On Fri, 5 May 2023 16:49:38 GMT, Roman Kennke <rken...@openjdk.org> wrote:
>> This change adds a fast-locking scheme as an alternative to the current >> stack-locking implementation. It retains the advantages of stack-locking >> (namely fast locking in uncontended code-paths), while avoiding the overload >> of the mark word. That overloading causes massive problems with Lilliput, >> because it means we have to check and deal with this situation when trying >> to access the mark-word. And because of the very racy nature, this turns out >> to be very complex and would involve a variant of the inflation protocol to >> ensure that the object header is stable. (The current implementation of >> setting/fetching the i-hash provides a glimpse into the complexity). >> >> What the original stack-locking does is basically to push a stack-lock onto >> the stack which consists only of the displaced header, and CAS a pointer to >> this stack location into the object header (the lowest two header bits being >> 00 indicate 'stack-locked'). The pointer into the stack can then be used to >> identify which thread currently owns the lock. >> >> This change basically reverses stack-locking: It still CASes the lowest two >> header bits to 00 to indicate 'fast-locked' but does *not* overload the >> upper bits with a stack-pointer. Instead, it pushes the object-reference to >> a thread-local lock-stack. This is a new structure which is basically a >> small array of oops that is associated with each thread. Experience shows >> that this array typcially remains very small (3-5 elements). Using this lock >> stack, it is possible to query which threads own which locks. Most >> importantly, the most common question 'does the current thread own me?' is >> very quickly answered by doing a quick scan of the array. More complex >> queries like 'which thread owns X?' are not performed in very >> performance-critical paths (usually in code like JVMTI or deadlock >> detection) where it is ok to do more complex operations (and we already do). >> The lock-stack is also a new set of GC roots, and would be scanned during >> thread scanning, possibly concurrently, via the normal protocols. >> >> The lock-stack is fixed size, currently with 8 elements. According to my >> experiments with various workloads, this covers the vast majority of >> workloads (in-fact, most workloads seem to never exceed 5 active locks per >> thread at a time). We check for overflow in the fast-paths and when the >> lock-stack is full, we take the slow-path, which would inflate the lock to a >> monitor. That case should be very rare. >> >> In contrast to stack-locking, fast-locking does *not* support recursive >> locking (yet). When that happens, the fast-lock gets inflated to a full >> monitor. It is not clear if it is worth to add support for recursive >> fast-locking. >> >> One trouble is that when a contending thread arrives at a fast-locked >> object, it must inflate the fast-lock to a full monitor. Normally, we need >> to know the current owning thread, and record that in the monitor, so that >> the contending thread can wait for the current owner to properly exit the >> monitor. However, fast-locking doesn't have this information. What we do >> instead is to record a special marker ANONYMOUS_OWNER. When the thread that >> currently holds the lock arrives at monitorexit, and observes >> ANONYMOUS_OWNER, it knows it must be itself, fixes the owner to be itself, >> and then properly exits the monitor, and thus handing over to the contending >> thread. >> >> As an alternative, I considered to remove stack-locking altogether, and only >> use heavy monitors. In most workloads this did not show measurable >> regressions. However, in