DIE STACK TEST ARCHITECTURE AND METHOD

Abstract

A test control port (TCP) includes a state machine SM, an instruction register IR, data registers DRs, a gating circuit and a TDO MX. The SM inputs TCI signals and outputs control signals to the IR and to the DR. During instruction or data scans, the IR or DRs are enabled to input data from TDI and output data to the TDO MX and the top surface TDO signal. The bottom surface TCI inputs may be coupled to the top surface TCO signals via the gating circuit. The top surface TDI signal may be coupled to the bottom surface TDO signal via TDO MX. This allows concatenating or daisy-chaining the IR and DR of a TCP of a lower die with an IR and DR of a TCP of a die stacked on top of the lower die.

Description

FIELD OF DISCLOSURE

This disclosure relates to die that are designed to be used in a 3D die stack and in particular to a common test architecture designed into each die of the 3D die stack for accessing and testing digital circuits of each die using parallel scan techniques.

BACKGROUND OF THE DISCLOSURE

Die manufactured for use in a die stack must be tested at the wafer level and then potentially retested after the die are singulated to ensure only known good die are used in the stack. Each time an upper die is stacked on top of a lower die that has been tested, the upper die needs to be retested to ensure it has not been damaged during the stacking process. Also, the interconnect between the lower and upper die needs to be tested and determined good. The interconnect between die in a 3D stack is provided by Through Silicon Vias (TSVs), which are vertical signaling paths between the bottom and top surfaces of each die. This testing process is repeated for each additional upper die assembled onto the stack. It is therefore advantageous to have a common test architecture in each die and a common method of accessing the test architecture in each die, independent of the stacked location of the die.

BRIEF SUMMARY OF THE DISCLOSURE

The disclosure provides a test architecture for providing access to and control of parallel scan paths within a single die or to parallel scan paths in dies that are stacked.

DESCRIPTIONS OF THE VIEWS OF THE DISCLOSURE

FIG. 1 illustrates a die containing the test architecture of the disclosure.

FIG. 2 illustrates a conventional parallel scan path arrangement that can be accessed and controlled by the test architecture of the disclosure.

FIG. 3 illustrates a conventional parallel scan compression arrangement that can be accessed and controlled by the test architecture of the disclosure.

FIG. 4 illustrates a test circuit of the disclosure for either outputting parallel test output data to a tester or comparing the parallel test output data to data input from a tester.

FIG. 5 illustrates a first example implementation of the comparator of FIG. 4 of the disclosure.

FIG. 6 illustrates a second example implementation of the comparator of FIG. 4 of the disclosure.

FIG. 7 illustrates a first example implementation of the test control port (TCP) of the disclosure.

FIG. 8 illustrates a second example implementation of the TCP of the disclosure.

FIG. 9 illustrates the test architecture in a die operating to input parallel test input data to a parallel scan test circuit in the die and outputting parallel test output data from the parallel test circuit, according to the disclosure.

FIG. 10 illustrates the test architecture operating to input parallel test input data to a parallel test circuit and comparing the parallel test output data with data input to the test circuit, according to the disclosure.

FIG. 11 illustrates a stack die arrangement where the lower die is tested as described in FIG. 9, according to the disclosure.

FIG. 12 illustrates a stack die arrangement where the lower die is tested as described in FIG. 10, according to the disclosure.

FIG. 13 illustrates a stack die arrangement where the upper die is tested as described in FIG. 9, via the lower die, according to the disclosure.

FIG. 14 illustrates a stack die arrangement where the upper die is tested as described in FIG. 10, via the lower die, according to the disclosure.

FIG. 15 illustrates a stack die arrangement where the parallel scan circuits of the lower and upper die are daisy-chained together and tested as described in FIG. 9, according to the disclosure.

FIG. 16 illustrates a stack die arrangement where the parallel scan circuits of the lower and upper die are daisy-chained together and tested as described in FIG. 10, according to the disclosure.

FIG. 17 illustrates a die containing multiple selectable parallel scan circuits that may be tested as described in FIG. 9 or FIG. 10, according to the disclosure.

FIG. 18 illustrates the die of FIG. 17 further illustrating the TCP of the disclosure for clarity.

FIG. 19 illustrates an alternate implementation of the TCP of FIG. 7.

