- Add a per-turnstile spinlock to solve potential priority propagation
deadlocks that are possible with thread_lock().
- The turnstile lock order is defined as the exact opposite of the
lock order used with the sleep locks they represent. This allows us
to walk in reverse order in priority_propagate and this is the only
place we wish to multiply acquire turnstile locks.
- Use the turnstile_chain lock to protect assigning mutexes to turnstiles.
- Change the turnstile interface to pass back turnstile pointers to the
consumers. This allows us to reduce some locking and makes it easier
to cancel turnstile assignment while the turnstile chain lock is held.
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)
These functions are intended to be used to drop a lock and then reacquire
it when doing an sleep such as msleep(9). Both functions accept a
'struct lock_object *' as their first parameter. The 'lc_unlock' function
returns an integer that is then passed as the second paramter to the
subsequent 'lc_lock' function. This can be used to communicate state.
For example, sx locks and rwlocks use this to indicate if the lock was
share/read locked vs exclusive/write locked.
Currently, spin mutexes and lockmgr locks do not provide working lc_lock
and lc_unlock functions.
and optimize away unused stack values. The 48 bytes that the lock_profile_object
adds to the stack evidently has a measurable performance impact on certain workloads.
- Fix missing initialization in kern_rwlock.c causing bogus times to be collected
- Move updates to the lock hash to after the lock is released for spin mutexes,
sleep mutexes, and sx locks
- Add new kernel build option LOCK_PROFILE_FAST - only update lock profiling
statistics when an acquisition is contended. This reduces the overhead of
LOCK_PROFILING to increasing system time by 20%-25% which on
"make -j8 kernel-toolchain" on a dual woodcrest is unmeasurable in terms
of wall-clock time. Contrast this to enabling lock profiling without
LOCK_PROFILE_FAST and I see a 5x-6x slowdown in wall-clock time.
- 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
- add cnt_hold cnt_lock support for spin mutexes
- make sure contested is initialized to zero to only bump contested when appropriate
- move initialization function to kern_mutex.c to avoid cyclic dependency between
mutex.h and lock_profile.h
wait (time waited to acquire) and hold times for *all* kernel locks. If
the architecture has a system synchronized TSC, the profiling code will
use that - thereby minimizing profiling overhead. Large chunks of profiling
code have been moved out of line, the overhead measured on the T1 for when
it is compiled in but not enabled is < 1%.
Approved by: scottl (standing in for mentor rwatson)
Reviewed by: des and jhb
panic, go ahead and do the longer DELAY(1) spin wait.
- If we panic due to spinning too long, print out a few more details
including the pointer to the mutex in question and the tid of the owning
thread.
a count of all non-spin locks, not just lockmgr locks. This can give us a
much cheaper way to see if we have any locks held (such as when returning
to userland via userret()) without requiring WITNESS.
MFC after: 1 week
all other mtx_lock() operations to block. Previously, when the mutex was
destroyed, it would still have a valid value in mtx_lock(): either the
unowned cookie, which would allow a subsequent mtx_lock() to succeed, or a
pointer to the thread who destroyed the mutex if the mutex was locked when
it was destroyed.
MFC after: 3 days
non-intuitive for the ~ to be built into the mask. All the users now
explicitly ~ the mask. In addition, add MTX_UNOWNED to the mask even
though it technically isn't a flag. This should unbreak mtx_owner().
Quickly spotted by: kris
compiler doesn't decide to cache td_state. Cachine the state would cause
the spinning thread to not notice when the owning thread stopped executing
(if it was preempted for example) which could result in livelock.
each turnstile. Also, allow for the owner thread pointer of a turnstile
to be NULL. This is needed for the upcoming reader/writer lock
implementation.
- Add a new ddb command 'show turnstile' that will look up the turnstile
associated with the given lock argument and display useful information
like the list of threads blocked on each queue, etc. If there isn't an
active turnstile for a lock at the specified address, then the function
will see if there is an active turnstile at the specified address and
display info about it if so.
- Adjust the mutex code to handle the turnstile API changes.
Tested on: i386 (all), alpha, amd64, sparc64 (1 and 3)
lock_obj objects:
- Add new lock_init() and lock_destroy() functions to setup and teardown
lock_object objects including KTR logging and registering with WITNESS.
- Move all the handling of LO_INITIALIZED out of witness and the various
lock init functions into lock_init() and lock_destroy().
- Remove the constants for static indices into the lock_classes[] array
and change the code outside of subr_lock.c to use LOCK_CLASS to compare
against a known lock class.
- Move the 'show lock' ddb function and lock_classes[] array out of
kern_mutex.c over to subr_lock.c.
and subsequently broke the build. This change is supposed to fix the
case where doing a mtx_destroy() off a spin mutex while you hold it fails.
