- Move all scheduler locking into the schedulers utilizing a technique
similar to solaris's container locking.
- A per-process spinlock is now used to protect the queue of threads,
thread count, suspension count, p_sflags, and other process
related scheduling fields.
- The new thread lock is actually a pointer to a spinlock for the
container that the thread is currently owned by. The container may
be a turnstile, sleepqueue, or run queue.
- thread_lock() is now used to protect access to thread related scheduling
fields. thread_unlock() unlocks the lock and thread_set_lock()
implements the transition from one lock to another.
- A new "blocked_lock" is used in cases where it is not safe to hold the
actual thread's lock yet we must prevent access to the thread.
- sched_throw() and sched_fork_exit() are introduced to allow the
schedulers to fix-up locking at these points.
- Add some minor infrastructure for optionally exporting scheduler
statistics that were invaluable in solving performance problems with
this patch. Generally these statistics allow you to differentiate
between different causes of context switches.
Tested by: kris, current@
Tested on: i386, amd64, ULE, 4BSD, libthr, libkse, PREEMPTION, etc.
Discussed with: kris, attilio, kmacy, jhb, julian, bde (small parts each)
unsigned char. Weirdly, casting the 1 constant to u_char still produces
a signed integer result that is then used in the % computation. This
avoids that mess all together and causes a 0 pri to turn into 255 % 64
as we expect.
Reported by: kkenn (about 4 times, thanks)
a thread is an idle thread, just see if it has the IDLETD
flag set. That flag will probably move to the pflags word
as it's permenent and never chenges for the life of the
system so it doesn't need locking.
- only collect timestamps when a lock is contested - this reduces the overhead
of collecting profiles from 20x to 5x
- remove unused function from subr_lock.c
- generalize cnt_hold and cnt_lock statistics to be kept for all locks
- NOTE: rwlock profiling generates invalid statistics (and most likely always has)
someone familiar with that should review
- Fix these types in ULE as well. This fixes bugs in priority index
calculations in certain edge cases. (int)-1 % 64 != (uint)-1 % 64.
Reported by: kkenn using pho's stress2.
minimize IPIs and rescheduling when scheduling like tasks while keeping
latency low for important threads.
1) An idle thread is running.
2) The current thread is worse than realtime and the new thread is
better than realtime. Realtime to realtime doesn't preempt.
3) The new thread's priority is less than the threshold.
support sched_4bsd.
- Rename the KTR level for non schedgraph parsed events. They take event
space from things we'd like to graph.
- Reset our slice value after we sleep. The slice is simply there to
prevent starvation among equal priorities. A thread which had almost
exhausted it's slice and then slept doesn't need to be rescheduled a
tick after it wakes up.
- Set the maximum slice value to a more conservative 100ms now that it is
more accurately enforced.
negative. Use unsigned integers for sleep and run time so this doesn't
disturb sched_interact_score(). This should fix the invalid interactive
priority panics reported by several users.
- Define our own maybe_preempt() as sched_preempt(). We want to be able
to preempt idlethread in all cases.
- Define our idlethread to require preemption to exit.
- Get the cpu estimation tick from sched_tick() so we don't have to worry
about errors from a sampling interval that differs from the time
domain. This was the source of sched_priority prints/panics and
inaccurate pctcpu display in top.
a power saving mode otherwise.
- If the thread is already bound in sched_bind() unbind it before
re-binding it to a new cpu. I don't like these semantics but they are
expected by some code in the tree. Patch by jkoshy.
the ipi settings. If NEEDRESCHED is set and an ipi is later delivered
it will clear it rather than cause extra context switches. However, if
we miss setting it we can have terrible latency.
- In sched_bind() correctly implement bind. Also be slightly more
tolerant of code which calls bind multiple times. However, we don't
change binding if another call is made with a different cpu. This
does not presently work with hwpmc which I believe should be changed.
- Switch back to direct modification of remote CPU run queues. This added
a lot of complexity with questionable gain. It's easy enough to
reimplement if it's shown to help on huge machines.
- Re-implement the old tdq_transfer() call as tdq_pickidle(). Change
sched_add() so we have selectable cpu choosers and simplify the logic
a bit here.
- Implement tdq_pickpri() as the new default cpu chooser. This algorithm
is similar to Solaris in that it tries to always run the threads with
the best priorities. It is actually slightly more complex than
solaris's algorithm because we also tend to favor the local cpu over
other cpus which has a boost in latency but also potentially enables
cache sharing between the waking thread and the woken thread.
