| Age | Commit message (Collapse) | Author |
|---|
|
Use the generic llist primitives.
We had a private lockless list implementation in the scheduler in the wake-list
code, now that we have a generic llist implementation that provides all required
operations, switch to it.
This patch is not expected to change any behavior.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1315836353.26517.42.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
So we don't have to expose the struct list_node member.
Cc: Huang Ying <ying.huang@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315836348.26517.41.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Use llist in irq_work instead of the lock-less linked list
implementation in irq_work to avoid the code duplication.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315461646-1379-6-git-send-email-ying.huang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Extend the llist_add*() functions to return a success indicator, this
allows us in the scheduler code to send an IPI if the queue was empty.
( There's no effect on existing users, because the list_add_xxx() functions
are inline, thus this will be optimized out by the compiler if not used
by callers. )
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315461646-1379-5-git-send-email-ying.huang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
If in llist_add()/etc. functions the first cmpxchg() call succeeds, it is
not necessary to use cpu_relax() before the cmpxchg(). So cpu_relax() in
a busy loop involving cmpxchg() should go after cmpxchg() instead of before
that.
This patch fixes this for all involved llist functions.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315461646-1379-4-git-send-email-ying.huang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Remove the nmi() checks spread around the code. in_nmi() is not available
on every architecture and it's a pretty obscure and ugly check in any case.
Cc: Huang Ying <ying.huang@intel.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315461646-1379-3-git-send-email-ying.huang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
In the pNFS obj-LD the device table at the layout level needs
to point to a device_cache node, where it is possible and likely
that many layouts will point to the same device-nodes.
In Exofs we have a more orderly structure where we have a single
array of devices that repeats twice for a round-robin view of the
device table
This patch moves to a model that can be used by the pNFS obj-LD
where struct ore_components holds an array of ore_dev-pointers.
(ore_dev is newly defined and contains a struct osd_dev *od
member)
Each pointer in the array of pointers will point to a bigger
user-defined dev_struct. That can be accessed by use of the
container_of macro.
In Exofs an __alloc_dev_table() function allocates the
ore_dev-pointers array as well as an exofs_dev array, in one
allocation and does the addresses dance to set everything pointing
correctly. It still keeps the double allocation trick for the
inodes round-robin view of the table.
The device table is always allocated dynamically, also for the
single device case. So it is unconditionally freed at umount.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
|
|
Because llist code will be used in performance critical scheduler
code path, make llist_add() and llist_del_all() inline to avoid
function calling overhead and related 'glue' overhead.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1315461646-1379-2-git-send-email-ying.huang@intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Merge reason: pick up the latest fixes.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Each event adds some points to its counters. By default it adds 1,
and a number of points may be transmited in event's parameters.
E.g. sched:sched_stat_runtime adds how long process has been running.
But this functionality was broken by v2.6.31-rc5-392-gf413cdb
and now the event's parameters doesn't affect on a number of points.
TP_perf_assign isn't defined, so __perf_count(c) isn't executed and
__count is always equal to 1.
Signed-off-by: Andrew Vagin <avagin@openvz.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1317052535-1765247-2-git-send-email-avagin@openvz.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
tx params should be configured per interface.
add ieee80211_vif param to the conf_tx callback,
and change all the drivers that use this callback.
The following spatch was used:
@rule1@
struct ieee80211_ops ops;
identifier conf_tx_op;
@@
ops.conf_tx = conf_tx_op;
@rule2@
identifier rule1.conf_tx_op;
identifier hw, queue, params;
@@
conf_tx_op (
- struct ieee80211_hw *hw,
+ struct ieee80211_hw *hw, struct ieee80211_vif *vif,
u16 queue,
const struct ieee80211_tx_queue_params *params) {...}
Signed-off-by: Eliad Peller <eliad@wizery.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
Recently mac80211 was changed to use nullfunc instead of probe
request for connection monitoring for tx ack status reporting
hardwares. Sometimes in congested network, STA got disconnected
quickly after the association. It was observered that the rate
control was not adopted to environment due to minimal transmission.
As the nullfunc are used for monitoring purpose, these frames should
not be sacrificed for rate control updation. So it is better to send
the monitoring null func frames at minimum rate that could help to
retain the connection.
Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
Add a gpio setup function which gives a chance to set up
platform specific configuration such as pin multiplexing,
input/output direction at the runtime or booting time.
Signed-off-by: Sangwook Lee <sangwook.lee@linaro.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
Allow injected unicast frames to be sent without having to wait
for an ACK.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
removing obsoleted sysctl,
ip_rt_gc_interval variable no longer used since 2.6.38
Signed-off-by: Vasily Averin <vvs@sw.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This repairs problem with compile library in userspace (libnl).
Signed-off-by: Jiří Župka <jzupka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Allows ss command (iproute2) to display "ecnseen" if at least one packet
with ECT(0) or ECT(1) or ECN was received by this socket.
"ecn" means ECN was negotiated at session establishment (TCP level)
"ecnseen" means we received at least one packet with ECT fields set (IP
level)
ss -i
...
ESTAB 0 0 192.168.20.110:22 192.168.20.144:38016
ino:5950 sk:f178e400
mem:(r0,w0,f0,t0) ts sack ecn ecnseen bic wscale:7,8 rto:210
rtt:12.5/7.5 cwnd:10 send 9.3Mbps rcv_space:14480
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The struct ore_striping_info will be used later in other
structures. And ore_calc_stripe_info as well. Rename them
make struct ore_striping_info public. ore_calc_stripe_info
is still static, will be made public on first use.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
|
|
The struct pnfs_osd_data_map data_map member of exofs_sb_info was
never used after mount. In fact all it's members were duplicated
by the ore_layout structure. So just remove the duplicated information.
Also removed some stupid, but perfectly supported, restrictions on
layout parameters. The case where num_devices is not divisible by
mirror_count+1 is perfectly fine since the rotating device view
will eventually use all the devices it can get.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: Benny Halevy <bhalevy@tonian.com>
|
|
ore_components already has a comps member so this leads
to things like comps->comps which is annoying. the name oc
was already used in new code. So rename all old usage of
ore_components comps => ore_components oc.
Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
|
|
Add zorder support on OMAP4, this feature allows deciding the visibility order
of the overlays based on the zorder value provided as an overlay info parameter
or a sysfs attribute of the overlay object.
Use the overlay cap OMAP_DSS_OVL_CAP_ZORDER to determine whether zorder is
supported for the overlay or not. Use dss feature FEAT_ALPHA_FREE_ZORDER
if the caps are not available.
Ensure that all overlays that are enabled and connected to the same manager
have different zorders. Swapping zorders of 2 enabled overlays currently
requires disabling one of the overlays.
Signed-off-by: Archit Taneja <archit@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
|
|
Add support for VIDEO3 pipeline on OMAP4:
- Add VIDEO3 pipeline information in dss_features and omapdss.h
- Add VIDEO3 pipeline register coefficients in dispc.h
- Create a new overlay structure corresponding to VIDEO3.
- Make changes in dispc.c for VIDEO3
Signed-off-by: Archit Taneja <archit@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
|
|
On OMAP3, in order to enable alpha blending for LCD and TV managers, we needed
to set LCDALPHABLENDERENABLE/TVALPHABLENDERENABLE bits in DISPC_CONFIG. On
OMAP4, alpha blending is always enabled by default, if the above bits are set,
we switch to an OMAP3 compatibility mode where the zorder values in the pipeline
attribute registers are ignored and a fixed priority is configured.
Rename the manager_info member "alpha_enabled" to "partial_alpha_enabled" for
more clarity. Introduce two dss_features FEAT_ALPHA_FIXED_ZORDER and
FEAT_ALPHA_FREE_ZORDER which represent OMAP3-alpha compatibility mode and OMAP4
alpha mode respectively. Introduce an overlay cap for ZORDER. The DSS2 user is
expected to check for the ZORDER cap, if an overlay doesn't have this cap, the
user is expected to set the parameter partial_alpha_enabled. If the overlay has
ZORDER cap, the DSS2 user can assume that alpha blending is already enabled.
Don't support OMAP3 compatibility mode for now. Trying to read/write to
alpha_blending_enabled sysfs attribute issues a warning for OMAP4 and does not
set the LCDALPHABLENDERENABLE/TVALPHABLENDERENABLE bits.
Change alpha_enabled to partial_alpha_enabled in the omap_vout driver. Use
overlay cap "OMAP_DSS_OVL_CAP_GLOBAL_ALPHA" to check if overlay supports alpha
blending or not. Replace this with checks for VIDEO1 pipeline.
