{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}



Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Test Generation 237 TABLE 4.12 ± AND Operation AND S 0 U 0 S 1 U 1 XX S 0 S 0 S 0 S 0 S 0 S 0 U 0 S 0 U 0 U 0 U 0 U 0 S 1 S 0 U 0 S 1 U 1 XX U 1 S 0 U 0 U 1 U 1 XX XX S 0 U 0 XX XX XX TABLE 4.13 ± NOT Operation NOT S 0 S 1 U 0 U 1 S 1 S 0 U 1 U 0 XX XX TABLE 4.14 ± OR Operation OR S 0 U 0 S 1 U 1 XX S 0 S 0 U 0 S 1 U 1 XX U 0 U 0 U 0 S 1 U 1 XX S 1 S 1 S 1 S 1 S 1 S 1 U 1 U 1 U 1 S 1 U 1 U 1 XX XX XX S 1 U 1 XX common. In RESIST [Fuchs 1994], this concept is taken into account such that a recursion-based ATPG algorithm searches starting from a primary input. The search progresses by targeting paths that differ only in the last segment; in other words, they share the same initial subpath. Essentially, RESIST starts at each primary input, and the circuit is traversed in a depth-first fashion. Once a complete path P from a primary input to primary output has been tested, a backtrack is invoked and a different path P 2 that differs from P in only one segment is tried. At the end, all the paths in the output cone of
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: the starting primary input would have been considered. In doing so, the decision tree can be shared among different paths and knowledge is reused. Likewise, if a subpath is found to be untestable, all paths that share the same initial subpath would be untestable. Then, RESIST repeats for another primary input until all primary inputs have been processed. RESIST is efficient because it incorporates knowledge into the ATPG process, and paths are handled such that much knowledge can be carried over from one path to the next....
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online