Paul E. McKenney [Tue, 16 Sep 2014 17:10:44 +0000 (10:10 -0700)]
Merge branch 'rcu-tasks.2014.09.10a' into HEAD
rcu-tasks.2014.09.10a: Add RCU-tasks flavor of RCU.
Paul E. McKenney [Tue, 16 Sep 2014 17:08:34 +0000 (10:08 -0700)]
Merge branches 'doc.2014.09.07a', 'fixes.2014.09.10a', 'nocb-nohz.2014.09.16b' and 'torture.2014.09.07a' into HEAD
doc.2014.09.07a: Documentation updates.
fixes.2014.09.10a: Miscellaneous fixes.
nocb-nohz.2014.09.16b: No-CBs CPUs and NO_HZ_FULL updates.
torture.2014.09.07a: Torture-test updates.
Paul E. McKenney [Tue, 12 Aug 2014 20:54:21 +0000 (13:54 -0700)]
rcu: Avoid misordering in nocb_leader_wait()
The NOCB follower wakeup ordering depends on the store to the tail
pointer happening before the wakeup. However, because atomic_long_add()
does not return a value, it does not provide ordering guarantees, and
the locking in wake_up() only guarantees that the store will happen
before the unlock, which might be too late. Even though this is only a
theoretical issue, this commit adds a smp_mb__after_atomic() after the
final atomic_long_add() to provide the needed ordering guarantee.
Reported-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Tue, 12 Aug 2014 18:27:31 +0000 (11:27 -0700)]
rcu: Handle NOCB callbacks from irq-disabled idle code
If an RCU callback is queued on a no-CBs CPU from idle code with irqs
disabled, and if that CPU stays idle forever after, the callback will
never be invoked. This commit therefore adds a check for this situation
in ____call_rcu_nocb(), invoking the RCU core solely for the purpose
of the ensuing return-to-idle transition. (If the CPU doesn't return
to idle, the next scheduling-clock interrupt will fix things up.)
Reported-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Tue, 12 Aug 2014 17:47:48 +0000 (10:47 -0700)]
rcu: Avoid misordering in __call_rcu_nocb_enqueue()
The NOCB leader wakeup ordering depends on the store to the header
happening before the check for the leader already being awake. However,
because atomic_long_add() does not return a value, it does not provide
ordering guarantees, the incorrect comment in wake_nocb_leader()
notwithstanding. This commit therefore adds a smp_mb__after_atomic()
after the final atomic_long_add() to provide the needed ordering
guarantee.
Reported-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Mon, 21 Jul 2014 18:34:33 +0000 (11:34 -0700)]
rcu: Don't track sysidle state if no nohz_full= CPUs
If there are no nohz_full= CPUs, then there is currently no reason to
track sysidle state. This commit therefore short-circuits this state
tracking if !tick_nohz_full_enabled().
Note that these checks will need to be revisited if nohz_full= state
can ever be changed at runtime.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Mon, 21 Jul 2014 18:26:54 +0000 (11:26 -0700)]
rcu: Eliminate redundant rcu_sysidle_state variable
Now that we have rcu_state_p, which references rcu_preempt_state for
TREE_PREEMPT_RCU and rcu_sched_state for TREE_RCU, we don't need a
separate rcu_sysidle_state variable. This commit therefore eliminates
rcu_preempt_state in favor of rcu_state_p.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Pranith Kumar [Fri, 18 Jul 2014 00:11:01 +0000 (20:11 -0400)]
rcu: Check for have_rcu_nocb_mask instead of rcu_nocb_mask
If we configure a kernel with CONFIG_NOCB_CPU=y, CONFIG_RCU_NOCB_CPU_NONE=y and
CONFIG_CPUMASK_OFFSTACK=n and do not pass in a rcu_nocb= boot parameter, the
cpumask rcu_nocb_mask can be garbage instead of NULL.
Hence this commit replaces checks for rcu_nocb_mask == NULL with a check for
have_rcu_nocb_mask.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Fri, 11 Jul 2014 18:30:24 +0000 (11:30 -0700)]
rcu: Create rcuo kthreads only for onlined CPUs
RCU currently uses for_each_possible_cpu() to spawn rcuo kthreads,
which can result in more rcuo kthreads than one would expect, for
example, derRichard reported 64 CPUs worth of rcuo kthreads on an
8-CPU image. This commit therefore creates rcuo kthreads only for
those CPUs that actually come online.
This was reported by derRichard on the OFTC IRC network.
Reported-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Sun, 13 Jul 2014 19:00:53 +0000 (12:00 -0700)]
rcu: Rationalize kthread spawning
Currently, RCU spawns kthreads from several different early_initcall()
functions. Although this has served RCU well for quite some time,
as more kthreads are added a more deterministic approach is required.
This commit therefore causes all of RCU's early-boot kthreads to be
spawned from a single early_initcall() function.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:17 +0000 (18:26 -0400)]
rcu: Return false instead of 0 in rcu_nocb_adopt_orphan_cbs()
Return false instead of 0 in rcu_nocb_adopt_orphan_cbs() as this has
bool as return type.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:16 +0000 (18:26 -0400)]
rcu: Use false for return in __call_rcu_nocb()
Return false instead of 0 in __call_rcu_nocb() as this has bool as
return type.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:15 +0000 (18:26 -0400)]
rcu: Use true/false for return in rcu_nocb_adopt_orphan_cbs()
Return true/false in rcu_nocb_adopt_orphan_cbs() instead of 0/1 as
this function has return type of bool.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:14 +0000 (18:26 -0400)]
rcu: Use true/false for return in __call_rcu_nocb()
Return true/false instead of 0/1 in __call_rcu_nocb() as this returns a
bool type.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Pranith Kumar [Fri, 25 Jul 2014 23:02:07 +0000 (16:02 -0700)]
rcu: Check the return value of zalloc_cpumask_var()
This commit checks the return value of the zalloc_cpumask_var() used for
allocating cpumask for rcu_nocb_mask.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Fri, 25 Jul 2014 18:21:47 +0000 (11:21 -0700)]
rcu: Fix attempt to avoid unsolicited offloading of callbacks
Commit
b58cc46c5f6b (rcu: Don't offload callbacks unless specifically
requested) failed to adjust the callback lists of the CPUs that are
known to be no-CBs CPUs only because they are also nohz_full= CPUs.
This failure can result in callbacks that are posted during early boot
getting stranded on nxtlist for CPUs whose no-CBs property becomes
apparent late, and there can also be spurious warnings about offline
CPUs posting callbacks.
This commit fixes these problems by adding an early-boot rcu_init_nohz()
that properly initializes the no-CBs CPUs.
Note that kernels built with CONFIG_RCU_NOCB_CPU_ALL=y or with
CONFIG_RCU_NOCB_CPU=n do not exhibit this bug. Neither do kernels
booted without the nohz_full= boot parameter.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Tested-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Paul E. McKenney [Wed, 10 Sep 2014 21:48:05 +0000 (14:48 -0700)]
rcutorture: Add RCU-tasks tests to default rcutorture list
Although the test cases have been added, they must be specified explicitly
via the kvm.sh --configs argument in order to run them. This commit
therefore adds the RCU-tasks tests to the CFLIST so that they will be
run automatically by default.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Thu, 14 Aug 2014 23:38:46 +0000 (16:38 -0700)]
rcu: Per-CPU operation cleanups to rcu_*_qs() functions
The rcu_bh_qs(), rcu_preempt_qs(), and rcu_sched_qs() functions use
old-style per-CPU variable access and write to ->passed_quiesce even
if it is already set. This commit therefore updates to use the new-style
per-CPU variable access functions and avoids the spurious writes.
