Method and apparatus for conducting input/output loop back tests using a local pattern generator and delay elements

[0002] The present invention relates to testing the correct operation of integrated circuits; more particularly, the present invention relates to testing an input/output circuit in an integrated circuit.

[0003] Before an integrated circuit (IC) may be used in an application, it is typically necessary to verify the proper functionality and timing of components within each input/output (I/O) circuit of the IC. Verifying the functionality of an IC is typically accomplished by placing the IC on a tester that includes a tester channel for each I/O pin on the IC. Subsequently, each I/O buffer coupled to an I/O pin is tested for functionality, timing, performance, etc. However, there are often problems associated with testing an IC in this manner. One problem is that testing each I/O pin on an IC is time consuming and often expensive due to test equipment costs. Another problem is that the speed of the test equipment is typically not fast enough to keep pace with the IC. Moreover, existing test equipment is not capable of testing high-speed source synchronous systems.

[0004] In order to solve the problems of exclusively using a tester to test an IC, several contemporary testing schemes use the IC to assist in the testing. This process is commonly referred to as an I/O loopback test. I/O loopback tests are typically carried out by providing data from a functional logic block (or FLB) within the IC (e.g., a microprocessor), and driving the data out through the output component of each I/O buffer. Subsequently, the data is driven back through the input component of the I/O buffer to the FLB in order to verify to that the correct data has been received. Consequently, the IC verifies whether the input and output components of each I/O buffer is functioning properly.

[0005] The problem with conventional I/O loopback tests, however, is that generating and verifying test data patterns at the FLB require the addition of special logic within the FLB. Further, since the FLB is required to generate and check data, it is not possible to conduct other tests within the IC (e.g., the effect of I/O data on the FLB or vice versa). Therefore, a method and apparatus for performing an I/O loopback test without using core logic within a FLB is desired.

[0006] An integrated circuit including a plurality of input/output (I/O) buffers is disclosed. In one embodiment, the integrated circuit contains a first I/O buffer that includes a first I/O pad, a first output driver coupled to the first I/O pad and a first input receiver coupled to the first I/O pad. In addition, the first I/O buffer includes a first I/O test circuit coupled to the first output driver and the first input receiver, wherein the first I/O test circuit generates test pattern signals whenever the first I/O test circuit is operating in a test mode.

[0007] The present invention will be understood more fully from the detailed description given below and from the accompanying drawings of various embodiments of the invention. The drawings, however, should not be taken to limit the invention to the specific embodiments, but are for explanation and understanding only.

[0008] FIG. 1 is a block diagram of one embodiment of an input/output buffer for common clock operation;

[0009] FIG. 2 is a block diagram of one embodiment of an input/output test circuit;

[0010] FIG. 3 is a block diagram of one embodiment of an input/output buffer for source synchronous operation;

[0011] FIG. 4 is a flow control diagram for one embodiment of the operation of a loopback test at an input/output buffer; and

[0012] FIG. 5 is a block diagram of one embodiment of an integrated circuit including a multitude of I/O buffers.

[0013] A method and apparatus for performing I/O loopback tests wherein I/O circuitry tests itself independent of a system processor is described. In the following detailed description of the present invention numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be apparent to one skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form, rather than in detail, in order not to unnecessarily obscure the present invention.

[0014] FIG. 1 is a block diagram of one embodiment of an input/output (I/O) buffer 100. I/O buffer 100 includes an I/O test circuit 110, a multiplexer (MUX) 115, and a register 120. Further, I/O buffer 100 includes pre-driver circuit 125, an output driver 130, an I/O pad 135, a voltage reference pad (V_REF) 140, an operational amplifier 145 and a register 150. According to one embodiment, I/O buffer 100 operates in a common clock mode.

[0015] I/O test circuit 110 generates test signals for the testing of other components within I/O buffer 100. Additionally, I/O test circuit 110 receives the test signals after the signals have propagated through I/O buffer 100 and compares them to the data that was initially transmitted. MUX 115 is coupled to I/O test circuit 110 and selects between test signals transmitted from I/O test circuit 110 and data received at I/O buffer 100 from the core of an integrated circuit (CORE). MUX 115 selects signals from I/O test circuit 110 whenever I/O buffer 110 is operating in a loopback or switching state (AC) loopback test mode. The CORE data path is selected whenever I/O buffer 110 is operating in a normal mode. One of ordinary skill in the art will appreciate that other selection circuits may be used to select between the CORE and test data paths, and that this MUX may be placed in other positions in the data path.

