Trace-based Just-in-Time Type Specialization for Dynamic Languages ∗+ ∗ ∗ ∗ ∗ Andreas Gal , Brendan Eich , Mike Shaver , David Anderson , David Mandelin , $ ∗ ∗ ∗ ∗ MohammadR.Haghighat ,BlakeKaplan ,GraydonHoare ,BorisZbarsky ,JasonOrendorff , ∗ # # + +# + Jesse Ruderman , Edwin Smith , Rick Reitmaier , Michael Bebenita , Mason Chang , Michael Franz ∗ Mozilla Corporation {gal,brendan,shaver,danderson,dmandelin,mrbkap,graydon,bz,jorendorff,jruderman}@mozilla.com AdobeCorporation# {edwsmith,rreitmai}@adobe.com Intel Corporation$ {mohammad.r.haghighat}@intel.com + University of California, Irvine {mbebenit,changm,franz}@uci.edu Abstract and is used for the application logic of browser-based productivity Dynamic languages such as JavaScript are more difficult to com- applications such as Google Mail, Google Docs and Zimbra Col- pile than statically typed ones. Since no concrete type information laboration Suite. In this domain, in order to provide a fluid user is available, traditional compilers need to emit generic code that can experienceandenableanewgenerationofapplications,virtualma- handle all possible type combinations at runtime. We present an al- chines must provide a low startup time and high performance. ternative compilation technique for dynamically-typed languages Compilers for statically typed languages rely on type informa- that identifies frequently executed loop traces at run-time and then tion to generate efficient machine code. In a dynamically typed pro- generates machine code on the fly that is specialized for the ac- gramming language such as JavaScript, the types of expressions tual dynamic types occurring on each path through the loop. Our may vary at runtime. This means that the compiler can no longer methodprovidescheapinter-procedural type specialization, and an easily transform operations into machine instructions that operate elegant and efficient way of incrementally compiling lazily discov- on one specific type. Without exact type information, the compiler ered alternative paths through nested loops. We have implemented must emit slower generalized machine code that can deal with all a dynamic compiler for JavaScript based on our technique and we potential type combinations. While compile-time static type infer- have measured speedups of 10x and more for certain benchmark ence might be able to gather type information to generate opti- programs. mized machine code, traditional static analysis is very expensive and hence not well suited for the highly interactive environment of Categories and Subject Descriptors D.3.4 [Programming Lan- a web browser. guages]: Processors — Incremental compilers, code generation. We present a trace-based compilation technique for dynamic General Terms Design, Experimentation, Measurement, Perfor- languages that reconciles speed of compilation with excellent per- mance. formanceofthegeneratedmachinecode.Oursystemusesamixed- modeexecutionapproach:thesystemstartsrunningJavaScriptina Keywords JavaScript, just-in-time compilation, trace trees. fast-starting bytecode interpreter. As the program runs, the system identifies hot (frequently executed) bytecode sequences, records 1. Introduction them, and compiles them to fast native code. We call such a se- DynamiclanguagessuchasJavaScript,Python,andRuby,arepop- quence of instructions a trace. ular since they are expressive, accessible to non-experts, and make Unlike method-based dynamic compilers, our dynamic com- deployment as easy as distributing a source file. They are used for piler operates at the granularity of individual loops. This design small scripts as well as for complex applications. JavaScript, for choice is based on the expectation that programs spend most of example, is the de facto standard for client-side web programming their time in hot loops. Even in dynamically typed languages, we expect hot loops to be mostly type-stable, meaning that the types of values are invariant. (12) For example, we would expect loop coun- ters that start as integers to remain integers for all iterations. When both of these expectations hold, a trace-based compiler can cover Permission to make digital or hard copies of all or part of this work for personal or the program execution with a small number of type-specialized, ef- classroom use is granted without fee provided that copies are not made or distributed ficiently compiled traces. for profit or commercial advantage and that copies bear this notice and the full citation Each compiled trace covers one path through the program with on the first page. To copy otherwise, to republish, to post on servers or to redistribute onemappingofvaluestotypes.WhentheVMexecutesacompiled to lists, requires prior specific permission and/or a fee. PLDI’09, June 15–20, 2009, Dublin, Ireland. trace, it cannot guarantee that the same path will be followed c Copyright 2009 ACM978-1-60558-392-1/09/06...$5.00 or that the same types will occur in subsequent loop iterations. Hence,recordingandcompilingatracespeculatesthatthepathand 1 for (var i = 2; i < 100; ++i) { typing will be exactly as they were during recording for subsequent 2 if (!primes[i]) iterations of the loop. 3 continue; Every compiled trace contains all the guards (checks) required 4 for (var k = i + i; i < 100; k += i) to validate the speculation. If one of the guards fails (if control 5 primes[k] = false; flow is different, or a value of a different type is generated), the 6 } trace exits. If an exit becomes hot, the VM can record a branch trace starting at the exit to cover the new path. In this way, the VM Figure 1. Sample program: sieve of Eratosthenes. primes is records a trace tree covering all the hot paths through the loop. initialized to an array of 100 false values on entry to this code Nested loops can be difficult to optimize for tracing VMs. In snippet. ¨ a naıve implementation, inner loops would become hot first, and the VM would start tracing there. When the inner loop exits, the VMwoulddetectthatadifferentbranchwastaken.TheVMwould try to record a branch trace, and find that the trace reaches not the Symbol Key inner loop header, but the outer loop header. At this point, the VM Interpret Bytecodes Overhead could continue tracing until it reaches the inner loop header again, thus tracing the outer loop inside a trace tree for the inner loop. loop Interpreting edge cold/blacklisted Butthis requires tracing a copy of the outer loop for every side exit loop/exit Native and type combination in the inner loop. In essence, this is a form abort Monitor compiled trace of unintended tail duplication, which can easily overflow the code recording ready cache. Alternatively, the VM could simply stop tracing, and give up Record hot Enter onever tracing outer loops. LIR Trace loop/exit Compiled Trace We solve the nested loop problem by recording nested trace finish at loop header loop edge with ¨ same types trees. Our system traces the inner loop exactly as the naıve version. Thesystem stops extending the inner tree when it reaches an outer Compile Execute loop, but then it starts a new trace at the outer loop header. When LIR Trace Compiled Trace the outer loop reaches the inner loop header, the system tries to call thetracetreefortheinnerloop.Ifthecallsucceeds,theVMrecords side exit, side exit to no existing trace existing trace the call to the inner tree as part of the outer trace and finishes Leave the outer trace as normal. In this way, our system can trace any Compiled Trace numberofloopsnestedtoanydepthwithoutcausing excessive tail duplication. These techniques allow a VM to dynamically translate a pro- Figure 2. State machine describing the major activities of Trace- gram to nested, type-specialized trace trees. Because traces can Monkey and the conditions that cause transitions to a new activ- cross function call boundaries, our techniques also achieve the ef- ity. In the dark box, TM executes JS as compiled traces. In the fects of inlining. Because traces have no internal control-flow joins, light gray boxes, TM executes JS in the standard interpreter. White they can be optimized in linear time by a simple compiler (10). boxes are overhead. Thus, to maximize performance, we need to Thus, our tracing VM efficiently performs the same kind of op- maximizetimespentinthedarkestboxandminimizetimespentin