This commit also eliminates the "cpu" argument to these functions because
they are always invoked on the indicated CPU.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Thu, 14 Aug 2014 23:01:53 +0000 (16:01 -0700)]
rcu: Remove local_irq_disable() in rcu_preempt_note_context_switch()
The rcu_preempt_note_context_switch() function is on a scheduling fast
path, so it would be good to avoid disabling irqs. The reason that irqs
are disabled is to synchronize process-level and irq-handler access to
the task_struct ->rcu_read_unlock_special bitmask. This commit therefore
makes ->rcu_read_unlock_special instead be a union of bools with a short
allowing single-access checks in RCU's __rcu_read_unlock(). This results
in the process-level and irq-handler accesses being simple loads and
stores, so that irqs need no longer be disabled. This commit therefore
removes the irq disabling from rcu_preempt_note_context_switch().
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Mon, 11 Aug 2014 02:47:12 +0000 (19:47 -0700)]
rcu: Additional information on RCU-tasks stall-warning messages
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 5 Aug 2014 12:23:35 +0000 (05:23 -0700)]
rcu: Remove redundant preempt_disable() from rcu_note_voluntary_context_switch()
In theory, synchronize_sched() requires a read-side critical section
to order against. In practice, preemption can be thought of as
being disabled across every machine instruction, at least for those
machine instructions that are not in the idle loop and not on offline
CPUs. So this commit removes the redundant preempt_disable() from
rcu_note_voluntary_context_switch().
Please note that the single instruction in question is the store of
zero to ->rcu_tasks_holdout. The "if" is simply a performance optimization
that avoids unnecessary stores. To see this, keep in mind that both
the "if" condition and the store are in a quiescent state. Therefore,
even if the task is preempted for a full grace period (presumably due
to its having done a context switch beforehand), the store will be
recording a legitimate quiescent state.
Reported-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Conflicts:
include/linux/rcupdate.h
Paul E. McKenney [Tue, 5 Aug 2014 12:10:24 +0000 (05:10 -0700)]
rcu: Make rcu_tasks_kthread()'s GP-wait loop allow preemption
The grace-period-wait loop in rcu_tasks_kthread() is under (unnecessary)
RCU protection, and therefore has no preemption points in a PREEMPT=n
kernel. This commit therefore removes the RCU protection and inserts
cond_resched().
Reported-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 5 Aug 2014 00:43:50 +0000 (17:43 -0700)]
rcu: Make TASKS_RCU handle nohz_full= CPUs
Currently TASKS_RCU would ignore a CPU running a task in nohz_full=
usermode execution. There would be neither a context switch nor a
scheduling-clock interrupt to tell TASKS_RCU that the task in question
had passed through a quiescent state. The grace period would therefore
extend indefinitely. This commit therefore makes RCU's dyntick-idle
subsystem record the task_struct structure of the task that is running
in dyntick-idle mode on each CPU. The TASKS_RCU grace period can
then access this information and record a quiescent state on
behalf of any CPU running in dyntick-idle usermode.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Mon, 4 Aug 2014 14:24:21 +0000 (07:24 -0700)]
rcu: Defer rcu_tasks_kthread() creation till first call_rcu_tasks()
It is expected that many sites will have CONFIG_TASKS_RCU=y, but
will never actually invoke call_rcu_tasks(). For such sites, creating
rcu_tasks_kthread() at boot is wasteful. This commit therefore defers
creation of this kthread until the time of the first call_rcu_tasks().
This of course means that the first call_rcu_tasks() must be invoked
from process context after the scheduler is fully operational.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 29 Jul 2014 16:49:23 +0000 (09:49 -0700)]
documentation: Add verbiage on RCU-tasks stall warning messages
This commit documents RCU-tasks stall warning messages and also describes
when to use the new cond_resched_rcu_qs() API.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Mon, 28 Jul 2014 21:39:25 +0000 (14:39 -0700)]
rcu: Improve RCU-tasks energy efficiency
The current RCU-tasks implementation uses strict polling to detect
callback arrivals. This works quite well, but is not so good for
energy efficiency. This commit therefore replaces the strict polling
with a wait queue.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Wed, 2 Jul 2014 01:16:30 +0000 (18:16 -0700)]
rcu: Add stall-warning checks for RCU-tasks
This commit adds a ten-minute RCU-tasks stall warning. The actual
time is controlled by the boot/sysfs parameter rcu_task_stall_timeout,
with values less than or equal to zero disabling the stall warnings.
The default value is ten minutes, which means that the tasks that have
not yet responded will get their stacks dumped every ten minutes, until
they pass through a voluntary context switch.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 1 Jul 2014 19:56:20 +0000 (12:56 -0700)]
rcutorture: Add RCU-tasks test cases
This commit adds the TASKS01 and TASKS02 Kconfig fragments, along with
the corresponding TASKS01.boot and TASKS02.boot boot-parameter files
specifying that rcutorture test RCU-tasks instead of the default flavor.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 1 Jul 2014 18:59:36 +0000 (11:59 -0700)]
rcutorture: Add torture tests for RCU-tasks
This commit adds torture tests for RCU-tasks. It also fixes a bug that
would segfault for an RCU flavor lacking a callback-barrier function.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Steven Rostedt [Thu, 3 Jul 2014 01:17:19 +0000 (18:17 -0700)]
rcu: Export RCU-tasks APIs to GPL modules
This commit exports the RCU-tasks synchronous APIs,
synchronize_rcu_tasks() and rcu_barrier_tasks(), to
GPL-licensed kernel modules.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Paul E. McKenney [Mon, 4 Aug 2014 13:10:23 +0000 (06:10 -0700)]
rcu: Make TASKS_RCU handle tasks that are almost done exiting
Once a task has passed exit_notify() in the do_exit() code path, it
is no longer on the task lists, and is therefore no longer visible
to rcu_tasks_kthread(). This means that an almost-exited task might
be preempted while within a trampoline, and this task won't be waited
on by rcu_tasks_kthread(). This commit fixes this bug by adding an
srcu_struct. An exiting task does srcu_read_lock() just before calling
exit_notify(), and does the corresponding srcu_read_unlock() after
doing the final preempt_disable(). This means that rcu_tasks_kthread()
can do synchronize_srcu() to wait for all mostly-exited tasks to reach
their final preempt_disable() region, and then use synchronize_sched()
to wait for those tasks to finish exiting.
Reported-by: Oleg Nesterov <oleg@redhat.com>
Suggested-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 1 Jul 2014 19:22:23 +0000 (12:22 -0700)]
rcu: Add synchronous grace-period waiting for RCU-tasks
It turns out to be easier to add the synchronous grace-period waiting
functions to RCU-tasks than to work around their absense in rcutorture,
so this commit adds them. The key point is that the existence of
call_rcu_tasks() means that rcutorture needs an rcu_barrier_tasks().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 1 Jul 2014 18:26:57 +0000 (11:26 -0700)]
rcu: Provide cond_resched_rcu_qs() to force quiescent states in long loops
RCU-tasks requires the occasional voluntary context switch
from CPU-bound in-kernel tasks. In some cases, this requires
instrumenting cond_resched(). However, there is some reluctance
to countenance unconditionally instrumenting cond_resched() (see
http://lwn.net/Articles/603252/), so this commit creates a separate
cond_resched_rcu_qs() that may be used in place of cond_resched() in
locations prone to long-duration in-kernel looping.
This commit currently instruments only RCU-tasks. Future possibilities
include also instrumenting RCU, RCU-bh, and RCU-sched in order to reduce
IPI usage.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Fri, 27 Jun 2014 20:42:20 +0000 (13:42 -0700)]
rcu: Add call_rcu_tasks()
This commit adds a new RCU-tasks flavor of RCU, which provides
call_rcu_tasks(). This RCU flavor's quiescent states are voluntary
context switch (not preemption!) and userspace execution (not the idle
loop -- use some sort of schedule_on_each_cpu() if you need to handle the
idle tasks. Note that unlike other RCU flavors, these quiescent states
occur in tasks, not necessarily CPUs. Includes fixes from Steven Rostedt.
This RCU flavor is assumed to have very infrequent latency-tolerant
updaters. This assumption permits significant simplifications, including
a single global callback list protected by a single global lock, along
with a single task-private linked list containing all tasks that have not
yet passed through a quiescent state. If experience shows this assumption
to be incorrect, the required additional complexity will be added.