- Add a bunch of tunables that can be used to measure effects of different
load balancing strategies. Most of these will go away once the
algorithm is more definite.
- Add a new mechanism to steal threads from busy cpus when we idle. This
is enabled with kern.sched.steal_busy and kern.sched.busy_thresh. The
threshold is the required length of a tdq's run queue before another
cpu will be able to steal runnable threads. This prevents most queue
imbalances that contribute the long latencies.
of max() when computing the divisor in SCHED_TICK_PRI(). This prevents
cases where rounding down would allow the quotient to exceed
SCHED_PRI_RANGE.
- Garbage collect some unused flags and fields.
- Replace TDF_HOLD with sched_pin_td()/sched_unpin_td() since it simply
duplicated this functionality.
- Re-enable the rebalancer by default and fix the sysctl so it can be
modified.
marked idle, thus breaking cpu load balancing.
- Change sched_interact_update() to fix cases where the stored history
has expanded significantly rather than handling them in the callers. This
fixes a case where sched_priority() could compute a bad value.
- Add a sysctl to disable the global load balancer for experimentation.
setting ftick = ltick = ticks in schedinit().
- Update the priority when we are pulled off of the run queue and when we
are inserted onto the run queue so that it more accurately reflects our
present status. This is important for efficient priority propagation
functioning.
- Move the frequency test into sched_pctcpu_update() so we don't repeat it
each time we'd like to call it.
- Put some temporary work-around code in sched_priority() in case the tick
mechanism produces a bad priority. Eventually this should revert to an
assert again.
the most recently chosen index. This significantly improves nice
behavior. This allows a lower priority thread to run some multiple of
times before the higher priority thread makes it to the front of
the queue. A nice +20 cpu hog now only gets ~5% of the cpu when running
with a nice 0 cpu hog and about 1.5% with a nice -20 hog. A nice
difference of 1 makes a 4% difference in cpu usage between two hogs.
- Track a seperate insert and removal index. When the removal index is
empty it is updated to point at the current insert index.
- Don't remove and re-add a thread to the runq when it is being adjusted
down in priority.
- Pull some conditional code out of sched_tick(). It's looking a bit
large now.
- Remove the double queue mechanism for timeshare threads. It was slow
due to excess cache lines in play, caused suboptimal scheduling behavior
with niced and other non-interactive processes, complicated priority
lending, etc.
- Use a circular queue with a floating starting index for timeshare threads.
Enforces fairness by moving the insertion point closer to threads with
worse priorities over time.
- Give interactive timeshare threads real-time user-space priorities and
place them on the realtime/ithd queue.
- Select non-interactive timeshare thread priorities based on their cpu
utilization over the last 10 seconds combined with the nice value. This
gives us more sane priorities and behavior in a loaded system as
compared to the old method of using the interactivity score. The
interactive score quickly hit a ceiling if threads were non-interactive
and penalized new hog threads.
- Use one slice size for all threads. The slice is not currently
dynamically set to adjust scheduling behavior of different threads.
- Add some new sysctls for scheduling parameters.
Bug fixes/Clean up:
- Fix zeroing of td_sched after initialization in sched_fork_thread() caused
by recent ksegrp removal.
- Fix KSE interactivity issues related to frequent forking and exiting of
kse threads. We simply disable the penalty for thread creation and exit
for kse threads.
- Cleanup the cpu estimator by using tickincr here as well. Keep ticks and
ltick/ftick in the same frequency. Previously ticks were stathz and
others were hz.
- Lots of new and updated comments.
- Many many others.
Tested on: up x86/amd64, 8way amd64.
Make part of John Birrell's KSE patch permanent..
Specifically, remove:
Any reference of the ksegrp structure. This feature was
never fully utilised and made things overly complicated.
All code in the scheduler that tried to make threaded programs
fair to unthreaded programs. Libpthread processes will already
do this to some extent and libthr processes already disable it.
Also:
Since this makes such a big change to the scheduler(s), take the opportunity
to rename some structures and elements that had to be moved anyhow.
This makes the code a lot more readable.
The ULE scheduler compiles again but I have no idea if it works.
The 4bsd scheduler still reqires a little cleaning and some functions that now do
ALMOST nothing will go away, but I thought I'd do that as a separate commit.
Tested by David Xu, and Dan Eischen using libthr and libpthread.
yield() and sched_yield() syscalls. Every scheduler has its own way
to relinquish cpu, the ULE and CORE schedulers have two internal run-
queues, a timesharing thread which calls yield() syscall should be
moved to inactive queue.
