This invention relates to programmable logic array integrated circuit devices (“programmable logic devices” or “PLDs”), and more particularly to interconnection resources for use on programmable logic devices that increase the speed at which those devices can be made to operate.
Programmable logic devices typically include (1) many regions of programmable logic, and (2) programmable interconnection resources for selectively conveying signals to, from, and/or between those logic regions. Each logic region is programmable to perform any of several different, relatively simple logic functions. The interconnection resources are programmable to allow the logic regions to work together to perform much more complex logic functions than can be performed by any individual logic region. Examples of known PLDs are shown in Wahlstrom U.S. Pat. No. 3,473,160, Freeman U.S. Pat. Re. 34,363, Cliff et al. U.S. Pat. No. 5,689,195, Cliff et al. U.S. Pat. No. 5,909,126, and Jefferson et al. U.S. patent application Ser. No. 09/266,235, all which are hereby incorporated by reference herein.
A frequent objective in the design of PLDs is to increase the speed at which the device can be operated. The speeds at which signals can travel through the interconnection resources between logic regions is particularly important to determining device speed. Overall, the interconnection resources must have the general-purpose capability of connecting any logic region to any other logic region. But in addition to this, it can be helpful to find ways to make faster interconnections between nearby logic regions. Many complex logic tasks can be broken down into parts, each of which can be performed by a respective cluster of logic regions. By providing interconnection resources that facilitate the flexible formation of clusters of logic regions with high-speed interconnection capabilities among the logic regions in such clusters, the ability of the PLD to perform various complex logic tasks at high speed in enhanced.
In view of the foregoing, it is an object of this invention to provide improved interconnection resources for programmable logic devices.
It is a more particular object of this invention to provide interconnection resources for programmable logic devices that facilitate the formation of extended clusters of nearby logic modules between which high-speed interconnections can be made.
These and other objects of the invention are accomplished in accordance with the principles of the invention by providing programmable logic devices with interconnection resources that facilitate the provision of interconnections between logic modules in adjacent rows of logic regions, as well as between nearby logic regions in each row. Typically the logic regions on a PLD are arranged in a two-dimensional array of intersecting rows and columns of such regions. Each logic region may include a plurality of subregions. Local feedback conductors may be provided for facilitating communication among the subregions in a region. In addition, these local feedback conductors may be interleaved between horizontally adjacent regions in a row, thereby facilitating high speed interconnection among the subregions of horizontally adjacent regions. In accordance with this invention such high speed local interconnection is additionally provided between adjacent rows in any of several ways. For example, output signals of subregions in each row may be additionally applied substantially directly (i.e., without making use of the more general-purpose interconnection resources of the device) to programmable logic connectors (e.g., multiplexers) feeding output drivers that are otherwise normally or nominally associated with subregions in an adjacent row. This makes it possible for the subregions in one row to optionally drive interconnection resources that are normally associated with an adjacent row, thereby facilitating clustering of logic regions in adjacent rows. As an alternative or addition to the foregoing, the interconnection resources that bring signals into the regions in each row can be partly shifted or extended relative to the rows so that some signals can be more readily and directly brought into each row from the adjacent rows, again without having to make use of the more general-purpose interconnection resources of the device. This again facilitates forming clusters of logic regions in adjacent rows. As still another alternative, the interconnection resources that bring signals into each row can be substantially directly driven by signals from similar resources in another row, thereby again facilitating the formation of clusters of logic regions in adjacent rows without needing to use the general-purpose interconnection resources.
As an alternative or addition to the foregoing, clustering of logic regions along a row may be facilitated by providing conductors associated with each logic region that extend adjacent a relatively small subplurality of the other adjacent logic regions in that row. For example, one of these conductors associated with each logic region may extend to the left from that logic region adjacent a relatively small number of other logic regions to the left of the associated logic region, and another of these conductors may extend to the right by approximately the same number of other logic regions. The same signal or different signals from the associated logic region can be applied to each of these conductors, and thereby to the other logic regions that these conductors are adjacent to. (The signals on these conductors can alternatively come from other sources.) The relatively short length, light loading, and other similar characteristics of these conductors make them especially suitable for use in providing high-speed interconnections from the associated logic region (or other signal source(s)) to the other logic regions that they are adjacent to, thereby again facilitating flexible clustering of nearby logic regions.