[0016] Register 120 is coupled to MUX 115 and captures data that is received from I/O test circuit 110 or the CORE in order to synchronize the received data with the system clock. Pre-driver 125 is coupled to register 120. Pre-driver 125 transmits data signals that are to be transmitted from I/O buffer 100. Driver 130 is coupled to pre-driver 125 and operates by amplifying data signals received from pre-driver 125 before they are transmitted from I/O buffer 100. I/O pad 135 is coupled to driver 130. In normal operation, I/O pad 135 receives data from other circuits coupled to I/O buffer 100 and transmits data from I/O buffer 100 to the other circuits.

[0017] V_REFpad 140 receives an external reference voltage. Differential amplifier (amp) 145 is coupled to I/O pad 135 and V_REFpad 140. Amp 145 aggregates signals received from I/O pad 135 and V_REFpad 140 into a single signal. According to one embodiment, amp 145 transmits a logical one whenever a signal received at I/O pad 135 is higher in magnitude than a signal received at V_REFpad 140. Additionally, amp 145 transmits a logical zero whenever a signal received at I/O pad 135 is lower in magnitude than a signal received at V_REFpad 140. One of ordinary skill in the art will appreciate that the operation of amp 145 may be reversed.

[0018] Register 150 is coupled to Amp 145 and captures data that is received from I/O pad 135 in order to synchronize the received data with the system core clock. Amp 145 and register 150 are typically referred to as the input (or receiver) portion of I/O buffer 100. In another embodiment, Amp 145 and register 150 may be combined to form a latched sense amp. However, one of ordinary skill in the art will appreciate that other input structures, such as a simple CMOS gate, may be used.

[0019] FIG. 2 is a block diagram of one embodiment of I/O test circuit 110. I/O test circuit 110 includes a delay unit 203, a MUX 205, a test pattern generator 210, a stage unit 215, a compare unit 220, a MUX 225 and a latch 230. Delay unit 203 provides a delay of core clock signals received at I/O test circuit 110 for operation in the AC loopback testing mode. The delayed clock signals are subsequently transmitted to test pattern generator 210. Additionally, delay unit 203 is programmable in order to vary the delay of the core clock signals. According to one embodiment, delay unit 203 receives one or more delay select signals that indicate the magnitude delay unit 203 is to delay the clock signals.

[0020] MUX 205 is coupled to delay unit 203. MUX 205 receives an AC select signal that enables MUX 205 to select between the core clock signals and the delayed clock signals. During an AC loopback test, the select signal indicates that MUX 205 select the delayed clock signals to drive test pattern generator 210. However, during non-test conditions, or during a standard loopback test, the select signal indicates that MUX 205 select core clock signals. According to another embodiment, delay unit 203 and MUX 205 may be external to I/O test circuit 110. Moreover, one of ordinary skill in the art will appreciate that delay unit 203 and MUX 205 may be placed in other locations within I/O test circuit 110. For example, delay unit 203 and MUX 205 may be placed between MUX 225 and latch 230 in order to delay the propagation of data with respect to clock signals.

[0021] Test pattern generator 210 is coupled to MUX 205 and MUX 115, and is used to generate test pattern signals for testing I/O test circuit 100 upon the initiation of a loopback test. According to one embodiment, test pattern generator 210 may be implemented with two or more flip flops. However, test pattern generator 210 may be implemented using other circuitry such as a linear feedback register, a random pattern generator or random access memory (RAM). According to a further embodiment, the test pattern signals may be loaded into test pattern generator 210 from an integrated circuit tester (not shown) via a test chain prior to conducting a loopback test.

[0022] Stage unit 215 is coupled between test pattern generator 210 and compare unit 220 and receives test pattern signals from test pattern generator 210. According to one embodiment, stage unit 215 provides a one cycle delay for the test pattern signals before they are transmitted to compare unit 220 be compared with test signals received from amp 145. However, it will be appreciated that stage unit 215 may provide a higher magnitude of delay.