Cc: linux-media@vger.kernel.org
Cc: Lajos Molnar <molnar@ti.com>
Signed-off-by: Archit Taneja <archit@ti.com>
Acked-by: Vaibhav Hiremath <hvaibhav@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
|
|
As request_percpu_irq() doesn't allow for a percpu interrupt to have
its type configured (it is generally impossible to configure it on all
CPUs at once), add a 'type' argument to enable_percpu_irq().
This allows some low-level, board specific init code to be switched to
a generic API.
[ tglx: Added WARN_ON argument ]
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Cc: Abhijeet Dharmapurikar <adharmap@codeaurora.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
|
The ARM GIC interrupt controller offers per CPU interrupts (PPIs),
which are usually used to connect local timers to each core. Each CPU
has its own private interface to the GIC, and only sees the PPIs that
are directly connect to it.
While these timers are separate devices and have a separate interrupt
line to a core, they all use the same IRQ number.
For these devices, request_irq() is not the right API as it assumes
that an IRQ number is visible by a number of CPUs (through the
affinity setting), but makes it very awkward to express that an IRQ
number can be handled by all CPUs, and yet be a different interrupt
line on each CPU, requiring a different dev_id cookie to be passed
back to the handler.
The *_percpu_irq() functions is designed to overcome these
limitations, by providing a per-cpu dev_id vector:
int request_percpu_irq(unsigned int irq, irq_handler_t handler,
const char *devname, void __percpu *percpu_dev_id);
void free_percpu_irq(unsigned int, void __percpu *);
int setup_percpu_irq(unsigned int irq, struct irqaction *new);
void remove_percpu_irq(unsigned int irq, struct irqaction *act);
void enable_percpu_irq(unsigned int irq);
void disable_percpu_irq(unsigned int irq);
The API has a number of limitations:
- no interrupt sharing
- no threading
- common handler across all the CPUs
Once the interrupt is requested using setup_percpu_irq() or
request_percpu_irq(), it must be enabled by each core that wishes its
local interrupt to be delivered.
Based on an initial patch by Thomas Gleixner.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/1316793788-14500-2-git-send-email-marc.zyngier@arm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
|
As proposed by Chris, Dave and Jan, don't start foreground writeback IO
inside balance_dirty_pages(). Instead, simply let it idle sleep for some
time to throttle the dirtying task. In the mean while, kick off the
per-bdi flusher thread to do background writeback IO.
RATIONALS
=========
- disk seeks on concurrent writeback of multiple inodes (Dave Chinner)
If every thread doing writes and being throttled start foreground
writeback, it leads to N IO submitters from at least N different
inodes at the same time, end up with N different sets of IO being
issued with potentially zero locality to each other, resulting in
much lower elevator sort/merge efficiency and hence we seek the disk
all over the place to service the different sets of IO.
OTOH, if there is only one submission thread, it doesn't jump between
inodes in the same way when congestion clears - it keeps writing to
the same inode, resulting in large related chunks of sequential IOs
being issued to the disk. This is more efficient than the above
foreground writeback because the elevator works better and the disk
seeks less.
- lock contention and cache bouncing on concurrent IO submitters (Dave Chinner)
With this patchset, the fs_mark benchmark on a 12-drive software RAID0 goes
from CPU bound to IO bound, freeing "3-4 CPUs worth of spinlock contention".
* "CPU usage has dropped by ~55%", "it certainly appears that most of
the CPU time saving comes from the removal of contention on the
inode_wb_list_lock" (IMHO at least 10% comes from the reduction of
cacheline bouncing, because the new code is able to call much less
frequently into balance_dirty_pages() and hence access the global
page states)
* the user space "App overhead" is reduced by 20%, by avoiding the
cacheline pollution by the complex writeback code path
* "for a ~5% throughput reduction", "the number of write IOs have
dropped by ~25%", and the elapsed time reduced from 41:42.17 to
40:53.23.
* On a simple test of 100 dd, it reduces the CPU %system time from 30% to 3%,
and improves IO throughput from 38MB/s to 42MB/s.
- IO size too small for fast arrays and too large for slow USB sticks
The write_chunk used by current balance_dirty_pages() cannot be
directly set to some large value (eg. 128MB) for better IO efficiency.