timizations that would require interprocedural analysis in a static the white boxes. The best case is a loop where the types at the loop optimization setting. This makes tracing an attractive and effective edge are the same as the types on entry–then TM can stay in native tool to type specialize even complex function call-rich code. code until the loop is done. Weimplementedthesetechniques for an existing JavaScript in- terpreter, SpiderMonkey. We call the resulting tracing VM Trace- Monkey. TraceMonkey supports all the JavaScript features of Spi- a set of industry benchmarks. The paper ends with conclusions in derMonkey, with a 2x-20x speedup for traceable programs. Section 9 and an outlook on future work is presented in Section 10. This paper makes the following contributions: • Weexplain an algorithm for dynamically forming trace trees to 2. Overview: Example Tracing Run coveraprogram,representingnestedloopsasnestedtracetrees. This section provides an overview of our system by describing how TraceMonkey executes an example program. The example • Weexplainhowtospeculativelygenerateefficienttype-specialized program, shown in Figure 1, computes the first 100 prime numbers code for traces from dynamic language programs. withnestedloops.ThenarrativeshouldbereadalongwithFigure2, • Wevalidate our tracing techniques in an implementation based which describes the activities TraceMonkey performs and when it on the SpiderMonkey JavaScript interpreter, achieving 2x-20x transitions between the loops. speedups on many programs. TraceMonkey always begins executing a program in the byte- code interpreter. Every loop back edge is a potential trace point. Theremainderofthispaperisorganizedasfollows.Section3is When the interpreter crosses a loop edge, TraceMonkey invokes a general overview of trace tree based compilation we use to cap- the trace monitor, which may decide to record or execute a native ture and compile frequently executed code regions. In Section 4 trace. At the start of execution, there are no compiled traces yet, so we describe our approach of covering nested loops using a num- thetracemonitorcountsthenumberoftimeseachloopbackedgeis ber of individual trace trees. In Section 5 we describe our trace- executeduntil a loop becomes hot, currently after 2 crossings. Note compilation based speculative type specialization approach we use that the way our loops are compiled, the loop edge is crossed before to generate efficient machine code from recorded bytecode traces. entering the loop, so the second crossing occurs immediately after Our implementation of a dynamic type-specializing compiler for the first iteration. JavaScript is described in Section 6. Related work is discussed in Here is the sequence of events broken down by outer loop Section 8. In Section 7 we evaluate our dynamic compiler based on iteration: v0 := ld state[748] // load primes from the trace activation record st sp[0], v0 // store primes to interpreter stack v1 := ld state[764] // load k from the trace activation record v2 := i2f(v1) // convert k from int to double st sp[8], v1 // store k to interpreter stack st sp[16], 0 // store false to interpreter stack v3 := ld v0[4] // load class word for primes v4 := and v3, -4 // mask out object class tag for primes v5 := eq v4, Array // test whether primes is an array xf v5 // side exit if v5 is false v6 := js_Array_set(v0, v2, false) // call function to set array element v7 := eq v6, 0 // test return value from call xt v7 // side exit if js_Array_set returns false. Figure 3. LIR snippet for sample program. This is the LIR recorded for line 5 of the sample program in Figure 1. The LIR encodes the semantics in SSA form using temporary variables. The LIR also encodes all the stores that the interpreter would do to its data stack. Sometimes these stores can be optimized away as the stack locations are live only on exits to the interpreter. Finally, the LIR records guards and side exits to verify the assumptions made in this recording: that primes is an array and that the call to set its element succeeds. mov edx, ebx(748) // load primes from the trace activation record mov edi(0), edx // (*) store primes to interpreter stack mov esi, ebx(764) // load k from the trace activation record mov edi(8), esi // (*) store k to interpreter stack mov edi(16), 0 // (*) store false to interpreter stack mov eax, edx(4) // (*) load object class word for primes and eax, -4 // (*) mask out object class tag for primes cmp eax, Array // (*) test whether primes is an array jne side_exit_1 // (*) side exit if primes is not an array sub esp, 8 // bump stack for call alignment convention push false // push last argument for call push esi // push first argument for call call js_Array_set // call function to set array element add esp, 8 // clean up extra stack space mov ecx, ebx // (*) created by register allocator test eax, eax // (*) test return value of js_Array_set je side_exit_2 // (*) side exit if call failed ... side_exit_1: mov ecx, ebp(-4) // restore ecx mov esp, ebp // restore esp jmp epilog // jump to ret statement Figure4. x86snippetforsampleprogram.Thisisthex86codecompiledfromtheLIRsnippetinFigure3.MostLIRinstructionscompile to a single x86 instruction. Instructions marked with (*) would be omitted by an idealized compiler that knew that none of the side exits would ever be taken. The 17 instructions generated by the compiler compare favorably with the 100+ instructions that the interpreter would execute for the same code snippet, including 4 indirect jumps. i=2. This is the first iteration of the outer loop. The loop on interpreter PC and the types of values match those observed when lines 4-5 becomes hot on its second iteration, so TraceMonkey en- trace recording was started. The first trace in our example, T , 45 ters recording mode on line 4. In recording mode, TraceMonkey covers lines 4 and 5. This trace can be entered if the PC is at line 4, records the code along the trace in a low-level compiler intermedi- iandkareintegers,andprimesisanobject.AftercompilingT , 45 ate representation we call LIR. The LIR trace encodes all the oper- TraceMonkeyreturns to the interpreter and loops back to line 1. ations performed and the types of all operands. The LIR trace also i=3. Now the loop header at line 1 has become hot, so Trace- encodes guards, which are checks that verify that the control flow Monkey starts recording. When recording reaches line 4, Trace- and types are identical to those observed during trace recording. Monkey observes that it has reached an inner loop header that al- Thus, on later executions, if and only if all guards are passed, the ready has a compiled trace, so TraceMonkey attempts to nest the trace has the required program semantics. inner loop inside the current trace. The first step is to call the inner TraceMonkey stops recording when execution returns to the trace as a subroutine. This executes the loop on line 4 to completion loop header or exits the loop. In this case, execution returns to the and then returns to the recorder. TraceMonkey verifies that the call loop header on line 4. wassuccessful and then records the call to the inner trace as part of After recording is finished, TraceMonkey compiles the trace to the current trace. Recording continues until execution reaches line native code using the recorded type information for optimization. 1, and at which point TraceMonkey finishes and compiles a trace The result is a native code fragment that can be entered if the for the outer loop, T . 