid and we didn't race with _unregister()
which called delete_uprobe() in between.
With this patch uprobe_register() simply checks uprobe_is_active()
and retries if it hits this very unlikely race. uprobes_mutex[] is
no longer needed and can be removed.
There is another reason for this change, prepare_uprobe() should be
folded into alloc_uprobe() and we do not want to hold the extra locks
around read_mapping_page/etc.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
The lifetime of uprobe->rb_node and uprobe->inode is not refcounted,
delete_uprobe() is called when we detect that uprobe has no consumers,
and it would be deadly wrong to do this twice.
Change delete_uprobe() to WARN() if it was already called. We use
RB_CLEAR_NODE() to mark uprobe "inactive", then RB_EMPTY_NODE() can
be used to detect this case.
RB_EMPTY_NODE() is not used directly, we add the trivial helper for
the next change.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
uprobe_events counts the number of uprobes in uprobes_tree but
it is used as a boolean. We can use RB_EMPTY_ROOT() instead.
Probably no_uprobe_events() added by this patch can have more
callers, say, mmf_recalc_uprobes().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Anton Arapov <anton@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Now that ->register_rwsem is safe under ->mmap_sem we can kill
->copy_mutex and abuse down_write(&uprobe->consumer_rwsem).
This makes prepare_uprobe() even more ugly, but we should kill
it anyway.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Simply remove UPROBE_RUN_HANDLER and the corresponding code.
It can only help if uprobe has a single consumer, and in fact
it is no longer needed after handler_chain() was changed to use
->register_rwsem, we simply can not race with uprobe_register().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Now that it safe to use ->consumer_rwsem under ->mmap_sem we can
almost finish the implementation of filter_chain(). It still lacks
the actual uc->filter(...) call but othewrwise it is ready, just
it pretends that ->filter() always returns true.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Introduce uprobe->register_rwsem. It is taken for writing around
__uprobe_register/unregister.
Change handler_chain() to use this sem rather than consumer_rwsem.
The main reason for this change is that we have the nasty problem
with mmap_sem/consumer_rwsem dependency. filter_chain() needs to
protect uprobe->consumers like handler_chain(), but they can not
use the same lock. filter_chain() can be called under ->mmap_sem
(currently this is always true), but we want to allow ->handler()
to play with the probed task's memory, and this needs ->mmap_sem.
Alternatively we could use srcu, but synchronize_srcu() is very
slow and ->register_rwsem allows us to do more. In particular, we
can teach handler_chain() to do remove_breakpoint() if this bp is
"nacked" by all consumers, we know that we can't race with the
new consumer which does uprobe_register().
See also the next patches. uprobes_mutex[] is almost ready to die.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To support the filtering uprobe_register() should do
register_for_each_vma(true) every time the new consumer comes,
we need to install the previously nacked breakpoints.
Note:
- uprobes_mutex[] should die, what it actually protects is
alloc_uprobe().
- UPROBE_RUN_HANDLER should die too, obviously it can't work
unless uprobe has a single consumer. The consumer should
serialize with _register/_unregister itself. Or this flag
should live in uprobe_consumer->state.
- Perhaps we can do some optimizations later. For example, if
filter_chain() never returns false uprobe can record this
fact and avoid the unnecessary register_for_each_vma().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
uprobe_unregister() removes the breakpoints only if the last consumer
goes away. To support the filtering it should do this every time, we
want to remove the breakpoints which nobody else want to keep.
Note: given that filter_chain() is not actually implemented, this patch
itself doesn't change the behaviour yet, register_for_each_vma(false)
is a heavy "nop" unless there are no more consumers.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Add the new helper filter_chain(). Currently it is only placeholder,
the comment explains what is should do. We will change it later to
consult every consumer to decide whether we need to install the swbp.
Until then it works as if any consumer returns true, this matches the
current behavior.
Change install_breakpoint() to call filter_chain() instead of checking
uprobe->consumers != NULL. We obviously need this, and this equally
closes the race with _unregister().
Change remove_breakpoint() to call this helper too. Currently this is
pointless because remove_breakpoint() is only called when the last
consumer goes away, but we will change this.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
uprobe_consumer->filter() is pointless in its current form, kill it.
We will add it back, but with the different signature/semantics. Perhaps
we will even re-introduce the callsite in handler_chain(), but not to
just skip uc->handler().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
register/unregister verifies that inode/uc != NULL. For what?
This really looks like "hide the potential problem", the caller
should pass the valid data.
register() also checks uc->next == NULL, probably to prevent the
double-register but the caller can do other stupid/wrong things.
If we do this check, then we should document that uc->next should
be cleared before register() and add BUG_ON().
Also add the small comment about the i_size_read() check.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cosmetic. __set_bit(UPROBE_SKIP_SSTEP) is the part of initialization,
it is not clear why it is set in insert_uprobe().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
This reverts commit ff0d05bf73.
Now that we have all the locking fixes in place, we can revert the
revert. This re-enables lockdep tracking for the console lock,
daee779718.
Signed-off-by: Dave Airlie <airlied@redhat.com>
defined(@array) is deprecated in Perl and gives off a warning.
Restructure the code to remove that warning.
[ hpa: it would be interesting to revert to the timeconst.bc script.
It appears that the failures reported by akpm during testing of
that script was due to a known broken version of make, not a problem
with bc. The Makefile rules could probably be restructured to avoid
the make bug, or it is probably old enough that it doesn't matter. ]
Reported-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: <stable@vger.kernel.org>
The old SRCU implementation loads sp->completed within an
RCU-sched section, courtesy of preempt_disable(). This was required
due to the use of synchronize_sched() in the old implemenation's
synchronize_srcu(). However, the new implementation does not rely
on synchronize_sched(), so it in turn does not require the load of
sp->completed and the ->c[] counter to be in a single preempt-disabled
region of code. This commit therefore moves the sp->completed access
outside of the preempt-disabled region and applies ACCESS_ONCE().
The resulting code is almost as the same as before, but it removes the
now-misleading rcu_dereference_index_check() call.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Because synchronize_srcu_expedited() no longer uses
synchronize_rcu_sched_expedited(), synchronize_srcu_expedited() no longer
indirectly acquires any CPU-hotplug-related locks. This commit therefore
updates the comments accordingly.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The core of SRCU is changed, but synchronize_srcu()'s comments describe
the old algorithm. This commit therefore updates them to match the
new algorithm.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pack six lines of code into two lines.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Although synchronize_srcu() can sleep, it will not sleep if the fast
path succeeds, which means that illegal use of synchronize_rcu()
might go unnoticed. This commit therefore adds might_sleep(), which
unconditionally catches illegal use of synchronize_rcu() from atomic
context.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
This commit replaces disabling of preemption and decrement of a per-CPU
variable with this_cpu_dec(), which avoids preemption disabling on x86
and shortens the code on all platforms.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Currently, __queue_work() chooses the pool to queue a work item to and
then determines cwq from the target wq and the chosen pool. This is a
bit backwards in that we can determine cwq first and simply use
cwq->pool. This way, we can skip get_std_worker_pool() in queueing
path which will be a hurdle when implementing custom worker pools.
Update __queue_work() such that it chooses the target cwq and then use
cwq->pool instead of the other way around. While at it, add missing
{} in an if statement.
This patch doesn't introduce any functional changes.
tj: The original patch had two get_cwq() calls - the first to
determine the pool by doing get_cwq(cpu, wq)->pool and the second
to determine the matching cwq from get_cwq(pool->cpu, wq).
Updated the function such that it chooses cwq instead of pool and
removed the second call. Rewrote the description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
get_work_pool_id() currently first obtains pool using get_work_pool()
and then return pool->id. For an off-queue work item, this involves
obtaining pool ID from worker->data, performing idr_find() to find the
matching pool and then returning its pool->id which of course is the
same as the one which went into idr_find().
Just open code WORK_STRUCT_CWQ case and directly return pool ID from
work->data.
tj: The original patch dropped on-queue work item handling and renamed
the function to offq_work_pool_id(). There isn't much benefit in
doing so. Handling it only requires a single if() and we need at
least BUG_ON(), which is also a branch, even if we drop on-queue
handling. Open code WORK_STRUCT_CWQ case and keep the function in
line with get_work_pool(). Rewrote the description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
As nr_running is likely to be accessed from other CPUs during
try_to_wake_up(), it was kept outside worker_pool; however, while less
frequent, other fields in worker_pool are accessed from other CPUs
for, e.g., non-reentrancy check. Also, with recent pool related
changes, accessing nr_running matching the worker_pool isn't as simple
as it used to be.
Move nr_running inside worker_pool. Keep it aligned to cacheline and
define CPU pools using DEFINE_PER_CPU_SHARED_ALIGNED(). This should
give at least the same cacheline behavior.
get_pool_nr_running() is replaced with direct pool->nr_running
accesses.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Joonsoo Kim <js1304@gmail.com>
Move rt scheduler definitions out of include/linux/sched.h into
new file include/linux/sched/rt.h
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094707.7b9f825f@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add a /proc/sys/kernel scheduler knob named
sched_rr_timeslice_ms that allows global changing of the
SCHED_RR timeslice value. User visable value is in milliseconds
but is stored as jiffies. Setting to 0 (zero) resets to the
default (currently 100ms).
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094704.13751796@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move the sysctl-related bits from include/linux/sched.h into
a new file: include/linux/sched/sysctl.h. Then update source
files requiring access to those bits by including the new
header file.
Signed-off-by: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20130207094659.06dced96@riff.lan
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With the recent is-work-queued-here test simplification, the nested
if() in try_to_grab_pending() can be collapsed. Collapse it.
This patch is purely cosmetic.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Currently, determining whether a work item is queued on a locked pool
involves somewhat convoluted memory barrier dancing. It goes like the
following.
* When a work item is queued on a pool, work->data is updated before
work->entry is linked to the pending list with a wmb() inbetween.