Because it could lead to more than 1 second user perceivable stalls.
Even the current 4MB write size may be too large for slow USB sticks.
The fact that balance_dirty_pages() starts IO on itself couples the
IO size to wait time, which makes it hard to do suitable IO size while
keeping the wait time under control.
Now it's possible to increase writeback chunk size proportional to the
disk bandwidth. In a simple test of 50 dd's on XFS, 1-HDD, 3GB ram,
the larger writeback size dramatically reduces the seek count to 1/10
(far beyond my expectation) and improves the write throughput by 24%.
- long block time in balance_dirty_pages() hurts desktop responsiveness
Many of us may have the experience: it often takes a couple of seconds
or even long time to stop a heavy writing dd/cp/tar command with
Ctrl-C or "kill -9".
- IO pipeline broken by bumpy write() progress
There are a broad class of "loop {read(buf); write(buf);}" applications
whose read() pipeline will be under-utilized or even come to a stop if
the write()s have long latencies _or_ don't progress in a constant rate.
The current threshold based throttling inherently transfers the large
low level IO completion fluctuations to bumpy application write()s,
and further deteriorates with increasing number of dirtiers and/or bdi's.
For example, when doing 50 dd's + 1 remote rsync to an XFS partition,
the rsync progresses very bumpy in legacy kernel, and throughput is
improved by 67% by this patchset. (plus the larger write chunk size,
it will be 93% speedup).
The new rate based throttling can support 1000+ dd's with excellent
smoothness, low latency and low overheads.
For the above reasons, it's much better to do IO-less and low latency
pauses in balance_dirty_pages().
Jan Kara, Dave Chinner and me explored the scheme to let
balance_dirty_pages() wait for enough writeback IO completions to
safeguard the dirty limit. However it's found to have two problems:
- in large NUMA systems, the per-cpu counters may have big accounting
errors, leading to big throttle wait time and jitters.
- NFS may kill large amount of unstable pages with one single COMMIT.
Because NFS server serves COMMIT with expensive fsync() IOs, it is
desirable to delay and reduce the number of COMMITs. So it's not
likely to optimize away such kind of bursty IO completions, and the
resulted large (and tiny) stall times in IO completion based throttling.
So here is a pause time oriented approach, which tries to control the
pause time in each balance_dirty_pages() invocations, by controlling
the number of pages dirtied before calling balance_dirty_pages(), for
smooth and efficient dirty throttling:
- avoid useless (eg. zero pause time) balance_dirty_pages() calls
- avoid too small pause time (less than 4ms, which burns CPU power)
- avoid too large pause time (more than 200ms, which hurts responsiveness)
- avoid big fluctuations of pause times
It can control pause times at will. The default policy (in a followup
patch) will be to do ~10ms pauses in 1-dd case, and increase to ~100ms
in 1000-dd case.
BEHAVIOR CHANGE
===============
(1) dirty threshold
Users will notice that the applications will get throttled once crossing
the global (background + dirty)/2=15% threshold, and then balanced around
17.5%. Before patch, the behavior is to just throttle it at 20% dirtyable
memory in 1-dd case.
Since the task will be soft throttled earlier than before, it may be
perceived by end users as performance "slow down" if his application
happens to dirty more than 15% dirtyable memory.
(2) smoothness/responsiveness
Users will notice a more responsive system during heavy writeback.
"killall dd" will take effect instantly.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
|
|
Add two fields to task_struct.
1) account dirtied pages in the individual tasks, for accuracy
2) per-task balance_dirty_pages() call intervals, for flexibility
The balance_dirty_pages() call interval (ie. nr_dirtied_pause) will
scale near-sqrt to the safety gap between dirty pages and threshold.
The main problem of per-task nr_dirtied is, if 1k+ tasks start dirtying
pages at exactly the same time, each task will be assigned a large
initial nr_dirtied_pause, so that the dirty threshold will be exceeded
long before each task reached its nr_dirtied_pause and hence call
balance_dirty_pages().
The solution is to watch for the number of pages dirtied on each CPU in
between the calls into balance_dirty_pages(). If it exceeds ratelimit_pages
(3% dirty threshold), force call balance_dirty_pages() for a chance to
set bdi->dirty_exceeded. In normal situations, this safeguarding
condition is not expected to trigger at all.