16 i=4. On this iteration, TraceMonkey calls T . Because i=4, the Atrace records all its intermediate values in a small activation 16 if statement on line 2 is taken. This branch was not taken in the record area. To make variable accesses fast on trace, the trace also original trace, so this causes T to fail a guard and take a side exit. imports local and global variables by unboxing them and copying 16 The exit is not yet hot, so TraceMonkey returns to the interpreter, them to its activation record. Thus, the trace can read and write which executes the continue statement. thesevariableswithsimpleloadsandstoresfromanativeactivation i=5. TraceMonkeycalls T ,whichinturncallsthenestedtrace recording, independently of the boxing mechanism used by the 16 T . T loops back to its own header, starting the next iteration interpreter. When the trace exits, the VM boxes the values from 45 16 without ever returning to the monitor. this native storage location and copies them back to the interpreter i=6. On this iteration, the side exit on line 2 is taken again. This structures. time, the side exit becomes hot, so a trace T is recorded that For every control-flow branch in the source program, the 23,1 covers line 3 and returns to the loop header. Thus, the end of T recorder generates conditional exit LIR instructions. These instruc- 23,1 jumps directly to the start of T . The side exit is patched so that tions exit from the trace if required control flow is different from 16 onfuture iterations, it jumps directly to T . what it was at trace recording, ensuring that the trace instructions 23,1 Atthispoint,TraceMonkeyhascompiledenoughtracestocover are run only if they are supposed to. We call these instructions the entire nested loop structure, so the rest of the program runs guard instructions. entirely as native code. Mostofourtracesrepresentloopsandendwiththespecialloop LIR instruction. This is just an unconditional branch to the top of 3. Trace Trees the trace. Such traces return only via guards. Now, we describe the key optimizations that are performed as In this section, we describe traces, trace trees, and how they are part of recording LIR. All of these optimizations reduce complex formed at run time. Although our techniques apply to any dynamic dynamic language constructs to simple typed constructs by spe- language interpreter, we will describe them assuming a bytecode cializing for the current trace. Each optimization requires guard in- interpreter to keep the exposition simple. structions to verify their assumptions about the state and exit the trace if necessary. 3.1 Traces Typespecialization. Atrace is simply a program path, which may cross function call All LIR primitives apply to operands of specific types. Thus, boundaries. TraceMonkey focuses on loop traces, that originate at LIR traces are necessarily type-specialized, and a compiler can a loop edge and represent a single iteration through the associated easily produce a translation that requires no type dispatches. A loop. typical bytecode interpreter carries tag bits along with each value, Similar to an extended basic block, a trace is only entered at and to perform any operation, must check the tag bits, dynamically the top, but may have many exits. In contrast to an extended basic dispatch, mask out the tag bits to recover the untagged value, block, a trace can contain join nodes. Since a trace always only perform the operation, and then reapply tags. LIR omits everything followsonesinglepaththroughtheoriginalprogram,however,join except the operation itself. nodes are not recognizable as such in a trace and have a single Apotential problem is that some operations can produce values predecessor node like regular nodes. of unpredictable types. For example, reading a property from an Atypedtrace is a trace annotated with a type for every variable object could yield a value of any type, not necessarily the type (includingtemporaries)onthetrace.Atypedtracealsohasanentry observed during recording. The recorder emits guard instructions type map giving the required types for variables used on the trace that conditionally exit if the operation yields a value of a different before they are defined. For example, a trace could have a type map type from that seen during recording. These guard instructions (x: int, b: boolean), meaning that the trace may be entered guarantee that as long as execution is on trace, the types of values only if the value of the variable x is of type int and the value of b match those of the typed trace. When the VM observes a side exit is of type boolean. The entry type map is much like the signature along such a type guard, a new typed trace is recorded originating of a function. at the side exit location, capturing the new type of the operation in In this paper, we only discuss typed loop traces, and we will question. refer to them simply as “traces”. The key property of typed loop Representation specialization: objects. In JavaScript, name traces is that they can be compiled to efficient machine code using lookup semantics are complex and potentially expensive because the same techniques used for typed languages. they include features like object inheritance and eval. To evaluate In TraceMonkey, traces are recorded in trace-flavored SSA LIR an object property read expression like o.x, the interpreter must (low-level intermediate representation). In trace-flavored SSA (or search the property map of o and all of its prototypes and parents. TSSA), phi nodes appear only at the entry point, which is reached Property maps can be implemented with different data structures both on entry and via loop edges. The important LIR primitives (e.g., per-object hash tables or shared hash tables), so the search are constant values, memory loads and stores (by address and process also must dispatch on the representation of each object offset), integer operators, floating-point operators, function calls, foundduringsearch.TraceMonkeycansimplyobservetheresultof and conditional exits. Type conversions, such as integer to double, the search process and record the simplest possible LIR to access are represented by function calls. This makes the LIR used by the property value. For example, the search might finds the value of TraceMonkey independent of the concrete type system and type o.xintheprototypeofo,whichusesasharedhash-tablerepresen- conversion rules of the source language. The LIR operations are tation that places x in slot 2 of a property vector. Then the recorded generic enough that the backend compiler is language independent. cangenerateLIRthatreadso.xwithjusttwoorthreeloads:oneto Figure 3 shows an example LIR trace. get the prototype, possibly one to get the property value vector, and Bytecode interpreters typically represent values in a various one more to get slot 2 from the vector. This is a vast simplification complex data structures (e.g., hash tables) in a boxed format (i.e., and speedup compared to the original interpreter code. Inheritance with attached type tag bits). Since a trace is intended to represent relationships and object representations can change during execu- efficient code that eliminates all that complexity, our traces oper- tion, so the simplified code requires guard instructions that ensure ate on unboxed values in simple variables and arrays as much as theobjectrepresentationisthesame.InTraceMonkey,objects’rep- possible. resentations are assigned an integer key called the object shape. Starting a tree. Tree trees always start at loop headers, because Thus, the guard is a simple equality check on the object shape. they are a natural place to look for hot paths. In TraceMonkey, loop Representation specialization: numbers. JavaScript has no headers are easy to detect–the bytecode compiler ensures that a integer type, only a Number type that is the set of 64-bit IEEE- bytecode is a loop header iff it is the target of a backward branch. 