A subsystem can maintain two subsystem vector tables. The two tables can describe different sets of functions to which the subsystem responds or identify different function routines to be invoked for the same function codes.

A subsystem would find it useful to maintain two subsystem vector tables if, for example, a subsystem must quiesce operations. This way, a subsystem can keep one full-function vector table and a second limited-function vector table, and swap so that it can continue to support some minimum set of function while shutting down.

The swap request of the IEFSSI macro allows the subsystem to deactivate the active vector table and activate the inactive table in a single operation. The swap request eliminates the need for separate deactivate and activate requests, which would result in a period of time when the subsystem cannot respond to requests.

Inputs

The swap request allows the user to specify a subsystem vector table token on input. The input token, which is named with the INTOKEN parameter, identifies the vector table that is to be activated (with the activate request or command). If INTOKEN is not specified, the inactive (previously created) vector table is activated.

Outputs:

The swap request allows the user to specify a subsystem vector table token on output. On completion of the swap, the output token, which is named with the OUTTOKEN parameter, identifies the outgoing (previously active) vector table.

If the subsystem is initially inactive, the swap request receives the IEFSSI_WARNING (4) return code and is treated as an activate request. The output token identified with the OUTTOKEN parameter contains a zero. If the outgoing (initially active) vector table is not managed by the SSI, the output token contains a zero and the request receives the IEFSSI_WARNING return code.