Abstract: A computing system having a processor employs an accelerated virtual subsystem architecture which may reside in either the processor or chipset logic circuitry disposed on the motherboard. The accelerated virtualization process employs at least one phantom read register that provides logical status information in response to an I/O read operation or operations--avoiding engagement of the system management mode as fulfillment of the virtualization process. The at least one phantom read register is updated by the virtualization process and supplies the expected response to an application/driver program running on the processor responsive to the execution of an I/O read operation without invocation of an SMI. Preferably, at least one latch is further provided to buffer writes of indexes of index/data write pairs to further avoid engagement of the system management mode as fulfillment for the virtualization process.

Abstract: A highly integrated central processing unit employs a single external physical bus having first and second protocols to support an L2 cache and a general purpose peripheral interface respectively, to avoid bond-out of the CPU bus to the external world and to steal unused bandwidth for L2 cache accesses while maintaining a standard peripheral bus interface.

Abstract: An integrated circuit package has a top die attach area and a bottom heat spreader thermally coupled to the die for conducting heat generated by the die through a thermal interface in the main circuit board to a heat sink or heat pipe mounted underneath the main circuit board. The preferred thermal interface is a thin, thermally conductive slug mounted through an opening formed in the main circuit board. The heat spreader spans the bottom surface of the integrated circuit package substantially parallel to the main circuit board and preferably extends substantially to the inner periphery of the pin arrangement which preferably, although not exclusively, is in a ball grid array. The opening formed in the main circuit board through which the thin, thermally conductive slug is fitted, is preferably substantially flush with the bottom surface of the main circuit board and juxtaposed against the heat spreader.

Abstract: A system and method for virtualizing external pins and their internal functions within a microprocessor employing an operating system independent interrupt and N subhandlers to virtual the equivalent functions of the pins ordinarily performed by extrinsic circuitry internal to the microprocessor.

Abstract: A processing system having a virtual subsystem architecture employs a reentrant system management mode mechanism and device handlers along with remappable hardware resources to simulate physical subsystems, all transparent to application programs executing on the processing system.

Abstract: A processor for outputting processor state information during idle bus cycles to facilitate the diagnosing and debugging of a processor. The processor includes a plurality of external pins for communicating data from the processor; a visibility register for selecting one of a plurality of modes which identifies processor state information to output onto the plurality of external pins; and a bus interface unit for communicating data to the external pins of the processor and for detecting an idle bus cycle. The bus interface unit outputs processor state information according to the identified mode onto the plurality of external pins in response to detecting an idle bus cycle.

Abstract: An up/dn read prioritization protocol is used to select between multiple hits in a set associative cache. Each set has associated with it an up/dn priority bit that controls read prioritization for multiple hits in the set--the up/dn bit designates either (i) up prioritization in which the up direction is used to select the entry with the lowest way number, or (ii) dn prioritization in which the down direction is used to select the entry with the highest way number. For each new entry allocated into the cache, the state of the up/dn priority bit is updated such that, for the next cache access resulting in multiple hits, the read prioritization protocol selects the new entry for output by the cache.

Abstract: A system and method of readily identifying and handling self-modifying variable length instructions in a pipelined processor is disclosed employing index tags associated with each stage of the execution pipeline wherein the index tags identify the cache line numbers in the instruction cache from which the instructions originate.

Abstract: A data compression apparatus and method of displaying graphics in a computer system employs a full frame buffer and compressed frame buffer wherein pixel data is sent to a display device and concurrently compressed and captured in parallel so that subsequent unchanged frames are regenerated directly from the compressed frame buffer.

Abstract: A prefetch unit is used, in an exemplary embodiment, in a superscalar, superpipelined microprocessor compatible with the x86 instruction set architecture. Normally, the prefetch unit performs split prefetching by generating low and high prefetch addresses in a single clock, with the high prefetch address being generated from the low prefetch address by incrementation. In cases where the low prefetch address is supplied to the prefetch unit too late in a clock period to generate the high prefetch address, such as where a branch instruction is not detected by a branch processing unit so that the target instruction address (i.e., the low prefetch address) is supplied by an address calculation stage, the prefetch unit generates a prefetch request consisting of only the low prefetch address.

