1. Field
This application relates to integrated circuits, and more particularly to interconnecting integrated circuits.
2. Related Art
There have been many reasons for interconnecting more than one integrated circuit die to form a single packaged device. One use has been to increase memory for a given package. Another has been to combine two die that are commonly used together but are difficult to make using a process that is effective for both. One example is a logic circuit and an RF circuit used for mobile phones. Sometimes there are interconnect issues or interference issues that must be addressed. In any case there are sometimes issues that are addressed because of the particular combination of die being implemented. Regardless of the reason for the combination of the multiple die, there are issues that arise in order to overcome the fact that there is a need to have multiple die. The ability to combine various functionalities on a single die remains limited so the issues associated with multiple die continue.
Accordingly there is a need for improved techniques for interconnecting multiple die.
The present invention is illustrated by way of example and is not limited by the accompanying figures, in which like references indicate similar elements. Elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale.
In one aspect, two integrated circuit die form a device using an intermediate substrate for electrical contact and physical support. The active sides of the die face the intermediate substrate. The intermediate substrate then provides contact externally through conductive contacts or through a package substrate. This is better understood by reference to the following description and the drawings.
Shown in
With regard to die 12, system interconnect 18 is connected to core 20 at a master port 21 of system interconnect 18, to DMA 22 at a master port 23 of system interconnect 18, to master circuit 24 at a master port 25 of system interconnect 18, to configuration register 26 at a master port 27 of system interconnect 18, to peripheral 28 at a slave port 29 of system interconnect 18, to NVM 30 at a slave port 31 of system interconnect 18, to SRAM 32 at a slave port 33 of system interconnect 18, and to slave circuit 34 at a slave port 35 of system interconnect 18. Master circuit 52 is connected to external terminals 66 and 68 which in this example are not connected externally to die 12. Configuration register 26 is shown connected directly to decoder 36 for clarity of function but is actually connected to decoder 36 through system interconnect 18. External terminal 42 is connected to slave circuit 34 and to intermediate substrate 16. External terminal 44 is connected to configuration register 26 and intermediate substrate 16. Slave circuit 34 is for connecting to the secondary die. Master circuit 24 is connected to core 20. Intermediate substrate 16 is for connecting die 12 and 14 together both electrically and structurally. The resources connected to what is shown as the upper portion of system interconnect 18 are connected to master ports and those resources on the lower portion of system interconnect 18 are connected to slave ports. Thus, core 20, DMA 22, and master circuit 24 are communicatively coupled to system interconnect 18 at master ports. Peripheral 28, NVM 30, SRAM 32, slave circuit 34, and configuration register 26 are communicatively coupled to system interconnect 18 at slave ports. Having a microcontroller with a system interconnect divided having slave ports and master ports is well known in the art.
With regard to die 14, system interconnect 46 is connected to core 48, DMA 50, master circuit 52, decoder 54, configuration register 56, peripheral 58, NVM 60, SRAM 62, slave circuit 64. Master circuit 52 is connected to external terminals 66 and 68. External terminals 66 and 68 are connected to intermediate substrate 16. Decoder 54 is shown being directly connected to configuration register 56 for clarity of function but is actually connected to configuration register 56 through system interconnect 46. Configuration register 56 is connected to external terminal 70. Slave circuit 64 is connected to external terminal 72. External terminals 70 and 72 are not connected to circuitry external to die 14. Slave circuit 34 and configuration register 26 being connected to master circuit 52 through intermediate substrate 16 establish die 12 as the primary and die 14 as the secondary. Core 48, DMA 50, master circuit 52 are communicatively coupled to system interconnect 18 at master ports. Peripheral 58, NVM 60, SRAM 62, slave circuit 64, and configuration register 56 are communicatively coupled to system interconnect 18 at slave ports.