FIG. 20 illustrates the TCP of FIG. 19 were the state machine is an IEEE standard 1149.1 Tap state machine.

FIG. 21 illustrates an alternate implementation of the TCP of FIG. 8.

FIG. 22 illustrates the TCP of FIG. 21 were the Test Control Inputs (TCI) inputs include IEEE standard 1500 signals.

FIG. 23 illustrates a die including the test architecture of the disclosure where the Parallel Test Input (PTI) and Parallel Test Inputs and Outputs (PTIO) are shared for both test and functional signaling.

FIG. 24 illustrates the die of FIG. 23 where a multiplexer is substituted for buffers “a” and “b”.

DETAILED DESCRIPTION OF THE DISCLOSURE

FIG. 1 illustrates a die 100 including the test architecture of the disclosure. The bottom surface of the die includes signal contact points (micro bumps or metal pads) for a Parallel Test Input (PTI) bus 102, a Test Data input (TDI) 104, Test Control Inputs (TCI) 106, a Test Data Output 108 and a Parallel Test Input/Output (PTIO) bus 110. The top surface of the die includes signal contact points for a Parallel Test Output (PTO) bus 112, TDO 114, Test Control Outputs (TCO) 116, TDI 118 and a PTIO bus 120. The architecture includes a Test Control Port (TCP) 122, a Test Circuit (TC) 124, at least one Parallel Scan Circuit 126 to be tested, a first signal coupling means illustrated in this example as signal buffers “a” 128 and a second signal coupling means illustrated in this example as signal buffers “b” 130. TCP is connected to the TDI, TCI and TDO bottom surface signals and the TDO, TCO and TDI top surface signals. TCP includes a control (CTL) output bus 131 that is connected to the buffers, Parallel Scan Circuit and TC.

The Parallel Scan Circuit inputs PTI signals from the PTI bus 102 via TSV signal connections 132 and outputs PTO signals to buffers “b” and to the TC. The TC is connected to the PTIO bus TSV signal connections 134 for outputting data onto the PTIO 110 bus or inputting data from the from the PTIO bus 120. The outputs of buffers “a” and “b” are connected together and to the top surface PTO bus 112 via TSV signal connections 136. If the top surface PTO bus 112 is to be driven from the bottom PTO bus, buffers “a” are enabled by an enable 3 (E3) signal from the TCP CTL bus 131 and buffers “b” are disabled by enable 4 (E4) from the TCP CTL bus 131. If the top surface PTO bus 112 is to be driven by the PTO output 138 of the Parallel Scan Circuit, buffers “b” are enabled by E4 and buffers “a” are disabled by E3.

While the first and second signal coupling means are shown to be buffers “a” 128 and “b” 130, the first and second coupling means could be realized using other types of circuitry such as but not limited to a multiplexer or transistor switches to allow the PTO bus 112 to be driven by either the PTI bus 102 or by the PTO bus 138 output from the Parallel Scan Circuit.

FIG. 2 illustrates one example implementation of the Parallel Scan Circuit 126 of FIG. 1. In this example, scan paths 202 input test data from PTI and output test data to PTO. The scan paths are controlled by the CTL inputs from the TCP.

FIG. 3 illustrates another example implementation of the Parallel Scan Circuit 126 of FIG. 1. In this example, a decompressor (D) 302 inputs compressed test data from PTI and outputs decompressed test data to parallel scan paths 304, and a compactor (C) 306 inputs test data from the parallel scan paths and outputs compacted data on PTO. The decompressor, scan paths and compactor are controlled by the CTL inputs from the TCP.

While not shown in FIGS. 2 and 3, system clock inputs and associated circuitry may be used in conjunction with the TCP CTL signals to improve the at-speed testing of the circuits, such as well known clock leaker system clock circuits.

