公开(公告)号：US20190082036A1

公开(公告)日：2019-03-14

申请号：US15704343

申请日：2017-09-14

Applicant: ARM LIMITED

Inventor： Mark Gerald LAVINE

IPC: H04L29/06 ,  H04L12/26 ,  H04L12/751

Abstract: An apparatus is provided, which includes monitor circuitry for monitoring the behaviour of processing circuitry and for generating at least a packet indicating the behaviour. Compare circuitry compares the packet to a base packet and generates a delta packet based on the comparison. Compress circuitry forms a compressed packet on the basis of the delta packet. Output circuitry outputs the compressed packet and update circuitry updates the base packet in dependence on one or more triggers.

公开(公告)号：US20210208968A1

公开(公告)日：2021-07-08

申请号：US16732465

申请日：2020-01-02

Applicant: Arm Limited

Inventor： Mark Gerald LAVINE ,  Simon John CRASKE

IPC: G06F11/10 ,  G11C29/36

Abstract: A data processing apparatus is provided, which includes storage circuitry comprising a plurality of lines, each of the plurality of lines comprising a data value. Access circuitry accesses a pair of the plurality of lines at a time, the pair of the plurality of lines comprising a further data value, distinct from the data value, and a plurality of error bits to detect or correct errors in the data value in each line in the pair of the plurality of lines.

公开(公告)号：US20220398145A1

公开(公告)日：2022-12-15

申请号：US17343820

申请日：2021-06-10

Applicant: Arm Limited

Inventor： Mark Gerald LAVINE

IPC: G06F11/07 ,  G06F13/40

Abstract: Apparatuses and methods for extracting ordered data sub-components from a data item are disclosed. A received data item has a data structure to accommodate multiple data sub-components. The data item indicates which data sub-components are valid. Adders sum respective subsets of indications of the valid data sub-component positions, with each adder covering one more position than the previous adder. Transitions of the counts generated by the respective adders are used to determine the ordinal valid data sub-component positions in the data item, which can then be output on the basis of the data item and the identified transition positions. Without requiring feedback paths from an identified earlier ordinal position to identify a later ordinal position, the set of ordered data sub-components can be extracted more quickly.