Method and apparatus for storage of test results within an integrated circuit

Abstract

An integrated circuit memory device has a plurality of nonvolatile programmable elements which are used to store a pass/fail status bit at selected milestones in a test sequence. At selected points in the test process an element may be programmed to indicate that the device has passed the tests associated with the selected point in the process. Prior to performing further tests on the device, the element is read to verify that it passed previous tests in the test process. If the appropriate elements are not programmed, the device is rejected. A rejected device may be retested according to the previous test steps. Laser fuses, electrically programmable fuses or antifuses are used to store test results. The use of electrically writeable nonvolatile memory elements allows for programming of the elements after the device has been packaged.

Description

FIELD OF THE INVENTION

This invention relates to integrated circuit test circuits and methods. The invention is particularly related to semiconductor memory devices having internal test circuits, and methods of testing semiconductor memory devices.

BACKGROUND OF THE INVENTION

Semiconductor integrated circuit devices are manufactured on a thin wafer of semiconductor material. Generally many devices are manufactured on a single wafer. These devices may be tested with electrical probes to verify functionality of the devices prior to singulation and/or packaging of individual devices. After a device is packaged, further tests are performed. Devices which do not receive a conventional device package may be tested in a temporary package or die holder, and later shipped in die form. Devices which are shipped in die form after passing a complete test flow may be termed Known Good Die (KGD). Some electrical tests may be performed in a Burn-In oven to weed out parts that would likely fail within a short period of time after being sold if the Burn-In step was not performed. In addition to Probe and Burn-In, parts may be tested both hot and cold to verify functionality over specified environmental conditions.

A complete test flow will often require that parts move from one piece of test equipment to another. A first piece of test equipment and test fixtures may be utilized for Probe, another for Burn-In, and yet another for packaged part testing after Burn-In. After being tested on a particular piece of test equipment, the parts may be sorted into bins according to the test results. Occasionally a part may be misbinned i.e., placed in an incorrect bin. This may occur as a result of machine malfunction, or human error. A failed part that is incorrectly placed in a passing bin has the potential of completing the test flow without further failures and may then be sold as a fully functional part.

SUMMARY OF THE INVENTION

An integrated circuit device has nonvolatile memory elements which are programmed to reflect successful completion of a test or series of tests, or completion of a formation step or a series of formation steps of a semiconductor process. Prior to performing additional tests, the nonvolatile memory element corresponding to successful completion of a previous test is read to verify that the part has indeed successfully completed the previous test. The nonvolatile memory elements may also be read prior to performing other production related activities such as packaging the device, and performing quality assurance checks. The nonvolatile memory elements may also be read prior to sale of the devices, and in the event that a device is returned from a customer. The integrated circuit may be a memory device such as a Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Video Random Access Memory (VRAM), etc. The invention is also applicable to other types of integrated circuit devices such as microprocessors, microcontrollers, memory controllers, Application Specific Integrated Circuits (ASICs), etc. Generally, any integrated circuit which requires multiple test steps and in which a nonvolatile memory element can be incorporated without prohibitively increasing the device manufacturing cost is a candidate for incorporation of the invention.

Large volumes of data may be recorded in the process of testing a single integrated circuit device. It is not feasible to store all of this information on the device itself. In accordance with the present invention, an antifuses is blown at predetermined test milestones to record a successful completion of tests associated with the milestone. Devices which do not successfully pass the tests do not have the fuse blown. Subsequent testing of each memory chip proceeds after verification of the blown fuse indicating that the device has passed previous test procedures. Memory devices which do not pass the tests or which do not have the fuse blown are removed from the test flow.

In an alternate embodiment of the present invention, multiple fuses are used to store information concerning the results of a test process. For example, three nonvolatile memory elements may be used to store one of eight bin numbers each of which represent a part classification based on the test results.

Nonvolatile memory elements associated with test results may be used by customers as part of their incoming inspection process. The data stored in these elements may represent standard test results, or results of tests that are specific to the customer.

After initial testing, devices which pass the tests are sent to a memory device assembly area to be packaged. Some devices may not require packaging as they may be destined to be shipped in die form as Known Good Die (KGD) devices.

Prior to encapsulation, nonvolatile memory elements may be read to verify that previous test sequences have been successfully completed. After packaging, the memory devices are tested further. Device Burn-in and other test procedures are performed to verify functionality of the devices after packaging and in adverse environmental conditions. These tests are also useful in detecting devices which are subject to infant mortality. In a preferred embodiment of the invention, devices which pass some or all of these remaining tests have antifuses which are associated with post packaging test milestones. These antifuses are programmed by application of electrical signals to device inputs and/or outputs. As a final inspection prior to shipping the devices, all appropriate fuses may be read to verify successful completion of all test procedures. While certain test procedures are described such as probe testing, Burn-In, etc., it should be noted that the invention is not limited to test flows which utilize these procedures. Alternate test flows in which some or all of these tests are not performed, in which additional tests are performed, or in which multiple tests are combined into a single test are equally valid. Furthermore, nonvolatile memory elements other than antifuses may be used to store the test or formation data.

