TW594991B - Integrated circuit with one metal layer for programming functionality of a logic operation module - Google Patents

Integrated circuit with one metal layer for programming functionality of a logic operation module Download PDF

Info

Publication number
TW594991B
TW594991B TW92120899A TW92120899A TW594991B TW 594991 B TW594991 B TW 594991B TW 92120899 A TW92120899 A TW 92120899A TW 92120899 A TW92120899 A TW 92120899A TW 594991 B TW594991 B TW 594991B
Authority
TW
Taiwan
Prior art keywords
circuit
operation module
logic
input
logic operation
Prior art date
Application number
TW92120899A
Other languages
Chinese (zh)
Other versions
TW200423404A (en
Inventor
Hsin Shih Wang
Shang-Jyh Shieh
Original Assignee
Faraday Tech Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Faraday Tech Corp filed Critical Faraday Tech Corp
Application granted granted Critical
Publication of TW594991B publication Critical patent/TW594991B/en
Publication of TW200423404A publication Critical patent/TW200423404A/en

Links

Landscapes

  • Semiconductor Integrated Circuits (AREA)
  • Design And Manufacture Of Integrated Circuits (AREA)

Abstract

An integrated circuit with one metal layer for programming functionality of a logic operation module formed on a semiconductor body. The logic operation module has an input circuit and a logic gate circuit. The input circuit has n output units. Each output unit is electrically connected a first predetermined voltage level or a corresponding input port of the input circuit. The logic gate circuit is electrically connected to the input circuit, and has 2 (n-1) input ports. At least an input port of the 2 (n-1) input ports is electrically connected to a first output port of a first output unit or a second output port of the first output unit. The logic gate circuit performs a combinational function according to connections of the n output units of the input circuit and the 2 (n-1) input ports of the logic gate circuit.

Description

594991 五、發明說明(1) 發明所屬之技術領域 本發明提供一種可規劃金屬層積體電路(metal programmable integrated circuit),尤指一種可使用 單一金屬層規劃其邏輯運算模組之積體電路。 先前技術 過去,電子元件(例如電容、電阻等)係經由硬體電路 板(rigid circuit board)來進行彼此間的連接,然 而,半導體技術的發展也進一步地促進積體電路 (integrated circuit, IC) 的應用,亦即於同一半導 體製程中,不僅習知的電子元件製作於一晶片上,且所 需的接線(trace)亦同時設置於該晶片上。近年來,半 導體製私應用了次微米製程(sub- micro process)或深 次微米製程(deep subiicro process)的製作技術而 大幅降低接線的線寬,所以,當規劃一更複雜的電路 時,同一晶片上所能容納的元件數便可大幅地增加。過 去,由於半導體技術的限制,一個系統電路需要多個晶 片經由適當外部連接來實施一預定邏輯運算,然而,隨 著半導體製程的精進,單一晶片可容納的電路規模越來 越大,所以組成一個系統電路的複數個晶片便可經由整 合而降低所需晶片數目與外部接線數量。舉例來=,目 如業界已k出系統早晶片(System On a Chip, s〇c)的概594991 V. Description of the invention (1) Technical field to which the invention belongs The present invention provides a metal programmable integrated circuit, especially a integrated circuit that can use a single metal layer to plan its logic operation module. In the past, electronic components (such as capacitors and resistors) were connected to each other via a rigid circuit board. However, the development of semiconductor technology has further promoted integrated circuits (ICs). In the same semiconductor process, not only the conventional electronic components are fabricated on a wafer, but the required traces are also set on the wafer at the same time. In recent years, semiconductor manufacturing has applied sub-micro process or deep subiicro process manufacturing technology to significantly reduce the wiring line width. Therefore, when planning a more complex circuit, the same The number of components that can be accommodated on a wafer can be greatly increased. In the past, due to the limitation of semiconductor technology, a system circuit required multiple chips to implement a predetermined logic operation through appropriate external connections. However, with the advancement of semiconductor processes, the size of circuits that can be accommodated by a single chip has become larger and larger. A plurality of chips of the system circuit can be integrated to reduce the number of required chips and external wiring. As an example, = In the industry, an overview of System On a Chip (SOC) has been released.

第6頁 594991 五、發明說明(2) 念來實現單一晶片即為一個系統的目標。然而,如此複 雜的晶片設計需要更有效率的設計方法(methodology )與 電細輔助設計(computer-aided design、CAD )軟體的協 助方得以完成。 積體電路的發展也逐漸帶動電子產品的發展,舉例來 說’原本以獨立單元出現於電路板上之電子元件改以積 體電路的型式來製作,相較於過去電子元件之間以金屬 導^ (例如銅線)來加以連接,由於積體電路具有較短 與較窄的接線而使内部因接線所產生的寄生電容大幅下 降,亦即使得積體電路運作的準確度大幅提升而優於以 習知電路板方式所構成的電路。此外,由於複數個電路 可經由半導體製程而整合於同一積體電路中,因此造 !:f品普,朝向輕、薄、短、小的趨勢發展,且該電 產 '的,率損耗(power dissipati〇n)與製作成本 』Γ: Γ電(Τ'!亦大幅降低。可攜性電子產品,例 ( ap computer)與個人數位助理 persona digital assistant, PDA) ^ ± H it ^ ^ 大的重視與歡迎,因此承推.^ , / a又判便用者廣 的蓬勃發展。理办Ϊ f 步帶動相關的可攜式裝置 路的大小並不一致,1二在ί知考置下,各類型積體電 最低,因此可大幅提;:件積體電路因其功率損耗 路,例如類比積體電===f,而其他類型的積體電 冤路,射頻積體電路因其功率損耗較Page 6 594991 V. Description of the invention (2) The idea of achieving a single chip is the goal of a system. However, such a complex chip design requires more efficient design methodology (computer-aided design, CAD) software partners to complete. The development of integrated circuits has also gradually promoted the development of electronic products. For example, 'electronic components that originally appeared on circuit boards as separate units were changed to integrated circuit types, compared to the use of metal guides between electronic components in the past. ^ (Such as copper wire) for connection, because the integrated circuit has shorter and narrower wiring, the internal parasitic capacitance generated by the wiring is greatly reduced, that is, the accuracy of the integrated circuit operation is greatly improved and is better than A circuit constructed by a conventional circuit board method. In addition, since multiple circuits can be integrated into the same integrated circuit through semiconductor manufacturing processes, the f: P product is developed towards the trend of lightness, thinness, shortness, and smallness. dissipati〇n) and production costs "Γ: Γ electricity (T '! is also greatly reduced. Portable electronics, such as (ap computer) and personal digital assistant (PDA)) ^ ± H it ^ ^ Great importance And welcome, so the extension of. ^, / A is judged that the users are flourishing. The size of the relevant portable device path driven by the f-step is not consistent. According to the knowledge, the various types of integrated circuits have the lowest power, so it can be greatly improved; because of the power loss circuit of the integrated circuit, For example, the analog integrated circuit === f, and other types of integrated circuits, the RF integrated circuit because of its power loss

594991594991

高或因金屬連線轉折在高頻訊號傳輸中造成能量損耗, 所以在散熱或信號傳輸的考量下無法提高其元件密度, 此外,隨著電路的改良,部份低功率而高速運作的類比 電路已與數位電路結合為混和式(Mixed-Mode)積體電 路,如上所述-,積體電路已經廣泛地應用於各種電子產 品中,而如何規劃積體電路内部元件之間的接線也成為 由於半導體製程的進步而擴展了積體電路的電路規模, 例如一超大型積體電路(very—large — scale integFated circuit, VLSI)的晶片設計工作已非單一工程師的人力 可以應付,因此一積體電路的製造流程進一步分化為製 程、光罩佈局、元件測試等工作。如業界所習知之佈局 獨立性(pattern independent),其係將光罩佈局的設計 與晶片實際半導體製程加以區隔,並根據所應用的半導 體製程技術(例如可容許的接線間隔)來進一步規範光 罩佈局中元件與接線的幾何配置,亦即透過習知設計規 範(design rule)來進行光罩上的圖樣(pattern)設 計,因此積體電路設計工程師便可以不需要了解實際晶 片的製造細節即可從事各式各樣積體電路晶片的設計; 同理,晶片製造者亦無需瞭解該積體電路的性能細節即 可從事製造該積體電路。由於該設計規範的使用,所以 積體電路設計工作便與半導體製程技術分離,減輕了電 路設計者與晶片製作者各別的工作負荷,因此兩者之間High or energy loss caused by high-frequency signal transmission due to the transition of metal wiring, so its component density cannot be increased in consideration of heat dissipation or signal transmission. In addition, with the improvement of circuits, some analog circuits with low power and high speed operation It has been combined with digital circuits into a mixed-mode integrated circuit. As mentioned above, integrated circuits have been widely used in various electronic products, and how to plan the wiring between internal components of integrated circuits has also become The progress of semiconductor manufacturing processes has expanded the circuit scale of integrated circuits. For example, a very large-scale integrated circuit (VLSI) chip design has not been handled by a single engineer, so an integrated circuit The manufacturing process is further differentiated into processes, mask layout, and component testing. As is known in the industry as pattern independence, it separates the design of the mask layout from the actual semiconductor process of the wafer, and further regulates the light according to the applied semiconductor process technology (such as the allowable wiring interval). The geometric configuration of the components and wiring in the mask layout, that is, the pattern design on the mask is performed through the known design rules, so the integrated circuit design engineer can eliminate the need to understand the actual manufacturing details of the chip. It can engage in the design of various integrated circuit chips; similarly, the chip maker does not need to know the performance details of the integrated circuit to manufacture the integrated circuit. Because of the use of this design specification, the integrated circuit design work is separated from the semiconductor process technology, which reduces the separate workload of circuit designers and chip makers.