* When trying to determine whether a work item is currently queued on
a pool pointed to by work->data, it locks the pool and looks at
work->entry. If work->entry is linked, we then do rmb() and then
check whether work->data points to the current pool.
This works because, work->data can only point to a pool if it
currently is or were on the pool and,
* If it currently is on the pool, the tests would obviously succeed.
* It it left the pool, its work->entry was cleared under pool->lock,
so if we're seeing non-empty work->entry, it has to be from the work
item being linked on another pool. Because work->data is updated
before work->entry is linked with wmb() inbetween, work->data update
from another pool is guaranteed to be visible if we do rmb() after
seeing non-empty work->entry. So, we either see empty work->entry
or we see updated work->data pointin to another pool.
While this works, it's convoluted, to put it mildly. With recent
updates, it's now guaranteed that work->data points to cwq only while
the work item is queued and that updating work->data to point to cwq
or back to pool is done under pool->lock, so we can simply test
whether work->data points to cwq which is associated with the
currently locked pool instead of the convoluted memory barrier
dancing.
This patch replaces the memory barrier based "are you still here,
really?" test with much simpler "does work->data points to me?" test -
if work->data points to a cwq which is associated with the currently
locked pool, the work item is guaranteed to be queued on the pool as
work->data can start and stop pointing to such cwq only under
pool->lock and the start and stop coincide with queue and dequeue.
tj: Rewrote the comments and description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
We plan to use work->data pointing to cwq as the synchronization
invariant when determining whether a given work item is on a locked
pool or not, which requires work->data pointing to cwq only while the
work item is queued on the associated pool.
With delayed_work updated not to overload work->data for target
workqueue recording, the only case where we still have off-queue
work->data pointing to cwq is try_to_grab_pending() which doesn't
update work->data after stealing a queued work item. There's no
reason for try_to_grab_pending() to not update work->data to point to
the pool instead of cwq, like the normal execution does.
This patch adds set_work_pool_and_keep_pending() which makes
work->data point to pool instead of cwq but keeps the pending bit
unlike set_work_pool_and_clear_pending() (surprise!).
After this patch, it's guaranteed that only queued work items point to
cwqs.
This patch doesn't introduce any visible behavior change.
tj: Renamed the new helper function to match
set_work_pool_and_clear_pending() and rewrote the description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
To avoid executing the same work item from multiple CPUs concurrently,
a work_struct records the last pool it was on in its ->data so that,
on the next queueing, the pool can be queried to determine whether the
work item is still executing or not.
A delayed_work goes through timer before actually being queued on the
target workqueue and the timer needs to know the target workqueue and
CPU. This is currently achieved by modifying delayed_work->work.data
such that it points to the cwq which points to the target workqueue
and the last CPU the work item was on. __queue_delayed_work()
extracts the last CPU from delayed_work->work.data and then combines
it with the target workqueue to create new work.data.
The only thing this rather ugly hack achieves is encoding the target
workqueue into delayed_work->work.data without using a separate field,
which could be a trade off one can make; unfortunately, this entangles
work->data management between regular workqueue and delayed_work code
by setting cwq pointer before the work item is actually queued and
becomes a hindrance for further improvements of work->data handling.
This can be easily made sane by adding a target workqueue field to
delayed_work. While delayed_work is used widely in the kernel and
this does make it a bit larger (<5%), I think this is the right
trade-off especially given the prospect of much saner handling of
work->data which currently involves quite tricky memory barrier
dancing, and don't expect to see any measureable effect.
Add delayed_work->wq and drop the delayed_work->work.data overloading.
tj: Rewrote the description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Currently, work_busy() first tests whether the work has a pool
associated with it and if not, considers it idle. This works fine
even for delayed_work.work queued on timer, as __queue_delayed_work()
sets cwq on delayed_work.work - a queued delayed_work always has its
cwq and thus pool associated with it.
However, we're about to update delayed_work queueing and this won't
hold. Update work_busy() such that it tests WORK_STRUCT_PENDING
before the associated pool. This doesn't make any noticeable behavior
difference now.
With work_pending() test moved, the function read a lot better with
"if (!pool)" test flipped to positive. Flip it.
While at it, lose the comment about now non-existent reentrant
workqueues.
tj: Reorganized the function and rewrote the description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Now that workqueue has moved away from gcwqs, workqueue no longer has
the need to have a CPU identifier indicating "no cpu associated" - we
now use WORK_OFFQ_POOL_NONE instead - and most uses of WORK_CPU_NONE
are gone.
The only left usage is as the end marker for for_each_*wq*()
iterators, where the name WORK_CPU_NONE is confusing w/o actual
WORK_CPU_NONE usages. Similarly, WORK_CPU_LAST which equals
WORK_CPU_NONE no longer makes sense.
Replace both WORK_CPU_NONE and LAST with WORK_CPU_END. This patch
doesn't introduce any functional difference.
tj: s/WORK_CPU_LAST/WORK_CPU_END/ and rewrote the description.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
All in-kernel users of class_find_device() don't really need mutable
data for match callback.
In two places (kernel/power/suspend_test.c, drivers/scsi/osd/osd_uld.c)
this patch changes match callbacks to use const search data.
The const is propagated to rtc_class_open() and power_supply_get_by_name()
parameters.
Note that there's a dev reference leak in suspend_test.c that's not
touched in this patch.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
TCP Appropriate Byte Count was added by me, but later disabled.
There is no point in maintaining it since it is a potential source
of bugs and Linux already implements other better window protection
heuristics.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Typical cputime stats infrastructure relies on the timer tick and
its periodic polling on the CPU to account the amount of time
spent by the CPUs and the tasks per high level domains such as
userspace, kernelspace, guest, ...
Now we are preparing to implement full dynticks capability on
Linux for Real Time and HPC users who want full CPU isolation.
This feature requires a cputime accounting that doesn't depend
on the timer tick.
To implement it, this new cputime infrastructure plugs into
kernel/user/guest boundaries to take snapshots of cputime and
flush these to the stats when needed. This performs pretty
much like CONFIG_VIRT_CPU_ACCOUNTING except that context location
and cputime snaphots are synchronized between write and read
side such that the latter can safely retrieve the pending tickless
cputime of a task and add it to its latest cputime snapshot to
return the correct result to the user.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=1Tzp
-----END PGP SIGNATURE-----
Merge tag 'full-dynticks-cputime-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks into sched/core
Pull full-dynticks (user-space execution is undisturbed and
receives no timer IRQs) preparation changes that convert the
cputime accounting code to be full-dynticks ready,
from Frederic Weisbecker:
"This implements the cputime accounting on full dynticks CPUs.
Typical cputime stats infrastructure relies on the timer tick and
its periodic polling on the CPU to account the amount of time
spent by the CPUs and the tasks per high level domains such as
userspace, kernelspace, guest, ...
Now we are preparing to implement full dynticks capability on
Linux for Real Time and HPC users who want full CPU isolation.
This feature requires a cputime accounting that doesn't depend
on the timer tick.
To implement it, this new cputime infrastructure plugs into
kernel/user/guest boundaries to take snapshots of cputime and
flush these to the stats when needed. This performs pretty
much like CONFIG_VIRT_CPU_ACCOUNTING except that context location
and cputime snaphots are synchronized between write and read
side such that the latter can safely retrieve the pending tickless
cputime of a task and add it to its latest cputime snapshot to
return the correct result to the user."
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In 7b270f6099 "sched: Bail out of yield_to when source and
target runqueue has one task" we changed this to store -ESRCH so
it needs to be signed.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: kbuild@01.org
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mike Galbraith <efault@gmx.de>
Link: http://lkml.kernel.org/r/20130205113751.GA20521@elgon.mountain
Signed-off-by: Ingo Molnar <mingo@kernel.org>
hrtimer_enqueue_reprogram contains a race which could result in
timer.base switch during unlock/lock sequence.
hrtimer_enqueue_reprogram is releasing the lock protecting the timer
base for calling raise_softirq_irqsoff() due to a lock ordering issue
versus rq->lock.
If during that time another CPU calls __hrtimer_start_range_ns() on
the same hrtimer, the timer base might switch, before the current CPU
can lock base->lock again and therefor the unlock_timer_base() call
will unlock the wrong lock.
[ tglx: Added comment and massaged changelog ]
Signed-off-by: Leonid Shatz <leonid.shatz@ravellosystems.com>
Signed-off-by: Izik Eidus <izik.eidus@ravellosystems.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1359981217-389-1-git-send-email-izik.eidus@ravellosystems.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Conflicts:
kernel/irq_work.c
Add support for printk in full dynticks CPU.
* Don't stop tick with irq works pending. This
fix is generally useful and concerns archs that
can't raise self IPIs.
* Flush irq works before CPU offlining.
* Introduce "lazy" irq works that can wait for the
next tick to be executed, unless it's stopped.
* Implement klogd wake up using irq work. This
removes the ad-hoc printk_tick()/printk_needs_cpu()
hooks and make it working even in dynticks mode.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
There's no reason kgdb.h itself needs to include the 8250 serial port
header file. So push it down to the _very_ limited number of individual
drivers that need the values in that file, and fix up the places where
people really wanted serial_core.h and platform_device.h.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull scheduler fixes from Ingo Molnar:
"Three small fixlets"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/debug: Fix format string for 32-bit platforms
sched: Fix warning in kernel/sched/fair.c
sched/rt: Use root_domain of rt_rq not current processor
Pull perf fixes from Ingo Molnar:
"Three fixlets and two small (and low risk) hw-enablement changes"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Fix event group context move
x86/perf: Add IvyBridge EP support
perf/x86: Fix P6 driver section warning
arch/x86/tools/insn_sanity.c: Identify source of messages
perf/x86: Enable Intel Lincroft/Penwell/Cloverview Atom support
Pull two small RCU fixlets from Ingo Molnar.