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Tue, 26 Aug 2014 18:35:43 +0000 (11:35 -0700)]
torture: Check for nul bytes in console output
When starting a new torture run while an old one is still running, both
qemu processes can be outputting to the same console.out file. This can
cause quite a bit of confusion, so this commit checks for this situation,
which is normally indicated by nul bytes in the console output. Yes,
if your new run uses up an exact number of blocks of the file, this
check will be ineffective, but the odds are not bad.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Paul E. McKenney [Tue, 26 Aug 2014 03:41:47 +0000 (20:41 -0700)]
torture: Print PID in hung-kernel diagnostic message
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Paul E. McKenney [Tue, 19 Aug 2014 04:12:17 +0000 (21:12 -0700)]
rcutorture: Add callback-flood test
Although RCU is designed to handle arbitrary floods of callbacks, this
capability is not routinely tested. This commit therefore adds a
cbflood capability in which kthreads repeatedly registers large numbers
of callbacks. One such kthread is created for each four CPUs (rounding
up), and the test may be controlled by several cbflood_* kernel boot
parameters, which control the number of bursts per flood, the number
of callbacks per burst, the time between bursts, and the time between
floods. The default values are large enough to exercise RCU's emergency
responses to callback flooding.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: David Miller <davem@davemloft.net>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Joe Perches [Mon, 14 Jul 2014 13:16:15 +0000 (09:16 -0400)]
rcu: Use pr_alert/pr_cont for printing logs
User pr_alert/pr_cont for printing the logs from rcutorture module directly
instead of writing it to a buffer and then printing it. This allows us from not
having to allocate such buffers. Also remove a resulting empty function.
I tested this using the parse-torture.sh script as follows:
$ dmesg | grep torture > log.txt
$ bash parse-torture.sh log.txt test
$
There were no warnings which means that parsing went fine.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Sun, 27 Jul 2014 04:38:09 +0000 (21:38 -0700)]
rcutorture: Specify CONFIG_CPUMASK_OFFSTACK=y for TREE07
This commit specifies offstack cpumasks in TREE07 in order to catch
references to unallocated cpumask_var_t variables.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Paul E. McKenney [Mon, 21 Jul 2014 20:35:10 +0000 (13:35 -0700)]
rcutorture: Specify MAXSMP=y for TREE01
Setting CONFIG_MAXSMP=y causes cpumasks to be moved offstack, which
introduces the possibility of NULL cpumask_var_t pointers. This commit
therefore enables CONFIG_MAXSMP=y in TREE01 to increase test coverage.
However, because CONFIG_MAXSMP=y implies 8192 CPUs, we need to use
the maxcpus= boot parameter to limit the number of CPUs to something
reasonable, which in turn requires updating the scripts to handle this.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Paul E. McKenney [Mon, 21 Jul 2014 20:13:35 +0000 (13:13 -0700)]
rcutorture: Test partial nohz_full= configuration
The current set of tests covers only cases where either all possible CPUs
are nohz_full= CPUs or none of them are. Because there have been some
recent bug escapes in cases where only some of the CPUs are nohz_full=
CPUs, this commit add a configuration where only half of the CPUs are
nohz_full= CPUs.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Paul E. McKenney [Fri, 18 Jul 2014 19:01:39 +0000 (12:01 -0700)]
rcu: Add step to initrd documentation
This commit tries to get people into the correct directory before
creating the initrd directory.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Pranith Kumar [Fri, 11 Jul 2014 23:47:35 +0000 (19:47 -0400)]
rcutorture: Set executable bit and drop bash from Usage
This commit sets the executable bit on test scripts config2frag.sh
and kvm.sh. Since #!/bin/bash is set in all the scripts, this commit
also drops it from all usage lines because the scripts can now all be
invoked directly.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Fri, 11 Jul 2014 21:31:27 +0000 (17:31 -0400)]
rcutorture: Use bash shell for all the test scripts
Some of the scripts encode a default /bin/sh shell. On systems which use
dash as default shell, these scripts fail as they are bash scripts. I
encountered this while testing the sprintf() changes on a Debian system
where dash is the default shell.
This commit changes all such uses to use bash explicitly.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Wed, 11 Jun 2014 20:39:43 +0000 (16:39 -0400)]
rcutorture: Fix a sparse warning by marking boost_mutex static
This commit fixes the following sparse warning by marking boost_mutex
static:
kernel/rcu/rcutorture.c:185:1: warning: symbol 'boost_mutex' was not declared. Should it be static?
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Paul E. McKenney [Thu, 14 Aug 2014 17:28:23 +0000 (10:28 -0700)]
rcu: Replace flush_signals() with WARN_ON(signal_pending())
Currently, when RCU awakens from a wait_event_interruptible() that
might have awakened prematurely, it does a flush_signals(). This is
done on the off-chance that someone figured out how to deliver a signal
to a kthread, which is supposed to be impossible. Given that this
is supposed to be impossible, this commit changes the flush_signals()
calls into WARN_ON(signal_pending()).
Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 12 Aug 2014 17:07:47 +0000 (13:07 -0400)]
rcu: Use rcu_gp_kthread_wake() to wake up grace period kthreads
The rcu_gp_kthread_wake() function checks for three conditions before
waking up grace period kthreads:
* Is the thread we are trying to wake up the current thread?
* Are the gp_flags zero? (all threads wait on non-zero gp_flags condition)
* Is there no thread created for this flavour, hence nothing to wake up?
If any one of these condition is true, we do not call wake_up().
It was found that there are quite a few avoidable wake ups both during
idle time and under stress induced by rcutorture.
Idle:
Total:66000, unnecessary:66000, case1:61827, case2:66000, case3:0
Total:68000, unnecessary:68000, case1:63696, case2:68000, case3:0
rcutorture:
Total:254000, unnecessary:254000, case1:199913, case2:254000, case3:0
Total:256000, unnecessary:256000, case1:201784, case2:256000, case3:0
Here case{1-3} are the cases listed above. We can avoid these wake
ups by using rcu_gp_kthread_wake() to conditionally wake up the grace
period kthreads.
There is a comment about an implied barrier supplied by the wake_up()
logic. This barrier is necessary for the awakened thread to see the
updated ->gp_flags. This flag is always being updated with the root node
lock held. Also, the awakened thread tries to acquire the root node lock
before reading ->gp_flags because of which there is proper ordering.
Hence this commit tries to avoid calling wake_up() whenever we can by
using rcu_gp_kthread_wake() function.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
CC: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Thu, 31 Jul 2014 23:02:33 +0000 (16:02 -0700)]
rcu: Make TINY_RCU tinier by putting error checks under #ifdef
The rcu_idle_enter_common() and rcu_idle_exit_common() functions contain
error checks that have to the best of my knowledge have never triggered
over the past several years. These are nevertheless valuable when
creating new architectures or doing other low-level changes, so the
checks should not be deleted. This commit instead places these checks
under #ifdef CONFIG_RCU_TRACE so that they are executed only when
specifically requested.
The savings are significant:
Before:
text data bss dec hex filename
1749 39 0 1788 6fc /tmp/b/kernel/rcu/tiny.o
632 152 0 784 310 /tmp/b/kernel/rcu/update.o
----
2572
After:
text data bss dec hex filename
1281 37 0 1318 526 /tmp/b/kernel/rcu/tiny.o
632 152 0 784 310 /tmp/b/kernel/rcu/update.o
----
2102
This amounts to 470 bytes, or 18% of the original.