Abstract: A computer system employs systems and methods that are transparent to the operating system and application programs, for interruption recovery and resynchronization of events including a playback FIFO buffer having an underrun counter that counts the number of audio samples that could not be read from the playback FIFO buffer because the playback FIFO buffer was empty. When the playback FIFO buffer goes empty, an interrupt is asserted to signal the processor to read the underrun counter to determine how many samples it missed and to advance its pointers forward to "re-sync" the data stream. The computer system further preferably includes a capture FIFO buffer to capture samples from an ADC and having an overrun counter that counts the number of audio samples that could not be written to the capture FIFO buffer because the capture FIFO buffer was full.

Abstract: A processing system includes a clock synchromesh that receives indicia of critical activity from various functional units within the processing system and responsive to the indicia, ratchets down/up the frequency of a clock output signal to at least one of the functional units to reduce power consumption. The determination of critical activity is preferably made according to a heuristic internal to a processor under software or hardware control.

Abstract: A processor includes a distributed clock generator employing a plurality of independently adjustable clocks reconstituted locally from multiple signals. A centralized generator is disposed substantially in the middle of the processing system with satellite reconstitutors being disposed around the periphery to service various functional units which collectively manifest the processing system. The distribution of the multiple signals to the satellite reconstitutors provides substantially equal wire length and local reconstitution mitigates R-C time constant skew problems.

Abstract: In a pipelined processor having at least one execution pipeline for executing instructions, the execution pipeline including ID (decode), AC (address calculation), and EX (execution) processing stages, the processor capable of addressing segments of system memory coupled thereto, a circuit for, and method of, setting a segment access indicator associated with a segment of the system memory being accessed by the processor. The circuit includes: (a) exception generating circuitry to generate an exception when the segment access indicator requires setting and (b) exception handling circuitry, invoked by the processor in response to generation of the exception, to flush the execution pipeline of instructions following a segment load instruction, set the segment access indicator and load an address pointer of the processor with an address corresponding to a specified location within the segment.

Abstract: A processing system includes a graphics subsystem that directly renders raster data to system memory and moves bitmaps between locations within system memory with the graphics subsystem providing the data and a processor providing the virtual-to-physical addresses with privilege and protection check mechanisms.

Abstract: A processor architecture and methodology for executing a condition dependent instruction over a plurality of execution stages in a microprocessor. The microprocessor includes a memory for storing microinstructions. The method involves various steps. In one step, an instruction is received. In another step, a first microinstruction is issued from the memory. This first microinstruction comprises a control and a base address. In yet another step, a secondary address is determined, external from the memory, by evaluating a plurality of predetermined data. In a still another step, the base address and the secondary address are combined to form a destination address in response to the control signal, wherein the destination address identifies a second microinstruction in the memory to execute a successive stage for the received instruction.

Abstract: A processing system removes the burden of maintaining legacy hardware by employing a system management mode mechanism to provide an environment for virtualizing preexisting memory and I/O space instructions into operations for high resolution raster graphics circuitry, thus maintaining functionality and backwards compatibility with preexisting software.

Abstract: In a pipelined processor having at least one execution pipeline for executing instructions, the execution pipeline including ID (decode), AC (address calculation), and EX (execution) processing stages to process instructions for the processor, the processor including a register translation system that controls a renaming of physical registers of the processor to logical registers thereof, a tracking circuit that tracks availability of the physical registers for the renaming, method of operation thereof and processor containing the same.

Abstract: Independent functional units are clocked by a clock source generator having at least two adjustable delay lines for independently adjusting the duty cycles of at least two clocks so that speed path margins are individually optimized for each functional unit.

Abstract: A microarchitecture in a complex instruction computer system is disclosed employing a sparse microROM array and concatenation address circuitry for forming microaddress entry points, avoiding the need for a programmable logic array to translate instruction opcodes and avoiding duplicative entry points, thus minimizing the microROM array size.