754 floating-pointer numbers (“doubles”). But many JavaScript TraceMonkey starts a tree when a given loop header has been exe- operators, in particular array accesses and bitwise operators, really cuted a certain number of times (2 in the current implementation). operate on integers, so they first convert the number to an integer, Starting a tree just means starting recording a trace for the current and then convert any integer result back to a double.1 Clearly, a point and type map and marking the trace as the root of a tree. Each JavaScript VM that wants to be fast must find a way to operate on tree is associated with a loop header and type map, so there may be integers directly and avoid these conversions. several trees for a given loop header. In TraceMonkey, we support two representations for numbers: Closing the loop. Trace recording can end in several ways. integers and doubles. The interpreter uses integer representations Ideally, the trace reaches the loop header where it started with as muchasitcan,switchingforresultsthat can only be represented the same type map as on entry. This is called a type-stable loop as doubles. When a trace is started, some values may be imported iteration. In this case, the end of the trace can jump right to the and represented as integers. Some operations on integers require beginning, as all the value representations are exactly as needed to guards. For example, adding two integers can produce a value too enter the trace. The jump can even skip the usual code that would large for the integer representation. copy out the state at the end of the trace and copy it back in to the Function inlining. LIR traces can cross function boundaries trace activation record to enter a trace. in either direction, achieving function inlining. Move instructions In certain cases the trace might reach the loop header with a need to be recorded for function entry and exit to copy arguments different type map. This scenario is sometime observed for the first in and return values out. These move statements are then optimized iteration of a loop. Some variables inside the loop might initially be away by the compiler using copy propagation. In order to be able undefined,beforetheyaresettoaconcretetypeduringthefirstloop to return to the interpreter, the trace must also generate LIR to iteration. When recording such an iteration, the recorder cannot record that a call frame has been entered and exited. The frame link the trace back to its own loop header since it is type-unstable. entry and exit LIR saves just enough information to allow the Instead, the iteration is terminated with a side exit that will always intepreter call stack to be restored later and is much simpler than fail and return to the interpreter. At the same time a new trace is the interpreter’s standard call code. If the function being entered recorded with the new type map. Every time an additional type- is not constant (which in JavaScript includes any call by function unstable trace is added to a region, its exit type map is compared to name), the recorder must also emit LIR to guard that the function the entry map of all existing traces in case they complement each is the same. other. With this approach we are able to cover type-unstable loop Guards and side exits. Each optimization described above iterations as long they eventually form a stable equilibrium. requires one or more guards to verify the assumptions made in Finally, the trace might exit the loop before reaching the loop doing the optimization. A guard is just a group of LIR instructions header, for example because execution reaches a break or return that performs a test and conditional exit. The exit branches to a statement. In this case, the VM simply ends the trace with an exit side exit, a small off-trace piece of LIR that returns a pointer to to the trace monitor. a structure that describes the reason for the exit along with the As mentioned previously, we may speculatively chose to rep- interpreter PC at the exit point and any other data needed to restore resent certain Number-typed values as integers on trace. We do so the interpreter’s state structures. when we observe that Number-typed variables contain an integer Aborts. Some constructs are difficult to record in LIR traces. value at trace entry. If during trace recording the variable is unex- For example, eval or calls to external functions can change the pectedly assigned a non-integer value, we have to widen the type program state in unpredictable ways, making it difficult for the of the variable to a double. As a result, the recorded trace becomes tracer to know the current type map in order to continue tracing. inherently type-unstable since it starts with an integer value but Atracing implementation can also have any number of other limi- ends with a double value. This represents a mis-speculation, since tations, e.g.,a small-memory device may limit the length of traces. at trace entry we specialized the Number-typed value to an integer, When any situation occurs that prevents the implementation from assumingthatattheloopedgewewouldagainfindanintegervalue continuingtracerecording,theimplementationabortstracerecord- in the variable, allowing us to close the loop. To avoid future spec- ing and returns to the trace monitor. ulative failures involving this variable, and to obtain a type-stable trace we note the fact that the variable in question as been observed 3.2 Trace Trees to sometimes hold non-integer values in an advisory data structure which we call the “oracle”. Especially simple loops, namely those where control flow, value Whencompiling loops, we consult the oracle before specializ- types, value representations, and inlined functions are all invariant, ing values to integers. Speculation towards integers is performed can be represented by a single trace. But most loops have at least only if no adverse information is known to the oracle about that some variation, and so the program will take side exits from the particular variable. Whenever we accidentally compile a loop that main trace. When a side exit becomes hot, TraceMonkey starts a is type-unstable due to mis-speculation of a Number-typed vari- newbranchtrace from that point and patches the side exit to jump able, we immediately trigger the recording of a new trace, which directly to that trace. In this way, a single trace expands on demand based on the now updated oracle information will start with a dou- to a single-entry, multiple-exit trace tree. ble value and thus become type stable. This section explains how trace trees are formed during execu- Extending a tree. Side exits lead to different paths through tion. The goal is to form trace trees during execution that cover all the loop, or paths with different types or representations. Thus, to the hot paths of the program. completelycovertheloop,theVMmustrecordtracesstartingatall side exits. These traces are recorded much like root traces: there is 1 acounterforeachsideexit,andwhenthecounterreachesahotness Arrays are actually worse than this: if the index value is a number, it must threshold, recording starts. Recording stops exactly as for the root be converted from a double to a string for the property access operator, and trace, using the loop header of the root trace as the target to reach. then to an integer internally to the array implementation. Our implementation does not extend at all side exits. It extends only if the side exit is for a control-flow branch, and only if the side T exit does not leave the loop. In particular we do not want to extend Tree
Anchor a trace tree along a path that leads to an outer loop, because we Trunk
Trace want to cover such paths in an outer tree through tree nesting. Trace
Anchor Branch
Trace 3.3 Blacklisting Guard Sometimes, a program follows a path that cannot be compiled Side
Exit into a trace, usually because of limitations in the implementation. TraceMonkey does not currently support recording throwing and catching of arbitrary exceptions. This design trade off was chosen, because exceptions are usually rare in JavaScript. However, if a program opts to use exceptions intensively, we would suddenly incur a punishing runtime overhead if we repeatedly try to record a trace for this path and repeatedly fail to do so, since we abort tracing every time we observe an exception being thrown. Figure 5. A tree with two traces, a trunk trace and one branch Asaresult,ifahotloopcontainstracesthatalwaysfail,theVM trace. The trunk trace contains a guard to which a branch trace was could potentially run much more slowly than the base interpreter: attached. The branch trace contain a guard that may fail and trigger the VMrepeatedly spends time trying to record traces, but is never a side exit. Both the trunk and the branch trace loop back to the tree able to run any. To avoid this problem, whenever the VM is about anchor, which is the beginning of the trace tree. to start tracing, it must try to predict whether it will finish the trace. Our prediction algorithm is based on blacklisting traces that havebeentriedandfailed.WhentheVMfailstofinishatracestart- Trace
1 Trace
2 Trace
1 Trace