第8頁 594991 五、發明說明(4) 只要依循該設計規範的原則,積體電路的製造便得到一 玎接受的良率(yield)。 現今的電子產品相繼採用特定用途積體電路 (application specific integrated circuit, ASIC) 以提咼該電子產品的功能以及降低其成本,許多電腦週 邊設備的產品,例如硬碟機與掃描機等,在其電路板上 均設置有各自的特定用途積體電路,該特定用途積體電 路的目的主要是一方面更有效率地整合所需的電路,而 另一方面可以保護其使電路設計而不易被競爭者輕易地 盜拷。然而,一雛塑系統(prototype)的開發係為該特 定用途積體電路製造的狀頸,近年來業界都在找尋一種 快速系統雛型的製程方法,其能在短時間内製作出該系 統雛型’以便驗證該特定用途積體電路的功能以及對該 特定用途積體電路進行除錯,並進而縮短該特定用途積 體電路的上市時間(time-to-ήι a rket)以增加其市場競 爭力(competitiveness)。一般而言,目前積體電路的 設计方法主要可區分為全客戶設計(full_cust〇Ifl design)間陣列设什(gate ar ray des i gn),與標準 細胞原,計(standard cell design),其中全客戶設 計指的是積體電路佈局的工作從設計基礎的電晶體開 始,所以積體電路設計工程師從元件的尺寸,元件的位 置以及元件間的連線等各項佈局細節均需從事設計,此 種設什方式可以獲得最佳的性能(如高運算速度與低消耗Page 8 594991 V. Description of the invention (4) As long as the principles of the design specification are followed, the manufacturing of integrated circuits will get a yield of acceptance. Today's electronic products successively use application specific integrated circuits (ASICs) to improve the functions of the electronic products and reduce their costs. Many computer peripheral products, such as hard disk drives and scanners, The circuit boards are provided with their respective special-purpose integrated circuits. The purpose of the special-purpose integrated circuits is mainly to integrate the required circuits more efficiently on the one hand, and to protect the circuit design from being competitive on the other. Easily stolen. However, the development of a prototype system is for the manufacture of integrated circuits for this specific purpose. In recent years, the industry is looking for a rapid system prototype process method that can produce prototypes of the system in a short time. Type 'in order to verify the function of the special-purpose integrated circuit and debug the special-purpose integrated circuit, and then shorten the time-to-price of a special-purpose integrated circuit to increase its market competition Force (competitiveness). Generally speaking, the current design methods of integrated circuits can be mainly divided into gate array ar ray des i gns (full_custoIfl design), and standard cell design, The all-customer design refers to the work of the integrated circuit layout starting from the design of the transistor. Therefore, the integrated circuit design engineer needs to design the layout details from the component size, component location, and connection between components. , This design method can get the best performance (such as high operation speed and low consumption

第9頁 594991Page 594 991

功率等)以及最高的元件密度(亦即最小的晶 在匕可因為較小的晶片面積而獲得最低的製作 而’全客戶設計的方式所耗費的人力最多, 局開發時間(1 e ad t i me)也較久。 片面積),因 成本。然 且所需的佈 ^準細胞原與閘陣列的設計方式則可用來適度減化設計 工^的複雜度’其中標準細胞原的設計方式則是利^1 =拼ί J cel1 library)中已設計好之小型元件模組加 =凑成一大型的電路,因此主要的工作内容是這些元 件模組的擺置(1)13(^1116111;)與元件模組間的繞線 (routing)。-該元件庫係由過去已開發之小型電路所構 成,由於該元件庫内所包含之小型電路的特性已經得 先前開發階段的驗證,因此組合後之大型電路且有 機率可正確地運作以及有較高的良率。此外,由於所 耗費Ϊ 2 ί Τ ί幅減少’因此開發時間亦得以大幅縮 :於ί t ^ ,每—元件模·組的電路結構不同,因此 當於一晶圓(wafer) 形成 λ η士 . 本身需要一獨特的光罩,亦n、,各兀件模組 造出所要的各種元且 + $ π相交而、& I 且各兀件杈組之光罩設計彼 此互不相合而蛻成晶片製造成本 組的幾何尺寸並不一致,所以x a 7 士 1 〔合70件模 ^ 致,所以不容易有效地縮小晶片面 積0 另一方面,對於閘陣列的設計方式來說,其是一晶圓廠Power, etc.) and the highest component density (that is, the smallest crystals can get the lowest production because of the small chip area, and the all-customer design method consumes the most labor, and the bureau development time (1 e ad ti me ) Is also longer. Sheet area) due to cost. However, the required design method for the layout of the quasi-cell and gate array can be used to moderately reduce the complexity of the design process. Among them, the design method of the standard cyto-gene is in the ^ 1 = spelling J cel1 library). The designed small component module plus = makes up a large circuit, so the main work content is the placement of these component modules (1) 13 (^ 1116111;) and the routing between the component modules. -The component library is composed of small circuits that have been developed in the past. Because the characteristics of the small circuits contained in the component library have been verified in the previous development stage, the combined large circuits can operate correctly and have the right rate. Higher yield. In addition, the development time can be shortened significantly due to the reduction in the cost of 2 ί Τ 幅: In ί t ^, the circuit structure of each element module group is different, so when a wafer (wafer) is formed λ η . Need a unique mask itself, n ,, each element module to create the various elements and + $ π intersect, & I, and the element mask design of each element group does not coincide with each other The geometric dimensions of the wafer manufacturing cost group are not consistent, so xa 7 ± 1 [combined with 70 molds, so it is not easy to effectively reduce the wafer area. On the other hand, for the design method of the gate array, it is a crystal Round factory

594991594991

五、發明說明(6) foundry)提供固定尺寸的標準電晶體以及許 的間距,因此該晶圓廠僅製作電晶體,亦即^門僅 形成包含有電晶體陣列的晶片半成品,但不進S電晶 體之間所需的接線,所以該積體電路設計者便可根據該 晶圓上電晶體的硬體規格來從事各電晶體間的繞線設 計,換句話說,該積體電路設計者的主要工作係規割 'program)該積體電路上層之金屬層(metal”Uyer)5. Description of the invention (6) foundry) Provides standard transistors with fixed dimensions and a small distance, so the wafer fab only manufactures transistors, that is, the gate only forms semi-finished wafers containing transistor arrays, but does not enter S The required wiring between transistors, so the integrated circuit designer can design the windings between the transistors according to the hardware specifications of the transistor on the wafer. In other words, the integrated circuit designer The main work is to 'program) the metal layer (metal) Uyer on the integrated circuit.

的光罩佈局,然後將該光罩佈局送到該晶圓廠來進行後 續的金屬層製程以元成各電晶體間的連線,最後即可製 造出包含該積體電路的晶片,如上所述,由於各電晶體 對應相同的硬體規格,因此同一光罩可重複地用來製造 電晶體於該晶片上,因此可大幅降低生產該電晶體所需 耗費的光罩成本。The photomask layout is then sent to the fab for subsequent metal layer processes to form the connections between the transistors, and finally a wafer containing the integrated circuit can be manufactured, as described above It is stated that since each transistor corresponds to the same hardware specification, the same photomask can be repeatedly used to manufacture the transistor on the wafer, so the cost of the photomask required to produce the transistor can be greatly reduced.

此外,於習知閘陣列的設計方式中,半導體基座經由半 導體製程而僅預先製造至接點層(contact layer),所 以該積體電路設計者必須依據該積體電路的功能來進 光罩圖樣設計’以便該半導體製程於該接點層上形成相 對應金屬層以設置所需接線來啟動積體電路的功能 (functionality)。然而,隨著半導體製程技術的快 發展,積體電路中的接線線寬亦越來越窄,相對地,半' 導體製程所需的光罩亦越來越精密,換句話說,光罩的 價格會隨著半導體製程的進步而大幅上升,所以,由於 習知閘陣列技術需要多層光罩來產生金屬層上的繞線,' 因此造成應用習知閘陣列技術所生產之積體電路的成本In addition, in the conventional design method of the gate array, the semiconductor base is only manufactured in advance to a contact layer through a semiconductor process, so the designer of the integrated circuit must enter the photomask according to the function of the integrated circuit. Pattern design 'so that the semiconductor process forms a corresponding metal layer on the contact layer to set the required wiring to activate the functionality of the integrated circuit. However, with the rapid development of semiconductor process technology, the width of wiring lines in integrated circuits has also become narrower. In contrast, the photomasks required for semi-conductor processes have become more and more precise. In other words, the The price will increase significantly with the progress of the semiconductor process. Therefore, because the conventional gate array technology requires multiple photomasks to generate windings on the metal layer, the cost of the integrated circuit produced by the conventional gate array technology is therefore increased.

