TWI363185B - - Google Patents

Download PDF

Info

Publication number
TWI363185B
TWI363185B TW94122698A TW94122698A TWI363185B TW I363185 B TWI363185 B TW I363185B TW 94122698 A TW94122698 A TW 94122698A TW 94122698 A TW94122698 A TW 94122698A TW I363185 B TWI363185 B TW I363185B
Authority
TW
Taiwan
Prior art keywords
test
machine
test machine
semiconductor integrated
integrated circuit
Prior art date
Application number
TW94122698A
Other languages
Chinese (zh)
Other versions
TW200702692A (en
Inventor
Satoh Masayuki
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 filed Critical
Priority to TW094122698A priority Critical patent/TW200702692A/en
Publication of TW200702692A publication Critical patent/TW200702692A/en
Application granted granted Critical
Publication of TWI363185B publication Critical patent/TWI363185B/zh

Links

Description

^185 4 九、發明說明: 【發明所屬之技術領域】 本發明是關於一種伴隨半導體積體電路研發的支援系 統。 【先前技術】 近年來,隨著半導體技術的發展,進展出LSI(Large Scale Integration)等半導體積體電路的高度化和多樣化 等’其研發效率愈顯重要。 _ +導體積體電路的研發步驟,亦即,從設計到製造的 步驟,一般如下。當研發半導體積體電路時,首先進行以 力月b動=為主的動作層級的設計,亦即,系統設計。接著, 進㈣輯間層、級的邏輯設計和以元件層級表現該設計的電 - 路設計。然後,轉換至稱為晶圓前製程的製造程序,該程 序為,製作光罩並且使用該光罩在晶圓上形成積體電路。 其後,使用探針測試對前製程中所製造的晶圓進行測 忒,判斷其是否為良品,將晶圓分割成晶片(半導體積體電 路)’找出藉由探針測試出為良品的良品晶片,進行針對封 裝的組裝。 此組裝藉由特性評估來除錯,當認定特性滿足所要的 值時’經由根據最後測試規格之測斌程式的最後測試,轉 移至量產。 在括些測試中,使用稱為測試機的裝置。測試機為每 "達數千萬日圓至數億曰圓的高價裝置,使用成本亦高, ’貝J試機,的選擇對使用者來說為一重要問題^[185] IX. Description of the Invention: [Technical Field of the Invention] The present invention relates to a support system developed with a semiconductor integrated circuit. [Prior Art] In recent years, with the development of semiconductor technology, the development and efficiency of semiconductor integrated circuits such as LSI (Large Scale Integration) have progressed, and the research and development efficiency has become more and more important. The development steps of the _ + volume body circuit, that is, the steps from design to manufacture are generally as follows. When developing a semiconductor integrated circuit, the design of the action level, that is, the system design, is first performed. Next, enter the (four) inter-layer, the logical design of the level and the electrical design of the design at the component level. Then, the process is switched to a manufacturing process called a pre-wafer process in which a photomask is fabricated and an integrated circuit is formed on the wafer using the photomask. Then, using the probe test to test the wafer fabricated in the pre-process, determine whether it is a good product, and divide the wafer into a wafer (semiconductor integrated circuit) to find out that it is good by the probe test. A good wafer is assembled for the package. This assembly is debugged by characterization and, when it is determined that the characteristic meets the desired value, is transferred to mass production via the final test of the bin program according to the final test specification. In some of these tests, a device called a test machine is used. The test machine is a high-priced device for every tens of thousands of yen to hundreds of millions of rounds, and the cost of use is high. The selection of the 'Be J test machine is an important issue for the user^

2036'7232-PF 5 ICO =試機由測試頭、職本體、控制器等所構成。所謂 端^遠,係指和作為測試對象之半導體積體電路的輸出入 、接,内建複數個對半導體積體電路輸出入信 為接腳電器的介面板。 稱 作為測試機的架構’所包括的方式有以複數個接腳共 疋時產生電路和圖樣產生電路的共享資源方式、在每個 /、有疋時產生電路且共用圖樣產生電路的針對各接 腳方式在每個接腳上具有定時產生電路和圖樣產生電路 的方式(以下稱「完全針對各接腳方式」)等。 k去夕使用共于資源的測試機,但最近,開始有很多 人使用針對各㈣方式。推料後會有更多人制完全針 對各接腳方式。各測試機具有對半導體積體電路供給電力 的電源、評估半冑體積體電路之輸出入端子之DC(Direct Current)特性的DC測量系統、產生對半導體積體電路供給 之 DC 層級的 DAC(Digital t〇Anal〇gC〇nverter)轉換器等。 這些測試機是以在(:叩的卯(作業系統)下運作的測試 程式來控制。此測試程式以所謂的測試機語言來撰寫。一 敖來說,不同的測试機架構,所支援的測試機語言也不同。 在此測試機語言方面’包括當初用來直接控制硬體且被稱 為機器語言的組合語言形式的語言。相對於此,在程式性 高的語言方面,已設計出測試機控制語言,亦使用卩⑽”… 形式和BASIC形式等。再者,結構化語言pASCAL亦曾經被廣 泛使用’目前則以C語§為主0 另外,近年來,在半導體積體電路的領域中,出現各 62036'7232-PF 5 ICO = Test machine consists of test head, job body, controller, etc. The term "terminal" refers to the input and output of the semiconductor integrated circuit as the test object, and a plurality of dielectric panels for inputting and receiving signals to the semiconductor integrated circuit are built in. The structure referred to as the "testing machine" includes a shared resource method for generating a circuit and a pattern generating circuit when a plurality of pins are shared, and a circuit for generating a circuit and sharing a pattern generating circuit for each /. The foot mode has a timing generation circuit and a pattern generation circuit on each of the pins (hereinafter referred to as "completely for each pin method"). k went to the test machine that used the resources together, but recently, many people started using the method for each (four). After pushing the material, there will be more people to fully handle each pin. Each of the test machines has a power supply for supplying power to the semiconductor integrated circuit, a DC measurement system for evaluating DC (Direct Current) characteristics of the input and output terminals of the semiconductor body circuit, and a DC level DAC for supplying the semiconductor integrated circuit (Digital) t〇Anal〇gC〇nverter) converters, etc. These test machines are controlled by a test program that operates under the operating system (the operating system). This test program is written in the so-called test machine language. In other words, different test machine architectures, supported by The test machine language is also different. In this test machine language, 'includes the language that was originally used to directly control the hardware and is called the combined language form of machine language. In contrast, in the highly fluent language, the test has been designed. The machine control language also uses the 卩(10)"... form and the BASIC form. In addition, the structured language pASCAL has also been widely used. 'Currently, the C language § is the main one. In addition, in recent years, in the field of semiconductor integrated circuits. In the middle, there are 6