2 ing at a given point, the VM records that a failure has occurred. The VMalsosetsacountersothatitwillnottrytorecordatracestarting Number Boolean Number Boolean at that point until it is passed a few more times (32 in our imple- mentation). This backoff counter gives temporary conditions that prevent tracing a chance to end. For example, a loop may behave differently during startup than during its steady-state execution. Af- Number Number Boolean Number ter a given number of failures (2 in our implementation), the VM Closed Linked Linked Linked marksthefragmentasblacklisted, which means the VM will never (a) (b) again start recording at that point. Trace
1 Trace
2 Trace
3 After implementing this basic strategy, we observed that for small loops that get blacklisted, the system can spend a noticeable Number Boolean String amountoftimejustfindingtheloopfragmentanddeterminingthat it has been blacklisted. We now avoid that problem by patching the bytecode. We define an extra no-op bytecode that indicates a loop String Number String header. The VM calls into the trace monitor every time the inter- String preter executes a loop header no-op. To blacklist a fragment, we Linked Linked Closed simply replace the loop header no-op with a regular no-op. Thus, Linked (c) the interpreter will never again even call into the trace monitor. Thereisarelatedproblemwehavenotyetsolved,whichoccurs Figure 6. We handle type-unstable loops by allowing traces to whenaloopmeetsalloftheseconditions: • The VMcanformatleastoneroottrace for the loop. compile that cannot loop back to themselves due to a type mis- match.Assuchtracesaccumulate,weattempttoconnecttheirloop • There is at least one hot side exit for which the VM cannot edges to form groups of trace trees that can execute without having complete a trace. to side-exit to the interpreter to cover odd type cases. This is par- • The loop body is short. ticularly important for nested trace trees where an outer tree tries to call an inner tree (or in this case a forest of inner trees), since inner In this case, the VM will repeatedly pass the loop header, search loops frequently have initially undefined values which change type for a trace, find it, execute it, and fall back to the interpreter. to a concrete value after the first iteration. With a short loop body, the overhead of finding and calling the trace is high, and causes performance to be even slower than the through the inner loop, {i ,i ,i ,α}. The α symbol is used to basic interpreter. So far, in this situation we have improved the 2 3 5 implementation so that the VM can complete the branch trace. indicate that the trace loops back the tree anchor. But it is hard to guarantee that this situation will never happen. Whenexecution leaves the inner loop, the basic design has two As future work, this situation could be avoided by detecting and choices. First, the system can stop tracing and give up on compiling blacklisting loops for which the average trace call executes few the outer loop, clearly an undesirable solution. The other choice is bytecodes before returning to the interpreter. to continue tracing, compiling traces for the outer loop inside the inner loop’s trace tree. For example, the program might exit at i and record a branch 4. Nested Trace Tree Formation 5 trace that incorporates the outer loop: {i ,i ,i ,i ,i ,i ,α}. 5 7 1 6 7 1 Figure 7 shows basic trace tree compilation (11) applied to a nested Later, the program might take the other branch at i and then 2 loop where the inner loop contains two paths. Usually, the inner exit, recording another branch trace incorporating the outer loop: loop (with header at i ) becomes hot first, and a trace tree is rooted {i ,i ,i ,i ,i ,i ,i ,i ,α}. Thus, the outer loop is recorded and 2 2 4 5 7 1 6 7 1 at that point. For example, the first recorded trace may be a cycle compiledtwice,andbothcopiesmustberetainedinthetracecache. i1 Outer
Tree i1 t1 t1 Nested
Tree Tree
Call i2 i2 t2 Nested
Tree i3 i6 i3 i4 t2 Exit
Guard i4 t4 i5 i5 i7 Exit
Guard i6 (a) (b) Figure8. Controlflowgraphofaloopwithtwonestedloops(left) and its nested trace tree configuration (right). The outer tree calls Figure 7. Control flow graph of a nested loop with an if statement the two inner nested trace trees and places guards at their side exit inside the inner most loop (a). An inner tree captures the inner locations. loop, and is nested inside an outer tree which “calls” the inner tree. The inner tree returns to the outer tree once it exits along its loop condition guard (b). loop is entered with m different type maps (on geometric average), then we compile O(mk) copies of the innermost loop. As long as In general, if loops are nested to depth k, and each loop has n paths miscloseto1,theresulting trace trees will be tractable. ¨ k Animportantdetailisthatthecalltotheinnertracetreemustact (on geometric average), this naıve strategy yields O(n ) traces, like a function call site: it must return to the same point every time. which can easily fill the trace cache. The goal of nesting is to make inner and outer loops independent; In order to execute programs with nested loops efficiently, a thus when the inner tree is called, it must exit to the same point tracing system needs a technique for covering the nested loops with in the outer tree every time with the same type map. Because we native code without exponential trace duplication. cannot actually guarantee this property, we must guard on it after 4.1 Nesting Algorithm the call, and side exit if the property does not hold. A common The key insight is that if each loop is represented by its own trace reason for the inner tree not to return to the same point would tree, the code for each loop can be contained only in its own tree, be if the inner tree took a new side exit for which it had never andouterlooppathswillnotbeduplicated.Anotherkeyfactisthat compiled a trace. At this point, the interpreter PC is in the inner wearenottracing arbitrary bytecodes that might have irreduceable tree, so we cannot continue recording or executing the outer tree. control flow graphs, but rather bytecodes produced by a compiler If this happens during recording, we abort the outer trace, to give for a language with structured control flow. Thus, given two loop the inner tree a chance to finish growing. A future execution of the edges, the system can easily determine whether they are nested outer tree would then be able to properly finish and record a call to and which is the inner loop. Using this knowledge, the system can the inner tree. If an inner tree side exit happens during execution of compileinnerandouterloopsseparately,andmaketheouterloop’s a compiled trace for the outer tree, we simply exit the outer trace traces call the inner loop’s trace tree. and start recording a new branch in the inner tree. The algorithm for building nested trace trees is as follows. We 4.2 Blacklisting with Nesting start tracing at loop headers exactly as in the basic tracing system. When we exit a loop (detected by comparing the interpreter PC The blacklisting algorithm needs modification to work well with with the range given by the loop edge), we stop the trace. The nesting. The problem is that outer loop traces often abort during key step of the algorithm occurs when we are recording a trace startup (because the inner tree is not available or takes a side exit), for loop LR (R for loop being recorded) and we reach the header which would lead to their being quickly blacklisted by the basic of a different loop LO (O for other loop). Note that LO must be an algorithm. inner loop of LR because we stop the trace when we exit a loop. The key observation is that when an outer trace aborts because • If L has a type-matching compiled trace tree, we call L as the inner tree is not ready, this is probably a temporary condition. O O Thus, we should not count such aborts toward blacklisting as long a nested trace tree. If the call succeeds, then we record the call as we are able to build up more traces for the inner tree. in the trace for LR. On future executions, the trace for LR will In our implementation, when an outer tree aborts on the inner call the inner trace directly. tree, we increment the outer tree’s blacklist counter as usual and • If LO does not have a type-matching compiled trace tree yet, back off on compiling it. When the inner tree finishes a trace, we we have to obtain it before we are able to proceed. In order decrement the blacklist counter on the outer loop, “forgiving” the to do this, we simply abort recording the first trace. The trace outer loop for aborting previously. We also undo the backoff so that monitor will see the inner loop header, and will immediately the outer tree can start immediately trying to compile the next time start recording the inner loop. 2 wereachit. If all the loops in a nest are type-stable, then loop nesting creates noduplication.Otherwise,ifloopsarenestedtoadepthk,andeach 5. Trace Tree Optimization 2 Instead of aborting the outer recording, we could principally merely sus- This section explains how a recorded trace is translated to an pend the recording, but that would require the implementation to be able optimized machine code trace. The trace compilation subsystem, to record several traces simultaneously, complicating the implementation, NANOJIT, is separate from the VM and can be used for other while saving only a few iterations in the interpreter. applications. 