BRIEF DESCRIPTION OF THE DRAWINGS

The features of the invention as well as objects and advantages will be best understood by reference to the appended claims, detailed description of particular embodiments and the accompanying drawings where:

FIG. 1 is a block schematic diagram of a computer system.

FIG. 2 is a front view and pin assignment of a memory module.

FIG. 3 is a schematic diagram of a memory device.

FIG. 4 is a schematic diagram of an antifuses circuit.

FIG. 5 is a flow chart of a test process.

FIG. 6 is a block diagram of a semiconductor device that incorporates a nonvolatile storage element in accordance with an embodiment of the present invention.

FIG. 7 is a top plan view of one embodiment of the storage-element bank of FIG. 6 with portions broken away.

FIG. 8 is a layout diagram of a portion of another embodiment of the storage-element bank of FIG. 6 .

FIG. 9 is a side view of the portion of the storage-element bank of FIG. 8 taken along lines 9 — 9 .

FIG. 10 is a schematic diagram of yet another embodiment of the storage-element bank of FIG. 6 .

FIG. 11 is a layout diagram of the storage-element bank of FIG. 10 .

FIG. 12 is a block diagram of an electronic system that incorporates a memory device formed in accordance with an embodiment of the present invention.

DESCRIPTION OF THE INVENTION

A computer system 10 as shown in FIG. 1 has a microprocessor 12 which is coupled to a memory module 14 by control circuit 16 and a data bus 18 . Memory module 14 may be, but is not limited to, being a removable module such as a Dual In-Line Memory Module (DIMM), a Single In-Line Memory Module (SIMM), or Multi-Chip Module (MCM). Alternatively, the memory module 14 may be a memory circuit which is directly coupled to the system such as a fast SRAM cache. Control circuit 16 receives microprocessor address and control signals 20 from the microprocessor 12 and sends memory module address and control signals 22 to memory module 14 . Video information may be sent to display 24 via signals 26 from control circuit 16 .

FIG. 2 is a front view and pin assignment of a DIMM which may be representative of the memory module 14 of FIG. 1 . The DIMM module 14 of FIG. 2 is made up of a plurality of memory devices 22 which are mounted on a printed circuit board 24 . Data stored in the DIMM is accessible to the microprocessor 12 of FIG. 1 through connector 26 . Address and control signals may be buffered through buffer 28 to reduce loading on these signals. Additional information concerning DIMM 14 as well as alternate memory modules may be found in the MICRON TECHNOLOGY INC. 1995 DRAM DATA BOOK, herein incorporated by reference.

FIG. 3 is a block schematic diagram of a DRAM which may be representative of one of the memory devices 22 of FIG. 2 . The DRAM of FIG. 2 may be any one of a wide variety which are currently available, and others which will become available as new memory devices are developed and manufactured. Within memory device 22 , are a plurality of nonvolatile memory elements 32 . The nonvolatile memory elements may be laser fuses, electrical fuses, antifuses, or a combination thereof Alternatively, the nonvolatile memory elements may be FLASH memory cells, FERAM memory cells, or any other nonvolatile type of memory cell. The preferred type of nonvolatile memory cell will depend largely on the integrated circuit device type and the process used to fabricate the integrated circuit. In a preferred embodiment of the present invention, the memory device is a DRAM, and the nonvolatile memory elements are electrically programmable antifuses. The use of nonvolatile memory elements within DRAMs is well known in the industry. Laser fuses for example, have been widely used for enabling redundant memory elements to replace nonfunctional elements within DRAM devices.

Nonvolatile memory elements 32 receive address 34 , data 36 and control 38 inputs either directly or through control circuitry such as DRAM control circuit 40 and data path control circuit 42 . These inputs may be used to program and/or read the nonvolatile memory elements. Outputs of the nonvolatile memory elements are used to enable redundant memory elements 44 and may also be read out of the device over data input/output lines 36 . Memory device 22 is primarily used for data storage in memory array 46 in which individual elements or groups of elements are addressed by address 34 through addressing circuitry 48 .

FIG. 4 is a schematic diagram of an antifuse circuit 50 which may be used as a single nonvolatile memory element which may be utilized to store test result information in an integrated circuit such as the memory device 22 of FIG. 2 . In one embodiment of the present invention, a plurality of antifuse nonvolatile memory elements 50 along with control circuitry, not shown, will make up the nonvolatile memory element circuit 32 of FIG. 3 .