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
rcu: Make rcu_nocb_poll an early_param instead of module_param
rcu: Prevent soft-lockup complaints about no-CBs CPUs
The uses of trace_clock_local() are dead code when CONFIG_RCU_TRACE=n,
but some compilers might nevertheless generate code calling this function.
This commit therefore ensures that trace_clock_local() is invoked only
when CONFIG_RCU_TRACE=y.
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If the previous CPU is cache affine and idle, select it.
The current implementation simply traverses the sd_llc domain,
taking the first idle CPU encountered, which walks buddy pairs
hand in hand over the package, inflicting excruciating pain.
1 tbench pair (worst case) in a 10 core + SMT package:
pre 15.22 MB/sec 1 procs
post 252.01 MB/sec 1 procs
Signed-off-by: Mike Galbraith <bitbucket@online.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1359371965.5783.127.camel@marge.simpson.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
As no one is using the return value of irq_work_queue(),
so it is better to just make it void.
Signed-off-by: anish kumar <anish198519851985@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
[ Fix stale comments, remove now unnecessary __irq_work_queue() intermediate function ]
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1359925703-24304-1-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
mips was the last architecture not using the generic variant.
Both native and compat variants switched to generic, which is
made unconditional now.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
usual "call force_sigsegv or signal_delivered" logics. Takes
ksignal instead of separate siginfo/k_sigaction.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Again, protected by a temporary config symbol (GENERIC_COMPAT_RT_SIGACTION);
will be gone by the end of series.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
... and make it unconditional - we want the sucker on all biarch
platforms, really. All kinds of wrappers and private implementations
can go now; fortunately, they don't cause name conflicts, so we can
do that one first without any bisect hazards.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
conditional on GENERIC_COMPAT_RT_SIGQUEUEINFO; by the end of that series
it will become the same thing as COMPAT and conditional will die out.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
conditional on GENERIC_COMPAT_RT_SIGPENDING; by the end of that series
it will become the same thing as COMPAT and conditional will die out.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
conditional on GENERIC_COMPAT_RT_SIGPROCMASK; by the end of that series
it will become the same thing as COMPAT and conditional will die out.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* pull compat version alongside with the native one
* make little-endian compat variant just call the native
* don't bother with separate conditional for compat (both native and
compat are going to become unconditional very soon).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Switch from __ARCH_WANT_SYS_RT_SIGACTION to opposite
(!CONFIG_ODD_RT_SIGACTION); the only two architectures that
need it are alpha and sparc. The reason for use of CONFIG_...
instead of __ARCH_... is that it's needed only kernel-side
and doing it that way avoids a mess with include order on many
architectures.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
again, strictly speaking we are in nasal daemon territory on ppc
and mips - we need to sign-extend int arguments.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Function next_prio() has been removed and pull_rt_task() is the
only user of pick_next_highest_task_rt() at the moment.
pull_rt_task is not interested in p->nr_cpus_allowed, its only
interest is the fact that cpu is allowed to execute p. If
nr_cpus_allowed == 1, cpu != task_cpu(p) and cpu is allowed then
it means that task p is in the middle of the migration
techniques; the task waits until it is moved by migration
thread. So, lets pull it earlier.
Signed-off-by: Kirill V Tkhai <tkhai@yandex.ru>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
CC: linux-rt-users <linux-rt-users@vger.kernel.org>
Link: http://lkml.kernel.org/r/70871359644177@web16d.yandex.ru
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When we have group with mixed events (hw/sw) we want to end up
with group leader being in hw context. So if group leader is
initialy sw event, we move all the events under hw context.
The move is done for each event by removing it from its context
and adding it back into proper one. As a part of the removal the
event is automatically disabled, which is not what we want at
this stage of creating groups.
The fix is to initialize event state after removal from sw
context.
This fix resulted from the following discussion:
http://thread.gmane.org/gmane.linux.kernel.perf.user/1144
Reported-by: Andreas Hollmann <hollmann@in.tum.de>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vince Weaver <vince@deater.net>
Link: http://lkml.kernel.org/r/1359714225-4231-1-git-send-email-jolsa@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On early boot up, when the ftrace ring buffer is initialized, the
static variable current_trace is initialized to &nop_trace.
Before this initialization, current_trace is NULL and will never
become NULL again. It is always reassigned to a ftrace tracer.
Several places check if current_trace is NULL before it uses
it, and this check is frivolous, because at the point in time
when the checks are made the only way current_trace could be
NULL is if ftrace failed its allocations at boot up, and the
paths to these locations would probably not be possible.
By initializing current_trace to &nop_trace where it is declared,
current_trace will never be NULL, and we can remove all these
checks of current_trace being NULL which never needed to be
checked in the first place.
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Hiraku Toyooka <hiraku.toyooka.gu@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Currently, the timer broadcast mechanism is defined by a function
pointer on struct clock_event_device. As the fundamental mechanism for
broadcast is architecture-specific, this means that clock_event_device
drivers cannot be shared across multiple architectures.
This patch adds an (optional) architecture-specific function for timer
tick broadcast, allowing drivers which may require broadcast
functionality to be shared across multiple architectures.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: nico@linaro.org
Cc: Will.Deacon@arm.com
Cc: Marc.Zyngier@arm.com
Cc: john.stultz@linaro.org
Link: http://lkml.kernel.org/r/1358183124-28461-3-git-send-email-mark.rutland@arm.com
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Currently the broadcast mechanism used for timers is abstracted by a
function pointer on struct clock_event_device. As the fundamental
mechanism for broadcast is architecture-specific, this ties each
clock_event_device driver to a single architecture, even where the
driver is otherwise generic.
This patch adds a standard path for the receipt of timer broadcasts, so
drivers and/or architecture backends need not manage redundant lists of
timers for the purpose of routing broadcast timer ticks.
[tglx: Made the implementation depend on the config switch as well ]
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: nico@linaro.org
Cc: Will.Deacon@arm.com
Cc: Marc.Zyngier@arm.com
Cc: john.stultz@linaro.org
Link: http://lkml.kernel.org/r/1358183124-28461-2-git-send-email-mark.rutland@arm.com
Tested-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
There are several places of consecutive calls of
dequeue_task_rt() and put_prev_task_rt() in the scheduler.
For example, function rt_mutex_setprio() does it.
The both calls lead to update_curr_rt(), the second of it
receives zeroed delta_exec. The only effective action in this
case is call of sched_rt_avg_update(), which can change
rq->age_stamp and rq->rt_avg. But it is possible in case of
""floating"" rq->clock. This fact is not reasonable to be
accounted. Another actions do nothing.
Signed-off-by: Kirill V Tkhai <tkhai@yandex.ru>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
CC: linux-rt-users <linux-rt-users@vger.kernel.org>
Link: http://lkml.kernel.org/r/931541359550236@web1g.yandex.ru
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull x86 fixes from Peter Anvin:
"This is a collection of miscellaneous fixes, the most important one is
the fix for the Samsung laptop bricking issue (auto-blacklisting the
samsung-laptop driver); the efi_enabled() changes you see below are
prerequisites for that fix.
The other issues fixed are booting on OLPC XO-1.5, an UV fix, NMI
debugging, and requiring CAP_SYS_RAWIO for MSR references, just as
with I/O port references."
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
samsung-laptop: Disable on EFI hardware
efi: Make 'efi_enabled' a function to query EFI facilities
smp: Fix SMP function call empty cpu mask race
x86/msr: Add capabilities check
x86/dma-debug: Bump PREALLOC_DMA_DEBUG_ENTRIES
x86/olpc: Fix olpc-xo1-sci.c build errors
arch/x86/platform/uv: Fix incorrect tlb flush all issue
x86-64: Fix unwind annotations in recent NMI changes
x86-32: Start out cr0 clean, disable paging before modifying cr3/4
This reverts commit daee779718.
I'll requeue this after the console locking fixes, so lockdep
is useful again for people until fbcon is fixed.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Ftrace has a snapshot feature available from kernel space and
latency tracers (e.g. irqsoff) are using it. This patch enables
user applictions to take a snapshot via debugfs.
Add "snapshot" debugfs file in "tracing" directory.
snapshot:
This is used to take a snapshot and to read the output of the
snapshot.
# echo 1 > snapshot
This will allocate the spare buffer for snapshot (if it is
not allocated), and take a snapshot.
# cat snapshot
This will show contents of the snapshot.
# echo 0 > snapshot
This will free the snapshot if it is allocated.
Any other positive values will clear the snapshot contents if
the snapshot is allocated, or return EINVAL if it is not allocated.
Link: http://lkml.kernel.org/r/20121226025300.3252.86850.stgit@liselsia
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: David Sharp <dhsharp@google.com>
Signed-off-by: Hiraku Toyooka <hiraku.toyooka.gu@hitachi.com>
[
Fixed irqsoff selftest and also a conflict with a change
that fixes the update_max_tr.
]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Currently the trace buffer read functions use a static variable
"old_tracer" for detecting if the current tracer changes. This
was suitable for a single trace file ("trace"), but to add a
snapshot feature that will use the same function for its file,
a check against a static variable is not sufficient.
To use the output functions for two different files, instead of
storing the current tracer in a static variable, as the trace
iterator descriptor contains a pointer to the original current
tracer's name, that pointer can now be used to check if the
current tracer has changed between different reads of the trace
file.
Link: http://lkml.kernel.org/r/20121226025252.3252.9276.stgit@liselsia
Signed-off-by: Hiraku Toyooka <hiraku.toyooka.gu@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
For systems with an unstable sched_clock, all cpu_clock() does is enable/
disable local irq during the call to sched_clock_cpu(). And for stable
systems they are same.
trace_clock_global() already disables interrupts, so it can call
sched_clock_cpu() directly.