On the sqrt in dirty_poll_interval():
It will serve as an initial guess when dirty pages are still in the
freerun area.
When dirty pages are floating inside the dirty control scope [freerun,
limit], a followup patch will use some refined dirty poll interval to
get the desired pause time.
thresh-dirty (MB) sqrt
1 16
2 22
4 32
8 45
16 64
32 90
64 128
128 181
256 256
512 362
1024 512
The above table means, given 1MB (or 1GB) gap and the dd tasks polling
balance_dirty_pages() on every 16 (or 512) pages, the dirty limit won't
be exceeded as long as there are less than 16 (or 512) concurrent dd's.
So sqrt naturally leads to less overheads and more safe concurrent tasks
for large memory servers, which have large (thresh-freerun) gaps.
peter: keep the per-CPU ratelimit for safeguarding the 1k+ tasks case
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: Andrea Righi <andrea@betterlinux.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
|
|
There are some imperfections in balanced_dirty_ratelimit.
1) large fluctuations
The dirty_rate used for computing balanced_dirty_ratelimit is merely
averaged in the past 200ms (very small comparing to the 3s estimation
period for write_bw), which makes rather dispersed distribution of
balanced_dirty_ratelimit.
It's pretty hard to average out the singular points by increasing the
estimation period. Considering that the averaging technique will
introduce very undesirable time lags, I give it up totally. (btw, the 3s
write_bw averaging time lag is much more acceptable because its impact
is one-way and therefore won't lead to oscillations.)
The more practical way is filtering -- most singular
balanced_dirty_ratelimit points can be filtered out by remembering some
prev_balanced_rate and prev_prev_balanced_rate. However the more
reliable way is to guard balanced_dirty_ratelimit with task_ratelimit.
2) due to truncates and fs redirties, the (write_bw <=> dirty_rate)
match could become unbalanced, which may lead to large systematical
errors in balanced_dirty_ratelimit. The truncates, due to its possibly
bumpy nature, can hardly be compensated smoothly. So let's face it. When
some over-estimated balanced_dirty_ratelimit brings dirty_ratelimit
high, dirty pages will go higher than the setpoint. task_ratelimit will
in turn become lower than dirty_ratelimit. So if we consider both
balanced_dirty_ratelimit and task_ratelimit and update dirty_ratelimit
only when they are on the same side of dirty_ratelimit, the systematical
errors in balanced_dirty_ratelimit won't be able to bring
dirty_ratelimit far away.
The balanced_dirty_ratelimit estimation may also be inaccurate near
@limit or @freerun, however is less an issue.
3) since we ultimately want to
- keep the fluctuations of task ratelimit as small as possible
- keep the dirty pages around the setpoint as long time as possible
the update policy used for (2) also serves the above goals nicely:
if for some reason the dirty pages are high (task_ratelimit < dirty_ratelimit),
and dirty_ratelimit is low (dirty_ratelimit < balanced_dirty_ratelimit),
there is no point to bring up dirty_ratelimit in a hurry only to hurt
both the above two goals.
So, we make use of task_ratelimit to limit the update of dirty_ratelimit
in two ways:
1) avoid changing dirty rate when it's against the position control target
(the adjusted rate will slow down the progress of dirty pages going
back to setpoint).
2) limit the step size. task_ratelimit is changing values step by step,
leaving a consistent trace comparing to the randomly jumping
balanced_dirty_ratelimit. task_ratelimit also has the nice smaller
errors in stable state and typically larger errors when there are big
errors in rate. So it's a pretty good limiting factor for the step
size of dirty_ratelimit.
Note that bdi->dirty_ratelimit is always tracking balanced_dirty_ratelimit.
task_ratelimit is merely used as a limiting factor.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
|
|
It's all about bdi->dirty_ratelimit, which aims to be (write_bw / N)
when there are N dd tasks.