Further features of the invention, its nature and various advantages will be more apparent from the accompanying drawings and the following detailed description of the preferred embodiments.
Although the principles of this invention are equally applicable to many different programmable logic device architectures, the invention will be fully understood from the following explanation of its application to programmable logic devices of the type shown in commonly assigned, co-pending, U.S. patent application Ser. No. ______, filed ______ (Docket No. 174/172), which is hereby incorporated by reference herein in its entirety. Because the last-mentioned reference is incorporated by reference herein, it will not be necessary to actually repeat the content of that reference here. Where elements described in that reference are mentioned again here, the same reference numbers will be used here to the greatest extent possible, even though such elements may be here diagrammed more simply or described more briefly.
Associated with each row of logic regions 20 is a plurality of global horizontal interconnection conductors 230a/b. (The “a/b” designation is from the last-mentioned reference and refers to “a” conductors having “normal” signal propagation speed characteristics and “b” conductors having higher signal propagation speed characteristics. This feature is at most only tangential to the present invention and therefore need not be further detailed herein.) Also associated with the left and right half of each row of logic regions 20 is a plurality of so-called direct horizontal interconnection conductors 240. Still further associated with subsets of horizontally adjacent regions 20 in each row are groups of so-called horizontal network of fast line (“HNFL”) interconnection conductors 250.
Associated with each column of logic regions 20 (and extending across I/O regions 40 and into memory regions 50) is a plurality of global vertical interconnection conductors 200a/b. (Again, the “a/b” designation is from the last-mentioned reference and refers to some conductors 200a having normal speed and other conductors 200b having higher speed.) Associated with the upper and lower half of each column of logic regions 20 (and the associated I/O and memory circuitry 40 and 50) is a plurality of so-called half vertical interconnection conductors 210a/b. (Once again, the “a/b” designation refers to different conductors having different signal propagation speed characteristics.) Associated with vertically adjacent pairs of regions 20 and 50 are pluralities of so-called interleaved vertical (“IV”) interconnection conductors 220. These conductors 220 form at least part of a first illustrative embodiment of this invention.
It will be understood that
Each subregion 30 may be constructed as shown in
The portions of
Both of the IV input signals 220 to the circuitry shown in
Other inputs to PLC 522 are (1) the signal from an adjacent global vertical conductor 200b, (2) the signal from an adjacent half vertical conductor 210b, (3) one of the global output signals (“OUT1”) of the upper left-hand subregion 30, (4) one of the global output signals (“OUT0)” of the lower right-hand subregion 30, (5) one of the global output signals (“OUT1”) of the lower left-hand subregion 30, (6) one of the global output signals (“OUT0”) of the upper right-hand subregion 30, and (7) the signal from a selected one of four adjacent global vertical conductors 200a. (PLC 520 makes the one-of-four selection referred to at the end of the preceding sentence.)
PLC 522 is programmable (e.g., by programmable function control elements (“FCEs”) that are not shown in
From the foregoing it will be seen that application to PLC 522 of IV output signals from the row above and the row below allows PLC 522 and its associated tri-state driver 524 to be used to apply one of those IV signals from an adjacent row to a global horizontal conductor 230b associated with the row partly shown in
Other possible routings of the IV signals 220 provided by the
As has been mentioned, one of the two IV signals received by the
The PLC 540 routing of one of the received IV signals 220 is similar to that just described for PLC 530, except that in the case of routing via elements 540, 542, and 544, one of the possible destinations is one of the adjacent normal-speed global vertical conductors 200a.
