Currently no-idle queues in cfq are not serviced fairly:
even if they can only dispatch a small number of requests at a time,
they have to compete with idling queues to be serviced, experiencing
large latencies.
We should notice, instead, that no-idle queues are the ones that would
benefit most from having low latency, in fact they are any of:
* processes with large think times (e.g. interactive ones like file
managers)
* seeky (e.g. programs faulting in their code at startup)
* or marked as no-idle from upper levels, to improve latencies of those
requests.
This patch improves the fairness and latency for those queues, by:
* separating sync idle, sync no-idle and async queues in separate
service_trees, for each priority
* service all no-idle queues together
* and idling when the last no-idle queue has been serviced, to
anticipate for more no-idle work
* the timeslices allotted for idle and no-idle service_trees are
computed proportionally to the number of processes in each set.
Servicing all no-idle queues together should have a performance boost
for NCQ-capable drives, without compromising fairness.
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
cfq can disable idling for queues in various circumstances.
When workloads of different priorities are competing, if the higher
priority queue has idling disabled, lower priority queues may steal
its disk share. For example, in a scenario with an RT process
performing seeky reads vs a BE process performing sequential reads,
on an NCQ enabled hardware, with low_latency unset,
the RT process will dispatch only the few pending requests every full
slice of service for the BE process.
The patch solves this issue by always performing idle on the last
queue at a given priority class > idle. If the same process, or one
that can pre-empt it (so at the same priority or higher), submits a
new request within the idle window, the lower priority queue won't
dispatch, saving the disk bandwidth for higher priority ones.
Note: this doesn't touch the non_rotational + NCQ case (no hardware
to test if this is a benefit in that case).
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
We use different service trees for different priority classes.
This allows a simplification in the service tree insertion code, that no
longer has to consider priority while walking the tree.
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
We embed a pointer to the service tree in each queue, to handle multiple
service trees easily.
Service trees are enriched with a counter.
cfq_add_rq_rb is invoked after putting the rq in the fifo, to ensure
that all fields in rq are properly initialized.
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
When the number of processes performing I/O concurrently increases,
a fixed time slice per process will cause large latencies.
This patch, if low_latency mode is enabled, will scale the time slice
assigned to each process according to a 300ms target latency.
In order to keep fairness among processes:
* The number of active processes is computed using a special form of
running average, that quickly follows sudden increases (to keep latency low),
and decrease slowly (to have fairness in spite of rapid decreases of this
value).
To safeguard sequential bandwidth, we impose a minimum time slice
(computed using 2*cfq_slice_idle as base, adjusted according to priority
and async-ness).
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
If active queue hasn't enough requests and idle window opens, cfq will not
dispatch sufficient requests to hardware. In such situation, current code
will zero hw_tag. But this is because cfq doesn't dispatch enough requests
instead of hardware queue doesn't work. Don't zero hw_tag in such case.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
cfq_queues are merged if they are issuing requests within the mean seek
distance of one another. This patch detects when the coopearting stops and
breaks the queues back up.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
The flag used to indicate that a cfqq was allowed to jump ahead in the
scheduling order due to submitting a request close to the queue that
just executed. Since closely cooperating queues are now merged, the flag
holds little meaning. Change it to indicate that multiple queues were
merged. This will later be used to allow the breaking up of merged queues
when they are no longer cooperating.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
When cooperating cfq_queues are detected currently, they are allowed to
skip ahead in the scheduling order. It is much more efficient to
automatically share the cfq_queue data structure between cooperating processes.
Performance of the read-test2 benchmark (which is written to emulate the
dump(8) utility) went from 12MB/s to 90MB/s on my SATA disk. NFS servers
with multiple nfsd threads also saw performance increases.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
async cfq_queue's are already shared between processes within the same
priority, and forthcoming patches will change the mapping of cic to sync
cfq_queue from 1:1 to 1:N. So, calculate the seekiness of a process
based on the cfq_queue instead of the cfq_io_context.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Add cciss_allow_hpsa module parameter. This parameter causes
the cciss driver to ignore any Smart Array devices known to be
supported by the hpsa driver.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Fix multiple calls to pci_release_regions. If cciss_pci_init
fails, it already does any necessary call to pci_release_regions,
so this does not need to be done again in cciss_init_one in that
case.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
It makes sense to do IOWAIT when someone is blocked
due to IO throttle, as suggested by Kame and Peter.