I picked it up again. The scheduler is forked from ULE, but the
algorithm to detect an interactive process is almost completely
different with ULE, it comes from Linux paper "Understanding the
Linux 2.6.8.1 CPU Scheduler", although I still use same word
"score" as a priority boost in ULE scheduler.
Briefly, the scheduler has following characteristic:
1. Timesharing process's nice value is seriously respected,
timeslice and interaction detecting algorithm are based
on nice value.
2. per-cpu scheduling queue and load balancing.
3. O(1) scheduling.
4. Some cpu affinity code in wakeup path.
5. Support POSIX SCHED_FIFO and SCHED_RR.
Unlike scheduler 4BSD and ULE which using fuzzy RQ_PPQ, the scheduler
uses 256 priority queues. Unlike ULE which using pull and push, the
scheduelr uses pull method, the main reason is to let relative idle
cpu do the work, but current the whole scheduler is protected by the
big sched_lock, so the benefit is not visible, it really can be worse
than nothing because all other cpu are locked out when we are doing
balancing work, which the 4BSD scheduelr does not have this problem.
The scheduler does not support hyperthreading very well, in fact,
the scheduler does not make the difference between physical CPU and
logical CPU, this should be improved in feature. The scheduler has
priority inversion problem on MP machine, it is not good for
realtime scheduling, it can cause realtime process starving.
As a result, it seems the MySQL super-smack runs better on my
Pentium-D machine when using libthr, despite on UP or SMP kernel.
a thread holding critical resource, e.g mutex or other implicit
synchronous flags. Give thread which exceeds nice threshold a minimum
time slice.
PR: kern/86087
as this happens via thread_switchout(). I don't particularly like the
structure of the code here. We twice call out to thread code when
a thread is voluntarily switching. Once to thread_switchout() and once
to slot_fill(), while sched_4BSD does even more work which is redundant
to select another thread to use our remaining slice. This should be
simplified in the future, but for now I'm only going to fix the bug not
the bad design.
slot for us. Previously, we would take two slots on every preempt, and
setrunqueue() would fix it up for us in the non threaded case. The
threaded case was simply broken.
- Clean up flags, prototypes, comments.
schedulers a bit to ensure more correct handling of priorities and fewer
priority inversions:
- Add two functions to the sched(9) API to handle priority lending:
sched_lend_prio() and sched_unlend_prio(). The turnstile code uses these
functions to ask the scheduler to lend a thread a set priority and to
tell the scheduler when it thinks it is ok for a thread to stop borrowing
priority. The unlend case is slightly complex in that the turnstile code
tells the scheduler what the minimum priority of the thread needs to be
to satisfy the requirements of any other threads blocked on locks owned
by the thread in question. The scheduler then decides where the thread
can go back to normal mode (if it's normal priority is high enough to
satisfy the pending lock requests) or it it should continue to use the
priority specified to the sched_unlend_prio() call. This involves adding
a new per-thread flag TDF_BORROWING that replaces the ULE-only kse flag
for priority elevation.
- Schedulers now refuse to lower the priority of a thread that is currently
borrowing another therad's priority.
- If a scheduler changes the priority of a thread that is currently sitting
on a turnstile, it will call a new function turnstile_adjust() to inform
the turnstile code of the change. This function resorts the thread on
the priority list of the turnstile if needed, and if the thread ends up
at the head of the list (due to having the highest priority) and its
priority was raised, then it will propagate that new priority to the
owner of the lock it is blocked on.
Some additional fixes specific to the 4BSD scheduler include:
- Common code for updating the priority of a thread when the user priority
of its associated kse group has been consolidated in a new static
function resetpriority_thread(). One change to this function is that
it will now only adjust the priority of a thread if it already has a
time sharing priority, thus preserving any boosts from a tsleep() until
the thread returns to userland. Also, resetpriority() no longer calls
maybe_resched() on each thread in the group. Instead, the code calling
resetpriority() is responsible for calling resetpriority_thread() on
any threads that need to be updated.
- schedcpu() now uses resetpriority_thread() instead of just calling
sched_prio() directly after it updates a kse group's user priority.
- sched_clock() now uses resetpriority_thread() rather than writing
directly to td_priority.
- sched_nice() now updates all the priorities of the threads after the
group priority has been adjusted.
Discussed with: bde
Reviewed by: ups, jeffr
Tested on: 4bsd, ule
Tested on: i386, alpha, sparc64
- Remove the sched_add wrapper that used sched_add_internal() as a backend.
Its only purpose was to interpret one flag and turn it into an int. Do
the right thing and interpret the flag in sched_add() instead.