In operation, core 20 can access resources connected to system interconnect 18 as well as peripheral 58, NVM 60, and SRAM 62 connected to system interconnect 46. Decoder 36 decodes the system interconnect to load configuration register with the control information that external terminal 44 will provide information that die 12 is the primary. This is received by external terminal 68 and thus master circuit 52 as a configuration signal C through intermediate substrate 16. Master circuit 52 is for receiving transaction requests from the primary die acting as the master. Slave circuit 34 controls transactions T with master circuit 52 through intermediate substrate 16 and external terminal 66. For example, if core 20 chooses to access SRAM 62, this is communicated to slave circuit 34 through system interconnect 18. Slave circuit communicates the transaction T to master circuit 52. Master circuit 52 then performs the transaction regarding SRAM 62 through system interconnect 46. The transaction is communicated back from master circuit 52 to slave circuit 34 and from slave circuit 34 to core 20 using system interconnect 18. This is further explained with reference to
Shown in
Core 20 has access to the resources connected to system interconnect 46 and thus has doubled the resources at its disposal. In the case of adding memory such as NVM 60 and SRAM 62, integrated circuit 12 must also be able to add corresponding address space compared to what is required for just using the memory connected to system interconnect 18. This is rarely a problem because the amount of system memory onboard a microcontroller is far less than the addressing capability of the core. Core 20 would be expected to have addressing capability of at least 32 bits and perhaps 64 or even 128. Even with the low addressing capability of only 32 bits, the number of memory locations being able to be addressed is in excess of 4 billion. If there was a byte in each location that would be a capability of addressing in excess of 4 gigabytes of memory. At the same time, however, the address space for the memory in integrated circuit 14 is the same as that for integrated circuit 12. Thus, in order to treat the memory of integrated circuit 14 as additional memory, there must be an address translation when core 20 is addressing the memory of integrated circuit 14. This is shown in
When a resource on the secondary die, such as SRAM 62, is treated as a duplicate resource, it replaces the identical resource SRAM 32 on the primary die. In operation, core 20 would access the address space associated with SRAM 32 across system interconnect 18, yet the access would be diverted to SRAM 62 via slave 1 circuit 34, intermediate substrate 16, master 2 circuit 52, and system interconnect 46. In this operation no address translation is required, however, the address decoding logic associated with SRAM 32 is disabled.
For an operational example, if an address for a write is to be communicated ultimately to SRAM 62, then communication handshake circuit 78 must be ready to receive it. Address translation 80, under the control of configuration register 26, performs necessary translations. In this example of die 12 and die 14 being the same design, the memory space allocated by decoder 36 for the memory, such as NVM 60 or SRAM 62, of die 14 is different than that recognized by die 14. Thus a translation is required. Configuration register 26 thus communicates what translation is needed. Address translation circuit 80 thus performs the translation that is commanded by configuration register 26. Master logic 82 receives the translated address from address translation circuit 80 and negotiates with system interconnect 46 to perform the commanded transaction. Core 48 is placed into a lower power mode under the command of configuration register 26. Core 48 may be active during start-up, but after start-up has been completed, core 48 may be powered down to save power. In this example, translation is performed by the secondary die, but the translation could instead be performed by the primary die. As shown in
In case of die 14 providing information back to die 12, master logic 82 receives the information from system interconnect 46 and couples the information to address translation circuit 80. Address translation circuit 80 performs any needed translation under the command of configuration register 26. Communication handshake circuit coordinates with handshake circuit 76 to properly communicate the information to logic 74. Logic 74 then negotiates with system interconnect to get the information through system interconnect to core 20.
This operation allows for core 20 to use resources of die 14 that are connected to system interconnect 46. Thus, a variety of experiments may be run to determine the optimum combination of resources for a next generation of integrated circuits. Because the experiments are being run with existing integrated circuits from which there is already, and probably improving, manufacturing capability, the time to market for an integrated circuit with a new combination of such resources is expected to be short.
