| 摘要 |
A computer system provides a vector monitor for monitoring a first instance of an error-handling vector in architected memory. The monitoring can involve repeatedly comparing the first instance with a second instance of the vector so as to detect a mismatch, should it occur. If a mismatch is detected, the vector monitor can notify an administrator, automatically initiate diagnostic routines, and/or correct the mismatch. As a result, potential fatal events in which firmware confronts a corrupted error-handling vector are less likely to occur.
|