- Pass the flag argument to sched_add() to kseq_runq_add() so that we can
get the SRQ_PREEMPT optimization too.
- Add a KEF_INTERNAL flag. If KEF_INTERNAL is set we don't adjust the SLOT
counts, otherwise the slot counts are adjusted as soon as we enter
sched_add() or sched_rem() rather than when the thread is actually placed
on the run queue. This greatly simplifies the handling of slots.
- Remove the explicit prevention of migration for ithreads on non-x86
platforms. This was never shown to have any real benefit.
- Remove the unused class argument to KSE_CAN_MIGRATE().
- Add ktr points for thread migration events.
- Fix a long standing bug on platforms which don't initialize the cpu
topology. The ksg_maxid variable was never correctly set on these
platforms which caused the long term load balancer to never inspect
more than the first group or processor.
- Fix another bug which prevented the long term load balancer from working
properly. If stathz != hz we can't expect sched_clock() to be called
on the exact tick count that we're anticipating.
- Rearrange sched_switch() a bit to reduce indentation levels.
nice of 0. Doing so can cause an infinite loop because they should be
running, but a nice -20 process could prevent them from doing so.
- Add a new flag KEF_PRIOELEV to flag a thread that has had its priority
elevated due to priority propagation. If a thread has had its priority
elevated, we assume that it must go on the current queue and it must
get a slice.
- In sched_userret() if our priority was elevated and we shouldn't have
a timeslice, yield here until we should.
Found/Tested by: glebius
outside of the nice threshold due to a recently awoken thread with a
lower nice value. This further reduces the amount of time a positively
niced thread gets while running in conjunction with a workload that has
many short sleeps (ie buildworld).
check for TD_ON_RUNQ() no longer means the thread is really on a run-
queue. I suspect this state should be re-evaluated as it must mean
something else now. This fixes ULE+KSE+PREEMPTION on UP x86.
fully initialed when the pmap layer tries to call sched_pini() early in the
boot and results in an quick panic. Use ke_pinned instead as was originally
done with Tor's patch.
Approved by: julian
scheduler specific extension to it. Put it in the extension as
the implimentation details of how the pinning is done needn't be visible
outside the scheduler.
Submitted by: tegge (of course!) (with changes)
MFC after: 3 days
but with slightly cleaned up interfaces.
The KSE structure has become the same as the "per thread scheduler
private data" structure. In order to not make the diffs too great
one is #defined as the other at this time.
The KSE (or td_sched) structure is now allocated per thread and has no
allocation code of its own.
Concurrency for a KSEGRP is now kept track of via a simple pair of counters
rather than using KSE structures as tokens.
Since the KSE structure is different in each scheduler, kern_switch.c
is now included at the end of each scheduler. Nothing outside the
scheduler knows the contents of the KSE (aka td_sched) structure.
The fields in the ksegrp structure that are to do with the scheduler's
queueing mechanisms are now moved to the kg_sched structure.
(per ksegrp scheduler private data structure). In other words how the
scheduler queues and keeps track of threads is no-one's business except
the scheduler's. This should allow people to write experimental
schedulers with completely different internal structuring.
A scheduler call sched_set_concurrency(kg, N) has been added that
notifies teh scheduler that no more than N threads from that ksegrp
should be allowed to be on concurrently scheduled. This is also
used to enforce 'fainess' at this time so that a ksegrp with
10000 threads can not swamp a the run queue and force out a process
with 1 thread, since the current code will not set the concurrency above
NCPU, and both schedulers will not allow more than that many
onto the system run queue at a time. Each scheduler should eventualy develop
their own methods to do this now that they are effectively separated.
Rejig libthr's kernel interface to follow the same code paths as
linkse for scope system threads. This has slightly hurt libthr's performance
but I will work to recover as much of it as I can.
Thread exit code has been cleaned up greatly.
exit and exec code now transitions a process back to
'standard non-threaded mode' before taking the next step.
Reviewed by: scottl, peter
MFC after: 1 week
FULL_PREEMPTION is defined. Add a runtime warning to ULE if PREEMPTION is
enabled (code inspired by the PREEMPTION warning in kern_switch.c). This
is a possible MT5 candidate.
preemption and/or the rev 1.79 kern_switch.c change that was backed out.
The thread was being assigned to a runq without adding in the load, which
would cause the counter to hit -1.
migration. Use this in sched_prio() and sched_switch() to stop us from
migrating threads that are in short term sleeps or are runnable. These
extra migrations were added in the patches to support KSE.
- Only set NEEDRESCHED if the thread we're adding in sched_add() is a
lower priority and is being placed on the current queue.
