Abstract: A microprocessor includes a plurality of semiconductor dies, a bus coupling the plurality of semiconductor dies, and a plurality of processing cores. A distinct subset of the processing cores is located on each of the semiconductor dies. Each die comprises a control unit configured to selectively control a respective clock signal to each of the subset of cores of the die. For each core of the subset, in response to the core writing a value to the control unit, the control unit is configured to turn off the respective clock signal to the core and to write the value over the bus to the control unit of the other die. Collectively all of the control units are configured to simultaneously turn on the clock signals to all of the processing cores after the clock signals have been turned off to all of the processing cores.

Abstract: An associative cache memory, comprising: an array of storage elements arranged as M sets by N ways; an allocation unit allocates the storage elements in response to memory accesses that miss in the cache memory. Each memory access selects a set. Each memory access has an associated memory access type (MAT) of a plurality of predetermined MATs. Each valid storage element has an associated MAT; a mapping that includes, for each MAT, a MAT priority. In response to a memory access that misses in the array, the allocation unit: determines a most eligible way and a second most eligible way of the selected set for replacement based on a replacement policy; and replaces the second most eligible way rather than the most eligible way when the MAT priority of the most eligible way is greater than the MAT priority of the second most eligible way.

Abstract: A microprocessor includes a plurality of cores, a shared cache memory, and a control unit that individually puts each core to sleep by stopping its clock signal. Each core executes a sleep instruction and responsively makes a respective request of the control unit to put the core to sleep, which the control unit responsively does, and detects when all the cores have made the respective request and responsively wakes up only the last requesting cores. The last core writes back and invalidates the shared cache memory and indicates it has been invalidated and makes a request to the control unit to put the last core back to sleep. The control unit puts the last core back to sleep and continuously keeps the other cores asleep while the last core writes back and invalidates the shared cache memory, indicates the shared cache memory was invalidated, and is put back to sleep.

Abstract: A hardware data compressor that compresses an input block of characters by replacing strings of characters in the input block with back pointers to matching strings earlier in the input block. A hash table is used in searching for the matching strings in the input block. A plurality of hash index generators each employs a different hashing algorithm on an initial portion of the strings of characters to be replaced to generate a respective index. The hardware data compressor also includes an indication of a type of the input block of characters. A selector selects the index generated by of one of the plurality hash index generators to index into the hash table based on the type of the input block.

Abstract: A microprocessor includes a predicting unit and a control unit. The control unit controls the predicting unit to accumulate a history of characteristics of executed instructions and makes predictions related to subsequent instructions based on the history while the microprocessor is running a first thread. The control unit also detects a transition from running the first thread to running a second thread and controls the predicting unit to selectively suspend accumulating the history and making the predictions using the history while running the second thread. The predicting unit makes static predictions while running the second thread. The selectivity may be based on the privilege level, identity or length of the second thread, static prediction effectiveness during a previous execution instance of the thread, whether the transition was made due to a system call, and whether the second thread is an interrupt handler.

Abstract: A hardware data compressor. A first hardware engine scans an input block of characters and uses a plurality of lists of nodes to produce back pointers to matching strings in the input block to compress the input block. Each node points to a character in the input block previously scanned and has an associated probability that a back pointer to a matching string that begins with the pointed-to character will be produced by the first hardware engine. A second hardware engine, for each list of nodes of the plurality of lists, sorts the list according to the probabilities of the nodes in the list so that higher probability nodes appear earlier in the list for use by the first hardware engine to search for matching strings during the scan of the input block of characters.

Abstract: A hardware data compressor that compresses an input block of characters by replacing strings of characters in the input block with back pointers to matching strings earlier in the input block. A hash table is used in searching for the matching strings in the input block. A plurality of hash index generators each employs a different hashing algorithm on an initial portion of the strings of characters to be replaced to generate a respective index. The hardware data compressor also includes an indication of a type of the input block of characters. A selector selects the index generated by of one of the plurality hash index generators to index into the hash table based on the type of the input block.

Abstract: A hardware data compressor. A hardware engine maintains first and second hash tables while it scans an input block of characters to be compressed. The first hash table is indexed by a hash of N characters of the input block. The second hash table is indexed by a hash of M characters of the input block. M is greater than two. N is greater than M. The engine uses the first hash table to search the input block behind a current search target location for a match of at least N characters at the current search target location, and uses the second hash table to search the input block behind the current search target location for a match of at least M characters at the current search target location when no match of at least N characters at the current search target location using the first hash table is found.

Abstract: A processor including an extended page table (EPT) translation mechanism that is enabled for virtualization, and a load EPT instruction. When executed by the processor, the load EPT instruction directly invokes the EPT translation mechanism to directly convert a provided guest physical address into a corresponding true physical address. The EPT translation mechanism may include an EPT paging structure and an EPT tablewalk engine. The EPT paging structure is generated and stored in an external system memory when the EPT translation mechanism is enabled. The EPT tablewalk engine is configured to access the EPT paging structure for the physical address conversion. The EPT tablewalk engine may perform relevant checks to trigger EPT misconfigurations and EPT violations during execution of the load EPT instruction.

Abstract: A hardware data compressor. An LZ77 compression engine scans an input block of characters and produces tokens, the tokens are either literal characters of the input block or a back pointer to replaced strings of characters of the input block. A Huffman encoding engine receives the tokens produced by the LZ77 compression engine and Huffman encodes the tokens using a Huffman code table to generate a compressed output block. The hardware data compressor is without a memory to intermediately store the back pointers and instead, for each token produced by the LZ77 engine, the Huffman encoding engine directly receives the token from the LZ77 engine and outputs Huffman codes corresponding in the Huffman code table to respective symbols associated with the token before the LZ77 engine produces the next token.