2036-7232-PF ^^185 ,公司’其中包括了 :稱為無製造廠公司的電路設計公司、 提供工具以在工作站等的電腦上評估電路設計公司所設叶 .之半導體積體電路之功能且稱為EDA(工程/設計/自動又化。) =賣商的公司、將有關電路設—公司所設計之半導體積體 f路㈣試程式轉換成測試機可以執行之程式且稱為測試 I驗至的料公司、根據電路設計公司所設計之電路設計, 貧料製作光罩的光罩製作公司、使用所製作之光罩製进半 導體積體電路且稱為製造廢公司的晶圓製作公司、使用由 测試公司所製作之測試程式承包所有測試機之測試且稱為 測試製造礙的公司等,這些專業公司的半導體積體電路的 研發進展至分工化。 “因此,在專利文獻1中,揭示一種和商業模式有關的技 術’其内容為’首先’測試公司對腸販賣商、電路設計公 司及晶圓製作公司提出一種測試機語言,接著,舰販賣商 對電路設計公司提供半導體積體電路的邏輯模擬,之後, 電路設計公司設計出具有所要之邏輯功能的半導體積體電 路,進行虛擬測試機(在電腦上表現出測試機的功能)的邏 軏功能的檢測’然後,透過網際網路將由测試機語言所製 作=測試程式傳送至測試公司,然後,測試公司根據該測 式程式選擇可使用之測試機。 專利文獻1··特.開2003-270305(段落0033〜〇〇35、第6圖) 【發明内容】 【發明所欲解決的課題】 然而,在專利文獻1的技術中有一個問題,那就是,在 2036-7232-PF 7 1363185 選擇測試機之前,有許多專業公司卡 得複雜或沒效率,因此,費用也跟著變多了。導致作業變 因此,本發明馨於上述問題,提供 之選擇等的半導髀 系統。 體積體電路的研發有關且具有效率的支援 【用以解決課題的手段】 為=上料題,本發明之半導體龍電路研發支援 以測試機作半導體積體電路之測試的受託者所 路::裝置'上述測試之委託者所擁有且透過網際網 控制裝置連線之終端裳置,上述控制裝置具有記 d:部,上述記憶部儲存測試機語言程式和測試機 試機語言程式在上述測試令指定並轉換必 並且使用於上述測試機上,測試機資料包含上 機的動作用資料的容許範圍,其和複數個上述測試 且測試中含有必要的參數’當上述處理部按照來自 迷<端裝置的請求而從上述終端裝置接收半導體積體電 路的測試程式時’可解析該測試程式的任意一部分並提取 上迹測試機的動作用資料,藉由比較該動作两資料和儲存 二上达記憶部之上述測試機資料,提取出可使用 【發明效果】 藉由本發明,在包含測試機之選取等的半導體積體電 路的研發方面,可實現有效率的支援系統。 【實施方式】 下面參照圖面說明本發明之半導體積體電路研發支援 2036-7232-PF 8 1363185 系統。首先,參照第1圖至第7圖來說明半導體積體電路 發支援系統的構造。 _ 第1圖為半導體積體電路研發支援系統的整體構、& 圖。半導體積體電路研發支援系統1 000包括測試公司(測$ 的受託者)1、電路設計公司(測試的委託者)2及晶圓製;^ ^ 司3而構成,這些公司所具有的各裝置相互以網際網路等二 路4來連線。 此外,電路設計公司2和晶圓製作公司3在圖中分刿僅 以一個來顯示,但亦可分別為複數個。 另外,雖未特別加以圖示,光罩製作公司和測試冶具 公司亦可和半導體積體電路的研發產生關係。 當測試公司1將要進行半導體積體電路的測試時 控制裝置10,另外,雖未圖心在此具有複數種測試機。 其具有控制裝置10為個人電腦等電腦裴置,用來和外呷裝 置進行通信的通信部U、由鍵盤等構成以進行資料輸入的、 輸入部12、由顯示部所構成以進行資料輸出的輸出部13、 :來儲存各種資料的記憶部14、由咖(—trai 及ΤΓΓβ It)等構成以進行各種運算處理的處理部15 作為處理部15之運算區域的記憶體16。 計公司2專門進行半導體積體電路的電路設 且有:I、裝置2〇。終端裝置20為個人電腦等電腦裝置, 八有進仃畫面顯示的顯示部21。 的廄:圓!:公司3為製作以半導體積體電路為主之晶圓 .匕括個人電腦等電腦裝置的終端裝置30。 2036-7232-pp 1363185 第2圖為電路設計公司2的終端裝置2()中的顯示部⑽ :、面顯示例(請斜酌參照第1圖)。在此’說明了各個書面(視 自)的概要,有關細節,則如後所述。此外,在第2—圖中 各畫面(視窗)同時顯示’-般亦可分別顯示。另外,即使 在未特別圖示的情況下,终端步晉川沾、s θ如 制裝置丨◦。: 、裝置—象全部為控 二證視^為終端裝置2G的使用者在通信時用來輸入 遇證用使用者名稱和·密碼的晝面。. 互動式編輯器視窗2 3是一晝面,此晝面使用測試公司i 所指示的互動式編輯器(測試機語言程式:藉由測試機。 來製作測試程式的編輯器),輸入半導體積體電路的測;: 需要的參數(施加電壓等),製作測試程式。 互動式編輯器若使用c語言之函數來撰寫,對於c語言 所運作的測試,會預先在測試機上準備c語言的函數定i °,. ,此,可在不作轉換的情況下使用該測試機。互動式編輯 器儲存於控制裝置1 〇的記憶部1 4。 另外具有-特徵,那就是,由互動式編輯器所製作的 測試程式中撰寫了必要的錢,所卩,對於以c語言以外的 語言來運作的測試機,亦可輕易進行該程式的轉換。 於是’藉由使用此互動式編輯器,對於那些對測試機 的相關知識了解不多的人來說’也可寫出適宜的測試程式。 測試圖樣視窗2 4為電路設計公司輸入用來檢驗半導體 積體電路之測試圖樣的畫面。在此’所謂測試圖樣,係指 根據時間順序來表現用來檢驗半導體積體電路而進行模^ 2036-7232-PF 10 時的各參數的設定值。 通常,在電路設計公司2,製作出一種稱為時間驅動方 ,弋來製作測s式圖樣,該方式為,藉由監視邏輯變化 7提^變化點較時資訊,以撰寫測試圖樣。在設計階 奴%間驅動方式雖然有用,但是直接使用這種方式會有 測試機無法讀取的問題。因此,必須轉換成一種稱為速率 方式的方式,該方式為,藉由一定的週期來區隔測試圖樣, 將測試圖樣步驟當成1/()的資訊來撰寫測試圖樣,以此方式 _在測S式機上可被讀取。 因此,在本發明的結構上,電路設計公司2以時間驅動 方式輸入測試圖樣,在測試公司丨轉換為速率方式。藉此, 可在測試公司1結合速率方式的測試程式和後述之RTL形式 之半導體積體電路資料,在電腦上進行虛擬測試,透過網 際網路等將該結果傳送至電路設計公司2,實現與半導體積 體電路之研發有關且有效率的支援系統。 此外,在下面的記載中,以時間驅動方式來表現的測 _試程式僅稱為「圖樣」來作區分。 返回第2圖的說明,測試圖樣視窗24為電路設計公司2 輸入圖樣檔名、選擇圖樣形式(文字、VCD(Value Change2036-7232-PF ^^185, the company's include: a circuit design company called a non-manufacturing company, providing tools to evaluate the function of the semiconductor integrated circuit set by the circuit design company on a computer such as a workstation. Also known as EDA (Engineering/Design/Automatic Reconstruction.) = The company of the seller, the semiconductor integrated circuit designed by the circuit design company, the four-way test program is converted into a program that the test machine can execute and is called test I. The company that has been inspected, the circuit design designed by the circuit design company, the mask manufacturing company that produces the mask for the poor material, and the wafer fabrication company that manufactures the semiconductor integrated circuit using the fabricated photomask and is called the manufacturing waste company. The company's semiconductor integrated circuit development has progressed to division of labor, using a test program produced by a test company to contract tests of all test machines and called test manufacturing. "Therefore, in Patent Document 1, a technology related to a business model is disclosed," which is a 'first' test company that proposes a test machine language to an enteric vendor, a circuit design company, and a wafer fabrication company, and then, a ship dealer. The circuit design company provides logic simulation of the semiconductor integrated circuit. After that, the circuit design company designs the semiconductor integrated circuit with the desired logic function, and performs the logic function of the virtual test machine (showing the function of the test machine on the computer). 'Detection' then transmits the test program made by the test machine language to the test company via the Internet, and then the test company selects the test machine that can be used according to the test program. Patent Document 1····· 270305 (Paragraphs 0033 to 〇〇35, FIG. 6) [Disclosure] [Problems to be Solved by the Invention] However, there is a problem in the technique of Patent Document 1, that is, selection at 2036-7232-PF 7 1363185 Before the test machine, there were many professional companies that were complicated or inefficient, so the cost has become more and more. Invented in the above-mentioned problems, providing a semi-conducting system such as selection. The development of volumetric circuits is related to and efficient. [Means for Solving the Problem] For the problem of loading, the semiconductor dragon circuit development support of the present invention is The test machine is used as a trustee of the semiconductor integrated circuit test: the device is installed by the client of the above test and connected through the Internet control device, and the control device has a memory portion: the memory portion The storage test language program and the test machine test language program are specified and converted in the above test order and used on the above test machine. The test machine data includes the allowable range of the action data of the upper machine, and the plurality of the above tests and tests Including the necessary parameter 'When the processing unit receives the test program of the semiconductor integrated circuit from the terminal device in response to a request from the terminal device, 'can analyze any part of the test program and extract the action of the track tester Using the data, by comparing the two data of the action and storing the above test machine data of the memory department, [Effect of the Invention] According to the present invention, an efficient support system can be realized in the development of a semiconductor integrated circuit including selection of a test machine, etc. [Embodiment] Hereinafter, a semiconductor integrated body of the present invention will be described with reference to the drawings. Circuit development support 2036-7232-PF 8 1363185 system. First, the structure of the semiconductor integrated circuit support system will be described with reference to Fig. 1 to Fig. 7. _ Fig. 1 shows the overall structure of the semiconductor integrated circuit development support system. & Figure. The semiconductor integrated circuit development support system 1 000 includes the test company (trustee of the measurement $) 1, the circuit design company (the tester's commissioner) 2, and the wafer system; ^ ^ division 3, these companies Each of the devices is connected to each other by an Internet 2 or the like. In addition, the circuit design company 2 and the wafer fabrication company 3 are displayed in one figure in the figure, but may be plural. In addition, although not specifically shown, the mask manufacturing company and the testing and tooling company can also have a relationship with the development of semiconductor integrated circuits. When the test company 1 is to perform the test of the semiconductor integrated circuit, the control device 10 is additionally provided with a plurality of test machines. The control device 10 is a computer device such as a personal computer, a communication unit U for communicating with the external device, an input unit 12 configured by a keyboard or the like for inputting data, and a display unit for data output. The output unit 13 is a memory unit 14 that stores various types of data, and a processing unit 15 that performs various types of arithmetic processing, such as a coffee (-trai and ΤΓΓβ It), as the memory 16 of the calculation area of the processing unit 15. The company 2 specializes in the circuit design of the semiconductor integrated circuit: I, device 2〇. The terminal device 20 is a computer device such as a personal computer, and has a display unit 21 that displays a screen.廄: Round!: Company 3 is a wafer that mainly uses semiconductor integrated circuits. It includes terminal devices 30 for computer devices such as personal computers. 2036-7232-pp 1363185 Fig. 2 is a display unit (10) in the terminal device 2 () of the circuit design company 2: an example of a surface display (please refer to FIG. 1 as appropriate). Here is a summary of each written (see), details of which are described later. In addition, in the second picture, each picture (window) is displayed simultaneously and can be displayed separately. In addition, even if it is not specifically shown, the terminal steps into Jinchuan and s θ as the device. : The device-like image is controlled by the user. The user of the terminal device 2G is used to input the user name and password of the witness during communication. The interactive editor window 2 3 is a face-to-face, using the interactive editor indicated by the test company i (test machine language program: the tester to make the test program by the test machine), input the semiconductor product Measurement of the body circuit;: Required parameters (applying voltage, etc.) to make a test program. If the interactive editor uses the function of the C language to write, for the test run by the C language, the function of the c language is pre-set on the test machine, i °,. , and the test can be used without conversion. machine. The interactive editor is stored in the memory unit 14 of the control unit 1 . In addition, there is a feature that the necessary test money is written in the test program created by the interactive editor. It is also easy to convert the program to a test machine operating in a language other than C. So by using this interactive editor, you can write a suitable test program for those who don't know much about the test machine. The test pattern window 2 4 is a screen for the circuit design company to input a test pattern for inspecting the semiconductor integrated circuit. Here, the term "test pattern" refers to a set value of each parameter when the module 2036-7232-PF 10 is used to inspect the semiconductor integrated circuit in accordance with the chronological order. Usually, in circuit design company 2, a type of time-driven party is created to produce a s-pattern, which is to update the point-time information by monitoring the logic change 7 to write a test pattern. Although it is useful to drive the design between the slaves and the slaves, there is a problem that the tester cannot read them directly. Therefore, it must be converted into a method called rate mode, which is to divide the test pattern by a certain period and write the test pattern as 1/() information to write the test pattern. It can be read on the S type machine. Therefore, in the structure of the present invention, the circuit design company 2 inputs the test pattern in a time-driven manner and converts it into a rate mode in the test company. In this way, the test company 1 can combine the rate mode test program and the RTL form of the semiconductor integrated circuit data described later to perform a virtual test on the computer, and transmit the result to the circuit design company 2 via the Internet, etc. A research and development-related and efficient support system for semiconductor integrated circuits. Further, in the following description, the test program expressed in a time-driven manner is simply referred to as "pattern" for distinction. Returning to the description of Figure 2, the test pattern window 24 is the circuit design company 2 input pattern file name, select the pattern form (text, VCD (Value Change

