Abstract: Circuitry for decomposing block floating-point numbers into lower precision floating-point numbers is provided. The circuitry may include a high precision storage circuit configured to provide high precision floating-point numbers, input selectors configured to receive the high precision floating-point numbers from the high precision storage circuit and to generate corresponding lower precision floating-point components with adjusted exponents, and a low precision block floating-point vector circuit configured to combine the various lower precision floating-point components generated by the input selectors. The lower precision floating-point components may be processed spatially or over multiple iterations over time.

Abstract: Circuitry for decomposing block floating-point numbers into lower precision floating-point numbers is provided. The circuitry may include a high precision storage circuit configured to provide high precision floating-point numbers, input selectors configured to receive the high precision floating-point numbers from the high precision storage circuit and to generate corresponding lower precision floating-point components with adjusted exponents, and a low precision block floating-point vector circuit configured to combine the various lower precision floating-point components generated by the input selectors. The lower precision floating-point components may be processed spatially or over multiple iterations over time.

Abstract: Circuitry for decomposing block floating-point numbers into lower precision floating-point numbers is provided. The circuitry may include a high precision storage circuit configured to provide high precision floating-point numbers, input selectors configured to receive the high precision floating-point numbers from the high precision storage circuit and to generate corresponding lower precision floating-point components with adjusted exponents, and a low precision block floating-point vector circuit configured to combine the various lower precision floating-point components generated by the input selectors. The lower precision floating-point components may be processed spatially or over multiple iterations over time.

Abstract: Circuitry for decomposing block floating-point numbers into lower precision floating-point numbers is provided. The circuitry may include a high precision storage circuit configured to provide high precision floating-point numbers, input selectors configured to receive the high precision floating-point numbers from the high precision storage circuit and to generate corresponding lower precision floating-point components with adjusted exponents, and a low precision block floating-point vector circuit configured to combine the various lower precision floating-point components generated by the input selectors. The lower precision floating-point components may be processed spatially or over multiple iterations over time.

Abstract: Circuitry for decomposing block floating-point numbers into lower precision floating-point numbers is provided. The circuitry may include a high precision storage circuit configured to provide high precision floating-point numbers, input selectors configured to receive the high precision floating-point numbers from the high precision storage circuit and to generate corresponding lower precision floating-point components with adjusted exponents, and a low precision block floating-point vector circuit configured to combine the various lower precision floating-point components generated by the input selectors. The lower precision floating-point components may be processed spatially or over multiple iterations over time.

Abstract: Devices and methods for filtering data include calculating intermediate input values from input elements using a transformation function. The transformation function is based at least in part on a size of the filter and a number of filter outputs. Intermediate filter values are calculated from filter elements of the filter using the transformation function. Each intermediate input value is multiplied with a respective intermediate filter value to form intermediate values. These intermediate values are combined with each other using the transformation function to determine one or more output values.

Abstract: Devices and methods for filtering data include calculating intermediate input values from input elements using a transformation function. The transformation function is based at least in part on a size of the filter and a number of filter outputs. Intermediate filter values are calculated from filter elements of the filter using the transformation function. Each intermediate input value is multiplied with a respective intermediate filter value to form intermediate values. These intermediate values are combined with each other using the transformation function to determine one or more output values.

Abstract: The present disclosure provides methods and apparatus for rapid interrupt look-ups for interrupts stored in memory. One embodiment relates to a method for providing interrupt lookups for a plurality of interrupt status vectors stored in random access memory on an integrated circuit. The plurality of interrupt status vectors in the random access memory are scanned to find activated interrupt status vectors that changed from null to non-null and dismissed interrupt status vectors that changed from non-null to null. A linked search list is maintained in the random access memory by inserting memory addresses of the activated interrupt status vectors into the linked search list and removing memory addresses of the dismissed interrupt status vectors from the linked search list. Interrupt status vectors for currently active interrupts are looked-up by transversing the linked search list in the random access memory. Other embodiments, aspects and features are also disclosed herein.