The PLC 502 routing of one of the received IV signals 220 allows that IV signal to be applied to inverting buffer 504 in lieu of the other inputs to PLC 502. The output signal of buffer 504 is applied to one of the adjacent HNFL conductors 250 that extends to the left adjacent several other logic regions 20 from the circuitry shown in
The PLC 506 and PLC 564 IV signal routing is similar to that described for PLCs 502 and 560, except that the other of the two IV signals 220 received by the
From the foregoing, it will be seen that the IV connections 220 between adjacent rows of logic regions 20 facilitate flexible formation of clusters of logic regions or subregions, as well as relatively high-speed communication within such clusters. For example, using an IV connection 220 and routing via elements like 522 and 524, a subregion 30 in one row can be clustered with (i.e., relatively directly coupled to) any of the subregions in one of the adjacent rows. The same is true (albeit using somewhat slower, normal-speed global horizontal conductors 230a) via elements 530/532/534 or 540/542/544. As another example, using an IV connection 220 and routing via elements like 502, 506, 560, and/or 564, a subregion 30 in one row can be clustered with any of the logic regions 20 in an adjacent row served by the associated HNFL conductors 250 in the adjacent row. In addition to providing more direct, and therefore higher speed interconnections between adjacent rows, the provision of IV conductors 220 reduces the need to use longer-haul and more general-purpose vertical conductors 200 and 210 for inter-row connections. This helps to reduce the numbers of conductors 200 and 210 that must be provided on the device.
It should be noted that the clustering options afforded by the above-described IV circuitry are preferably boundary-less within the array of logic regions 20. By this it is meant that any logic region 20 can serve as a member of a cluster, and the cluster can extend from that logic region in substantially the same way regardless of the logic region that is chosen as the cluster member being considered. Only the physical edges of the logic region array bound the possible clusters.
An alternative embodiment of the invention which facilitates flexible clustering of subregions 30 in one row with logic regions 20 (and their subregions 30) in another adjacent row is shown in
In
Similarly, the PLCs 270/276 that feed the conductors 300 to the right of the left-most logic region 20 shown in row N in
It should be noted that because each PLC group 270/276 is interleaved between two logic regions 20 (one region to the left and one region to the right) as is described more fully above in connection with
A possible alternative to shifting the groups of PLCs 270/276 as shown in
Like the arrangements shown in earlier FIGS., arrangements of the type shown in
The HNFL conductors 250 that have already been occasionally mentioned also facilitate flexible clustering of logic regions 20 without recourse to the general interconnection conductor resource network in accordance with this invention. Accordingly the HNFL conductors will now be considered in more detail in connection with
As has been mentioned,
To briefly recapitulate the foregoing discussion of HNFL conductors 250, these conductors facilitate flexible high-speed clustering of nearby logic regions due to such characteristics as the following: (1) there is only a single source for the signal on each HNFL conductor (i.e., the driver block 500 associated with the pair of logic regions 20 near the midpoint of a left- and right-extending pair of HNFL conductors 250), (2) the HNFL conductors are relatively short, (3) the HNFL conductor signals are applied to region-feeding conductors 300 via downstream PLCs 276 that bypass other upstream PLCs 270, and (4) the PLC groups 270/276 connect the HNFL conductors to only certain groups of region-feeding conductors 300 that the HNFL conductors pass.
From
Connections 276a allow a signal applied to a region-feeding conductor 300 associated with the center row (e.g., from any of the conductors 230/240/250/260 associated with the center row) to be applied not only to the center row logic regions 20 to the left and/or right of that conductor 300, but also to the adjacent logic regions 20 above and/or below those logic regions (i.e., in the adjacent rows above and below the center row). Similarly, connections 276b allow a signal applied to a local feedback conductor 310 associated with the center row (i.e., from the associated subregion 30 in a region to the left or right of that conductor 310) to be applied not only to the center row logic regions to the left and/or right of that conductor 310, but also to the adjacent logic regions 20 above and/or below those logic regions (i.e., in the adjacent rows above and below the center row). Connections 276a and 276b therefore facilitate rapid and close association of logic regions that are generally vertically adjacent to one another. For example, connections 276a facilitate application of the same signals (from the conductors 230/240/250/260 associated with the center row) to vertically adjacent logic regions 20. Connections 276b facilitate application of signals from center row logic regions 20 to other vertically adjacent logic regions 20.
Although both types of connections 276a and 276b are shown in
System 1002 can be used in a wide variety of applications, such as computer networking, data networking, instrumentation, video processing, digital signal processing, or any other application where the advantage of using programmable or reprogrammable logic is desirable. Programmable logic device 10 can be used to perform a variety of different logic functions. For example, programmable logic device 10 can be configured as a processor or controller that works in cooperation with processor 1004. Programmable logic device 10 may also be used as an arbiter for arbitrating access to a shared resource in system 1002. In yet another example, programmable logic device 10 can be configured as an interface between processor 1004 and one of the other components in system 1002. It should be noted that system 1002 is only exemplary, and that the true scope and spirit of the invention should be indicated by the following claims.
Various technologies can be used to implement programmable logic devices 10 having the features of this invention, as well as the various components of those devices (e.g., the above-described PLCs and the FCEs that control the PLCs). For example, each PLC can be a relatively simple programmable connector such as a switch or a plurality of switches for connecting any one of several inputs to an output. Alternatively, each PLC can be a somewhat more complex element which is capable of performing logic (e.g., by logically combining several of its inputs) as well as making a connection. In the latter case, for example, each PLC can be product term logic, implementing functions such as AND, NAND, OR, or NOR. Examples of components suitable for implementing PLCs are EPROMs, EEPROMs, pass transistors, transmission gates, antifuses, laser fuses, metal optional links, etc. As has been mentioned, the various components of PLCs can be controlled by various, programmable, function control elements (“FCEs”). (With certain PLC implementations (e.g., fuses and metal optional links) separate FCE devices are not required.) FCEs can also be implemented in any of several different ways. For example, FCEs can be SRAMs, DRAMs, first-in first-out (“FIFO”) memories, EPROMs, EEPROMs, function control registers (e.g., as in Wahlstrom U.S. Pat. No. 3,473,160), ferro-electric memories, fuses, antifuses, or the like. From the various examples mentioned above it will be seen that this invention is applicable to both one-time-only programmable and reprogrammable devices.
It will be understood that the forgoing is only illustrative of the principles of the invention, and that various modifications can be made by those skilled in the art without departing from the scope and spirit of the invention. For example, the numbers of the various types of resources on device 10 can be different from the numbers present in the depicted and described illustrative embodiments. This applies to such parameters as the numbers of rows and columns of the various types of circuitry, the number of subregions 30 in each region 20, the numbers of the various types of interconnection conductors, the numbers and sizes of the PLCs provided for making interconnections between various types of interconnection conductors, etc. It will also be understood that various directional and orientational terms such as “vertical” and “horizontal,” “left” and “right,” “above” and “below,” “row” and “column,” and the like are used herein only for convenience, and that no fixed or absolute directional or orientational limitations are intended by the use of these words. For example, the devices of this invention can have any desired orientation. If reoriented, different directional or orientational terms may need to be used in their description, but that will not alter their fundamental nature as within the scope and spirit of this invention. Terms like “region” and “subregion” are also used only as generic, relative terms, and other terms may be used for generally similar circuitry. Indeed, these terms may be used interchangeably herein in contexts in which a region/subregion hierarchy is not important. Alternatively, devices within the scope of this invention may have regions of programmable logic that are not divided into subregions. Although look-up table logic is employed in the illustrative embodiments shown and described herein, it will be understood that other types of logic may be used instead if desired. For example, sum-of-products logic, such as is the primary example considered in references like Pederson et al. U.S. Pat. No. 5,241,224 and Patel et al. U.S. Pat. No. 5,371,422 (both of which are hereby incorporated by reference herein in their entireties), may be used instead of look-up table logic.
This application claims the benefit of U.S. provisional patent application No. 60/122,788, filed Mar. 4, 1999, and U.S. provisional patent application No. 60/142,431, filed Jul. 6, 1999, both of which are hereby incorporated by reference herein in their entireties.
Number | Date | Country | |
---|---|---|---|
60122788 | Mar 1999 | US | |
60142431 | Jul 1999 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 10797484 | Mar 2004 | US |
Child | 11087377 | Mar 2005 | US |
Parent | 10299572 | Nov 2002 | US |
Child | 10797484 | Mar 2004 | US |
Parent | 10017199 | Dec 2001 | US |
Child | 10299572 | Nov 2002 | US |
Parent | 09517146 | Mar 2000 | US |
Child | 10017199 | Dec 2001 | US |