Switched from #ifdef to IS_ENABLED() on Josh Triplett's advice.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Paul E. McKenney [Tue, 29 Jul 2014 21:50:47 +0000 (14:50 -0700)]
rcu: Break more call_rcu() deadlock involving scheduler and perf
Commit
96d3fd0d315a9 (rcu: Break call_rcu() deadlock involving scheduler
and perf) covered the case where __call_rcu_nocb_enqueue() needs to wake
the rcuo kthread due to the queue being initially empty, but did not
do anything for the case where the queue was overflowing. This commit
therefore also defers wakeup for the overflow case.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Thu, 17 Jul 2014 02:20:33 +0000 (22:20 -0400)]
rcu: Remove stale comment in tree.c
This commit removes a stale comment in rcu/tree.c which was left
out when some code was moved around previously in commit
2036d94a7b61
("rcu: Rework detection of use of RCU by offline CPUs") For reference,
the following updated comment exists a few lines below this which means
the same:
/* Remove the outgoing CPU from the masks in the rcu_node hierarchy. */
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 15 Jul 2014 22:31:47 +0000 (18:31 -0400)]
rcu: Update tiny.c references to tree.c
This commit updates the references to rcutree.c which is now rcu/tree.c
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Ard Biesheuvel [Sat, 12 Jul 2014 17:01:49 +0000 (19:01 +0200)]
rcu: Define tracepoint strings only if CONFIG_TRACING is set
Commit
f7f7bac9cb1c ("rcu: Have the RCU tracepoints use the tracepoint_string
infrastructure") unconditionally populates the __tracepoint_str input section,
but this section is not assigned an output section if CONFIG_TRACING is not set.
This results in the __tracepoint_str turning up in unexpected places, i.e.,
after _edata.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Oleg Nesterov [Tue, 8 Jul 2014 22:17:59 +0000 (15:17 -0700)]
rcu: Uninline rcu_read_lock_held()
This commit uninlines rcu_read_lock_held(). According to "size vmlinux"
this saves 28549 in .text:
-
5541731 3014560 14757888 23314179
+
5513182 3026848 14757888 23297918
Note: it looks as if the data grows by 12288 bytes but this is not true,
it does not actually grow. But .data starts with ALIGN(THREAD_SIZE) and
since .text shrinks the padding grows, and thus .data grows too as it
seen by /bin/size. diff System.map:
-
ffffffff81510000 D _sdata
-
ffffffff81510000 D init_thread_union
+
ffffffff81509000 D _sdata
+
ffffffff8150c000 D init_thread_union
Perhaps we can change vmlinux.lds.S to .data itself, so that /bin/size
can't "wrongly" report that .data grows if .text shinks.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Wed, 9 Jul 2014 04:08:17 +0000 (00:08 -0400)]
rcu: Use true/false instead of 1/0 for a bool type
This commit uses true/false instead of 1/0 for bool types in rcu_gp_fqs()
and force_qs_rnp().
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:18 +0000 (18:26 -0400)]
rcu: Return bool type in rcu_lockdep_current_cpu_online()
Return true instead of 1 in rcu_lockdep_current_cpu_online() as this
has bool as return type.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:13 +0000 (18:26 -0400)]
rcu: Return bool type for rcu_try_advance_all_cbs()
Return a bool type instead of 0 in rcu_try_advance_all_cbs().
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:11 +0000 (18:26 -0400)]
rcu: Use bool type for return value in rcu_is_watching()
Use a bool type for return in rcu_is_watching().
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 8 Jul 2014 22:26:10 +0000 (18:26 -0400)]
rcu: Fix sparse warning about rcu_batches_completed_preempt() being non-static
fix sparse warning about rcu_batches_completed_preempt() being non-static by
marking it as static
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Pranith Kumar [Tue, 8 Jul 2014 21:46:50 +0000 (17:46 -0400)]
rcu: Remove remaining read-modify-write ACCESS_ONCE() calls
Change the remaining uses of ACCESS_ONCE() so that each ACCESS_ONCE() either does a load or a store, but not both.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Wed, 13 Aug 2014 22:40:02 +0000 (15:40 -0700)]
memory-barriers: Fix description of 2-legged-if-based control dependencies
Sad to say, current compilers really will hoist identical stores from both
branches of an "if" statement to precede the conditional. This commit
therefore updates the description of control dependencies to reflect this
ugly reality.
Reported-by: Pranith Kumar <bobby.prani@gmail.com>
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Mon, 4 Aug 2014 18:49:34 +0000 (11:49 -0700)]
memory-barriers: Retain barrier() in fold-to-zero example
The transformation in the fold-to-zero example incorrectly omits the
barrier() directive. This commit therefore adds it back in.
Reported-by: Pranith Kumar <pranith@gatech.edu>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Paul E. McKenney [Sat, 26 Jul 2014 00:05:24 +0000 (17:05 -0700)]
memory-barriers: Fix control-ordering no-transitivity example
The control-ordering example demonstrating lack of transitivity had
multiple problems. This commit fixes them.
Reported-by: Nikolay Samofatov <nikolay.samofatov@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
Pranith Kumar [Wed, 27 Aug 2014 20:43:40 +0000 (16:43 -0400)]
rcu: Make nocb leader kthreads process pending callbacks after spawning
The nocb callbacks generated before the nocb kthreads are spawned are
enqueued in the nocb queue for later processing. Commit
fbce7497ee5af ("rcu:
Parallelize and economize NOCB kthread wakeups") introduced nocb leader kthreads
which checked the nocb_leader_wake flag to see if there were any such pending
callbacks. A case was reported in which newly spawned leader kthreads were not
processing the pending callbacks as this flag was not set, which led to a boot
hang.
The following commit ensures that the newly spawned nocb kthreads process the
pending callbacks by allowing the kthreads to run immediately after spawning
instead of waiting. This is done by inverting the logic of nocb_leader_wake
tests to nocb_leader_sleep which allows us to use the default initialization of
this flag to 0 to let the kthreads run.
Reported-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Link: http://www.spinics.net/lists/kernel/msg1802899.html
[ paulmck: Backported to v3.17-rc2. ]
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Amit Shah <amit.shah@redhat.com>
Linus Torvalds [Mon, 25 Aug 2014 22:36:20 +0000 (15:36 -0700)]
Linux 3.17-rc2
Linus Torvalds [Mon, 25 Aug 2014 22:34:28 +0000 (15:34 -0700)]
Merge tag 'nfs-for-3.17-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client fixes from Trond Myklebust:
"Highlights:
- more fixes for read/write codepath regressions
* sleeping while holding the inode lock
* stricter enforcement of page contiguity when coalescing requests
* fix up error handling in the page coalescing code
- don't busy wait on SIGKILL in the file locking code"
* tag 'nfs-for-3.17-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
nfs: Don't busy-wait on SIGKILL in __nfs_iocounter_wait
nfs: can_coalesce_requests must enforce contiguity
nfs: disallow duplicate pages in pgio page vectors
nfs: don't sleep with inode lock in lock_and_join_requests
nfs: fix error handling in lock_and_join_requests
nfs: use blocking page_group_lock in add_request
nfs: fix nonblocking calls to nfs_page_group_lock
nfs: change nfs_page_group_lock argument
Linus Torvalds [Mon, 25 Aug 2014 22:29:33 +0000 (15:29 -0700)]
Merge tag 'renesas-sh-drivers-for-v3.17' of git://git./linux/kernel/git/horms/renesas
Pull SH driver fix from Simon Horman:
"Confine SH_INTC to platforms that need it"
* tag 'renesas-sh-drivers-for-v3.17' of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:
sh: intc: Confine SH_INTC to platforms that need it
Linus Torvalds [Mon, 25 Aug 2014 22:28:57 +0000 (15:28 -0700)]
Merge branch 'upstream' of git://git.linux-mips.org/ralf/upstream-linus
Pull MIPS fixes from Ralf Baechle:
"Pretty much all across the field so with this we should be in
reasonable shape for the upcoming -rc2"
* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
MIPS: OCTEON: make get_system_type() thread-safe
MIPS: CPS: Initialize EVA before bringing up VPEs from secondary cores
MIPS: Malta: EVA: Rename 'eva_entry' to 'platform_eva_init'
MIPS: EVA: Add new EVA header
MIPS: scall64-o32: Fix indirect syscall detection
MIPS: syscall: Fix AUDIT value for O32 processes on MIPS64
MIPS: Loongson: Fix COP2 usage for preemptible kernel
MIPS: NL: Fix nlm_xlp_defconfig build error
MIPS: Remove race window in page fault handling
MIPS: Malta: Improve system memory detection for '{e, }memsize' >= 2G
MIPS: Alchemy: Fix db1200 PSC clock enablement
MIPS: BCM47XX: Fix reboot problem on BCM4705/BCM4785
MIPS: Remove duplicated include from numa.c
MIPS: Add common plat_irq_dispatch declaration
MIPS: MSP71xx: remove unused plat_irq_dispatch() argument
MIPS: GIC: Remove useless parens from GICBIS().