第11頁 594991 五、發明說明(7) 上升而影響該積體電路於市場上的競爭力。 發明内容 因此本發明之主要目的在於提供一種可使用單一金屬層 規劃其邏輯運算模組之積體電路,以解決上述問題。 本發明之申請專利範圍提供一種形成於積體電路之半導 體基座(semiconductor body)上之邏輯運算模組,其 包含有一輸入電路以及一邏輯閘電路。該輸入電路包含 有η個輸出單元,每一輸出單元電連接於一第一預定電壓 準位或該輸入電路之相對應輸入端。該邏輯閘電路係電 連接於該輸入電路,其包含有2( η~υ個輸入端,該2 ( η-υ個 輸入端中至少一輸入端係電連接於該輸入電路之第一輸 出單元之第一輸出端或第二輸出端。該邏輯運算模組可 依據該輸入電路之η個輸出單元及該邏輯閘電路之2 ( η—υ個 輸入端所電連接的位置而對應一組合邏輯函數 (combinational function) 〇 本發明之申請專利範圍另提供一種形成於積體電路之半 導體基座(semiconductor body)上之邏輯運算模組, 其包含有一輸入電路以及一邏輯閘電路。該輸入電路包 含有η個輸出單元,每一輸出單元電連接於一第一預定電 壓準位或該輸入電路之相對應輸入端。該邏輯閘電路係Page 11 594991 V. Description of the invention (7) The rise affects the competitiveness of the integrated circuit in the market. SUMMARY OF THE INVENTION Therefore, the main object of the present invention is to provide an integrated circuit that can use a single metal layer to plan its logic operation module to solve the above problems. The patent application scope of the present invention provides a logic operation module formed on a semiconductor body of a semiconductor circuit, which includes an input circuit and a logic gate circuit. The input circuit includes n output units, and each output unit is electrically connected to a first predetermined voltage level or a corresponding input terminal of the input circuit. The logic gate circuit is electrically connected to the input circuit and includes 2 (η ~ υ input terminals, and at least one of the 2 (η-υ input terminals) is electrically connected to the first output unit of the input circuit. The first output terminal or the second output terminal. The logic operation module may correspond to a combination of logic according to the positions of the n output units of the input circuit and the 2 (n—υ input terminals of the logic gate circuit). Function (combinational function) 〇 The scope of patent application of the present invention also provides a logic operation module formed on a semiconductor body of a semiconductor circuit, which includes an input circuit and a logic gate circuit. The input circuit includes There are n output units, and each output unit is electrically connected to a first predetermined voltage level or a corresponding input terminal of the input circuit. The logic gate circuit is

第12頁 594991 五、發明說明(8) 電連接於該輸入電路,其僅包含有2( nM)個輸入端,該2 ( 1 υ個輸入端中之每一輸入端電連接於該第一預定電壓準 位,一第二預定電壓準位,或者該輸入電路之第一輸出 單元之第一輸出端或第二輸出端。其中該邏輯運算模組 可依據該輸入電路之η個輸出單元及該邏輯閘電路之2 ( ηΜ) 個輸入端所電連接的位置形成一組合邏輯函數 (combinational function) 〇 本發明積體電路之半導體基座上係預先經由一晶圓廠製 造複數個運算區塊,而每一運算區塊上設置有邏輯運算 模組,且邏輯運算模組包含有複數個規劃節點用來規劃 該邏輯運算模組的功能。因此,對於邏輯運算模組來 說,一積體電路設計者僅需一層光罩圖樣設計即可決定 該複數個規劃節點的規劃來設計邏輯運算模組的功能, 因此該晶圓廠隨後便可依據該光罩圖樣設計於一金屬層 上設置實施該邏輯運算模組之功能所需的繞線。此外, 本發明邏輯運算模組中,談複數個規劃節點係佈局於最 少的水平軌道中,因此於後續半導體製程所形成的金屬 層中,便可擁有較大空間來設置各運算區塊之間的繞 線。換句話說,本發明積體電路之邏輯運算模組於該積 體電路的製造過程中可使用較少的光罩,因而可大幅地 降低生產成本。 實施方式Page 12 594991 V. Description of the invention (8) Electrically connected to the input circuit, which contains only 2 (nM) input terminals, each of the 2 (1 υ input terminals) is electrically connected to the first A predetermined voltage level, a second predetermined voltage level, or a first output terminal or a second output terminal of a first output unit of the input circuit. The logic operation module may be based on the n output units of the input circuit and The positions where the 2 (ηM) input terminals of the logic gate circuit are electrically connected form a combined logic function. The semiconductor base of the integrated circuit of the present invention is manufactured in advance by a fab with a plurality of operation blocks. , And each operation block is provided with a logical operation module, and the logical operation module includes a plurality of planning nodes for planning the functions of the logical operation module. Therefore, for a logical operation module, a product The circuit designer only needs a layer of mask pattern design to determine the planning of the plurality of planning nodes to design the function of the logic operation module. Therefore, the fab can then design the The windings required to implement the functions of the logic operation module are set on the subordinate layer. In addition, in the logic operation module of the present invention, a plurality of planning nodes are arranged in the least horizontal track, so they are formed in subsequent semiconductor processes. In the metal layer, it is possible to have a larger space for setting the windings between the various operation blocks. In other words, the logic operation module of the integrated circuit of the present invention can use less in the manufacturing process of the integrated circuit. Mask, which can significantly reduce production costs.

第13頁 594991 五、發明說明(9) 請參閱圖一,圖一為本發明積體電路之半導體基座 (semiconductor body) 10的示意圖。半導體基座10包 含有複數個運算區塊(b a s i c u n i t) 1 2,於本實施例 中,運算區塊1 2係以矩陣方式排列於半導體基座丨〇上, 所以可使運算區塊1 2的設置對應較高的元件密度,亦即 縮小半導體基座1 0所需的面積而降低積體電路的尺寸。 然而,運算區塊1 2亦可以其他排列方式設置於半導體基 座10上,例如排列於同一行(row)或同一列(column) 而形成一陣列(array)。半導體基座1 〇係經由一晶圓廠 預先製造,然後一積體電路設計者便可對半導體基座1 〇 中的每一運算區塊1 2來進行相關光罩佈局以規劃每一運 鼻區塊1 2的繞線,最後該晶圓廠便依據該積體電路設計 者所提供的光罩佈局而於半導體基座10上形成至少—金 屬層以設置每一運算區塊1 2實施其相對應功能所需的接 線,然後每一運算區塊1 2便可執行其功能來使該積體電 路依據該積體電路設計者的設計而正常地運作。 請參閱圖二,圖二為圖一所示之運算區塊1 2的功能方塊 圖。運算區塊1 2包含有複數個邏輯運算模組1 4。本實施 例中,邏輯運算模組1 4係用來執行一邏輯函數運算,'此 外,運算區塊1 2亦可包含有一.驅動模組1 6以及一儲存模 組1 8。儲存模組1 8可用來執行資料儲存的功能,而驅動 模組1 6可用來驅動一預定訊號,例如一資料訊號(dataPage 13 594991 V. Description of the invention (9) Please refer to FIG. 1. FIG. 1 is a schematic diagram of a semiconductor body 10 of the integrated circuit of the present invention. The semiconductor base 10 includes a plurality of basic unit blocks 12. In this embodiment, the operation blocks 12 are arranged on the semiconductor base in a matrix manner. Setting a correspondingly higher component density, that is, reducing the area required for the semiconductor base 10 and reducing the size of the integrated circuit. However, the operation blocks 12 can also be arranged on the semiconductor base 10 in other arrangements, for example, arranged in the same row (row) or the same column (column) to form an array. Semiconductor pedestal 10 is pre-manufactured by a fab, and then an integrated circuit designer can perform relevant mask layouts for each operation block 12 in semiconductor pedestal 10 to plan each shipping nose. The winding of block 12 is completed. Finally, the fab will form at least a metal layer on the semiconductor base 10 according to the photomask layout provided by the integrated circuit designer to set each computing block 12 to implement its operation. Corresponding to the required function, each computing block 12 can then execute its function to make the integrated circuit operate normally according to the design of the integrated circuit designer. Please refer to FIG. 2, which is a functional block diagram of the operation block 12 shown in FIG. 1. The operation block 12 includes a plurality of logic operation modules 14. In this embodiment, the logic operation module 14 is used to perform a logical function operation. In addition, the operation block 12 may also include a drive module 16 and a storage module 18. The storage module 18 can be used to perform data storage functions, and the drive module 16 can be used to drive a predetermined signal, such as a data signal (data