On write() syscall, use bdi->dirty_ratelimit
============================================
balance_dirty_pages(pages_dirtied)
{
task_ratelimit = bdi->dirty_ratelimit * bdi_position_ratio();
pause = pages_dirtied / task_ratelimit;
sleep(pause);
}
On every 200ms, update bdi->dirty_ratelimit
===========================================
bdi_update_dirty_ratelimit()
{
task_ratelimit = bdi->dirty_ratelimit * bdi_position_ratio();
balanced_dirty_ratelimit = task_ratelimit * write_bw / dirty_rate;
bdi->dirty_ratelimit = balanced_dirty_ratelimit
}
Estimation of balanced bdi->dirty_ratelimit
===========================================
balanced task_ratelimit
-----------------------
balance_dirty_pages() needs to throttle tasks dirtying pages such that
the total amount of dirty pages stays below the specified dirty limit in
order to avoid memory deadlocks. Furthermore we desire fairness in that
tasks get throttled proportionally to the amount of pages they dirty.
IOW we want to throttle tasks such that we match the dirty rate to the
writeout bandwidth, this yields a stable amount of dirty pages:
dirty_rate == write_bw (1)
The fairness requirement gives us:
task_ratelimit = balanced_dirty_ratelimit
== write_bw / N (2)
where N is the number of dd tasks. We don't know N beforehand, but
still can estimate balanced_dirty_ratelimit within 200ms.
Start by throttling each dd task at rate
task_ratelimit = task_ratelimit_0 (3)
(any non-zero initial value is OK)
After 200ms, we measured
dirty_rate = # of pages dirtied by all dd's / 200ms
write_bw = # of pages written to the disk / 200ms
For the aggressive dd dirtiers, the equality holds
dirty_rate == N * task_rate
== N * task_ratelimit_0 (4)
Or
task_ratelimit_0 == dirty_rate / N (5)
Now we conclude that the balanced task ratelimit can be estimated by
write_bw
balanced_dirty_ratelimit = task_ratelimit_0 * ---------- (6)
dirty_rate
Because with (4) and (5) we can get the desired equality (1):
write_bw
balanced_dirty_ratelimit == (dirty_rate / N) * ----------
dirty_rate
== write_bw / N
Then using the balanced task ratelimit we can compute task pause times like:
task_pause = task->nr_dirtied / task_ratelimit
task_ratelimit with position control
------------------------------------
However, while the above gives us means of matching the dirty rate to
the writeout bandwidth, it at best provides us with a stable dirty page
count (assuming a static system). In order to control the dirty page
count such that it is high enough to provide performance, but does not
exceed the specified limit we need another control.
The dirty position control works by extending (2) to
task_ratelimit = balanced_dirty_ratelimit * pos_ratio (7)
where pos_ratio is a negative feedback function that subjects to
1) f(setpoint) = 1.0
2) df/dx < 0
That is, if the dirty pages are ABOVE the setpoint, we throttle each
task a bit more HEAVY than balanced_dirty_ratelimit, so that the dirty
pages are created less fast than they are cleaned, thus DROP to the
setpoints (and the reverse).
Based on (7) and the assumption that both dirty_ratelimit and pos_ratio
remains CONSTANT for the past 200ms, we get
task_ratelimit_0 = balanced_dirty_ratelimit * pos_ratio (8)
Putting (8) into (6), we get the formula used in
bdi_update_dirty_ratelimit():
write_bw
balanced_dirty_ratelimit *= pos_ratio * ---------- (9)
dirty_rate
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
|
|
No behavior change.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
|
|
Introduce the BDI_DIRTIED counter. It will be used for estimating the
bdi's dirty bandwidth.
CC: Jan Kara <jack@suse.cz>
CC: Michael Rubin <mrubin@google.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
|
|
Based on a patch from Arnd Bergmann this fixes up the build
problem of assigning a non-existing global when the ux500 PRCMU
timer is not linked in by passing its base address to the init
function. We also add a missing <linux/errno.h> inclusion and
staticize the dummy function.
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
|
|
Allow the sas-transport-class to update events for local phys via a new
PHY_FUNC_GET_EVENTS command to ->lldd_control_phy(). Fixup drivers that
are not prepared for new enum phy_func values, and unify
->lldd_control_phy() error codes.
These are the SAS defined phy events that are reported in a
smp-report-phy-error-log command:
* /sys/class/sas_phy/<phyX>/invalid_dword_count
* /sys/class/sas_phy/<phyX>/running_disparity_error_count
* /sys/class/sas_phy/<phyX>/loss_of_dword_sync_count
* /sys/class/sas_phy/<phyX>/phy_reset_problem_count
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
Based on original implementation from Jiangbi Liu and Maciej Trela.