Dump) 、 WGL(Wavef〇rm Generation Language)、 STIIXStandard Tester Interface Unguage))並以時間驅 動方式輸入圖樣的畫面。 潘1覽視窗25為一畫面,該畫面可根據從測試公司1的控 制裝置10傳送過來的資料,顯示可使用之測試機的一覽表 2036-7232-PF 11 1363185 和使用各測試機時的 ^ 叶的列4成本,稭由選取財務情況(進帳) 項目,找出將實際使用的測試機。 虛擬測試視窗26是-畫面,該畫面可在希望執行虛擬 測。式(在電腦裝置上所進行的虛擬測試)的情況下,在半導 體積體電路以RTL(Register “仏如·直接連接 型暫存器邏輯電路)形式來撰寫時,輸人該半導體積體電路 的RTL檔名並進行該指示^ ^ ^ ^ 1隹此以手導體積體電路為 Κ T L木式為月’j提’不過,芒蚀爾甘 个、右使用其他形式,亦可配合著進杆 畫面顯示。 設計視窗27是-晝面,該畫面可根據從測試公5Ίι的控 制裝置i。所接收的資料,顯示對半導體積體電路採用Dump), WGL (Wavef〇rm Generation Language), STIIXStandard Tester Interface Unguage)) and input the pattern of the pattern in a time-driven manner. The Pan 1 view window 25 is a screen which can display a list of usable test machines 2036-7232-PF 11 1363185 and the use of each test machine based on the data transmitted from the control device 10 of the test company 1. The cost of column 4, the straw is selected by the financial situation (accounting) project, to find out which test machine will actually be used. The virtual test window 26 is a screen that can be used to perform a virtual test. In the case of a virtual test performed on a computer device, when the semiconductor integrated circuit is written in the form of RTL (Register, for example, a direct-connected register logic circuit), the semiconductor integrated circuit is input. The RTL file name and the indication ^ ^ ^ ^ 1 隹 以 以 手 手 体积 体积 TL TL TL TL TL TL TL TL TL TL TL TL TL TL TL 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过 不过The screen display is displayed. The design window 27 is - the surface, which can be displayed on the semiconductor integrated circuit according to the received data from the test device i.