第14頁 594991Page 594 991

signal)或是一時 模組1 6可用來驅動 該資料訊號或該時 1 8 ’或者驅動模組 算結果或儲存模組 1 2之另一邏輯運算 動模組1 6亦可用來 存模組1 8的儲存資 异模組1 4或儲存模 脈5孔號(c 1 〇 c k s 該資料訊號或該時 脈訊號至邏輯運算 1 6亦可用來驅動邏 18的儲存資料而後 模組1 4或另一儲存 驅動邏輯運算模組 料而後輪入另一運 組1 8 〇 l gnal),亦即驅動 脈訊號,而後輪入 模組1 4或儲存模組 輯運算模組1 4的運 輸入同一運算區塊 模組18 ’此外,•辱區 1 4的運算結果或儲 鼻區塊1 2之邏輯運 『=參閱圖二與圖三,圖三為圖二所示之邏輯運 功能方塊ϊ。邏輯運算模組14包含有複數個運算 二々a\15b,每一運算電路15a、15b係分別用來執杆 二 j a 邏輯函數(c〇mbinati〇nal functi〇n),例如 包含有AND邏輯閘,。R邏輯閘,以及X〇R邏輯 ^寺構成的組合邏輯函數來對輸入運算電路15&的 行處理’並將處理後的運算資料輸出。本實施例中、,’ 積體電路設計者可經由一光罩佈局來規劃邏輯運算模= 1 4所執行的組合邏輯函數,舉例來說,邏輯運算模組、1 $含有η個輸入端用來接收n個變數資料,而該積體電 設計者可規劃該光罩圖樣佈局以設計邏輯運算模組14 處理的變數資料與對應該變數資料的組合邏輯函數, 即該積體電路設計者可選取運算電路1 5a,其係用 声、 k個變數資料(η),或是該積體電路設計者可選=^signal) or a temporary module 16 can be used to drive the data signal or 1 18 'at that time or to drive the module calculation result or store another logical operation module 16 of the module 12 can also be used to store the module 1. 8 storage resource module 1 4 or storage mold pulse 5 hole number (c 1 0cks the data signal or the clock signal to the logical operation 16 can also be used to drive the storage data of logic 18 and then module 14 or another One storage drives the logic operation module and then turns into another operation group (18 glnal), that is, the driving pulse signal, and the operation of the rear rotation module 14 or the storage module series operation module 14 enters the same operation. Block module 18 'In addition, the operation results of the shame zone 1 4 or the logical operation of the nose block 12 are shown in Figure 2 and Figure 3. Figure 3 is the logical operation function block shown in Figure 2. The logic operation module 14 includes a plurality of operation two 々a \ 15b, and each operation circuit 15a, 15b is used to perform a two-ja logic function (c〇mbinati〇nal functi〇n), for example, including AND logic gate . R logic gate and XOR logic logic combined logic function to process the input operation circuit 15 & and output the processed operation data. In this embodiment, the designer of the integrated circuit can plan the logic operation mode = 1 through a photomask layout. The combinational logic function executed by 1 4 is, for example, a logic operation module, 1 $ contains n input terminals. To receive n variable data, and the integrated circuit designer can plan the mask pattern layout to design the combination of the variable data processed by the logic operation module 14 and the corresponding logical function of the variable data, that is, the integrated circuit designer can Select the arithmetic circuit 1 5a, which uses sound, k variable data (η), or can be selected by the designer of the integrated circuit = ^

1 _ 594991 五、發明說明(11) - 算電路1 5 b,其係用來處理k + m個變數資料(k +仏n)。 換句話說,該積體電路設計者可依據所要的組合邏輯函 數來規劃適當光罩圖樣佈局以決定邏輯運算模組i 4所要 啟動的運算電路15a或運算電路15b。請注意,本實施例 中,β亥晶圓薇係預先於邏輯運算模組1 4中製作各運算電 路1 5 a、1 5 b所需的基礎元件,因此該積體電路設計者僅 需使用一層光罩即可快速地設定對應該基礎元件的相關 繞線以選取邏輯運算模組1 4中運算電路1 5a或運算電路 1 5 b所對應的組合邏輯函數。 請參閱圖四,圖四為圖二所示之邏輯運算模組1 4的第一 種電路示意圖。邏輯運算模組14包含有一輸入電路20以 及一邏輯閘電路2 2。輸入電路.2 0包含有複數個規劃節點 24a、24b、24c,以及複數個反相器( inverter) 25a、 25b、 25c、 26a、 26b、 26c,其中規劃節點 24a、 24b、 24c係用來規劃(program)相對應反相器25a、25b、25c 的輸入端係連接於接地電壓Gnd或是輸入訊號la、lb、1 _ 594991 V. Description of the invention (11)-Calculation circuit 1 5 b, which is used to process k + m variable data (k + 仏 n). In other words, the designer of the integrated circuit can plan an appropriate mask pattern layout according to the desired combinational logic function to determine the operation circuit 15a or operation circuit 15b to be started by the logic operation module i4. Please note that in this embodiment, the β-Haiwa wafer is a basic component required for each calculation circuit 1 5 a and 1 5 b in the logic operation module 14 in advance, so the integrated circuit designer only needs to use One layer of mask can quickly set the relevant windings corresponding to the basic components to select the combinational logic function corresponding to the operation circuit 15a or the operation circuit 15b in the logic operation module 14. Please refer to FIG. 4, which is a schematic diagram of the first circuit of the logic operation module 14 shown in FIG. The logic operation module 14 includes an input circuit 20 and a logic gate circuit 22. The input circuit .2 0 includes a plurality of planning nodes 24a, 24b, and 24c, and a plurality of inverters (inverters) 25a, 25b, 25c, 26a, 26b, and 26c. The planning nodes 24a, 24b, and 24c are used for planning. (Program) The input terminals of the corresponding inverters 25a, 25b, 25c are connected to the ground voltage Gnd or the input signals la, lb,

Ic,而反相器 25a、25b、25c、26a、26b、26c主要係用 來輸出互為反相的訊號至邏輯閘電路2 2,例如反相器2 5 a 可自端點A輸出與輸入訊號I a反相的訊號I A,而反相器 2 6 a可自端點B輸出與輸入訊號I a同相的訊號I B ;若規劃 節點2 4a規劃接地電壓Gnd输入反相器2 5 a,則訊號I A、I B 當然均為接地電壓Gnd。此外,其他反相器25b、26b、 25c、26 c的運作與反相器25a、26 a相同而於此不再重複Ic, and the inverters 25a, 25b, 25c, 26a, 26b, 26c are mainly used to output signals that are inverted to each other to the logic gate circuit 2 2, for example, the inverter 2 5 a can be output and input from the terminal A The signal I a is an inverted signal IA, and the inverter 2 6 a can output a signal IB in the same phase as the input signal I a from the terminal B. If the planning node 2 4a plans the ground voltage Gnd and inputs it to the inverter 2 5 a, then The signals IA and IB are of course ground voltage Gnd. In addition, the operation of the other inverters 25b, 26b, 25c, 26c is the same as that of the inverters 25a, 26a, and will not be repeated here.

第16頁 594991 五、發明說明(12) 地贅述。邏輯閘電路22包含有複數個規劃節點24d、 2 4 e、2 4 f、2 4 g與複數個電晶體2 8 a、2 8 b,其t規劃節點 2 4d、24e、24f、24g係用來規劃輸入邏輯閘電路22的訊 號為接地電壓G n d,操作電壓V c c,訊號I A,或者訊號 1 B。本實施例中,電晶體2 8 a係為η型金屬氧化半導體電 晶體(NMOS transistor),以及電晶體28b係為ρ型金屬 乳化半導體電晶體(PMOS transistor),且複數個電晶 體2 8 a、2 8 b可經由習知互補金屬氧化半導體電晶體製程 所形成,另外,η型金屬氧化半導體電晶體與ρ型金屬氧 化半導體電晶體互相並聯(例如兩電晶體28a、28b)而 用來作為電晶體開關(transistor switch) 30a、30b。 因此,複數個電晶體28a、28b可依據規劃節點24d、 2 4 e、2 4 f、2 4 g的不同規劃而對應不同的組合邏輯函數。 舉例來說,規劃節點2 4 f可規劃操作電壓v c c輸入邏輯閘 電路2 2,規劃節點24b規劃輸入訊號I b輸入邏輯閘電路 2 2 ’以及規劃節點2 4 c規劃輸入訊號I c輸入邏輯閘電路 2 2 ’因此電晶體開關3 0 a、3 0 b的串接電路對於輸入訊號 1 b、I c而言即為一 AND邏輯閘,亦即僅有當輸入訊號I b、 Ic均為高邏輯準位” 1”時,電晶體開關3〇a、3〇1)才會同時 導通。當電晶體開關3 0 a、3 0 b的輸出經由一反相器3 2後 產生一輸出訊號lout,亦即輸出訊號i〇ut與輸入訊號 b、Ic之間最後會對應一 NAND邏輯運算。換句話說,圖 三之邏輯運算模組14中,可經由規劃節點24a、24b、 2 4 c、2 4 d、2 4 e、2 4 f、2 4 g來規劃邏輯運算模組1 4所使用Page 16 594991 V. Explanation of the invention (12). The logic gate circuit 22 includes a plurality of planning nodes 24d, 2 4e, 2 4 f, 2 4 g and a plurality of transistors 2 8 a, 2 8 b. The t planning node 2 4d, 24e, 24f, 24g is used for Let the signal of the input logic gate circuit 22 be a ground voltage G nd, an operating voltage V cc, a signal IA, or a signal 1 B. In this embodiment, the transistor 2 8 a is an n-type metal oxide semiconductor transistor (NMOS transistor), and the transistor 28 b is a p-type metal emulsion semiconductor transistor (PMOS transistor), and the plurality of transistors 2 8 a , 2 8 b can be formed through the conventional complementary metal oxide semiconductor transistor process. In addition, the η-type metal oxide semiconductor transistor and the p-type metal oxide semiconductor transistor are connected in parallel with each other (for example, two transistors 28a, 28b) and used as Transistor switches 30a, 30b. Therefore, the plurality of transistors 28a and 28b can correspond to different combination logic functions according to different plans of the planning nodes 24d, 2e, 2e, and 4g. For example, the planning node 2 4 f can plan the operating voltage vcc input logic gate circuit 2 2, the planning node 24 b plans the input signal I b input logic gate circuit 2 2 ′ and the planning node 2 4 c plans the input signal I c input logic gate Circuit 2 2 'Therefore, the series connection circuit of transistor switches 3 0 a and 3 0 b is an AND logic gate for input signals 1 b and I c, that is, only when the input signals I b and Ic are both high When the logic level is "1", the transistor switches 30a, 301) will be turned on at the same time. When the outputs of the transistor switches 30 a and 30 b pass through an inverter 32, an output signal lout is generated, that is, a NAND logic operation is finally corresponding between the output signal iout and the input signals b and Ic. In other words, in the logical operation module 14 of FIG. 3, the logical operation module 14 can be planned through the planning nodes 24a, 24b, 2 4 c, 2 4 d, 2 4 e, 2 4 f, 2 4 g. use