ATAPI transfers happen in two-to-three stages. The two stage atapi
commands are those that include a dma data transfer. The data transfer
portion of these operations is handled by the hardware packet-dma
acceleration. The three-stage commands do not have a data transfer and
are handled without hardware assistance in raw frame mode.
stage1: transmit host-to-device fis to notify the device of an incoming
atapi cdb. Upon reception of the pio-setup-fis repost the task_context
to perform the dma transfer of the cdb+data (go to stage3), or repost
the task_context to transmit the cdb as a raw frame (go to stage 2).
stage2: wait for hardware notification of the cdb transmission and then
go to stage 3.
stage3: wait for the arrival of the terminating device-to-host fis and
terminate the command.
To keep the implementation simple we only support ATAPI packet-dma
protocol (for commands with data) to avoid needing to handle the data
transfer manually (like we do for SATA-PIO). This may affect
compatibility for a small number of devices (see
ATA_HORKAGE_ATAPI_MOD16_DMA).
If the data-transfer underruns, or encounters an error the
device-to-host fis is expected to arrive in the unsolicited frame queue
to pass to libata for disposition. However, in the DONE_UNEXP_FIS (data
underrun) case it appears we need to craft a response. In the
DONE_REG_ERR case we do receive the UF and propagate it to libsas.
Signed-off-by: Maciej Trela <maciej.trela@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
cache aligned xid and ex_lock beside
removing holes.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
Re-arrange its fields to avoid padding and have better
cacheline alignments.
Removed not used start_time, end_time and last_pkt_time
fields.
This all reduced this struct size to 448 from 480 and
that also reduced one cacheline on x86_64 beside
eliminating 8 pads. However kept logical fields together.
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Yi Zou <yi.zou@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
Several sas drivers legitimately check the protocol against the union of
SAS_PROTOCOL_SATA and SAS_PROTOCOL_STP. Provide a SAS_PROTOCOL_STP_ALL
to silence warnings like:
drivers/scsi/pm8001/pm8001_sas.c:438:3: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/mvsas/mv_sas.c:798:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/mvsas/mv_sas.c:1783:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/mvsas/mv_sas.c:1886:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
drivers/scsi/isci/request.c:3565:2: warning: case value ‘5’ not in enumerated type ‘enum sas_protocol’ [-Wswitch]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
If the user has disabled CONFIG_SCSI_SAS_HOST_SMP then libsas drivers
will not be receiving smp-gpio frames and do not need this lookup code.
Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Tested-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
Pass queue_depth change requests to libata, and prevent queue_type
changes for ATA devices.
Otherwise:
1/ we do not honor the libata specific restrictions on the queue depth
2/ libsas drivers that do not set sdev->tagged_supported are unable to
change the queue_depth of ata devices via sysfs
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
Allow expander table-to-table attachments for
expanders that support it.
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
|
|
Four cpufreq-like governors are provided as examples.
powersave: use the lowest frequency possible. The user (device) should
set the polling_ms as 0 because polling is useless for this governor.
performance: use the highest freqeuncy possible. The user (device)
should set the polling_ms as 0 because polling is useless for this
governor.
userspace: use the user specified frequency stored at
devfreq.user_set_freq. With sysfs support in the following patch, a user
may set the value with the sysfs interface.
simple_ondemand: simplified version of cpufreq's ondemand governor.
When a user updates OPP entries (enable/disable/add), OPP framework
automatically notifies devfreq to update operating frequency
accordingly. Thus, devfreq users (device drivers) do not need to update
devfreq manually with OPP entry updates or set polling_ms for powersave
, performance, userspace, or any other "static" governors.
Note that these are given only as basic examples for governors and any
devices with devfreq may implement their own governors with the drivers
and use them.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Mike Turquette <mturquette@ti.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
|
|
With OPPs, a device may have multiple operable frequency and voltage
sets. However, there can be multiple possible operable sets and a system
will need to choose one from them. In order to reduce the power
consumption (by reducing frequency and voltage) without affecting the
performance too much, a Dynamic Voltage and Frequency Scaling (DVFS)
scheme may be used.
This patch introduces the DVFS capability to non-CPU devices with OPPs.
