Abstract: A neural network unit convolves an H×W×C input with F R×S×C filters to generate F Q×P outputs. N processing units (PU) each have a register receiving a respective word of an N-word row of a second memory and multiplexed-register selectively receiving a respective word of an N-word row of a first memory or word rotated from an adjacent PU multiplexed-register. H first memory rows hold input blocks of B words each of channels of respective 2-dimensional input row slices. R×S×C second memory rows hold filter blocks of B words each holding P copies of a filter weight. B is the smallest factor of N greater than W. The PU blocks multiply-accumulate input blocks and filter blocks in column-channel-row order; they read a row of input blocks and rotate it around the N PUs while performing multiply-accumulate operations so each PU block receives each input block before reading another row.

Abstract: A microprocessor includes an indicator and a plurality of processing cores. Each of the plurality of processing cores is configured to sample the indicator. When the indicator indicates a first predetermined value, the plurality of processing cores are configured to collectively designate multiple of the plurality of processing cores to be a bootstrap processor. When the indicator indicates a second predetermined value distinct from the first predetermined value, the plurality of processing cores are configured to collectively designate a single processing core of the plurality of processing cores to be the bootstrap processor.

Abstract: A secure memory, key expansion logic, and decryption logic are provided for a microprocessor that executes encrypted instructions. The secure memory stores a plurality of decryption key primitives. The key expansion logic selects two or more decryption key primitives from the secure memory and then derives a decryption key from them. The decryption logic uses the decryption key to decrypt an encrypted instruction fetched from the instruction cache. The decryption key primitives are selected on the basis of an encrypted instruction address, one of them is rotated by an amount also determined by the encrypted instruction address, and then they are additively or subtractively accumulated, also on the basis of the encrypted instruction address.

Abstract: A multi-core microprocessor supports a plurality of operating states that provide different levels of performance and power consumption to the microprocessor and its cores. A control unit puts selected cores into selected operating states at selected times. A core-specific synchronization register is provided for each core external to the core and readable by the control unit. Each core responds to an instruction to target an operating state by writing a value identifying the target operating state to the synchronization register. The control unit causes power saving actions that affect shared resources provided that the actions do not reduce performance of any core sharing the resources below the core's target operating state.

Abstract: An apparatus includes a first reservation station and a second reservation station. The first reservation station dispatches a first load micro instruction, and detects and indicates on a hold bus if the first load micro instruction is a specified load micro instruction directed to retrieve an operand from a prescribed resource other than on-core cache memory. The second reservation station is coupled to the hold bus, and dispatches one or more younger micro instructions therein that depend on the first load micro instruction for execution after a first number of clock cycles following dispatch of the first load micro instruction, and if it is indicated on the hold bus that the first load micro instruction is the specified load micro instruction, the second reservation station is configured to stall dispatch of the one or more younger micro instructions until the first load micro instruction has retrieved the operand.

Abstract: A microprocessor is provided in which an encrypted program can replace the decryption keys that are used to decrypt sections of the encrypted program. The microprocessor may be decrypting and executing a first section of the encrypted program when it encounters, decrypts, and executes an encrypted store-key instruction to store a new set of decryption keys. After executing the store-key instruction, the microprocessor decrypts and executes a subsequent section of the encrypted program using the new set of decryption keys. On-the-fly key switching may occur numerous times with successive encrypted store-key instructions and successive sets of encrypted instructions.

Abstract: An apparatus is provided for protecting a basic input/output system (BIOS) in a computing system. The apparatus includes a BIOS read only memory (ROM), an event detector, and a tamper detector. The BIOS ROM has BIOS contents that are stored as plaintext, and an encrypted message digest, where the encrypted message digest comprises an encrypted version of a first message digest that corresponds to the BIOS contents, and where and the encrypted version is generated via a symmetric key algorithm and a key. The event detector is configured to generate a BIOS check interrupt that interrupts normal operation of the computing system upon the occurrence of an event, where the event includes one or more occurrences of an operating system call.

Abstract: A microprocessor includes a plurality of processing cores, a resource shared by the plurality of processing cores, and a hardware semaphore readable and writeable by each of the plurality of processing cores within a non-architectural address space. Each of the plurality of processing cores is configured to write to the hardware semaphore to request ownership of the shared resource and to read from the hardware semaphore to determine whether or not the ownership was obtained. Each of the plurality of processing cores is configured to write to the hardware semaphore to relinquish ownership of the shared resource.

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 microprocessor a plurality of processing cores, wherein each of the plurality of processing cores instantiates a respective architecturally-visible storage resource. A first core of the plurality of processing cores is configured to encounter an architectural instruction that instructs the first core to update the respective architecturally-visible storage resource of the first core with a value specified by the architectural instruction. The first core is further configured to, in response to encountering the architectural instruction, provide the value to each of the other of the plurality of processing cores and update the respective architecturally-visible storage resource of the first core with the value. Each core of the plurality of processing cores other than the first core is configured to update the respective architecturally-visible storage resource of the core with the value provided by the first core without encountering the architectural instruction.

