1. Field of the Invention
The present invention generally relates to a method and apparatus for bypassing memory arrays and in particular a method and apparatus for bypassing memory arrays in a very large scale integration (VLSI) circuit without an additional logic element.
2. Description of the Related Art
Memory arrays occupy a large portion of conventional very large scale integration (VLSI) circuits. Besides the basic functionality of read and write operations, additional (supporting) functions are needed, especially for an in-system test of the memory itself or of the logic attached to it. One of these features is conventionally referred to as a write-around or bypass circuit. It is used to display data from a separate port at the output of the memory, without affecting the array content.
When arrays get involved in a logic cone (e.g., multiple arrays or logic elements feeding into the same receiving flip-flop), performing the logic test becomes more difficult. This difficulty increases even more so in the case the array does not contain a scannable latch at the output but is directly connected to some downstream logic. In this configuration, it becomes more difficult to apply appropriate test-vectors to the logic cone. For this case, the “bypass” or “write-around” feature is required. The bypass feature is used to display data from a separate port at the output of the memory, without affecting the array content. A well-known and widely used method for realizing this function is a multiplexor circuit, which is connected to the memory output.
Adding a multiplexor to the VLSI circuit has the disadvantages in that an additional logic element has to be placed in the timing critical output path of the memory. The increased delay caused by the multiplexor limits the achievable maximum operating frequency. Timing behavior of the bypassed data is different than that of the read data. This timing difference makes time sensitive testing of the down-stream logic almost impossible.
In view of the foregoing, and other, exemplary problems, drawbacks, and disadvantages of the conventional systems, it is an exemplary feature of the present invention to provide a method for bypassing a memory array in a circuit having a global bit line, a test port configured to output a logic test, a memory portion connected to the global bit line via a word line, a header device connected to the global bit line via a pre-charge signal, the header device configured to recharge the global bit line, the method including sending a gating signal to a gating device connected to the header device, switching the gating device to a test mode upon receipt of the gating signal, sending a bypass data signal to an evaluating device connected to the gating device, the evaluating device configured to output a logic test, and outputting the logic test through the test port upon receipt of bypass data signal.
The present invention provides a bypass feature for memory arrays. The proposed circuits minimize the impact on the read delay present in conventional circuit. Furthermore, the present invention provides a circuit that helps to mimic the regular functional mode read delay, because large portions of the regular read path are also used for the bypassed data.
This feature is important when the bypass feature is used for performance testing of the downstream logic which is attached to the array output. Finally, the present invention can be easily attached to existing memory designs, minimizing the required device count and device overhead.
The foregoing and other purposes, aspects and advantages will be better understood from the following detailed description of an exemplary embodiment of the invention with reference to the drawings, in which:
Referring now to the drawings, and more particularly to
In some embodiments, keeper circuit 370 would exemplarily be added to the dynamic node for leakage protection. Global bitline readout circuit 380 may exemplarily be used to detect and store the global bitline level and would complete the circuit.
Exemplarily, in a regular array read operation the global bitline 310 will be discharged using the evaluation device 330 based on the selected memory cell 340 content using word-line 314, local bitline 312 and the local evaluation circuit 334. At the end of the read cycle, readout circuit 380 stores the state of the global bitline 310 and passes it to the output 390.
Exemplarily, evaluation leg 400 would be connected to global bitline 310 and would exemplarily be used for the memory bypass functionality. Evaluation leg 400 would exemplarily include gating device 410, an evaluation device 420 and a footer device 430.
Footer device 430 is connected to pre-charge device 320 and avoids excessive cross-currents, in case pre-charge device 320 and the gating device 410 and evaluation device 420 are switched on at the same time. The to-be-bypassed data is exemplarily applied to the evaluation device 420. Where a bypass operation occurs, gating device 410 would exemplarily be switched on, such as when a gating signal is sent and received by gating device 410, whereas all word-lines for the VLSI, such as word-line 314, would exemplarily be in an OFF state. That is, exemplarily, none of the memory cells (such as memory cell 340) is selected. Footer device 430 would be connected to global bit 310 line via a pre-charge signal sent by pre-charge device 320.
A bypass data signal is sent to evaluating device 420 and the evaluating device 420 outputs a logic test signal. Depending on the bypass input data, global bitline 310 is then discharged through evaluation device 420, the gating device 410, and the footer device 430.
Exemplarily, transistor 512 would possess both an evaluation and gating function of the exemplary embodiment illustrated by the gating device 410 and footer device 430. Device 514 and transistor 512 would form bypass device 510. The gating signal would be applied to the gate of transistor 512 and the to-be-bypassed data is applied to the source of transistor 512, or vice-versa. In the bypass mode, bypass device 510 would be switched on using the gating signal.
While the invention has been described in terms of exemplary embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.
Further, it is noted that, Applicants' intent is to encompass equivalents of all claim elements, even if amended later during prosecution.
Number | Name | Date | Kind |
---|---|---|---|
6650582 | Matsumoto et al. | Nov 2003 | B2 |
7088627 | Bajwa et al. | Aug 2006 | B1 |
20080056044 | Tomita | Mar 2008 | A1 |
20080291763 | Mori et al. | Nov 2008 | A1 |