Link: http://lkml.kernel.org/r/1356576585-28782-2-git-send-email-namhyung@kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Add a stat about the number of events read from the ring buffer:
# cat /debug/tracing/per_cpu/cpu0/stats
entries: 39869
overrun: 870512
commit overrun: 0
bytes: 1449912
oldest event ts: 6561.368690
now ts: 6565.246426
dropped events: 0
read events: 112 <-- Added
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
During kdump kernel's booting stage, it need to find low ram for
swiotlb buffer when system does not support intel iommu/dmar remapping.
kexed-tools is appending memmap=exactmap and range from /proc/iomem
with "Crash kernel", and that range is above 4G for 64bit after boot
protocol 2.12.
We need to add another range in /proc/iomem like "Crash kernel low",
so kexec-tools could find that info and append to kdump kernel
command line.
Try to reserve some under 4G if the normal "Crash kernel" is above 4G.
User could specify the size with crashkernel_low=XX[KMG].
-v2: fix warning that is found by Fengguang's test robot.
-v3: move out get_mem_size change to another patch, to solve compiling
warning that is found by Borislav Petkov <bp@alien8.de>
-v4: user must specify crashkernel_low if system does not support
intel or amd iommu.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Link: http://lkml.kernel.org/r/1359058816-7615-31-git-send-email-yinghai@kernel.org
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Rob Landley <rob@landley.net>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Jason pointed out the HAS_PERSISTENT_CLOCK name isn't
quite accurate for the config, as some systems may have
the persistent_clock in some cases, but not always.
So change the config name to the more clear
ALWAYS_USE_PERSISTENT_CLOCK.
Signed-off-by: John Stultz <john.stultz@linaro.org>
While debugging the virtual cputime with the function graph tracer
with a max_depth of 1 (most common use of the max_depth so far),
I found that I was missing kernel execution because of a race condition.
The code for the return side of the function has a slight race:
ftrace_pop_return_trace(&trace, &ret, frame_pointer);
trace.rettime = trace_clock_local();
ftrace_graph_return(&trace);
barrier();
current->curr_ret_stack--;
The ftrace_pop_return_trace() initializes the trace structure for
the callback. The ftrace_graph_return() uses the trace structure
for its own use as that structure is on the stack and is local
to this function. Then the curr_ret_stack is decremented which
is what the trace.depth is set to.
If an interrupt comes in after the ftrace_graph_return() but
before the curr_ret_stack, then the called function will get
a depth of 2. If max_depth is set to 1 this function will be
ignored.
The problem is that the trace has already been called, and the
timestamp for that trace will not reflect the time the function
was about to re-enter userspace. Calls to the interrupt will not
be traced because the max_depth has prevented this.
To solve this issue, the ftrace_graph_return() can safely be
moved after the current->curr_ret_stack has been updated.
This way the timestamp for the return callback will reflect
the actual time.
If an interrupt comes in after the curr_ret_stack update and
ftrace_graph_return(), it will be traced. It may look a little
confusing to see it within the other function, but at least
it will not be lost.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Bring in the 'net' tree so that we can get some ipv4/ipv6 bug
fixes that some net-next work will build upon.
Signed-off-by: David S. Miller <davem@davemloft.net>
In case of undercomitted scenarios, especially in large guests
yield_to overhead is significantly high. when run queue length of
source and target is one, take an opportunity to bail out and return
-ESRCH. This return condition can be further exploited to quickly come
out of PLE handler.
(History: Raghavendra initially worked on break out of kvm ple handler upon
seeing source runqueue length = 1, but it had to export rq length).
Peter came up with the elegant idea of return -ESRCH in scheduler core.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Raghavendra, Checking the rq length of target vcpu condition added.(thanks Avi)
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
Acked-by: Andrew Jones <drjones@redhat.com>
Tested-by: Chegu Vinod <chegu_vinod@hp.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
doctorture.2013.01.11a: Changes to rcutorture and to RCU documentation.
fixes.2013.01.26a: Miscellaneous fixes.
tagcb.2013.01.24a: Tag RCU callbacks with grace-period number to
simplify callback advancement.
tiny.2013.01.29b: Enhancements to uniprocessor handling in tiny RCU.
A number of kthreads have been added to rcutorture, but the shuffler
task was not informed of them, and thus did not shuffle them. This
commit therefore adds the requisite shuffling, and, while in the area
fixes up some whitespace issues.
However, the shuffling is intended to keep randomly selected CPUs
idle, which means that the RCU priority boosting kthreads need to
avoid waking up every jiffy. This commit also makes that fix.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tiny RCU has historically omitted RCU CPU stall warnings in order to
reduce memory requirements, however, lack of these warnings caused
Thomas Gleixner some debugging pain recently. Therefore, this commit
adds RCU CPU stall warnings to tiny RCU if RCU_TRACE=y. This keeps
the memory footprint small, while still enabling CPU stall warnings
in kernels built to enable them.
Updated to include Josh Triplett's suggested use of RCU_STALL_COMMON
config variable to simplify #if expressions.
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
I get the following warning every day with v3.7, once or
twice a day:
[ 2235.186027] WARNING: at /mnt/sda7/kernel/linux/arch/x86/kernel/apic/ipi.c:109 default_send_IPI_mask_logical+0x2f/0xb8()
As explained by Linus as well:
|
| Once we've done the "list_add_rcu()" to add it to the
| queue, we can have (another) IPI to the target CPU that can
| now see it and clear the mask.
|
| So by the time we get to actually send the IPI, the mask might
| have been cleared by another IPI.
|
This patch also fixes a system hang problem, if the data->cpumask
gets cleared after passing this point:
if (WARN_ONCE(!mask, "empty IPI mask"))
return;
then the problem in commit 83d349f35e ("x86: don't send an IPI to
the empty set of CPU's") will happen again.
Signed-off-by: Wang YanQing <udknight@gmail.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Jan Beulich <jbeulich@suse.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: peterz@infradead.org
Cc: mina86@mina86.org
Cc: srivatsa.bhat@linux.vnet.ibm.com
Cc: <stable@kernel.org>
Link: http://lkml.kernel.org/r/20130126075357.GA3205@udknight
[ Tidied up the changelog and the comment in the code. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While remotely reading the cputime of a task running in a
full dynticks CPU, the values stored in utime/stime fields
of struct task_struct may be stale. Its values may be those
of the last kernel <-> user transition time snapshot and
we need to add the tickless time spent since this snapshot.
To fix this, flush the cputime of the dynticks CPUs on
kernel <-> user transition and record the time / context
where we did this. Then on top of this snapshot and the current
time, perform the fixup on the reader side from task_times()
accessors.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
[fixed kvm module related build errors]
Signed-off-by: Sedat Dilek <sedat.dilek@gmail.com>
Do some ground preparatory work before adding guest_enter()
and guest_exit() context tracking callbacks. Those will
be later used to read the guest cputime safely when we
run in full dynticks mode.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Gleb Natapov <gleb@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
This is in preparation for the full dynticks feature. While
remotely reading the cputime of a task running in a full
dynticks CPU, we'll need to do some extra-computation. This
way we can account the time it spent tickless in userspace
since its last cputime snapshot.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Allow to dynamically switch between tick and virtual based
cputime accounting. This way we can provide a kind of "on-demand"
virtual based cputime accounting. In this mode, the kernel relies
on the context tracking subsystem to dynamically probe on kernel
boundaries.
This is in preparation for being able to stop the timer tick in
more places than just the idle state. Doing so will depend on
CONFIG_VIRT_CPU_ACCOUNTING_GEN which makes it possible to account
the cputime without the tick by hooking on kernel/user boundaries.
Depending whether the tick is stopped or not, we can switch between
tick and vtime based accounting anytime in order to minimize the
overhead associated to user hooks.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
If we want to stop the tick further idle, we need to be
able to account the cputime without using the tick.
Virtual based cputime accounting solves that problem by
hooking into kernel/user boundaries.
However implementing CONFIG_VIRT_CPU_ACCOUNTING require
low level hooks and involves more overhead. But we already
have a generic context tracking subsystem that is required
for RCU needs by archs which plan to shut down the tick
outside idle.
This patch implements a generic virtual based cputime
accounting that relies on these generic kernel/user hooks.
There are some upsides of doing this:
- This requires no arch code to implement CONFIG_VIRT_CPU_ACCOUNTING
if context tracking is already built (already necessary for RCU in full
tickless mode).
- We can rely on the generic context tracking subsystem to dynamically
(de)activate the hooks, so that we can switch anytime between virtual
and tick based accounting. This way we don't have the overhead
of the virtual accounting when the tick is running periodically.
And one downside:
- There is probably more overhead than a native virtual based cputime
accounting. But this relies on hooks that are already set anyway.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
If the architecture doesn't provide an implementation of
nsecs_to_cputime(), the cputime accounting core uses a
default one that converts the nanoseconds to jiffies. However
this only makes sense if we use the jiffies based cputime.
For now it doesn't matter much because this API is only
called on code that uses jiffies based cputime accounting.
But the code may evolve and this API may be used more
broadly in the future. Keeping this default implementation
around is very error prone as it may introduce a bug and
hide it on architectures that don't override this API.
Fix this by moving this definition to the jiffies based
cputime headers as it is the only place where it belongs to.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
We scale stime, utime values based on rtime (sum_exec_runtime
converted to jiffies). During scaling we multiple rtime * utime,
which seems to be fine, since both values are converted to u64,
but it's not.
Let assume HZ is 1000 - 1ms tick. Process consist of 64 threads,
run for 1 day, threads utilize 100% cpu on user space. Machine
has 64 cpus.
Process rtime = utime will be 64 * 24 * 60 * 60 * 1000 jiffies,
which is 0x149970000. Multiplication rtime * utime result is
0x1a855771100000000, which can not be covered in 64 bits.
Result of overflow is stall of utime values visible in user
space (prev_utime in kernel), even if application still consume
lot of CPU time.