DmDesb fQr Testability:測試簡易化設計)時的 體積體電路的面積增加率,另外’若希望實現附 RTL檀名並進行指示。 ' 接送器視窗28為希望算出使用各接送服務之測 時進行該指示的畫面。 + 不良閑極映射視窗29為顯示晶圓上之不良 極的位置)的畫面。 第3圖為顯示每個測試機之規格範例的表格。此表格 存於記憶部U(參照第旧),顯示最高運作頻率 各個測試機的規格。 腳數4 第4圖顯示在各種方式的測試機中所具有的妓通美本 構造。如第4圖所示,測試機糊在構造上包括對 對象之半導體積體電路L供給電源電壓的電源單元3〇ι對 2036-7232-PF 12 1363185 半導體積體電路L之輪入诚羊势λ彳_^站 — 叛入编子輸入6谠的驅動器302、比較 從半導體積體電路丨 貝瓶电峪L之輸出鈿子輸出之信號和期望值 的比較器303、產生輸入至半導體積體 王干守體槓體電路L之測試資料及 期望值的圖樣產生器3〇4、產生輪 主 、 座生输入至丰導體積體電路輸入 之信號之施加時序的{時產生器3G5、以測試程式來進行各 控制的控制器3〇6、用來進行輸.出接腳之㈣層級檢測等直 流測试的DC測試電路3 〇 7。 此外,電源單元301及DC測試電路3〇7如同測試程式, 並不會因各個測試機不同而不同,另夕卜,作為測定半導體 積體電路的技術,變更是不需要的,可使料通的程式碼。 第5圖顯示測試機語言TL的範例。在此測試機語士几 中’有測試機程式碼5Q中的各測試機資源(包含測試機運作 時所需要之參數的測試機運作用資料)和各個c語言函數形 式5 2的支援》 夕 例如,當測試機資源為裝置電源51時,c語言函數形式 52為「VS{單元數,施加電壓,f壓範圍,測定電流範圍 上限鉗位電流,下限鉗位電流}」。在此情況下,若要使用 此測試機語言而輸入測試機之測試所使用的參數,可在其 顯示的位置輸入單元數(電源之單元編號)等各參數。若未 輸入,會輸入半導體積體電路研發支援系統丨〇〇〇所指定的 預设參數,所以,設計測試程式的工程師不需要了解全部 的參數。有關其他測試機資源的情況亦相同。 第6(a)圖為顯示RTL形式之.半導體積體電路的模式 圖。在半導體積體電路L的輸入端子68和輸出端子69之間, 2036-7232-PF 13 丄为3185 具有既>定邏輯運作功能的組合電路6?和複數個正反器 (FF)61交互以組合電路線65(圖示中為g,其他的線亦相 同)來連接,另外,所有的正反器61藉由時基線66來連接。 藉由此構造,可在半導體積體電路^,使正反器61 同步,以良好順序處理從輸入端子68輸入的信號,從輸出 端子6 9輸出信號.。 . 第(b)為模式圖,顯示當在第6(a)圖所示之半導體積 體電路L使用SCAN技術時,在各種正反器61中&加多工器 _ (MUX)的情況。 針對1個正反器61,附加丨個多工器62。多工器62由·^ 閘極621,622及OR閘極623這3個閘極所構成,測試電路線 63、SCAN電路線64及組合電路線65分別以圖中所示的方式 來連接。 藉此,在第6(a)圖的半導體積體電路,可連鎖各個 正反器61以使之聯合運作,可在測試時找出不良間極的位 置。此外,有關此SCAN技術的細節,記載於特開 攀號公報中。 第7圖顯示接送服務晶片的構造。放大圖⑽大晶圓 的-部分。在放大圖71中’晶片共有―(Α晶片等)、 5_1〇mm(U #、β晶片等)、—m(C晶片等)這3種。 近年來,半導體積體電路的製作發展至微細化加工, 使用高科技裝置、多層配線等,所以,價格變得非常高。 於是,在同-晶圓上製作複數種半導體積體電路的接送服 務能有效降低半導體積體電路的製作成本。 2036-7232-PF 14 1363185 以上完成了有關在半導體積體電路研發支援系統1000 之結構的說明,接著,將一邊參照第8圖至第14圖,一邊說 明其運作(請適度參照第1圖)。 第8圖為流程圖’顯示關於半導體積體電路研發支援系 統1 0 0 0的整體動作流程。 首先’測-試公司1使用控制裝置1 〇公開與測試半導體積 •體電•路之測試機有關的介紹的網頁(Hp)(步驟S8〇1)。 電路設計公司2使用終端裝置20製作與預先製作之乎 鲁導體積體電路有關的電路設計資料(步驟S8〇2),根據該電 路設計資料,委託晶圓製作公司3製作晶圓,換言之,將該 資訊傳送至晶圓製作公司3的終端裝置3〇(步驟S8〇3)。 接著,電路設計公司2可使用終端裝置2〇搜尋網際網路 以找出用來測試預先製作之半導體積體電路的測試機,進 入測試公司1的網站,透過網際網路來使用互動式編輯器。 另外,可從控制裝置10下載互動式編輯器(步驟s8〇4)。此 外,即使沒有互動式編輯器,也可下載其他測試程式製作 另外,當進入測試公司!的網站時,需要認證,電路言 計公司2藉由顯示部21的認證視窗22(參照第⑼),輸 用者名稱和密鳴’可使用本環境。藉此, 的安全層級。 示部=著一路設計公司2使用終端農置2〇 ’ -方面參照馬 •互動式編輯„。面藉由互動式編輯器視窗23(參照第2圖)使用 工’自’另-方面再根據電路設計資料,製作出用DmDesb fQr Testability: The area increase rate of the bulk circuit when testing the simplified design. In addition, if you want to implement the RTL name and indicate it. The pick-up window 28 is a screen for wishing to calculate the instruction to use each of the pick-up services. + The defective idle mapping window 29 is a screen for displaying the position of the defective pole on the wafer. Figure 3 is a table showing an example of the specifications of each test machine. This table is stored in the memory unit U (refer to the old one) and shows the specifications of each test machine at the highest operating frequency. Number of feet 4 Figure 4 shows the configuration of the 妓通美本 in various types of test machines. As shown in FIG. 4, the tester paste is configured to include a power supply unit for supplying a power supply voltage to the semiconductor integrated circuit L of the object. 〇ι pairs 2036-7232-PF 12 1363185 The semiconductor integrated circuit L is in turn. λ彳_^ station - a driver 302 that invades the editor input 6谠, a comparator 303 that compares the signal output from the output of the semiconductor integrated circuit, the output of the dice, and an expected value, and generates an input to the semiconductor integrated body. The test data generator of the body bar circuit L and the expected value of the pattern generator 3〇4, the timing generator 3G5 for generating the timing of the signal input from the wheel main body and the input to the rich volume body circuit, each of which is tested by a program The controller 3〇6 for controlling, the DC test circuit 3〇7 for performing DC test such as (4) level detection of the output pin. In addition, the power supply unit 301 and the DC test circuit 3〇7 are similar to the test program, and are not different for each test machine. In addition, as a technique for measuring the semiconductor integrated circuit, the change is unnecessary, and the material can be passed. The code. Figure 5 shows an example of the test machine language TL. In this test machine, there are several test machine resources in the test machine code 5Q (test machine operation data including parameters required for the test machine operation) and support for each c language function form 5 2 When the test machine resource is the device power supply 51, the c language function form 52 is "VS {unit number, applied voltage, f voltage range, measured current range upper limit clamp current, lower limit clamp current}". In this case, if you want to use the tester language and enter the parameters used for the test of the tester, you can enter the parameters such as the number of cells (unit number of the power supply) at the position where it is displayed. If it is not input, the preset parameters specified by the semiconductor integrated circuit development support system are input, so the engineer who designs the test program does not need to know all the parameters. The same is true for other tester resources. Figure 6(a) is a schematic diagram showing a semiconductor integrated circuit in the form of RTL. Between the input terminal 68 and the output terminal 69 of the semiconductor integrated circuit L, 2036-7232-PF 13 3 is 3185. The combination circuit 6 having both the logical operation function and the plurality of flip-flops (FF) 61 interacts. The combined circuit lines 65 (g in the figure, the other lines are also the same) are connected, and all the flip-flops 61 are connected by the time base 66. With this configuration, the flip-flop 61 can be synchronized in the semiconductor integrated circuit, the signal input from the input terminal 68 can be processed in a good order, and the signal can be output from the output terminal 69. (b) is a pattern diagram showing the case of adding & multiplexer _ (MUX) in various flip-flops 61 when the semiconductor integrated circuit L shown in Fig. 6(a) uses the SCAN technique. . One multiplexer 62 is added to one flip-flop 61. The multiplexer 62 is composed of three gates, a gate 621, 622 and an OR gate 623. The test circuit line 63, the SCAN circuit line 64, and the combined circuit line 65 are connected in the manner shown in the figure. Thereby, in the semiconductor integrated circuit of Fig. 6(a), the respective flip-flops 61 can be interlocked to operate in cooperation, and the position of the defective interpole can be found during the test. In addition, the details of this SCAN technology are described in the special open bulletin. Figure 7 shows the construction of the shuttle service wafer. Enlarge the (10) part of the large wafer. In the enlarged view of Fig. 71, there are three types of "wafer sharing" (Α wafer, etc.), 5_1 mm (U #, β wafer, etc.), and -m (C wafer, etc.). In recent years, the production of semiconductor integrated circuits has progressed to miniaturization, and high-tech devices and multilayer wiring have been used. Therefore, the price has become very high. Therefore, the shuttle service for fabricating a plurality of semiconductor integrated circuits on the same wafer can effectively reduce the manufacturing cost of the semiconductor integrated circuit. 2036-7232-PF 14 1363185 The above description of the structure of the semiconductor integrated circuit development support system 1000 is completed. Next, the operation will be described with reference to Figs. 8 to 14 (please refer to Fig. 1 as appropriate). . Fig. 8 is a flow chart showing the overall operation flow of the semiconductor integrated circuit development support system 100. First, the test-test company 1 uses the control device 1 to disclose a web page (Hp) of the introduction of the tester for testing the semiconductor power supply circuit (step S8〇1). The circuit design company 2 uses the terminal device 20 to fabricate circuit design data related to the pre-fabricated volume body circuit (step S8〇2), and according to the circuit design data, commissions the wafer fabrication company 3 to fabricate the wafer, in other words, This information is transmitted to the terminal device 3 of the wafer fabrication company 3 (step S8〇3). Next, Circuit Design Company 2 can use the terminal device 2 to search the Internet to find the test machine used to test the pre-made semiconductor integrated circuit, enter the test company 1 website, and use the interactive editor through the Internet. . In addition, an interactive editor can be downloaded from the control device 10 (step s8〇4). In addition, you can download other test programs even without an interactive editor. Also, when you enter the test company! In the case of the website, authentication is required, and the circuit company 2 can use the environment by the authentication window 22 (see (9)) of the display unit 21, the user name and the secret sound. By this, the security level. The display department = the design company 2 uses the terminal farm 2 〇 ' - the aspect refers to the horse · interactive editing „. The surface is edited by the interactive editor window 23 (refer to the second figure) using the work 'self' another aspect Circuit design information, production and use

