Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
48 Cards in this Set
- Front
- Back
Motorola 68000 |
this CISC microprocessor has 16 programmer-visible registers available and was used in the original Apple Macintosh
|
|
NaN (Not a Number)
|
this is the result of the operation 0.0/0.0 is performed on a system with IEEE-754 floating-point arithmetic
|
|
Delayed branch
|
a technique used in pipelined CPUs where the instruciton immediately following a control transfer instruction is executed as though it came before the control transfer
|
|
One-operand instructions
|
these would be a characteristic of an accumulator-based machine
|
|
Vertical microprogramming
|
technique used to reduce the size of microprogram memory by encoding bit fields to represent mutually exclusive control signals
|
|
MIPS R4000
|
this was one of the first superpipelined microprocessors using one 8-stage pipeline
|
|
shift-add algorithm
|
this is the description of Booth’s algorithm
|
|
Program counter
|
register to keep track of instruction execution
|
|
Pipeline register
|
this is used to separate one stage of the pipeline from the other
|
|
Two-operand instructions
|
instructions where two operands are needed; one source operand can double as a destination operand being overwritten by the result of the operation
|
|
Carry lookahead adder (CLA)
|
this type of addition circuit develops all carries in logic, directly from the inputs, rather than waiting for them to propagate from less significant bit positions
|
|
Op code
|
the portion of the machine code language instruction which specifies the operation to be done by the CPU
|
|
Positive infinity
|
this is the result when the operation +1.0/0.0 is performed on a system with IEEE-754 floating-point arithmetic
|
|
RISC
|
Reduced Instruction Set Computer; The idea was to make the hardware as fast and as simple as possible by eliminating microcode and explicity encouraging pipelining of the hardware. Any task that cannot be quickly and conveniently done in hardware is left for the compiler to implement by combining simpler functions
|
|
DEC Alpha 21064
|
this microprocessor family is both superscalar and superpipelined in its design
|
|
Zero
|
number in floating-point standard which cannot be normalized, and is not negative infinity
|
|
Zero-operand instructions
|
These would be characteristics of a stack based instruction set
|
|
Booth’s algorithm
|
this can be efficiently used to multiply signed integers
|
|
Wallace tree
|
a structure comprised of multiple levels of carry-save adders (CSAs)
|
|
microPC
|
this keeps track of the location of the next microword to be retrieved from microcode storage
|
|
Delayed load
|
a feature of some instruction sets that is used to cover up a potential performance penalty associated with the pipelined information
|
|
Carry Save Adder
|
a set of full adders, one per bit position; it is similar to a ripple-carry adder with the important exception that the carry-out of each adder is not connected to the carry-in of the next more significant adder
|
|
Horizontal microprogramming
|
a technique used in microprogrammed control unit design in which mutually-exclusive control signals are not encoded into bit fields, thus eliminating the need for microinstructions
|
|
VLIW
|
a type of computer architecture that relies heavily on the compiler technology to exploit instruction-level parallelism
|
|
Overlapping Register Windows
|
a technique used in SPARC architecture to speed up procedure calling by avoiding the use of stack frames to pass parameters
|
|
Negative infinity
|
this is the result of the operation -1.0/0.0 is performed on a system with IEEE-754 floating-point arithmetic
|
|
Super-pipelined architecture
|
this type of processor architecture maximizes temporal parallelism by using a very deep pipeline with very fast stages
|
|
Load-Store Architecture
|
a feature of some computer architectures where “operate” instructions don’t have memory operands; their operands are found in CPU registers
|
|
EPIC
|
the architecture technology used in Intel’s IA-64 (Itanium) chips
|
|
superscalar architecture
|
a type of architecture that uses multiple, pipelined instructions execution units with resolution of inter-instruction dependencies done at run-time by the machine’s control unit
|
|
machine code
|
definition is not found in the book
|
|
denormalized number
|
exception in the IEEE-754 standards for numbers that are nonzerio but is too small to be normalized (that is less than 1.0 time 2 to the smallest exponent)
|
|
one
|
number of a single quantity used to confuse people; between +-infinity, NaN, and 0
|
|
non-restoring division
|
not found in book; is NOT restoring dvision, which is the algorithm for dividing binary numbers shown in the book
|
|
CISC
|
Complex Instruction Set Computer; a design philosophy which states that machine language instructions should have high-level functionality
|
|
MAL
|
Minimum Average Latency; this is the smallest mean number of clock cycles between initiation of operations into a pipeline
|
|
minimum latency
|
the smallest constant interval at which new operations can be started
|
|
Hardwired Control unit
|
directly wired sequential and combinational logic for handling controls
|
|
Advantages of Hardwired
|
faster run time, smaller chip space
|
|
Disadvantages of Hardwired
|
more monetary cost, slower compilation, complex design
|
|
Microcoded control unit
|
each individual operation is treated as a task to be implemented using programmable steps
|
|
Advantages of Microcoded
|
cheaper to implement, faster compilation, more dynamic (coded), compatibility-lower cost
|
|
Disadvantages of Microcoded
|
slower runtime, larger chip space
|
|
Advantages of Superscalar
|
factors small transistors, easier to write compilers, easier for diff. hardware imp. to be software imp.
|
|
Disadvantages of Superscalar
|
complicates CPU hardware, may not exploit all of ILP
|
|
Advantages of VLIW
|
each very long instruction process multiple instructions at the same time, large shared register set, keeps hardware simpler and typically faster
|
|
Disadvantages of VLIW
|
makes compilers slower and harder to write, not compatible of x86
|
|
Similarities of Superscalar & VLIW
|
both are multi-pipelined and use spatial paralellism |