A solution to solve this is to perform the multiplication on
stime instead of utime. It's easy to grow the utime value fast
with a CPU bound thread in userspace for example. Now we assume
that doing so with stime is much harder. In most cases a task
shouldn't ever spend much time in kernel space as it tends to
sleep waiting for jobs completion when they take long to
achieve. IO is the typical example of that.
Hence scaling the cputime by performing the multiplication on
stime instead of utime should considerably reduce the chances of
an overflow on most workloads.
This is largely inspired by a patch from Stanislaw Gruszka:
http://lkml.kernel.org/r/20130107113144.GA7544@redhat.com
Inspired-by: Stanislaw Gruszka <sgruszka@redhat.com>
Reported-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1359217182-25184-1-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When I initially wrote the code for /proc/<pid>/uid_map. I was lazy
and avoided duplicate mappings by the simple expedient of ensuring the
first number in a new extent was greater than any number in the
previous extent.
Unfortunately that precludes a number of valid mappings, and someone
noticed and complained. So use a simple check to ensure that ranges
in the mapping extents don't overlap.
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
When freeing a deeply nested user namespace free_user_ns calls
put_user_ns on it's parent which may in turn call free_user_ns again.
When -fno-optimize-sibling-calls is passed to gcc one stack frame per
user namespace is left on the stack, potentially overflowing the
kernel stack. CONFIG_FRAME_POINTER forces -fno-optimize-sibling-calls
so we can't count on gcc to optimize this code.
Remove struct kref and use a plain atomic_t. Making the code more
flexible and easier to comprehend. Make the loop in free_user_ns
explict to guarantee that the stack does not overflow with
CONFIG_FRAME_POINTER enabled.
I have tested this fix with a simple program that uses unshare to
create a deeply nested user namespace structure and then calls exit.
With 1000 nesteuser namespaces before this change running my test
program causes the kernel to die a horrible death. With 10,000,000
nested user namespaces after this change my test program runs to
completion and causes no harm.
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Pointed-out-by: Vasily Kulikov <segoon@openwall.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
This subsystem lacks many explanations on its purpose and
design. Add these missing comments.
v4: Document function parameter to be more kernel-doc
friendly, as per Namhyung suggestion.
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
As context tracking subsystem evolved, it stopped using ignore_user_qs
and in_user defined in the rcu_dynticks structure. This commit therefore
removes them.
Signed-off-by: Li Zhong <zhong@linux.vnet.ibm.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Small grammar fix in rcutree comment regarding 'rcu_scheduler_active'
var.
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Export the context state: whether we run in user / kernel
from the context tracking subsystem point of view.
This is going to be used by the generic virtual cputime
accounting subsystem that is needed to implement the full
dynticks.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
__this_cpu_inc_return() or __this_cpu_dec generates a single instruction,
which is faster than __get_cpu_var operation.
Link: http://lkml.kernel.org/r/50A9C1BD.1060308@gmail.com
Reviewed-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Shan Wei <davidshan@tencent.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The text in Documentation said it would be removed in 2.6.41;
the text in the Kconfig said removal in the 3.1 release. Either
way you look at it, we are well past both, so push it off a cliff.
Note that the POWER_CSTATE and the POWER_PSTATE are part of the
legacy tracing API. Remove all tracepoints which use these flags.
As can be seen from context, most already have a trace entry via
trace_cpu_idle anyways.
Also, the cpufreq/cpufreq.c PSTATE one is actually unpaired, as
compared to the CSTATE ones which all have a clear start/stop.
As part of this, the trace_power_frequency also becomes orphaned,
so it too is deleted.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
There's no need to test whether a (delayed) work item is pending
before queueing, flushing or cancelling it, so remove work_pending()
tests used in those cases.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
9fdb04cdc5 ("async: replace list of active domains with global list
of pending items") added a struct list_head global_list in struct
async_entry, which isn't initialised. This means that if
!domain->registered at __async_schedule(), then list_del_init() will
be called on the list head in async_run_entry_fn with both pointers
NULL, causing a crash. This is fixed by initialising both the
global_list and domain_list list_heads after kzalloc'ing the entry.
This was noticed due to dapm_power_widgets() which uses
ASYNC_DOMAIN_EXCLUSIVE, which initialises the domain->registered to 0.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: James Hogan <james.hogan@imgtec.com>
Reported-by: Stephen Warren <swarren@wwwdotorg.org>
The type returned from atomic64_t can be either unsigned
long or unsigned long long, depending on the architecture.
Using a cast to unsigned long long lets us use the same
format string for all architectures.
Without this patch, building with scheduler debugging
enabled results in:
kernel/sched/debug.c: In function 'print_cfs_rq':
kernel/sched/debug.c:225:2: warning: format '%ld' expects argument of type 'long int', but argument 4 has type 'long long int' [-Wformat]
kernel/sched/debug.c:225:2: warning: format '%ld' expects argument of type 'long int', but argument 3 has type 'long long int' [-Wformat]
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Turner <pjt@google.com>
Cc: linux-arm-kernel@list.infradead.org
Link: http://lkml.kernel.org/r/1359123276-15833-7-git-send-email-arnd@arndb.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
a4c96ae319 "sched: Unthrottle rt runqueues in
__disable_runtime()" turned the unthrottle_offline_cfs_rqs
function into a static symbol, which now triggers a warning
about it being potentially unused:
kernel/sched/fair.c:2055:13: warning: 'unthrottle_offline_cfs_rqs' defined but not used [-Wunused-function]
Marking it __maybe_unused shuts up the gcc warning and lets the
compiler safely drop the function body when it's not being used.
To reproduce, build the ARM bcm2835_defconfig.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Peter Boonstoppel <pboonstoppel@nvidia.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Turner <pjt@google.com>
Cc: linux-arm-kernel@list.infradead.org
Link: http://lkml.kernel.org/r/1359123276-15833-6-git-send-email-arnd@arndb.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The issue below was found in 2.6.34-rt rather than mainline rt
kernel, but the issue still exists upstream as well.
So please let me describe how it was noticed on 2.6.34-rt:
On this version, each softirq has its own thread, it means there
is at least one RT FIFO task per cpu. The priority of these
tasks is set to 49 by default. If user launches an RT FIFO task
with priority lower than 49 of softirq RT tasks, it's possible
there are two RT FIFO tasks enqueued one cpu runqueue at one
moment. By current strategy of balancing RT tasks, when it comes
to RT tasks, we really need to put them off to a CPU that they
can run on as soon as possible. Even if it means a bit of cache
line flushing, we want RT tasks to be run with the least latency.
When the user RT FIFO task which just launched before is
running, the sched timer tick of the current cpu happens. In this
tick period, the timeout value of the user RT task will be
updated once. Subsequently, we try to wake up one softirq RT
task on its local cpu. As the priority of current user RT task
is lower than the softirq RT task, the current task will be
preempted by the higher priority softirq RT task. Before
preemption, we check to see if current can readily move to a
different cpu. If so, we will reschedule to allow the RT push logic
to try to move current somewhere else. Whenever the woken
softirq RT task runs, it first tries to migrate the user FIFO RT
task over to a cpu that is running a task of lesser priority. If
migration is done, it will send a reschedule request to the found
cpu by IPI interrupt. Once the target cpu responds the IPI
interrupt, it will pick the migrated user RT task to preempt its
current task. When the user RT task is running on the new cpu,
the sched timer tick of the cpu fires. So it will tick the user
RT task again. This also means the RT task timeout value will be
updated again. As the migration may be done in one tick period,
it means the user RT task timeout value will be updated twice
within one tick.
If we set a limit on the amount of cpu time for the user RT task
by setrlimit(RLIMIT_RTTIME), the SIGXCPU signal should be posted
upon reaching the soft limit.
But exactly when the SIGXCPU signal should be sent depends on the
RT task timeout value. In fact the timeout mechanism of sending
the SIGXCPU signal assumes the RT task timeout is increased once
every tick.
However, currently the timeout value may be added twice per
tick. So it results in the SIGXCPU signal being sent earlier
than expected.
To solve this issue, we prevent the timeout value from increasing
twice within one tick time by remembering the jiffies value of
last updating the timeout. As long as the RT task's jiffies is
different with the global jiffies value, we allow its timeout to
be updated.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Fan Du <fan.du@windriver.com>
Reviewed-by: Yong Zhang <yong.zhang0@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1342508623-2887-1-git-send-email-ying.xue@windriver.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When the system has multiple domains do_sched_rt_period_timer()
can run on any CPU and may iterate over all rt_rq in
cpu_online_mask. This means when balance_runtime() is run for a
given rt_rq that rt_rq may be in a different rd than the current
processor. Thus if we use smp_processor_id() to get rd in
do_balance_runtime() we may borrow runtime from a rt_rq that is
not part of our rd.
This changes do_balance_runtime to get the rd from the passed in
rt_rq ensuring that we borrow runtime only from the correct rd
for the given rt_rq.
This fixes a BUG at kernel/sched/rt.c:687! in __disable_runtime
when we try reclaim runtime lent to other rt_rq but runtime has
been lent to a rt_rq in another rd.
Signed-off-by: Shawn Bohrer <sbohrer@rgmadvisors.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Mike Galbraith <bitbucket@online.de>
Cc: peterz@infradead.org
Cc: <stable@kernel.org>
Link: http://lkml.kernel.org/r/1358186131-29494-1-git-send-email-sbohrer@rgmadvisors.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since commit 48ddbe1946
("cgroup: make css->refcnt clearing on cgroup removal optional"),
each css holds a ref on cgroup's dentry, so cgroup_diput() won't be
called until all css' refs go down to 0, which invalids the comments.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Free cgroup via call_rcu(). The actual work is done through
workqueue.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
When destroying a cgroup, though in cgroup_diput() we've called
synchronize_rcu(), we then still have to free it via call_rcu().