2036-7232-PF 15 丄: 來撰寫半導體積體電路测試項目的測試程式,另外,以文 :T,STiL中任一形式來製作(步㈣⑹,將測 试程式和圖樣傳送至測試公司i的㈣ _。在步_6中,使用測試圖樣視窗爾昭 亦以圖樣檔案的形式從4個中作出選擇並傳送。 子二卜=互動式編輯器製作成的測試程式為根據電 ::十置技術來撰寫的敘述方式程式碼,具有可從程式法内 谷輕易辨識指令涵義的特徵。 在步驟S8°7中’在測試公司1.這裡,藉由控制裝置10 的處理部15,從來自電路設計公羯圖樣展產生速率方式 的測試圖樣。'然後’預先從儲存於控制裝置的硬體中提取 可以測试中所要求之速度或規格從這個測試圖樣產生定時 信號的硬體(以下稱定時資源)。再者,可在圖樣產生器的 使用硬體上提取測試時所需要的各圖樣的長度等。 在提取定時資源的情況下,處理部15藉由分析整㈣ 試圖樣或其中任意一部分(例如1〇%),提取出需要的資料。 在二近年來之RTL同步設計所產生的電路設計資料有關的 測4圖樣中’可藉由百分之幾到百分之幾十的分析來保證 必要資訊的取得’所以,可在該情況下進行-部分的分析。 j後,刀析整個測試圖樣花費很多時間和費用,但是 右可以進仃一部分的分析,對測試公司1來說,可實現相當 省力的效果。 其-人,測試公司1使用控制裝置ίο搜尋(提取)可使用之 測忒機,算出每個測試機的測試成本(步驟S808 :有關細2036-7232-PF 15 丄: To write the test program for the semiconductor integrated circuit test project, and to make it in any form of T:STiL (step (4) (6), transfer the test program and pattern to the test company i (4) _. In step _6, using the test pattern window Erzhao also selects and transmits from the four in the form of a pattern file. Sub-2 = interactive editor made the test program based on electricity:: ten The narrative mode code written by the technique has a feature that can easily recognize the meaning of the command from the valley of the program. In step S8°7, 'in the test company 1. Here, the processing unit 15 of the control device 10 comes from The circuit design shows the test pattern of the rate mode. ' Then' pre-extracts the hardware that is stored in the control device to extract the timing signal from the test pattern by the speed or specification required for the test (hereinafter referred to as Further, the length of each pattern required for the test can be extracted on the use hardware of the pattern generator, etc. In the case of extracting the timing resource, the processing unit 15 analyzes the whole (four) by trying Any part of it (for example, 1%) extracts the required data. In the second 4th of the RTL synchronization design, the circuit design data related to the 4 patterns can be used by a few percent to tens of percent. The analysis to ensure the necessary information is obtained 'so, in this case - part of the analysis. After j, it takes a lot of time and expense to analyze the entire test pattern, but the right can go into part of the analysis, to the test company 1 It can be said that it can achieve a very labor-saving effect. It-person, test company 1 uses the control device ίο to search (extract) the usable tester, and calculate the test cost of each test machine (step S808: relevant