- Fix some minor whitespace problems.
contributed to the transferable load count. This prevents any potential
problems with sched_pin() being used around calls to setrunqueue().
- Change the sched_add() load balancing algorithm to try to migrate on
wakeup. This attempts to place threads that communicate with each other
on the same CPU.
- Don't clear the idle counts in kseq_transfer(), let the cpus do that when
they call sched_add() from kseq_assign().
- Correct a few out of date comments.
- Make sure the ke_cpu field is correct when we preempt.
- Call kseq_assign() from sched_clock() to catch any assignments that were
done without IPI. Presently all assignments are done with an IPI, but I'm
trying a patch that limits that.
- Don't migrate a thread if it is still runnable in sched_add(). Previously,
this could only happen for KSE threads, but due to changes to
sched_switch() all threads went through this path.
- Remove some code that was added with preemption but is not necessary.
specify "us" as the thread not the process/ksegrp/kse.
You can always find the others from the thread but the converse is not true.
Theorotically this would lead to runtime being allocated to the wrong
entity in some cases though it is not clear how often this actually happenned.
(would only affect threaded processes and would probably be pretty benign,
but it WAS a bug..)
Reviewed by: peter
since they are only accessed by curthread and thus do not need any
locking.
- Move pr_addr and pr_ticks out of struct uprof (which is per-process)
and directly into struct thread as td_profil_addr and td_profil_ticks
as these variables are really per-thread. (They are used to defer an
addupc_intr() that was too "hard" until ast()).
takes an argument to specify if it should preempt or not. Don't preempt
when sched_add_internal() is called from kseq_idled() or kseq_assign()
as in those cases we are about to call mi_switch() anyways. Also, doing
so during the first context switch on an AP leads to a NULL pointer deref
because curthread is NULL.
- Reenable preemption for ULE.
Submitted by: Taku YAMAMOTO taku at tackymt.homeip.net
hangs due to recent preemption changes. This change appears to remove
the panic that I was running into, but at the cost of increasing
ithread scheduling latency, and as such is a temporary band-aid until
jhb has a chance to resolve the ule<->preemption interaction that is
the source of the problem. If it doesn't fix the problem for others--
sorry!
introduced a KSE_CAN_MIGRATE() invocation with one argument
missing (class). Either this is a genuine forget or it crept
in from JHB's repo where he may have modified it. If it's
the latter then it may require more attention. For now fix
the make depend.
than as one-off hacks in various other parts of the kernel:
- Add a function maybe_preempt() that is called from sched_add() to
determine if a thread about to be added to a run queue should be
preempted to directly. If it is not safe to preempt or if the new
thread does not have a high enough priority, then the function returns
false and sched_add() adds the thread to the run queue. If the thread
should be preempted to but the current thread is in a nested critical
section, then the flag TDF_OWEPREEMPT is set and the thread is added
to the run queue. Otherwise, mi_switch() is called immediately and the
thread is never added to the run queue since it is switch to directly.
When exiting an outermost critical section, if TDF_OWEPREEMPT is set,
then clear it and call mi_switch() to perform the deferred preemption.
- Remove explicit preemption from ithread_schedule() as calling
setrunqueue() now does all the correct work. This also removes the
do_switch argument from ithread_schedule().
- Do not use the manual preemption code in mtx_unlock if the architecture
supports native preemption.
- Don't call mi_switch() in a loop during shutdown to give ithreads a
chance to run if the architecture supports native preemption since
the ithreads will just preempt DELAY().
- Don't call mi_switch() from the page zeroing idle thread for
architectures that support native preemption as it is unnecessary.
- Native preemption is enabled on the same archs that supported ithread
preemption, namely alpha, i386, and amd64.
This change should largely be a NOP for the default case as committed
except that we will do fewer context switches in a few cases and will
avoid the run queues completely when preempting.
Approved by: scottl (with his re@ hat)
switch to. If a non-NULL thread pointer is passed in, then the CPU will
switch to that thread directly rather than calling choosethread() to pick
a thread to choose to.
- Make sched_switch() aware of idle threads and know to do
TD_SET_CAN_RUN() instead of sticking them on the run queue rather than
requiring all callers of mi_switch() to know to do this if they can be
called from an idlethread.
- Move constants for arguments to mi_switch() and thread_single() out of
the middle of the function prototypes and up above into their own
section.
sched_clock() rather than using callouts. This means we no longer have to
take the load of the callout thread into consideration while balancing and
should make the balancing decisions simpler and more accurate.