Antifuses element 52 is manufactured in a manner which is similar to the manufacture of a capacitor. An unblown antifuse has two electrodes separated by a dielectric. If the antifuse element remains unprogrammed, there is no current path through the element. To program the antifuse, input PGND on signal node 54 is elevated to a programming potential, and signal PROG 56 is taken to a logic high potential. While signal 56 is high, device 58 will provide a path to ground through device 60 to node 62 . The potential difference between the programming potential and ground is sufficient to cause the dielectric of the antifuse structure to break down and the two nodes 54 and 62 become shorted together. Thus, when programmed, a current path does exist between nodes 54 and 62 . The antifuse is a binary signal storage device since a binary logic level can be associated with the blown and unblown states of the antifuse through the presence or absence of a current path through the device. After programming selected devices, the PGND signal 54 and the PROG signal 56 are both coupled to logic low potentials. While PROG is at a logic low potential, inverter 64 activates device 66 which provides a path for a logic low level at node 62 to be coupled to node 68 if antifuse 52 is conductive. To read the antifuse, signal READ* 70 is coupled to a logic low potential which turns on device 72 . Device 72 couples a high potential from Vcc node 73 through a resistive device 74 to node 68 . If the antifuse is conductive, node 68 will remain low since device 74 is designed to be more resistive than the programmed antifuse. If the antifuse is not blown (nonconductive), node 68 will go high. Output node 78 will be driven to the opposite state of node 68 by inverter 76 . If antifuse 52 is blown, a low potential will remain at node 68 , and node 78 will remain high. If the antifuse is unblown, node 68 will go high when the antifuse is read, node 78 will go low and device 80 will be turned on to provide a path from Vcc to node 68 through resistive device 74 . Device 80 is used to latch the high level logic state at node 68 after the READ* signal is brought back high after a read function. If the READ* is not brought back high after the read function, a DC current path from Vcc to ground may exist through a blown antifuse.

Programming signals READ* 70 , PROG and PGND may be logically derived through the use of a test sequence of control signals such as a Write-CAS-Before-RAS (WCBR) programming cycle in a DRAM, the use of a super voltage signal, or other methods of performing test functions within an integrated circuit. The programming potential may be applied to a device input, or may be internally generated through the use of a charge pump circuit.

Laser fuses may be programmed through the use of a laser. In nonvolatile memory devices such as FLASH memory devices, a number of memory elements of the array of memory elements may be utilized to store the test information. For example, a piece of test equipment may read a set of bits representing test results from a predetermined memory location, perform a test of the memory device and then rewrite the test results back into a predetermined location. The test results that are written back into the device may be updated to reflect the results of the most current test. Since this data would be lost when the device is erased, it is not as desirable as adding a number of dedicated nonvolatile memory elements which may be maintained throughout the life of the part. Similarly, test status bits could be maintained in a DRAM cell as long as power is applied to the DRAiM, but would also be lost once power is removed. Reading the nonvolatile memory elements may be by optical means, normal device read cycles, or special test mode cycles such as super voltage or WCBR cycles. Other methods of programming and reading nonvolatile storage elements within an integrated circuit may be appropriate depending on the particular characteristics of the integrated circuit.

While the antifuse circuit of FIG. 4 is drawn towards a DRAM, it should be understood that the invention is not limited to DRAM applications. The microprocessor of FIG. 1 for example, may have a nonvolatile memory storage area in which test results are stored at various milestones within the test process. Other device types may also take advantage of this invention to store test results within a tested device.

Referring to FIG. 5 , nonvolatile memory elements, such as those discussed below in conjunction with FIGS. 7-11 , can be used to store data during the manufacture of the circuit or device that contains them, such as the memory device 10 of FIG. 1 . The stored data may include information regarding the device's formation, such as the identity of a particular process step or sequence of process steps used to form a portion or portions of the integrated device, or may include the identity or result of a test performed on the integrated device during its manufacture. This information may be used after the manufacture of the device for troubleshooting and other purposes. Alternatively, this type of data or other characteristics of the integrated device may be stored in nonvolatile memory elements during the manufacture of the integrated device, and may be read at a subsequent time during the manufacture to control the manufacturing flow. For example, based on the stored data value, a particular process step or sequence may be chosen or omitted, or a particular test may be performed or omitted. This allows electronic checking of the integrated-device dies during the manufacture of the device to ensure that the correct processing steps or tests are performed, and that these steps or tests are performed in the correct sequence.

More specifically, FIG. 5 is a flowchart of a manufacturing process flow according to one embodiment of the invention. In this embodiment, a test flow is discussed, it being understood that the same principles apply to a formation flow that includes formation steps for forming various portions of the integrated device. An integrated circuit is tested at step A. The results of step A are characterized as pass or fail at step 82 . Devices which fail are removed from the test flow to point 84 . Devices that pass the test have a nonvolatile memory element, such as those described in FIGS. 7-11 , programmed at step 86 to indicate that the device has successfully completed step A. At step 88 , the nonvolatile memory element associated with successful completion of step A is read. Devices that have been misbinned, i.e., put in a pass bin that indicates the passing of the test when in reality the device has failed the test and should have been put in a fail bin, will not have a test-pass-status bit programmed, and will be removed from the test flow to point 90 . Devices that do have the process step A bit programmed will proceed to step B.