MIPS: perf: Mark pmu interupt IRQF_NO_THREAD
Linus Torvalds [Mon, 25 Aug 2014 22:11:53 +0000 (15:11 -0700)]
Merge tag 'trace-fixes-v3.17-rc1' of git://git./linux/kernel/git/rostedt/linux-trace
Pull fix for ftrace function tracer/profiler conflict from Steven Rostedt:
"The rewrite of the ftrace code that makes it possible to allow for
separate trampolines had a design flaw with the interaction between
the function and function_graph tracers.
The main flaw was the simplification of the use of multiple tracers
having the same filter (like function and function_graph, that use the
set_ftrace_filter file to filter their code). The design assumed that
the two tracers could never run simultaneously as only one tracer can
be used at a time. The problem with this assumption was that the
function profiler could be implemented on top of the function graph
tracer, and the function profiler could run at the same time as the
function tracer. This caused the assumption to be broken and when
ftrace detected this failed assumpiton it would spit out a nasty
warning and shut itself down.
Instead of using a single ftrace_ops that switches between the
function and function_graph callbacks, the two tracers can again use
their own ftrace_ops. But instead of having a complex hierarchy of
ftrace_ops, the filter fields are placed in its own structure and the
ftrace_ops can carefully use the same filter. This change took a bit
to be able to allow for this and currently only the global_ops can
share the same filter, but this new design can easily be modified to
allow for any ftrace_ops to share its filter with another ftrace_ops.
The first four patches deal with the change of allowing the ftrace_ops
to share the filter (and this needs to go to 3.16 as well).
The fifth patch fixes a bug that was also caused by the new changes
but only for archs other than x86, and only if those archs implement a
direct call to the function_graph tracer which they do not do yet but
will in the future. It does not need to go to stable, but needs to be
fixed before the other archs update their code to allow direct calls
to the function_graph trampoline"
* tag 'trace-fixes-v3.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Use current addr when converting to nop in __ftrace_replace_code()
ftrace: Fix function_profiler and function tracer together
ftrace: Fix up trampoline accounting with looping on hash ops
ftrace: Update all ftrace_ops for a ftrace_hash_ops update
ftrace: Allow ftrace_ops to use the hashes from other ops
Linus Torvalds [Sun, 24 Aug 2014 23:17:41 +0000 (16:17 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"A couple of EFI fixes, plus misc fixes all around the map"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
efi/arm64: Store Runtime Services revision
firmware: Do not use WARN_ON(!spin_is_locked())
x86_32, entry: Clean up sysenter_badsys declaration
x86/doc: Fix the 'tlb_single_page_flush_ceiling' sysconfig path
x86/mm: Fix sparse 'tlb_single_page_flush_ceiling' warning and make the variable read-mostly
x86/mm: Fix RCU splat from new TLB tracepoints
Linus Torvalds [Sun, 24 Aug 2014 23:16:55 +0000 (16:16 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"A kprobes and a perf compat ioctl fix"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Handle compat ioctl
kprobes: Skip kretprobe hit in NMI context to avoid deadlock
Linus Torvalds [Sun, 24 Aug 2014 22:57:00 +0000 (15:57 -0700)]
Merge tag 'fixes-for-linus' of git://git./linux/kernel/git/arm/arm-soc
Pull ARM SoC fixes from Olof Johansson:
"A collection of fixes from this week, it's been pretty quiet and
nothing really stands out as particularly noteworthy here -- mostly
minor fixes across the field:
- ODROID booting was fixed due to PMIC interrupts missing in DT
- a collection of i.MX fixes
- minor Tegra fix for regulators
- Rockchip fix and addition of SoC-specific mailing list to make it
easier to find posted patches"
* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
bus: arm-ccn: Fix warning message
ARM: shmobile: koelsch: Remove non-existent i2c6 pinmux
ARM: tegra: apalis/colibri t30: fix on-module 5v0 supplies
MAINTAINERS: add new Rockchip SoC list
ARM: dts: rockchip: readd missing mmc0 pinctrl settings
ARM: dts: ODROID i2c improvements
ARM: dts: Enable PMIC interrupts on ODROID
ARM: dts: imx6sx: fix the pad setting for uart CTS_B
ARM: dts: i.MX53: fix apparent bug in VPU clks
ARM: imx: correct gpu2d_axi and gpu3d_axi clock setting
ARM: dts: imx6: edmqmx6: change enet reset pin
ARM: dts: vf610-twr: Fix pinctrl_esdhc1 pin definitions.
ARM: imx: remove unnecessary ARCH_HAS_OPP select
ARM: imx: fix TLB missing of IOMUXC base address during suspend
ARM: imx6: fix SMP compilation again
ARM: dt: sun6i: Add #address-cells and #size-cells to i2c controller nodes
Linus Torvalds [Sun, 24 Aug 2014 22:54:23 +0000 (15:54 -0700)]
Merge tag 'gpio-v3.17-2' of git://git./linux/kernel/git/linusw/linux-gpio
Pull gpio fixes from Linus Walleij:
- a largeish fix for the IRQ handling in the new Zynq driver. The
quite verbose commit message gives the exact details.
- move some defines for gpiod flags outside an ifdef to make stub
functions work again.
- various minor fixes that we can accept for -rc1.
* tag 'gpio-v3.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio-lynxpoint: enable input sensing in resume
gpio: move GPIOD flags outside #ifdef
gpio: delete unneeded test before of_node_put
gpio: zynq: Fix IRQ handlers
gpiolib: devres: use correct structure type name in sizeof
MAINTAINERS: Change maintainer for gpio-bcm-kona.c
Linus Torvalds [Sun, 24 Aug 2014 22:48:12 +0000 (15:48 -0700)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull drm fixes from Dave Airlie:
"Intel and radeon fixes.
Post KS/LC git requests from i915 and radeon stacked up. They are all
fixes along with some new pci ids for radeon, and one maintainers file
entry.
- i915: display fixes and irq fixes
- radeon: pci ids, and misc gpuvm, dpm and hdp cache"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (29 commits)
MAINTAINERS: Add entry for Renesas DRM drivers
drm/radeon: add additional SI pci ids
drm/radeon: add new bonaire pci ids
drm/radeon: add new KV pci id
Revert "drm/radeon: Use write-combined CPU mappings of ring buffers with PCIe"
drm/radeon: fix active_cu mask on SI and CIK after re-init (v3)
drm/radeon: fix active cu count for SI and CIK
drm/radeon: re-enable selective GPUVM flushing
drm/radeon: Sync ME and PFP after CP semaphore waits v4
drm/radeon: fix display handling in radeon_gpu_reset
drm/radeon: fix pm handling in radeon_gpu_reset
drm/radeon: Only flush HDP cache for indirect buffers from userspace
drm/radeon: properly document reloc priority mask
drm/i915: don't try to retrain a DP link on an inactive CRTC
drm/i915: make sure VDD is turned off during system suspend
drm/i915: cancel hotplug and dig_port work during suspend and unload
drm/i915: fix HPD IRQ reenable work cancelation
drm/i915: take display port power domain in DP HPD handler
drm/i915: Don't try to enable cursor from setplane when crtc is disabled
drm/i915: Skip load detect when intel_crtc->new_enable==true
...
Benjamin LaHaise [Sun, 24 Aug 2014 17:14:05 +0000 (13:14 -0400)]
aio: fix reqs_available handling
As reported by Dan Aloni, commit
f8567a3845ac ("aio: fix aio request
leak when events are reaped by userspace") introduces a regression when
user code attempts to perform io_submit() with more events than are
available in the ring buffer. Reverting that commit would reintroduce a
regression when user space event reaping is used.