FIG. 4 illustrates an example implementation of the TC 124 of FIG. 1. In a first mode of operation, the TC inputs PTO data 138 from the Parallel Scan Circuit and outputs it to the PTIO bus 134, via 3-state buffers 4021-N. In a second mode of operation, the TC inputs PTO data from the Parallel Scan Circuit and compares it to data input from the PTIO bus, using comparators (C) 4041-N. The modes are controlled by enable (E1 and E2) inputs from the CTL bus of the TCP. The comparators (C) have fail 4051-N outputs that are set whenever a comparison failure occurs. The fail outputs are input to a gating circuit 406 that can detect when one or more fail outputs are set to indicate a failure. The output of the gating circuit passes through a 3-state buffer 408 to a Compare Fail Output (CFO) 410 that is connected to a signal path in the PTIO bus. The second mode of operation is advantageous at wafer probe since it enables multiple dies to be tested using a common PTI and PTIO bus from a tester to input test data to multiple die commonly connected to the PTI and PTIO buses.

FIG. 5 illustrates a first example comparator (C) 404 of FIG. 4 for comparing PTO data to expected (EXP) data from the PTIO bus.

FIG. 6 illustrates a second example maskable comparator (C) 404 of FIG. 4 for comparing PTO data to expected (EXP) data from the PTIO bus or masking the compare operation off using mask (MSK) data input from the PTIO bus. When masked, the Fail signal output does not output a failure indication when a comparison fails.

FIG. 7 illustrates a first example implementation of the TCP 122 of FIG. 1 which includes a state machine (SM) 702, an instruction register (IR) 704, data registers (DRs) 706, a gating circuit 708 and a TDO multiplexer (MX) 710. The SM inputs TCI signals 714 and outputs instruction register control (IRC) signals to the IR and data register control (DRC) signals to the DR. During instruction scans, the IR is enabled by IRC inputs to input data from TDI 712 and output data to the TDO multiplexer 710 and the top surface TDO signal 718. During data scans, a DR is enabled by the DRC inputs to input data from TDI 712 and output data to the TDO multiplexer 710 and the top surface TDO signal 718.

The bottom surface TCI inputs 714 may be coupled to the top surface TCO signals 720 via the gating circuit 708. The top surface TDI signal 722 may be coupled to the bottom surface TDO signal 716 via TDO multiplexer 710. This allows concatenating or daisy-chaining the IR and DR of a TCP of a lower die with an IR and DR of a TCP of a die stacked on top of the lower die. The instruction register outputs (IRO) include an enable (ENA) signal for the gating circuit 708 and a second select (SEL) signal for TDO multiplexer 710. The CTL bus output from the TCP includes IRO signals and DRC signals. The gating circuit 708 may gate one, more than one or all of the TCI signals. Non-gated TCI signals are coupled to appropriate top surface TCO signals, as shown in dotted line. The SM, IR, and DRs of this implementation could be the TAP SM, IR and DRs as defined in the IEEE 1149.1 boundary scan standard.

FIG. 8 illustrates a second example implementation 800 of the TCP 122 of FIG. 1. This implementation is identical to the FIG. 7 implementation with the exception that it does not include a SM 702. The TCI inputs 802 are coupled to the IRC inputs of IR 704 and DRC inputs of DR 706. During instruction scans, the IR is enabled by the IRC signals from the TCI inputs 802 to input data from TDI 712 and output data to the TDO multiplexer 710 and the top surface TDO signal 718. During data scans, a DR is enabled by DRC signals from the TCI inputs 802 to input data from TDI 712 and output data to the TDO multiplexer 710 and the top surface TDO signal 718.

The bottom surface TCI inputs 802 may be coupled to the top surface TCO signals 720 via the gating circuit 708. The top surface TDI signal 722 may be coupled to the bottom surface TDO signal 716 via the TDO multiplexer 710. This allows concatenating or daisy-chaining the IR and DR of a TCP 122 of a lower die with an IR and DR of a TCP 122 of a die stacked on top of the lower die. The instruction register outputs (IRO) include an enable (ENA) signal for the gating 708 circuit and the SEL signal to TDO multiplexer 710. The CTL bus output 724 from the TCP includes IRO signals and DRC signals. The gating circuit 708 may gate one, more than one or all of the TCI signals. Non-gated TCI signals are coupled to appropriate top surface TCO signals, as shown in dotted line. The TCI input bus 802, IR 704 and DRs 706 of this implementation could be the control inputs, IR and DRs as defined in IEEE 1500 core wrapper test standard.