The results of step B are characterized as pass or fail at step 92 . Devices that fail are removed from the test flow to point 94 . Devices that pass have a nonvolatile memory element programmed at step 96 to indicate that the device has successfully passed the testing step B. At the step 98 , the nonvolatile memory element associated with the successful completion of the testing steps A and B are read. Devices that have been misbinned will not have a process-pass-status bit programmed for at least one of the steps A and B, and will be removed from the manufacturing flow to the point 99 . Devices that do have the process steps A and B bits programmed will proceed to the process step C.

It should be noted that any manufacturing process step may include one or more individual formation or test steps. Also, it may be unnecessary to read all of the status memory elements before proceeding through a particular process sequence. For example, some process steps may be performed on all devices regardless of whether or not they have successfully completed all prior process steps. In this case, certain failure bits or codes may be detected at a later point in the manufacturing flow, or at an inspection point during the manufacturing process. In one embodiment, the devices that pass steps A and B have the associated memory elements A and B programmed. Before performing the step C, the memory elements A and B are checked to verify that out-of-specification parts have not inadvertently been allowed to pass through the process sequence. For the purposes of this description, the memory elements A and B may each represent one or more nonvolatile memory elements such as, but not limited to, those shown in FIGS. 7-11 , antifuses, laser fuses, electrical fuses, FLASH elements, and FERAM elements. Where multiple nonvolatile memory elements are used, the step of programming an element may be accomplished by programming an appropriate one or ones of the elements. Likewise, the step of reading a memory element may include reading multiple elements and verifying the presence of a completion value associated with the data read from the elements. Devices that are removed from the manufacturing flow to the points 84 , 90 , 94 , and 99 may be further processed according to the particular process steps not completed, or may simply be scrapped. In this way, the stored data affects, i.e., controls, the manufacturing process flow. Furthermore, this stored data need not represent an uncompleted/completed process step, but may represent other information, such as the part number or mask revision of the device, that can affect the manufacturing process flow.

The ability to record manufacturing process data, such as test results or formation steps, on a chip-by-chip basis also allows for improved fault and defect tracking through the life of the product line. Furthermore, the test results of individually failed parts can be read to determine whether the part is potentially repairable, or if the part can be utilized as a partially functional or out-of-specification device. For example, memory parts that have failed certain tests may be useful as audio random access memory (RAM) devices where a limited number of memory-bit errors can be tolerated.

FIG. 6 is a block diagram of a semiconductor device 100 , such as a dynamic random access memory (DRAM), which includes a bank 102 of read-only storage elements 104 , such as fuses (not shown in FIG. 6 ). A storage-element interface circuit 106 is coupled to the bank 102 and includes circuitry for determining the state of each storage element and for providing this state to an external device (not shown). Circuits that are suitable for use as the interface circuit 106 are discussed in U.S. Pat. No. 5,301,143, which is incorporated above, and U.S. Pat. No. 5,345,110, entitled “Low-Power Fuse Detect and Latch Circuit”, which issued to Renfro et al. on Sep. 6, 1994 and is incorporated herein by reference.

In operation, the storage elements 104 in the bank 102 are formed such that they store information that identifies the device 100 . For example, the identification information may include, e.g., the manufacturing information discussed above in conjunction with FIG. 5 , the version number of the mask set used to form the device 100 , the date the device 100 was manufactured, the position that the device 100 , in die form, occupied in a wafer (not shown), and the part number of the device 100 . The stored part number is particularly useful where the printed part number has worn off or has been intentionally removed from the package (not shown) of the device 100 . Techniques for encoding the elements 104 in the bank 102 are discussed below in conjunction with FIGS. 7-9 .

FIG. 7 is a top plan view of a portion of one embodiment of the bank 102 of FIG. 6 according to the present invention. The bank 102 includes a number of read-only storage elements 104 a-n . Although multiple storage elements are shown, it is understood that the bank 102 may include only one storage element 104 . Typically, the storage elements 104 are each formed in a layer of conductive material within the device 100 . Not all the storage elements 104 need be formed in the same conductive layer. Conductive materials that may be used to form the storage elements 104 include, e.g., polysilicon, aluminum, and other metals. For example purposes, storage elements 104 a and 104 n are shown in a closed state, and the storage elements 104 b and 104 c are shown in an opened state.

In operation. the read-only storage elements 104 are formed in their desired states. That is, the mask or masks (not shown) used to form the storage elements 104 are such that during the formation, i.e., processing, of the conductive layer or layers containing the storage elements 104 , the elements 104 a and 104 n that are to be in a closed state to provide the desired information are formed in a closed state, and the elements 104 b and 104 c that are to be in an opened state to provide the desired information are formed in an opened state. Thus, the storage elements 104 are formed to store the desired information during the processing of the semiconductor device 100 (specifically, during the processing of the conductive layer containing the storage element), and not in a separate step after the processing of the device 100 as in the known techniques. Formin, the storage elements 104 in their desired state during the processing of the device 100 is referred to as “preprogramming,” the storage elements 104 . Such preprogramming eliminates many error sources to which the prior-art techniques are susceptible, such as programming and software errors and laser alignment errors that may cause the prior-art fuse elements to be programmed in an incorrect state.