第17頁 594991 五、發明說明(13) 的輸入訊號I a、I b、I c及對應該輸入訊號丨a、! b、丨⑽ 組合邏輯函數’因此邏輯運算模組丨4可應用於任何需要 一個輸入訊號的組合邏輯函數,任何需要兩個輸入訊號 的組合邏輯函數’以及任何需要三個輸入訊號的組合邏 輯函數。 請參閱圖五,圖五為圖二所示之邏輯運算模組1 4的第二 種電路示意圖。邏輯運算模組1 4包含有一輸入電路40以 及一邏輯閘電路42。輸入電路4 0包含有複數個規劃節點 44a、44b、44c、44d,以及複數個反相器45a、45b、 45c、 45d、 46a、 46b、 46c、 46d,其中規劃節點 44a、 4 4 b、4 4 c、4 4 d係用來規劃相對應反相器4 5 a、4 5 b、 4 5 c、4 5 d的輸入端係連接於接地電壓G n d或是輸入訊號 la、 lb、 Ic、 Id,而反相器 45a、 45b、 45c、 45d、 46a、 4 6 b、4 6 c、4 6 d則主要用來產生互為反相的訊號至邏輯閘 電路42。舉例來說,反相器45a可自端點A輸出與輸入訊 號I a反相的訊號I A,而反相器4 6 b可自端點B輸出與輸入 訊號I a同相的訊號I B,若規劃節點44a規劃接地電壓Gnd 輸入反相器45a,則訊號ΙΑ、IB均為接地電壓Gnd。此 外,其他反相器45b、46b、45c、46c、45d、46d的運作 與反相器4 5 a、4 6 a相同而於此不再重複地贅述。邏輯閘 電路4 2包含有複數個規劃節點4 4 e、4 4 f、4 4 g、4 4 h、 44i、44j、44k、441以及複數個電晶體48a、48b。規劃 節點 44e、 44f、 44g、44h、 44i、 44j、 44k、 441係用來Page 17 594991 V. Input signals I a, I b and I c of the invention description (13) and corresponding input signals 丨 a ,! b 、 丨 ⑽ Combined logic function 'so the logic operation module 丨 4 can be applied to any combined logic function that requires one input signal, any combined logic function that requires two input signals' and any combined logic function that requires three input signals . Please refer to FIG. 5, which is a schematic diagram of the second circuit of the logic operation module 14 shown in FIG. The logic operation module 14 includes an input circuit 40 and a logic gate circuit 42. The input circuit 40 includes a plurality of planning nodes 44a, 44b, 44c, and 44d, and a plurality of inverters 45a, 45b, 45c, 45d, 46a, 46b, 46c, and 46d. Among them, the planning nodes 44a, 4 4b, 4 4 c, 4 4 d are used to plan the corresponding inverters 4 5 a, 4 5 b, 4 5 c, 4 5 d. The input terminals are connected to the ground voltage G nd or the input signals la, lb, Ic, Id, and the inverters 45a, 45b, 45c, 45d, 46a, 4 6b, 4 6 c, 4 6 d are mainly used to generate signals that are inverted to each other to the logic gate circuit 42. For example, the inverter 45a can output the signal IA inverted from the input signal I a from the terminal A, and the inverter 4 6 b can output the signal IB in phase with the input signal I a from the terminal B. The node 44a plans the ground voltage Gnd to be input to the inverter 45a, and the signals IA and IB are both ground voltage Gnd. In addition, the operation of the other inverters 45b, 46b, 45c, 46c, 45d, and 46d is the same as that of the inverters 4 5a and 4 6a and will not be repeated here. The logic gate circuit 42 includes a plurality of planning nodes 4 4 e, 4 4 f, 4 4 g, 4 4 h, 44i, 44j, 44k, 441, and a plurality of transistors 48a, 48b. Planning nodes 44e, 44f, 44g, 44h, 44i, 44j, 44k, 441 are used to

第18頁 594991 五、發明說明(14) -—- 規劃輸入邏輯閘電路4 2的訊號為接地φPage 18 594991 V. Description of the invention (14) ----- Planning the input logic gate circuit 4 2 The signal is ground φ

Vcc,訊號ΙΑ,或者訊號IB。本實施例堅Gnd’操作電壓 為η型金屬氧化半導體電晶體,以及電’電晶體28痛 屬氧化半導體電晶體,且複數個電晶\128b係為ρ型金 習知互補金屬氧化半導體電晶體製程戶斤,a、281)可經由 金屬氧化半導體電晶體與P型金屬氧化^成’另外’ η型 並聯(例如兩電晶體48a、48b)而用决導體電晶體互相 ^ 〇 用朿作為電晶體開 圖五所示 1 4的電路 輯運算模 輸入訊號 輯運算模 個輸入訊 之邏輯運 體開關, 晶體開關 個規劃節 之相對應 組1 4來說 規劃節點 •規劃節 此圖四所 之邏輯運算模組14與圖四所示之 架構類似’主要的不同之處在於圖五所邏 組14最多可接收四個輪入訊號,並 來規劃所需的、组合邏輯函數’而圖三ΐ示之邏 組1 4最多僅可接收二個輸入訊號,並依該三 號來規劃所需的組合邏輯函數。對於圖四所^ 算模組U來說’規劃節點24c係對應於兩個電晶 而另一規劃節點24b則對應於四(亦即2 2)個電 ’因此圖三所示之邏輯運算模組14必須使用四 點24d、24e、24f、24g來控制對應規劃節點24b 四個電晶體開關;對於圖五所示之邏輯運算模 ,規劃節點44d係對應於兩個電晶體開關,%另一 44c則對應於四(亦即個電晶體開關,而另 點44b則對應於八(亦即2,個電晶體開關,因 示之邏輯運算模組14必須使用八個規Vcc, signal ΙΑ, or signal IB. In this embodiment, the Gnd 'operating voltage is an n-type metal oxide semiconductor transistor, and the transistor 28 is an oxide semiconductor transistor, and a plurality of transistors \ 128b are p-type gold conventional complementary metal oxide semiconductor transistors. (A, 281) can be oxidized into P type metal via metal oxide semiconductor transistor and P-type metal ^ in parallel (for example, two transistors 48a, 48b) and the conductor transistors are used to connect each other ^ 〇 Use 朿 as a transistor to open Figure 5 shows the logic operation switch of the input signal of the circuit series operation mode, and the input signal of the operation mode of the operation mode of the crystal series. The corresponding group of the planning section of the crystal switch is the planning node of the planning section. Module 14 is similar to the architecture shown in Figure 4. 'The main difference is that the logic group 14 in Figure 5 can receive up to four round-in signals and plan the required combination of logical functions.' The logic shown in Figure 3 is shown. Group 1 4 can only receive at most two input signals and plan the required combinational logic function according to the three numbers. For the computing module U shown in Figure 4, 'planning node 24c corresponds to two transistors and the other planning node 24b corresponds to four (that is, 2 2) electricity'. Therefore, the logical operation module shown in Figure 3 Group 14 must use four points 24d, 24e, 24f, and 24g to control the four transistor switches corresponding to the planning node 24b. For the logic operation mode shown in Figure 5, the planning node 44d corresponds to two transistor switches. 44c corresponds to four (that is, a transistor switch, and the other point 44b corresponds to eight (that is, 2, transistor switches, because the logic operation module 14 shown must use eight rules)