FIG. 9 illustrates the die test architecture of FIG. 1 when an instruction has been loaded into the TCP's IR to enable the Parallel Scan Circuit to be tested by inputting PTI data from the PTI bus 102 and outputting PTO data to the PTIO bus 110, as seen in darkened line. The CTL outputs 131 from the TCP 122 controls the capture and input and output shift operations of Parallel Scan Circuit 126 during test.

FIG. 10 illustrates the die test architecture of FIG. 1 when an instruction has been loaded into the TCP's IR to enable the Parallel Scan Circuit 126 to be tested by inputting PTI data from the PTI bus 102 and comparing the PTO output data from the Parallel Scan Path 126 with data input from the PTIO bus 110, as seen in darkened line. The CTL outputs from the TCP 122 controls the capture and input and output shift operations of Parallel Scan Circuit 126 during test. During this test, the CFO output 410 (dotted line) of the TC 124 of FIG. 4 is enabled to output comparison failures on the PTIO bus 110.

FIG. 11 illustrates a stack 1100 including a bottom die 100 and a top die 1102, both including the die test architecture of FIG. 1. In this example, the Parallel Scan Circuit 126 of the bottom die 100 is being tested as described in FIG. 9. The IR of the TCP 1122 of the top die has been loaded with an instruction which place the top die in a quiescent mode that ignores the testing of the bottom die, does not interfere with the testing of the bottom die and disables the tri-state buffers of TC 1124 of the top die. The TCO outputs 116 from the bottom die to the TCI inputs 1106 of the top die will typically be gated off during testing of the bottom die.

FIG. 12 illustrates a stack 1100 including a bottom die 100 and a top die 1102, both including the die test architecture of FIG. 1. In this example, the Parallel Scan Circuit 126 of the bottom die 100 is being tested as described in FIG. 10. The IR of the TCP 1122 of the top die 1102 has been loaded with an instruction which places the top die in a quiescent mode that ignores the testing of the bottom die, does not interfere with the testing of the bottom die and disables the tri-state buffers of TC 1124 of the top die. The TCO outputs 116 from the bottom die to the TCI inputs 1106 of the top die will typically be gated off during testing of the bottom die.

FIG. 13 illustrates a stack 1100 including a bottom die 100 and a top die 1102, both including the die test architecture of FIG. 1. In this example, the Parallel Scan Circuit 1126 of the top die 1102 is being tested as described in FIG. 9. The IR of the TCP 122 of the bottom die 100 has been loaded with an instruction which place the bottom die 100 in a quiescent mode that ignores the testing of the top die 1102, does not interfere with the testing of the top die, disables the tri-state buffers of TC 124 of the bottom die and enables buffers “a” 128 to pass the PTI signals up to the top die. The TCO outputs from the bottom die 116 to the top die 1106 will be gated on during testing of the top die to control the top die's TCP 1122.

FIG. 14 illustrates a stack 1100 including a bottom die 100 and a top die 1102, both including the die test architecture of FIG. 1. In this example, the Parallel Scan Circuit 1126 of the top die 1102 is being tested as described in FIG. 10. The IR of the TCP 122 of the bottom die 100 has been loaded with an instruction which place the bottom die in a quiescent mode that ignores the testing of the top die, does not interfere with the testing of the top die, disables the tri-state buffers of TC 124 of the bottom die and enables buffers “a” 128 to pass the PTI signals up to the top die. The TCO outputs from the bottom die 116 to the top die 1106 will be gated on during testing of the top die to control the top die's TCP 1122.

FIG. 15 illustrates a stack 1100 including a bottom die 100 and a top die 1102, both including the die test architecture of FIG. 1. In this example, instructions have been loaded into the IRs of the TCPs 122, 1122 of the bottom and top die to daisy-chain the Parallel Scan Circuits 126, 1126 of the bottom and top die together to form a PTI to PTO path through both Parallel Scan Circuits, from the PTI 102 of the bottom die to the PTIO 110 of the bottom die. The instruction loaded in the bottom die TCP 122 enables buffers “b” 130 to drive the PTI inputs to the top die, disables the tri-state buffers of TC 124 of the bottom die and gates on the TCO outputs 116 to the TCI inputs 1116 of the top die. The instruction in the TCP of the top die enables the tri-state buffers of TC 1124 of the top die to drive the PTIO bus 1134, 134, 110.