Tested on: x86/UP, amd64/SMP
sched_ule, in January 2004. Looking at this, "pagezero" is (one of) the
culprit(s). We had no provision for processes with P_NOLOAD set. With
pagezero not running at PRI_ITHD, kseq_load_{add,rem} count pagezero as
another-normal-process, thus the "expected-plus-one" load reported in
the above thread.
Submitted by: Nikos Ntarmos <ntarmos@ceid.upatras.gr>
SCHED_INTERACT_MAX was used where SCHED_SLP_RUN_MAX was needed. This was
causing the interactivity scaler to lose history at a more dramatic rate
than intended.
long as there are still explicit uses of int, whether in types or
in function names (such as atomic_set_int() in sched_ule.c), we can
not change cpumask_t to be anything other than u_int. See also the
commit log for sys/sys/types.h, revision 1.84.
activation (i.e., applications are using libpthread). This is because
SCHED_ULE sometimes puts P_SA processes into ksq_next unnecessarily.
Which doesn't give fair amount of CPU time to processes which are
using scheduler-activation-based threads when other (semi-)CPU-intensive,
non-P_SA processes are running.
Further work will no doubt be done by jeffr at a later date.
Submitted by: Taku YAMAMOTO <taku@cent.saitama-u.ac.jp>
Reviewed by: rwatson, freebsd-current@
sleep queue interface:
- Sleep queues attempt to merge some of the benefits of both sleep queues
and condition variables. Having sleep qeueus in a hash table avoids
having to allocate a queue head for each wait channel. Thus, struct cv
has shrunk down to just a single char * pointer now. However, the
hash table does not hold threads directly, but queue heads. This means
that once you have located a queue in the hash bucket, you no longer have
to walk the rest of the hash chain looking for threads. Instead, you have
a list of all the threads sleeping on that wait channel.
- Outside of the sleepq code and the sleep/cv code the kernel no longer
differentiates between cv's and sleep/wakeup. For example, calls to
abortsleep() and cv_abort() are replaced with a call to sleepq_abort().
Thus, the TDF_CVWAITQ flag is removed. Also, calls to unsleep() and
cv_waitq_remove() have been replaced with calls to sleepq_remove().
- The sched_sleep() function no longer accepts a priority argument as
sleep's no longer inherently bump the priority. Instead, this is soley
a propery of msleep() which explicitly calls sched_prio() before
blocking.
- The TDF_ONSLEEPQ flag has been dropped as it was never used. The
associated TDF_SET_ONSLEEPQ and TDF_CLR_ON_SLEEPQ macros have also been
dropped and replaced with a single explicit clearing of td_wchan.
TD_SET_ONSLEEPQ() would really have only made sense if it had taken
the wait channel and message as arguments anyway. Now that that only
happens in one place, a macro would be overkill.
track the load for the sched_load() function. In the SMP case this member
is not defined because it would be redundant with the ksg_load member
which already tracks the non ithd load.
- For sched_load() in the UP case simply return ksq_sysload. In the SMP
case traverse the list of kseq groups and sum up their ksg_load fields.
SW_INVOL. Assert that one of these is set in mi_switch() and propery
adjust the rusage statistics. This is to simplify the large number of
users of this interface which were previously all required to adjust the
proper counter prior to calling mi_switch(). This also facilitates more
switch and locking optimizations.
- Change all callers of mi_switch() to pass the appropriate paramter and
remove direct references to the process statistics.
cpu could have been bogged down with non-transferable load and still not
migrated a new thread to an idle cpu. This required some benchmarking and
tuning to get right as the comment above it suggests.
- In sched_add(), do the idle check prior to the transfer check so that we
don't try to transfer load from an idle cpu. This fixes panics caused by
IPIs on UP machines running SMP kernels.
Reported/Debugged by: seanc
- The new sched_balance_groups() function does intra-group balancing while
sched_balance() balances the available groups.
- Pick a random time between 0 ticks and hz * 2 ticks to restart each
balancing process. Each balancer has its own timeout.
- Pick a random place in the list of groups to start the search for lowest
and highest group loads. This prevents us from prefering a group based on
numeric position.
- Use a nasty hack to stop us from preferring cpu 0. The problem is that
softclock always runs on cpu 0, so it always has a little extra load. We
ignore this load in the balancer for now. In the future softclock should
run on a random cpu and these hacks can go away.
cpu are added to a group.
- Don't place a cpu into the kseq_idle bitmask until all cpus in that group
have idled.
- Prefer idle groups over idle group members in the new kseq_transfer()
function. In this way we will prefer to balance load across full cores
rather than add further load a partial core.