FIG. 8 is a layout diagram of a portion of another embodiment of the storage-element bank 102 of FIG. 6 . The bank 102 includes one or more read-only storage modules 108 . Each storage module 108 includes a number of read-only storage elements 104 that are electrically connected in parallel. Each of the storage elements 104 includes a number of read-only storage sections or links 110 that are electrically connected in series. In the embodiment illustrated in FIG. 8 , each of the storage modules 108 includes four storage elements 104 , and each of the storage elements 104 includes three storage links 110 , although the modules 108 may include more or fewer storage elements 104 , and the storage elements 104 may each include more or fewer links 110 . Each of the storage links 110 of a storage element 104 is formed in a different one of the conductive layers 112 , 114 , and 116 ( FIG. 9 ) of the semiconductor device 100 . Although the storage elements 104 are each shown having one link 110 per conductive layer, each storage element 104 may have more or fewer links 1 I 0 per conductive layer. In a preferred embodiment of the invention, each storage element 104 has one storage link 110 formed in each conductive layer of the semiconductor device 100 . Thus, the number of links 110 in each storage element 104 equals the number of conductive layers in the semiconductor device 100 . Such a structure provides advantages that are described below.

FIG. 9 is a side view of the device 100 taken along lines 9 — 9 of FIG. 8 . As shown, the storage element 104 b includes three storage links 110 2a - 110 2c , which are conventionally formed in respective conductive layers 112 , 114 , and 116 . As shown, the links 110 2a and 110 2b are formed, i.e., preprogrammed, in a closed state, and the link 110 2b is preprogrammed in an opened state. Each of the links 110 2a - 110 2c are serially coupled to one another with vertical vias 118 a - 118 d . The vias 118 may be formed from any conductive material, such as aluminum or another metal, in a conventional manner. The uppermost layer 112 and the inner layers 114 and 116 are formed on a conventional substrate 120 , and a conventional passivation layer 122 is formed over the uppermost conductive layer 112 . In one embodiment of the invention, the layers 112 and 116 are formed from polysilicon, and the layer 114 is formed from a metal such as aluminum. One or more insulator layers 124 electrically isolate the layers 112 , 114 , and 116 from one another. The insulator layers 124 may be formed from, e.g, silicon dioxide. The other storage elements 104 a , 104 c , and 104 d are typically formed in a manner similar to that in which the storage element 104 b is formed. For example, referring to FIG. 8 , the storage links 110 1a , 110 3a , and 110 4a are formed in the layer 112 , the links 110 1b , 110 3b , and 110 4b are formed in the layer 114 , and the links 110 1c , 110 3c , and 110 4c are formed in the layer 116 .

Referring to FIGS. 8-9 , in operation, the read-only storage module 108 is preprogrammed in either an opened or a closed state. That is, if at least one of the storage elements 104 is preprogrammed in a closed state, then the storage module 108 is preprogrammed in a closed state. But, if all of the storage elements 104 are preprogrammed in an opened state, then the storage module 108 is preprogrammed in an opened state.

Still referring to FIGS. 8-9 , the storage module 108 is typically preprogrammed in one state such that it can be toggled to the other state by changing the state of a storage link or links 110 in only one of the layers 112 , 114 , or 116 . For example, when the storage module 108 is used to store a bit of data that identifies the version number of the mask set used to form a particular device 100 , in the first version of the mask set, the module 108 is typically preprogrammed in a closed state, with one of the storage elements, here 104 a , preprogrammed in a closed state, and the remaining storage elements, here 104 b-d , preprogrammed in an opened state. As shown, each of the opened storage elements 104 b-d are opened in a different link 110 , and, therefore, in a different layer 112 , 114 , or 116 . For example and as shown, the links 110 2b , 110 3c , and 110 4a , of the storage elements 104 b-d respectively are preprogrammed in an opened state. In a subsequent version of the mask set for the device 100 , suppose one wishes to toggle the state of the storage module 108 from the closed state shown in FIG. 8 to an opened state by changing only one or more of the storage links 110 1c , 110 2c , 110 3c , 110 4c , in the layer 116 . Such may be the case when circuitry or other semiconductor structures that exist only in the layer 116 need to be changed to effect the revision of the device 100 . To preprogram the storage module 108 in an opened state, the storage link 110 1c is preprogrammed in an opened state instead of a closed state. Thus, in the subsequent revision of the device 100 , because all of the storage elements 104 a-d are preprogrammed in an opened state, the storage module 108 is preprogrammed in an opened state. Furthermore, such toggling of the state of the module 108 is accomplished by making changes thereto in only a desired one of the conductive layers 112 , 114 , or 116 . The structure of and the techniques associated with the module 108 allow one to keep track of the mask-set version numbers by changing only the mask or masks for layers in which there are circuitry or other changes, and render unnecessary mask alterations to other conductive layers (not shown) in which there are no changes. One can also appreciate that in a further revision of the device 100 , the storage module 108 can be toggled back to a closed state by closing an opened link 110 in any desired one of the layers 112 , 114 , and 116 . In the preferred embodiment described above, the structure of module 108 allows one to toggle the state of the module 108 by making changes in any one of the conductive layers. Thus, when making mask changes to a conductive layer, one can easily toggle the state of the storage module 108 to effect a change in the stored mask-set version number. Of course, one can understand that multiple storage modules 108 can be used to store such information, with only one module 108 required to be toggled each time a mask change is made.