If it had been tested I would just leave it in, but it hasn't been tested
yet, so it will have to wait until later.
struct sx). Instead of storing a direct pointer to a our lock_class
struct in lock_object, reserve 4 bits in the lo_flags field to serve as an
index into a global lock_classes array that contains pointers to the lock
classes. Only debugging code such as WITNESS or INVARIANTS checks and KTR
logging need to access the lock_class member, so this shouldn't add any
overhead to production kernels. It might add some slight overhead to
kernels using those debug options however.
As with the previous set of changes to lock_object, this is going to
completely obliterate the kernel ABI, so be sure to recompile all your
modules.
class, then it displays various information about the lock and calls a
new function pointer in lock_class (lc_ddb_show) to dump class-specific
information about the lock as well (such as the owner of a mutex or
xlock'ed sx lock). This is easier than staring at hex dumps of locks to
figure out who owns the lock, etc. Note that extending lock_class doesn't
affect the ABI for any kernel modules as the only code that deals with
lock_class structures directly is kern_mutex.c, kern_sx.c, and witness.
MFC after: 1 week
called during early init before cninit().
Tested on: i386, alpha, sparc64
Reviewed by: phk, imp
Reported by: Divacky Roman xdivac02 at stud dot fit dot vutbr dot cz
MFC after: 1 week
mutex.
- Don't panic if a spin lock is held too long inside _mtx_lock_spin() if
panicstr is set (meaning that we are already in a panic). Just keep
spinning forever instead.
variables rather than void * variables. This makes it easier and simpler
to get asm constraints and volatile keywords correct.
MFC after: 3 days
Tested on: i386, alpha, sparc64
Compiled on: ia64, powerpc, amd64
Kernel toolchain busted on: arm
critical_enter() and critical_exit() are now solely a mechanism for
deferring kernel preemptions. They no longer have any affect on
interrupts. This means that standalone critical sections are now very
cheap as they are simply unlocked integer increments and decrements for the
common case.
Spin mutexes now use a separate KPI implemented in MD code: spinlock_enter()
and spinlock_exit(). This KPI is responsible for providing whatever MD
guarantees are needed to ensure that a thread holding a spin lock won't
be preempted by any other code that will try to lock the same lock. For
now all archs continue to block interrupts in a "spinlock section" as they
did formerly in all critical sections. Note that I've also taken this
opportunity to push a few things into MD code rather than MI. For example,
critical_fork_exit() no longer exists. Instead, MD code ensures that new
threads have the correct state when they are created. Also, we no longer
try to fixup the idlethreads for APs in MI code. Instead, each arch sets
the initial curthread and adjusts the state of the idle thread it borrows
in order to perform the initial context switch.
This change is largely a big NOP, but the cleaner separation it provides
will allow for more efficient alternative locking schemes in other parts
of the kernel (bare critical sections rather than per-CPU spin mutexes
for per-CPU data for example).
Reviewed by: grehan, cognet, arch@, others
Tested on: i386, alpha, sparc64, powerpc, arm, possibly more
turn it back on. Specifically, the actual changes are now less intrusive
in that the _get_spin_lock() and _rel_spin_lock() macros now have their
contents changed for UP vs SMP kernels which centralizes the changes.
Also, UP kernels do not use _mtx_lock_spin() and no longer include it. The
UP versions of the spin lock functions do not use any atomic operations,
but simple compares and stores which allow mtx_owned() to still work for
spin locks while removing the overhead of atomic operations.
Tested on: i386, alpha
- Add a new _lock() call to each API that locks the associated chain lock
for a lock_object pointer or wait channel. The _lookup() functions now
require that the chain lock be locked via _lock() when they are called.
- Change sleepq_add(), turnstile_wait() and turnstile_claim() to lookup
the associated queue structure internally via _lookup() rather than
accepting a pointer from the caller. For turnstiles, this means that
the actual lookup of the turnstile in the hash table is only done when
the thread actually blocks rather than being done on each loop iteration
in _mtx_lock_sleep(). For sleep queues, this means that sleepq_lookup()
is no longer used outside of the sleep queue code except to implement an
assertion in cv_destroy().
- Change sleepq_broadcast() and sleepq_signal() to require that the chain
lock is already required. For condition variables, this lets the
cv_broadcast() and cv_signal() functions lock the sleep queue chain lock
while testing the waiters count. This means that the waiters count
internal to condition variables is no longer protected by the interlock
mutex and cv_broadcast() and cv_signal() now no longer require that the
interlock be held when they are called. This lets consumers of condition
variables drop the lock before waking other threads which can result in
fewer context switches.
MFC after: 1 month
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.
macros and pass the value to the associated _mtx_*() functions to avoid
more curthread dereferences in the function implementations. This provided
a very modest perf improvement in some benchmarks.
Suggested by: rwatson
Tested by: scottl
