This patch favours moving tasks towards NUMA node that recorded a higher
number of NUMA faults during active load balancing. Ideally this is
self-reinforcing as the longer the task runs on that node, the more faults
it should incur causing task_numa_placement to keep the task running on that
node. In reality a big weakness is that the nodes CPUs can be overloaded
and it would be more efficient to queue tasks on an idle node and migrate
to the new node. This would require additional smarts in the balancer so
for now the balancer will simply prefer to place the task on the preferred
node for a PTE scans which is controlled by the numa_balancing_settle_count
sysctl. Once the settle_count number of scans has complete the schedule
is free to place the task on an alternative node if the load is imbalanced.
[srikar@linux.vnet.ibm.com: Fixed statistics]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
[ Tunable and use higher faults instead of preferred. ]
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-23-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
NUMA hinting fault counts and placement decisions are both recorded in the
same array which distorts the samples in an unpredictable fashion. The values
linearly accumulate during the scan and then decay creating a sawtooth-like
pattern in the per-node counts. It also means that placement decisions are
time sensitive. At best it means that it is very difficult to state that
the buffer holds a decaying average of past faulting behaviour. At worst,
it can confuse the load balancer if it sees one node with an artifically high
count due to very recent faulting activity and may create a bouncing effect.
This patch adds a second array. numa_faults stores the historical data
which is used for placement decisions. numa_faults_buffer holds the
fault activity during the current scan window. When the scan completes,
numa_faults decays and the values from numa_faults_buffer are copied
across.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-22-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The NUMA PTE scan rate is controlled with a combination of the
numa_balancing_scan_period_min, numa_balancing_scan_period_max and
numa_balancing_scan_size. This scan rate is independent of the size
of the task and as an aside it is further complicated by the fact that
numa_balancing_scan_size controls how many pages are marked pte_numa and
not how much virtual memory is scanned.
In combination, it is almost impossible to meaningfully tune the min and
max scan periods and reasoning about performance is complex when the time
to complete a full scan is is partially a function of the tasks memory
size. This patch alters the semantic of the min and max tunables to be
about tuning the length time it takes to complete a scan of a tasks occupied
virtual address space. Conceptually this is a lot easier to understand. There
is a "sanity" check to ensure the scan rate is never extremely fast based on
the amount of virtual memory that should be scanned in a second. The default
of 2.5G seems arbitrary but it is to have the maximum scan rate after the
patch roughly match the maximum scan rate before the patch was applied.
On a similar note, numa_scan_period is in milliseconds and not
jiffies. Properly placed pages slow the scanning rate but adding 10 jiffies
to numa_scan_period means that the rate scanning slows depends on HZ which
is confusing. Get rid of the jiffies_to_msec conversion and treat it as ms.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-18-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
PTE scanning and NUMA hinting fault handling is expensive so commit
5bca2303 ("mm: sched: numa: Delay PTE scanning until a task is scheduled
on a new node") deferred the PTE scan until a task had been scheduled on
another node. The problem is that in the purely shared memory case that
this may never happen and no NUMA hinting fault information will be
captured. We are not ruling out the possibility that something better
can be done here but for now, this patch needs to be reverted and depend
entirely on the scan_delay to avoid punishing short-lived processes.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1381141781-10992-16-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Avoiding marking PTEs pte_numa because a particular NUMA node is migrate rate
limited sees like a bad idea. Even if this node can't migrate anymore other
nodes might and we want up-to-date information to do balance decisions.
We already rate limit the actual migrations, this should leave enough
bandwidth to allow the non-migrating scanning. I think its important we
keep up-to-date information if we're going to do placement based on it.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1381141781-10992-15-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With a trace_printk("working\n"); right after the cmpxchg in
task_numa_work() we can see that of a 4 thread process, its always the
same task winning the race and doing the protection change.
This is a problem since the task doing the protection change has a
penalty for taking faults -- it is busy when marking the PTEs. If its
always the same task the ->numa_faults[] get severely skewed.
Avoid this by delaying the task doing the protection change such that
it is unlikely to win the privilege again.