a few workloads, I have observed severe >> regressions. All of them have been using old synchronized Java collections >> (Vector, Stack), StringBuffer or similar code. The combination of two >> conditions leads to regressions without stack- or fast-locking: 1. The >> workload synchronizes on uncontended locks (e.g. single-threaded use of >> Vector or StringBuffer) and 2. The workload churns such locks. IOW, >> uncontended use of Vector, StringBuffer, etc as such is ok, but creating >> lots of such single-use, single-threaded-locked objects leads to massive >> ObjectMonitor churn, which can lead to a significant performance impact. But >> alas, such code exists, and we probably don't want to punish it if we can >> avoid it. >> >> This change enables to simplify (and speed-up!) a lot of code: >> >> - The inflation protocol is no longer necessary: we can directly CAS the >> (tagged) ObjectMonitor pointer to the object header. >> - Accessing the hashcode could now be done in the fastpath always, if the >> hashcode has been installed. Fast-locked headers can be used directly, for >> monitor-locked objects we can easily reach-through to the displaced header. >> This is safe because Java threads participate in monitor deflation protocol. >> This would be implemented in a separate PR >> >> Also, and I might be mistaken here, this new lightweight locking would make >> synchronized work better with Loom: Because the lock-records are no longer >> scattered across the stack, but instead are densely packed into the >> lock-stack, it should be easy for a vthread to save its lock-stack upon >> unmounting and restore it when re-mounting. However, I am not sure about >> this, and this PR does not attempt to implement that support. >> >> Testing: >> - [x] tier1 x86_64 x aarch64 x +UseFastLocking >> - [x] tier2 x86_64 x aarch64 x +UseFastLocking >> - [x] tier3 x86_64 x aarch64 x +UseFastLocking >> - [x] tier4 x86_64 x aarch64 x +UseFastLocking >> - [x] tier1 x86_64 x aarch64 x -UseFastLocking >> - [x] tier2 x86_64 x aarch64 x -UseFastLocking >> - [x] tier3 x86_64 x aarch64 x -UseFastLocking >> - [x] tier4 x86_64 x aarch64 x -UseFastLocking >> - [x] Several real-world applications have been tested with this change in >> tandem with Lilliput without any problems, yet >> >> ### Performance >> >> #### Simple Microbenchmark >> >> The microbenchmark exercises only the locking primitives for monitorenter >> and monitorexit, without contention. The benchmark can be found >> (here)[https://github.com/rkennke/fastlockbench]. Numbers are in ns/ops. >> >> | | x86_64 | aarch64 | >> | -- | -- | -- | >> | -UseFastLocking | 20.651 | 20.764 | >> | +UseFastLocking | 18.896 | 18.908 | >> >> >> #### Renaissance >> >> | x86_64 | | | | aarch64 | | >> -- | -- | -- | -- | -- | -- | -- | -- >> | stack-locking | fast-locking | | | stack-locking | fast-locking | >> AkkaUct | 841.884 | 836.948 | 0.59% | | 1475.774 | 1465.647 | 0.69% >> Reactors | 11041.427 | 11181.451 | -1.25% | | 11381.751 | 11521.318 | >> -1.21% >> Als | 1367.183 | 1359.358 | 0.58% | | 1678.103 | 1688.067 | -0.59% >> ChiSquare | 577.021 | 577.398 | -0.07% | | 986.619 | 988.063 | -0.15% >> GaussMix | 817.459 | 819.073 | -0.20% | | 1154.293 | 1155.522 | -0.11% >> LogRegression | 598.343 | 603.371 | -0.83% | | 638.052 | 644.306 | -0.97% >> MovieLens | 8248.116 | 8314.576 | -0.80% | | 7569.219 | 7646.828 | -1.01%% >> NaiveBayes | 587.607 | 581.608 | 1.03% | | 541.583 | 550.059 | -1.54% >> PageRank | 3260.553 | 3263.472 | -0.09% | | 4376.405 | 4381.101 | -0.11% >> FjKmeans | 979.978 | 976.122 | 0.40% | | 774.312 | 771.235 | 0.40% >> FutureGenetic | 2187.369 | 2183.271 | 0.19% | | 2685.722 | 2689.056 | >> -0.12% >> ParMnemonics | 2434.551 | 2468.763 | -1.39% | | 4278.225 | 4263.863 | 0.34% >> Scrabble | 111.882 | 111.768 | 0.10% | | 151.796 | 153.959 | -1.40% >> RxScrabble | 210.252 | 211.38 | -0.53% | | 310.116 | 315.594 | -1.74% >> Dotty | 750.415 | 752.658 | -0.30% | | 1033.636 | 1036.168 | -0.24% >> ScalaDoku | 3072.05 | 3051.2 | 0.68% | | 3711.506 | 3690.04 | 0.58% >> ScalaKmeans | 211.427 | 209.957 | 0.70% | | 264.38 | 265.788 | -0.53% >> ScalaStmBench7 | 1017.795 | 1018.869 | -0.11% | | 1088.182 | 1092.266 | >> -0.37% >> Philosophers | 6450.124 | 6565.705 | -1.76% | | 12017.964 | 11902.559 | >> 0.97% >> FinagleChirper | 3953.623 | 3972.647 | -0.48% | | 4750.751 | 4769.274 | >> -0.39% >> FinagleHttp | 3970.526 | 4005.341 | -0.87% | | 5294.125 | 5296.224 | -0.04% > > Roman Kennke has updated the pull request incrementally with one additional > commit since the last revision: > > Only allow lock-stack verification for owning Java threads or at safepoints Slowdebug had a better stack trace: --------------- T H R E A D --------------- Current thread (0x00007fe4ad0062d0): WorkerThread "GC Thread#0" [id=19715, stack(0x0000700004416000,0x0000700004516000) (1024K)] Stack: [0x0000700004416000,0x0000700004516000], sp=0x00007000045152b0, free space=1020k Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code) V [libjvm.dylib+0x133a8a6] VMError::report_and_die(int, char const*, char const*, __va_list_tag*, Thread*, unsigned char*, void*, void*, char const*, int, unsigned long)+0x906 (javaThread.hpp:983) V [libjvm.dylib+0x133af59] VMError::report_and_die(Thread*, void*, char const*, int, char const*, char const*, __va_list_tag*)+0x89 V [libjvm.dylib+0x6d4e5c] report_vm_error(char const*, int, char const*, char const*, ...)+0x1ac V [libjvm.dylib+0xd33f] JavaThread::cast(Thread*)+0x4f V [libjvm.dylib+0x111911] JavaThread::current()+0x11 V [libjvm.dylib+0xdeffe9] LockStack::is_owning_thread() const+0x19 V [libjvm.dylib+0xdefe14] LockStack::verify(char const*) const+0x134 V [libjvm.dylib+0xac9367] LockStack::oops_do(OopClosure*)+0x27 V [libjvm.dylib+0xac92da] JavaThread::oops_do_no_frames(OopClosure*, CodeBlobClosure*)+0x2da V [libjvm.dylib+0x1287210] Thread::oops_do(OopClosure*, CodeBlobClosure*)+0x40 V [libjvm.dylib+0x129f395] ParallelOopsDoThreadClosure::do_thread(Thread*)+0x25 V [libjvm.dylib+0x129b6cc] Threads::possibly_parallel_threads_do(bool, ThreadClosure*)+0xfc V [libjvm.dylib+0x129dfad] Threads::possibly_parallel_oops_do(bool, OopClosure*, CodeBlobClosure*)+0x3d V [libjvm.dylib+0x99b3b6] G1RootProcessor::process_java_roots(G1RootClosures*, G1GCPhaseTimes*, unsigned int)+0xc6 V [libjvm.dylib+0x99b217] G1RootProcessor::evacuate_roots(G1ParScanThreadState*, unsigned int)+0x77 V [libjvm.dylib+0x9ac68e] G1EvacuateRegionsTask::scan_roots(G1ParScanThreadState*, unsigned int)+0x2e V [libjvm.dylib+0x9ac568] G1EvacuateRegionsBaseTask::work(unsigned int)+0x78 V [libjvm.dylib+0x13f75b4] WorkerTaskDispatcher::worker_run_task()+0x74 V [libjvm.dylib+0x13f7c14] WorkerThread::run()+0x34 V [libjvm.dylib+0x12868ee] Thread::call_run()+0x15e V [libjvm.dylib+0xfeafa7] thread_native_entry(Thread*)+0x117 C [libsystem_pthread.dylib+0x68fc] _pthread_start+0xe0 C [libsystem_pthread.dylib+0x2443] thread_start+0xf JavaThread 0x00007fe4af015610 (nid = 22019) was being processed Java frames: (J=compiled Java code, j=interpreted, Vv=VM code) j java.lang.ref.Reference.waitForReferencePendingList()V+0 java.base j java.lang.ref.Reference.processPendingReferences()V+0 java.base j java.lang.ref.Reference$ReferenceHandler.run()V+8 java.base v ~StubRoutines::call_stub 0x000000011fd08d21 Does that still match up with your theory? ------------- PR Comment: https://git.openjdk.org/jdk/pull/10907#issuecomment-1536554672