FIG. 10 is a schematic diagram of portions of the storage-element bank 102 and the storage-element interface circuit 106 of FIG. 6 . The bank 102 includes one or more sets of read-only storage elements per each of a number of conductive layers of the memory device 100 of FIG. 6 . The number of conductive layers that include such storage elements may range from one to the total number of conductive layers in the memory device 100 . In the illustrated embodiment, the bank 102 includes a set of elements 128 a - 128 c in a conductive layer 130 , a set of elements 132 a - 132 d in a conductive layer 134 , a set of elements 136 a - 136 d in a conductive layer 138 , and a set of elements 140 a - 140 e in a conductive layer 142 . For example, in one embodiment of the invention, the conductive layer 130 is the second polysilicon layer, i.e., the “poly2” layer, the conductive layer 134 is the poly1 layer, the conductive layer 138 is the metal-contact layer, and the conductive layer 142 is the metal-interconnect layer. Each element 128 a - 128 c , 132 a - 132 d , 136 a - 136 d , and 140 a - 140 e is formed, i.e., preprogrammed, in a predetermined state, either conductive or nonconductive, during the fabrication of the memory device 100 . Thus, the predetermined states of the elements in each set permanently store a digital value associated with the conductive layer in which the set of elements is formed. In one embodiment of the invention, the digital value identifies the version number of the mask used to form the associated conductive layer. Where there are n elements in a set, the set can track 2 n versions of the associated mask. For example, the set of elements 128 a - 128 c includes three elements. Thus, this set can track 2 3 =8 versions of the mask associated with the conductive layer 130 . Likewise, the elements 132 a - 132 d , 136 a - 136 d , and 140 a - 140 e can respectively track 16, 16, and 32 versions of the masks used to form the conductive layers 134 , 138 , and 142 . Thus, such a bank 102 can track a change in the version of a single mask, even if the version number of the mask set, i.e., the number associated with the set of all the masks, remains the same. Such a bank 102 is useful when specific masks from different mask sets or version numbers are combined.

FIG. 10 also includes one embodiment of a read circuit 144 for the storage-element interface circuit 106 . The read circuit 144 is constructed as shown in FIG. 10 , receives a storage-element position address A 1 ( 0 : 3 ) and a storage-element set address A 2 ( 0 : 3 ), and provides a serial read signal FBANK. The read circuit 144 also includes load transistors 146 , position access transistors 148 a - 148 d , 150 a - 150 d , 152 a - 152 d , 154 a - 154 d , set access transistors 156 a - 156 d , 158 a - 158 d , 160 a - 160 d , and 162 a - 162 d , and a weak latch 164 . For balancing purposes, set transistors 156 d , 158 d , and 160 d are associated with elements 132 a , 136 a , and 140 a respectively, and position transistors 148 d , 150 d , and 152 d are associated with elements 132 a , 136 a , and 140 a respectively.

In operation, before reading a storage element 128 a-c , 132 a - 132 d , 136 a - 136 d , or 140 a - 140 e , the address bits of the position address A 1 ( 0 : 3 ) are driven to inactive logic 0's, which activate the serially coupled PMOS transistors 146 . The active transistors 146 conduct, and thus precharge FBANK to Vcc, i.e., a logic 1. The weak latch 164 then maintains the logic 1 for FBANK. Next, corresponding address bits from the two addresses A 1 ( 0 : 3 ) and A 2 ( 0 : 3 ) are driven active high, and the associated pair of n-channel position and set transistors conduct and thus provide the state of the element being read as the signal FBANK. The signals FBANK from each set of storage elements form the digital value stored by the set. For example, when both address bits A 1 ( 0 ) and A 2 ( 0 ) are driven to active logic 1's, the serially coupled n-channel position and set transistors 148 a and 156 a , respectively, turn on, i.e., conduct, and provide the state of element 128 a as the signal FBANK. If the element 128 a is nonconductive, i.e., opened, then FBANK remains at its precharged value of logic 1. If the element 128 a is conductive, ie., closed, as is shown in FIG. 10 , then the active transistors 148 a and 156 a overcome the output signal of the weak latch 164 and pull FBANK to ground, i.e., a logic 0. The read circuit 144 then precharges FBANK to Vcc before reading each element, and reads the elements 128 b - 128 c . The signals FBANK generated by the set of elements 128 a - 128 c represent the digital value stored by this set of elements. In a manner similar to that discussed for elements 128 a - 128 c , the read circuit 144 precharges FBANK and reads the elements 132 a - 132 d , 136 a - 136 d , and 140 a - 140 e.