FIG. 16 illustrates a stack 1100 including a bottom die 100 and a top die 1102, both including the die test architecture of FIG. 1. In this example, instructions have been loaded into the IRs of the TCPs 122, 1122 of the bottom and top die to daisy-chain the Parallel Scan Circuits 126, 1126 of the bottom and top die together to form a PTI to PTO path through both Parallel Scan Circuits, from the PTI 102 of the bottom die to the PTO inputs to TC 1124 of the top die. The instruction loaded in the bottom die TCP 122 enables buffers “b” 130 to drive the PTI inputs to the top die, disables the tri-state buffers of TC 124 of the bottom die and gates on the TCO 116 outputs to the TCI inputs 1106 of the top die. The instruction in the TCP 1122 of the top die disables the tri-state buffers of TC 1124 of the top die so that data from PTIO 110 can be input to the comparators of TC 1124 to compare against the PTO data being input to the comparator from the Parallel Scan Circuit 1126 of the top die.

While the stack die examples of FIGS. 11-16 show only two dies including the test architecture of the disclosure in the stack, the stack could contain any number of stacked dies including the test architecture of the disclosure.

While in the preceding Figure examples, only one Parallel Scan Circuit 126 was shown in a die, any number of Parallel Scan Circuits may be included in a die and tested using the test architecture of the disclosure. FIG. 17 below shows one preferred way of accessing plural Parallel Scan Circuits in a die.

FIG. 17 illustrates multiple Parallel Scan Circuits 126a, 126b in a die 1700 outputting their PTO outputs to a multiplexer 1702. The PTO outputs of the Parallel Scan Circuit to be tested are selected by the multiplexer 1702 to be output on the PTO bus to the TC 124 and to the inputs of buffers “b” 130. Multiplexer 1702 is controlled by PTO select (PTOSEL) signals from the IRO part of the CTL bus of the TCP 122. The multiplexer 1702 allows multiple Parallel Scan Circuits to share a common TC 124, which reduces test circuit overhead in the die 100.

FIG. 18 is the same as FIG. 17 but it includes the TCP 1822 to more clearly illustrate the TCP CTL outputs being connected buffers “a” 128 and “b” 130, Parallel Scan Circuits 126a, 126b, multiplexer 1702 and TC 124.

FIG. 19 illustrates a third example implementation of the TCP 122 of FIG. 1, similar to the first example implementation of FIG. 7. This third example implementation TCP 1900 includes a state machine (SM) 702, an instruction register (IR) 704, data registers (DRs) 706, a gating circuit 708, first TDO multiplexer (MX) 712 and a second TDO multiplexer (MX) 710. The SM inputs TCI signals and outputs instruction register control (IRC) signals to the IR, data register control (DRC) signals to the DR and a first select (SEL) signal to the first multiplexer 712. During instruction scans, the IR inputs data from TDI and outputs data to the TDO multiplexer 710 and the top surface TDO signal 718, via TDO multiplexer 712. During data scans, a DR inputs data from TDI and outputs data to the TDO multiplexer 710 and the top surface TDO signal 718, via TDO multiplexer 712.

The bottom surface TCI inputs 714 may be coupled to the top surface TCO signals 720 via the gating circuit 708. The top surface TDI signal 722 may be coupled to the bottom surface TDO signal 716 via TDO multiplexer 710. This allows concatenating or daisy-chaining the IR and DR of a TCP of a lower die with an IR and DR of a TCP of a die stacked on top of the lower die. The instruction register outputs (IRO) include an enable (ENA) signal for the gating circuit and a second select (SEL) signal for TDO multiplexer 710. The CTL bus output from the TCP includes IRO signals and DRC signals. The gating circuit 708 may gate one, more than one or all of the TCI signals. Non-gated TCI signals are coupled to appropriate top surface TCO signals, as shown in dotted line. The SM, IR, and DRs of this implementation could be the TAP SM, IR and DRs as defined in the IEEE 1149.1 boundary scan standard.

FIG. 20 illustrates the TCP 2000 of FIG. 19 where the SM 702 is realized as an IEEE standard 1149.1 Tap State Machine (TSM) 2002. The TCI inputs 714 to the TSM 2002 include a test clock (TCK) 2004 and test mode select (TMS) signal 2006. The gating circuit 708 may gate the TMS signal, the TCK signal or both the TMS and TCK signals. Non-gated TCI signals are coupled to appropriate top surface TCO signals, as shown in dotted line.