Shown in
Shown in
In other applications where the die can be different, the symmetry may not be of concern and the approach shown in
Shown in
Shown in
Shown in
Shown in
Thus, a variety of variations for packaging die 12 and 14 are available as shown in
By now it should be appreciated that there has been provided a semiconductor device. The semiconductor device includes an intermediate substrate having a first surface and a second surface. The semiconductor device further includes a first die attached to the first surface of the intermediate substrate in which the first die has a first active surface and the first active surface faces the intermediate substrate, a second die attached to the second surface of the intermediate substrate in which the second die has a second active surface, the second active surface faces the intermediate substrate, and the second die is coupled to the first die through an electrically conductive material in the intermediate substrate. The semiconductor device further includes an organic material encapsulating at least an edge of the intermediate substrate and an edge the second die. The semiconductor device may be further characterized by the first die further including a master circuit and a master port, in which the master circuit is coupled to the master port, the second die further including a slave circuit and a slave port in which the slave circuit is coupled to the slave port. The semiconductor device may be further characterized by the second die being over the first die. The semiconductor device may further comprise a substrate in which the first die is closer to the substrate than the second die and the intermediate substrate is wirebonded to the substrate. The semiconductor device may further comprise solder balls attached to the substrate, wherein the intermediate substrate is coupled to the solder balls. The semiconductor device may further comprise a heat spreader over the substrate and in contact with a non-active surface of the first die, wherein the non-active surface is parallel to the first active surface. The semiconductor device may be further characterized by the organic material being over a non-active surface of the second die wherein the non-active surface is parallel to the second active surface. The semiconductor device may be further characterized by at least a portion of the second die being exposed. The semiconductor device may be further characterized by the first die being over the second die. The semiconductor device may be further characterized by at least a portion of the first die being exposed. The semiconductor device may be further characterized by at least a portion of the second die being exposed. The semiconductor device may be further characterized by the second die being attached to a leadframe. The semiconductor device may further include a substrate, wherein the second die is closer to the substrate than the first die. The semiconductor device may further comprise a via, wherein the via comprises the electrically conductive material. The semiconductor device of claim may be further characterized by the first die further comprising a first master port and a first slave port, the first master port and the first slave port being symmetrically located around a line of symmetry of the first die, the second die further comprising a second master port and a second slave port, the second master port and the second slave port being symmetrically located around a line of symmetry of the second die, and the first master port being coupled to the second slave port.
Also described is a semiconductor device. The semiconductor device includes a first die having a first die active surface and a first die non-active surface, wherein the first die active surface and the first die non-active surface are parallel to each other. The semiconductor device further includes a second die over the first die, wherein the second die has a second die active surface and a second die non-active surface, wherein the second die active surface and the second die non-active surface are parallel to each other. The semiconductor device further includes an intermediate substrate between the first die and the second die in which the first die active surface is closer to the intermediate substrate than the first die non-active surface and the second die active surface is closer to the intermediate substrate than the second die non-active surface. The semiconductor device further includes an organic material encapsulating an edge of the intermediate substrate. The semiconductor device may further comprise plurality of vias in the intermediate substrate, wherein the plurality of vias couple the first die to the second die. The semiconductor may be further characterized by the first die further comprising a first master port and a first slave port, the first master port and the first slave port being symmetrically located around a line of symmetry of the first die, the second die further comprising a second master port and a second slave port, the second master port and the second slave port being symmetrically located around a line of symmetry of the second die, and the first master port being coupled to the second slave port through the plurality of vias. The semiconductor device may be further characterized by the first die being closer to a substrate than the second die and the intermediate substrate being wirebonded to the substrate.
Described also is a method of forming a semiconductor device. The method includes attaching a first die to an intermediate substrate, wherein a first die active surface is facing the intermediate substrate. The method further includes attaching a second die to the intermediate substrate in which a second die active surface is facing the intermediate substrate and the second die is coupled to the first die via the intermediate substrate. The method further includes encapsulating at least a portion of the intermediate substrate with an organic material after attaching the second die to the intermediate substrate.
Moreover, the terms “front,” “back,” “top,” “bottom,” “over,” “under” and the like in the description and in the claims, if any, are used for descriptive purposes and not necessarily for describing permanent relative positions. It is understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein are, for example, capable of operation in other orientations than those illustrated or otherwise described herein.
Although the invention is described herein with reference to specific embodiments, various modifications and changes can be made without departing from the scope of the present invention as set forth in the claims below. For example, the materials used may differ from those described. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of the present invention. Any benefits, advantages, or solutions to problems that are described herein with regard to specific embodiments are not intended to be construed as a critical, required, or essential feature or element of any or all the claims.
The term “coupled,” as used herein, is not intended to be limited to a direct coupling or a mechanical coupling.
Furthermore, the terms “a” or “an,” as used herein, are defined as one or more than one. Also, the use of introductory phrases such as “at least one” and “one or more” in the claims should not be construed to imply that the introduction of another claim element by the indefinite articles “a” or “an” limits any particular claim containing such introduced claim element to inventions containing only one such element, even when the same claim includes the introductory phrases “one or more” or “at least one” and indefinite articles such as “a” or “an.” The same holds true for the use of definite articles.
Unless stated otherwise, terms such as “first” and “second” are used to arbitrarily distinguish between the elements such terms describe. Thus, these terms are not necessarily intended to indicate temporal or other prioritization of such elements.
This application is related to U.S. application docket number AC50050HH, titled “Technique for Interconnecting integrated circuits,” by Gary L. Miller and Ronald W. Stence,” filed on even date herewith, and assigned to the assignee hereof.