The technical field relates generally to software analysis of voltage drops in semiconductor devices.
Computer modeling of semiconductor devices, e.g., integrated circuits, is commonly used in various development stages of such devices. Determining voltage drops, commonly referred to as IR drops, in semiconductor devices is one analysis that may be performed.
Unfortunately, in large and/or complex semiconductor devices, the voltage drop analysis may be overly cumbersome and time consuming. For example, a 64-megabit (“Mbit”) static random access memory (“SRAM”) module may include more than 300 million sinks, e.g., transistors, resistors, or other primitive devices. The voltage drop analysis of such a module may take over 512 gigabytes (“GB”) of computer memory and several days of run time to process.
Accordingly, it is desirable to provide systems and methods of providing voltage drop analysis of semiconductor devices that are expedient and consume fewer computer resources. Furthermore, other desirable features and characteristics of the present invention will become apparent from the subsequent detailed description of the invention and the appended claims, taken in conjunction with the accompanying drawings and this background of the invention.
A method, according to one embodiment, is provided for computing IR drop in a semiconductor device. The semiconductor device is divided into a plurality of transistors. The method includes generating a modeling element corresponding to the plurality of transistors. The method further includes replacing at least one of the plurality of the transistors with a current source in the modeling element. The method also includes performing an IR drop analysis of the modeling element utilizing a software program.
A system, according to one embodiment, is provided for computing IR drop in a semiconductor device. The semiconductor device is divided into a plurality of blocks and each block includes a plurality of transistors. The system includes a processor and a memory in communication with the processor. The memory is configured to receive a fine modeling element corresponding to at least one fine block of the plurality of blocks. The fine modeling element models at least one of the plurality of transistors of the at least one fine block as a transistor. The memory is also configured to receive a coarse modeling element corresponding to at least one coarse block of the plurality of blocks. The coarse modeling element models at least one of the plurality of transistors of the at least one coarse block as a current source. The processor is configured to perform IR drop analysis of the fine modeling element and the coarse modeling element utilizing a software program.
The present invention will hereinafter be described in conjunction with the following drawing figures, wherein like numerals denote like elements, and
The following detailed description of the invention is merely exemplary in nature and is not intended to limit the invention or the application and uses of the invention. Furthermore, there is no intention to be bound by any theory presented in the preceding background of the invention or the following detailed description of the invention.
One embodiment of a semiconductor device 100 is shown in
However, in other embodiments, the semiconductor device 100 may perform other functions, besides storing data. For instance, the semiconductor devices 100 described herein may include, but are certainly not limited to, logic devices, embedded circuits, and mixed signal designs.
The semiconductor device 100 of the exemplary embodiment includes a plurality of transistors 102. The transistors 102 are field-effect transistors (“FETs”) and more specifically metal-oxide-semiconductor field-effect transistors (“MOSFETs”). However, the semiconductor device 100 may include other types of transistors 102 as well as other types of electronic components, as is well appreciated by those skilled in the art.
The transistors 102 of the semiconductor device 100 may be grouped into a plurality of blocks 104. The blocks 104 may correspond to the physical layout and/or placement of the transistors 102 as part of the semiconductor device 100, as shown in
Referring now to
The computer system 300 includes a processor 302 capable of performing calculations and/or executing instructions, i.e., running a software program. The computer system 300 may also include a memory 304 in communication with the processor 302 for storing data and transferring such data to and from the processor 302. Examples of a suitable memory 304 include, but are not limited to, random-access memory, read only memory, flash memory, and a hard disk drive. The computer system 300 may further include at least one input device 306 and at least one output device 308. Examples of the input device 306 include, but are not limited to, a network communications apparatus, a keyboard, a mouse, and a touchscreen display. Examples of the output device 308 include, but are not limited to, the network communications apparatus, a display, and a printer.
One embodiment of a method 400 of computing IR drop in a semiconductor device 100 is shown in
The method 400 includes, at label 402, generating at least one modeling element (not shown) corresponding to the plurality of transistors 102. The modeling element may be implemented as a computer data file (not shown). The computer data file provides information related to the transistors 102 or other components of the semiconductor device 100 for use by a computer program. For example, the computer data file may include data regarding interconnection of the transistors 102 and/or electrical characteristics of the transistors 102. These electrical characteristics may include voltage drop across the transistors 102 in various states, resistance, etc., as is appreciated by those skilled in the art. The method 400 may also include (not shown) generating at least one modeling element corresponding to electronic components other than the transistors 102. Once generated, the at least one modeling element may be stored in the memory 304 for use by the processor 302.
One problem with processing the modeling elements, e.g., performing an IR drop analysis, is that such processing may be overly cumbersome and time consuming. Some of the fine detail required in modeling each and every transistor 102 of the semiconductor device 100 may not be absolutely necessary. Therefore, in one embodiment, the method 400 includes, at label 404, replacing at least one of the plurality of the transistors 102 with a current source 500 in the modeling element. A schematic representation of such a replacement is shown in
Other possible techniques for replacing transistors 102 are shown in
Referring again to
By replacing one or more transistors 102 with the current source 500 and/or resistors 700, processing of the modeling element, i.e., performing the IR drop analysis, may be performed faster and with less computer resources, e.g., processor 302 time and memory 304, then processing the IR drop analysis with each and every transistor 102 modeled in the modeling element.