2036-7232-PF 16 1363185 節’在之後的第ίο圖中會再加以敘述)。 此外’在步驟S808中’若使用DFT,可提取出更便宜的 測試機以降低測試成本,另外,亦可同時使用DFT算出曰片 的面積增加率,對電路設計公司2提出該資訊,有關=細 節’在之後的第11圖中會再加以敘述。 一 另外,在步驟S808冲,有關在使用接送服務時算出測 試成本的情況,在之後的第12圖及第13圖中會再加以敘述、。 接著,測試公司1使用控制裝置1〇將可使用之测試機及 各個測試機的測試成本傳送至電路設計公司2的終端妒置 2〇(步驟 S809)。 、 接收到之後,在電路設計公司2這裡,藉由終端裝置 20,在顯示部21顯示所接收之可使用之測試機及各測試成 本(參照第2圖的瀏覽視窗25)。 當電路設計公司2希望執行虛擬測試時,使用終端裝置 20透過顯示部21的虛擬測試視窗26來指定電路設計資料 UTL檔)名稱,在虛擬測試的項目進行勾選,傳送至控制裝 置1 〇,藉此,委託測試公司丨執行虛擬測試(步驟S8丨〇)。 測試公司1使用控制裝置10,根據測試程式、測試圖樣 及電路設計資料,#由處理部15,執行虛擬測試,並將該 結果傳送至電路設計公司2的終端裝置20(步驟811)。 藉由此虛擬測試的執#,可發現無法適用力各測試機 的測試圖樣、定時設定等,接收該結果之後,電路設計公 司2可適當修正測試程式、測試圖樣等。 此外,當對半導體積體電路的邏輯功能進行虛擬測試 2036-7232-PF 17 時,該檢測宜以速率方式來進行 係指與稱為時間驅動方式或時間事件方式==, 相對的圖樣產生方式。具體而言,時間生方式 邏輯變化點爽m 吟門事件方式猎由監控 雙化點來如取變化點的定時資訊和 樣,相對於此,速率方式以—定的 產生圖 樣步帮當作的資訊,以產生圖樣。來^圖樣,將圖 也可所使用的測試圖樣即使不是整個圖樣, 充为發現不良情況的範圍内,僅使用任意 刀(百刀之幾到百分之幾十),如此,可縮短各作聿的時門 以增加效率。 合忭菜的打間 其-人’電路設計公司2使用终 測試的測試機,藉由將之值…^ ^裝置2選擇使用於實際 傳运至測試公司1的控制f置丨^ 委託測試(步驟S812)。 t制裝置10, 然後,測試公司丨根據需, ^ 試機之動作用程式並提…::⑽%式轉換為測 S813)〇 挺供化具專的测試作業準備(步驟 另外’晶圓製作公司3接收來自電路 =委託(步驟S_,製作晶圓(步驟_,再= 圓又給測5式公司1 (步驟S81 5 )。 ,“ 然後,測試公司!可使用控制裝置1〇、測 式(根據需要轉換為動作用程 測5式程 Μ、▲且莖, 飞)測忒圖樣、電路設計資 料荨,執行晶圓的測試(步驟_)。從 將測試結果傳送至終端裝置20。 』裝置1〇’ 最後,在控制裝置10上,將測試機資源、剛試結果等2036-7232-PF 16 Section 1363185 will be described later in the figure ίο). In addition, 'in step S808', if DFT is used, a cheaper test machine can be extracted to reduce the test cost. In addition, the area increase rate of the cymbal can be calculated by using DFT at the same time, and the information is proposed to circuit design company 2, The details will be described later in Figure 11. Further, in the case of step S808, the case where the test cost is calculated when the shuttle service is used will be described later in Fig. 12 and Fig. 13. Next, the test company 1 transmits the test cost of the usable test machine and each test machine to the terminal device 2 of the circuit design company 2 using the control device 1 (step S809). After receiving the data, the circuit design company 2 displays the received tester and each test cost on the display unit 21 by the terminal device 20 (see the browsing window 25 of Fig. 2). When the circuit design company 2 wishes to perform the virtual test, the terminal device 20 is used to specify the circuit design data UTL file name through the virtual test window 26 of the display unit 21, and the virtual test item is checked and transmitted to the control device 1 Thereby, the test company is commissioned to perform the virtual test (step S8). The test company 1 uses the control device 10 to execute a virtual test by the processing unit 15 based on the test program, the test pattern, and the circuit design data, and transmits the result to the terminal device 20 of the circuit design company 2 (step 811). By means of the virtual test, it is found that the test pattern, timing setting, etc. of each test machine cannot be applied. After receiving the result, the circuit design company 2 can appropriately correct the test program, test pattern, and the like. In addition, when the logic function of the semiconductor integrated circuit is subjected to the virtual test 2036-7232-PF 17, the detection should be performed in a rate mode and the pattern generation method is called a time-driven mode or a time event mode ==. . Specifically, the time-based mode logic changes the point m 吟 事件 event mode hunting by monitoring the doubled point to take the timing information and sample of the change point, in contrast, the rate mode is determined by the pattern Information to produce a pattern. In the case of the pattern, the test pattern that can be used in the figure can be shortened even if it is not the entire pattern, and only a random knife (a few hundred to a few hundred percent) can be used. The time of the door increases efficiency. The combination of the leeks and the people's circuit design company 2 uses the final test machine, by the value of ... ^ ^ device 2 selected for the actual transfer to the test company 1 control f set 委托 ^ commission test ( Step S812). t device 10, then, the test company 丨 according to the need, ^ test machine action program and mention ...:: (10)% conversion to test S813) 〇 供 供 供 供 供 供 供 供 ( ( ( ( Production company 3 receives from the circuit = commission (step S_, making the wafer (step _, then = round again to test 5 company 1 (step S81 5).," Then, test company! can use the control device 1 〇, test The test (step _) is performed by transferring the test result to the terminal device 20 by converting the test pattern into 5 steps ▲, ▲ and stem, flying, and measuring the circuit design data. 』Device 1〇' Finally, on the control device 10, test machine resources, test results, etc.