The story is, long ago to fix a race between reading /proc/sched_debug
and freeing cgroup, the code was changed to utilize call_rcu(). See
commit a47295e6bc ("cgroups: make
cgroup_path() RCU-safe")
As we've fixed cpu cgroup that cpu_cgroup_offline_css() is used
to unregister a task_group so there won't be concurrent access
to this task_group after synchronize_rcu() in diput(). Now we can
just kfree(cgrp).
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
A task_group won't be online (thus no one can see it) until
cpu_cgroup_css_online(), and at that time tg->css.cgroup has
been initialized, so this NULL check is redundant.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
This is a preparaton for later patches.
- What do we gain from cpu_cgroup_css_online():
After ss->css_alloc() and before ss->css_online(), there's a small
window that tg->css.cgroup is NULL. With this change, tg won't be seen
before ss->css_online(), where it's added to the global list, so we're
guaranteed we'll never see NULL tg->css.cgroup.
- What do we gain from cpu_cgroup_css_offline():
tg is freed via RCU, so is cgroup. Without this change, This is how
synchronization works:
cgroup_rmdir()
no ss->css_offline()
diput()
syncornize_rcu()
ss->css_free() <-- unregister tg, and free it via call_rcu()
kfree_rcu(cgroup) <-- wait possible refs to cgroup, and free cgroup
We can't just kfree(cgroup), because tg might access tg->css.cgroup.
With this change:
cgroup_rmdir()
ss->css_offline() <-- unregister tg
diput()
synchronize_rcu() <-- wait possible refs to tg and cgroup
ss->css_free() <-- free tg
kfree_rcu(cgroup) <-- free cgroup
As you see, kfree_rcu() is redundant now.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Ingo Molnar <mingo@kernel.org>
With this change, we're guaranteed that cgroup_path() won't see NULL
cgrp->dentry, and thus we can remove the NULL check in it.
(Well, it's not strictly true, because dummptop.dentry is always NULL
but we already handle that separately.)
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
We checked for uprobe==NULL earlier, no need to redo that.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1356030701-16284-22-git-send-email-sasha.levin@oracle.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Remove remaining references to gcwq.
* __next_gcwq_cpu() steals __next_wq_cpu() name. The original
__next_wq_cpu() became __next_cwq_cpu().
* s/for_each_gcwq_cpu/for_each_wq_cpu/
s/for_each_online_gcwq_cpu/for_each_online_wq_cpu/
* s/gcwq_mayday_timeout/pool_mayday_timeout/
* s/gcwq_unbind_fn/wq_unbind_fn/
* Drop references to gcwq in comments.
This patch doesn't introduce any functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Rename per-cpu and unbound nr_running variables such that they match
the pool variables.
This patch doesn't introduce any functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
global_cwq is now nothing but a container for per-cpu standard
worker_pools. Declare the worker pools directly as
cpu/unbound_std_worker_pools[] and remove global_cwq.
* ____cacheline_aligned_in_smp moved from global_cwq to worker_pool.
This probably would have made sense even before this change as we
want each pool to be aligned.
* get_gcwq() is replaced with std_worker_pools() which returns the
pointer to the standard pool array for a given CPU.
* __alloc_workqueue_key() updated to use get_std_worker_pool() instead
of open-coding pool determination.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
v2: Joonsoo pointed out that it'd better to align struct worker_pool
rather than the array so that every pool is aligned.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Joonsoo Kim <js1304@gmail.com>
The only remaining user of pool->gcwq is std_worker_pool_pri().
Reimplement it using get_gcwq() and remove worker_pool->gcwq.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
for_each_std_worker_pool() takes @cpu instead of @gcwq.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Instead of holding locks from both pools and then processing the pools
together, make freezing/thwaing per-pool - grab locks of one pool,
process it, release it and then proceed to the next pool.
While this patch changes processing order across pools, order within
each pool remains the same. As each pool is independent, this
shouldn't break anything.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Instead of holding locks from both pools and then processing the pools
together, make hotplug processing per-pool - grab locks of one pool,
process it, release it and then proceed to the next pool.
rebind_workers() is updated to take and process @pool instead of @gcwq
which results in a lot of de-indentation. gcwq_claim_assoc_and_lock()
and its counterpart are replaced with in-line per-pool locking.
While this patch changes processing order across pools, order within
each pool remains the same. As each pool is independent, this
shouldn't break anything.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Move gcwq->lock to pool->lock. The conversion is mostly
straight-forward. Things worth noting are
* In many places, this removes the need to use gcwq completely. pool
is used directly instead. get_std_worker_pool() is added to help
some of these conversions. This also leaves get_work_gcwq() without
any user. Removed.
* In hotplug and freezer paths, the pools belonging to a CPU are often
processed together. This patch makes those paths hold locks of all
pools, with highpri lock nested inside, to keep the conversion
straight-forward. These nested lockings will be removed by
following patches.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Move gcwq->cpu to pool->cpu. This introduces a couple places where
gcwq->pools[0].cpu is used. These will soon go away as gcwq is
further reduced.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
There's no functional necessity for the two pools on the same CPU to
share the busy hash table. It's also likely to be a bottleneck when
implementing pools with user-specified attributes.
This patch makes busy_hash per-pool. The conversion is mostly
straight-forward. Changes worth noting are,
* Large block of changes in rebind_workers() is moving the block
inside for_each_worker_pool() as now there are separate hash tables
for each pool. This changes the order of operations but doesn't
break anything.
* Thre for_each_worker_pool() loops in gcwq_unbind_fn() are combined
into one. This again changes the order of operaitons but doesn't
break anything.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Currently, when a work item is off-queue, work->data records the CPU
it was last on, which is used to locate the last executing instance
for non-reentrance, flushing, etc.
We're in the process of removing global_cwq and making worker_pool the
top level abstraction. This patch makes work->data point to the pool
it was last associated with instead of CPU.
After the previous WORK_OFFQ_POOL_CPU and worker_poo->id additions,
the conversion is fairly straight-forward. WORK_OFFQ constants and
functions are modified to record and read back pool ID instead.
worker_pool_by_id() is added to allow looking up pool from ID.
get_work_pool() replaces get_work_gcwq(), which is reimplemented using
get_work_pool(). get_work_pool_id() replaces work_cpu().
This patch shouldn't introduce any observable behavior changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Add worker_pool->id which is allocated from worker_pool_idr. This
will be used to record the last associated worker_pool in work->data.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Currently, when a work item is off queue, high bits of its data
encodes the last CPU it was on. This is scheduled to be changed to
pool ID, which will make it impossible to use WORK_CPU_NONE to
indicate no association.
This patch limits the number of bits which are used for off-queue cpu
number to 31 (so that the max fits in an int) and uses the highest
possible value - WORK_OFFQ_CPU_NONE - to indicate no association.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Make GCWQ_FREEZING a pool flag POOL_FREEZING. This patch doesn't
change locking - FREEZING on both pools of a CPU are set or clear
together while holding gcwq->lock. It shouldn't cause any functional
difference.
This leaves gcwq->flags w/o any flags. Removed.
While at it, convert BUG_ON()s in freeze_workqueue_begin() and
thaw_workqueues() to WARN_ON_ONCE().
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Make GCWQ_DISASSOCIATED a pool flag POOL_DISASSOCIATED. This patch
doesn't change locking - DISASSOCIATED on both pools of a CPU are set
or clear together while holding gcwq->lock. It shouldn't cause any
functional difference.
This is part of an effort to remove global_cwq and make worker_pool
the top level abstraction, which in turn will help implementing worker
pools with user-specified attributes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
There are currently two worker pools per cpu (including the unbound
cpu) and they are the only pools in use. New class of pools are
scheduled to be added and some pool related APIs will be added
inbetween. Call the existing pools the standard pools and prefix them
with std_. Do this early so that new APIs can use std_ prefix from
the beginning.
This patch doesn't introduce any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
This function no longer has any external users. Unexport it. It will
be removed later on.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
init_task.cgroups is initialized at boot phase, and whenver a ask
is forked, it's cgroups pointer is inherited from its parent, and
it's never set to NULL afterwards.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The MSI specification has several constraints in comparison with
MSI-X, most notable of them is the inability to configure MSIs
independently. As a result, it is impossible to dispatch
interrupts from different queues to different CPUs. This is
largely devalues the support of multiple MSIs in SMP systems.
Also, a necessity to allocate a contiguous block of vector
numbers for devices capable of multiple MSIs might cause a
considerable pressure on x86 interrupt vector allocator and
could lead to fragmentation of the interrupt vectors space.
This patch overcomes both drawbacks in presense of IRQ remapping
and lets devices take advantage of multiple queues and per-IRQ
affinity assignments.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Jeff Garzik <jgarzik@pobox.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/c8bd86ff56b5fc118257436768aaa04489ac0a4c.1353324359.git.agordeev@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reschedule rq->curr if the first RT task has just been
pulled to the rq.
Signed-off-by: Kirill V Tkhai <tkhai@yandex.ru>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tkhai Kirill <tkhai@yandex.ru>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/118761353614535@web28f.yandex.ru
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The last remaining user was oprofile and its use has been
removed a while ago in commit bc078e4eab
("oprofile: convert oprofile from timer_hook to hrtimer").
There doesn't seem to be any upstream user of this hook
for about two years now. And I'm not even aware of any out of
tree user.
Let's remove it.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@kernel.org>
Cc: Avi Kivity <avi@redhat.com>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Geoff Levand <geoff@infradead.org>
Cc: Gilad Ben Yossef <gilad@benyossef.com>
Cc: Hakan Akkan <hakanakkan@gmail.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1356191991-2251-1-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The caller of sched_sliced() should pass se.cfs_rq and se as the
arguments, however in sched_rr_get_interval() we gave it
rq.cfs_rq and se, which made the following computation obviously
wrong.
The change was introduced by commit:
77034937dc sched: fix crash in sys_sched_rr_get_interval()
... 5 years ago, while it had been the correct 'cfs_rq_of' before
the commit. The change seems to be irrelevant to the commit
msg, which was to return a 0 timeslice for tasks that are on an
idle runqueue. So I believe that was just a plain typo.