Fixing this bug is a bit more involved than the previous attempts to fix
this regression. Since we do not have a single point at which we can
count events as being reaped by user space and io_getevents(), we have
to track event completion by looking at the number of events left in the
event ring. So long as there are as many events in the ring buffer as
there have been completion events generate, we cannot call
put_reqs_available(). The code to check for this is now placed in
refill_reqs_available().
A test program from Dan and modified by me for verifying this bug is available
at http://www.kvack.org/~bcrl/
20140824-aio_bug.c .
Reported-by: Dan Aloni <dan@kernelim.com>
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
Acked-by: Dan Aloni <dan@kernelim.com>
Cc: Kent Overstreet <kmo@daterainc.com>
Cc: Mateusz Guzik <mguzik@redhat.com>
Cc: Petr Matousek <pmatouse@redhat.com>
Cc: stable@vger.kernel.org # v3.16 and anything that f8567a3845ac was backported to
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pawel Moll [Mon, 18 Aug 2014 17:20:49 +0000 (18:20 +0100)]
bus: arm-ccn: Fix warning message
A message warning a user about wrong vc value was printing
out port instead.
Reported-by: Drew Richardson <drew.richardson@arm.com>
Signed-off-by: Pawel Moll <pawel.moll@arm.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
Geert Uytterhoeven [Thu, 14 Aug 2014 22:19:10 +0000 (07:19 +0900)]
ARM: shmobile: koelsch: Remove non-existent i2c6 pinmux
On r8a7791, i2c6 (aka iic3) doesn't need pinmux, but the koelsch dts
refers to non-existent pinmux configuration data:
pinmux core: sh-pfc does not support function i2c6
sh-pfc
e6060000.pfc: invalid function i2c6 in map table
Remove it to fix this.
Fixes: commit
1d41f36a68c0f4e9b01d563ce33bab5201858b54 ("ARM: shmobile:
koelsch dts: Add VDD MPU regulator for DVFS")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: Olof Johansson <olof@lixom.net>
Marcel Ziswiler [Fri, 22 Aug 2014 19:25:10 +0000 (13:25 -0600)]
ARM: tegra: apalis/colibri t30: fix on-module 5v0 supplies
Working on Gigabit/PCIe support in U-Boot for Apalis T30 I realised
that the current device tree source includes for our modules only
happen to work due to referencing the on-carrier 5v0 supply from USB
which is not at all available on-module. The modules actually contain
TPS60150 charge pumps to generate the PMIC required 5 volts from the
one and only 3.3 volt module supply. This patch fixes this.
(Note: When back-porting this to v3.16 stable releases, simply drop the
change to tegra30-apalis.dtsi; that file was added in v3.17)
Cc: <stable@vger.kernel.org> #v3.16+
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
Olof Johansson [Sun, 24 Aug 2014 18:19:37 +0000 (11:19 -0700)]
Merge tag 'v3.17-rockchip-fixes1' of git://git./linux/kernel/git/mmind/linux-rockchip into fixes
Merge "ARM: rockchip: fix for 3.17" from Heiko Stubner:
Pinctrl that got accidentially dropped when reorganizing the
dts files and addition of the new Rockchip list to MAINTAINERS.
* tag 'v3.17-rockchip-fixes1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
MAINTAINERS: add new Rockchip SoC list
ARM: dts: rockchip: readd missing mmc0 pinctrl settings
Signed-off-by: Olof Johansson <olof@lixom.net>
Laurent Pinchart [Tue, 19 Aug 2014 21:04:31 +0000 (23:04 +0200)]
MAINTAINERS: Add entry for Renesas DRM drivers
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Dave Airlie <airlied@gmail.com>
Dave Airlie [Sun, 24 Aug 2014 05:47:46 +0000 (15:47 +1000)]
Merge branch 'drm-fixes-3.17' of git://people.freedesktop.org/~agd5f/linux into drm-next
This pull just contains some new pci ids.
* 'drm-fixes-3.17' of git://people.freedesktop.org/~agd5f/linux:
drm/radeon: add additional SI pci ids
drm/radeon: add new bonaire pci ids
drm/radeon: add new KV pci id
Heiko Stuebner [Wed, 20 Aug 2014 10:31:03 +0000 (12:31 +0200)]
MAINTAINERS: add new Rockchip SoC list
Add the new list that Rockchip-specific patches should also be directed to.
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Heiko Stuebner [Mon, 11 Aug 2014 17:51:44 +0000 (19:51 +0200)]
ARM: dts: rockchip: readd missing mmc0 pinctrl settings
During the restructuring of the Rockchip Cortex-A9 dtsi files it seems
like the pinctrl settings vanished at some point from the mmc0 support.
This of course renders them unusable, so readd the necessary pinctrl
properties.
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Olof Johansson [Sat, 23 Aug 2014 05:57:22 +0000 (22:57 -0700)]
Merge tag 'sunxi-dt-for-3.17-2' of git://git./linux/kernel/git/mripard/linux into fixes
Merge "Allwinner DT changes, take 2" from Maxime Ripard:
Only a single patch in here that fixes a DTC warning.
* tag 'sunxi-dt-for-3.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mripard/linux:
ARM: dt: sun6i: Add #address-cells and #size-cells to i2c controller nodes
Signed-off-by: Olof Johansson <olof@lixom.net>
Steven Rostedt (Red Hat) [Mon, 18 Aug 2014 00:59:10 +0000 (20:59 -0400)]
ftrace: Use current addr when converting to nop in __ftrace_replace_code()
In __ftrace_replace_code(), when converting the call to a nop in a function
it needs to compare against the "curr" (current) value of the ftrace ops, and
not the "new" one. It currently does not affect x86 which is the only arch
to do the trampolines with function graph tracer, but when other archs that do
depend on this code implement the function graph trampoline, it can crash.
Here's an example when ARM uses the trampolines (in the future):
------------[ cut here ]------------
WARNING: CPU: 0 PID: 9 at kernel/trace/ftrace.c:1716 ftrace_bug+0x17c/0x1f4()
Modules linked in: omap_rng rng_core ipv6
CPU: 0 PID: 9 Comm: migration/0 Not tainted
3.16.0-test-10959-gf0094b28f303-dirty #52
[<
c02188f4>] (unwind_backtrace) from [<
c021343c>] (show_stack+0x20/0x24)
[<
c021343c>] (show_stack) from [<
c095a674>] (dump_stack+0x78/0x94)
[<
c095a674>] (dump_stack) from [<
c02532a0>] (warn_slowpath_common+0x7c/0x9c)
[<
c02532a0>] (warn_slowpath_common) from [<
c02532ec>] (warn_slowpath_null+0x2c/0x34)
[<
c02532ec>] (warn_slowpath_null) from [<
c02cbac4>] (ftrace_bug+0x17c/0x1f4)
[<
c02cbac4>] (ftrace_bug) from [<
c02cc44c>] (ftrace_replace_code+0x80/0x9c)
[<
c02cc44c>] (ftrace_replace_code) from [<
c02cc658>] (ftrace_modify_all_code+0xb8/0x164)
[<
c02cc658>] (ftrace_modify_all_code) from [<
c02cc718>] (__ftrace_modify_code+0x14/0x1c)
[<
c02cc718>] (__ftrace_modify_code) from [<
c02c7244>] (multi_cpu_stop+0xf4/0x134)
[<
c02c7244>] (multi_cpu_stop) from [<
c02c6e90>] (cpu_stopper_thread+0x54/0x130)
[<
c02c6e90>] (cpu_stopper_thread) from [<
c0271cd4>] (smpboot_thread_fn+0x1ac/0x1bc)
[<
c0271cd4>] (smpboot_thread_fn) from [<
c026ddf0>] (kthread+0xe0/0xfc)
[<
c026ddf0>] (kthread) from [<
c020f318>] (ret_from_fork+0x14/0x20)
---[ end trace
dc9ce72c5b617d8f ]---
[ 65.047264] ftrace failed to modify [<
c0208580>] asm_do_IRQ+0x10/0x1c
[ 65.054070] actual: 85:1b:00:eb
Fixes:
7413af1fb70e7 "ftrace: Make get_ftrace_addr() and get_ftrace_addr_old() global"
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Steven Rostedt (Red Hat) [Fri, 15 Aug 2014 21:18:46 +0000 (17:18 -0400)]
ftrace: Fix function_profiler and function tracer together
The latest rewrite of ftrace removed the separate ftrace_ops of
the function tracer and the function graph tracer and had them
share the same ftrace_ops. This simplified the accounting by removing
the multiple layers of functions called, where the global_ops func
would call a special list that would iterate over the other ops that
were registered within it (like function and function graph), which
itself was registered to the ftrace ops list of all functions
currently active. If that sounds confusing, the code that implemented
it was also confusing and its removal is a good thing.