5.1 Optimizations Tag JS Type Description Because traces are in SSA form and have no join points or φ- xx1 number 31-bit integer representation nodes, certain optimizations are easy to implement. In order to 000 object pointer to JSObject handle get good startup performance, the optimizations must run quickly, 010 number pointer to double handle so we chose a small set of optimizations. We implemented the 100 string pointer to JSString handle optimizations as pipelined filters so that they can be turned on and 110 boolean enumeration for null, undefined, true, false off independently, and yet all run in just two loop passes over the null, or trace: one forward and one backward. undefined Every time the trace recorder emits a LIR instruction, the in- Figure 9. Tagged values in the SpiderMonkey JS interpreter. struction is immediately passed to the first filter in the forward Testing tags, unboxing (extracting the untagged value) and boxing pipeline. Thus, forward filter optimizations are performed as the (creating tagged values) are significant costs. Avoiding these costs trace is recorded. Each filter may pass each instruction to the next is a key benefit of tracing. filter unchanged, write a different instruction to the next filter, or write no instruction at all. For example, the constant folding filter can replace a multiply instruction like v13 := mul3,1000 with a constant instruction v13 = 3000. heuristic selects v with minimum vm. The motivation is that this Wecurrently apply four forward filters: frees up a register for as long as possible given a single spill. If we need to spill a value vs at this point, we generate the • On ISAs without floating-point instructions, a soft-float filter restore code just after the code for the current instruction. The converts floating-point LIR instructions to sequences of integer corresponding spill code is generated just after the last point where instructions. v wasused.Theregisterthatwasassignedtov ismarkedfreefor s s • CSE(constant subexpression elimination), the preceding code, because that register can now be used freely • expression simplification, including constant folding and a few without affecting the following code algebraic identities (e.g., a − a = 0), and 6. Implementation • source language semantic-specific expression simplification, To demonstrate the effectiveness of our approach, we have im- primarily algebraic identities that allow DOUBLE to be replaced with INT. For example, LIR that converts an INT to a DOUBLE plemented a trace-based dynamic compiler for the SpiderMonkey and then back again would be removed by this filter. JavaScript Virtual Machine (4). SpiderMonkey is the JavaScript When trace recording is completed, nanojit runs the backward VMembeddedinMozilla’s Firefox open-source web browser (2), optimization filters. These are used for optimizations that require whichisusedbymorethan200millionusersworld-wide.Thecore backward program analysis. When running the backward filters, of SpiderMonkey is a bytecode interpreter implemented in C++. nanojit reads one LIR instruction at a time, and the reads are passed In SpiderMonkey, all JavaScript values are represented by the through the pipeline. type jsval. A jsval is machine word in which up to the 3 of the Wecurrently apply three backward filters: least significant bits are a type tag, and the remaining bits are data. See Figure 6 for details. All pointers contained in jsvals point to • Deaddata-stackstoreelimination. The LIR trace encodes many GC-controlled blocks aligned on 8-byte boundaries. stores to locations in the interpreter stack. But these values are JavaScript object values are mappings of string-valued property never read back before exiting the trace (by the interpreter or namestoarbitrary values. They are represented in one of two ways another trace). Thus, stores to the stack that are overwritten in SpiderMonkey. Most objects are represented by a shared struc- before the next exit are dead. Stores to locations that are off tural description, called the object shape, that maps property names the top of the interpreter stack at future exits are also dead. to array indexes using a hash table. The object stores a pointer to the shape and the array of its own property values. Objects with • Deadcall-stackstore elimination. This is the same optimization large, unique sets of property names store their properties directly as above, except applied to the interpreter’s call stack used for in a hash table. function call inlining. The garbage collector is an exact, non-generational, stop-the- • Dead code elimination. This eliminates any operation that world mark-and-sweep collector. stores to a value that is never used. In the rest of this section we discuss key areas of the TraceMon- key implementation. After a LIR instruction is successfully read (“pulled”) from 6.1 Calling Compiled Traces the backward filter pipeline, nanojit’s code generator emits native machine instruction(s) for it. Compiled traces are stored in a trace cache, indexed by intepreter 5.2 Register Allocation PC and type map. Traces are compiled so that they may be called as functions using standard native calling conventions (e.g., We use a simple greedy register allocator that makes a single FASTCALLonx86). backward pass over the trace (it is integrated with the code gen- The interpreter must hit a loop edge and enter the monitor in erator). By the time the allocator has reached an instruction like order to call a native trace for the first time. The monitor computes v3 = add v1,v2, it has already assigned a register to v3. If v1 and the current type map, checks the trace cache for a trace for the v2 have not yet been assigned registers, the allocator assigns a free current PC and type map, and if it finds one, executes the trace. register to each. If there are no free registers, a value is selected for To execute a trace, the monitor must build a trace activation spilling. We use a class heuristic that selects the “oldest” register- record containing imported local and global variables, temporary carried value (6). stack space, and space for arguments to native calls. The local and Theheuristic considers the set R of values v in registers imme- global values are then copied from the interpreter state to the trace diately after the current instruction for spilling. Let vm be the last activation record. Then, the trace is called like a normal C function instruction before the current where each v is referred to. Then the pointer. When a trace call returns, the monitor restores the interpreter Recording is activated by a pointer swap that sets the inter- state. First, the monitor checks the reason for the trace exit and preter’s dispatch table to call a single “interrupt” routine for ev- applies blacklisting if needed. Then, it pops or synthesizes inter- ery bytecode. The interrupt routine first calls a bytecode-specific preter JavaScript call stack frames as needed. Finally, it copies the recording routine. Then, it turns off recording if necessary (e.g., imported variables back from the trace activation record to the in- the trace ended). Finally, it jumps to the standard interpreter byte- terpreter state. codeimplementation.Somebytecodeshaveeffectsonthetypemap At least in the current implementation, these steps have a non- that cannot be predicted before executing the bytecode (e.g., call- negligible runtime cost, so minimizing the number of interpreter- ing String.charCodeAt, which returns an integer or NaN if the to-trace and trace-to-interpreter transitions is essential for perfor- index argument is out of range). For these, we arrange for the inter- mance. (see also Section 3.3). Our experiments (see Figure 12) preter to call into the recorder again after executing the bytecode. show that for programs we can trace well such transitions hap- Since such hooks are relatively rare, we embed them directly into pen infrequently and hence do not contribute significantly to total the interpreter, with an additional runtime check to see whether a runtime. In a few programs, where the system is prevented from recorder is currently active. recording branch traces for hot side exits by aborts, this cost can Whileseparating the interpreter from the recorder reduces indi- rise to up to 10% of total execution time. vidual code complexity, it also requires careful implementation and extensive testing to achieve semantic equivalence. 