An IC module, such as a Multi-Chip Module (MCM), includes multiple IC dice, each having a test mode enable bond pad, such as an output enable pad. A fuse incorporated into the MCM's substrate connects each die's test mode enable bond pad to one of the MCM's no-connection (N/C) pins, and a resistor incorporated into the substrate connects the test mode enable bond pads to one of the MCM's ground pins. By applying a supply voltage to the test mode enable bond pads through the N/C pin, a test mode is initiated in the dice. Once testing is complete, the fuse may be blown, and a ground voltage applied to the test mode enable bond pads through the ground pins so the resistor disables the test mode in the dice and initiates an operational mode. As a result, dice packaged in IC modules may be tested after packaging.

Device and method for testing integrated circuit dice in an integrated circuit module

Control is switched from a first server to a second server in a fault tolerant system. The first and second servers are coupled with an expansion bus in an expansion box for communication with the expansion bus. An indication is provided to the second server to indicate the activity state of the first server. Communication between the first server and the expansion box is disabled if the indication indicates the first server is inactive. Communication between the second server and the expansion bus is disabled if the indication indicates that the first server is active. Communication between the second server is enabled if the indication indicates that the first server is inactive. The indication includes a heartbeat message transmitted periodically to the second server. The expansion bus includes a PCI bus.

Switching control in a fault tolerant system

A device causing a faulty condition in a computer system having devices is isolated by detecting for a faulty condition associated with the devices and identifying the device causing the faulty condition. The devices are coupled to a bus. The faulty condition includes a bus hang condition. The devices are turned off when a bus hang condition is detected. The devices are then turned back on to test the devices. Each device is tested by writing and reading its configuration space. Information on the bus associated with the faulty condition is stored. The stored information is retrieved after the faulty condition has occurred, with the stored information including address, data, and bus control information.

Method and apparatus for identifying faulty devices in a computer system

A method of determining which of a plurality of electrical devices connected to a node of a circuit under test is causing a failure at that node by causing individual devices to drive the node one device at a time, taking passive voltage measurements at the node at separate times when individual devices are controlled to drive the node, and analyzing measured voltages resulting when different devices are driving the node to identify which is causing the failure.

Diagnosis of faults on circuit board

A method for testing a circuit board having both boundary-scan and non-boundary-scan devices is provided. The test method distinguishes boundary-scan nodes from non-boundary-scan nodes and uses cartesian coordinates (X,Y) of every pin of every device on the circuit board to determine a number of sets of non-boundary-scan nodes that are within a predetermined distance "R" from a device pin coupled to a boundary-scan node. The number of sets of non-boundary-scan nodes are grouped into "independent" groups which can be tested in parallel. A test cycle is performed by testing independent non-boundary-scan nodes in parallel by forcing drivers in the boundary-scan devices to a first logic state, and forcing each of the non-boundary-scan nodes to another logic state for a brief interval. Receivers on the boundary-scan devices capture a response vector during the brief interval, which is scanned out of the circuit board for evaluation. The response vector reflects which of the nodes has failed, and cartesian coordinate data giving the precise location of any faulty node(s) is returned to user.

Ronald E. Roetzer

Papers

Method of and apparatus for the automatic diagnosis of the failure of electrical devices connected to common bus nodes and the like