There is an old comment for not doing IOWAIT on throttle,
however it has been mismatching the code for a long time.
If we stop accounting IOWAIT for 2.6.32, it could be an
undesirable behavior change. So restore the io_schedule.
CC: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
If the average think time is larger than the remaining time slice
for any given queue, don't allow it to idle. A succesful idle also
means that we need to dispatch and complete a request, so if we don't
even have time left for the idle process, we would overrun the slice
in any case.
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Saves 16 bytes of text, woohoo. But the more important point is
that it makes the code more readable when returning bool for 0/1
cases.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
CFQ enables idle only for processes that think less than the allowed
idle time. Since idle time is lower for seeky queues, we should use the
correct value in the comparison.
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
drbd_int.h uses __ratelimit(), so it needs to #include ratelimit.h:
drivers/block/drbd/drbd_int.h:1765: error: implicit declaration of function '__ratelimit'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: drbd-dev@lists.linbit.com
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
We should subtract the slice residual from the rb tree key, since
a negative residual count indicates that the cfqq overran its slice
the last time. Hence we want to add the overrun time, to position
it a bit further away in the service tree.
Reported-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Commit a9327cac44 added seperate read
and write statistics of in_flight requests. And exported the number
of read and write requests in progress seperately through sysfs.
But Corrado Zoccolo <czoccolo@gmail.com> reported getting strange
output from "iostat -kx 2". Global values for service time and
utilization were garbage. For interval values, utilization was always
100%, and service time is higher than normal.
So this was reverted by commit 0f78ab9899
The problem was in part_round_stats_single(), I missed the following:
if (now == part->stamp)
return;
- if (part->in_flight) {
+ if (part_in_flight(part)) {
__part_stat_add(cpu, part, time_in_queue,
part_in_flight(part) * (now - part->stamp));
__part_stat_add(cpu, part, io_ticks, (now - part->stamp));
With this chunk included, the reported regression gets fixed.
Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>
--
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Now we have the capabilities of the sending process available,
use them to enforce CAP_SYS_ADMIN.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
It was briefly introduced to allow CFQ to to delayed scheduling,
but we ended up removing that feature again. So lets kill the
function and export, and just switch CFQ back to the normal work
schedule since it is now passing in a '0' delay from all call
sites.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
The RR service tree is indexed by a key that is relative to current jiffies.
This can cause problems on jiffies wraparound.
The patch fixes it using time_before comparison, and changing
the add_front path to use a relative number, too.
Signed-off-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
cfq uses rq->start_time as the fifo indicator, but that field may
get modified prior to cfq doing it's fifo list adjustment when
a request gets merged with another request. This can cause the
fifo list to become unordered.
Reported-by: Corrado Zoccolo <czoccolo@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
I'm skipping -rc2 because the -rc1 Makefile mistakenly said -rc2, so in
order to avoid confusion, I'm jumping from -rc1 to -rc3. That way, when
'uname' (or an oops report) says 2.6.32-rc2, there's no confusion about
whether people perhaps meant -rc1 or -rc2.
* 'acpi-pad' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
acpi_pad: build only on X86
ACPI: create Processor Aggregator Device driver
Fixup trivial conflicts in MAINTAINERS file.
* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
ACPI: EC: Don't parse DSDT for EC early init on Compal
ACPI: EC: Rewrite DMI checks
ACPI: dock: fix "sibiling" typo
ACPI: kill overly verbose "throttling states" log messages
ACPI: Fix bound checks for copy_from_user in the acpi /proc code
ACPI: fix bus scanning memory leaks
ACPI: EC: Restart command even if no interrupts from EC
sony-laptop: Don't unregister the SPIC driver if it wasn't registered
sony-laptop: remove _INI call at init time
sony-laptop: SPIC unset IRQF_SHARED, set IRQF_DISABLED
sony-laptop: remove device_ctrl and the SPIC mini drivers
If i2c device probing fails, then there is no driver to dereference
after calling i2c_new_device(). Stop assuming that probing will always
succeed, to avoid NULL pointer dereferences. We have an easier access
to the driver anyway.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Tested-by: Tim Shepard <shep@alum.mit.edu>
Cc: Colin Leroy <colin@colino.net>
When an ACPI resource conflict is detected, error messages are already
printed by ACPI. There's no point in causing the driver core to print
more error messages, so return one of the error codes for which no
message is printed.