6.2 Trace Stitching In some cases achieving this equivalence is difficult since Spi- Transitions from a trace to a branch trace at a side exit avoid the derMonkey follows a fat-bytecode design, which was found to be costs of calling traces from the monitor, in a feature called trace beneficial to pure interpreter performance. stitching. At a side exit, the exiting trace only needs to write live In fat-bytecode designs, individual bytecodes can implement register-carried values back to its trace activation record. In our im- complex processing (e.g., the getprop bytecode, which imple- plementation, identical type maps yield identical activation record mentsfullJavaScript property value access, including special cases layouts, so the trace activation record can be reused immediately for cached and dense array access). bythe branch trace. Fat bytecodes have two advantages: fewer bytecodes means In programs with branchy trace trees with small traces, trace lower dispatch cost, and bigger bytecode implementations give the stitching has a noticeable cost. Although writing to memory and compiler more opportunities to optimize the interpreter. then soon reading back would be expected to have a high L1 Fat bytecodes are a problem for TraceMonkey because they cache hit rate, for small traces the increased instruction count has require the recorder to reimplement the same special case logic a noticeable cost. Also, if the writes and reads are very close in the same way. Also, the advantages are reduced because (a) in the dynamic instruction stream, we have found that current dispatch costs are eliminated entirely in compiled traces, (b) the x86 processors often incur penalties of 6 cycles or more (e.g., if traces contain only one special case, not the interpreter’s large the instructions use different base registers with equal values, the chunk of code, and (c) TraceMonkey spends less time running the processor may not be able to detect that the addresses are the same base interpreter. right away). Onewaywehavemitigatedtheseproblemsisbyimplementing The alternate solution is to recompile an entire trace tree, thus certain complex bytecodes in the recorder as sequences of simple achieving inter-trace register allocation (10). The disadvantage is bytecodes.Expressingtheoriginalsemanticsthiswayisnottoodif- that tree recompilation takes time quadratic in the number of traces. ficult, and recording simple bytecodes is much easier. This enables We believe that the cost of recompiling a trace tree every time us to retain the advantages of fat bytecodes while avoiding some of a branch is added would be prohibitive. That problem might be their problems for trace recording. This is particularly effective for mitigated by recompiling only at certain points, or only for very fat bytecodes that recurse back into the interpreter, for example to hot, stable trees. convert an object into a primitive value by invoking a well-known In the future, multicore hardware is expected to be common, methodontheobject, since it lets us inline this function call. making background tree recompilation attractive. In a closely re- It is important to note that we split fat opcodes into thinner op- lated project (13) background recompilation yielded speedups of codes only during recording. When running purely interpretatively up to 1.25x on benchmarks with many branch traces. We plan to (i.e. code that has been blacklisted), the interpreter directly and ef- apply this technique to TraceMonkey as future work. ficiently executes the fat opcodes. 6.3 Trace Recording ThejobofthetracerecorderistoemitLIRwithidenticalsemantics 6.4 Preemption to the currently running interpreter bytecode trace. A good imple- SpiderMonkey,likemanyVMs,needstopreempttheuserprogram mentation should have low impact on non-tracing interpreter per- periodically. The main reasons are to prevent infinitely looping formance and a convenient way for implementers to maintain se- scripts from locking up the host system and to schedule GC. mantic equivalence. In the interpreter, this had been implemented by setting a “pre- In our implementation, the only direct modification to the inter- empt now” flag that was checked on every backward jump. This preter is a call to the trace monitor at loop edges. In our benchmark strategy carried over into TraceMonkey: the VM inserts a guard on results (see Figure 12) the total time spent in the monitor (for all the preemption flag at every loop edge. We measured less than a activities) is usually less than 5%, so we consider the interpreter 1%increase in runtime on most benchmarks for this extra guard. impact requirement met. Incrementing the loop hit counter is ex- In practice, the cost is detectable only for programs with very short pensivebecauseitrequiresustolookuptheloopinthetracecache, loops. but we have tuned our loops to become hot and trace very quickly We tested and rejected a solution that avoided the guards by (on the second iteration). The hit counter implementation could be compiling the loop edge as an unconditional jump, and patching improved, which might give us a small increase in overall perfor- the jump target to an exit routine when preemption is required. mance, as well as more flexibility with tuning hotness thresholds. This solution can make the normal case slightly faster, but then Once a loop is blacklisted we never call into the trace monitor for preemption becomes very slow. The implementation was also very that loop (see Section 3.3). complex,especiallytryingtorestartexecutionafterthepreemption. 6.5 Calling External Functions ?>9@AJ.D2.@A:0>#3$4,56# Like most interpreters, SpiderMonkey has a foreign function inter- ?>9@AJ.0A:9@AJ.>9@AJ.B<#3$4(56# (e.g., web browser control and DOM access). The FFI has a stan- ?>9@AJ.1;.?:2/>9;.:<9I;./89-@/#3'4,56# FFIinteract with the program state through an interpreter API (e.g., -<>2.B897<>.5:<91#3$4!56# to read a property from an argument). There are also certain inter- -<>2.B897<>.>8H2#3$4$56# /9=:>8.?;<$#3(4,56# preter builtins that do not use the FFI, but interact with the program /9=:>8.7-(#3%4&56# state in the same way, such as the CallIteratorNext function /9=:>8.<2?#3$4)56# /8A>98FG8E.92/09?@D2#3$4!56# used with iterator objects. TraceMonkey must support this FFI in 1@>8:?.A?@2D2.1@>?#3%4*56# order to speed up code that interacts with the host system inside hot 1@>8:?.1@>E@?2.8:?.1@>?.@A.1=>2#3+4*56# loops. 1@>8:?.&1@>.1@>?.@A.1=>2#3%(4(56# Calling external functions from TraceMonkey is potentially dif- 922?#3!4,56# &-.9<=>929:92># L?J+B:F>*:?7-<#0(1923# mance. =<6>?J+-?7:,A+,5*/#0(1$23# =<6>?J+<:J,F5-*#0(1(23# Call threading, also known as context threading (8), compiles =<6>?J+@:=<:#0(1C23# methods by generating a native call instruction to an interpreter =<6>?J+.:=/&%#0$1C23# 6/J/27+*?:#0%1$23# method for each interpreter bytecode. A call-return pair has been 4:<8+=7/,<6:F+?564#0D1(23# 4:<8+7:6I:F+=-4=#0C1923# showntobeapotentially much more efficient dispatch mechanism 4:<8+,56*>,#0%1923# than the indirect jumps used in standard bytecode interpreters. *:B/#0(1!23# dispatch overhead. .><57=+?=>/B/+.><=#0$1D23# .><57=+.>=/+:?*#0$C1$23# Neithercallthreadingnorinlinethreadingperformtypespecial- .><57=+.><=+>?+.;<57=+).><+.><=+>?