Mechanism handling race conditions in FRC-enabled processors
AssigneeIntel Corporation
InventorsSteven J. Tu, Alexander J. Honcharik, Hang T. Nguyen, Sujat Jamil, Quinn W. Merrell
Life of this patent
Estimating the term…
Abstract
An processor includes first and second execution cores that operate in an FRC mode, an FRC check unit to compare results from the first and second execution cores, and an error check unit to detect recoverable errors in the first and second cores. The FRC check unit temporarily stores results from the first or second core, and a timer is activated if a mismatch is detected. If the error detector detects a recoverable error before the timer interval expires, a recovery routine is activated. If the timer interval expires first, a reset routine is activated.
Claims20
An apparatus comprising:
- first and second execution cores to operate in an FRC mode;
- an FRC check unit to compare results from the first and second execution cores and to store at least one result and a status to indicate if the results match;
- an error check unit to assert a signal to the FRC checker if a recoverable error is detected in the first or second execution cores;
- and
- a timer to trigger an FRC recovery routine if the status indicates the results do not match and the error check unit does not assert the signal within a specified interval.
The apparatus of claim 4, wherein the specified interval represents a time for a recoverable error to be signaled to the error detector.
An system comprising:
- first and second execution cores to operate in an FRC mode;
- an FRC checker to compare results from the first and second execution cores and to trigger a countdown interval if the results do not match;
- and
- an error detector to monitor error signals during the countdown interval and to disable the FRC checker if a recoverable error is detected before the countdown interval expires.
A method comprising:
- comparing results from a first and second execution core to detect an FRC error;
- if the results do not match, setting a first flag and initiating a countdown interval;
- monitoring an error signal for a recoverable error;
- and
- initiating a recovery routine if the error signal is asserted before the countdown interval expires.
Dependencies are inferred from the claim text.