[0023] Compare unit 220 is coupled to stage unit 215 and amp 145. Compare unit 220 compares test signals received from stage unit 215 with test signals received from amp 145, after having passed through the components of I/O buffer 100. If compare unit 220 detects a difference between the signals received from amp 145 and those received from test pattern generator 210, an error signal is transmitted from compare unit 220. According to one embodiment, compare unit 220 may be implemented using an Exclusive-Or-Gate. Alternatively, other comparison logic may be used to implement compare unit 220.

[0024] MUX 225 is coupled to compare unit 220 and latch 230. MUX 225 selects signals from compare unit 220 whenever a loopback test is in progress at I/O buffer 110, and selects the scan chain path after the test pattern is complete. According to one embodiment, test pattern generator 210 generates an infinite repeating pattern. Nevertheless, test pattern generator 210 may generate patterns according to other methods.

[0025] Latch 230 is coupled to compare unit 220 and receives the error signal transmitted by compare unit 220 in cases where a difference is detected between the amp 145 signals and the signals received from stage unit 215. According to one embodiment, latch 230 is a boundary scan latch as described in the Institute of Electrical and Electronics Engineers (IEEE) 1149.1 Specification. Once an error signal is received, latch 230 remains set until the loopback test is completed. Latch 230 is subsequently examined as a part of a boundary scan chain.

[0026] FIG. 5 is a block diagram of one embodiment of an integrated circuit (IC) 500. IC 500 includes input/output (I/O) buffers 100(1)-100(n). I/O buffers 100(1)-100(n) make up a data block of I/O circuitry for transmitting to and receiving data from other IC 100 devices. According to one embodiment, a data block includes sixteen (16) I/O buffers 200. However, in other embodiments, a data block may include other multiples (e.g., 2, 4, 8, 12, 18, 32, 40, 64, etc.) of I/O buffers 100.

[0027] According to one embodiment, the data block of I/O buffers 100 operate as a source synchronous interface. Core clock and select signals are distributed to each I/O buffer 100 in a data block. In addition, a strobe clock is generated at I/O buffer 100(1) and distributed along with I/O data. According to one embodiment, the strobe clock generated at I/O buffer 100(1) is distributed to the remaining I/O buffers 100 (e.g., 100(2)-100(n)) in the data block during an AC I/O loopback test mode. According to one embodiment, I/O buffer 100(1) is placed in the center of the data group to minimize the routing length. Nevertheless, one of ordinary skill in the art will appreciate that I/O buffer 100(1) can be placed at any physical location.

[0028] During the normal mode of operation, the strobe clock is used for source synchronous data transactions. Source synchronous refers to interfaces wherein a receiving I/O buffer captures data based upon a strobe clock that is provided by another IC device driving the data. With careful control over the signal paths, the transfer rate of information can be maximized because the strobe edge can be accurately placed to minimize the skew relative to the data.

[0029] FIG. 3 is a block diagram of one embodiment of I/O buffer 100 in a source synchronous interface. In this embodiment, I/O buffer 100 includes all of the components described above with respect to FIGS. 1 and 2 designated as 100(x), with the addition of a patch march unit 370. I/O buffer 100(x) represents any of I/O buffers 100(2)-100(n) in the data group described above with respect to FIG. 5. I/O buffer 100(1) includes a strobe pattern generator 310, a register 320, a delay unit 322, a MUX 324, a pre-driver 325, a driver 330, a strobe pad 335, a strobe distribution unit 365.

[0030] Strobe pattern generator 310 generates strobe pulses for source synchronous operation. Register 320 is coupled to strobe pattern generator 310. Register 320 stores data that is received from strobe pattern generator 310 in order to synchronize the strobe pulses with a strobe clock. Delay unit 322 provides a delay of strobe pulses generated at strobe pattern generator 310 for operation in the AC loopback testing mode. The delayed strobe low pulses are subsequently transmitted from I/O buffer 100(1). Additionally, delay unit 322 is programmable in order to vary the delay. According to one embodiment, delay unit 322 receives one or more delay select signals that indicate the magnitude delay unit 322 is to delay the clock signals.