This fixes bug #14293:
http://bugzilla.kernel.org/show_bug.cgi?id=14293
Signed-off-by: Jean Delvare <khali@linux-fr.org>
The I2C_CLIENT_INSMOD_1 macro is only useful for i2c drivers which
implement device detection. The ab3100 driver doesn't, so there is no
point in calling it.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Cc: Samuel Ortiz <sameo@linux.intel.com>
The I2C_CLIENT_INSMOD macro is only useful for i2c drivers which
implement device detection. The tsl2561 driver doesn't, so there
is no point in calling it.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Jonathan Cameron <jic23@cam.ac.uk>
The I2C_CLIENT_INSMOD_1 macro is only useful for i2c drivers which
implement device detection. The leds-pca9532 driver doesn't, so there
is no point in calling it.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Richard Purdie <rpurdie@rpsys.net>
Cc: Riku Voipio <riku.voipio@iki.fi>
There is no point in implementing a detect callback for the LTC4215
and LTC4245, as these devices can't be detected. It was there solely
to handle "force" module parameters to instantiate devices, but now
we have a better sysfs interface that can do the same.
So we can get rid of the ugly module parameters and the detect
callbacks. This shrinks the binary module sizes by 36% and 46%,
respectively.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ira W. Snyder <iws@ovro.caltech.edu>
There is no point in implementing a detect callback for the DS2482, as
this device can't be detected. It was there solely to handle "force"
module parameters to instantiate devices, but now we have a better sysfs
interface that can do the same.
So we can get rid of the ugly module parameters and the detect callback.
This shrinks the binary module size by 21%.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Ben Gardner <gardner.ben@gmail.com>
There is no point in implementing a detect callback for the MAX6875, as
this device can't be detected. It was there solely to handle "force"
module parameters to instantiate devices, but now we have a better sysfs
interface that can do the same.
So we can get rid of the ugly module parameters and the detect callback.
This basically divides the binary module size by 2.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Wolfram Sang <w.sang@pengutronix.de>
Acked-by: Ben Gardner <gardner.ben@gmail.com>
Some times ago the eeprom and max6875 drivers moved to
drivers/misc/eeprom, but their documentation did not follow. It's
finally time to get rid of Documentation/i2c/chips.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ben Gardner <gardner.ben@gmail.com>
Acked-by: Wolfram Sang <w.sang@pengutronix.de>
* 'for-linus' of git://git.kernel.dk/linux-2.6-block: (41 commits)
Revert "Seperate read and write statistics of in_flight requests"
cfq-iosched: don't delay async queue if it hasn't dispatched at all
block: Topology ioctls
cfq-iosched: use assigned slice sync value, not default
cfq-iosched: rename 'desktop' sysfs entry to 'low_latency'
cfq-iosched: implement slower async initiate and queue ramp up
cfq-iosched: delay async IO dispatch, if sync IO was just done
cfq-iosched: add a knob for desktop interactiveness
Add a tracepoint for block request remapping
block: allow large discard requests
block: use normal I/O path for discard requests
swapfile: avoid NULL pointer dereference in swapon when s_bdev is NULL
fs/bio.c: move EXPORT* macros to line after function
Add missing blk_trace_remove_sysfs to be in pair with blk_trace_init_sysfs
cciss: fix build when !PROC_FS
block: Do not clamp max_hw_sectors for stacking devices
block: Set max_sectors correctly for stacking devices
cciss: cciss_host_attr_groups should be const
cciss: Dynamically allocate the drive_info_struct for each logical drive.
cciss: Add usage_count attribute to each logical drive in /sys
...
We cannot delay for the first dispatch of the async queue if it
hasn't dispatched at all, since that could present a local user
DoS attack vector using an app that just did slow timed sync reads
while filling memory.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Corrado Zoccolo