Self-diagnostic scheme for detecting errors

A method for processing memory-related errors on a multi-chip package (MCP), comprising:

• executing a write or read command from a controller located in the MCP to access data in a volatile memory chip;
• detecting an error associated with execution of the write or read command;
• storing fail signature data associated with the error in a fail signature storage element located in the volatile memory chip;
• obtaining, by the controller, the fail signature data from the fail signature storage element;
• and
• storing the fail signature data in a non-volatile memory chip of the MCP.

A system, comprising:

• a printed-circuit board (PCB);
• and
• a multi-chip package, comprising:
• a first chip, comprising error detection logic and a first fail signature storage element;
• wherein the first chip is a volatile memory chip;
• a second chip, comprising a second fail signature storage element;
• wherein the second chip is a non-volatile memory chip;
• and
• a controller chip configured to copy the fail signature data from the first chip to the second chip;
• wherein the first and second fail signature storage elements are configured for storing fail signature data related to memory errors detected in a multi-chip package;
• and wherein the chips are encapsulated in packaging thereby forming the multi-chip package;
• wherein the multi-chip package is disposed on the PCB.