[0031] MUX 324 is coupled to delay unit 322. MUX 324 receives an AC select signal that enables MUX 324 to select between the strobe pulses and the delayed pulses. During an AC loopback test, the select signal indicates that MUX 324 select the delayed pulses. However, during non-test conditions, or during a standard loopback test, the select signal indicates that MUX 324 select the strobe pulses. According to another embodiment, delay unit 322 and MUX 324 may be placed in other locations within I/O buffer 100(1). Pre-driver 325 is coupled to MUX 324. Pre-driver 325 transmits strobe signals that are to be transmitted from I/O buffer 100. Driver 330 is coupled to pre-driver 325 and operates by amplifying data signals received from pre-driver 325.

[0032] Strobe pad 335 is coupled to driver 330. In normal operation, strobe pad 335 transmits strobe pulses from I/O buffer 100 to other integrated circuit components. Strobe pad 335 also receives strobe pulses at I/O buffer 100 from other components. Strobe distribution unit 365 is coupled to driver 330 and distributes the test strobe pulses the other I/O buffers in the data block (e.g., buffers 100(2)-100(n). Strobe distribution unit 365 may also distributes strobe pulses received during the normal source synchronous receive operation.

[0033] Path match unit 370 in buffer 100(x) is coupled to driver amp 145, strobe distribution unit 365 and compare unit 220. Path match unit 370 clocks data received from amp 145 into compare unit 220 using the test strobe pulses. In addition, path match unit 370 matches the delay provided by strobe distribution unit 365 in order to prevent a relative change in position between the test signals and the test strobe pulses.

[0034] FIG. 4 is a flow diagram for one embodiment of the operation of I/O buffer 100 while conducting a standard loopback test. At process block 405, test patterns are loaded into test pattern generator 210. At process block 410, a test signal is transmitted from test pattern generator 210 to MUX 115. Additionally, the test signal is transmitted to stage unit 215. Since a loopback test is being conducted at I/O buffer 100, MUX 115 selects the test data path, thus, enabling data to be transmitted from I/O test circuit 110 to the components of I/O buffer 100. The test signal received at MUX 115 is subsequently received at driver 130 after having passed through register 120 and pre-driver 125.

[0035] Further, a test strobe pulse may be transmitted from strobe pattern generator 310 through register 320 and pre-driver 325 to driver 330 if I/O buffer 100 is a source synchronous buffer. At process block 415, the test signal is transmitted from driver 130 to I/O pad 135 and amp 145. Also, the test strobe pulse may be transmitted from driver 330 to strobe output 335 and strobe distribution unit 365 in the source synchronous mode. At process block 420, the test signal that was transmitted to stage unit 215 (duplicate signal) is transmitted from stage unit 215 to compare unit 220. At process block 425, the test signal is transmitted from amp 145 to compare unit 220.

[0036] In the source synchronous mode, the test signal is transmitted from amp 145 to path match unit 370 and the test strobe pulse is transmitted from strobe distribution unit 365 to path match unit 370. Subsequently, the test signal is transmitted from path match unit 370 to compare unit 220. The duplicate signal received from stage unit 215 is compared with the test signal received from amp 145, or path match unit 370. At process block 430, it is determined whether the duplicate signal matches the test signal received from amp 145.

[0037] If it is determined that there is a mismatch between the two signals, a defect exists in I/O buffer 100 and an error signal is transmitted from compare unit 220 to latch 230, process block 435. At process block 440, it is determined whether the loopback test has been completed. After a sufficient number of test cycles have passed, latch 230 may be examined to determine whether a defect has been detected in I/O buffer 100, process block 445. If a sufficient number of test cycles have not passed, control is returned to process block 410 wherein a subsequent test signal is transmitted from test pattern generator 210 to MUX 115.

[0038] If it is determined that the test is complete, it is determined whether the loopback test has been completed, process block 440. If all of the test pattern signals have been issued (e.g., a sufficient number of test cycles have been run), latch 230 may be examined to determine whether a defect was detected in I/O buffer 100 during the transmission of previous test signals, process block 445. However, if all of the test pattern signals have not been issued (e.g., a sufficient number of test cycles have not been run), control is returned to process block 405 wherein a subsequent test signal is transmitted from test pattern generator 210.

[0039] One of ordinary skill in the art will recognize that process blocks 415 and 425 may be processed in a variety of different sequences. For example, the process disclosed in process block 425 may be executed before the process in process block 415. Alternatively, process blocks 415 and 425 may be executed in parallel.