Abstract: A microprocessor includes a plurality of processing cores and an uncore random access memory (RAM) readable and writable by each of the plurality of processing cores. Each core of the plurality of processing cores comprises microcode run by the core that implements architectural instructions of an instruction set architecture of the microprocessor. The microcode is configured to both read and write the uncore RAM to accomplish inter-core communication between the plurality of processing cores.

Abstract: A microprocessor and method are provided for securely decrypting and executing encrypted instructions within a microprocessor. A plurality of master keys are stored in a secure memory. Encrypted instructions are fetched from an instruction cache. A set of one or more master keys are selected from the secure memory based upon an encrypted instruction fetch address. The selected set of master keys or a decryption key derived therefrom is used to decrypt the encrypted instructions fetched from the instruction cache. The decrypted instructions are then securely executed within the microprocessor. In one implementation, the master keys are intervolved with each other to produce a new decryption key with every fetch quantum. Moreover, a new set of master keys is selected with every new block of instructions.

Abstract: A microprocessor performs an If-Then (IT) instruction and an associated IT block by extracting condition information from the IT instruction and for each instruction of the IT block: determining a respective condition for the instruction using the extract condition information, translating the instruction into a microinstruction, and conditionally executing the microinstruction based on the respective condition. For a first instruction, the translating comprises fusing the IT instruction with the first IT block instruction. A hardware instruction translation unit performs the extracting, determining and translating. Execution units conditionally execute the microinstructions. The hardware instruction translation unit and execution units are distinct hardware elements and are coupled together.

Abstract: An apparatus is provided for protecting a basic input/output system (BIOS) in a computing system. The apparatus includes a BIOS read only memory (ROM), an event detector, and a tamper detector. The BIOS ROM has BIOS contents that are stored as plaintext, and an encrypted message digest, where the encrypted message digest comprises an encrypted version of a first message digest that corresponds to the BIOS contents, and where and the encrypted version is generated via a symmetric key algorithm and a key. The event detector is configured to generate a BIOS check interrupt that interrupts normal operation of the computing system upon the occurrence of an event, where the event includes one or more occurrences of a hard disk access.

Abstract: An apparatus is provided for protecting a basic input/output system (BIOS) in a computing system. The apparatus includes a BIOS read only memory (ROM), an event detector, and a tamper detector. The BIOS ROM has BIOS contents that are stored as plaintext, and an encrypted message digest, where the encrypted message digest comprises an encrypted version of a first message digest that corresponds to the BIOS contents, and where and the encrypted version is generated via a symmetric key algorithm and a key. The event detector is configured to generate a BIOS check interrupt that interrupts normal operation of the computing system upon the occurrence of an event, where the event includes one or more occurrences of a change in virtual memory mapping.

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 multi-die package for a microprocessor provides a power management synchronization system. The package has a plurality of dies. Each die has a plurality of cores, including a single master core. A plurality of sideband non-system-bus inter-die communication wires communicatively couple the dies to each other for a purpose of synchronizing power management. The master core of each die is configured to use one and only one of the inter-die communication wires to transmit power management synchronization messages to each of the other master cores. The master core of each die is also configured to receive power management synchronization messages from each of the other master cores via one or more inter-die communication wires. The cores use this system of inter-die communication wires to synchronize management of resources that affect both the performance and power consumption of the cores.

Abstract: A microprocessor includes an indicator and a plurality of processing cores. Each of the plurality of processing cores is configured to generate a default core ID and to sample the indicator. When the indicator indicates a first predetermined value, the default core ID generated by a default one of the plurality of processing cores designates the default processing core to be a bootstrap processor. When the indicator indicates a second predetermined value distinct from the first predetermined value, the plurality of processing cores are configured to generate alternate core IDs that are different from the default core IDs. One of the alternate core IDs designates an alternate processing core, other than the default processing core, to be the bootstrap processor.

Abstract: An apparatus is provided for protecting a basic input/output system (BIOS) in a computing system. The apparatus includes a BIOS read only memory (ROM), an event detector, and a tamper detector. The BIOS ROM has BIOS contents that are stored as plaintext, and an encrypted message digest, where the encrypted message digest comprises an encrypted version of a first message digest that corresponds to the BIOS contents, and where and the encrypted version is generated via a symmetric key algorithm and a key. The event detector is configured to generate a BIOS check interrupt that interrupts normal operation of the computing system upon the occurrence of an event, where the event includes one or more occurrences of a PCI Express access.

Abstract: An apparatus including first and second reservation stations. The first reservation station dispatches a load micro instruction, and indicates on a hold bus if the load micro instruction is a specified load micro instruction directed to retrieve an operand from a prescribed resource other than on-core cache memory, where the specified load instruction comprises a load instruction resulting from execution of an x86 special bus cycle. The second reservation station is coupled to the hold bus, and dispatches one or more younger micro instructions therein that depend on the load micro instruction for execution after a number of clock cycles following dispatch of the first load micro instruction, and if it is indicated on the hold bus that the load micro instruction is the specified load micro instruction, the second reservation station is configured to stall dispatch of the one or more younger micro instructions until the load micro instruction has retrieved the operand.