The problem with this change was that it assumed that the function
and function graph tracer can never be used at the same time.
This is mostly true, but there is an exception. That is when the
function profiler uses the function graph tracer to profile.
The function profiler can be activated the same time as the function
tracer, and this breaks the assumption and the result is that ftrace
will crash (it detects the error and shuts itself down, it does not
cause a kernel oops).
To solve this issue, a previous change allowed the hash tables
for the functions traced by a ftrace_ops to be a pointer and let
multiple ftrace_ops share the same hash. This allows the function
and function_graph tracer to have separate ftrace_ops, but still
share the hash, which is what is done.
Now the function and function graph tracers have separate ftrace_ops
again, and the function tracer can be run while the function_profile
is active.
Cc: stable@vger.kernel.org # 3.16 (apply after 3.17-rc4 is out)
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
David Jeffery [Tue, 5 Aug 2014 15:19:42 +0000 (11:19 -0400)]
nfs: Don't busy-wait on SIGKILL in __nfs_iocounter_wait
If a SIGKILL is sent to a task waiting in __nfs_iocounter_wait,
it will busy-wait or soft lockup in its while loop.
nfs_wait_bit_killable won't sleep, and the loop won't exit on
the error return.
Stop the busy-wait by breaking out of the loop when
nfs_wait_bit_killable returns an error.
Signed-off-by: David Jeffery <djeffery@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Thu, 14 Aug 2014 21:39:33 +0000 (17:39 -0400)]
nfs: can_coalesce_requests must enforce contiguity
Commit
6094f83864c1d1296566a282cba05ba613f151ee
"nfs: allow coalescing of subpage requests" got rid of the requirement
that requests cover whole pages, but it made some incorrect assumptions.
It turns out that callers of this interface can map adjacent requests
(by file position as seen by req_offset + req->wb_bytes) to different pages,
even when they could share a page. An example is the direct I/O interface -
iov_iter_get_pages_alloc may return one segment with a partial page filled
and the next segment (which is adjacent in the file position) starts with a
new page.
Reported-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Thu, 14 Aug 2014 21:39:32 +0000 (17:39 -0400)]
nfs: disallow duplicate pages in pgio page vectors
Adjacent requests that share the same page are allowed, but should only
use one entry in the page vector. This avoids overruning the page
vector - it is sized based on how many bytes there are, not by
request count.
This fixes issues that manifest as "Redzone overwritten" bugs (the
vector overrun) and hangs waiting on page read / write, as it waits on
the same page more than once.
This also adds bounds checking to the page vector with a graceful failure
(WARN_ON_ONCE and pgio error returned to application).
Reported-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Fri, 8 Aug 2014 15:00:57 +0000 (11:00 -0400)]
nfs: don't sleep with inode lock in lock_and_join_requests
This handles the 'nonblock=false' case in nfs_lock_and_join_requests.
If the group is already locked and blocking is allowed, drop the inode lock
and wait for the group lock to be cleared before trying it all again.
This should fix warnings found in peterz's tree (sched/wait branch), where
might_sleep() checks are added to wait.[ch].
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Reviewed-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Fri, 8 Aug 2014 15:00:56 +0000 (11:00 -0400)]
nfs: fix error handling in lock_and_join_requests
This fixes handling of errors from nfs_page_group_lock in
nfs_lock_and_join_requests. It now releases the inode lock and the
reference to the head request.
Reported-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Reviewed-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Fri, 8 Aug 2014 15:00:55 +0000 (11:00 -0400)]
nfs: use blocking page_group_lock in add_request
__nfs_pageio_add_request was calling nfs_page_group_lock nonblocking, but
this can return -EAGAIN which would end up passing -EIO to the application.
There is no reason not to block in this path, so change the two calls to
do so. Also, there is no need to check the return value of
nfs_page_group_lock when nonblock=false, so remove the error handling code.
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Reviewed-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Fri, 8 Aug 2014 15:00:54 +0000 (11:00 -0400)]
nfs: fix nonblocking calls to nfs_page_group_lock
nfs_page_group_lock was calling wait_on_bit_lock even when told not to
block. Fix by first trying test_and_set_bit, followed by wait_on_bit_lock
if and only if blocking is allowed. Return -EAGAIN if nonblocking and the
test_and_set of the bit was already locked.
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Reviewed-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Weston Andros Adamson [Fri, 8 Aug 2014 15:00:53 +0000 (11:00 -0400)]
nfs: change nfs_page_group_lock argument
Flip the meaning of the second argument from 'wait' to 'nonblock' to
match related functions. Update all five calls to reflect this change.
Signed-off-by: Weston Andros Adamson <dros@primarydata.com>
Reviewed-by: Peng Tao <tao.peng@primarydata.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Linus Torvalds [Fri, 22 Aug 2014 21:50:21 +0000 (14:50 -0700)]
Merge tag 'pwm/for-3.17-rc2' of git://git./linux/kernel/git/thierry.reding/linux-pwm
Pull pwm fix from Thierry Reding:
"Just one bugfix for the PWM lookup table code that would cause a PWM
channel to be set to the wrong period and polarity for non-perfect
matches"
* tag 'pwm/for-3.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm:
pwm: Fix period and polarity in pwm_get() for non-perfect matches
Michal Kazior [Mon, 18 Aug 2014 11:19:09 +0000 (13:19 +0200)]
mac80211: fix channel switch for chanctx-based drivers
The new_ctx pointer is set only for non-chanctx drivers. This yielded a
crash for chanctx-based drivers during channel switch finalization:
BUG: unable to handle kernel NULL pointer dereference at
0000000000000020
IP: ieee80211_vif_use_reserved_switch+0x71c/0xb00 [mac80211]
Use an adequate chanctx pointer to fix this.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Fri, 22 Aug 2014 21:33:18 +0000 (14:33 -0700)]
Merge git://git./linux/kernel/git/davem/net
Pull networking fixes from David Miller:
"Here are some bug fixes that have piled up during ksummit/linuxcon.
1) Fix endian problems in ibmveth, from Anton Blanchard.
2) IPV6 routing code does GFP_KERNEL allocation in atomic, fix from
Benjamin Block.
3) SCTP association fixes from Daniel Borkmann.
4) When multiple VLAN headers are present we have to make sure the
second and subsequent ones are pullable in the SKB otherwise we
blindly dereference garbage. From Jiri Benc.
5) The argument adjustment of the signature of hlist_add_after*()
introduced a regression in the batman-adv code, fix from Sven
Eckelmann.
6) Fix TX hang handling to avoid a panic in i40e, from Anjali Singhai
Jain.
7) PTP flag test is inverted in i40e driver, from Jesse Brandeburg.
8) ATM LEC driver needs to hold RTNL mutex over MTU changes, from
Chas Williams.
9) Truncate packets larger then the TPACKET_V3 format configured
buffers, otherwise we overwrite past the end of said buffers.
From Eric Dumazet.
10) Fix endianness bugs in qlcnic firmware handling, from Rajesh
Borundia and Shahed Shaikh.
11) CXGB4 sometimes doesn't get all of the TX completion events it
should resulting in SKBs getting stuck in the TX queue, from
Hariprasad Shenai.