594991594991

44e、44f、44g、44h、44i、44j、44k、441 來控制對應 規劃節點44b之相對應八個電晶體開關。換句話說,本實 施例中,邏輯運算模組1 4每增加一個輸入訊號,邏輯運 算模組1 4之邏輯閘電路所需之規劃節點亦需增加兩倍以 便可規劃邏輯運算模組1 4所執行之組合邏輯函數,亦即 若邏輯運算模組1 4可處理η個輸入訊號,則其邏輯閘電路 所需之規劃節點數目即為2 ( η1)。综合上述,本發明邏輯 運算模組1 4並未侷限輸入訊號的數量,亦即其可處理趨 輸入訊號所對應的各種組合邏輯函數,舉例來說,圖五 所示之邏輯運算模組14可經由規劃節點24a、24b、24c、 24d、24e、24f、24g、24h、24i、24j、24k、241 來實現 對應一個輸入訊號、二個輸入訊號、三個輸入訊號、或 四個輸入訊號的任意組合邏輯函數。 請同時參閱圖六,圖七,圖八,以及圖九。圖六係為圖 四所示之邏輯運算模組1 4的第一佈局圖,圖七係為圖四 所示之邏輯運算模組1 4的第二佈局圖,圖八係為圖四所 示之邏輯運算模組1 4的第三佈局圖,以及圖九係為圖四 所示之邏輯運算模組14的第四佈局圖。如圖六所示,圖 四所示之邏輯運算模組1 4的第一佈局圖係為一元件層, 其包含有氧化層(oxide) 92,多晶矽(p〇lySilicon) 94,連接墊(contact) 96,以及 η型井(n-well) 98, 而於該元件層之上層則逐一設置有第一金屬層(顯示於 第二佈局圖),第二金屬層(顯示於第三佈局圖),以44e, 44f, 44g, 44h, 44i, 44j, 44k, 441 to control the corresponding eight transistor switches corresponding to the planning node 44b. In other words, in this embodiment, each time an input signal is added to the logic operation module 14, the planning node required for the logic gate circuit of the logic operation module 14 also needs to be doubled so that the logic operation module 14 can be planned. The combined logic function that is executed, that is, if the logic operation module 14 can process n input signals, the number of planned nodes required for its logic gate circuit is 2 (η1). To sum up, the logic operation module 14 of the present invention does not limit the number of input signals, that is, it can process various combined logic functions corresponding to the input signals. For example, the logic operation module 14 shown in FIG. 5 can Via planning nodes 24a, 24b, 24c, 24d, 24e, 24f, 24g, 24h, 24i, 24j, 24k, 241 to achieve any one of the corresponding input signals, two input signals, three input signals, or four input signals Combining logic functions. Please refer to FIG. 6, FIG. 7, FIG. 8, and FIG. 9 at the same time. Figure 6 is the first layout of the logical operation module 14 shown in Figure 4, Figure 7 is the second layout of the logical operation module 14 shown in Figure 4, and Figure 8 is shown in Figure 4 The third layout diagram of the logic operation module 14 and FIG. 9 are the fourth layout diagram of the logic operation module 14 shown in FIG. As shown in FIG. 6, the first layout of the logic operation module 14 shown in FIG. 4 is an element layer, which includes an oxide layer 92, polysilicon 94, and a contact pad. ) 96 and n-well 98, and the first metal layer (shown in the second layout diagram) and the second metal layer (shown in the third layout diagram) are arranged one by one above the element layer. To

第20頁 594991 五、發明說明(16) 及第三金屬層(顯示於第四佈局圖)來進行繞線的規 劃。於該第一、二金屬層中分別包含有繞缘1 〇 〇、2 0 0來 定義邏輯運算模組14中各元件的基礎連接,而於第三金 屬層中則設置有對應各規劃節點2 4 a、2 4 b、2 4 c、2 4 d、 24e、24f、24g,接地電壓Gnd,以及操作電壓Vcc的連接 墊。本發明積體電路之半導體基座1 〇係經由一晶圓廠預 先依據上述第一至第四佈局圖來製作邏輯運算模組14, 因此一積體電路設計者便可依據邏第三金屬層上之各規 劃節點 24a、 24b、 24c、 24d、 24e、 24f、 24g,接地電壓Page 20 594991 V. Description of the invention (16) and the third metal layer (shown in the fourth layout diagram) to plan the winding. The first metal layer and the second metal layer respectively include the edges 100 and 2000 to define the basic connection of each element in the logic operation module 14, and the third metal layer is provided with corresponding planning nodes 2 Connection pads for 4 a, 2 4 b, 2 4 c, 2 4 d, 24e, 24f, 24g, ground voltage Gnd, and operating voltage Vcc. The semiconductor pedestal 10 of the integrated circuit of the present invention is prepared in advance by a wafer fab according to the first to fourth layout diagrams described above, so that a designer of an integrated circuit can design a third metal layer based on logic. Each of the planning nodes 24a, 24b, 24c, 24d, 24e, 24f, 24g, ground voltage

Gnd,以及操作電壓Vcc的連接墊來定義邏輯運算模組14 所實施的組合邏輯函數。請參閱圖四與圖十,圖十為圖 九所示之第四佈局圖上之第三金屬層30 0的示意圖。第三 金屬層30 0包含有複數個連接墊l〇6a、106b、106c、 106d> 106e^ 106f^ 106g^ 107a> 107b> 107c^ 107d^ 107e、107f、107g、107h、107i、107j,其中連接墊 106a、106b、106c、106d、10 6e、106f、l〇6g對應於各 規劃節點 24a、24b、24c、24d、24e、24f、24g,連接螯 l〇7a、107b、107 j則對應接地電壓Gnd,連接墊i〇7 i則對 應操作電壓V c c,連接塾1 Ο 7 d、1 Ο 7 f、1 Ο 7 h對應端點B, 以及連接墊1 Ο 7 e、1 Ο 7 g則對應端點A。另外,連接墊1 〇 7 c 則用來作為一輸出端以輸出該輸出訊號I ou t。 如圖四所示,規劃節點2 4 a係用來規劃輸入一輸入訊號I a 或一接地電壓G n d,因此第三金屬層3 0 0上對應規劃節點Gnd and the connection pad of the operating voltage Vcc define the combinational logic function implemented by the logic operation module 14. Please refer to FIG. 4 and FIG. 10. FIG. 10 is a schematic diagram of the third metal layer 300 on the fourth layout shown in FIG. The third metal layer 300 includes a plurality of connection pads 106a, 106b, 106c, 106d > 106e ^ 106f ^ 106g ^ 107a > 107b > 107c ^ 107d ^ 107e, 107f, 107g, 107h, 107i, 107j, among which the connection The pads 106a, 106b, 106c, 106d, 106e, 106f, and 106g correspond to the planning nodes 24a, 24b, 24c, 24d, 24e, 24f, and 24g, and the connection cheques 107a, 107b, and 107j correspond to the ground voltage. Gnd, connection pad i〇7 i corresponds to operating voltage V cc, connection 塾 1 〇 7 d, 1 〇 7 f, 1 〇 7 h corresponds to terminal B, and connection pad 1 〇 7 e, 1 〇 7 g corresponds to Endpoint A. In addition, the connection pad 107c is used as an output terminal to output the output signal Iout. As shown in FIG. 4, the planning node 2 4 a is used to plan an input signal I a or a ground voltage G n d. Therefore, the corresponding planning node on the third metal layer 3 0 0