+.;/B/#0)1!23# Apple’s SquirrelFish Extreme (5) is a JavaScript implementa- :,,/==+?.5*;#0%1$23# tion based on call threading with selective inline threading. Com- :,,/==+@:??A-,8#0$1$23# :,,/==+.>?:6;+<6//=#0!1923# bined with efficient interpreter engineering, these threading tech- )*+6:;<6:,/#0(1$23# niques have given SFX excellent performance on the standard Sun- )*+45678#0$1923# )*+,-./#0$1$23# Spider benchmarks. !"# $!"# %!"# &!"# '!"# (!!"# Google’s V8 is a JavaScript implementation primarily based on inline threading, with call threading only for very complex K?<56# M/,56*# N547>F/# N:FF#O6:,/# M-?#O6:,/# operations. Figure 12. Fraction of time spent on major VM activities. The 9. Conclusions speedup vs. interpreter is shown in parentheses next to each test. This paper described how to run dynamic languages efficiently by Most programs where the VM spends the majority of its time run- recording hot traces and generating type-specialized native code. ning native code have a good speedup. Recording and compilation Our technique focuses on aggressively inlined loops, and for each costs can be substantial; speeding up those parts of the implemen- loop, it generates a tree of native code traces representing the tation would improve SunSpider performance. paths and value types through the loop observed at run time. We explained how to identify loop nesting relationships and generate nested traces in order to avoid excessive code duplication due to the many paths through a loop nest. We described our type specialization algorithm. We also described our trace compiler, inner loops become hot first), leading to much greater tail duplica- which translates a trace from an intermediate representation to tion. optimized native code in two linear passes. YETI, from Zaleski et al. (19) applied Dynamo-style tracing Our experimental results show that in practice loops typically to Java in order to achieve inlining, indirect jump elimination, are entered with only a few different combinations of value types and other optimizations. Their primary focus was on designing an of variables. Thus, a small number of traces per loop is sufficient interpreter that could easily be gradually re-engineered as a tracing to run a program efficiently. Our experiments also show that on VM. programs amenable to tracing, we achieve speedups of 2x to 20x. Suganumaetal.(18)describedregion-basedcompilation(RBC), a relative of tracing. A region is an subprogram worth optimizing 10. Future Work that can include subsets of any number of methods. Thus, the com- Work is underway in a number of areas to further improve the piler has more flexibility and can potentially generate better code, performance of our trace-based JavaScript compiler. We currently buttheprofilingandcompilationsystemsarecorrespondinglymore do not trace across recursive function calls, but plan to add the complex. support for this capability in the near term. We are also exploring Type specialization for dynamic languages. Dynamic lan- adoption of the existing work on tree recompilation in the context guage implementors have long recognized the importance of type of the presented dynamic compiler in order to minimize JIT pause specialization for performance. Most previous work has focused on times and obtain the best of both worlds, fast tree stitching as well methods instead of traces. as the improved code quality due to tree recompilation. Chambers et. al (9) pioneered the idea of compiling multiple We also plan on adding support for tracing across regular ex- versions of a procedure specialized for the input types in the lan- pression substitutions using lambda functions, function applica- guage Self. In one implementation, they generated a specialized tions and expression evaluation using eval. All these language methodonlineeachtimeamethodwascalledwithnewinputtypes. constructs are currently executed via interpretation, which limits In another, they used an offline whole-program static analysis to our performance for applications that use those features. infer input types and constant receiver types at call sites. Interest- ingly, the two techniques produced nearly the same performance. Acknowledgments Salib (17) designed a type inference algorithm for Python based ontheCartesian Product Algorithm and used the results to special- Parts of this effort have been sponsored by the National Science ize on types and translate the program to C++. Foundation under grants CNS-0615443 and CNS-0627747, as well McCloskey (14) has work in progress based on a language- as by the California MICRO Program and industrial sponsor Sun independent type inference that is used to generate efficient C Microsystems under Project No. 07-127. implementations of JavaScript and Python programs. The U.S. Government is authorized to reproduce and distribute Native code generation by interpreters. The traditional inter- reprints for Governmental purposes notwithstanding any copyright preter design is a virtual machine that directly executes ASTs or annotation thereon. Any opinions, findings, and conclusions or rec- machine-code-likebytecodes.Researchershaveshownhowtogen- ommendations expressed here are those of the author and should not be interpreted as necessarily representing the official views, [10] A. Gal. Efficient Bytecode Verification and Compilation in a Virtual policies or endorsements, either expressed or implied, of the Na- Machine Dissertation. PhD thesis, University Of California, Irvine, tional Science foundation(NSF),anyotheragencyoftheU.S.Gov- 2006. ernment, or any of the companies mentioned above. [11] A. Gal, C. W. Probst, and M. Franz. HotpathVM: An effective JIT compiler for resource-constrained devices. In Proceedings of the References International Conference on Virtual Execution Environments, pages [1] LuaJIT roadmap 2008 - http://lua-users.org/lists/lua-l/2008- 144–153. ACMPress, 2006. 02/msg00051.html. [12] C. Garrett, J. Dean, D. Grove, and C. Chambers. Measurement and [2] Mozilla — Firefox web browser and Thunderbird email client - Application of Dynamic Receiver Class Distributions. 1994. http://www.mozilla.com. [13] J. Ha, M. R. Haghighat, S. Cong, and K. S. McKinley. A concurrent [3] SPECJVM98-http://www.spec.org/jvm98/. trace-based just-in-time compiler for javascript. Dept.of Computer Sciences, The University of Texas at Austin, TR-09-06, 2009. [4] SpiderMonkey (JavaScript-C) Engine - [14] B. McCloskey. Personal communication. http://www.mozilla.org/js/spidermonkey/. [15] I. Piumarta and F. Riccardi. Optimizing direct threaded code by selec- [5] Surfin’ Safari - Blog Archive - Announcing SquirrelFish Extreme - tive inlining. In Proceedings of the ACM SIGPLAN 1998 conference http://webkit.org/blog/214/introducing-squirrelfish-extreme/. on Programming language design and implementation, pages 291– [6] A. Aho, R. Sethi, J. Ullman, and M. Lam. Compilers: Principles, 300. ACMNewYork,NY,USA,1998. techniques, and tools, 2006. [16] A. Rigo. Representation-Based Just-In-time Specialization and the [7] V. Bala, E. Duesterwald, and S. Banerjia. Dynamo: A transparent Psyco Prototype for Python. In PEPM, 2004. dynamic optimization system. In Proceedings of the ACM SIGPLAN [17] M. Salib. Starkiller: A Static Type Inferencer and Compiler for Conference on Programming Language Design and Implementation, Python. In Master’s Thesis, 2004. pages 1–12. ACM Press, 2000. [18] T. Suganuma, T. Yasue, and T. Nakatani. A Region-Based Compila- [8] M. Berndl, B. Vitale, M. Zaleski, and A. Brown. Context Threading: tion Technique for Dynamic Compilers. ACM Transactions on Pro- a Flexible and Efficient Dispatch Technique for Virtual Machine In- grammingLanguagesandSystems(TOPLAS),28(1):134–174,2006. terpreters. In Code Generation and Optimization, 2005. CGO 2005. [19] M. Zaleski, A. D. Brown, and K. Stoodley. YETI: A graduallY International Symposium on, pages 15–26, 2005. Extensible Trace Interpreter. In Proceedings of the International [9] C. Chambers and D. Ungar. Customization: Optimizing Compiler Conference on Virtual Execution Environments, pages 83–93. ACM Technology for SELF, a Dynamically-Typed O bject-Oriented Pro- Press, 2007. gramming Language. In Proceedings of the ACM SIGPLAN 1989 Conference on Programming Language Design and Implementation, pages 146–160. ACM New York, NY, USA, 1989.