Optimize unaligned vanilla vec iteration

[hiphop-php.git] / hphp / doc / ir.specification

*******************************************
* HipHop Intermediate Representation (HHIR)
*******************************************



Introduction
------------

The HipHop Intermediate Representation (IR) is a typed, in-memory,
static-single-assignment, intermediate-level representation of HHBC programs
used for just in time compilation, with these goals:

  1. Complete. The IR represents a program or program fragment entirely,
     without reference to HHBC or other upstream forms of the program.

  2. Type-Safe. Since the IR deals directly with refined types and internal VM
     types, all operations are typesafe. All instruction parameters have a
     parameter type P, and all variables have a type S. Given an instruction
     with source parameter type P and variable type S, S must be equal to or
     more refined than P (S == P or S <: P).

  3. Machine Independent. Since this IR is intended to be used in a JIT
     compiler, it will always be used in a machine specific context.
     Nevertheless, we rely on machine independence in order to separate
     concerns and increase portability of the VM. Passes which manipulate IR
     based on PHP or HHBC semantics should be portable. Passes which deal with
     machine specifics (such as register allocation) should be done in the
     lower level IR (vasm). Types are machine independent.

The unit of compilation is the IRUnit, which is a collection of Blocks
containing IRInstructions that produce and consume SSATmp values. Blocks are
single-entry, single-exit sequences of instructions (i.e. basic
blocks). Instructions may be annotated with Type parameter which modifies the
instruction's behavior, or with additional compile-time constant data (see
extra-data.h). Each SSATmp has a Type which describes the set of values it may
hold, over its entire live range. Instructions may have side effects, which
occur in execution order.

The static single assignment form guarantees the following two invariants for a
well-formed compilation unit:

  1. Each SSATmp is assigned to by exactly one IRInstruction.

  2. Definitions dominate uses. Every path to an IRInstruction using an SSATmp
     first executes the IRInstruction defining the SSATmp.

Any pass that generates or manipulates IR must preserve these invariants,
however it is possible and expected for the invariants to be temporarily broken
during IR generation or during an optimization pass.


Control Flow
------------

IRUnits have one entry block, zero or more exit blocks, and zero or more catch
blocks. Exit blocks leave the compilation unit in the middle of the same PHP
function using one of several instructions that exit a compilation unit
(e.g. ReqBindJmp). Catch blocks are blocks that are reachable from exceptional
control flow edges, and are executed during unwinding if an exception
propagates through the instruction that had it as a `taken' edge.

No SSATmps are defined on entry to the main Block.

Blocks which are join points may start with a DefLabel with destination
SSATmps. In that case, each predecessor must be a Jmp passing a matching number
of sources. In this case the Jmp acts as a tail-call, passing arguments the
same way a plain call would.

Together, the sources of the Jmp instructions and the destinations of the
DefLabel instructions act as traditional SSA Phi pseudo-functions; The type of
the DefLabel's destination is the type-union of the corresponding sources.
Because the Jmp sources are at the ends of blocks, they do not violate the SSA
dominator rule (rule 2, above).


Types
-----

For an overview of the HHIR type system, see the "Type System" section in
hackers-guide/jit-core.md.


SSATmps
-------

An SSATmp represents a virtual register. Since HHIR uses SSA, an SSATmp may
only be assigned to by one instruction. The type of an SSATmp represents the
set of values it may hold at the point it is defined, which is invariant over
the lifetime of the variable (from the definition point to the last use).


IRInstructions
--------------

An instruction is an executable operation with zero or more inputs (sources),
zero or one result (destination), and possible side effects such as accessing
memory, doing I/O, and which may branch or throw an exception. Some
instructions have a Type parameter which modifies its behavior, or other "extra
data" in an arbitrary C++ struct (see extra-data.h).

Each instruction has a signature which describes its effect, parameter types,
and return type, for example:

  IsType<T>, D(Bool), S(Cell), NF

The first column is the instruction name (and optional Type parameter in <>).

The second column describes the result (destination) using one of the D*
macros documented in hphp/runtime/vm/jit/ir-opcode.h, or ND for no destination.

The third column describes the sources, separated by whitespace, using macros
documented in hphp/runtime/vm/jit/ir-opcode.h, or NA if there are no sources.

The fourth column contains the flags, described below. The short name of the
flag (used in this file) is given first, with the long name that it expands to
in hphp/runtime/vm/jit/ir-opcode.cpp in parentheses after it.

NF

  The instruction has no flags.

PRc (ProducesRC)

  The instruction produces a value with an unconsumed reference that must be
  consumed, either by DecRefing it or storing it somewhere in memory.

CRc (ConsumesRC)

  The instruction consumes a reference to one or more of its sources, either by
  decreasing its refcount or storing the reference to memory.

T (Terminal)

  The instruction has no next instruction; it either jumps, returns, or throws.

B (Branch)

  The instruction has a (sometimes optional) taken edge. Instructions that are
  conditional branches (i.e. a Branch that is not Terminal) will also have a
  next edge.

P (Passthrough)

  The value of the instruction's dest is the same as one of its inputs; it
  differs only in the type of the variable, or some other property that doesn't
  affect the value of the variable itself.

LA (Layout-Agnostic)

  The instruction is generic over array-like inputs and outputs. Most ops that
  deal with array-like types can only handle their default ("Vanilla") layouts.
  We whitelist those ops that are generic over layout.

LP (Layout-preserving)

  The instruction is closed under vanilla array layouts: if the first argument
  is vanilla, so too is the destination. The first argument must be an
  array-like type. Layout-preserving implies layout-agnostic.


Instruction set
---------------

1. Checks and Asserts

| CheckType<T>, DRefineS(0), S(Cell), B|P

  Check that the type of the src S0 is T, and if so copy it to D, and
  fallthrough. If S0 cannot be proven to be T, branch to block B. Note that
  this means S0 still /may/ be a subtype of T in block B in some circumstances.

  Specifically, subtypes of Type::Static may not be checked precisely,
  depending on the type of the source.  This means the instruction may take the
  branch in some circumstances even when S0 is a subtype of T, if T has a
  non-empty intersection with Type::Static.

  Also note that many types are not supported as the typeParam right now.

| CheckNullptr, ND, S(Cls|StaticStr|Nullptr), B

  If S0 is not a null pointer, branch to block B. This is used to check the
  return value of a native helper that returns a potentially null StringData*.

| AssertType<T>, DRefineS(0), S(Cell,Mem), P

  Assert that the type of S0 is T, copying it to D.

| CheckTypeMem<T>, ND, S(Mem), B

  If the value pointed to by S0 is not type T, branch to the block B.

| CheckIter<iterId,iterType>, ND, S(FramePtr), B

  Check that specialization type of the given iterator `iterId` on the frame S0
  is `iterType`; if it is not, branch to block B.

| CheckLoc<T,localId>, ND, S(FramePtr), B

  Check that type of the given localId on the frame S0 is T; if not, branch to
  block B.

| CheckStk<T,offset>, ND, S(StkPtr), B

  Check that the type of the cell on the stack pointed to by S0 at offset (in
  cells) is T; if not, branch to block B.

| CheckMBase<T>, ND, S(Lval), B

  Check that the value pointed to by the member base register S0 has type T; if
  not, branch to block B.  This is functionally the same as CheckTypeMem.

| AssertLoc<T,localId>, ND, S(FramePtr), NF

  Asserts that type of the supplied local on the frame S0 is T. This is used
  for local type information, and is similar to CheckLoc except it doesn't
  imply a runtime check (the assertion must've already been proven to be true)
  and cannot cause control flow.

| AssertStk<T,offset>, ND, S(StkPtr), NF

  Assert that stack element at `offset' (in cells) from S0 has type T. This is
  similar to a CheckStk except that it does not imply a runtime check and
  cannot cause control flow.

| AssertMBase<T>, ND, NA, NF

  Assert that the value pointed to by the member base register has type T.
  This is similar to a CheckMBase except that it does not imply a runtime check
  and cannot cause control flow.

| CheckInit, ND, S(Cell), B

  If S0's type is Uninit, branch to block B.

| CheckInitMem, ND, S(Mem), B

  If the value pointed to by S0 has type Uninit, branch to block B.

| CheckCold<TransID>, ND, NA, B

  Check if the counter associated with translation TransID is cold (i.e. within
  a fixed threshold). If it's not (i.e. such translation has reached the
  "hotness threshold"), then branch to block B.

| EndGuards, ND, NA, NF

  A no-op at runtime, this instruction serves to mark the end of the initial
  sequence of guards in a trace.

| CheckNonNull, DSubtract(0, Nullptr), S(Cls|Func|Obj|Str|Mem|TCA|Nullptr), B

  If the value in S0 is Nullptr, branch to block B. If S0 cannot be Nullptr, or
  always is Nullptr, this check may be optimized away.

| AssertNonNull, DSubtract(0, Nullptr), S(StaticStr|Nullptr), P

  Returns S0, with Nullptr removed from its type. This instruction currently
  supports a very limited range of types but can be expanded if needed.

| CheckSmashableClass, ND, S(Smashable) S(Cls), B

  If the lower 32 bits of S0 does not match class pointer S1, branch to block B.

2. Arithmetic

| AddInt, D(Int), S(Int) S(Int), NF

| AddOffset, D(VoidPtr), S(VoidPtr) C(Int), NF

| SubInt, D(Int), S(Int) S(Int), NF

| MulInt, D(Int), S(Int) S(Int), NF

| AddIntO, D(Int), S(Int) S(Int), B

| SubIntO, D(Int), S(Int) S(Int), B

| MulIntO, D(Int), S(Int) S(Int), B

| AddDbl, D(Dbl), S(Dbl) S(Dbl), NF

| SubDbl, D(Dbl), S(Dbl) S(Dbl), NF

| MulDbl, D(Dbl), S(Dbl) S(Dbl), NF

| DivDbl, D(Dbl), S(Dbl) S(Dbl), NF

| DivInt, D(Int), S(Int) S(Int), NF

| Floor, D(Dbl), S(Dbl), NF

| Ceil, D(Dbl), S(Dbl), NF

| AbsDbl, D(Dbl), S(Dbl), NF

| Sqrt, D(Dbl), S(Dbl), NF

| AndInt, D(Int), S(Int) S(Int), NF

| OrInt, D(Int), S(Int) S(Int), NF

| XorInt, D(Int), S(Int) S(Int), NF

| Shl, D(Int), S(Int) S(Int), NF

| Shr, D(Int), S(Int) S(Int), NF

| Lshr, D(Int), S(Int) S(Int), NF

  Double arithmetic, integer arithmetic, and integer bitwise operations.
  Performs the operation described by the opcode name on S0 and S1, and puts
  the result in D.

  Undefined behavior occurs if Mod is given a divisor of zero, or if the
  divisor is -1 and the dividend is the minimum representable integer.

  AbsDbl computes the absolute value of a double-precision value.

  DivDbl conforms to IEEE 754. In particular, division by zero returns +/- INF
  or NAN depending on the dividend; and should the result of a division be zero
  the sign will follow the normal sign rules for division.

  DivInt will perform integer division of S1 by S0. S0 should not be zero and
  must divide S1.

  Note that Shr is an arithmetic right shift: The MSB is sign-extended.

  Lshr is logical right shift.

  Floor and Ceil will return an integral value not greater, or not less
  than their input respectively. Their use requires SSE 4.1, availability
  should be checked before they are emitted.

  AddIntO, SubIntO, MulIntO perform integer arithmetic on S0 and S1, but will
  branch to block B on integer overflow.

| XorBool, D(Bool), S(Bool) S(Bool), NF

  Logical XOR of the two sources. (Note that && and || do not have
  corresponding opcodes because they're handled at the bytecode level, to
  implement short-circuiting.)

| Mod, D(Int), S(Int) S(Int), NF

  Compute S0 mod S1. If S1 is -1 or 0 the results are undefined.


3. Type conversions

To vec conversions:

| ConvArrLikeToVec,             D(Vec), S(ArrLike),                  PRc|CRc|LP

| ConvObjToVec,                 D(Vec), S(Obj),                      PRc|CRc


To dict conversions:

| ConvArrLikeToDict,            D(Dict), S(ArrLike),                 PRc|CRc|LP

| ConvObjToDict,                D(Dict), S(Obj),                     PRc|CRc


To keyset conversions:

| ConvArrLikeToKeyset,          D(Keyset), S(ArrLike),               PRc|CRc|LP

| ConvObjToKeyset,              D(Keyset), S(Obj),                   PRc|CRc


To bool conversions:

| ConvDblToBool,               D(Bool), S(Dbl),                           NF

| ConvIntToBool,               D(Bool), S(Int),                           NF

| ConvStrToBool,               D(Bool), S(Str),                           NF

| ConvObjToBool,               D(Bool), S(Obj),                           NF

| ConvTVToBool,                D(Bool), S(Cell),                          NF


To double conversions:

| ConvBoolToDbl,                D(Dbl), S(Bool),                          NF

| ConvIntToDbl,                 D(Dbl), S(Int),                           NF

| ConvObjToDbl,                 D(Dbl), S(Obj),                           NF

| ConvStrToDbl,                 D(Dbl), S(Str),                           NF

| ConvResToDbl,                 D(Dbl), S(Res),                           NF

| ConvTVToDbl,                  D(Dbl), S(Cell),                          NF


To int conversions:

| ConvBoolToInt,                D(Int), S(Bool),                          NF

| ConvDblToInt,                 D(Int), S(Dbl),                           NF

| ConvObjToInt,                 D(Int), S(Obj),                           NF

| ConvStrToInt,                 D(Int), S(Str),                           NF

| ConvResToInt,                 D(Int), S(Res),                           NF

| ConvTVToInt,                  D(Int), S(Cell),                          NF


To string conversions:

| ConvDblToStr,                 D(Str), S(Dbl),                           PRc

| ConvIntToStr,                 D(Str), S(Int),                           PRc

| ConvObjToStr,                 D(Str), S(Obj),                           PRc

| ConvTVToStr,                  D(Str), S(Cell),                          PRc


  All the above opcodes convert S0 from its current type to the destination
  type, according to the PHP semantics of such a conversion.

| DblAsBits,                    D(Int), S(Dbl),                           NF

  Reinterpret a double as an integer with the same bit pattern.

| OrdStr,                       D(Int), S(Str),                           NF

  Convert the first byte in a string to an unsigned integer.
  Intended as an optimization for ord($str)

| OrdStrIdx,                    D(Int), S(Str) S(Int),                    NF

  Convert the character at position S1 in base string S0 to an unsigned
  integer.  Raises a notice if the position is out of bounds.
  Intended as an optimization for ord($str[$idx]).

| ChrInt,                      D(StaticStr), S(Int),                      NF

  Convert the integer S0 to a the one character string with ascii code
  S0 & 255.

| StrictlyIntegerConv,         D(Str|Int), S(Str),                        PRc

  If S0 is a string representing an integer value (same criteria as array key
  conversion), return that value as an integer. Otherwise return S0.

| ConvPtrToLval,               DLvalOfPtr, S(Ptr),                        NF

  Convert S0 to an equivalent lval.

| VoidPtrAsDataType<T>,        DParam(Cell), S(VoidPtr),                  NF

  Take VoidPtr S0 and convert it to heap type TParam.

4. Boolean predicates

| GtInt,                       D(Bool), S(Int) S(Int),                    NF

| GteInt,                      D(Bool), S(Int) S(Int),                    NF

| LtInt,                       D(Bool), S(Int) S(Int),                    NF

| LteInt,                      D(Bool), S(Int) S(Int),                    NF

| EqInt,                       D(Bool), S(Int) S(Int),                    NF

| NeqInt,                      D(Bool), S(Int) S(Int),                    NF

| CmpInt,                      D(Int),  S(Int) S(Int),                    NF

  Perform 64-bit integer comparisons.

| GtDbl,                       D(Bool), S(Dbl) S(Dbl),                    NF

| GteDbl,                      D(Bool), S(Dbl) S(Dbl),                    NF

| LtDbl,                       D(Bool), S(Dbl) S(Dbl),                    NF

| LteDbl,                      D(Bool), S(Dbl) S(Dbl),                    NF

| EqDbl,                       D(Bool), S(Dbl) S(Dbl),                    NF

| NeqDbl,                      D(Bool), S(Dbl) S(Dbl),                    NF

| CmpDbl,                      D(Int),  S(Dbl) S(Dbl),                    NF

  Perform comparisons of doubles. Comparisons that are unordered according to
  IEEE 754 (such as when at least one operand is NaN) result in false.

| GtStr,                       D(Bool), S(Str) S(Str),                    NF

| GteStr,                      D(Bool), S(Str) S(Str),                    NF

| LtStr,                       D(Bool), S(Str) S(Str),                    NF

| LteStr,                      D(Bool), S(Str) S(Str),                    NF

| EqStr,                       D(Bool), S(Str) S(Str),                    NF

| NeqStr,                      D(Bool), S(Str) S(Str),                    NF

| SameStr,                     D(Bool), S(Str) S(Str),                    NF

| NSameStr,                    D(Bool), S(Str) S(Str),                    NF

| CmpStr,                      D(Int),  S(Str) S(Str),                    NF

  Performs comparison of strings

| GtBool,                      D(Bool), S(Bool) S(Bool),                  NF

| GteBool,                     D(Bool), S(Bool) S(Bool),                  NF

| LtBool,                      D(Bool), S(Bool) S(Bool),                  NF

| LteBool,                     D(Bool), S(Bool) S(Bool),                  NF

| EqBool,                      D(Bool), S(Bool) S(Bool),                  NF

| NeqBool,                     D(Bool), S(Bool) S(Bool),                  NF

| CmpBool,                     D(Int),  S(Bool) S(Bool),                  NF

  Performs comparison of booleans.

| GtObj,                       D(Bool), S(Obj) S(Obj),                    NF

| GteObj,                      D(Bool), S(Obj) S(Obj),                    NF

| LtObj,                       D(Bool), S(Obj) S(Obj),                    NF

| LteObj,                      D(Bool), S(Obj) S(Obj),                    NF

| EqObj,                       D(Bool), S(Obj) S(Obj),                    NF

| NeqObj,                      D(Bool), S(Obj) S(Obj),                    NF

| SameObj,                     D(Bool), S(Obj) S(Obj),                    NF

| NSameObj,                    D(Bool), S(Obj) S(Obj),                    NF

| CmpObj,                      D(Int),  S(Obj) S(Obj),                    NF

  Perform comparison of object. All versions except for SameObj and NSameObj may
  re-enter the VM and therefore may throw exceptions. SameObj and NSameObj never
  re-enter or throw.

| GtArrLike,                   D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| GteArrLike,                  D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| LtArrLike,                   D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| LteArrLike,                  D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| EqArrLike,                   D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| NeqArrLike,                  D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| SameArrLike,                 D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| NSameArrLike,                D(Bool), S(ArrLike) S(ArrLike),            NF|LA

| CmpArrLike,                  D(Int),  S(ArrLike) S(ArrLike),            NF|LA

  Perform comparison of array-likes. All versions except for SameArrLike and
  NSameArrLike may re-enter the VM and therefore may throw exceptions.
  SameArrLike and NSameArrLike never re-enter or throw. Relational comparisons
  for dicts and keysets are not supported. As keysets only contain ints and
  strings, comparisons never re-enter or throw.

| GtRes,                       D(Bool), S(Res) S(Res),                    NF

| GteRes,                      D(Bool), S(Res) S(Res),                    NF

| LtRes,                       D(Bool), S(Res) S(Res),                    NF

| LteRes,                      D(Bool), S(Res) S(Res),                    NF

| EqRes,                       D(Bool), S(Res) S(Res),                    NF

| NeqRes,                      D(Bool), S(Res) S(Res),                    NF

| CmpRes,                      D(Int),  S(Res) S(Res),                    NF

  Perform comparison of resources using PHP semantics. Resource comparisons
  never re-enter or throw.

| EqCls,                       D(Bool), S(Cls) S(Cls),                    NF

  Checks if two Class values are equal.

| EqLazyCls,                   D(Bool), S(LazyCls) S(LazyCls),            NF

  Checks if two Lazy class values are equal.

| EqFunc,                      D(Bool), S(Func) S(Func),                  NF

  Checks if two Func values are equal.

| EqStrPtr,                    D(Bool), S(Str) S(Str),                    NF

  Checks if two string values represent the same underlying string. That is,
  that they point at the same underlying storage.

| EqArrayDataPtr,              D(Bool), S(ArrLike) S(ArrLike),            LA

  Checks if the two arguments represent the same underlying ArrayData. That is,
  that they point at the same underlying storage.

| ProfileInstanceCheck, ND, C(StaticStr), NF

  Profile that S0 has been used as the RHS of an instance check.

| InstanceOf, D(Bool), S(Cls) S(Cls|Nullptr), NF

  Sets D based on whether S0 is a descendant of the class, interface, or trait
  in S1. (Note that this is always false for a trait). S1 may be null at
  runtime if the class is not defined.

| InstanceOfIface, D(Bool), S(Cls) CStr, NF

  Fast path for interface checks. Sets D based on whether S0 implements S1, but
  S1 must be a unique interface. This should only be used in repo-authoritative
  mode.

| InstanceOfIfaceVtable<iface,canOptimize>, D(Bool), S(Cls), NF

  Faster path for interface checks. Sets D based on whether S0 implements
  iface, which must be a unique interface with an assigned vtable slot.  In
  some circumstances, this instruction is ensuring the presence of the
  vtableVec; in those cases, canOptimize is false to avoid eliminating the
  guard.

| ExtendsClass<cls,strictLikely>, D(Bool), S(Cls), NF

  A fast-path for instanceof checks. Sets D based on whether S0 is a descendant
  of cls, where cls must be a unique class that is not an interface or a trait.

  If strictLikely is true, optimize for the case where S0 is not equal to S1.

| InstanceOfBitmask,           D(Bool), S(Cls) CStr,                      NF

| NInstanceOfBitmask,          D(Bool), S(Cls) CStr,                      NF

  A fast-path for instanceof checks. Sets D based on whether S0 is a descendant
  of the class named by S1, where S1 must have a bit allocated for it in the
  fast instance check bitvector (see class.h).

| InterfaceSupportsArrLike,    D(Bool), S(Str), NF

| InterfaceSupportsStr,        D(Bool), S(Str), NF

| InterfaceSupportsInt,        D(Bool), S(Str), NF

| InterfaceSupportsDbl,        D(Bool), S(Str), NF

  Returns whether t instanceof S0 returns true when t is of the given type.

| ResolveTypeStruct<class,suppress,offset,size,isOrAsOp>,
|   D(Dict), S(StkPtr) S(Cls|Nullptr), NF

  Applies class/alias resolution on the type structure that is at the stack
  offset given by S0 and offset. If size > 1, combine the type structures on
  the stack into the first one's denoted holes. Returns a copy.
  S1 is the calling class, used to resolve the this typehint.
  If isOrAsOp is set, raises an error if S0 contains traits, function types or
  typevars.
  If there is an error during type structure resolution, this instruction raises
  an error. If suppress is set, this error is demoted to a warning.

| IsTypeStruct<handle>, D(Bool), S(Dict) S(Cell), LA

  Returns whether S1 matches the type structure of a defined type in S0 and S1
  is a subtype of S0. The input type structure (S0) must be resolved.
  Handle is used for caching purposes.

| IsTypeStructCached, D(Bool), S(Dict) S(Cell), B|LA

  Checks if S0 is cached in TSClassCache and if so, returns whehter S1 is a
  subtype of S0. Otherwise, it branches.

| ProfileIsTypeStruct<handle>, ND, S(Dict), LA|ND

  Profile S0 to determine whether S0 is a type structure holding a reference to
  a Class*.

| ThrowAsTypeStructException, ND, S(Dict) S(Cell), LA|T

  Throws an exception indicating why S1 does not match the type structure of a
  defined type in S0 or why S1 is not a subtype of S0. The input type structure
  (S0) must be resolved.

| RaiseErrorOnInvalidIsAsExpressionType, D(Dict), S(Dict), LA

  Raises an error if the type hint for is/as expression contains an invalid
  type such as callables, erased type variables and trait type hints.
  The input type structure (S0) must be resolved.

| HasToString, D(Bool), S(Obj), NF

  Returns whether the object S0 has a toString method.

| IsType<T>, D(Bool), S(Cell), NF

  Sets D to true iff S0 holds a value that is of type T. T must not be a
  specialized type.

| IsNType<T>, D(Bool), S(Cell), NF

  Sets D to true iff S0 holds a value that is not of type T. T must not be a
  specialized type.

| IsTypeMem<T>, D(Bool), S(Mem), NF

  Sets D to true iff the value referenced by S0 is of type T. T must not be a
  specialized type.

  The value in S0 must not be a pointer into the evaluation stack or frame
  locals.

| IsNTypeMem<T>, D(Bool), S(Mem), NF

  Sets D to true iff the value referenced by S0 is not of type T. T must not be
  a specialized type.

| IsWaitHandle, D(Bool), S(Obj), NF

  Sets D to true iff S0 is a subclass of WaitHandle.

| IsCol, D(Bool), S(Obj), NF

  Sets D to true iff S0 is a collection.

| HasReifiedGenerics, D(Bool), S(Func), NF

  Set D to true iff S0 is a reified function.

5. Branches

| JmpZero,                          ND, S(Int,Bool),                     B

| JmpNZero,                         ND, S(Int,Bool),                     B

  Conditionally jump to based on S0.

| JmpSSwitchDest, ND, S(TCA) S(StkPtr) S(FramePtr), T

  Jump to the target of a sswitch statement, leaving the region, where the
  target TCA is S0.

| JmpSwitchDest, ND, S(Int) S(StkPtr) S(FramePtr), T

  Jump to the target of a switch statement, leaving the region, using table
  metadata <JmpSwitchData> and index S0, which must be a valid index in the
  jump table.

| ProfileSwitchDest<handle,nCases>, ND, S(Int), NF

  Profile a switch statement target.

| CheckSurpriseFlags, ND, S(FramePtr,StkPtr), B

  Tests the implementation-specific surprise flags. If they're true, branches
  to block B. This is done by comparing an evaluation stack pointer to the RDS
  stackLimitAndSurprise word. Note that in a resumed, the frame pointer is not
  pointing into the eval stack, so S0 should be a StkPtr in that case.

| HandleRequestSurprise, ND, NA, NF

  Generate exceptions based on surprise flags on a per request basis.
  Make sure CheckSurpriseFlags is true before calling HandleRequestSurprise.

| ReturnHook, ND, S(FramePtr) S(Cell), NF

  Surprise flag hook for function returns.

| SuspendHookAwaitEF, ND, S(FramePtr) S(FramePtr) S(Obj), NF

  Surprise flag hook for suspending eagerly executing async functions. The S0
  frame was already teleported into S1. Decrefs S2 if it throws an exception.

| SuspendHookAwaitEG, ND, S(FramePtr) S(Obj), NF

  Surprise flag hook for suspending eagerly executing async generators. The S0
  frame has an associated AG, which is already linked to the newly constructed
  AGWH in the blocked state. Decrefs S1 if it throws an exception.

| SuspendHookAwaitR, ND, S(FramePtr) S(Obj), NF

  Surprise flag hook for suspending async functions and async generators resumed
  at Await. The S0 frame has an associated AFWH/AGWH still in the running state,
  S1 points to the child WH we are going to block on.

| SuspendHookCreateCont, ND, S(FramePtr) S(FramePtr) S(Obj), NF

  Surprise flag hook for suspending generators and async generators during their
  invocation. The S0 frame was already teleported into S1. Decrefs S2 if it
  throws an exception.

| SuspendHookYield, ND, S(FramePtr), NF

  Surprise flag hook for suspending generators and async generators at Yield.

| Unreachable<AssertReason>, ND, NA, T

  Indicates an unreachable code path. Any instructions that are post-dominated
  by an Unreachable may be treated as unreachable by the optimizer, and the
  behavior of a program that attempts to execute an Unreachable is undefined.

| EndBlock<AssertReason>, ND, NA, T

  Halt execution, without implying anything about the reachability of
  instructions preceding this. Intended for use in internal tests or other code
  not meant to be executed.

| Jmp, ND, SVar(Top), B|T

  Unconditional jump to block B. In the second form, the target block must
  start with a DefLabel with the same number of destinations as Jmp's number of
  sources. Jmp parallel-copies its sources to the DefLabel destinations.

| DefLabel, DMulti, NA, NF

  DefLabel defines variables received from a previous Jmp. A DefLabel with zero
  destinations is a no-op, and the predecessor blocks may not necessarily end
  in Jmp. A DefLabel with one or more destinations may only be reached by a Jmp
  instruction with the same number of sources. Ordinary branch instructions may
  not pass values to a DefLabel.

| Select, DUnion(1,2), S(Bool,Int) S(Top) S(Top), NF

  If S0 is true/non-zero, return S1, otherwise return S2.


6. Loads

| LdStk<T,offset>, DParam(Cell), S(StkPtr), NF

  Loads from S0 at offset (in cells), and puts the value in D as type T.

| LdLoc<T,localId>, DParam(Cell), S(FramePtr), NF

  Loads local slot localId from the frame S0 and puts the value in D as type T.

| LdLocForeign<T>, DParam(Cell), S(FramePtr) S(Int), NF

  Loads local slot S1 from the frame S0 and puts the value in D as type T.
  Note that it does not perform the local optimizations that LdLoc does.
  Users of this opcode need to ensure that the local is not optimized away.

| LdStkAddr<offset>, D(PtrToStk), S(StkPtr), NF

  Loads the address of the stack slot given by the pointer in S0 at the given
  stack offset (measured in cells).

| LdLocAddr<localId>, D(PtrToFrame), S(FramePtr), NF

  Loads the address of the local slot localId from the frame S0 into D.

| LdRDSAddr<T,RDSHandle>, DParam(Ptr), NA, NF

  Load the address of a Cell that lives at the specified RDS handle.

| LdInitRDSAddr<T,RDSHandle>, DParam(Ptr), NA, B

  Load the address of a Cell that lives at the specified RDS handle. Branch if
  the value at that address is Uninit.

| LdPairElem, D(InitCell), S(Obj) S(Int), NF

  Load the element at S1 out of the Pair collection at S0.

| LdMem<T>, DParam(Cell), S(Mem), NF

  Loads from S0 and puts the value in D.

| LdTVFromRDS<T,RDSHandle,includeAux>, DParam(InitCell), NA, NF

  Load the TypedValue from the specified RDS handle. Must load the aux bits if
  `includeAux` is true.

| LdContField<T>, DParam(Cell), S(Obj) C(Int), NF

  Loads a property from the object referenced by S0 at the offset given by S1
  and puts the value in D. S0 must be a Generator.

| LdClsInitElem<idx>, D(Cell), S(PtrToClsInit), NF

  Load the cell at index `idx` from the class init vector at S0 into D0.

| LdColVec, D(Vec), S(Obj), NF

  Load the vec array backing a collection instance in S0, which must be a
  Vector or ImmVector, and that specific object type must be known at compile
  time.

| LdColDict, D(Dict), S(Obj), NF

  Load the dict array backing a collection instance in S0, which must be a
  Map, Set, ImmMap, or ImmSet, and that specific object type must be known at
  compile time.

| LdIterBase<T,iterId>, DParam(ArrLike), S(FramePtr), LA

  Load the base of the iterator with type `T` at `iterId`. `T` must be a valid,
  DataTypeSpecific-or-better type for the iterator's base; for example, it may
  be based on an earlier call to CheckIter.

| LdIterPos<T,iterId>, DParam(Int|PtrToElem), S(FramePtr), NF

| LdIterEnd<T,iterId>, DParam(Int|PtrToElem), S(FramePtr), NF

  Load the specified field of the iterator at `iterId`. These ops should only
  be generated for iterators known to have a specialized type (via CheckIter).
  The type param `T` should be compatible with this type - i.e. `T` should be
  either an int or a pointer based on whether it's an index or pointer iter.

| LdFrameThis, DParam(Obj), S(FramePtr), NF

  Loads into D the value of m_this from S0.

| LdFrameCls, DParam(Cls), S(FramePtr), NF

  Loads into D the value of m_cls from S0.

| LdClsCtor, D(Func), S(Cls) S(Cls|Nullptr), NF

  Loads into D the constructor of class S0. If the constructor cannot be called
  from the context S1, raise an error.

| LdSmashable, D(Smashable), NA, NF

  Loads a smashable value. The initial value is set to (1 << addr) + 1, where
  addr is a pointer pointing to the value in TC. The lowest bit is set for
  convenience of checking whether the value was already smashed.

| LdSmashableFunc, D(Func), S(Smashable), NF

  Loads into D the func pointer stored in the higher 32 bits of S0.

| DefConst<T>, DParam(Top), NA, NF

  Define a constant value of type T. D is presumed to be globally available and
  the DefConst instruction will not actually appear in the IR instruction
  stream.

| Conjure<T>, DParam(Top), NA, NF

  Define a value of type T. This instruction aborts at runtime; it is meant to
  be used in tests or code that is known to be unreachable.

| ConjureUse, ND, S(Cell), NF

  Define a "use" of S0 effectively keeping the value alive. As with Conjure it
  should not appear in reachable code.

| LdCls, D(Cls), S(Str) C(Cls|Nullptr), NF

  Loads the class named S0 in the context of the class S1. Invokes autoload and
  may raise an error if the class is not defined. The explicit context
  parameter allows the compiler to simplify this instruction to a DefConst in
  some cases. If S0 is constant, this instruction may be simplified to a
  LdClsCached.

| LdClsCached, D(Cls), CStr, NF

  Loads the class named S0 via the RDS. Invokes autoload and may raise an error
  if the class is not defined.

| LdClsCachedSafe, D(Cls), CStr, B

  Loads the class whose name is S0 out of the RDS. If the class is not defined,
  branch to B.

| LdClsInitData, D(PtrToClsInit), S(Cls), NF

  Loads the pointer to the property initializer array for class S0.

| LookupClsRDS, D(Cls|Nullptr), S(Str), NF

  Lookup the cached-class RDS handle for a given class name. Dereference that
  handle and return the associated Class, or null if not present.

| LdCns, D(InitCell), CStr, B

  Load the constant named S0, branching to B if isn't present.

| LookupCnsE<T,constName>,  D(InitCell), CStr, PRc

  Load a constant via the RDS. Raises a fatal error if it cannot define the
  constant.  This should only be executed if LdCns on the same constant has
  failed.

| LdClsCns<className,constantName>, DParam(InitCell), NA, B

  Load the constant 'constantName' for the class 'className'. If not
  initialized, branch to B.

| LookupClsCns, D(InitCell), S(Cls) S(Str), NF

  Lookup a class constant. May raise errors if either the class or constant are
  not defined.

| LookupClsCtxCns, D(Int), S(Cls) S(Str), NF

  Lookup a class context constant. May raise errors if the context constant
  is not defined, is abstract or is a type/value constant.

| LdSubClsCns<constantName,slot>, D(Uncounted), S(Cls), NF

  Load the constant 'constantName' for the class S0. The constant is known to
  be in the given slot. If the returned value is Uninit, the constant is not
  initialized and it must be loaded using InitSubClsCns. Counted constants will
  never have their slot be initialized and thus always take the slow path.

| InitSubClsCns<constantName,slot>, D(InitCell), S(Cls), NF

  Slow path for LdSubClsCns. Used when LdSubClsCns indicates the slot is
  uninitialized.

| CheckSubClsCns<constantName,slot>, ND, S(Cls), B

  Check that the constant 'constantName' lives in the given slot for the class
  S0, and branch if not. S0 must have at least slot+1 constants.

| ProfileSubClsCns<constantName,handle>, D(Cell), S(Cls), NF

  Load the constant 'constantName' for the class S0, profiling the observed
  slots. If the returned value is Uninit, the constant does not exist or is
  abstract or is a type-constant.

| LdClsCnsVecLen, D(Int), S(Cls), NF

  Load the size of S0's constant table.

| LdResolvedTypeCns<slot>, D(StaticDict), S(Cls), B

  Loads the resolved type constant in the slot `slot' on S0 or branches to B if
  the type constant is not resolved.

| LdResolvedTypeCnsNoCheck<slot>, D(StaticDict), S(Cls), NF

  Loads the resolved type constant in the slot `slot' on S0. The type constant
  is assumed to be present and resolved.

| LdTypeCns, D(StaticDict), S(Cls) S(Str), NF

  Loads type constant with name S1 from class S0, performing any necessary
  resolution. Raises an error if no such constant could be found, if S0::S1 is
  not a type constant, if resolution fails, or if S0::S1 is abstract.

| LdTypeCnsNoThrow, D(StaticDict), S(Cls) S(Str), NF

  Loads type constant with name S1 from class S0, performing any necessary
  resolution. If no such constant can be found, if S0::S1 is not a type
  constant, if resolution fails, or if S0::S1 is abstract, then a "fake"
  invalid type-structure is returned.

| LdResolvedTypeCnsClsName<slot>, DTypeCnsClsName, S(Cls), NF

  Loads the cached 'classname' field from the resolved type constant in the
  slot `slot' on S0. If there is no cached field, returns nullptr.

| LdTypeCnsClsName, D(StaticStr), S(Cls) S(Str), NF

  Loads and resolved type constant with name S1 from class S0, as if by
  LdTypeCns. Returns the 'classname' field from that type constant. Raises an
  error if the type constant loading fails, or if there is no 'classname' field
  present.

| LdClsMethodFCacheFunc<clsName,methodName>, D(Func), NA, B

  Loads the target cache entry for a forwarding call to clsName::methodName.
  If the method does not exist, or the cache hasn't been filled yet, branch to
  B.

| LookupClsMethodFCache<clsName,methodName>,
|    D(Func|Nullptr), C(Cls) S(FramePtr),
|    NF

  Lookup clsName::methodName in the forwarding class method cache. S0 should be
  the Class named by clsName and S1 should be the current vm frame pointer. May
  return Nullptr if lookup fails using a subset of the required lookup paths,
  indicating that a more complete lookup path should be taken. May throw if the
  method does not exist.

| LdClsMethodCacheFunc<clsName,methodName>, D(Func), NA, B

  Loads the target cache entry for the method clsName::methodName. If the
  method does not exist or the cache hasn't been filled yet, branch to B.

| LdClsMethodCacheCls<clsName,methodName>, D(Cls), NA, NF

  Loads the target cache class context entry for a call to clsName::methodName
  from the current context. This instruction must only be used when the value
  is known to not be empty (i.e., LdClsMethodCacheFunc must have succeeded, or
  LookupClsMethodCache returned a non-null value).

| LookupClsMethodCache<clsName,methodName>, D(Func|Nullptr), S(FramePtr), NF

  Lookup a function in the class method targetcache. The class name and method
  name are clsName and methodName, respectively. S0 is the current vm frame
  pointer. Returns Nullptr if the method cannot be found using a subset of the
  required lookup paths, indicating that a more complete lookup path should be
  taken. May throw if the method does not exist.

| LdIfaceMethod<vtableIdx,methodIdx>, D(Func), S(Cls), NF

  Load the Func* at methodIdx from the vtable at vtableIdx in S0.

  If methodIdx represents an instance method, S0 must be a non-abstract class.
  This requirement is naturally satisfied if S0 comes from LdObjClass.

| LdFuncVecLen, D(Int), S(Cls), NF

  Load the funcVecLen field from S0.

| LdClsMethod, D(Func), S(Cls) C(Int), NF

  Load the Func* in slot S1 of the class method table for class S0. (Note that
  this vector is located before the class in memory, so the non-negative slots
  will map to negative offset loads.)

| LookupClsMethod, D(Func|Nullptr), S(Cls) S(Str) S(Obj|Nullptr) S(Cls|Nullptr),
|                  NF

  Lookup a pointer to a class method for a given class S0 and method name S1,
  assuming caller's $this is S2 and caller is defined in a class S3. Throws or
  fatals if the method does not exist, is not accessible, or is not a static
  method. Returns nullptr if it is an instance method defined in S2's class
  hierarchy, indicating that this legacy call should be handled by interpreter.

| LdPropAddr<T,index>, D(LvalToProp), S(Obj), NF

  Load the address of the object property at physical index `index`.

| LdInitPropAddr<T,index>, D(LvalToProp), S(Obj), B

  Load the address of the object property at physical index `index`. Branch if
  the value at that address is Uninit.

| LdGblAddr, D(LvalToGbl|Nullptr), S(Str), NF

  Loads a pointer to a global with name S0, or a null lval if that global is
  not already defined.

| LdGblAddrDef, D(LvalToGbl), S(Str), NF

  Loads a pointer to a global with name S0, defining it to be InitNull if it
  is not already defined.

| ProfileGlobal, ND, S(Str), NF

  Profile an (attempted) access of a global with name S0.

| LdClsPropAddrOrNull<readonly op>, D(PtrToSProp|Nullptr), S(Cls) S(Str)
|                     C(Cls|Nullptr) C(Bool) C(Bool), NF

  Loads a pointer to a static class property. S0 points to the class, S1 is the
  property name, and S2 is the class representing the context of the code
  accessing the property. If class S0 does not have a visible and accessible
  static property named S1, then nullptr is returned. An exception
  will be thrown if the property is marked LateInit and its value is
  Uninit, unless S3 is true. An exception is thrown if S4 is true,
  and the property is constant. An exception is also thrown if there is a
  readonly violation with the readonly op.

| LdClsPropAddrOrRaise<readonly op>, D(PtrToSProp), S(Cls) S(Str)
|                       C(Cls|Nullptr) C(Bool) C(Bool), NF

  Loads a pointer to a static class property. S0 points to the class, S1 is the
  property name, and S2 is the class representing the context of the code
  accessing the property. If class S0 does not have a visible and accessible
  static property named S1, then nullptr is returned. An exception
  will be thrown if the property is marked LateInit and its value is
  Uninit, unless S3 is true. An exception is thrown if S4 is true,
  and the property is constant. An exception is also thrown if there is a
  readonly violation with the readonly op.

| LookupSPropSlot, D(Int), S(Cls) S(Str), NF

  Lookup the slot index of the static property with the name S1 on the class
  S0, returning -1 if not found.

| LdObjMethodD, D(Func), S(Cls) S(Str), NF

  Loads a func pointer pointing to the instance method that would be called
  if a method named S1 is invoked on an instance of S0. Raises a fatal if the
  class does not have an accessible method with the given name.

| LdObjMethodS<methodName>, D(Func), S(Cls) S(Smashable), NF

  Loads a func pointer pointing to the instance method that would be called
  if a `methodName` is invoked on an instance of S0. Caches the mapping in
  the TC cache (using S1) and target cache. Raises a fatal if the class does
  not have an accessible method with the given name.

| LdObjInvoke, D(Func), S(Cls), B

  Try to load a cached non-static __invoke Func from the Class in S0, or branch
  to block B if it is not present.

| LdObjClass, DLdObjCls, S(Obj), NF

  Load the class out of the object in S0 and put it in D.

| LdClsName, D(StaticStr), S(Cls), NF

  Load the name of the Class* in S0.

| LdLazyCls, D(LazyCls), S(Cls), NF

  Load a lazy class corresponding to the Class* in S0.

| LdLazyClsName, D(StaticStr), S(LazyCls), NF

  Load the name of the LazyClass in S0.

| LdFunc, D(Func|Nullptr), S(Str), NF

  Loads the Func whose name is S0, invoking autoloader if it is not defined yet.
  Fatals if the named function is not defined, and the autoloader fails to
  define it. Returns nullptr if S0 contained '::', indicating that this legacy
  call should be handled by interpreter.

| LdFuncCached<funcName>, D(Func), NA, NF

  Loads the Func whose name is funcName from the RDS, invoking autoload if it
  not defined yet. Fatal if function autoloader fails to define it.

| LookupFuncCached<funcName>, D(Func), NA, NF

  Loads the Func whose name is given from %1, invoking autoload if it is not
  defined yet. Fatal if the function autoload fails to define it. This
  instruction does not assume the loaded function will be called immediately,
  so it will raise a resolution failure error instead of a call failure error.

| LdFuncNumParams, D(Int), S(Func), NF

  Returns the value of func->numParams().

| LdFuncInOutBits, D(Int), S(Func), NF

  Loads the Func::m_inoutBits field.

| LdARFunc, D(Func), S(FramePtr), NF

  Loads the result of ar->func() where ar is S0.
  Users of this opcode need to ensure that the writing of m_funcId to S0 is
  not optimized away.

| LdFuncName, D(StaticStr), S(Func), NF

  Loads the full name of S0.

| LdMethCallerName<isCls>, D(StaticStr), S(Func), NF

  Loads the meth_caller cls or func name.

| LdFuncCls, D(Cls|Nullptr), S(Func), NF

  Loads the Class* of S0.

| LdStrLen, D(Int), S(Str), NF

  Load the length of the string in S0.

| FuncHasAttr<attr>, D(Bool), S(Func), NF

  Tests for Func::m_attrs & attr.

| ClassHasAttr<attr>, D(Bool), S(Cls), NF

  Tests for Class::m_attrsCopy & attr.

| LdFuncRequiredCoeffects, D(Int), S(Func), NF

  Returns the value of func->requiredCoeffects().

| LdClsFromClsMeth, D(Cls), S(ClsMeth), NF

  Load the Class* of the ClsMethDataRef in S0.

| LdFuncFromClsMeth, D(Func), S(ClsMeth), NF

  Load the Func* of the ClsMethDataRef in S0.

| LdClsFromRClsMeth, D(Cls), S(RClsMeth), NF

  Load the Class* of the RClsMeth* in S0.

| LdFuncFromRClsMeth, D(Func), S(RClsMeth), NF

  Load the Func* of the RClsMeth* in S0.

| LdGenericsFromRClsMeth, D(Vec), S(RClsMeth), NF

  Load the ArrayData* containing the generics attached to the RClsMeth in S0

| LdFuncFromRFunc, D(Func), S(RFunc), NF

  Load the Func* of the RFuncData in S0

| LdGenericsFromRFunc, D(Vec), S(RFunc), NF

  Load the ArrayData* containing the generics attached to the RFuncData in S0

| LdImplicitContext, D(Obj|InitNull), NA, NF

  Loads implicit context from RDS.

7. Allocation

| AllocInitROM<rom>, D(VoidPtr), NA, NF

  Allocate a block of memory for `rom`, and initialize it's contents with the
  ROM data.  Dest is the base of the newly allocated block.

| AllocObj, DAllocObj, S(Cls), PRc

  Allocates a new object of class S1.

| AllocObjReified, DAllocObj, S(Cls) S(Vec), LA|PRc

  Allocates a new object of class S1 and sets S2 as the reified generics of
  this class. If this class is not reified, this instruction raises an error.

| InitProps<class>, ND, NA, NF

  Calls the property initializer function (86pinit) for class.  May throw.

| InitSProps<class>, ND, NA, NF

  Calls the static property initializer functions (86sinit and/or 86linit)
  for class. May throw.

| CheckRDSInitialized<RDSHandle>, ND, NA, B

  Check if the RDS entry at the specified handle is initialized, and branches
  if not.

| MarkRDSInitialized<RDSHandle>, ND, NA, NF

  Mark the given RDS entry as being initialized.

| MarkRDSAccess<RDSHandle>, ND, NA, NF

  Mark the given RDS entry as being accessed for profiling.

| PropTypeRedefineCheck, ND, C(Cls) C(Int), NF

  Check that the specified property at the slot S1 on S0, which redeclares a
  property in the parent, has a declared type equivalent to the parent
  declaration.

| DebugBacktrace, D(Vec), S(Int), PRc

  Obtain stack trace by calling the debug_backtrace() method.

| InitThrowableFileAndLine, ND, S(Obj), NF

  Initialize Throwable's file name and line number assuming the stack trace
  was already initialized and the current vmfp() is a built-in.

| NewInstanceRaw<class>, DAllocObj, NA, PRc

  Allocates an instance of class.

| InitObjProps<class>, ND, S(Obj), NF

  Initializes properties of object S0.

| ConstructInstance<class>, DAllocObj, NA, PRc

  Call the custom instance constructor of an extension class.

| ConstructClosure<class>, DAllocObj, S(Cls|Obj|Nullptr), CRc|PRc

  Call the custom instance constructor of a Closure.

  Store the context represented by S0 into the newly constructed closure object.
  S0 may be a Nullptr when there is no context (i.e. the closure is being used
  in a non-method).

| LockObj, ND, S(Obj), NF

  Clear the IsBeingConstructed flag on the object.

| NewLoggingArray, DLoggingArrLike, S(Vec,Dict,Keyset), PRc|CRc|LA

  Maybe create a LoggingArray wrapping the vanilla array-like S0. The decision
  may be based on flags or sampling. If we don't create an array, we return S0
  unchanged. The result will match S0 in type. For static array constructors
  (including property initial values), the result will be static.

| ProfileArrLikeProps, ND, S(Obj), NF

  Maybe create LoggingArrays wrapping each of the vanilla array-likes in the
  properties of the newly-initialized object S0. The decision may be based on
  flags or sampling. This op mutates the object's props in place.

| LogArrayReach<TransSrcKey>, ND, S(ArrLike), LA

  If the supplied array is a LoggingArray, log the current tracelet and SrcKey
  for use during specialization.

| LogGuardFailure<T>, ND, S(ArrLike), LA

  Log that we failed to guard input S0 to the given type. Useful for catching
  performance bugs in bespoke layout selection, but may have more general use.

| NewDictArray, D(Dict), C(Int), PRc

  Allocate a new dict with the expected capacity S0.

| NewKeysetArray<offset,keys>, D(Keyset), S(StkPtr), PRc|CRc

  Allocate a new keyset containing N elements off the stack given by S0, at
  `offset'. This instruction moves the elements off the stack without
  manipulating their reference counts.

| AllocVec<size>, D(Vec), NA, PRc

  Allocate a new uninitialized vector array with space for size elements in it.
  The array will be initialized with values using either InitVecElem or
  InitVecElemLoop.

| InitVecElem<index>, ND, S(Vec) S(Cell), CRc

  Store the S1 into the slot at index in array S0. This instruction assumes
  that it doesn't have to incref the value being stored. Used to initialize an
  array allocated with AllocVec.

| InitVecElemLoop<offset,size>, ND, S(Vec) S(StkPtr), CRc

  Move `size' elements from the stack given by S1, at `offset', into the array
  S0.  Assumes that the first element on the stack is the last element in the
  array.  Used to initialize an array allocated with or AllocVec that was too
  big for us to use a series of InitVecElem ops.

| AllocStructDict<keys...>,   D(Dict), NA, PRc

  Allocate a new key/value dict or mixed-layout array, given N string literal
  immediates for keys. This op initializes the header and hash table of the
  new array-like, but does not its elements; use InitDictElem for that.

| InitDictElem<index,key>, ND, S(Dict) S(Cell), NF

  Initialize the element at position `index` in array S0 to have the string
  literal `key` as its key and S1 as its value. This instruction assumes that
  S1 has already been inc-reffed. Used with the result of AllocStructDict.

| NewStructDict<offset,keys...>,   D(Dict), S(StkPtr), PRc|CRc

  Allocate a new key/value dict, given N string literal immediates for keys
  and N stack elements at `offset` on stack S0 for values.  This op assumes it
  can move values from the stack without inc-reffing them.

| AllocBespokeStructDict<layout>, DStructDict, NA, PRc

  Allocate a new, empty dict with the bespoke StructLayout `layout`.

| InitStructPositions<slots>, ND, SStructDict, NF

  Initialize the size and iterator positions of the bespoke StructDict S0 for
  the fields specified by `slots`. This instruction does not set the elements
  of the dict; use InitStructElem for that.

| InitStructElem<slot,key>, ND, SStructDict S(InitCell), CRc

  Initialize the element at slot `slot` corresponding to `key` in StructDict S0
  with S1. This instruction assumes that S1 has already been inc-reffed.
  Used with the result of AllocUninitBespokeStructDict.

| NewBespokeStructDict<offset,layout,slots...>, DStructDict, S(StkPtr), PRc|CRc

  Allocate a new dict with the bespoke StructLayout `layout`.  Then, init the
  N fields with the given `slots` with N stack elements starting at `offset`.
  This op moves stack values with no refcounting.

| NewCol<type>, DCol, NA, PRc

  Create an empty new collection of `type'. `type' cannot be Pair.

| NewPair<offset>, DCol, S(Cell) S(Cell), PRc|CRc

  Allocate a new Pair and fill it with the given cells. Ownership of the cells
  is transferred from $1 and $2 to the pair without manipulating the refcounts.

| NewColFromArray<type>, DCol, S(Vec,Dict), PRc|CRc

  Create a collection of `type` from a Vec or Dict kind. `type` cannot be
  Pair. S0 must be vec kind when `type` is Vector or ImmVector, and must be
  dict kind otherwise. Ownership of S0 is transferred from $1 to the
  collection, without manipulating the refcount.

| Clone, DofS(0), S(Obj), PRc

  Allocate an object by cloning S0.

| NewClsMeth, D(ClsMeth), S(Cls) S(Func), NF

  Allocate a new ClsMethDataRef.

| NewRClsMeth, D(RClsMeth), S(Cls) S(Func) S(Vec), LA|PRc

  Allocate a new reified class method by storing the Class in S0,
  the Func in S1 and the reified generics in S2.

| NewRFunc, D(RFunc), S(Func) S(Vec), LA|PRc

  Allocate a new reified function pointer given the Func pointer in S0
  and reified generics in S1.

| FuncCred, DAllocObj, S(Func), PRc

  Allocate a new FunctionCredential

8. Call & Return

| BeginInlining<func, offset>, D(FramePtr), S(StkPtr), NF

  Marks the start of an inlined function whose stack resides offset cells below
  the SP.

  Defines a new frame pointer for an ActRec at callBCOff for use in an inlined
  region. In resumed contexts the new frame is computed relative to S0 as S1 is
  not a stack location.

| EndInlining, ND, S(FramePtr), NF

  Marks the end of an inlined function. S0 is no longer a valid frame location.

| InlineCall<retSPOff>, ND, S(FramePtr) S(FramePtr), NF

  Sets the current vmfp for an inlined call to S0, and the m_sfp of S0 to the
  previous vmfp, S1.

  This instruction is primarily used to represent an inlined frame in the IR
  when it cannot be eliminated during a side-exit. It performs callee-side
  responsibilities for setting up an activation record (i.e. setting the return
  ip, m_func and m_callOff, storing the frame pointer into D).

  The caller frame pointer is passed as S1. This is used to keep track of the
  call chain of inlined functions for simplification and dead code elimination.

| DefFuncEntryFP<func>,
|   D(FramePtr),
|   S(FramePtr) S(StkPtr) S(Int) S(Cls|Obj|Nullptr) S(Int),
|   NF

  Writes a callee's frame and defines a frame pointer.

  `func': the function being called
  S0: frame pointer of the caller
  S1: stack pointer pointing to the space reserved for the callee's frame
  S2: prologue flags (see PrologueFlags)
  S3: prologue context
  S4: callee func id

| Call<offset,numParams,callOff,destroyLocals>,
|   DCall,
|   S(StkPtr) S(FramePtr) S(Func) S(Cls|Obj|Nullptr) S(Int),
|   CRc|PRc

  Transfer control to a callee S2, based on the pre-live activation record and
  set of args on the stack pointed to by S0 at `offset'. S1 is the current
  caller frame pointer. S3 is the context (nullptr, $this or static::class).
  S4 is the ambient coeffects.

| NativeImpl<func>, ND, S(FramePtr) S(StkPtr), NF

  Execute a call to the native builtin specified by the current function. S0
  and S1 should be the current vmfp and vmsp, respectively.

| CallBuiltin, DBuiltin, S(FramePtr) S(StkPtr) SVar(Mem,Cell,Nullptr), PRc

  Call builtin function with N arguments. S0 and S1 should be the current vmfp
  and vmsp, respectively.

  The source and destination types correspond to C++ parameter and return types
  as follows:

    C++ type            HHIR type         Position
    -----------------   ---------         --------
    bool                Bool              source, destination
    int64_t             Int               source, destination
    double              Dbl               source, destination
    const String&       Ptr               source
    const Array&        Ptr               source
    const Object&       Ptr               source
    const Variant&      Ptr               source
    Variant&            Ptr               source (ref param)
    String              {Str|InitNull}    destination
    Array               {Arr|InitNull}    destination
    Object              {Obj|InitNull}    destination
    Variant             {Cell-UninitNull} destination

| RetCtrl<spOff,suspendingResumed>, ND, S(StkPtr) S(FramePtr) S(Cell), T

  Ensure that S0 + `spOff' (in cells) is stored in rvmsp and that S1's saved
  frame pointer is stored in rvmfp, then return to the saved return address in
  S1.  The return value is S2, which is passed via the rret_*() registers to
  the caller.  The `suspendingResumed' flag indicates when this instruction is
  suspending a resumable rather than performing a normal function return.

| AsyncFuncRetPrefetch, ND, S(FramePtr), NF

  Prefetch the WaitHandle that the current (async, resume-mode) frame's AFWH
  is blocking. If this AFWH is blocking multiple WaitHandles, we will only
  prefetch the first one; in any case, this IR op has no visible semantics.

| AsyncFuncRet<spOff>, ND, S(StkPtr) S(FramePtr) S(Cell), T

  Return from a resumed async function, assuming no surprise. Ensures that
  S0 + `spOff` (in cells) is stored in rvmsp and that S1 is stored in rvmfp,
  packs return value S2 into registers and calls the `asyncFuncRet` unique
  stub. The stub stores the result into the wait handle associated with the
  frame pointer, marks it as finished, unblocks its parents and if possible,
  directly resumes the first parent (fast path), or a pending fast runnable
  ResumableWaitHandle (slower path). Otherwise, it will exit VM and return
  control to the asio scheduler (slow path). The stack must contain exactly one
  cell containing uninitialized garbage, which will be populated by the stub
  either to pass the return value to the resumed function, or to return null
  to the scheduler.

| AsyncFuncRetSlow<spOff>, ND, S(StkPtr) S(FramePtr) S(Cell), T

  Return from a resumed async function, assuming unknown surprise flag state
  after the previous surprise was handled by executing "return" event hook.
  Calls the `asyncFuncRetSlow` stub, which re-checks the surprise flag and
  transfers control to the AsyncFuncRet if it was clear, or performs the slow
  path of AsyncFuncRet if it was not, without resuming another function, as
  we are not able to call a potential "resume await" event hook from the stub.

| AsyncGenRetR<spOff>, ND, S(StkPtr) S(FramePtr), T

  Return from an async generator resumed at Await, assuming unknown surprise
  flag state. Ensures that S0 + `spOff` (in cells) is stored in rvmsp and that
  S1 is stored in rvmfp and calls the `asyncGenRetR` unique stub. The stub
  stores `null` result indicating end of iteration into the async generator
  wait handle associated with the generator associated with the frame pointer,
  marks the wait handle as finished and unblocks its parents. If the surprise
  flag was not set and there is a pending fast runnable ResumableWaitHandle,
  it resumes it directly (slow path). Otherwise, it will exit VM and return
  control to the asio scheduler (turtle path). The stack must contain exactly
  one cell containing uninitialized garbage, which will be populated by the
  stub either to pass the return value to the resumed function, or to return
  null to the scheduler.

| AsyncGenYieldR<spOff>, ND, S(StkPtr) S(FramePtr) S(Cell) S(Cell), T

  Yield from an async generator resumed at Await, assuming unknown surprise
  flag state. Works the same as `AsyncGenRetR`, except the async generator
  wait handle is populated with `tuple($key, $value)` result of the iteration
  step, where `$key` is given by S2 and `$value` by S3.

| AsyncSwitchFast<spOff>, ND, S(StkPtr) S(FramePtr), T

  Switch control to another ResumableWaitHandle. Ensures that S0 + `spOff`
  (in cells) is stored in rvmsp and that S1 is stored in rvmfp and calls the
  `asyncSwitchCtrl` unique stub, which tries to resume a pending fast runnable
  ResumableWaitHandle (fast path) if possible, otherwise it will exit VM and
  return control to the asio scheduler (slow path). As with AsyncRetFast, the
  stack must contain exactly one cell containing uninitialied garbage.

| LdRetVal<T>, DParam(InitCell), S(FramePtr), NF

  Load the return value from the already-returned-from ActRec pointed to by S0
  into the dest.  This is used by NativeImpl.  TODO(#7150575): We want to make
  NativeImpl return a TypedValue in the C++ ABI registers.

| GenericRetDecRefs, ND, S(FramePtr), NF

  Does decrefs of all the current function's locals, where S0 is a pointer to
  the relevant activation record. This instruction may not occur in an inlined
  call.

9. Stores

| StPtrAt, ND, S(VoidPtr) S(VoidPtr), NF

  Store the pointer address S1 to pointer S0.

| StTypeAt, ND, S(VoidPtr) S(Cell), NF

  Store the type of S1 to pointer S0.

| StClsInitElem<idx>, ND, S(PtrToClsInit) S(Cell), NF

  Store S1 into the slot at index `idx` in the class init vector at S0.

| StMem, ND, S(Mem) S(Cell), NF

  Store S1 into the location pointed to by S0.

| StMemMeta, ND, S(Mem) S(Cell), NF

  Logically store S1 into the location pointer to by S0. This is like StMem,
  but emits no code. It exists to update frame state tracking.

| StTVInRDS<RDSHandle,includeAux>, ND, S(InitCell), NF

  Store the TypedValue in S0 into the specified RDS handle. Must store the aux
  bits if `includeAux` is true.

| StImplicitContext, ND, S(Obj|InitNull), NF

  Sets the implicit context to S0.

| StImplicitContextWH, ND, S(Obj), NF

  Stores the implicit context from RDS to the WaitHandle in S0.

| StLoc<localId>, ND, S(FramePtr) S(Cell), NF

  Store S1 to local number localId on the frame pointed to by S0.

| StLocMeta<localId>, ND, S(FramePtr) S(Cell), NF

  Logically store S1 to local number localId on the frame pointed to by
  S0. This is like StLoc, but emits no code. It exists to update frame state
  tracking.

| StLocRange<localIds>, ND, S(FramePtr) S(Cell), NF

  Store S1 to the local variables corresponding to localIds, on the frame
  pointed to by S0.

| StIterBase<iterId>, ND, S(FramePtr) S(ArrLike,Nullptr), LA

  Sets the base of the iterator at `iterId` to the pointer S1. The array must
  match the specialized type of the iterator, or be null (for local iterators).

| StIterType<iterId,iterType>, ND, S(FramePtr), NF

  Sets the type of the iterator at `iterId` to `iterType`. This type must be a
  specialized type. Also sets auxiliary fields (like next helper index).

| StIterPos<iterId>, ND, S(FramePtr) S(Int|PtrToElem), NF

| StIterEnd<iterId>, ND, S(FramePtr) S(Int|PtrToElem), NF

  Store S1 to the given field of the iterator at `iterId`. S1 must be an int if
  we're doing index iteration and a pointer if we're doing pointer iteration.

| StStk<offset>, ND, S(StkPtr) S(Cell), NF

  Store S1 to the stack pointed to by S0, at a given offset (in cells).

| StStkMeta<offset>, ND, S(StkPtr) S(Cell), NF

  Logically store S1 to the stack pointed to by S0, at a given offset (in
  cells). This is like StStk, but emits no code. It exists to update frame
  state tracking.

| StStkRange<offsets>, ND, S(StkPtr) S(Cell), NF

  Store S1 to the stack slots at the given offsets from S0.

| StOutValue<index>, ND, S(FramePtr) S(Cell), NF

  Store S1 in a caller allocated out-value vm stack cell index cells above
  S0 on the stack.

| LdOutAddr<index>, D(PtrToOther), S(FramePtr), NF

  Load the address of the storage for out parameter `index` provided by the
  callee (the address will be a location on the callee stack).

| RecordReifiedGenericsAndGetTSList, D(Vec), S(Vec), CRc|LA

  Takes a varray of reified generics from the stack and adds them to the reified
  generics table and returns the input varray of reified generics, possibly
  static

| StFrameCtx, ND, S(FramePtr) S(Obj,Cls), NF

  Store object or class S1 on frame S0.

| StFrameFunc<func>, ND, S(FramePtr), NF

  Store the func in m_funcId/m_func for S0.

| StFrameMeta<callBCOff,flags>, ND, S(FramePtr), NF

  Store the m_callOffAndFlags field of S0.

10. Trace exits

| StVMFP, ND, S(FramePtr), NF

| StVMSP, ND, S(StkPtr), NF

| StVMPC, ND, C(Int), NF

| StVMReturnAddr, ND, C(Int), NF

| StVMRegState, ND, C(Int), NF

  Sets the VM register state to the specified state. This is used
  to track whether the register state is clean or dirty as
  EagerSyncVMRegs is moved around by store elimination.

| ReqBindJmp<target,invSPOff,irSPOff>, ND, S(StkPtr) S(FramePtr), T

  Emit a jump to a translation starting at `target'. If it does not exist, jump
  to a service request that will translate the code and bind this jump to it.

| ReqRetranslate<irSPOff>, ND, S(StkPtr) S(FramePtr), T

  Emit a jump to a service request that will chain to a retranslation of this
  tracelet.

  This instruction is used in exit traces for a type prediction that occurs at
  the first bytecode offset of a tracelet.

| ReqRetranslateOpt<irSPOff>, ND, S(StkPtr) S(FramePtr), T

  Emit a service request to retranslate the current function with a higher
  optimization gear. This instruction is used in exit traces that trigger
  profile-guided optimizations.

| ReqInterpBBNoTranslate<target,invSPOff,irSPOff>, ND, S(StkPtr) S(FramePtr), T

  Jump to the `interpHelperNoTranslate' stub, which first interprets the basic
  block starting at `target' and then continues interpreting basic blocks until
  an already existing translation is found.


11. Refcounting and copies

| Mov, DofS(0), S(Top), P

  Defines D as S0. May imply register-to-register moves at code generation
  time. Does not imply an incref or any other manipulation of S0.

| IncRef, ND, S(Cell), NF

  If S0 is a refcounted type, increment its refcount.

| DecRef<locId>, ND, S(Cell), CRc

  Decrease the reference count of S0 by one, and call a destructor for types
  that require it if it goes to zero.

  The locId is just a hint to the runtime indicating which local variable is
  being DecRef'd, if any.

| DecRefNZ<locId>, ND, S(Cell), CRc

  Decrease the reference count of S0 by one, do not check if it goes to zero.
  This instruction can be used for more efficient code when it is provable that
  the reference count cannot go to zero.

| ProfileDecRef<locId>, ND, S(Cell), NF

  Update the DecRefProfile for the given input as if it were dec-ref-ed, but do
  not actually dec-ref it. We can use this op for e.g. iterator output locals,
  because we don't specialize iterators in profiling translations.

| ReleaseShallow<locId>, ND, S(ArrLike), CRc|LA

  Release memory associated with array S0 without performing any other
  operations.

| DecReleaseCheck<locId>, ND, S(Cell), B|P

  Decrease the refcount of S0 and fall through if the refcount goes to zero.
  Otherwise, branch to B.

12. Misc

| DefFP<offset>, D(FramePtr), NA, NF

  Creates a temporary D representing the current VM frame pointer.

  If the offset is specified, the frame is located at that offset from the SP,
  otherwise the frame either lives on the heap, or its position on the stack
  is unknown.

| DefFrameRelSP<irSPOff, bcSPOff>, D(StkPtr), S(FramePtr), NF

| DefRegSP<irSPOff, bcSPOff>, D(StkPtr), NA, NF

  Defines a stack positioned relative to the frame or the rvmsp register.

  Creates a temporary D representing the current VM stack pointer:
  - DefFrameRelSP: D points to the same location as the frame pointer S0
  - DefRegSP: D is the memory address given by the rvmsp register

  The logical stack starts at the stack base and its position and size is
  defined by these values:
  - `irSPOff' is the offset from the stack base to D
  - `bcSPOff' is the offset from the stack base to the top of the stack

  DefFrameRelSP is used at the beginning of translations of non-resumed
  functions to represent the state of the stack on entry.

  DegRegSP is used at the beginning of translations of prologues and
  resumed functions to represent the state of the stack on entry.

  In prologues, the stack base represents a stack without any func arguments,
  i.e. it is pointing to the empty space reserved for an ActRec.

  In resumables, the stack base represents an empty VM stack.

| LoadBCSP<irSPOff>, D(StkPtr), S(StkPtr), NF

  Gets the bytecode stack pointer from IR stack pointer, for use in eager
  syncing.

| DefFuncPrologueFlags, D(Int), NA, NF

| DefFuncPrologueCallee, D(Func), NA, NF

| DefFuncPrologueNumArgs, D(Int), NA, NF

| DefFuncPrologueCtx, DParam(Cls|Obj), NA, NF

  Creates a temporary D representing prologue flags, callee function pointer,
  number of arguments, or context passed to the prologue.

  May be used only at the beginning of a prologue or a stub used in a prologue
  context.

| ConvFuncPrologueFlagsToARFlags, D(Int), S(Int), NF

  Convert a function prologue flags to function entry flags.

| Count, D(Int), S(Cell), NF

  Computes the number of elements in S0. The count of an array is the number of
  elements it contains, without recursing into containers in the array.
  Subtypes of Bool|Int|Dbl|Str|Res have a count of 1, subtypes of Null have a
  count of 0. The count of objects that implement the Countable interface is
  computed by returning the value of their count method. Objects that do not
  implement Countable have a count of 1.

| CountVec,        D(Int), S(Vec),    LA

| CountDict,       D(Int), S(Dict),   LA

| CountKeyset,     D(Int), S(Keyset), LA

| CountCollection, D(Int), S(Obj),    NF

  Computes the number of elements in S0 using the same definition as Count,
  but with a restriction on the input type that allows for optimizations.

| Nop, ND, NA, NF

  Does nothing. It's sometimes useful for the simplifier to insert one of these
  in the instruction stream.

| JmpPlaceholder, ND, NA, B

  Does nothing if executed. Semantically, this instruction carries a taken edge
  to a block of speculatively-generated code during initial IR generation, such
  as specialized code for an iterator init or next which we may or may not use.
  If it survives irgen, it should be eliminated in the first DCE pass.

| CheckFuncNeedsCoverage<Func>, ND, NA, B

  Checks if Func needs to have coverage information recorded for the current
  request.

| RecordFuncCall<Func>, ND, NA, NF

  Records a call to Func for a function call based code coverage report.

13. Runtime helpers

| VerifyParamCls<func,param,tc>, ND, S(Obj) S(Cls) S(Cls|Nullptr), NF

  Verify parameter type for classes or traits. If S1 does not extend (if S2 is
  a class) or implement (if S2 is an interface) S2, this instruction will raise
  a recoverable fatal error describing the type mismatch.

| VerifyParamCallable<func,param>, ND, S(Cell), NF

  If S0 is not callable, as defined by the php function is_callable, this
  instruction will raise a recoverable fatal error describing the type
  mismatch.

| VerifyParamFail<func,param,tc>, DVerifyParamFail, S(Cell) S(Cls|Nullptr), NF

  Assumes that parameter specified in extra-data in the current function
  has failed a its type check. Depending on the typehint being verified and
  a number of runtime options, may coerce the parameter to the correct type i
  or raise a recoverable fatal error describing the type mismatch.

| VerifyParamFailHard<func,param,tc>, ND, S(Cell) S(Cls|Nullptr), T

  A terminal version of VerifyParamFail, to be used when the compiler can
  statically prove that this failure will result in a fatal error rather than a
  type coercion.

| VerifyRetCallable<func,param>, ND, S(Cell), NF

  Verify a return type hint.

| VerifyRetCls<func,param,tc>, ND, S(Obj) S(Cls) S(Cls|Nullptr), NF

  Verify a return type hint for a class.

| VerifyRetFail<func,param,tc>, DVerifyParamFail, S(Cell) S(Cls|Nullptr), NF

  Failure to verify a return type hint.

| VerifyRetFailHard<func,param,tc>, ND, S(Cell) S(Cls|Nullptr), T

  Terminal version of VerifyRetFail, to be used when the compiler can prove
  that this failure will result in a fatal error.

| VerifyPropCls, ND, S(Cls) S(Int) S(Cls|Nullptr) S(Obj) C(Bool), NF

  Verify a property type hint with AnnotType::Object against an object
  value. S0 is the class of the object containing the property. S1 is the slot
  of the property on the class. S3 is the object which is being set in the
  property. If S2 is not nullptr, than the type-hint refers to that Class, and
  S3 will be checked to see if its an instance of S2. Otherwise, the type-hint
  refers to a type-alias, and the alias will be resolved and checked against
  S3. S4 is true if this is a static property, false otherwise.

| VerifyPropFail, ND, S(Cls) S(Int) S(Cell) C(Bool), NF

  Failure to verify a property type hint. S0 is the class of the object
  containing the property. S1 is the slot of the property on the class. S2 is
  the value which was being set in the property. S3 is true if this is a static
  property, false otherwise.

| VerifyPropFailHard, ND, S(Cls) S(Int) S(Cell) C(Bool), T

  Terminal version of VerifyPropFail, to be used when the compiler can prove
  that this failure will result in a fatal error.

| VerifyProp, ND, S(Cls) S(Int) S(Cell) C(Bool), NF

  Verify that S2 is compatible with the type hint for the property at slot S1
  on S0. S3 is true if this is a static property, false otherwise.

| VerifyPropAll, ND, S(Cls) S(Int) S(Cell) C(Bool), NF

  Verify that S2 is compatible with the all type hints for the property at slot
  S1 on S0 including upper-bounds. S3 is true if this is a static property,
  false otherwise.

| VerifyPropCoerce, D(InitCell), S(Cls) S(Int) S(Cell) C(Bool), PRc|CRc

  Verify that S2 is compatible with the type hint for the property at slot S1
  on S0. S3 is true if this is a static property, false otherwise. Once support
  for coercing class_meth types is removed this ir instruction can also be
  removed (T61738946).

| VerifyPropCoerceAll, D(InitCell), S(Cls) S(Int) S(Cell) C(Bool), PRc|CRc

  Verify that S2 is compatible with all type hints for the property at slot S1
  on S0, including upper-bounds. S3 is true if this is a static property,
  false otherwise. Once support for coercing class_meth types is removed this
  ir instruction can also be removed (T61738946).

| VerifyReifiedLocalType<func,paramId>, ND, S(Cell) S(Dict) S(Cls|Nullptr), LA

  Raises a catchable type hint error if the reified generics of function
  parameter id does not match the type structure given on S0.

| VerifyReifiedReturnType<func>, ND, S(Cell) S(Dict) S(Cls|Nullptr), LA

  Raises a catchable type hint error if the reified generics of S0 does not
  match the type structure given on S1.

| ThrowUninitLoc<localId>, ND, S(Str), T

  Throws an UndefinedVariableException on an uninitialized local variable.

| ThrowUndefPropException, ND, S(Obj) CStr, T

  Throws an UndefinedPropertyException on an undefined property named S1 on the
  class of S0.

| RaiseTooManyArg<func>, ND, S(Vec), CRc|LA

  Raise a too many argument warning because extra arguments stored in S0 were
  passed to function func.

| RaiseError, ND, S(Str), T

  Raises a fatal error with the text in S0 as its message.

| RaiseWarning, ND, S(Str), NF

  Raises a warning with the text in S0 as its message.

| RaiseNotice, ND, S(Str), NF

  Raises a notice with the text in S0 as its message.

| ThrowHasThisNeedStatic, ND, S(Func), T

  Throws a BadMethodCallException to indicate that func was called on an object
  but is a static method.

| ThrowMissingArg<func,argc>, ND, NA, T

  Throws a RuntimeExceptionObject to indicate that only argc arguments were
  passed to function func.

| ThrowMissingThis, ND, S(Func), T

  Throws a BadMethodCallException to indicate that an instance method was called
  with null $this.

| ThrowCallReifiedFunctionWithoutGenerics, ND, S(Func), T

  Throws a BadMethodCallException to indicate that S0 was called without reified
  generics.

| CheckInOutMismatch<numArgs, inoutArgs>, ND, S(Func), NF

| ThrowInOutMismatch<numArgs, inoutArgs>, ND, S(Func), T

  Throw an exception if the inout-ness of passed arguments given by `inoutArgs'
  does not match the inout-ness of parameters of the callee S0. There must be
  a mismatch if ThrowInOutMismatch is used.

  These don't verify that if there are packed args they are not in a position
  that require inout. You will still need verify that when unpacking.

| CheckReadonlyMismatch<numArgs, readonlyArgs>, ND, S(Func), NF

  Checks whether readonly-ness of the caller matches that of callees

| ThrowReadonlyMismatch<numArgs, readonlyArgs>, ND, S(Func), T

  Throw an exception if the readonly-ness of passed arguments given by `readonlyArgs'
  does not match the readonly-ness of parameters of the callee S0. There must be
  a mismatch if ThrowReadonlyMismatch is used.

  These don't verify that packed args are not in a position that require
  readonly. You will still need verify that when unpacking.

| RaiseForbiddenDynCall, ND, S(Func), NF

  Depending on the setting of `ForbidDynamicCallsToFunc`,
  `ForbidDynamicCallsToClsMeth` and `ForbidDynamicCallsToInstMeth` runtime
  options, either raise a warning or throw an exception indicating that the
  func specified in S0 was called dynamically (and should not be).

| RaiseForbiddenDynConstruct, ND, S(Cls), NF

  Depending on the setting of the `ForbidDynamicConstructs` runtime option, either
  raise a warning or throw an exception indicating that the class specified in
  S0 was constructed dynamically (and should not be).

| RaiseCoeffectsCallViolation<func>, ND, S(Int) S(Int), NF

  Depending on the setting of the `CoeffectEnforcementLevels` runtime option,
  either raise a warning or throw an exception indicating that the caller
  was violating coeffects enforcement when calling the callee
  specified by func. S0 is used to pass the coeffects provided by the caller.
  S1 is used to pass the required coeffects of the callee.

| RaiseCoeffectsFunParamTypeViolation<paramIdx>, ND, S(Cell), NF

  Raises a warning to indicate a violation for the expected type for S0
  at position paramIdx for the coeffect rule FunParam.

| RaiseCoeffectsFunParamCoeffectRulesViolation, ND, S(Func), NF

  Raises a warning to indicate a violation for the input S0 to coeffect rule
  FunParam uses polymorphic coeffect rules.

| RaiseStrToClassNotice, ND, S(Str), NF

  Raise a notice if a string is implicitly converted to a class.

| CheckClsMethFunc, ND, S(Func), NF

  Raises runtime errors if the func in S0 is not a callable static method.

| CheckClsReifiedGenericMismatch<cls>, ND, S(Vec), LA

  Raises a runtime error unless whether each generic in S0 is reified or erased
  matches exactly to the expectations of the cls.

| CheckFunReifiedGenericMismatch, ND, S(Func) S(Vec), LA

  Raises a runtime error unless whether each generic in S0 is reified or erased
  matches exactly to the expectations of the func.

| IsFunReifiedGenericsMatched<func>, D(Bool), S(Int), NF

  Load the generics bitmap from prologue flags given by S0 and check whether the
  bitmap proves that the number of given generics and positions of reified vs
  erased generics matches the expectations of the callee `func' (which must use
  reified generics). If this opcode returned false, further checks implemented
  by CheckFunReifiedGenericMismatch are needed.

| InitClsCns<className,constName>, DParam(InitCell), NA, NF

  Initialize the RDS entry for a constant for a class, invoking autoload if it
  is not defined. The initialized value is returned. This instruction may raise
  an undefined constant error if autoload cannot define the constant.

| PrintStr,  ND, S(Str),  CRc

| PrintInt,  ND, S(Int),  CRc

| PrintBool, ND, S(Bool), CRc

  Print for various types.

| ConcatIntStr, D(Str), S(Int) S(Str), PRc

  Concatenate S0 and S1 after converting S0 to String.

| ConcatStrInt, D(Str), S(Str) S(Int), CRc|PRc

  Concatenate S0 and S1 after converting S1 to String.

| ConcatStrStr, D(Str), S(Str) S(Str), CRc|PRc

  Concatenate S0 and S1.

| ConcatStr3, D(Str), S(Str) S(Str) S(Str), CRc|PRc

  Concatenate S0, S1, and S2.

| ConcatStr4, D(Str), S(Str) S(Str) S(Str) S(Str), CRc|PRc

  Concatenate S0, S1, S2, and S3.

| AddNewElemKeyset, D(Keyset),    S(Keyset) S(InitCell), CRc|PRc

| AddNewElemVec,    DModified(0), S(Vec) S(InitCell),    CRc|PRc

  Add S1 as a new element to the array/keyset/vec S0.  (Note: S1 must actually
  be a subtype of InitCell for array invariants, but we can't assert this yet
  in the IR because many eval stack slots are not entirely typed wrt initness
  right now.)

| AKExistsDict, D(Bool), S(Dict) S(Int,Str), NF

  Has the effects of array_key_exists(S0, S1).

| AKExistsKeyset, D(Bool), S(Keyset) S(Int,Str), NF

  Has the effects of array_key_exists(S0, S1).

| AKExistsObj, D(Bool), S(Obj) S(Int,Str), NF

  Has the effects of array_key_exists(S0, S1) on an object S0. This does
  collection accesses.

| GetMemoKey, DMemoKey, S(Cell), PRc

  Given a cell, produces a string or an int that can be used as a memoize cache
  key. Valid values for the input include all basic types, arrays and
  collections, and objects that implement IMemoizeParam. Any other type will
  cause GetMemoKey to throw. This op can only be used within functions marked
  as memoize wrappers.

| GetMemoKeyScalar, DMemoKey, S(Uncounted,Str), PRc

  Identical to GetMemoKey but only accepts scalar types and cannot produce
  errors.

| DictIdx<sizeHint>, DDictElem, S(Dict) S(Int,Str) S(Cell), NF

  Checks if S0 contains the key S1 and returns the result if found. Otherwise
  S2 is returned. The optimization data `sizeHint` doesn't affect semantics.
  (`sizeHint` describes S0; it's one of {Default, SmallStatic}. Default is a
  hash lookup. For SmallStatic, we'll do a linear scan for static string keys.)

| KeysetIdx, DKeysetElem, S(Keyset) S(Int,Str) S(Cell), NF

  Checks if S0 contains the key S1 and returns the result if found. Otherwise
  S2 is returned.

| MethodExists, D(Bool), S(Cls) S(Str), NF

  Checks if the method named S1 exists on class S0.  S0 must be a normal class
  that is not abstract.

| LdBindAddr<SrcKey,spOff>, D(TCA), NA, NF

  Creates a service request to bind the given target address. Returns a TCA
  pointing either to the service request (before the service request is
  satisfied) or to the native code for the given target address (once the
  service request is satisfied).

| LdSSwitchDest<numCases, cases, defaultOffset, bcSPOff>, D(TCA), S(Str), NF

  Using the cases in the extra data, create a hash table for destination lookup.
  Then, lookup the destination for the switched value in the table and yield the
  default if not present.

| InterpOne<T,spOff,bcOff,numPopped,numPushed>, ND,
|                                               S(StkPtr) S(FramePtr),
|                                               NF

  Call the interpreter implementation function for one opcode. S0 + `spOff' (in
  cells) and S1 are, respectively, the VM stack and frame pointers before this
  instruction. T is only present if the instruction pushes to the stack, in
  which case it is the type of the top stack element after the call. `bcOff' is
  the bytecode offset. `numPopped' is the number of stack cells consumed by the
  instruction, and `numPushed' is the number of stack cells produced by the
  instruction.

| InterpOneCF<T,bcOff,numPopped,numPushed>, ND,
|                                           S(StkPtr) S(FramePtr),
|                                           T

  Similar to InterpOne, but for instructions that may modify vmpc. This is
  implemented as a tail call to a stub, so any exceptions thrown will be thrown
  in the context of the stub, not the InterpOneCF instruction.

| OODeclExists<kind>, D(Bool), S(Str) S(Bool), NF

  Returns a bool indicating whether the class, interface, or trait named by S0
  exists. Invokes autoload if S1 is true.

| SetOpTV<op>, ND, S(Lval) S(Cell), NF

  Performs S0 <op>= S1.

| OutlineSetOp<op>, D(InitCell), S(Cell) S(Cell), NF

  Similar to SetOpTV, but does not write back the result to S0, instead it is
  returned as the dest.  This is useful to handle ops that may require a
  type check before writing the value back.

| GetTime, D(Dbl), NA, NF

  Returns a double of the current time in seconds.

| GetTimeNs, D(Int), C(Int), NF

  Returns the current time of the given clock id specified as clockid_t in
  nanoseconds as integer. This will call kernel's clock_gettime_ns() API. Note
  that this cannot be used for CLOCK_THREAD_CPUTIME_ID, as HHVM provides
  different semantics for that counter.

| LdUnitPerRequestFilepath<handle>, D(StaticStr), NA, NF

  Returns the filepath currently bound to the current unit (stored in the given
  RDS handle). The RDS handle must be an initialized normal handle. Only valid
  when Eval.ReuseUnitsByHash is enabled.

| DirFromFilepath, D(StaticStr), S(StaticStr), NF

  Given a string representing a filepath in S0, return only the directory
  portion of the path.

14. Generators & Closures

| LdClosureCls, DParam(Cls), S(Obj), NF

| LdClosureThis, DParam(Obj), S(Obj), NF

  Load the context from the closure object S0 into D, assuming `func' is
  S0's closure Func.

| StClosureArg<index>, ND, S(Obj) S(Cell), CRc

  Store one of the closure environment arguments (i.e. from the closure's use
  clause) from S1 into the closure object S0.

| CreateGen, DAllocObj, S(FramePtr) C(Int) S(TCA,Nullptr) C(Int), PRc

  Create a Generator object and suspend the ActRec provided by S0 into its
  embedded ActRec, allocating S1 slots for locals/iterators. Set the native
  resume address to S2 and resume offset to S3.

| CreateAGen, DAllocObj, S(FramePtr) C(Int) S(TCA,Nullptr) C(Int), PRc

  Create an AsyncGenerator object and suspend the ActRec provided by S0 into its
  embedded ActRec, allocating S1 slots for locals/iterators. Set the native
  resume address to S2 and resume offset to S3.

| CreateAFWH, DAllocObj,
|             S(FramePtr) C(Int) S(TCA,Nullptr) C(Int) S(Obj),
|             CRc|PRc

  Create an AsyncFunctionWaitHandle object and suspend the ActRec provided by
  S0 into its embedded ActRec, allocating S1 slots for locals/iterators.  Set
  the native resume address to S2, resume offset to S3, and mark it blocked on
  non-finished child S4.

| CreateAGWH, DAllocObj,
|             S(FramePtr) S(TCA,Nullptr) C(Int) S(Obj),
|             CRc|PRc

  Create an AsyncGeneratorWaitHandle object and link it to the AsyncGenerator
  associated with the ActRec provided by S0.  Set the native resume address
  to S1, resume offset to S2, and mark it blocked on non-finished child S3.

| CreateAAWH<local,count>, DAllocObj, S(FramePtr) S(Int), PRc

  Create an AwaitAllWaitHandle and add the count elements from frame contiguous
  frame locals beginning at local and extending count locals. S1 denotes the
  total number of non-completed waithandles. All locals must be subclasses of
  WaitHandle.

| CreateSSWH, DAllocObj, S(Cell), CRc|PRc

  Call c_StaticWaitHandle::CreateSucceeded.

| AFWHPrepareChild, ND, S(FramePtr) S(Obj), NF

  Prepare unfinished WaitableWaitHandle object specified by S1 for getting
  awaited by an AsyncFunctionWaitHandle object specified by its ActRec
  provided by S0.

  Injects S1 into the currently running scheduler instance and performs
  cross-scheduler and intra-scheduler cycle detection. Throws if the
  dependency cannot be established.

| AFWHPushTailFrame, ND, S(Obj) C(Int), B

  If S0 is eligible for the tail frame optimization and has any free tail
  frame ID slots, pushes S1 as a new tail frame ID. Otherwise, branches to B.
  This IR op assumes that S0 is an Awaitable in the blocked state.

  S0 is eligible for the optimization if it is an AsyncFunctionWaitHandle,
  if this site "owns" it (i.e. if it has a refcount of exactly 2 - this site
  and its child's back pointer), and if it has space in its tail-frames list.

| StArResumeAddr<offset>, ND, S(FramePtr) S(TCA), NF

  Store the resume address S1 into the Resumable whose ActRec is given by S0,
  marking the offset to resume at as `offset'.

| ContEnter<spOffset,callBCOffset>,
|   DGenIter,
|   S(StkPtr) S(FramePtr) S(FramePtr) S(TCA) S(Cell),
|   CRc

  Enters a generator body. S0 + `spOffset' (in cells) is a pointer to the
  stack, S1 is the current frame pointer, S2 is the generator frame pointer
  embedded in the Generator object, S3 is the address to jump to, and S4 is
  the value that will be pushed onto the stack to send it to the output of
  the yield statement. The `callBCOffset' will be stored to the m_callOff
  field of the ActRec in the generator.

| ContCheckNext, ND, S(Obj) C(Bool), B

  Check whether the generator S0 can be iterated. If the generator is already
  running or finished, or it was not started yet and the S1 check-started flag
  is set, the branch B is taken.

| ContValid, D(Bool), S(Obj), NF

  Return true if a generator is not done, false otherwise.

| ContArIncKey, ND, S(FramePtr), NF

  Special-case key update for generator, ActRec of which is S0, which
  increments the key of a generator if that generator's key is an Int.
  This will cause undefined behavior if the generator's key is not an Int.

| ContArIncIdx, D(Int), S(FramePtr), NF

  Increment the internal index in the Generator in S0, and return the new index
  value.

| ContArUpdateIdx, ND, S(FramePtr) S(Int), NF

  Updates the internal index of generator with S1 if necessary, i.e. if S1
  is larger than the index. S0 is the pointer to the embedded ActRec.

| LdContActRec, D(FramePtr), S(Obj), NF

  Loads the Generator object's ActRec, given a pointer to the generator
  object in S0.

| LdContResumeAddr, D(TCA|Nullptr), S(Obj), NF

  Load the resume addr from the Generator in S0.

| StContArState<state>, ND, S(FramePtr), NF

  Change the state of the Generator object which has frame pointer S0.

| LdContArValue, DParam(InitCell), S(FramePtr), PRc

  Loads 'value' from the Generator object ActRec of which is S0.

| StContArValue, ND, S(FramePtr) S(InitCell), CRc

  Stores 'value' into the Generator object ActRec of which is S0. S1 is the
  new value.

| LdContArKey, DParam(InitCell), S(FramePtr), PRc

  Loads 'key' from the Generator object ActRec of which is S0.

| StContArKey, ND, S(FramePtr) S(InitCell), CRc

  Stores 'key' into the Generator object ActRec of which is S0. S1 is the
  new value.

| AFWHBlockOn, ND, S(FramePtr) S(Obj), CRc

  Establish dependency between parent AsyncFunctionWaitHandle object, whose
  ActRec is given by S0, and child WaitableWaitHandle object referenced by S1.

| LdWHState, D(Int), S(Obj), NF

  Loads the state of the WaitHandle in S0, which is a value from the wait
  handle states in ext_asio.h. This instruction has undefined behavior if S0 is
  not a WaitHandle.

| LdWHResult, DParam(InitCell), S(Obj), NF

  Loads the result of the WaitHandle in S0. This instruction has undefined
  behavior if S0 is not a WaitHandle, or if S0 is not finished.

| LdWHNotDone, D(Int), S(Obj), NF

  Returns 1 if S0 is not finished, and 0 if S0 is finished.

| CountWHNotDone<local,count>, D(Int), S(FramePtr), B

  Returns the number of unfinished awaitables contained in the contiguous
  locals beginning at local and extending count, skipping all nulls. A branch
  is taken if a non-Awaitable non-null value is encountered.

| LdAFWHActRec, D(FramePtr), S(Obj), NF

  Loads the AsyncFunctionWaitHandle object's ActRec, given a pointer to the
  AsyncFunctionWaitHandle object in S0.


15. Debugging, instrumentation, and profiling

| IncStat, ND, C(Int), NF

  Increment stat counter. S0 is the implementation defined stat counter index.

| IncProfCounter<TransID>, ND, NA, NF

  Increment the profiling counter associated with translation TransID.

| IncCallCounter, ND, S(FramePtr), NF

  Increment the counter associated associated with the last call, namely from
  the function containing the previous translation in the call stack into the
  current function.

| DbgAssertRefCount<AssertReason>, ND, S(Cell), NF

  Assert that S0 has a valid refcount. If S0 has a reference counted type and
  its count is implausible then execute a hardware trap instruction.

| DbgTraceCall<spOffset>, ND, S(FramePtr) S(StkPtr), NF

  When EvalHHIRGenerateAsserts is on, this instruction is inserted at the
  start of each region, to emit some sanity checking code.

| DbgAssertFunc, ND, S(FramePtr), NF

  Assert that the func on the srckey is the current function in Frame S0.
  If the assertion fails, execution is aborted via a hardware exception.

| DbgCheckLocalsDecRefd, ND, S(FramePtr), NF

  In debug builds, if LocalsDecRefd flag is set on S0, causes runtime failure by
  emitting a trap instruction. Otherwise, this instruction does nothing.

| DbgTrashStk<offset>, ND, S(StkPtr), NF

  For debugging purposes.  Store kTVTrashJITStk to the stack slot pointed to
  by S0, at a given offset (in cells).

| DbgTrashFrame<offset>, ND, S(StkPtr), NF

  For debugging purposes.  Store kTVTrashJITFrame to kNumActRecCells stack
  slots starting at the offset (in cells), and going toward higher memory
  addresses.

| DbgTrashMem, ND, S(Mem), NF

  For debugging purposes.  Store kTVTrashJITHeap to a heap slot pointed to by
  S0.

| RBTraceEntry, ND, NA, NF

| RBTraceMsg, ND, NA, NF

  Ring buffer tracing.

| ZeroErrorLevel, D(Int), NA, NF

| RestoreErrorLevel, ND, S(Int), NF

  Helper instructions for fast implementation of the PHP error silencing
  operator (@foo()).


16. Iterators

| IterInit<IterData>,   D(Bool), S(ArrLike,Obj) S(FramePtr), CRc|LA

| IterInitK<IterData>,  D(Bool), S(ArrLike,Obj) S(FramePtr), CRc|LA

| LIterInit<IterData>,  D(Bool), S(ArrLike) S(FramePtr),     LA

| LIterInitK<IterData>, D(Bool), S(ArrLike) S(FramePtr),     LA

  <IterData> consists of three indices, iterId, keyId and valId. iterId is
  the index of the iterator variable, keyId and valId are indices of local
  variables.

  Initializes the iterator variable whose index is given by iterId.
  This instruction creates the appropriate iterator for the array or object that
  S0 references, and rewinds the new iterator to its start. S0 points to the
  stack frame containing the iterator and local variables with the indices
  iterId, keyId and valId.

  If the new iterator is at its end (i.e., has no elements to iterate over),
  this instruction decrements the refcount of S0 and returns false; otheriwse,
  it stores a reference to S0 in the new iterator and returns true. If the
  iterator is not at its end, then this instruction stores the iterator's first
  value (and key) into the local variable with index valId (and keyId,
  respectively).

  The LIter variations only accept arrays and do not take ownership of their
  base. Instead, the base is provided on each operation on the iterator, so
  that we can avoid inc-ref-ing the base (in cases where that's safe).

  This instruction has the ConsumesRC property because it either decrements the
  reference count of S0 or stores a reference to S0 into the new iterator.

| IterNext<IterData>,   D(Bool), S(FramePtr), NF

| IterNextK<IterData>,  D(Bool), S(FramePtr), NF

| LIterNext<IterData>,  D(Bool), S(ArrLike) S(FramePtr), LA

| LIterNextK<IterData>, D(Bool), S(ArrLike) S(FramePtr), LA

  <IterData> consists of three indices, iterId, keyId and valId. iterId is
  the index of the iterator variable, keyId and valId are indices of local
  variables.  S0 points to the stack frame containing the iterator and local
  variables with the indices iterId, keyId and valId.

  Advances the iterator variable whose index is given by iterId.

  If the iterator has reached the end, this instruction frees the iterator
  variable and returns false; otherwise, it returns true. If the iterator has
  not reached its end, then this instruction stores the iterator's next value
  (and key) into the local variable with index valId (and keyId, respectively).

| IterFree<iterId>, ND, S(FramePtr), NF

  Free the iterator variable with id `iterId` in the stack frame of S0.
  For non-local iterators, this instruction will dec-ref the stored base.

| KillActRec, ND, S(FramePtr), NF

| KillLoc<localId>, ND, S(FramePtr), NF

| KillIter<iterId>, ND, S(FramePtr), NF

  Mark a given ActRec, local, or iterator as being dead for the purposes of
  memory effects. It no longer contains a meaningful value.

  These operations are used for both correctness and performance. Inserting a
  kill can help us do store-elimination; if a write is followed by a kill on
  all paths before any reads, then it can be eliminated. However, we also use
  kills to mark when a given location cannot be reused by load-elimination or
  similar optimizations, in which case, the kill ops are required.

  In debug builds, these ops will write poison values to these fields.

| GetDictPtrIter, D(PtrToElem), S(Dict) S(Int), NF

| GetVecPtrIter, D(PtrToElem), S(Vec) S(Int), NF

  Returns a pointer to the elm S1 of a vanilla dict or vec S0. S1 does not need
  to be a valid array position; for example, it may equal the size of the array
  (so that the "elm" returned is the pointer-iteration end for S0).

| AdvanceDictPtrIter<offset>, D(PtrToElem), S(PtrToElem), NF

| AdvanceVecPtrIter<offset>, D(PtrToElem), S(PtrToElem), NF

  Increments the pointer S0 to the array element with the given layout `offset`
  positions forward. `offset` is allowed to be negative.

| LdPtrIterKey<T>, DParam(Int|Str), S(PtrToElem), NF

| LdPtrIterVal<T>, DParam(InitCell), S(PtrToElem), NF

  Loads the key or val from the array element pointed to by S0. S0 must be a
  valid elm; that is, it can't point to the end of the array data. LdPtrIterKey
  can only be used for mixed elms, but LdPtrIterVal supports mixed and packed.

  T is used to type the result. For LdPtrIterKey, it must be a valid type for
  the array's keys - i.e., a subtype of TInt|TStr.

| EqPtrIter, D(Bool), S(PtrToElem) S(PtrToElem), NF

  Compares two pointer iterators for equality.


17. Member instruction support

| LdMIStateTempBaseAddr, D(PtrToMISTemp), NA, NF

  Returns a pointer to the tvTempBase field within the current MInstrState.

| LdMBase, DParam(Lval), NA, NF

  Load the current value of the member base register.

| StMBase, ND, S(Lval), NF

  Store a new value to the member base register. It is illegal for any
  instruction other than StMBase or InterpOne (when interpreting a member
  instruction) to modify the member base register.

| CheckMROProp, ND, NA, B

  If the roProp field of MInstrState is false, branch.

| StMROProp, ND, S(Bool), NF

  Set the roProp field of MInstrState to S0.

| FinishMemberOp, ND, NA, NF

  Mark the end of a member operation. This has no effect at runtime but exists
  to provide information for certain optimizations.

All of the remaining opcodes in this section are simple wrappers around helper
functions (specified in S0) to perform the corresponding vector operation. If
S1 is a ConstCls it represents the context class for the operation.

SetElem, SetProp, and SetNewElem are used to implement part of the SetM hhbc
opcode, which almost always pushes its first stack input or a StaticStr as its
stack result. The combinations of input types that cause SetM to push anything
other than those two values are vanishingly rare in correct PHP programs, so
these three instructions have been optimized for the common cases. SetNewElem
and SetProp have no destination, allowing the compiler to predict that the
SetM's output will be the same as its input (and optimize accordingly). If that
turns out to not be the case at runtime, the instruction will throw an
InvalidSetMException. The exception will hold a Cell containing the value the
SetM should push on the stack instead of its input value. The runtime is
responsible for catching this exception, finishing execution of the SetM
instruction, pushing the value from the exception on the stack, and proceeding
as appropriate (most likely with a side exit to the next bytecode instruction,
since it has pushed an unexpected type onto the stack).

SetElem is similar to SetProp and SetNewElem but can also be used for setting
characters within strings. When given a string base and a valid offset, SetElem
returns a string representation of the newly inserted character. In all other
cases it returns nullptr or throws an InvalidSetMException. It will throw this
exception when it detects invalid input types, or when trying to set a string
offset that would grow the string beyond the maximum supported size.

The input types that will cause the errors described above are listed here:

SetNewElem will fail if the base is not a subtype of {Null|Str|Arr|Obj} and not
           Bool<false>.
SetElem has the same base constraint as SetNewElem. In addition, the key must
        not be a subtype of {Arr|Obj}.
SetProp will fail if the base is not a subtype of {Obj|Null}.

Any instructions that take a pointer to an MInstrState struct use the various
fields of that struct for holding intermediate values.

| BaseG, D(LvalToGbl|LvalToConst), S(Str), NF

  Get a base from global named S0. If it is not a defining BaseG it can also
  return the init_null_variant

| PropX, DPropLval, S(Cell) S(Cell) S(PtrToMISTemp|Nullptr), NF

  Lookup intermediate property in S0, with key S1. An exception is thrown if
  the property is Readonly when it was required to be Mutable.

| PropQ, DPropLval, S(Cell) S(StaticStr) S(PtrToMISTemp|Nullptr), NF

  A nullsafe version of PropX, returns null if the base S0 is null. An exception
  is thrown if the property is Readonly when it was required to be Mutable.

| PropDX, DPropLval, S(Cell) S(Cell) S(PtrToMISTemp|Nullptr), NF

  Like PropX, but used for intermediate element lookups that may modify the
  base. An exception is thrown if the property is Readonly when it was
  required to be Mutable.

| CGetProp, D(InitCell), S(Cell) S(Cell), PRc

  Get property with key S1 from S0.

| CGetPropQ, D(InitCell), S(Cell) S(StaticStr), PRc

  A nullsafe version of CGetProp, returns null if the base S0 is null.

| SetProp, ND, S(Cell) S(Cell) S(Cell), NF

  Set property with key S1 in S0 to S2.

| UnsetProp, ND, S(Cell) S(Cell), NF

  Unset an object property.

| SetOpProp<op>, D(InitCell), S(Cell) S(Cell) S(Cell), PRc

  Set op propery with key S1 in base S0, using S2 as the right hand side.

| IncDecProp<op>, D(InitCell), S(Cell) S(Cell), PRc

  Increment/decrement property with key S1 in base S0.

| IssetProp, D(Bool), S(Cell) S(Cell), NF

  Returns true iff the property with key S1 in base S0 is set.

| ElemX, D(InitCell), S(Cell) S(Cell), NF

  Get intermediate element with key S1 from base S0. The base will not be
  modified and the result will not be inc-ref-ed.

| CheckDictKeys<T>, ND, S(Dict), B

  Check that the given mixed array is free of tombstones and that all of its
  elements' keys match the type T. If any check fails, branch to block B.
  Like CheckMixedArrayOffset, this check is allowed to have false negatives -
  it may fail even if the array has keys of the given type.

| CheckArrayCOW, DCOW, S(ArrLike), B|PRc|CRc|LP

  Check that S0 has a refcount of exactly 1; if not, branch to B.

| CopyArray, DCOW, S(ArrLike), PRc|LP

  Make a copy of S0 (regardless of its refcount). The resultant copy will
  always be counted with a ref-count of 1. The array must be of a kind that
  allows for counted variants.

| ProfileDictAccess, ND, S(Dict) S(Int,Str), NF

  Profile access of the element keyed by S1 in S0, tracking sizes and offsets.

| CheckDictOffset<pos>, ND, S(Dict) S(Int,Str), B

  Check that `pos' is within the usage bounds of S0 (including tombstones), and
  that S1 exactly matches the element key of S0 at `pos'.  If any of the checks
  fail, branch to B.  This check is allowed to have false negatives.

| ProfileKeysetAccess, ND, S(Keyset) S(Int,Str), NF

  Profile access of the element keyed by S1 in S0, tracking sizes and offsets.

| CheckKeysetOffset<pos>, ND, S(Keyset) S(Int,Str), B

  Check that `pos' is within the usage bounds of S0 (including tombstones), and
  that S1 exactly matches the element key of S0 at `pos'.  If any of the checks
  fail, branch to B.  This check is allowed to have false negatives.

| ProfileArrayCOW, ND, S(ArrLike), LA

  Profile whether S0 would require a COW before being mutated.

| CheckMissingKeyInArrLike, ND, S(ArrLike) S(StaticStr), B

  Uses the StrKeyTable to check if S1 is guaranteed to be missing in S0.
  If S1 may be present, branches to block B. If we branch here, the key may or
  may not be present.

| ElemDictD, DElemLval, S(Lval) S(Int,Str), NF

| ElemDictU, DElemLval, S(Lval) S(Int,Str), NF

  Similar to ElemDX and ElemUX, but specialized for when the base S0 is a
  dict and the key S1 is an int/str.

| ElemDictK, DElemLvalPos, S(Dict) S(Int,Str) S(Int) S(Dict), NF

  Returns an lval to the element of dict S0 at the known position S2
  (corresponding to the key S1).  S3 is S0 with a potentially more refined type
  and used to calculate the elem type. It is not used at runtime.

| ElemDX, D(LvalToElemOrConst), S(Lval) S(Cell), NF

  Like ElemX, but used for intermediate element lookups that may modify the
  base.

| ElemUX, D(LvalToElemOrConst), S(Lval) S(Cell), NF

  Like ElemX, but used for intermediate element lookups that may modify the
  base as part of an unset operation.

| DictGet, DDictElem, S(Dict) S(Int,Str), NF

  Get element with key S1 from base S0, throwing if the element is not present.

| DictGetQuiet, DDictElem, S(Dict) S(Int,Str), NF

  Get element with key S1 from base S0, returning null if the element is not
  present.

| DictGetK, DDictElem, S(Dict) S(Int,Str) S(Int), NF

  Like DictGet, but the element for S1 is at a known position S2 in S0.

| KeysetGet, DKeysetElem, S(Keyset) S(Int,Str), NF

  Get element with key S1 from base S0, throwing if the element is not present.

| KeysetGetQuiet, DKeysetElem, S(Keyset) S(Int,Str), NF

  Get element with key S1 from base S0, returning null if the element is not
  present.

| KeysetGetK, DKeysetElem, S(Keyset) S(Int,Str) S(Int), NF

  Like KeysetGet, but the element for S1 is at a known position S2 in S0.

| StringGet, D(StaticStr), S(Str) S(Int), PRc

  Get string representing character at position S1 from base string S0.  Raises
  a notice if the position is out of bounds.

| MapGet, D(InitCell), S(Obj) S(Int,Str), PRc

  Get element with key S1 from base S0.

| CGetElem, D(InitCell), S(Cell) S(Cell), PRc

  Get element with key S1 from S0.

| MemoGetStaticValue<func,T>, DParam(InitCell), NA, B

  Get the memo value associated with the static function "func". If the value
  is not present, branch. The returned value is not inc-reffed. This op can
  only be used inside a memoize wrapper.

| MemoGetStaticCache<func,keys,T>, DParam(InitCell), S(FramePtr), B

  Perform a lookup on the memo cache associated with the static function
  "func". The keys for the lookup are read from the locals on the frame pointed
  to by S0 (which must be ints or strings). If the lookup fails, branch. The
  returned value is not inc-reffed. This op can only be used inside a memoize
  wrapper.

| MemoGetLSBValue<func,T>, DParam(InitCell), S(Cls), B

  Get the memo value associated with the static function "func" and late static
  bound class S0. If the value is not present, branch. The returned value is not
  inc-reffed. This op can only be used inside a memoize wrapper.

| MemoGetLSBCache<func,keys,T>, DParam(InitCell), S(FramePtr) S(Cls), B

  Perform a lookup on the memo cache associated with the static function
  "func" and late static bound class S1. The keys for the lookup are read from
  the locals on the frame pointed to by S0 (which must be ints or strings). If
  the lookup fails, branch. The returned value is not inc-reffed. This op can
  only be used inside a memoize wrapper.

| MemoGetInstanceValue<slot,func,T>, DParam(InitCell), S(Obj), B

  Get the memo value at the specified memo slot on S0. If the value is not
  present, branch. The returned value is not inc-reffed. This op can only be
  used inside a memoize wrapper.

| MemoGetInstanceCache<func,keys,T>, DParam(InitCell), S(FramePtr) S(Obj), B

  Perform a lookup on the memo cache at the specified memo slot on S1. The keys
  for the lookup are read from the locals on the frame pointed to by S0 (which
  must be ints or strings). If the lookup fails, branch. The returned value is
  not inc-reffed. This op can only be used inside a memoize wrapper.

| MemoSetStaticValue<func>, ND, S(InitCell), NF

  Set S0 as the memo value associated with the static function "func". Store
  the value, overwriting any previous value, with appropriate ref-count
  manipulations. This op can only be used inside a memoize wrapper.

| MemoSetStaticCache<func,keys>, ND, S(FramePtr) S(InitCell), NF

  Store S1 in the memo cache associated with the static function "func". The
  keys for the lookup are read from the locals on the frame pointed to be S0
  (which must be ints or strings). Store the value, overwriting any previous
  value, with appropriate ref-count manipulations. This op can only be used
  inside a memoize wrapper.

| MemoSetLSBValue<func>, ND, S(InitCell) S(Cls), NF

  Set S0 as the memo value associated with the static function "func" and
  late static bound class S1. Store the value, overwriting any previous
  value, with appropriate ref-count manipulations. This op can only be used
  inside a memoize wrapper.

| MemoSetLSBCache<func,keys>, ND, S(FramePtr) S(Cls) S(InitCell), NF

  Store S2 in the memo cache associated with the static function "func" and
  late static bound class S1. The keys for the lookup are read from the
  locals on the frame pointed to be S0 (which must be ints or strings).
  Store the value, overwriting any previous value, with appropriate
  ref-count manipulations. This op can only be used inside a memoize wrapper.

| MemoSetInstanceValue<slot,func>, ND, S(Obj) S(InitCell), NF

  Set S2 as the memo value at the specified memo slot on S1. Store the value,
  overwriting any previous value, with appropriate ref-count
  manipulations. This op can only be used inside a memoize wrapper.

| MemoSetInstanceCache<slot,func,keys>, ND, S(FramePtr) S(Obj) S(InitCell), NF

  Store S2 in the memo cache at the specified memo slot on S1. Store the value,
  overwriting any previous value, with appropriate ref-count
  manipulations. This op can only be used inside a memoize wrapper.

| InitObjMemoSlots<class>, ND, S(Obj), NF

  Initialize the memoization instance slots for object S0 of the given class.

| DictSet, DArrLikeSet, S(Dict) S(Int,Str) S(InitCell), PRc|CRc

  Set element with key S1 in S0 to S2. The dest will be a new Dict that should
  replace S0.

| BespokeGet, DBespokeElemUninit, S(Vec,Dict,Keyset) S(Int,Str), LA

  Get element with key S1 in the array S0, which may have an arbitrary layout.
  This op returns TUninit if the key is not in the array.

| BespokeGetThrow, DBespokeElem, S(Vec,Dict,Keyset) S(Int,Str), LA

  Get element with key S1 in the array S0, which may have an arbitrary layout.
  This op throws if the key is not in the array.

| BespokeElem, DElemLval, S(Lval) S(Int,Str) C(Bool), LA

  Get a half-lval to an element of type T with key S1 in the array S0, which
  may have an arbitrary layout. The op will copy or escalate the S0 as needed,
  with ElemInt/ElemStr semantics.

  If the key is missing in the array, we'll use S2 to determine what to do.
  If S2 is true, we'll throw; else, we'll return an immutable lval to null.

| BespokeSet, DArrLikeSet, S(Vec,Dict,Keyset) S(Int,Str) S(InitCell), LA|PRc|CRc

  Set element with key S1 in the array S0, which may have an arbitrary layout.
  This op has SetMove semantics; it consumes a refcount on the input array and
  produces one on the output, and does no refcounting to the value S2.

| BespokeUnset, DArrLikeUnset, S(Vec,Dict,Keyset) S(Int,Str), LA|PRc|CRc

  Unset element with key S1 in the array S0, which may have an arbitrary layout.

| BespokeAppend, DArrLikeAppend, S(Vec,Dict,Keyset) S(InitCell), LA|PRc|CRc

  Set element with key S1 in the array S0, which may have an arbitrary layout.
  This op has AppendMove semantics; it consumes a refcount on the input array
  and produces one on the output, and does no refcounting to the value S2.

| BespokeIterFirstPos, D(Int), S(ArrLike), LA

  Obtain the pos coresponding to the first valid element in the non-empty
  array S0, which may have an arbitrary layout.

| BespokeIterLastPos, D(Int), S(ArrLike), LA

  Obtain the pos coresponding to the last valid element in the non-empty
  array S0, which may have an arbitrary layout.

| BespokeIterEnd, D(Int), S(ArrLike), LA

  Returns the "end" iterator position for the given array. Unlike the "last"
  iterator position, the end is never a valid iterator position.

| BespokeIterGetKey, DBespokePosKey, S(ArrLike) S(Int), LA

  Obtain the key at the valid pos S1 in the array S0.

| BespokeIterGetVal, DBespokePosVal, S(ArrLike) S(Int), LA

  Obtain the value at the valid pos S1 in the array S0.

| BespokeEscalateToVanilla<Layout>, DEscalateToVanilla, SBespokeArr CStr, PRc

  Escalate the bespoke array S0 to a vanilla array for the reason S1.

| LdMonotypeDictTombstones, D(Int), SMonotypeDict, NF

  Returns the number of tombstones in the given MonotypeDict's value array.
  The MonotypeDict's IterEnd is equal to its size plus its tombstone count.

| LdMonotypeDictKey, DBespokePosKey, SMonotypeDict S(Int), NF

| LdMonotypeDictVal, DBespokePosVal, SMonotypeDict S(Int), NF

  Specializations of BespokeIterGetKey and BespokeIterGetVal for when the base
  is a monotype dict. The GetKey and GetVal ops require that S1 is a valid
  iterator position for S0.

| LdMonotypeVecElem, DBespokeElem, SMonotypeVec S(Int), NF

  Loads the element of the monotype vec in S0 at offset S1. This instruction
  assumes that the vec actually contains an element at that offset (i.e. the
  index is within bounds).

| StructDictUnset, DArrLikeUnset, SStructDict C(StaticStr), PRc|CRc

  Unset element with key S1 in the struct-layout dict S0. This op has move
  semantics: it consumes a refcount on S0 and produces one on the output.

| StructDictSlot, D(Int), SStructDict S(Str), B

  Calculate the slot corresponding to key S1 in struct dict S0, branching if
  that key is not present.

| StructDictElemAddr, D(LvalToElem), SStructDict S(Str) S(Int) SStructDict, NF

  Calculate the pointer to the value in struct dict S0 corresponding to slot
  S2. S1 is the corresponding key used to calculate the slot, and S3 is the
  struct dict to use for type calculations (if the struct dict has been COWed,
  this will correspond to the pre-COW version). S1 and S3 are only used for
  type calculations and not used at runtime.

| StructDictAddNextSlot, ND, SStructDict S(Int), NF

  Mark the slot S1 as being used in struct dict S0. The slot must not have been
  previously in use (its corresponding value must have been Uninit).

| StructDictTypeBoundCheck, DStructTypeBound, S(InitCell) SStructDict S(Int), B|P

  Check if S0 passes the type bound associated with slot S2 in struct dict
  S1. If not, branch. The returned value is a passthrough of S0, possibly
  refined with what we know statically of the type-bound. This is like a
  CheckType, but the type isn't known statically.

| MapSet, ND, S(Obj) S(Int,Str) S(InitCell), CRc

  Set element with key S1 in S0 to S2.

| VectorSet, ND, S(Obj) S(Int,Str) S(InitCell), CRc

  Set element with key S1 in S0 to S2.

| SetElem<T>, DSetElem, S(Lval) S(Cell) S(InitCell), NF

  Set element with key S1 in S0 to S2. SetElem returns a Nullptr in the common
  case, where the logical result of the hhbc SetM is its right hand side. In
  the case of string bases, the SetM returns a new string containing the newly
  inserted character. So the return value of this instruction is Nullptr unless
  SetM needed to return a static string. The type param is the known type of
  the base that S0 points to.

  Furthermore, in the case of "invalid offsets", SetElem may throw an
  InvalidSetMException (see discussion above).

| SetRange, ND, S(Lval) S(Int) S(Cell) S(Int) S(Int), NF

| SetRangeRev, ND, S(Lval) S(Int) S(Cell) S(Int) S(Int), NF

  Perform a range set or reverse range set operation, with the same arguments
  and semantics as the RangeSet bytecode instruction.

| UnsetElem, ND, S(Lval) S(Cell), NF

  Unsets the element at key S1 in the base S0.

| SetOpElem<op>, D(InitCell), S(Lval) S(Cell) S(Cell), PRc

  Set op elem with key S1 in base S0, using S2 as the right hand side.

| IncDecElem, D(InitCell), S(Lval) S(Cell), PRc

  Increment/decrement element with key S1 in base S0.

| SetNewElem, ND, S(Lval) S(InitCell), NF

  Append the value in S1 to S0.

| SetNewElemDict,   ND, S(Lval) S(InitCell), NF

| SetNewElemVec,    ND, S(Lval) S(InitCell), NF

| SetNewElemKeyset, ND, S(Lval) S(Int,Str), NF

  Specializations of SetNewElem for pointers to dicts, vecs, and keysets.

| DictIsset, D(Bool), S(Dict) S(Int,Str), NF

  Returns true iff the element at key S1 in the base S0 is set.

| KeysetIsset, D(Bool), S(Keyset) S(Int,Str), NF

  Returns true iff the element at key S1 in the base S0 is set.

| StringIsset, D(Bool), S(Str) S(Int), NF

  Returns true iff the string S0 has a character at position S1.

| VectorIsset, D(Bool), S(Obj) S(Int), NF

  Returns true iff the element at key S1 in the base S0 is set.

| PairIsset, D(Bool), S(Obj) S(Int), NF

  Returns true iff the element at key S1 in the base S0 is set.

| MapIsset, D(Bool), S(Obj) S(Int,Str), NF

  Returns true iff the element at key S1 in the base S0 is set.

| IssetElem, D(Bool), S(Cell) S(Cell), NF

  Returns true iff the element at key S1 in S0 is set.

| CheckRange, D(Bool), S(Int) S(Int), NF

  Returns true iff S0 is in the range [0, S1).

| ThrowArrayIndexException, ND, S(ArrLike) S(Int), T

  Throws an OutOfBoundsException if S0 is an undefined index for an array.

| ThrowArrayKeyException, ND, S(Dict) S(Str), T|LA

  Throws an OutOfBoundsException if S0 is an undefined key for a darray or dict.

| ThrowOutOfBounds, ND, S(ArrLike|Obj) S(Cell), T|LA

  Throws an OutOfBoundsException corresponding to an access of S0 with the key
  S1.

| ThrowInvalidArrayKey, ND, S(ArrLike) S(Cell), T|LA

  Throws an InvalidArgumentException corresponding to an access of S0 with the
  key S1, which has a type invalid for that array.

| ThrowInvalidOperation, ND, S(Str), T

  Throws an InvalidOperationException with a message indicating S0.

| ThrowDivisionByZeroException, ND, NA, T

  Throws a DivisionByZeroException.

| ThrowLateInitPropError, ND, S(Cls) S(Str) S(Bool), T

  Throws an InvalidOperationException indicating an access of a unset LateInit
  property. S0 is the class the property was declared on. S1 is the property
  name. S2 is true if its a static property, false otherwise.

| ThrowParameterWrongType<expectedType, func, argNum>, ND, S(Cell), T

  Throws a RuntimeException if calling a function with an argument that has the
  wrong type

| ThrowMustBeMutableException<cls>, ND, S(Str), T

  Throws an InvalidOperationException indicating a readonly semantics violation
  where the property was required to be mutable. cls is the class the property
  was declared on. S0 is the property name.

| ThrowMustBeReadonlyException<cls>, ND, S(Str), T

  Throws an InvalidOperationException indicating a readonly semantics violation
  where the property was required to be readonly. cls is the class the property
  was declared on. S0 is the property name.

| ThrowLocalMustBeValueTypeException, ND, S(Str), T

  Throws an InvalidOperationException indicating a readonly semantics violation
  where the local was required to be a value type. S0 is the local name.

| ThrowMustBeValueTypeException<cls>, ND, S(Str), T

  Throws an InvalidOperationException indicating a readonly semantics violation
  where the property was required to be a value type. S0 is the property
  name. cls is the class the property was declared on.

| ThrowMustBeEnclosedInReadonly<cls>, ND, S(Str), T

  Throws an InvalidOperationException indicating a readonly semantics violation
  where the property was required to be enclosed in a readonly expression. cls
  is the class the property was declared on. S0 is the property name.

| ThrowCannotModifyReadonlyCollection, ND, NA, T

  Throws an InvalidOperationException indicating a readonly semantics violation
  where a collection is modified.

| ProfileType, ND, S(Cell), NF

  Profile the type of S0.

| ProfileCall<rdsHandle>, ND, S(Func), NF

  Profile the call to a function S0.

| ProfileMethod<rdsHandle>, ND, S(Cls) S(Func), NF

  Profile the method S1 called with class context S0.

| ProfileProp, ND, C(StaticStr) C(StaticStr), NF

  Profile the access to property S(1) with base class S(0).

| CheckVecBounds, ND, S(Vec) S(Int), B|LA

  Checks that the index in S1 is within the bounds of the packed array or
  vector array in S0. Branches to B if the index is out of bounds.

| LdVecElemAddr, DElemLvalPos, S(Vec) S(Int) S(Vec), NF

  Loads the address of the element at index S1 of the vec array in S0. This
  instruction assumes the array actually contains an element at that offset
  (IE, the array has the proper length). S2 is S0 with a potentially more
  refined type and used to calculate the elem type. It is not used at runtime.

| ReserveVecNewElem, D(Int), S(CountedVec), B

  If there is room in the packed or vec array (which is assumed to be mutable),
  increments the array size and returns the index of the new last element
  (which you must initialize); else jumps to the taken branch.

| LdVecElem, DVecElem, S(Vec) S(Int), NF

  Loads the element of the vec array in S0 at offset S1. This instruction
  assumes that the vec actually contains an element at that offset (IE, the vec
  has the proper length).

| LdVectorSize, D(Int), S(Obj), NF

  Returns the size of the given Vector collection in S0.

| ColIsEmpty, D(Bool), S(Obj), NF

| ColIsNEmpty, D(Bool), S(Obj), NF

  Returns whether a collection instance is empty or not.  S0 must be known to
  be an instance of a collection class at compile time.

| VecFirst, DFirstElem, S(Vec), NF

  Returns the first value from the packed or vec array in S0.
  If the array is empty, it will return NULL.

| VecLast, DLastElem, S(Vec), NF

  Returns the last value from the packed or vec array in S0.
  If the array is empty, it will return NULL.

| DictFirst, DFirstElem, S(Dict), NF

  Returns the first value from the mixed or dict array in S0.
  If the array is empty, it will return NULL.

| DictLast, DLastElem, S(Dict), NF

  Returns the last value from the mixed or dict array in S0.
  If the array is empty, it will return NULL.

| DictFirstKey, DFirstKey, S(Dict), NF

  Returns the first key from the mixed or dict array in S0.
  If the array is empty, it will return NULL.

| DictLastKey, DLastKey, S(Dict), NF

  Returns the last key from the mixed or dict array in S0.
  If the array is empty, it will return NULL.

| KeysetFirst, DFirstElem, S(Keyset), NF

  Returns the first value(key) from the keyset in S0.
  If the array is empty, it will return NULL.

| KeysetLast, DLastElem, S(Keyset), NF

  Returns the first value(key) from the keyset in S0.
  If the array is empty, it will return NULL.

| IsLegacyArrLike, D(Bool), S(Vec|Dict), LA

  Returns true iff the given array has the legacy bit set.

| ArrayMarkLegacyShallow,     DModified(0), S(Vec,Dict), CRc|PRc|LA

| ArrayMarkLegacyRecursive,   DModified(0), S(Vec,Dict), CRc|PRc|LA

| ArrayUnmarkLegacyShallow,   DModified(0), S(Vec,Dict), CRc|PRc|LA

| ArrayUnmarkLegacyRecursive, DModified(0), S(Vec,Dict), CRc|PRc|LA

  Set or unset the legacy bit on the array-like in S0, copying it if needed.
  The recursive variants traverse through the array-like's transitive elements
  but stop traversal at any non-array-like.


18. Exception/unwinding support

| BeginCatch, ND, NA, NF

  Marks the beginning of a catch region. Exact behavior is implementation and
  architecture specific.

| EndCatch<spOffset,mode,stublogue>, ND, S(FramePtr) S(StkPtr), T

  Marks the end of a catch region and returns control to the unwinder.  The
  `spOffset' field represents a logical adjustment to S1 (in cells) to yield
  the vm stack pointer, however the stack pointer is not actually adjusted
  before this instruction returns control to the unwinder.  The unwinder
  instead relies on fixup map information to find the appropriate stack
  pointers.  Instead it's part of this instruction to facilitate assertions and
  memory effect analysis.

  If the `stublogue' flag is set, the native stack pointer is updated to reflect
  the state prior to entering the stublogue context.

| UnwindCheckSideExit, ND, S(FramePtr) S(StkPtr), B

  Branches to B if the currently executing catch region should return control
  to the unwinder rather than side exiting.  Used to control behavior in catch
  traces for the InvalidSetMException and TVCoercionException situations.

| LdUnwinderValue<T>, DParam(Cell), NA, PRc

  Loads the value contained by the current unwinder exception.

| EnterTCUnwind<spOff,teardown>, ND, S(Obj), CRc|T

  Enters tc_unwind_resume by doing a side enter, i.e. skipping
  itanium ABI. Stores the exception given by S0 to UnwindRDS's exn
  field as well as true to sideEnter field.

  If teardown is set, it notifies tc_unwind_resume to also teardown the locals.

19. Function prologues

| EnterPrologue, ND, NA, NF

  Enter prologue context, which operates in the same mode as unique stubs.
  Makes sure the native stack is properly aligned.

| CheckStackOverflow, ND, S(StkPtr), NF

  Check if the stack depth has exceeded its limit.  If it has, jump to the
  stack overflow helper stub, which will throw.

| CheckSurpriseFlagsEnter<func,argc>, ND, S(FramePtr), NF

  Test the implementation-specific surprise flags.  If they're nonzero, call
  the function enter helper.

| CheckSurpriseAndStack<func,args>, ND, S(FramePtr), NF

  Test surprise flags and stack overflow at the same time.

| LdARFlags, D(Int), S(FramePtr), NF

  Load the flags stored on the ActRec pointed to by the frame
  pointer S0. Bits not defined as flags may contain arbitrary garbage.

| LdTVAux<ValidBits>, D(Int), S(Cell), NF

  Load the value of m_aux from the TypedValue S0.  ValidBits is a mask
  specifying which bits are allowed to be set.  The runtime may ignore it.

  Note that when we pass TypedValues around in registers, we usually use a byte
  register for the m_type member, and thus ignore m_aux.  LdTVAux is only valid
  when we know that S0's m_type and m_aux were both materialized into the same
  64-bit register.

/* Local Variables: */
/* fill-column: 79 */
/* End: */
vim:textwidth=80