FIG. 11 is a layout diagram of the circuitry shown in FIG. 10 .

FIG. 12 is a block diagram of an electronic system 170 that uses the semiconductor device 100 of FIG. 6 . For example, the semiconductor device 100 may be a memory device. The computer system 170 includes computer circuitry 172 for performing computer fulnctions, such as executing software to perform desired calculations and tasks. The circuitry 172 typically includes a processor (not shown) and the device 100 . One or more input devices 174 , such as a keypad or a mouse, are coupled to the computer circuitry 172 and allow an operator (not shown) to manually input data thereto. One or more output devices 176 are coupled to the computer circuitry 172 to provide to the operator data generated by the computer circuitry 172 . Examples of output devices 176 include a printer and a video display unit. One or more data storage devices 178 are coupled to the computer circuitry 172 to store data on or retrieve data from external storage media (not shown). Examples of the storage devices 178 and corresponding storage media include drives that accept hard and floppy disks, tape cassettes, and compact disk read-only memories (CD-ROMs). If the device 100 is a memory device, the computer circuitry 172 is coupled to the ADDRESS and DATA buses (not shown) and the control lines (not shown) of the memory device 100 . Furthermore, although shown as part of the computer circuitry 172 , the device 100 may also be used in one or more of the input devices 174 , output devices 176 , or storage devices 178 .

It will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without departing from the spirit and scope of the invention. For example, the fuse elements 104 and the fuse module 108 may be encoded with and thus store data other than data that identifies the semiconductor device 100 . For example, the fuse elements 104 and the fuse module 108 may store the types of data discussed in conjunction with FIGS. 1-5 . Accordingly, the invention is not limited except as by the appended claims.

Claims

1. A method of testing an integrated circuit, comprising:performing a test of the integrated circuit; and programming a nonvolatile storage element within the integrated circuit in accordance with a result of said step of performing a test, the nonvolatile storage element being programmed by selectively altering the conductive state of the non-volatile storage element.

2. The method according to claim 1, further comprising:reading a value of the nonvolatile storage element; determining a subsequent test step of the integrated circuit device based upon the value read from the nonvolatile storage element.

3. The method according to claim 2, wherein the subsequent test step is ejection of the device from a test flow.

4. The method according to claim 1, further comprising a step of packaging the integrated circuit.

5. The method according to claim 4, wherein said step of programming the nonvolatile storage element occurs after said step of packaging the integrated circuit.

6. A method of testing an integrated circuit, comprising:performing a first test of the integrated circuit; programming a plurality of nonvolatile storage elements within the integrated circuit in accordance with a result of said step of performing a test, the plurality of nonvolatile storage elements being programmed by selectively altering the conductive state of the nonvolatile storage elements, reading the result from the plurality of nonvolatile storage elements; and performing a second test of the integrated circuit only when the result indicates successful completion of the first test.

7. A method of processing/handling an integrated circuit, comprising:programming a nonvolatile storage element within the integrated circuit in accordance with a predetermined designation, the nonvolatile storage element being programmed by selectively altering the conductive state of the nonvolatile storage element; reading a value from the nonvolatile storage element; and selectively affecting a process/manufacturing/handling flow of the integrated circuit in accordance with the value read from the nonvolatile storage element.

8. The method accoiding to claim 7, further comprising preliminary steps of:performing a test of the integrated circuit; and determining said predetermined designation in accordance with a result of said step of performing a test.

9. The method according to claim 7, further comprising a preliminary step of determining said predetermined designation representative of a predetermined product type.

10. The method according to claim 9 wherein said step of selectively affecting a process/manufacturing/handling flow of the IC comprises channeling said integrated circuit to a given packaging flow in accordance with the value read from the nonvolatile storage element.

11. The method according to claim 10 wherein said predetermined product type comprises one of either a Known Good Die or an alternative given packaging type.

12. The method according to claim 10 wherein said given packaging flow comprises a packaging flow of the group consisting of a KGD process flow or an alternative packaging type process flow.

13. The method according to claim 9 wherein said step of deterning said predetermined designation comprises providing a pait classification as said predetermined designation.

14. A method, comprising:reading a value stored in a nonvolatile memory cell of an integrated circuit, the value stored in the nonvolatile memory cell being determined by selectively altering the conductive state of the nonvolatile memory cell; selecting a process step for continuing the manufacture of the integrated circuit, the selecting based on the stored value; and performing the process step with respect to the integrated circuit.

15. The method of claim 14 wherein the process step comprises a formation step for forming a portion of the integrated circuit.

16. The method of claim 14 wherein the process step comprises a testing step for testing the integrated circuit.

17. The method of claim 14 wherein the integrated circuit comprises a semiconductor die that includes the nonvolatile memory cell.