2036-7232-PF 18 1363185 各種資訊登錄於記憶部14( 田一 +次 、歹鄉6818)。洌試公司1可藉由活 用迫些資訊,掌握盥測撸 一蛀次“ H式機之頻率、接腳數、圖樣長度、 定時負源等有關f % π ▲丄\ ^ Λ ,關電路故s十公司2所期望的方向,適當進行與 將來購入測試機等 .又貝冲里、低成本的測試機的研 發等。 第9圖為机程圖,顯示在第8圖的步驟^術中提取測試 機資源時的處理。 在控制裝置10中,處理部15輸入測試程式,亦即,將 籲從電路設計公司2的終端裝置2〇接收的測試程式(參照第2 圖的互動式編輯器視窗M)儲存至記憶部14(步驟S901)。 處理。卩15從測試程式的行數n = 〇開始(步驟sg〇2),以 n=n + l來搜尋第丨行(步驟S9〇3),以判斷該行是否為pin程式 碼亦即,該行疋否為與接腳數有關的程式(步驟§904)。 當該行為Pi η程式碼時(在步驟S9 04中為Yes),處理部 15從Pin程式碼中提取接腳數的值,儲存於記憶部14(步驟 S905)。 • 當該行不為Pin程式碼時(在步驟S904中為No),處理部 15前進至步驟S906,判斷該行是否為VS程式碼,亦即,該 行是否為與裝置電源有關的程式。 當該行為VS程式碼時(在步驟S906中為Yes),處理部15 從VS程式碼中提取電源數的值’儲存於記憶部1 & (步驟 S907)。 此外,在此,在測試機資源之117雖然僅提取並儲存接 腳數和電源數的值,但實際上同樣地亦提取並儲存其他測2036-7232-PF 18 1363185 Various information is registered in the memory unit 14 (Tianyi + Times, Xiangxiang 6818). The test company 1 can use the information to control the frequency of the H-type machine, the number of pins, the length of the pattern, the timing of the negative source, etc. related to f % π ▲ 丄 \ ^ Λ, off the circuit s ten companies in the desired direction, appropriate and future purchase of test machines, etc., and the development of low-cost test machines, etc. Figure 9 is a machine diagram, shown in the step of Figure 8 extraction In the control device 10, the processing unit 15 inputs a test program, that is, a test program that is received from the terminal device 2 of the circuit design company 2 (refer to the interactive editor window of FIG. 2). M) is stored in the memory unit 14 (step S901). Processing: 卩15 starts from the number of lines of the test program n = ( (step sg 〇 2), searches for the 丨 line with n = n + l (step S9 〇 3) To determine whether the line is a pin code, that is, whether the line is a program related to the number of pins (step § 904). When the behavior is Pi η code (Yes in step S9 04), processing The unit 15 extracts the value of the number of pins from the Pin code and stores it in the storage unit 14 (step S905). • When the line is not When the pin code is No (No in step S904), the processing unit 15 proceeds to step S906 to determine whether the line is a VS code, that is, whether the line is a program related to the power of the device. When the behavior is VS code When it is Yes (Yes in step S906), the processing unit 15 extracts the value of the number of power sources from the VS code in the memory unit 1 & (step S907). Further, here, only 117 of the test machine resources are extracted. And store the value of the number of pins and the number of power supplies, but in fact also extract and store other tests