FIG. 21 illustrates a fourth example implementation 2100 of the TCP 122 of FIG. 1. This implementation is identical to the FIG. 19 implementation with the exception that it does not include a SM 122. The TCI inputs are coupled to the IRC inputs of IR 704, the DRC inputs of DRs 706 and SEL input of TDO multiplexer 712. During instruction scans, the IR is controlled by IRC signals from the TCI inputs to input data from TDI 712 and output data to the TDO multiplexer 710 and the top surface TDO signal 718, via TDO multiplexer 712. During data scans, a DR is controlled by the DRC signals from the TCI inputs 714 to input data from TDI 712 and output data to the TDO multiplexer 710 and the top surface TDO signal 718, via TDO multiplexer 712.

The bottom surface TCI inputs 714 may be coupled to the top surface TCO signals 720 via the gating circuit 708. The top surface TDI signal 722 may be coupled to the bottom surface TDO signal 716 via the TDO multiplexer 710. This allows concatenating or daisy-chaining the IR and DR of a TCP of a lower die with an IR and DR of a TCP of a die stacked on top of the lower die. The instruction register outputs (IRO) include an enable (ENA) signal for the gating circuit and the SEL signal to TDO multiplexer 710. The CTL bus output from the TCP includes IRO signals and DRC signals. The gating circuit 708 may gate one, more than one or all of the TCI signals. Non-gated TCI are coupled to appropriate top surface TCO signals, as shown in dotted line. The TCI input bus 714, IR and DRs of this implementation could be the control inputs, IR and DRs as defined in IEEE 1500 core wrapper test standard.

FIG. 22 illustrates the TCP 2200 of FIG. 21 where the TCI inputs 714 are defined to be the wrapper control input signals defined in IEEE standard 1500, which includes a wrapper clock (WRCK) signal, select wrapper instruction register (SelectWIR) signal, shift wrapper (ShiftWR) signal, capture wrapper (CaptureWR) signal, update wrapper (UpdateWR) signal and reset wrapper (ResetWR) signal. The wrapper instruction register (WIR) 2202 is the same as the IR of FIG. 8, it is just labeled WIR by IEEE 1500. The gating circuit 708 may gate one, more than one or all of the TCI signals. Non-gated TCI signals a coupled to appropriate top surface TCO signals, as shown in dotted line.

FIG. 23 illustrates a die 2302 including the test architecture of the disclosure. Die 2302 differs from die 100 of FIG. 1 as follows; (1) the PTI TSVs 132 are shared between inputting PTI signals 102 to the test architecture and inputting functional data input (FDI) signals 2310 to functional circuitry (FC) 2304 of the die, (2) the PTO TSVs 136 are shared between outputting PTO signals 112 from the test architecture and outputting function data output (FDO) signals 2312 from the FC 2304 and (3) the PTIO TSVs 134 are shared between inputting or outputting PTIO 110, 120 signals to the test architecture and inputting or outputting functional data input/output (FDIO) signals 2314, 2316 to functional circuitry (FC) 2306 of the die. When the die is in functional mode, an instruction in the TCP 122 will disable buffers “a” 128 and “b” 130 and disable the buffers in the TC 124.

In functional mode, FDI can be input to FC 2304 via TSVs 132, FDO can be output from FC 2304 via TSVs 136 and FDIO can be input and output to FC 2006 via TSVs 134. When the die is in test mode, an instruction in the TCP 122 will disable the outputs of FC 2004 and 2006 using an enable signal (E5) from the CTL output of TCP 122. In test mode, PTI can be input to the test architecture via TSVs 132, PTO can be output from the test architecture via TSVs 136 and PTIO can be input to or output from the test architecture via TSVs 134, as previously described.

FIG. 24 illustrates a die 2402 include the test architecture of the disclosure. The only difference between the die 2302 of FIG. 23 and die 2402 of FIG. 24 is that a 3-state multiplexer 2404 is shown being used in die 2402 in place of buffers “a” and “b”. The E3 input to the multiplexer selects either PTI data from TSVs 132 or the PTO data from the Parallel Scan Circuit to be output to TSVs 136. The E4 input to the multiplexer enables or disables the multiplexer outputs to TSVs 136. The multiplexer operates the same as the buffers “a” and “b” described in FIG. 23.