12) When the FEC chip's PTP clock is disabled, you can't access the
register. Add necessary checks to avoid the resulting hang, from
Fugang Duan"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (37 commits)
drivers: isdn: eicon: xdi_msg.h: Fix typo in #ifndef
net: sctp: fix suboptimal edge-case on non-active active/retrans path selection
net: sctp: spare unnecessary comparison in sctp_trans_elect_best
net: ethernet: broadcom: bnx2x: Remove redundant #ifdef
ibmveth: Fix endian issues with rx_no_buffer statistic
net: xgene: fix possible NULL dereference in xgene_enet_free_desc_rings()
openvswitch: fix panic with multiple vlan headers
net: ipv6: fib: don't sleep inside atomic lock
net: fec: ptp: avoid register access when ipg clock is disabled
cxgb4: Free completed tx skbs promptly
cxgb4: Fix race condition in cleanup
sctp: not send SCTP_PEER_ADDR_CHANGE notifications with failed probe
bnx2x: Revert UNDI flushing mechanism
qlcnic: Fix endianess issue in firmware load from file operation
qlcnic: Fix endianess issue in FW dump template header
qlcnic: Fix flash access interface to application
MAINTAINERS: Add section for MRF24J40 IEEE 802.15.4 radio driver
macvlan: Allow setting multicast filter on all macvlan types
packet: handle too big packets for PACKET_V3
MAINTAINERS: add entry for ec_bhf driver
...
Steven Rostedt (Red Hat) [Thu, 21 Aug 2014 03:57:04 +0000 (23:57 -0400)]
ftrace: Fix up trampoline accounting with looping on hash ops
Now that a ftrace_hash can be shared by multiple ftrace_ops, they can dec
the rec->flags by more than once (one per those that share the ftrace_hash).
This means that the tramp_hash may not have a hash item when it was added.
For example, if two ftrace_ops share a hash for a ftrace record, and the
first ops has a trampoline, when it adds itself it will set the rec->flags
TRAMP flag and increments its nr_trampolines counter. When the second ops
is added, it must clear that tramp flag but also decrement the other ops
that shares its hash. As the update to the function callbacks has not yet
been performed, the other ops will not have the tramp hash set yet and it
can not be used to know to decrement its nr_trampolines.
Luckily, the tramp_hash does not need to be used. As the ftrace_mutex is
held, a ops with a trampoline to a record during an update of another ops
that shares the record will have its func_hash pointing to it. Since a
trampoline can only be set for a record if only one ops is attached to it,
we can just check if the record has a trampoline (the FTRACE_FL_TRAMP flag
is set) and then find the ops that has this record in its hashes.
Also added some output to help debug when things go wrong.
Cc: stable@vger.kernel.org # 3.16+ (apply after 3.17-rc4 is out)
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Rasmus Villemoes [Fri, 22 Aug 2014 16:39:14 +0000 (18:39 +0200)]
drivers: isdn: eicon: xdi_msg.h: Fix typo in #ifndef
Test for definedness of the macro which is actually defined (the
change is hard to see: it is s/SSS/SSA/).
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 22 Aug 2014 11:03:30 +0000 (13:03 +0200)]
net: sctp: fix suboptimal edge-case on non-active active/retrans path selection
In SCTP, selection of active (T.ACT) and retransmission (T.RET)
transports is being done whenever transport control operations
(UP, DOWN, PF, ...) are engaged through sctp_assoc_control_transport().
Commits
4c47af4d5eb2 ("net: sctp: rework multihoming retransmission
path selection to rfc4960") and
a7288c4dd509 ("net: sctp: improve
sctp_select_active_and_retran_path selection") have both improved
it towards a more fine-grained and optimal path selection.
Currently, the selection algorithm for T.ACT and T.RET is as follows:
1) Elect the two most recently used ACTIVE transports T1, T2 for
T.ACT, T.RET, where T.ACT<-T1 and T1 is most recently used
2) In case primary path T.PRI not in {T1, T2} but ACTIVE, set
T.ACT<-T.PRI and T.RET<-T1
3) If only T1 is ACTIVE from the set, set T.ACT<-T1 and T.RET<-T1
4) If none is ACTIVE, set T.ACT<-best(T.PRI, T.RET, T3) where
T3 is the most recently used (if avail) in PF, set T.RET<-T.PRI
Prior to above commits, 4) was simply a camp on T.ACT<-T.PRI and
T.RET<-T.PRI, ignoring possible paths in PF. Camping on T.PRI is
still slightly suboptimal as it can lead to the following scenario:
Setup:
<A> <B>
T1: p1p1 (10.0.10.10) <==> .'`) <==> p1p1 (10.0.10.12) <= T.PRI
T2: p1p2 (10.0.10.20) <==> (_ . ) <==> p1p2 (10.0.10.22)
net.sctp.rto_min = 1000
net.sctp.path_max_retrans = 2
net.sctp.pf_retrans = 0
net.sctp.hb_interval = 1000
T.PRI is permanently down, T2 is put briefly into PF state (e.g. due to
link flapping). Here, the first time transmission is sent over PF path
T2 as it's the only non-INACTIVE path, but the retransmitted data-chunks
are sent over the INACTIVE path T1 (T.PRI), which is not good.
After the patch, it's choosing better transports in both cases by
modifying step 4):
4) If none is ACTIVE, set T.ACT_new<-best(T.ACT_old, T3) where T3 is
the most recently used (if avail) in PF, set T.RET<-T.ACT_new
This will still select a best possible path in PF if available (which
can also include T.PRI/T.RET), and set both T.ACT/T.RET to it.
In case sctp_assoc_control_transport() *just* put T.ACT_old into INACTIVE
as it transitioned from ACTIVE->PF->INACTIVE and stays in INACTIVE just
for a very short while before going back ACTIVE, it will guarantee that
this path will be reselected for T.ACT/T.RET since T3 (PF) is not
available.
Previously, this was not possible, as we would only select between T.PRI
and T.RET, and a possible T3 would be NULL due to the fact that we have
just transitioned T3 in sctp_assoc_control_transport() from PF->INACTIVE
and would select a suboptimal path when T.PRI/T.RET have worse properties.
In the case that T.ACT_old permanently went to INACTIVE during this
transition and there's no PF path available, plus T.PRI and T.RET are
INACTIVE as well, we would now camp on T.ACT_old, but if everything is
being INACTIVE there's really not much we can do except hoping for a
successful HB to bring one of the transports back up again and, thus
cause a new selection through sctp_assoc_control_transport().
Now both tests work fine:
Case 1:
1. T1 S(ACTIVE) T.ACT
T2 S(ACTIVE) T.RET
2. T1 S(ACTIVE) T.ACT, T.RET
T2 S(PF)
3. T1 S(ACTIVE) T.ACT, T.RET
T2 S(INACTIVE)
5. T1 S(PF) T.ACT, T.RET
T2 S(INACTIVE)
[ 5.1 T1 S(INACTIVE) T.ACT, T.RET
T2 S(INACTIVE) ]
6. T1 S(ACTIVE) T.ACT, T.RET
T2 S(INACTIVE)
7. T1 S(ACTIVE) T.ACT
T2 S(ACTIVE) T.RET
Case 2:
1. T1 S(ACTIVE) T.ACT
T2 S(ACTIVE) T.RET
2. T1 S(PF)
T2 S(ACTIVE) T.ACT, T.RET
3. T1 S(INACTIVE)
T2 S(ACTIVE) T.ACT, T.RET
5. T1 S(INACTIVE)
T2 S(PF) T.ACT, T.RET
[ 5.1 T1 S(INACTIVE)
T2 S(INACTIVE) T.ACT, T.RET ]
6. T1 S(INACTIVE)
T2 S(ACTIVE) T.ACT, T.RET
7. T1 S(ACTIVE) T.ACT
T2 S(ACTIVE) T.RET
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 22 Aug 2014 11:03:29 +0000 (13:03 +0200)]
net: sctp: spare unnecessary comparison in sctp_trans_elect_best
When both transports are the same, we don't have to go down that
road only to realize that we will return the very same transport.
We are guaranteed that curr is always non-NULL. Therefore, just
short-circuit this special case.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>