DVFS is a techique whereby the frequency and supplied voltage of a
device is adjusted on-the-fly. DVFS usually sets the frequency as low
as possible with given conditions (such as QoS assurance) and adjusts
voltage according to the chosen frequency in order to reduce power
consumption and heat dissipation.
The generic DVFS for devices, devfreq, may appear quite similar with
/drivers/cpufreq. However, cpufreq does not allow to have multiple
devices registered and is not suitable to have multiple heterogenous
devices with different (but simple) governors.
Normally, DVFS mechanism controls frequency based on the demand for
the device, and then, chooses voltage based on the chosen frequency.
devfreq also controls the frequency based on the governor's frequency
recommendation and let OPP pick up the pair of frequency and voltage
based on the recommended frequency. Then, the chosen OPP is passed to
device driver's "target" callback.
When PM QoS is going to be used with the devfreq device, the device
driver should enable OPPs that are appropriate with the current PM QoS
requests. In order to do so, the device driver may call opp_enable and
opp_disable at the notifier callback of PM QoS so that PM QoS's
update_target() call enables the appropriate OPPs. Note that at least
one of OPPs should be enabled at any time; be careful when there is a
transition.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Mike Turquette <mturquette@ti.com>
Acked-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
|
|
'sched-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip
* 'irq-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
irq: Fix check for already initialized irq_domain in irq_domain_add
irq: Add declaration of irq_domain_simple_ops to irqdomain.h
* 'x86-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
x86/rtc: Don't recursively acquire rtc_lock
* 'sched-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
posix-cpu-timers: Cure SMP wobbles
sched: Fix up wchan borkage
sched/rt: Migrate equal priority tasks to available CPUs
|
|
|
|
The patch enables to register notifier_block for an OPP-device in order
to get notified for any changes in the availability of OPPs of the
device. For example, if a new OPP is inserted or enable/disable status
of an OPP is changed, the notifier is executed.
This enables the usage of opp_add, opp_enable, and opp_disable to
directly take effect with any connected entities such as cpufreq or
devfreq.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Mike Turquette <mturquette@ti.com>
Reviewed-by: Kevin Hilman <khilman@ti.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
|
|
With the addition of uAPSD and driver buffering
the powersave handling has gotten quite complex.
Add a section to the documentation to explain it
for anyone wanting to implement it.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
iwlwifi has a separate EOSP notification from
the device, and to make use of that properly
it needs to be passed to mac80211. To be able
to mix with tx_status_irqsafe and rx_irqsafe
it also needs to be an "_irqsafe" version in
the sense that it goes through the tasklet,
the actual flag clearing would be IRQ-safe
but doing it directly would cause reordering
issues.
This is needed in the case of a P2P GO going
into an absence period without transmitting
any frames that should be driver-released as
in this case there's no other way to inform
mac80211 that the service period ended. Note
that for drivers that don't use the _irqsafe
functions another version of this function
will be required.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
iwlwifi needs to know the number of frames that are
going to be sent to a station while it is asleep so
it can properly handle the uCode blocking of that
station.
Before uAPSD, we got by by telling the device that
a single frame was going to be released whenever we
encountered IEEE80211_TX_CTL_POLL_RESPONSE. With
uAPSD, however, that is no longer possible since
there could be more than a single frame.
To support this model, add a new callback to notify
drivers when frames are going to be released.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
If a PS-poll frame is retried (but was received)
there is no way to detect that since it has no
sequence number. As a consequence, the standard
asks us to not react to PS-poll frames until the
response to one made it out (was ACKed or lost).
Implement this by using the WLAN_STA_SP flags to
also indicate a PS-Poll "service period" and the
IEEE80211_TX_STATUS_EOSP flag for the response
packet to indicate the end of the "SP" as usual.
We could use separate flags, but that will most
likely completely confuse drivers, and while the
standard doesn't exclude simultaneously polling
using uAPSD and PS-Poll, doing that seems quite
problematic.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
|
Add uAPSD support to mac80211. This is probably not
possible with all devices, so advertising it with
the cfg80211 flag will be left up to drivers that
want it.
Due to my previous patches it is now a fairly
straight-forward extension. Drivers need to have
accurate TX status reporting for the EOSP frame.
For drivers that buffer themselves, the provided
APIs allow releasing the right number of frames,
but then drivers need to set EOSP and more-data
themselves. This is documented in more detail in
the new code itself.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
|