[0040] According to one embodiment, I/O buffers 100(1) and 100(n) support AC I/O loopback testing. Although I/O buffer 100 may be determined to be functional after a standard I/O loopback test, an AC loopback test provides the capability of detecting more subtle defects in the components of I/O buffer 100 that may effect timing. Thus, conducting an AC I/O loopback test examines the AC I/O loopback of the input and output paths of I/O buffer 100(x).

[0041] According to one embodiment, the AC loopback test includes a first fail test, an all fail test and a difference test. The first fail test determines whether the path delay of the I/O buffers 100(x) meet the designed time valid after (Tva) output specification and hold time (Th), input specification. According to one embodiment, the first fail test is implemented by delaying the strobe pulses generated at strobe pattern generator 310. As described above, the delay is implemented by programming delay unit 322) until a first I/O buffer 100(x) of a particular data group (e.g., 100(2)-100(n)) fails. First a predetermined delay is programmed into delay unit 322. Subsequently, test data generated at test pattern generator 210 is received at register 320. The test data is propagated through the output path back to amp 145 where it is to be latched into path match unit 370.

[0042] If the delay of the test data through the I/O paths to path match unit 370 exceeds I/O buffer 100(x) specifications the test data will not be latched into path match unit 370. Accordingly, the data will not be received at latch 230. A particular I/O buffer 100(x) will be considered to have failed the test if the expected test data is not received at latch 230. The delay at which the first I/O buffer 100(x) fails is then compared to a criteria that is based on the Tva+Th specifications for the particular I/O buffer 100(x). If the delay time at which the buffer 100(x) failed is less than the criteria, the buffer 100(x) and IC 500 is considered to be defective.

[0043] The all fail test determines whether the path delay of the I/O buffers 100(x) is so large that it exceeds the receiver's setup time (Tsu) specification or the time valid before (Tvb) output specification for a subsequent data cycle. According to one embodiment, the all fail test is implemented by delaying the strobe pulses generated at output strobe generator 320 until all I/O buffers 100(x) of a data group fail. The point at which all of the buffers 100(x) fail, in effect, indicates the time at which the last I/O buffer 100(x) failed. Again, the time at which the last I/O buffer 100(x) fails is compared to a criteria that is based on the design specifications (Tvb +Tsu). If the delay time is greater than the criteria, the time at which the last buffer 100(x) has completed a data transaction will interfere with the setup time for the ensuing transaction. As a result, the buffer 100(x) and IC 500 is considered to be defective.

[0044] The difference test is applied to measure the difference in delay between the first fail test and the all fail test. The window of time between the first fail test and the all fail test must also meet design specifications for the data group as failing this could indicate the presence of defect that may go undetected using only the first fail and all fail tests detailed above.

[0045] A common clock AC loopback test at I/O buffer 100 described in FIG. 1, is conducted by first programming the fixed delay into delay unit 203 such that all outputs of I/O buffer 100 fail the test (all fail test). In this embodiment, latch 230 may be configured to capture multiple test signals in order to determine whether an error has been detected for each test signal generated at test pattern generator 210. Subsequently, the delay is reduced to the point at which the first output passes from I/O buffer 100 (first fail test). Afterward, the delay is further reduced until all of the outputs pass. The range of delay between the first fail test and the all fail test discloses the relative health of the fastest and slowest buffers in I/O buffer 100.

[0046] Referring back to FIG. 5, is a block diagram of one embodiment of an integrated circuit 500 including a multitude of I/O buffers 100. According to one embodiment, a loopback test is conducted at I/O buffers 100(1)-100(n) simultaneously. After all of the loopback tests have been completed, latch 230 of each I/O buffer may be examined to determine whether an error has been detected in any of the buffers. Alternatively, all of the I/O buffers 100 may be tested sequentially.

[0047] Whereas many alterations and modifications of the present invention will no doubt become apparent to a person of ordinary skill in the art after having read the foregoing description, it is to be understood that any particular embodiment shown and described by way of illustration is in no way intended to be considered limiting. Therefore, references to details of various embodiments are not intended to limit the scope of the claims which in themselves recite only those features regarded as the invention.

[0048] Therefore, a method and apparatus for performing an I/O loopback test has been described.