To further illustrate practical techniques of implementing the method 400, the blocks 104 of transistors 102 may be categorized depending on the whether a more detailed IR drop analysis of the transistors 102 in the block 104 is required or whether a less detailed IR drop analysis may suffice. Accordingly, with reference again to
As such, and with reference to
For instance, selecting at least one coarse block 504 may include selecting at least one of the plurality of blocks 104 disposed adjacent a voltage rail 106. For instance, if a block 104 is disposed in close proximity to the voltage rail 106, which typically provides a steady voltage to the blocks 102, then the block 102, in this instance, may be selected as a coarse block 504.
Selecting at least one coarse block 504 may also include selecting at least one block 102 of a lowest design level. For instance, those skilled in the art appreciate that certain blocks 102 of semiconductor devices may be categorized based on their functionality, commonly referred to as their “design level”. A block of minimum importance, i.e., a block 102 at the lowest design level, may be selected as a coarse block 504.
In one embodiment, selecting the at least one fine block 502 may include selecting at least one block 104 where a highest IR drop is expected, i.e., where the highest magnitude of voltage drop is expected to occur. Utilizing an understanding of IR drops in a semiconductor device 100 and/or past experience, those skilled in the art are able to determine which block of the semiconductor device 100 will likely incur the largest IR drop. However, the selection of at least one fine block 502 is not limited to merely one block 104. It should be appreciated that multiple fine blocks 502 may be selected. Furthermore, it should also be appreciated that the selection of the at least one fine block 502 need not occur where the highest IR drop is expected, but at any location, depending on the needs and nature of the particular IR drop analysis being performed.
The at least one fine block 502 may be surrounded by a plurality of coarse blocks 504. As such, the method 400, in one embodiment, may also include selecting at least one fine block 502 as a block 104 that is surrounded by a plurality of coarse blocks 504. For example, the fine blocks 502 of the embodiment shown in
Referring again to
The method 400 of this embodiment also includes, at label 804, generating a coarse modeling element (not shown) corresponding to at least one coarse block 504 of the plurality of blocks 104. The coarse modeling element models at least one of the plurality of transistors 102 of the at least one coarse block 504 as a current source 500.
In the exemplary embodiments, the coarse modeling element models a plurality of the plurality of transistors 102 of the at least one coarse block 504 as at least one current source 500. Said another way, multiple transistors 102 are modeled as a current source 500. More specifically, in the exemplary embodiment shown in
Once generated, the fine and coarse modeling elements may be received for analysis by the processor 302. For instance, the modeling elements may be received and stored in the memory 304.
Referring again to
As stated above, utilizing the method 400 provides a more efficient and faster running IR drop analysis. For example, in one experiment, an IR drop analysis was performed on a semiconductor device 100 where all of the transistors 102 are modeled as transistors. The exemplary semiconductor device 100 included 128 blocks. In this experiment, the IR drop analysis ran for 8 hours and 59 minutes, required 288 gigabytes (GB) of memory 304 at peak usage, and calculated a maximum IR drop of 15.1 mV. In contrast, in another experiment, the IR drop analysis was performed in accordance with the method 400 of
The method 400 may further include (not shown), calculating an IR drop of the semiconductor device 100 based on the IR drop analysis of the fine modeling element and the coarse modeling element. That is, the IR drop calculated for each block 104 may be combined together to determine the IR drop for the entire device 100.
The method 400 may also include reporting the IR drop of the individual blocks 104 and/or the semiconductor device 100 to a user. For instance, the IR drop of the semiconductor device 100 may be displayed on a display of the output device 308 for use by the user. The IR drop may also be written to the memory 304 or transferred to an external database (not shown), such that IR drops of multiple semiconductor devices 100 that are being simulated may be compared.
A computer program product, i.e., a software program, may also be realized to implement the methods described above. In one embodiment, the computer program product includes a fine modeling element corresponding to at least one fine block of the plurality of blocks. As stated previously, the fine modeling element models at least one of the plurality of transistors of the at least one fine block as a transistor. The computer program product may further include a coarse modeling element corresponding to at least one coarse block of the plurality of blocks. As also stated previously, the coarse modeling element models at least one of the plurality of transistors of the at least one coarse block as a current source. The computer program product may also include an IR drop analysis routine. The IR drop analysis routine is configured to compute the IR drop of the fine modeling element and the IR drop of the coarse modeling element. The IR drop analysis routine is further configured to compute the IR drop of the entire semiconductor device based on the IR drops of the fine and coarse modeling elements.
While at least one exemplary embodiment has been presented in the foregoing detailed description of the invention, it should be appreciated that a vast number of variations exist. It should also be appreciated that the exemplary embodiment or exemplary embodiments are only examples, and are not intended to limit the scope, applicability, or configuration of the invention in any way. Rather, the foregoing detailed description will provide those skilled in the art with a convenient road map for implementing an exemplary embodiment of the invention, it being understood that various changes may be made in the function and arrangement of elements described in an exemplary embodiment without departing from the scope of the invention as set forth in the appended claims and their legal equivalents.
This application claims the benefit of U.S. Provisional Application No. 61/830,384, filed Jun. 3, 2013, which is hereby incorporated by reference.
Number | Name | Date | Kind |
---|---|---|---|
20020045995 | Shimazaki et al. | Apr 2002 | A1 |
20040073878 | Hasegawa et al. | Apr 2004 | A1 |
20050035893 | Horsky et al. | Feb 2005 | A1 |
20060026540 | Bhooshan et al. | Feb 2006 | A1 |
20080052654 | Rahmat et al. | Feb 2008 | A1 |
20110249770 | Bellaouar et al. | Oct 2011 | A1 |
Number | Date | Country | |
---|---|---|---|
20140359551 A1 | Dec 2014 | US |
Number | Date | Country | |
---|---|---|---|
61830384 | Jun 2013 | US |