In this disclosure the words connected and coupled both mean a “link” formed between elements mentioned in this disclosure. The elements could be, but are not limited to circuits, buses and contact points. The links may be direct links such as links formed between two elements by a wire or they may be indirect links such as a links formed between elements through intermediate circuitry, registered circuitry or buffered circuitry for example.

Claims

1. An integrated circuit die comprising: (a) a first surface including Parallel Test Input (PTI) contact points, a first Test Data input (TDI) contact point, a first Test Clock (TCK) contact point, a first Test Mode Select (TMS) contact point, a first Test Data Output (TDO) contact point, and first Parallel Test Input/Output (PTIO) contact points;(b) a second surface including Parallel Test Output contact points, a second TDO contact point, a second TCK contact point, a second TMS contact point, a second TDI contact point, and second PTIO contact points;(c) state machine circuitry having inputs coupled to the first TCK contact point, and the first TMS contact point, and having a control bus output that includes an instruction register control output, and a data register control output;(d) an instruction register having a control input coupled to the instruction register control output, a serial input coupled to the first TDI contact point, a serial output coupled to the second TDO contact point, and having instruction register control outputs;(e) a data register having a control input coupled to the data register control output, a control input coupled to an instruction register control output, a serial input coupled to the first TDI contact point, and a serial output coupled to the second TDO contact point;(f) first switching circuitry having inputs coupled to the PTI contact points, having outputs coupled to the PTO contact points, and having a control input;(g) parallel scan circuitry having PTI inputs coupled to the first surface PTI contact points, having control inputs coupled to the control bus outputs, and having parallel test outputs (PTO); and(h) comparator circuitry having inputs coupled to the PTO outputs of the parallel scan circuitry, having input and output leads coupled to the PTIO contact points of the first and second surfaces, having a control input coupled to a control bus output, and having a fail output.

2. The integrated circuit die of claim 1 in which the switching circuitry includes tri-state buffers.

3. The integrated circuit die of claim 1 in which the instruction register includes a switching circuitry control output coupled to the switching circuitry control input.

4. The integrated circuit die of claim 1 in which the first surface is a bottom surface of the die and the second surface is a top surface of the die.

5. The integrated circuit die of claim 1 in which the first TCK contact point is coupled to the second TCK contact point.

6. The integrated circuit die of claim 1 in which in which the first TMS contact point is coupled to the second TMS contact point.

7. The integrated circuit die of claim 1 in which the serial output of the instruction register is also coupled to the first TDO contact point.

8. The integrated circuit die of claim 1 in which the serial output of the data register is also coupled to the first TDO contact point.

9. The integrated circuit die of claim 1 in which the instruction register control outputs include a SEL output, and including TDO multiplexer circuitry having an input coupled to the second TDI contact point, having an input coupled to the second TDO contact point, having a control output coupled to the SEL output, and having an output coupled to the first TDO.

10. The integrated circuit die of claim 1 in which the instruction register control outputs include an ENA output, and including gating circuitry having an input coupled to the first Test Clock (TCK) contact point, having a control input coupled to the ENA output, and having an output coupled to the second TCK contact point.

11. The integrated circuit die of claim 1 in which the instruction register control outputs include an ENA output, and including gating circuitry having an input coupled to the first Test Mode Select (TMS) contact point, having a control input coupled to the ENA output, and having an output coupled to the second TMS contact point.

12. The integrated circuit die of claim 11 in which the instruction register control outputs include an ENA output, and including gating circuitry having inputs coupled to the first TCK contact point and the first TMS contact point, having a control input coupled to the ENA output, and having outputs coupled to the second TCK contact point and the second TMS contact point.

13. The integrated circuit die of claim 1 including a test control port that includes the state machine, the instruction register, and the data register, the test control port having a control bus output coupled to the switching circuitry control input.

14. The integrated circuit die of claim 1 including a test control port that includes the state machine, the instruction register, and the data register.

15. The integrated circuit die of claim 14 in which the control bus output includes an E2 control output and the comparator circuitry control input is coupled to the E2 control output.