摘要:
A mechanism for recycling error bits in a floating point unit is disclosed. A system of the disclosure includes a memory and a processing device communicably coupled to the memory. In one embodiment, the processing device comprising a floating point unit (FPU) to generate a result value from applying an operation on floating point number inputs to the FPU and generate an error value using the result value. The FPU also writes the result value to a first register of the processing device dedicated to storing results from the operation of the FPU and writes the error value to a second register of the processing device dedicated to storing errors from the operation of the FPU.
摘要:
A method, device and system for representing numbers in a computer including storing a floating-point number M in a computer memory; representing the floating-point number M as an interval with lower and upper bounds A and B when it is accessed by using at least two floating-point numbers in the memory; and then representing M as an interval with lower and upper bounds A and B when it is used in a calculation by using at least three floating-point numbers in the memory. Calculations are performed using the interval and when the data is written back to the memory it may be stored as an interval if the size of the interval is significant, i.e. larger than a first threshold value. A warning regarding the suspect accuracy of any data stored as an interval may be issued if the interval is too large, i.e. larger than a second threshold value.
摘要:
In a memory device where writing a memory cell to a first bit state takes longer than writing to the second bit state, selectively executing the write operation can amortize the performance cost of writing the bit state that takes longer to write. Write logic dequeues multiple cachelines from a write buffer and sets all bits of all cachelines to the first bit state in a single write operation. The write logic then executes separate write operations on each cacheline separately to selectively write memory cells of each respective cacheline to the second bit state.
摘要:
Embodiments of the invention relate to a mobile computing device with ambient energy harvesting capability. Embodiments of the invention, when manually operated by a user, convert the kinetic motion of a part of the user's hand, applied onto a controller of the device, to electrical energy. The energy can be used to power the device, or to charge the battery of the device. Embodiments of the invention include an electrical power storage device disposed in a housing, a display screen attached to the housing to display a plurality of user-interactive interfaces, and a manually operable input controller interactable with the interfaces and being coupled to an energy transformer in the housing to electrically charge the power storage device when operated.
摘要:
A method for constructing and addressing a nanoscale memory with known addresses and for tolerating defects which may arise during manufacture or device operational lifetime. During construction, nanoscale wires with addresses are stochastically assembled. During a programming phase, nanoscale wires are stochastically selected using their stochastic addresses through microscale inputs and a desired address code is associated with the selected nanoscale wires. Memory addresses are associated to the codes and then selected using the known codes during read/write operations from/to the memory.
摘要:
A method for constructing and addressing a nanoscale memory with known addresses and for tolerating defects which may arise during manufacture or device operational lifetime. During construction, nanoscale wires with addresses are stochastically assembled. During a programming phase, nanoscale wires are stochastically selected using their stochastic addresses through microscale inputs and a desired address code is associated with the selected nanoscale wires. Memory addresses are associated to the codes and then selected using the known codes during read/write operations from/to the memory.
摘要:
In one embodiment, the present invention includes a method for generating a hybrid error correction code for a data block. The hybrid code, which may be a residual arithmetic-Hamming code, includes a first residue code based on the data block and a first parity code based on the data block and a Hamming matrix. Then the generated code along with the data block can be communicated through at least a portion of a datapath of a processor. Other embodiments are described and claimed.
摘要:
Generally, this disclosure provides error management across hardware and software layers to enable hardware and software to deliver reliable operation in the face of errors and hardware variation due to aging, manufacturing tolerances, etc. In one embodiment, an error management module is provided that gathers information from the hardware and software layers, and detects and diagnoses errors. A hardware or software recovery technique may be selected to provide efficient operation, and, in some embodiments, the hardware device may be reconfigured to prevent future errors and to permit the hardware device to operate despite a permanent error.
摘要:
A semiconductor chip is described having ECC decoder circuitry disposed along any of: i) an interconnect path that resides between an instruction execution core and a cache; ii) an interconnect path that resides between an instruction execution core and a memory controller; and, iii) an interconnect path that resides between a cache and a memory controller. The ECC decoder circuitry has an input register to receive data, CRC values associated with the data and residue information associated with the data.