18. The method of claim 14 wherein the integrated circuit comprises:a package; and a semiconductor die that includes the nonvolatile memory cell and that is disposed within the package.

19. The method of claim 14, further comprising:before the reading, performing a preceding process step with respect to the integrated circuit; determining the value based on the preceding process step; and storing the value in the nonvolatile memory cell.

20. The method of claim 14, further comprising before the reading:forming a portion of the integrated circuit; determining the value based on the forming of the poition; and storing the value in the nonvolatile memory cell.

21. The method of claim 14, further comprising before the reading:testing the integrated circuit; determining the value based on a result of the testing; and storing the value in the nonvolatile memory cell.

22. The method of claim 14, further comprising before the reading:determining the value based on a characteristic of the integrated circuit; and storing the value in the nonvolatile memory cell.

23. A method, comprising:forming a layer of conductive material on a substrate of a semiconductor structure; and during the forming of the layer, forming in the layer a read-only storage element having a stored state that describes a characteristic of the semiconductor structure, the stored state being identified by selectively altering the conductive state of the read-only storage element.

24. The method of claim 23 wherein the characteristic comprises a process used during the manufacture of the semiconductor structure.

25. The method of claim 23 wherein the characteristic comprises a process step that forms a portion of the semiconductor structure.

26. The method of claim 23 wherein the characteristic comprises a process step that forms a portion of the semiconductor structure before or concurrently with the layer.

27. The method of claim 23 wherein the characteristic comprises a process step that tests the semiconductor structure before or concurrently with the forming of the layer.

28. The method of claim 23, further comprising:reading the read-only storage element; and controlling a subsequent portion of the process in response to the state of the read-only storage element.

29. A method for forming an integrated circuit, comprising:forming on a substrate a plurality of conductive layers; forming in a first of the layers a first portion of a storage element; forming in a second of the layers a second portion of the storage element such that first and second portions of the storage element together define a state that describes a characteristic of the integrated circuit, the state being identified by selectively altering the conductive state of the first or second portions of the storage element.

30. The method of claim 29 wherein the connecting comprises serially connecting the first portion to the second portion.

31. The method of claim 29, further comprising forming the storage element having more than two portions, no more than one of the portions formed in each of the conductive layers.

32. The method of claim 29, further comprising:reading the storage element; and forming a portion of the integrated circuit in response to the state of the read-only storage element.

33. A method for forming a semiconductor device, comprising:forming on a substrate a first plurality of conductive layers; and forming in one or more of the conductive layers a storage module that stores data that describes a characteristic of the semiconductor device, the module including a second plurality of electrically and parallel coupled storage elements, each of the storage elements including multiple electrically and serially intercoupled storage links, the data being stored by selectively altering the conductive state of the storage link.

34. The method of claim 33, further comprising:forming each of the storage links of each of the storage elements in a different one of the conductive layers; and selectively altering the conductive state in one or more of the storage links in only one layer during the formation of the layer to change the data.

35. The method of claim 34 wherein the forming the storage module comprises:forming one of the storage elements in a conducting state; and forming remaining ones of the storage elements in a nonconducting state by forming one nonconducting storage link in each of the remaining storage elements, each nonconducting storage link formed in a different one of the layers such that each layer contains a nonconducting storage link from at most one of the storage elements.

36. The method of claim 34 wherein the forming the storage module comprises forming each of said storage elements in a nonconducting state by forming one nonconducting storage link in each of the storage elements, each nonconducting storage link formed in a different one of said layers such that each layer contains a nonconducting storage link from no more than one of the storage elements.

37. The method of claim 34, further comprising:reading the storage module; and testing the semiconductor device in response to the data stored in the storage module.

38. A method for forming a semiconductor structure, comprising:forming a conductive layer on a substrate using a mask; forming in the conductive layer one or more read-only storage elements each having a predetermined digital state, the predetermined digital states being determined by selectively altering the conductive state of the storage elements together identifying the mask; reading the storage elements; and controlling a portion of a process used to manufacture the semiconductor structure in response to the predetermined digital states, the portion being before or concurrent with the forming the conductive layer.

39. The method of claim 38, further comprising:forming each of a plurality of conductive layers on the substrate using a mask that is different from masks used to form other of the conductive layers; and forming in each conductive layer one or more read-only storage elements each having a respective predetermined digital state, the predetermined digital states of the storage elements in each respective conductive layer together identifying the mask used to form the respective conductive layer.

40. A method for forming an integrated device, comprising:forming on a substrate a plurality of conductive layers; forming in each conductive layer one or more read-only storage elements that together permanently store a desired data value pertaining to the conductive layer, the value stored in the one or more read-only storage elements being determined by selectively altering the conductive state of the one or more read-only storage elements; reading the data value; and controlling a process used during the manufacture of the integrated device in response to the data value.

41. The method of claim 40 wherein the data value comprises a digital version number of a mask used to form the conductive layer.