- Before a cpu goes idle, check the other group members for threads. Since
SMT cpus may freely share threads, this is cheap.
- SMT cores may be individually pinned and bound to now. This contrasts the
old mechanism where binding or pinning would have allowed a thread to run
on any available cpu.
- Remove some unnecessary logic from sched_switch(). Priority propagation
should be properly taken care of in sched_prio() now.
Be sure to shift (long)1 << 33 and higher, not (int)1. Otherwise bad
things happen(TM). This is why beast.freebsd.org paniced with ULE.
Reviewed by: jeff
1) mp_maxid is a valid FreeBSD CPU ID in the range 0 .. MAXCPU - 1.
2) For all active CPUs in the system, PCPU_GET(cpuid) <= mp_maxid.
Approved by: re (scottl)
Tested on: i386, amd64, alpha
kses from the run queues. Also, on SMP, we track the transferable
count here. Threads are transferable only as long as they are on the
run queue.
- Previously, we adjusted our load balancing based on the transferable count
minus the number of actual cpus. This was done to account for the threads
which were likely to be running. All of this logic is simpler now that
transferable accounts for only those threads which can actually be taken.
Updated various places in sched_add() and kseq_balance() to account for
this.
- Rename kseq_{add,rem} to kseq_load_{add,rem} to reflect what they're
really doing. The load is accounted for seperately from the runq because
the load is accounted for even as the thread is running.
- Fix a bug in sched_class() where we weren't properly using the PRI_BASE()
version of the kg_pri_class.
- Add a large comment that describes the impact of a seemingly simple
conditional in sched_add().
- Also in sched_add() check the transferable count and KSE_CAN_MIGRATE()
prior to checking kseq_idle. This reduces the frequency of access for
kseq_idle which is a shared resource.
idle. They figure out that we're idle fast enough that the cache pollution
introduces by scanning their run queue is more expensive than waiting
a little longer.
- Add kseq_setidle() to mark us as being idle. Use this in place of
kseq_find().
- Remove kseq_load_highest(), kseq_find() was the only consumer of this
interface. kseq_balance() has it's own customized version that finds the
lowest and highest loads simultaneously.
Continuously told that this would be faster by: terry
the total load, the timeshare load, and the number of threads that can
be migrated to another cpu. Account for these seperately.
- Introduce a KSE_CAN_MIGRATE() macro which determines whether or not a KSE
can be migrated to another CPU. Currently, this only checks to see if
we're an interrupt handler. Eventually this will also be used to support
CPU binding.
slice assignment. Add a comment describing what it does.
- Remove a stale XXX comment, the nice should not impact the interactivity,
nice adjustments only effect non-interactive tasks in ULE.
- Don't allow nice -20 tasks to totally starve nice 0 tasks. Give them at
least SCHED_SLICE_MIN ticks. We still allow nice 0 tasks to starve nice
+20 tasks as intended.
- SCHED_PRI_NRESV does not have the off by one error in PRIO_TOTAL so we
do not have to account for it in the few places that we use it.
Requested by: bde
0 and SCHED_SLP_RUN_MAX * 2. This allows us to simplify the algorithm
quite a bit. Before, it dealt with arbitrary values which required us
to do nasty integer division tricks that didn't quite work out correctly.
- Chnage sched_wakeup() to detect conditions where the slp+runtime could
exceed SCHED_SLP_RUN_MAX * 2. This can happen if we go to sleep for
longer than 6 seconds. In this case, we'll just clear the runtime and
set the sleep time to the max.
- Define a new function, sched_interact_fork() which updates the slp+runtime
of a newly forked thread. We want to limit the amount of history retained
from the parent so that we learn the child's behavior quickly. We don't,
however want to decay it to nothing. Previously, we would simply divide
each parameter by 100 whenever we forked. After a few forks the values
would reach 0 and tasks would not be considered interactive.
- Add another KTR entry, cleanup some existing entries.
- Remove a useless sched_interact_update() from sched_priority(). This is
already done by the callers that require it.
- Add an IPI based mechanism for migrating kses. This mechanism is
broken down into several components. This is intended to reduce cache
thrashing by eliminating most cases where one cpu touches another's
run queues.
- kseq_notify() appends a kse to a lockless singly linked list and
conditionally sends an IPI to the target processor. Right now this is
protected by sched_lock but at some point I'd like to get rid of the
global lock. This is why I used something more complicated than a
standard queue.