Before:
root@interlagos:~# grep "thread 0/.*working" /debug/tracing/trace | tail -15
thread 0/0-3232 [022] .... 212.787402: task_numa_work: working
thread 0/0-3232 [022] .... 212.888473: task_numa_work: working
thread 0/0-3232 [022] .... 212.989538: task_numa_work: working
thread 0/0-3232 [022] .... 213.090602: task_numa_work: working
thread 0/0-3232 [022] .... 213.191667: task_numa_work: working
thread 0/0-3232 [022] .... 213.292734: task_numa_work: working
thread 0/0-3232 [022] .... 213.393804: task_numa_work: working
thread 0/0-3232 [022] .... 213.494869: task_numa_work: working
thread 0/0-3232 [022] .... 213.596937: task_numa_work: working
thread 0/0-3232 [022] .... 213.699000: task_numa_work: working
thread 0/0-3232 [022] .... 213.801067: task_numa_work: working
thread 0/0-3232 [022] .... 213.903155: task_numa_work: working
thread 0/0-3232 [022] .... 214.005201: task_numa_work: working
thread 0/0-3232 [022] .... 214.107266: task_numa_work: working
thread 0/0-3232 [022] .... 214.209342: task_numa_work: working
After:
root@interlagos:~# grep "thread 0/.*working" /debug/tracing/trace | tail -15
thread 0/0-3253 [005] .... 136.865051: task_numa_work: working
thread 0/2-3255 [026] .... 136.965134: task_numa_work: working
thread 0/3-3256 [024] .... 137.065217: task_numa_work: working
thread 0/3-3256 [024] .... 137.165302: task_numa_work: working
thread 0/3-3256 [024] .... 137.265382: task_numa_work: working
thread 0/0-3253 [004] .... 137.366465: task_numa_work: working
thread 0/2-3255 [026] .... 137.466549: task_numa_work: working
thread 0/0-3253 [004] .... 137.566629: task_numa_work: working
thread 0/0-3253 [004] .... 137.666711: task_numa_work: working
thread 0/1-3254 [028] .... 137.766799: task_numa_work: working
thread 0/0-3253 [004] .... 137.866876: task_numa_work: working
thread 0/2-3255 [026] .... 137.966960: task_numa_work: working
thread 0/1-3254 [028] .... 138.067041: task_numa_work: working
thread 0/2-3255 [026] .... 138.167123: task_numa_work: working
thread 0/3-3256 [024] .... 138.267207: task_numa_work: working
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1381141781-10992-14-git-send-email-mgorman@suse.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge Linux v3.12-rc4 to fix a conflict and also to refresh the tree
before applying more scheduler patches.
Conflicts:
arch/avr32/include/asm/Kbuild
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While both the nr and total times are showed, having the avg
lock hold and wait times show in the report is quite useful when
working on performance related issues. Furthermore, I find
myself constantly doing the calculations manually.
In addition, some of the documentation examples were changed to
easily update them to show the two new columns. No textual
change otherwise, as descriptions match the lockstat output.
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Cc: aswin@hp.com
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1380746928.2313.14.camel@buesod1.americas.hpqcorp.net
[ Fixlets: changed a seq_printf() to seq_puts(), converted spaces to tabs. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull perf fixes from Ingo Molnar:
"Various fixlets:
On the kernel side:
- fix a race
- fix a bug in the handling of the perf ring-buffer data page
On the tooling side:
- fix the handling of certain corrupted perf.data files
- fix a bug in 'perf probe'
- fix a bug in 'perf record + perf sched'
- fix a bug in 'make install'
- fix a bug in libaudit feature-detection on certain distros"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf session: Fix infinite loop on invalid perf.data file
perf tools: Fix installation of libexec components
perf probe: Fix to find line information for probe list
perf tools: Fix libaudit test
perf stat: Set child_pid after perf_evlist__prepare_workload()
perf tools: Add default handler for mmap2 events
perf/x86: Clean up cap_user_time* setting
perf: Fix perf_pmu_migrate_context
Pull ACPI and power management fixes from Rafael Wysocki:
- The resume part of user space driven hibernation (s2disk) is now
broken after the change that moved the creation of memory bitmaps to
after the freezing of tasks, because I forgot that the resume utility
loaded the image before freezing tasks and needed the bitmaps for
that. The fix adds special handling for that case.
- One of recent commits changed the export of acpi_bus_get_device() to
EXPORT_SYMBOL_GPL(), which was technically correct but broke existing
binary modules using that function including one in particularly
widespread use. Change it back to EXPORT_SYMBOL().
- The intel_pstate driver sometimes fails to disable turbo if its
no_turbo sysfs attribute is set. Fix from Srinivas Pandruvada.
- One of recent cpufreq fixes forgot to update a check in cpufreq-cpu0
which still (incorrectly) treats non-NULL as non-error. Fix from
Philipp Zabel.
- The SPEAr cpufreq driver uses a wrong variable type in one place
preventing it from catching errors returned by one of the functions
called by it. Fix from Sachin Kamat.