Signed-off-by: Zhu Yanhai <gaoyang.zyh@taobao.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Turner <pjt@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1357621012-15039-1-git-send-email-gaoyang.zyh@taobao.com
[ Since this is an ABI and an old bug, we'll test this via a
slow upstream route, to hopefully discover any app breakage. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Dan's smatch found a compare bug with the result of the
trace_test_and_set_recursion() and comparing to less than
zero. If the function fails, it returns -1, but was saved in
an unsigned int, which will never be less than zero and will
ignore the result of the test if a recursion did happen.
Luckily this is the last of the recursion tests, as the
infrastructure of ftrace would catch recursions before it
got here, except for some few exceptions.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Pull tracing updates from Steve Rostedt.
This commit:
tracing: Remove the extra 4 bytes of padding in events
changes the ABI. All involved parties seem to agree that it's safe to
do now, but the devil is in the details ...
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
iQEcBAABAgAGBQJQ+MDOAAoJEHm+PkMAQRiGFL8H/39bXl3xs/ZkBct3ciICtaAt
UeRcdjvBdUrObQmmw8FEFJA62Az74g1UYyNyc0yVJoB6DwL2oawXhGXQXqDXMBMs
xJa/dpx41gmcrvsLOJ1v2Tmokk32hgdpLNWwzcn36xZMsrhYUxjxPhVNK4aDbwrW
3uzBj83dK1rvi2PEz3hgHizRhmdD1cNgLXKrczNbT6y6rLp6Jkng2q5UhNfaZUvE
A7Ub5/nW+X6gb1PctAjulz2CoySq0FbH4viUc+zLj3NN0L9ImeJvpREauE2QPVgJ
a8Sg3E6/cL4wIXBEV1E1KapVTd1XIEGa8yB5X6q8HWNKI8nf7Kwuj65VgvwJIMc=
=+VR8
-----END PGP SIGNATURE-----
Merge tag 'v3.8-rc4' into irq/core
Merge Linux 3.8-rc4 before pulling in new commits - we were on an old v3.7 base.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Global synchronization - async_synchronize_full() - is currently
implemented by keeping a list of all active registered domains and
syncing them one by one until no domain is active.
While this isn't necessarily a complex scheme, it can easily be
simplified by keeping global list of the pending items of all
registered active domains instead of list of domains and simply using
the globl pending list the same way as domain syncing.
This patch replaces async_domains with async_global_pending and update
lowest_in_progress() to use the global pending list if @domain is
%NULL. async_synchronize_full_domain(NULL) is now allowed and
equivalent to async_synchronize_full(). As no one is calling with
NULL domain, this doesn't affect any existing users.
async_register_mutex is no longer necessary and dropped.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Dan Williams <djbw@fb.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Async kept single global pending list and per-domain running lists.
When an async item is queued, it's put on the global pending list.
The item is moved to the per-domain running list when its execution
starts.
At this point, this design complicates execution and synchronization
without bringing any benefit. The list only matters for
synchronization which doesn't care whether a given async item is
pending or executing. Also, global synchronization is done by
iterating through all active registered async_domains, so the global
async_pending list doesn't help anything either.
Rename async_domain->running to async_domain->pending and put async
items directly there and remove when execution completes. This
simplifies lowest_in_progress() a lot - the first item on the pending
list is the one with the lowest cookie, and async_run_entry_fn()
doesn't have to mess with moving the item from pending to running.
After the change, whether a domain is empty or not can be trivially
determined by looking at async_domain->pending. Remove
async_domain->count and use list_empty() on pending instead.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Dan Williams <djbw@fb.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Currently, next_cookie is used as the infinity value. In most cases,
this should work fine but it theoretically could bring subtle behavior
difference between async_synchronize_full() and
async_synchronize_full_domain().
async_synchronize_full() keeps waiting until there's no registered
async_entry left regardless of what next_cookie was when the function
was called. It guarantees that the queue is completely drained at
least once before returning.
However, async_synchronize_full_domain() doesn't. It synchronizes
upto next_cookie and if further async jobs are queued after the
next_cookie value to synchronize is decided, they won't be waited for.
For unrelated async jobs, the behavior difference doesn't matter;
however, if async jobs which are related (nested or otherwise) to the
executing ones are queued while sychronization is in progress, the
resulting behavior difference could be problematic.
This can be easily fixed by using ULLONG_MAX as the infinity value
instead. Define ASYNC_COOKIE_MAX as ULLONG_MAX and use it as the
infinity value for synchronization. This makes
async_synchronize_full_domain() fully drain the domain at least once
before returning, making its behavior match async_synchronize_full().
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Dan Williams <djbw@fb.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
In the beginning, running lists were literal struct list_heads. Later
on, struct async_domain was added. For some reason, while the
conversion substituted list_heads with async_domains, the variable
names weren't fully converted. In more places, "running" was used for
struct async_domain while other places adopted new "domain" name.
The situation is made much worse by having async_domain's running list
named "domain" and async_entry's field pointing to async_domain named
"running".
So, we end up with mix of "running" and "domain" for variable names
for async_domain, with the field names of async_domain and async_entry
swapped between "running" and "domain".
It feels almost intentionally made to be as confusing as possible.
Bring some sanity by
* Renaming all async_domain variables "domain".
* s/async_running/async_dfl_domain/
* s/async_domain->domain/async_domain->running/
* s/async_entry->running/async_entry->domain/
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Dan Williams <djbw@fb.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
ring_buffer.c use to require declarations from trace.h, but
these have moved to the generic header files. There's nothing
in trace.h that ring_buffer.c requires.
There's some headers that trace.h included that ring_buffer.c
needs, but it's best that it includes them directly, and not
include trace.h.
Also, some things may use ring_buffer.c without having tracing
configured. This removes the dependency that may come in the
future.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Using context bit recursion checking, we can help increase the
performance of the ring buffer.
Before this patch:
# echo function > /debug/tracing/current_tracer
# for i in `seq 10`; do ./hackbench 50; done
Time: 10.285
Time: 10.407
Time: 10.243
Time: 10.372
Time: 10.380
Time: 10.198
Time: 10.272
Time: 10.354
Time: 10.248
Time: 10.253
(average: 10.3012)
Now we have:
# echo function > /debug/tracing/current_tracer
# for i in `seq 10`; do ./hackbench 50; done
Time: 9.712
Time: 9.824
Time: 9.861
Time: 9.827
Time: 9.962
Time: 9.905
Time: 9.886
Time: 10.088
Time: 9.861
Time: 9.834
(average: 9.876)
a 4% savings!
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The function tracer had two different versions of function tracing.
The disabling of irqs version and the preempt disable version.
As function tracing in very intrusive and can cause nasty recursion
issues, it has its own recursion protection. But the old method to
do this was a flat layer. If it detected that a recursion was happening
then it would just return without recording.
This made the preempt version (much faster than the irq disabling one)
not very useful, because if an interrupt were to occur after the
recursion flag was set, the interrupt would not be traced at all,
because every function that was traced would think it recursed on
itself (due to the context it preempted setting the recursive flag).
Now that we have a recursion flag for every context level, we
no longer need to worry about that. We can disable preemption,
set the current context recursion check bit, and go on. If an
interrupt were to come along, it would check its own context bit
and happily continue to trace.
As the preempt version is faster than the irq disable version,
there's no more reason to keep the preempt version around.
And the irq disable version still had an issue with missing
out on tracing NMI code.
Remove the irq disable function tracer version and have the
preempt disable version be the default (and only version).
Before this patch we had from running:
# echo function > /debug/tracing/current_tracer
# for i in `seq 10`; do ./hackbench 50; done
Time: 12.028
Time: 11.945
Time: 11.925
Time: 11.964
Time: 12.002
Time: 11.910
Time: 11.944
Time: 11.929
Time: 11.941
Time: 11.924
(average: 11.9512)
Now we have:
# echo function > /debug/tracing/current_tracer
# for i in `seq 10`; do ./hackbench 50; done
Time: 10.285
Time: 10.407
Time: 10.243
Time: 10.372
Time: 10.380
Time: 10.198
Time: 10.272
Time: 10.354
Time: 10.248
Time: 10.253
(average: 10.3012)
a 13.8% savings!
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When function tracing occurs, the following steps are made:
If arch does not support a ftrace feature:
call internal function (uses INTERNAL bits) which calls...
If callback is registered to the "global" list, the list
function is called and recursion checks the GLOBAL bits.
then this function calls...
The function callback, which can use the FTRACE bits to
check for recursion.
Now if the arch does not suppport a feature, and it calls
the global list function which calls the ftrace callback
all three of these steps will do a recursion protection.
There's no reason to do one if the previous caller already
did. The recursion that we are protecting against will
go through the same steps again.
To prevent the multiple recursion checks, if a recursion
bit is set that is higher than the MAX bit of the current
check, then we know that the check was made by the previous
caller, and we can skip the current check.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Currently for recursion checking in the function tracer, ftrace
tests a task_struct bit to determine if the function tracer had
recursed or not. If it has, then it will will return without going
further.
But this leads to races. If an interrupt came in after the bit
was set, the functions being traced would see that bit set and
think that the function tracer recursed on itself, and would return.
Instead add a bit for each context (normal, softirq, irq and nmi).
A check of which context the task is in is made before testing the
associated bit. Now if an interrupt preempts the function tracer
after the previous context has been set, the interrupt functions
can still be traced.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
There is lots of places that perform:
op = rcu_dereference_raw(ftrace_control_list);
while (op != &ftrace_list_end) {
Add a helper macro to do this, and also optimize for a single
entity. That is, gcc will optimize a loop for either no iterations
or more than one iteration. But usually only a single callback
is registered to the function tracer, thus the optimized case
should be a single pass. to do this we now do:
op = rcu_dereference_raw(list);
do {
[...]