- kseq_assign() processes our list of kses that have been assigned to us
by other processors. This simply calls sched_add() for each item on the
list after clearing the new KEF_ASSIGNED flag. This flag is used to
indicate that we have been appeneded to the assigned queue but not
added to the run queue yet.
- In sched_add(), instead of adding a KSE to another processor's queue we
use kse_notify() so that we don't touch their queue. Also in sched_add(),
if KEF_ASSIGNED is already set return immediately. This can happen if
a thread is removed and readded so that the priority is recorded properly.
- In sched_rem() return immediately if KEF_ASSIGNED is set. All callers
immediately readd simply to adjust priorites etc.
- In sched_choose(), if we're running an IDLE task or the per cpu idle thread
set our cpumask bit in 'kseq_idle' so that other processors may know that
we are idle. Before this, make a single pass through the run queues of
other processors so that we may find work more immediately if it is
available.
- In sched_runnable(), don't scan each processor's run queue, they will IPI
us if they have work for us to do.
- In sched_add(), if we're adding a thread that can be migrated and we have
plenty of work to do, try to migrate the thread to an idle kseq.
- Simplify the logic in sched_prio() and take the KEF_ASSIGNED flag into
consideration.
- No longer use kseq_choose() to steal threads, it can lose it's last
argument.
- Create a new function runq_steal() which operates like runq_choose() but
skips threads based on some criteria. Currently it will not steal
PRI_ITHD threads. In the future this will be used for CPU binding.
- Create a kseq_steal() that checks each run queue with runq_steal(), use
kseq_steal() in the places where we used kseq_choose() to steal with
before.
begin with sched_lock held but not recursed, so this variable was
always 0.
Removed fixup of sched_lock.mtx_recurse after context switches in
sched_switch(). Context switches always end with this variable in the
same state that it began in, so there is no need to fix it up. Only
sched_lock.mtx_lock really needs a fixup.
Replaced fixup of sched_lock.mtx_recurse in fork_exit() by an assertion
that sched_lock is owned and not recursed after it is fixed up. This
assertion much match the one in mi_switch(), and if sched_lock were
recursed then a non-null fixup of sched_lock.mtx_recurse would probably
be needed again, unlike in sched_switch(), since fork_exit() doesn't
return to its caller in the normal way.
Contributed by: Thomaswuerfl@gmx.de
- In sched_prio(), adjust the run queue for threads which may need to move
to the current queue due to priority propagation .
- In sched_switch(), fix style bug introduced when the KSE support went in.
Columns are 80 chars wide, not 90.
- In sched_switch(), Fix the comparison in the idle case and explicitly
re-initialize the runq in the not propagated case.
- Remove dead code in sched_clock().
- In sched_clock(), If we're an IDLE class td set NEEDRESCHED so that threads
that have become runnable will get a chance to.
- In sched_runnable(), if we're not the IDLETD, we should not consider
curthread when examining the load. This mimics the 4BSD behavior of
returning 0 when the only runnable thread is running.
- In sched_userret(), remove the code for setting NEEDRESCHED entirely.
This is not necessary and is not implemented in 4BSD.
- Use the correct comparison in sched_add() when checking to see if an idle
prio task has had it's priority temporarily elevated.
rounding errors. This was the source of the majority of the
interactivity problems. Reintroduce the old algorithm and its XXX.
- Up the interactivity threshold to 30. It really could stand to be even
a tiny bit higher.
- Let the sleep and run time accumulate up to 5 seconds of history rather
than two. This helps stop XFree86 from becoming non-interactive during
bursts of activity.
elevated either due to priority propagation or because we're in the
kernel in either case, put us on the current queue so that we dont
stop others from using important resources. At some point the priority
elevations from sleeping in the kernel should go away.
- Remove an optimization in sched_userret(). Before we would only set
NEEDRESCHED if there was something of a higher priority available. This
is a trivial optimization and it breaks priority propagation because it
doesn't take threads which we may be blocking into account. Notice that
the thread which is blocking others gets up to one tick of cpu time before
we honor this NEEDRESCHED in sched_clock().
you on the current queue. In the future, it would be nice if priority
propagation could deterministicly pluck a thread off of the next queue
and put it on the current queue. Until then this hack stops us from
holding up our entire current queue, including interrupt handlers, while
a thread on the next queue is blocked while holding Giant.
- Inherit our pctcpu information from our parent.
- Associate logical CPUs on the same physical core with the same kseq.
- Adjust code that assumed there would only be one running thread in any
kseq.
- Wrap the HTT code with a ULE_HTT_EXPERIMENTAL ifdef. This is a start
towards HyperThreading support but it isn't quite there yet.