第21頁 594991 五、發明說明(17) 2 4 a之連接墊1 0 6 a可用來接收該輸入訊號I a,或經由一光 罩圖樣设计以經由另一半導體製程而形成連接塾1 〇 6 &與 連接墊107a (對應接地電壓Gnd)之間的繞線。同樣地, 對於規劃節點24g而言,其可規劃接地電壓Gnd,操作電 壓V c c,端點A,或端點B連接邏輯閘電路2 2,因此第三金 屬層3 0 0上之相對應連接墊1 〇 6g可經由該光罩圖樣設計而 經由後續半導體製程來形成連接墊1 〇 6 g與連接墊丨〇 7 i . (對應操作電壓Vcc)、連接墊l〇7d (對應端點B)、連 接墊107e (對應端點A)、或連接墊1〇7 j (對應接地電壓 G nd)之間的繞線。關於對應其他規劃節點2 4b、2 4 c、 2 4d、2 4e、24f之連接墊 l〇6b、106c、l〇6d、106e、106f 來說’其繞線的設置原理均連接墊1 〇 6 g (對應規劃節點 2 4 a)相同,因此在此不再重複贅述。本實施例中,對應 規劃節點24a、24b、24c之連接墊i〇6a、i〇6b、l〇6c係設 置於第三金屬層3 〇 〇上同一水平軌道,而對應規劃節點 24d、24e、24f、24g 之連接墊106(1、i〇6e、1〇6f、i〇6g 亦a又置於第二金屬層3〇 〇上同一水平執道,因此第三金屬 層3 0 0僅使用四個水平執道即可設置對應於規劃節^ 接 地電壓,操作電壓,及端點A、B的連接墊。因此,第三 金屬層3 0 0本身金屬層便可擁有較大空間來設置運算區塊 1 2中邏輯運异模組1 4,驅動模組1 6,以及儲存模組1 &之 f 間的繞線。此外,本實施例中,僅需使用一層光罩即可 規劃邏輯運算模組1 4的組合邏輯函數,因此對於應用半 導體基座10的積體電路而言,製造該積體電路所需光罩Page 21 594991 V. Description of the invention (17) The connection pad 1 0 6 a of 2 4 a can be used to receive the input signal I a, or can be designed through a photomask pattern to form a connection through another semiconductor process 塾 1 〇6 & Winding between the connection pad 107a (corresponding to the ground voltage Gnd). Similarly, for the planning node 24g, the planned ground voltage Gnd, the operating voltage V cc, the terminal A, or the terminal B are connected to the logic gate circuit 22, so the corresponding connection on the third metal layer 300 The pad 1 〇6g can be formed by the photomask pattern design and subsequent semiconductor processes to form the connection pad 1 〇6g and the connection pad 丨 〇7 i. (Corresponding to the operating voltage Vcc), the connection pad 107d (corresponding to the end point B) , Connection pad 107e (corresponding to terminal A), or connection pad 107j (corresponding to ground voltage G nd). With regard to the connection pads 106b, 106c, 106d, 106e, and 106f corresponding to other planning nodes 2 4b, 2 4c, 2 4d, 2 4e, and 24f, the principles of their winding settings are all connected to the pad 1 〇6 g (corresponding to the planning node 2 4 a) is the same, so it will not be repeated here. In this embodiment, the connection pads i06a, i06b, and 106c corresponding to the planning nodes 24a, 24b, and 24c are arranged on the same horizontal track on the third metal layer 300, and correspond to the planning nodes 24d, 24e, 24f, 24g connection pads 106 (1, 〇6e, 〇6f, 〇6g are also placed on the second metal layer 300 to perform the same level, so the third metal layer 300 uses only four The grounding voltage, operating voltage, and connection pads of endpoints A and B can be set for each horizontal execution. Therefore, the third metal layer 300 can have a large space to set up the computing area. The windings among the logic different module 14, the drive module 16 and the storage module 1 & f in block 12 are used. In addition, in this embodiment, only one layer of mask is needed to plan the logic operation. The combined logic function of modules 14 and 4, therefore, for an integrated circuit using the semiconductor base 10, a photomask required for manufacturing the integrated circuit

第22頁 594991 五、發明說明(18) 的成本可大幅地降低。 本發明積體電路之半導體基座上係預先經由一晶圓廠製 造複數個運算區塊,而每一運算區塊上設置有邏輯運算 模組,而邏輯運算模組包含有複數個規劃節點。因此, 對於邏輯運算模組來說,一積體電路設計者僅需一層光 罩圖樣設計即可規劃邏輯運算模組的功能,因此當該晶 圓廠依據該光罩圖樣設計即可於另一金屬層上完成實施 該邏輯運算模組之功能所需的繞線。此外,本發明邏輯 運算模組中,該複數個規劃節點係佈局於最少的水平軌 道中,因此於同一金屬層中,便可擁有較大空間來設置 各運算區塊之間所需的繞線。換句話說,本發明積體電 路之邏輯運算模組於該積體電路的製造過程中可使用較 少的光罩,因而大幅地降低生產成本。此外,本發明積 體電路之邏輯運算模組可應用習知特定用途積體電路的 設計流程(d e s i g n f 1 〇 w)來進行邏輯運算模組的相關邏 輯合成(synthesis),時序修正,以及自動化佈局等操 作,因此當依據具有本發明邏輯運算模組之半導體基座 來設計一積體電路時,不需修改習知特定用途積體電路 的設計流程,所以本發明邏輯運算模組便可透過習知設 計流程而輕易地實作。 以上所述僅為本發明之較佳實施例,凡依本發明申請專 利範圍所做之均等變化與修飾,皆應屬本發明專利之涵Page 22 594991 5. The cost of the invention description (18) can be greatly reduced. The semiconductor base of the integrated circuit of the present invention is manufactured in advance with a plurality of arithmetic blocks through a wafer fab. Each arithmetic block is provided with a logic operation module, and the logic operation module includes a plurality of planning nodes. Therefore, for a logic operation module, an integrated circuit designer only needs one layer of mask design to plan the function of the logic operation module. Therefore, when the fab is designed based on the mask pattern, it can be used in another The windings on the metal layer to complete the functions of the logic operation module. In addition, in the logic operation module of the present invention, the plurality of planning nodes are arranged in a minimum horizontal orbit, so that in the same metal layer, there can be a large space to set the windings required between the operation blocks. . In other words, the logic operation module of the integrated circuit of the present invention can use fewer photomasks in the manufacturing process of the integrated circuit, thereby greatly reducing the production cost. In addition, the logic operation module of the integrated circuit of the present invention can apply the design flow (designf 1 0w) of the integrated circuit of a specific application to perform related logic synthesis, timing correction, and automatic layout of the logical operation module. And other operations, so when designing an integrated circuit based on the semiconductor base with the logical operation module of the present invention, there is no need to modify the design flow of the integrated circuit of a specific purpose, so the logical operation module of the present invention can Know the design process and implement it easily. The above description is only a preferred embodiment of the present invention. Any equivalent changes and modifications made in accordance with the scope of the patent application for the present invention shall all belong to the scope of the invention patent.

第23頁 594991 五、發明說明(19) 蓋範圍。 !! 第24頁 594991 圖式簡單說明 圖式之簡單說明 圖一為本發明積體電路之半導體基座的示意圖。 圖二為圖一所示之運算區塊的功能方塊圖。 圖三為圖二所示之邏輯運算模組的功能方塊圖。 圖四為圖二所示之邏輯運算模組的第一種電路示意圖。 圖五為圖二所示之邏輯運算模組的第二種電路示意圖。 圖六係為圖四所示之邏輯運算模組的第一佈局圖。 圖七係為圖四所示之邏輯運算模組的第二佈局圖。 圖八係為圖四所示之邏輯運算模組的第三佈局圖。 圖九係為圖四所示之邏輯運算模組的第四佈局圖。 圖十為圖九所示之第四佈局圖上之第三金屬層的示意 圖。 圖式之符號說明 10 半導體基座 12 運算區塊 14 邏輯運算模組 16 驅動模組 18 儲存模組 15a、15b 運算電路 20 ^ 40 輸入電路 22 、4 2 邏輯閘電路 24a 、24b、 24c、 24d、 24e" 24f、 24g、 44a、 44b、 44c 、44d、44d、44e、 44f、 44g、 44h、 44i、 44j、 44k 規劃 25a 節點 、25b、 25c、 26a、 26b、 26c、 45a、 45b、 45c、Page 23 594991 V. Description of the invention (19) Coverage. !! Page 24 594991 Brief description of the drawings Brief description of the drawings Figure 1 is a schematic diagram of a semiconductor base of an integrated circuit of the present invention. FIG. 2 is a functional block diagram of the operation block shown in FIG. 1. FIG. 3 is a functional block diagram of the logic operation module shown in FIG. 2. FIG. 4 is a schematic diagram of a first circuit of the logic operation module shown in FIG. 2. FIG. 5 is a schematic diagram of a second circuit of the logic operation module shown in FIG. 2. FIG. 6 is a first layout diagram of the logic operation module shown in FIG. 4. FIG. 7 is a second layout diagram of the logic operation module shown in FIG. 4. FIG. 8 is a third layout diagram of the logic operation module shown in FIG. 4. FIG. 9 is a fourth layout diagram of the logic operation module shown in FIG. 4. FIG. 10 is a schematic view of the third metal layer on the fourth layout shown in FIG. Explanation of Symbols of the Drawings 10 Semiconductor base 12 Calculation block 14 Logic operation module 16 Drive module 18 Storage module 15a, 15b Operation circuit 20 ^ 40 Input circuit 22, 4 2 Logic gate circuits 24a, 24b, 24c, 24d 24e &24; 24f, 24g, 44a, 44b, 44c, 44d, 44d, 44e, 44f, 44g, 44h, 44i, 44j, 44k Planning 25a nodes, 25b, 25c, 26a, 26b, 26c, 45a, 45b, 45c,

594991 圖式簡單說明 45d、 4 6a、 46b、 46c Λ 4 6 d 28a、 28b、 48a、 48b φ 日 电a曰 92 氧化層 96^ 106a、 106b, • 106c、1 107a 、107b 、107c、 107d、 107i 、l〇7j 連接墊 98 n型井594991 Schematic description of 45d, 4 6a, 46b, 46c Λ 4 6 d 28a, 28b, 48a, 48b φ NEC a 92 oxide layer 96 ^ 106a, 106b, 106c, 1 107a, 107b, 107c, 107d, 107i, l07j connection pads, 98n wells