* tag 'pm+acpi-3.12-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPI: Use EXPORT_SYMBOL() for acpi_bus_get_device()
intel_pstate: fix no_turbo
cpufreq: cpufreq-cpu0: NULL is a valid regulator, part 2
cpufreq: SPEAr: Fix incorrect variable type
PM / hibernate: Fix user space driven resume regression
Add a generic qualifier for transaction events, as a new sample
type that returns a flag word. This is particularly useful
for qualifying aborts: to distinguish aborts which happen
due to asynchronous events (like conflicts caused by another
CPU) versus instructions that lead to an abort.
The tuning strategies are very different for those cases,
so it's important to distinguish them easily and early.
Since it's inconvenient and inflexible to filter for this
in the kernel we report all the events out and allow
some post processing in user space.
The flags are based on the Intel TSX events, but should be fairly
generic and mostly applicable to other HTM architectures too. In addition
to various flag words there's also reserved space to report an
program supplied abort code. For TSX this is used to distinguish specific
classes of aborts, like a lock busy abort when doing lock elision.
Flags:
Elision and generic transactions (ELISION vs TRANSACTION)
(HLE vs RTM on TSX; IBM etc. would likely only use TRANSACTION)
Aborts caused by current thread vs aborts caused by others (SYNC vs ASYNC)
Retryable transaction (RETRY)
Conflicts with other threads (CONFLICT)
Transaction write capacity overflow (CAPACITY WRITE)
Transaction read capacity overflow (CAPACITY READ)
Transactions implicitely aborted can also return an abort code.
This can be used to signal specific events to the profiler. A common
case is abort on lock busy in a RTM eliding library (code 0xff)
To handle this case we include the TSX abort code
Common example aborts in TSX would be:
- Data conflict with another thread on memory read.
Flags: TRANSACTION|ASYNC|CONFLICT
- executing a WRMSR in a transaction. Flags: TRANSACTION|SYNC
- HLE transaction in user space is too large
Flags: ELISION|SYNC|CAPACITY-WRITE
The only flag that is somewhat TSX specific is ELISION.
This adds the perf core glue needed for reporting the new flag word out.
v2: Add MEM/MISC
v3: Move transaction to the end
v4: Separate capacity-read/write and remove misc
v5: Remove _SAMPLE. Move abort flags to 32bit. Rename
transaction to txn
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1379688044-14173-2-git-send-email-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While auditing the list_entry usage due to a trinity bug I found that
perf_pmu_migrate_context violates the rules for
perf_event::event_entry.
The problem is that perf_event::event_entry is a RCU list element, and
hence we must wait for a full RCU grace period before re-using the
element after deletion.
Therefore the usage in perf_pmu_migrate_context() which re-uses the
entry immediately is broken. For now introduce another list_head into
perf_event for this specific usage.
This doesn't actually fix the trinity report because that never goes
through this code.
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-mkj72lxagw1z8fvjm648iznw@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This patch adds a kgdb_nmicallin() interface that can be used by
external NMI handlers to call the KGDB/KDB handler. The primary
need for this is for those types of NMI interrupts where all the
CPUs have already received the NMI signal. Therefore no
send_IPI(NMI) is required, and in fact it will cause a 2nd
unhandled NMI to occur. This generates the "Dazed and Confuzed"
messages.
Since all the CPUs are getting the NMI at roughly the same time,
it's not guaranteed that the first CPU that hits the NMI handler
will manage to enter KGDB and set the dbg_master_lock before the
slaves start entering. The new argument "send_ready" was added
for KGDB to signal the NMI handler to release the slave CPUs for
entry into KGDB.
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Jason Wessel <jason.wessel@windriver.com>
Reviewed-by: Dimitri Sivanich <sivanich@sgi.com>
Reviewed-by: Hedi Berriche <hedi@sgi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Link: http://lkml.kernel.org/r/20131002151417.928886849@asylum.americas.sgi.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull (mostly) ARM clocksource driver updates from Daniel Lezcano:
" - Soren Brinkmann added FEAT_PERCPU to a clock device when it is local
per cpu. This feature prevents the clock framework to choose a per cpu
timer as a broadcast timer. This problem arised when the ARM global
timer is used when switching to the broadcast timer which is the case
now on Xillinx with its cpuidle driver.
- Stephen Boyd extended the generic sched_clock code to support 64bit
counters and removes the setup_sched_clock deprecation, as that causes
lots of warnings since there's still users in the arch/arm tree. He
added also the CLOCK_SOURCE_SUSPEND_NONSTOP flag on the architected
timer as they continue counting during suspend.
- Uwe Kleine-König added some missing __init sections and consolidated the
code by moving the of_node_put call from the drivers to the function
clocksource_of_init. "
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge updated full dynticks support from Frederic Weisbecker:
- support 32-bit systems (full dynticks was 64-bit only before)
- support ARM
Signed-off-by: Ingo Molnar <mingo@kernel.org>