} while (likely(op = rcu_dereference_raw((op)->next)) &&
unlikely((op) != &ftrace_list_end));
An op is always registered (ftrace_list_end when no callbacks is
registered), thus when a single callback is registered, the link
list looks like:
top => callback => ftrace_list_end => NULL.
The likely(op = op->next) still must be performed due to the race
of removing the callback, where the first op assignment could
equal ftrace_list_end. In that case, the op->next would be NULL.
But this is unlikely (only happens in a race condition when
removing the callback).
But it is very likely that the next op would be ftrace_list_end,
unless more than one callback has been registered. This tells
gcc what the most common case is and makes the fast path with
the least amount of branches.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The function tracing recursion self test should not crash
the machine if the resursion test fails. If it detects that
the function tracing is recursing when it should not be, then
bail, don't go into an infinite recursive loop.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If one of the function tracers set by the global ops is not recursion
safe, it can still be called directly without the added recursion
supplied by the ftrace infrastructure.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The test that checks function recursion does things differently
if the arch does not support all ftrace features. But that really
doesn't make a difference with how the test runs, and either way
the count variable should be 2 at the end.
Currently the test wrongly fails for archs that don't support all
the ftrace features.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
There's a race condition between the setting of a new tracer and
the update of the max trace buffers (the swap). When a new tracer
is added, it sets current_trace to nop_trace before disabling
the old tracer. At this moment, if the old tracer uses update_max_tr(),
the update may trigger the warning against !current_trace->use_max-tr,
as nop_trace doesn't have that set.
As update_max_tr() requires that interrupts be disabled, we can
add a check to see if current_trace == nop_trace and bail if it
does. Then when disabling the current_trace, set it to nop_trace
and run synchronize_sched(). This will make sure all calls to
update_max_tr() have completed (it was called with interrupts disabled).
As a clean up, this commit also removes shrinking and recreating
the max_tr buffer if the old and new tracers both have use_max_tr set.
The old way use to always shrink the buffer, and then expand it
for the next tracer. This is a waste of time.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Synchronous requet_module() from an async worker can lead to deadlock
because module init path may invoke async_synchronize_full(). The
async worker waits for request_module() to complete and the module
loading waits for the async task to finish. This bug happened in the
block layer because of default elevator auto-loading.
Block layer has been updated not to do default elevator auto-loading
and it has been decided to disallow synchronous request_module() from
async workers.
Trigger WARN_ON_ONCE() on synchronous request_module() from async
workers.
For more details, please refer to the following thread.
http://thread.gmane.org/gmane.linux.kernel/1420814
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Alex Riesen <raa.lkml@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
If cgroup_create() failed and cgroup_destroy_locked() is called to
do cleanup, we'll see a bunch of warnings:
cgroup_addrm_files: failed to remove 2MB.limit_in_bytes, err=-2
cgroup_addrm_files: failed to remove 2MB.usage_in_bytes, err=-2
cgroup_addrm_files: failed to remove 2MB.max_usage_in_bytes, err=-2
cgroup_addrm_files: failed to remove 2MB.failcnt, err=-2
cgroup_addrm_files: failed to remove prioidx, err=-2
cgroup_addrm_files: failed to remove ifpriomap, err=-2
...
We failed to remove those files, because cgroup_create() has failed
before creating those cgroup files.
To fix this, we simply don't warn if cgroup_rm_file() can't find the
cft entry.
Signed-off-by: Li Zefan <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Commit 083b804c4d ("async: use workqueue for worker pool") made it
possible that async jobs are moved from pending to running out-of-order.
While pending async jobs will be queued and dispatched for execution in
the same order, nothing guarantees they'll enter "1) move self to the
running queue" of async_run_entry_fn() in the same order.
Before the conversion, async implemented its own worker pool. An async
worker, upon being woken up, fetches the first item from the pending
list, which kept the executing lists sorted. The conversion to
workqueue was done by adding work_struct to each async_entry and async
just schedules the work item. The queueing and dispatching of such work
items are still in order but now each worker thread is associated with a
specific async_entry and moves that specific async_entry to the
executing list. So, depending on which worker reaches that point
earlier, which is non-deterministic, we may end up moving an async_entry
with larger cookie before one with smaller one.
This broke __lowest_in_progress(). running->domain may not be properly
sorted and is not guaranteed to contain lower cookies than pending list
when not empty. Fix it by ensuring sort-inserting to the running list
and always looking at both pending and running when trying to determine
the lowest cookie.
Over time, the async synchronization implementation became quite messy.
We better restructure it such that each async_entry is linked to two
lists - one global and one per domain - and not move it when execution
starts. There's no reason to distinguish pending and running. They
behave the same for synchronization purposes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
is placed on a function mcount/nop location, and the arch supports it,
instead of adding a breakpoint, kprobes will register a function callback
as that is much more efficient.
The function tracer requires to update modules before they run, and
uses the module notifier to do so. But if something else in the module
notifiers registers a kprobe at one of these locations, before ftrace
can get to it, then the system could fail.
The function tracer must be initialized early, otherwise module notifiers
that probe will only work by chance.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
iQEcBAABAgAGBQJQ/ZaaAAoJEOdOSU1xswtMQsoH/02S5ngidMgIXwYCGzwxR6Ym
f2krSz76rBkFn9ivFPJFfq2Wi05UhhVfFr4gpKzoK+/rBqwVkV5tRIQ4AhJ1/Q3d
wCXB2mcRI6ky/kB/ts8q6gjj+rlJ18NgZf79TA5y5q7FEYc6DvB2dZ+vE+rvCnXk
q/Bx6q4phEdLbVqet+Ga36qzi9u+pW0P+ntZmi0EQLVnz9p+mtdVaRz32qKp0FYi
XhHtPMHQDZoOu8utPNtPcfSOZp1sNN8tXqjEAJ4/Ba54badUfg4WJ+RXGPsYH+4T
lOwpVv7Xp0Sp2b1HivEVfCs1bx2RNzluZisPahBEwdBv+XssvqMbAfG+X3EVk3w=
=6EYI
-----END PGP SIGNATURE-----
Merge tag 'trace-3.8-rc4-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull ftrace fix from Steven Rostedt:
"Kprobes now uses the function tracer if it can. That is, if a probe
is placed on a function mcount/nop location, and the arch supports it,
instead of adding a breakpoint, kprobes will register a function
callback as that is much more efficient.
The function tracer requires to update modules before they run, and
uses the module notifier to do so. But if something else in the
module notifiers registers a kprobe at one of these locations, before
ftrace can get to it, then the system could fail.
The function tracer must be initialized early, otherwise module
notifiers that probe will only work by chance."
* tag 'trace-3.8-rc4-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Be first to run code modification on modules
wake_up_process() should never wakeup a TASK_STOPPED/TRACED task.
Change it to use TASK_NORMAL and add the WARN_ON().
TASK_ALL has no other users, probably can be killed.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
putreg() assumes that the tracee is not running and pt_regs_access() can
safely play with its stack. However a killed tracee can return from
ptrace_stop() to the low-level asm code and do RESTORE_REST, this means
that debugger can actually read/modify the kernel stack until the tracee
does SAVE_REST again.
set_task_blockstep() can race with SIGKILL too and in some sense this
race is even worse, the very fact the tracee can be woken up breaks the
logic.
As Linus suggested we can clear TASK_WAKEKILL around the arch_ptrace()
call, this ensures that nobody can ever wakeup the tracee while the
debugger looks at it. Not only this fixes the mentioned problems, we
can do some cleanups/simplifications in arch_ptrace() paths.
Probably ptrace_unfreeze_traced() needs more callers, for example it
makes sense to make the tracee killable for oom-killer before
access_process_vm().
While at it, add the comment into may_ptrace_stop() to explain why
ptrace_stop() still can't rely on SIGKILL and signal_pending_state().
Reported-by: Salman Qazi <sqazi@google.com>
Reported-by: Suleiman Souhlal <suleiman@google.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As trace_clock is used by other things besides tracing, and it
does not require anything from trace.h, it is best not to include
the header file in trace_clock.c.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Cleanup and preparation for the next change.
signal_wake_up(resume => true) is overused. None of ptrace/jctl callers
actually want to wakeup a TASK_WAKEKILL task, but they can't specify the
necessary mask.
Turn signal_wake_up() into signal_wake_up_state(state), reintroduce
signal_wake_up() as a trivial helper, and add ptrace_signal_wake_up()
which adds __TASK_TRACED.
This way ptrace_signal_wake_up() can work "inside" ptrace_request()
even if the tracee doesn't have the TASK_WAKEKILL bit set.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Due to a userspace issue with PowerTop v2beta, which hardcoded
the offset of event fields that it was using, it broke when
we removed the Big Kernel Lock counter from the event header.
(commit e6e1e2593 "tracing: Remove lock_depth from event entry")
Because this broke userspace, it was determined that we must
keep those 4 bytes around.
(commit a3a4a5acd "Regression: partial revert "tracing: Remove lock_depth from event entry"")
This unfortunately wastes space in the ring buffer. 4 bytes per
event, where a lot of events are just 24 bytes. That's 16% of the
buffer wasted. A million events will add 4 megs of white space
into the buffer.
It was later noticed that PowerTop v2beta could not work on systems
where the kernel was 64 bit but the userspace was 32 bits.
The reason was because the offsets are different between the
two and the hard coded offset of one would not work with the other.
With PowerTop v2 final, it implemented the same interface that both
perf and trace-cmd use. That is, it reads the format file of
the event to find the offsets of the fields it needs. This fixes
the problem with running powertop on a 32 bit userspace running
on a 64 bit kernel. It also no longer requires the 4 byte padding.
As PowerTop v2 has been out for a while, and is included in all
major distributions, it is time that we can safely remove the
4 bytes of padding. Users of PowerTop v2beta should upgrade to
PowerTop v2 final.
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