Abstract: A microprocessor includes a plurality of processing cores each comprises a corresponding memory physically located inside the core and readable by the core but not readable by the other cores (“core memory”). The microprocessor also includes a memory physically located outside all of the cores and readable by all of the cores (“uncore memory”). For each core, the uncore memory and corresponding core memory collectively provide M words of storage for microcode instructions fetchable by the core as follows: the uncore memory provides J of the M words of microcode instruction storage, and the corresponding core memory provides K of the M words of microcode instruction storage. J, K and M are counting numbers, and M=J+K. The memories are non-architecturally-visible and accessed using a fetch address provided by a non-architectural program counter, and the microcode instructions are non-architectural instructions that implement architectural instructions.

Abstract: A microprocessor includes a control unit configured to selectively control a respective clock signal to each of a plurality of processing cores. Each of the processing cores is configured to separately write a value to the control unit. For each core of the plurality of processing cores, the control unit is configured to turn off the respective clock signal to the core in response to the core writing a value to the control unit. The control unit is configured to detect a condition has occurred when all of the processing cores have written a value to the control unit and the control unit has turned off the respective clock signal to all of the processing cores. The control unit is configured to simultaneously turn on the respective clock signal to all of the processing cores in response to detecting the condition has occurred.

Abstract: A method for encrypting a program for subsequent execution by a microprocessor configured to decrypt and execute the encrypted program includes receiving an object file specifying an unencrypted program that includes conventional branch instructions whose target address may be determined pre-run time. The method also includes analyzing the program to obtain chunk information that divides the program into a sequence of chunks each comprising a sequence of instructions and that includes encryption key data associated with each of the chunks. The encryption key data associated with each of the chunks is distinct. The method also includes replacing each of the conventional branch instructions that specifies a target address that is within a different chunk than the chunk in which the conventional branch instruction resides with a branch and switch key instruction. The method also includes encrypting the program based on the chunk information.

Abstract: A microprocessor includes functional units and control registers writeable to cause the functional units to institute actions that reduce the instructions-per-clock rate of the microprocessor to reduce power consumption when the microprocessor is operating in its lowest performance running state. Examples of the actions include in-order vs. out-of-order execution, serial vs. parallel cache access and single vs. multiple instruction issue, retire, translation and/or formatting per clock cycle. The actions may be instituted only if additional conditions exist, such as residing in the lowest performance running state for a minimum time, not running in a higher performance state for more than a maximum time, a user did not disable the feature, the microprocessor supports multiple running states and the operating system supports multiple running states.

Abstract: A microprocessor includes compressed and uncompressed microcode memory storages, having N-bit wide and M-bit wide addressable words, respectively, where N<M. The microprocessor also includes a fetch unit, an instruction translator, and an execution stage. When the instruction translator receives an architectural instruction, it writes information identifying source and destination registers specified by the architectural instruction to an indirection register. It also issues one or more fetch addresses to retrieve a sequence of one or more microcode instructions from one of the uncompressed microcode memory storage and the compressed microcode memory storage to implement the architectural instruction. It merges information in the indirection register with the sequence of one or more microcode instructions to generate a sequence of one or more implementing microinstructions.

Abstract: A microprocessor natively translates and executes instructions of both the x86 instruction set architecture (ISA) and the Advanced RISC Machines (ARM) ISA. An instruction formatter extracts distinct ARM instruction bytes from a stream of instruction bytes received from an instruction cache and formats them. ARM and x86 instruction length decoders decode ARM and x86 instruction bytes, respectively, and determine instruction lengths of ARM and x86 instructions. An instruction translator translates the formatted x86 ISA and ARM ISA instructions into microinstructions of a unified microinstruction set architecture of the microprocessor. An execution pipeline executes the microinstructions to generate results defined by the x86 ISA and ARM ISA instructions.

Abstract: A processor includes a decoder that decodes an instruction that instructs the processor to perform subsequent computations in an approximate manner and a functional unit that performs the subsequent computations in the approximate manner in response to the instruction. An instruction instructs the processor to clear an error amount associated with a value stored in a general purpose register of the processor. The error amount indicates an amount of error associated with a result of a computation performed by the processor in an approximate manner. The processor also clears the error amount in response to the instruction. Another instruction specifies a computation to be performed and includes a prefix that indicates the processor is to perform the computation in an approximate manner. The functional unit performs the computation specified by the instruction in the approximate manner specified by the prefix.

Abstract: A microprocessor instruction translator translates a conditional load instruction into at least two microinstructions. An out-of-order execution pipeline executes the microinstructions. To execute a first microinstruction, an execution unit receives source operands from the source registers of a register file and responsively generates a first result using the source operands. To execute a second the microinstruction, an execution unit receives a previous value of the destination register and the first result and responsively reads data from a memory location specified by the first result and provides a second result that is the data if a condition is satisfied and that is the previous destination register value if not. The previous value of the destination register comprises a result produced by execution of a microinstruction that is the most recent in-order previous writer of the destination register with respect to the second microinstruction.

Abstract: A microprocessor includes a plurality of memories each configured to hold microcode instructions. At least a first of the plurality of memories is configured to provide M-bit wide words of compressed microcode instructions, and at least a second of the plurality of memories is configured to provide N-bit wide words of uncompressed microcode instructions. M and N are integers greater than zero and N is greater than M. The microprocessor also includes a decompression unit configured to decompress the compressed microcode instructions after being fetched from the at least a first of the plurality of memories and before being executed.

Abstract: An asymmetric multi-core processor uses at least two asymmetric cores to collectively support the instructions of an instruction set architecture (ISA). A general-feature core and a special feature core that support different instruction subsets of the ISA. A switch manager detects whether a thread includes an instruction that is not supported by the currently-executing core and, after detecting such an instruction, switches the thread to the other core.