94 多晶矽 06d、 106e、 106f、 106g、 107e、 107f、 107g、 107h 第26頁94 Polycrystalline silicon 06d, 106e, 106f, 106g, 107e, 107f, 107g, 107h page 26

Claims (1)

594991 六、申請專利範圍 1. 一種形成於積體電路之半導體基座(semiconductor body)上之邏輯運算模組,其包含有: 一輸入電路,其包含有η個輸出單元,每一輸出單元電連 接於一第一預定電壓準位或該輸入電路之相對應輸入 端;以及 一邏輯閘電路,電連接於該輸入電路,其包含有2( 個 輸入端,該2 ( η-υ個輸入端令至少一輸入端係電連接於該 輸入電路之第一輸出單元之第一輸出端或第二輸出端; 其中該邏輯運算模組可依據該輸入電路之η個輸出單元及 該邏輯閘電路之2 ( 個輸入端所電連接的位置而對應一 組合邏輯函數(combinational function)。 2. 如申請專利範圍第1項所述之邏輯運算模組,其中該半 導體基座包含有複數個運算區塊,且每一運算區塊設置 有至少一邏輯運算模組。 3. 如申請專利範圍第1項所述之邏輯運算模組,其係使用 一層光罩(photomask)形成該η個輸出單元與該2( n_1)個 輸入端之接線(trace)以執行該組合邏輯函數。 4. 如申請專利範圍第1項所述之邏輯運算模組,其中該邏 輯閘電路包含有複數個金屬氧化半導體電晶體(M0S transistor) °594991 VI. Scope of patent application 1. A logic operation module formed on a semiconductor body of a semiconductor circuit, which includes: an input circuit including n output units, each output unit is electrically Connected to a first predetermined voltage level or a corresponding input terminal of the input circuit; and a logic gate circuit electrically connected to the input circuit, which includes 2 (input terminals, the 2 (η-υ input terminals) The at least one input terminal is electrically connected to the first output terminal or the second output terminal of the first output unit of the input circuit; wherein the logic operation module may be based on the n output units of the input circuit and the logic gate circuit. 2 (The positions where the input terminals are electrically connected correspond to a combinatorial function. 2. The logic operation module described in item 1 of the scope of patent application, wherein the semiconductor base includes a plurality of operation blocks And each operation block is provided with at least one logical operation module. 3. The logical operation module described in item 1 of the scope of patent application, which uses a layer of photomask (photomas k) forming a trace between the n output units and the 2 (n_1) input terminals to execute the combined logic function. 4. The logic operation module according to item 1 of the scope of patent application, wherein the logic gate The circuit contains a plurality of metal oxide semiconductor transistors (M0S transistor) ° 第27頁 594991 六、申請專利範圍 5. 如申請專利範圍第4項所述之邏輯運算模組,其中該邏 輯閘電路係經由一互補金屬氧化半導體電晶體(CMOS) 製程所製造。 6. 如申請專利範圍第1項所述之邏輯運算模組,其中該第 一預定電壓準位係為一接地電壓(ground voltage)。 7. —種形成於積體電路之半導體基座(semiconductor body)上之邏輯運算模組,其包含有: 一輸入電路,其包含有η個輸出單元,每一輸出單元電連 接於一第一預定電壓準位或該輸入電路之相對應輸入 端;以及 一邏輯閘電路,電連接於該輸入電路,其僅包含有2( ηΜ) 個輸入端,該2 ( ηΜ)個輸入端中之每一輸入端電連接於該 第一預定電壓準位,一第二預定電壓準位’或者該輸入 電路之第一輸出單元之第一輸出端或第二輸出端; 其中該邏輯運算模組可依據該輸入電路之η個輸出單元及 該邏輯閘電路之2 ( ηΜ)個輸入端所電連接的位置形成一組 合邏輯函數(combinational function)。 8. 如申請專利範圍第7項所述之邏輯運算模組,其中該半 導體基座包含有複數侗運算區塊,且每一運算區塊設置 有至少一邏輯運算模組。Page 27 594991 6. Scope of patent application 5. The logic operation module described in item 4 of the scope of patent application, wherein the logic gate circuit is manufactured through a complementary metal oxide semiconductor transistor (CMOS) process. 6. The logic operation module according to item 1 of the scope of patent application, wherein the first predetermined voltage level is a ground voltage. 7. —A logic operation module formed on a semiconductor body of a semiconductor circuit, comprising: an input circuit including n output units, each output unit is electrically connected to a first A predetermined voltage level or a corresponding input terminal of the input circuit; and a logic gate circuit electrically connected to the input circuit, which includes only 2 (ηΜ) input terminals, each of the 2 (ηΜ) input terminals An input terminal is electrically connected to the first predetermined voltage level, a second predetermined voltage level 'or the first output terminal or the second output terminal of the first output unit of the input circuit; wherein the logic operation module may be based on The positions where the n output units of the input circuit and the 2 (ηM) input terminals of the logic gate circuit are electrically connected form a combined national function. 8. The logical operation module according to item 7 in the scope of the patent application, wherein the semiconductor base includes a complex unitary operation block, and each operation block is provided with at least one logical operation module. 第28頁 594991 六、申請專利範圍 9.如申請專利範圍第7項所述之邏輯運算模組,其係使用 一光罩(photomask)形成該η個輸出單元與該2( nl)個輸 入端之接線(trace)以執行該組合邏輯函數。Page 28 594991 6. Scope of patent application 9. The logic operation module described in item 7 of the scope of patent application, which uses a photomask to form the n output units and the 2 (nl) input terminals. Trace to perform the combinational logic function. 第29頁Page 29
TW92120899A 2003-04-29 2003-07-30 Integrated circuit with one metal layer for programming functionality of a logic operation module TW594991B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US24967503A 2003-04-29 2003-04-29

Publications (2)

Publication Number Publication Date
TW594991B true TW594991B (en) 2004-06-21
TW200423404A TW200423404A (en) 2004-11-01

Family

ID=34078803

Family Applications (1)

Application Number Title Priority Date Filing Date
TW92120899A TW594991B (en) 2003-04-29 2003-07-30 Integrated circuit with one metal layer for programming functionality of a logic operation module

Country Status (1)

Country Link
TW (1) TW594991B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8653857B2 (en) 2006-03-09 2014-02-18 Tela Innovations, Inc. Circuitry and layouts for XOR and XNOR logic
US9563733B2 (en) 2009-05-06 2017-02-07 Tela Innovations, Inc. Cell circuit and layout with linear finfet structures
US8658542B2 (en) 2006-03-09 2014-02-25 Tela Innovations, Inc. Coarse grid design methods and structures
US7763534B2 (en) 2007-10-26 2010-07-27 Tela Innovations, Inc. Methods, structures and designs for self-aligning local interconnects used in integrated circuits
US7956421B2 (en) 2008-03-13 2011-06-07 Tela Innovations, Inc. Cross-coupled transistor layouts in restricted gate level layout architecture
US7446352B2 (en) 2006-03-09 2008-11-04 Tela Innovations, Inc. Dynamic array architecture
US8541879B2 (en) 2007-12-13 2013-09-24 Tela Innovations, Inc. Super-self-aligned contacts and method for making the same
TWI569160B (en) * 2007-03-05 2017-02-01 泰拉創新股份有限公司 Method for defining layout, generating cell library and designing integrated circuit, and set of masks for multiple patterning
US8667443B2 (en) 2007-03-05 2014-03-04 Tela Innovations, Inc. Integrated circuit cell library for multiple patterning
US8902625B2 (en) * 2011-11-22 2014-12-02 Marvell World Trade Ltd. Layouts for memory and logic circuits in a system-on-chip

Also Published As

Publication number Publication date
TW200423404A (en) 2004-11-01

Similar Documents

Publication Publication Date Title
US10339245B2 (en) Timing exact design conversions from FPGA to ASIC
US8589847B2 (en) Circuits and methods for programmable transistor array
US11461528B2 (en) Integrated circuit, system for and method of forming an integrated circuit
US6747478B2 (en) Field programmable gate array with convertibility to application specific integrated circuit
US7463059B2 (en) Alterable application specific integrated circuit (ASIC)
US7451426B2 (en) Application specific configurable logic IP
TW594991B (en) Integrated circuit with one metal layer for programming functionality of a logic operation module
US20240037309A1 (en) Multiplexer
US11392743B2 (en) Multiplexer
US6902957B2 (en) Metal programmable integrated circuit capable of utilizing a plurality of clock sources and capable of eliminating clock skew
JP5630856B2 (en) Semiconductor device
TW201917821A (en) An integrated circuit and a cell structure
TWI227057B (en) Method of fabricating an integrated circuit through utilizing metal layers to program randomly positioned basic units
TW202429325A (en) Compressor circuit and semiconductor integrated circuit including the same
JPH058576B2 (en)
JP2005197529A (en) Semiconductor integrated circuit and manufacturing method thereof
JPH01287946A (en) Gate array

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees