Package ghidra.pcode.emu.jit.gen.op

Enum Class IntMultOpGen

java.lang.Object
java.lang.Enum<IntMultOpGen>
ghidra.pcode.emu.jit.gen.op.IntMultOpGen
All Implemented Interfaces:
BinOpGen<JitIntMultOp>, IntOpBinOpGen<JitIntMultOp>, OpGen<JitIntMultOp>, Serializable, Comparable<IntMultOpGen>, Constable
public enum IntMultOpGen extends Enum<IntMultOpGen> implements IntOpBinOpGen<JitIntMultOp>
The generator for a int_mult.

This uses the binary operator generator and simply emits imul or lmul depending on the type.

For multi-precision multiplication, this emits code to invoke JitCompiledPassage.mpIntMultiply(int[], int[], int[])

  • Enum Constant Details

    • GEN

      public static final IntMultOpGen GEN
      The generator singleton

  • Method Details

    • values

      public static IntMultOpGen[] values()
      Returns an array containing the constants of this enum class, in the order they are declared.
      Returns:
      an array containing the constants of this enum class, in the order they are declared

    • valueOf

      public static IntMultOpGen valueOf(String name)
      Returns the enum constant of this class with the specified name. The string must match exactly an identifier used to declare an enum constant in this class. (Extraneous whitespace characters are not permitted.)
      Parameters:
      name - the name of the enum constant to be returned.
      Returns:
      the enum constant with the specified name
      Throws:
      IllegalArgumentException - if this enum class has no constant with the specified name
      NullPointerException - if the argument is null

    • isSigned

      public boolean isSigned()
      Description copied from interface: BinOpGen
      Whether this operator is signed

      In many cases, the operator itself is not affected by the signedness of the operands; however, if size adjustments to the operands are needed, this can determine how those operands are extended.

      Specified by:
      isSigned in interface BinOpGen<JitIntMultOp>
      Returns:
      true for signed, false if not

    • opForInt

      public <N2 extends Emitter.Next, N1 extends Emitter.Ent<N2, Types.TInt>, N0 extends Emitter.Ent<N1, Types.TInt>> Emitter<Emitter.Ent<N2,Types.TInt>> opForInt(Emitter<N0> em, JitType.IntJitType type)
      Description copied from interface: IntOpBinOpGen
      Emit the JVM bytecode to perform the operator with intF operands on the stack.
      Specified by:
      opForInt in interface IntOpBinOpGen<JitIntMultOp>
      Type Parameters:
      N2 - the tail of the incoming stack
      N1 - the tail of the incoming stack including the right operand
      N0 - the incoming stack with the right and left operands on top
      Parameters:
      em - the emitter typed with the incoming stack
      type - the p-code type of the operands
      Returns:
      the emitter typed with the resulting stack, i.e., the tail with the result pushed

    • opForLong

      public <N2 extends Emitter.Next, N1 extends Emitter.Ent<N2, Types.TLong>, N0 extends Emitter.Ent<N1, Types.TLong>> Emitter<Emitter.Ent<N2,Types.TLong>> opForLong(Emitter<N0> em, JitType.LongJitType type)
      Description copied from interface: IntOpBinOpGen
      Emit the JVM bytecode to perform the operator with long operands on the stack.
      Specified by:
      opForLong in interface IntOpBinOpGen<JitIntMultOp>
      Type Parameters:
      N2 - the tail of the incoming stack
      N1 - the tail of the incoming stack including the right operand
      N0 - the incoming stack with the right and left operands on top
      Parameters:
      em - the emitter typed with the incoming stack
      type - the p-code type of the operands
      Returns:
      the emitter typed with the resulting stack, i.e., the tail with the result pushed

    • genRunMpInt

      public <THIS extends JitCompiledPassage> Emitter<Emitter.Bot> genRunMpInt(Emitter<Emitter.Bot> em, Local<Types.TRef<THIS>> localThis, JitCodeGenerator<THIS> gen, JitIntMultOp op, JitType.MpIntJitType type, Scope scope)
      Generate the mp-int multiply code.

      NOTE: I'd really like to know how many legs of the input operands are actually relevant. Very often, the following idiom is used: 

       temp: 16 = zext(r1) * zext(r2);
 r0 = temp(0);

      That ensures all the operand sizes match, which is often (at least conventionally) required by the Sleigh compiler. However, if r1 and r2 are each only 64 bits, and I can keep track of that fact, then I could perform about half as many multiplies and adds. It also be nice if I can look ahead and see that only 64 bits of temp is actually used. The same is true of IntDivOpGen, IntRemOpGen, IntSDivOpGen, and IntSRemOpGen.

      IDEA: It would be quite a change, but perhaps generating a temporary JVM-level DFG would be useful for culling. The difficulty here is knowing whether or not a temp (unique) is used by a later cross-build. Maybe with the right API calls, I could derive that without additional Sleigh compiler support. If used, I should not cull any computations, so that the retired value is the full value.

      Specified by:
      genRunMpInt in interface IntOpBinOpGen<JitIntMultOp>
      Type Parameters:
      THIS - the type of the generated passage
      Parameters:
      em - the code emitter with an empty stack
      localThis - a handle to the owning compiled passage
      gen - the code generator
      op - the p-code op
      type - the (uniform) type of the inputs and output operands
      scope - a scope for op-temporary variables
      Returns:
      the emitter typed with the empty stack