2036-7232-PF 19 13631852036-7232-PF 19 1363185

試機資源的值。 處理部㈣斷該行是否為最後一行(步驟別⑻若不 :取後-行⑽),返回步霸⑽,反覆進行處理若為最 後一行(Yes),結束處理。 _ .如此,可在控制裝置丨〇上,從 資源m大提取出測試機 :1〇圖為流程圖’顯示在第8圖的步驟咖8中搜尋測試 機和算出測試成本時的處理。 在控制裝置10中,處理部15輸入測試機資源I,亦即, 將在第8圖的步驟漏中所提取的測試機資源和定時資源 儲存至記憶部14 (步驟S1 0 01)。 巧其-人,處理部15從記憶部14取得測試機資訊,亦即, 取得與測試機規格有關的資料(參照第3圖)(步驟議2)。 —處理部15從測試機的機種11=0開始(步驟sl〇〇3),以 ㈣+ 1來搜尋第以測試機的規格(步驟训⑷,判斷測試機 貝源的各值是否滿足該測試機的測試機種資源(測試機資 料)的條件(步驟S1 005)。 處理部15若在步驟51〇〇5中滿足條件(Yes),在輸出部 顯不該測試機(可使用之測試機)(步驟Sl〇〇6),若不滿足 條件(No) ’前進至步驟S1 007。 ,處理部丨5在步驟S1007中,判斷有關於測試機所有機種 =處理疋否結束,若未結束(N〇),返回步驟S1004,反覆進 仃處理,若結束(Yes),算出各個可使用之測試機的測試成 本並將之顯示於輸出部13(步驟S1008)。 2〇36*7232-pf 20 异出夠試成本時,首先求 時間可考慮撰寫測試程式之測·:的:類:試需要 不受所使用之測試機的種類求出, 出之測試需要時間乘上各測試機二/式成本可錯由對算 圓/秒等)來求出。_ 早料时用價⑽日 、J式機的早位時間你 部15從儲存於記,_ 格亦可由處理 購入價格、使科F1 w)之表格巾的測試機的 用4間、償還期限、人事費用等 如此,可搜尋(選取)可使用之測試 各個測試機的測試成本。 可异出 之satr圖為流程圖’顯示當算出使用在第6(b)圖中說明 ,術之DFT的晶片面積增加率時的處理。 百先’在控制裝置10中,處理部15輸入RTL,亦即,將 Μ式的電路設計資料儲存至記憶部14(步驟S1101)。 …其次,處理部15進行邏輯合成’亦即,將其轉換成RTL 形式的電路设計資料的邏輯電路(步驟S1I02)。 接著,處理部15在邏輯電路中,計數含有正反器的總 閑極數(步驟S1103),另外,計數其中的正反器數口(步驟 S1104) 〇 處理部15以m=3xn來計算在多工器(Μυχ)中所使用的閘 極數m(步驟sii〇5)。如在第6(b)圖中所說明,當使用scan 技術的DFT時’針對1個正反器附加3個多工器,所以,可以 m=3xn來求出閘極數^ 最後’處理部15以m/N . 100來算出DFT的晶片的面積增 加率(%)(步驟S1106)。 2036-7232-PF 21 1363185 λ 如此 增加率。 可在控制裝置ίο上算出 使用DFT時的晶片 的面積 當使用謂技術的DFT時,隨著晶片面積的择 加〜片製作的成本亦跟著增加,但可在測試中減少所: 要的接腳數或減少測試頻率,所以,可在第8圖的步驟而 可搜尋(提取)出便宜的測試機,作為可使用之測 藉此’可降低測試成本。 於是,電路設計公司2在第8圖的步驟別12中,即使在 鲁與DFT有關的知識方面不充分,也可根據晶片製作成本隨著 晶片面積增加而增加、.測試成本下降等各種有關成本的事 實,來適當判斷是否應該使用DFT。此外,若電路設計公司 2希望使用DFT,可透過顯示部21的設計視窗27(參照第2 圖),輸入DFT的使用,將之傳送至測試公司i的控制裝置1〇。 第12圖為流程圖,顯示利用接送服務時的算出測試成 本的處理。此外,以下所謂樣本數,係指以除法算出單位 面積所具有之作為對象之晶片的數目,例如,在第7圖中, ®將單位面積設定為25mm2,則A晶片、B晶片、c晶片及D晶片 的樣本數分別為4, 2, 1及2。 另外,當使用在第12圖中所進行的接送服務時算出測 試成本的處理與第8圖的步驟S8 08中的測試成本算出的處 理對應,在此情況下’從第8圖的電路設計公司2到晶圓製 作公司3的晶圓製作委託(步驟S803)在步驟S808以後的時 點進行。 首先,電路設計公司2使用終端裝置20選擇接送器(參 2036-7232-PF 22 13*63185The value of the test resource. The processing unit (4) breaks whether the line is the last line (step (8) if not: take the back line (10)), returns to step (10), and repeats the processing to the last line (Yes), and ends the processing. Thus, the test machine can be extracted from the resource m on the control device: 1 is a flowchart showing the process of searching for the test machine and calculating the test cost in the step 8 of Fig. 8. In the control device 10, the processing unit 15 inputs the test machine resource I, that is, the test machine resource and the timing resource extracted in the step leakage of Fig. 8 are stored in the storage unit 14 (step S1 01). The processing unit 15 acquires the test machine information from the storage unit 14, that is, acquires the data related to the test machine specifications (see FIG. 3) (step 2). - The processing unit 15 starts from the model 11 = 0 of the test machine (step s1 〇〇 3), searches for the specification of the first test machine with (4) + 1 (step training (4), and judges whether the values of the test machine source meet the test. The condition of the test machine resource (test machine data) of the machine (step S1 005). If the processing unit 15 satisfies the condition (Yes) in step 51〇〇5, the test machine is not displayed in the output unit (a test machine that can be used) (Step S1〇〇6), if the condition (No) is not satisfied, the process proceeds to step S1 007. In step S1007, the processing unit 丨5 determines whether or not all the test machines of the test machine are processed or not, and if not, (N) Then, the process returns to step S1004, and the processing is repeated. If the result is (Yes), the test cost of each tester that can be used is calculated and displayed on the output unit 13 (step S1008). 2〇36*7232-pf 20 When you have enough test costs, you can first consider the time to write the test program. ·: Class: The test needs to be determined by the type of test machine used. The test takes time to multiply the cost of each test machine. Can be determined by the calculation of the circle / second, etc.). _ Early use price (10) day, J type machine early time, your department 15 from the storage, _ grid can also be processed by the purchase price, make the F1 w) table towel test machine 4, repayment period As for personnel costs, etc., the test cost of each test machine that can be used can be searched (selected). The dissipative satr map is a flowchart ’ shows the processing when calculating the wafer area increase rate of the DFT described in Fig. 6(b). In the control device 10, the processing unit 15 inputs the RTL, that is, stores the circuit design data in the memory unit 14 (step S1101). Next, the processing unit 15 performs logical synthesis 'that is, a logic circuit that converts it into circuit design data in the form of RTL (step S1I02). Next, the processing unit 15 counts the total number of idle poles including the flip-flops in the logic circuit (step S1103), and counts the number of flip-flops therein (step S1104). The processing unit 15 calculates at m=3xn. The number of gates m used in the multiplexer (Μυχ) (step sii 〇 5). As described in the 6th (b) diagram, when using the DFT of the scan technique, '3 multiplexers are added to one flip-flop, so the number of gates can be obtained by m=3xn^ The last 'processing section The area increase rate (%) of the wafer of the DFT is calculated by m/N. 100 (step S1106). 2036-7232-PF 21 1363185 λ This increases the rate. The area of the wafer when the DFT is used can be calculated on the control device ίο. When the DFT using the technology is used, the cost of the wafer area is increased as the wafer area is increased, but the test can be reduced: Counting or reducing the test frequency, so you can search for (extract) a cheap test machine at the step of Figure 8, which can be used as a test to reduce the test cost. Therefore, in the step 12 of the eighth drawing, even if the knowledge related to the DFT is insufficient, the circuit design company 2 can increase the cost of the wafer as the wafer area increases, and the test cost decreases. The fact of the matter, to properly determine whether DFT should be used. Further, if the circuit design company 2 wishes to use the DFT, the design of the DFT can be input through the design window 27 (see FIG. 2) of the display unit 21, and transmitted to the control device 1 of the test company i. Fig. 12 is a flow chart showing the process of calculating the test cost when using the shuttle service. In addition, the number of samples referred to below refers to the number of wafers to be used for calculating the unit area by division. For example, in FIG. 7, when the unit area is set to 25 mm 2 , the A wafer, the B wafer, the c wafer, and The number of samples of the D wafer is 4, 2, 1, and 2, respectively. In addition, the process of calculating the test cost when using the pick-up service performed in FIG. 12 corresponds to the process of calculating the test cost in step S8 08 of FIG. 8, in which case the circuit design company from FIG. 8 2 The wafer creation request from the wafer fabrication company 3 (step S803) is performed at a time point after step S808. First, the circuit design company 2 selects the shuttle device using the terminal device 20 (refer to 2036-7232-PF 22 13*63185

A 將該資料傳送至測試公司〗的控 照第2圖的接送器視窗28), 制裝置10。 其次,在測試公司丨的控制裝置1〇中,處理部15將以來 自電路設計公司2之終端裝置2()之接送器(接送器s)作為胃 象之晶片的樣本數輸入至記憶部】4(步驟s〗2〇l)。 接著’處理部15參照與儲存於記憶部14之接送服務有 關的資料庫,對接送器S判斷該樣本數是否空出( S1202)。- • #在接送以上的樣本數未空出時(在步驟S1202中為 ㈣,處理部15結束處理’當在接送器5上的樣本數空出時 (在〆驟51 202中為Yes),將該晶月登錄於與記憶部η之接 送服務有關的資料庫(步驟sl2〇3)。藉由此登錄,減少 送器S的空出面積。 其次,處理部15算出整個接送器8的測試成本(步轉 SI 204··有關細節料第13圖中再加以敘述),將整個接送器 S的測試成本除以總樣本數,算出每單位面積的測試成本, 藉由將該值乘以特;t晶片的樣本數,#出該晶片的測試成 本(步驟S1 205)。 藉此 試成本》 可在控制裝置10上算出該晶片在接送器s上的測 第13圖為流程圖,顯示算出第12圖的步驟312〇4中的整 個接送機S的測試成本的處理。 首先,處理部15將接送器S的測試中所使用的測試機機 種數M設為U步驟S13G1)。 2036-7232-PF 23 Ι3ύ3185 其次’處理部15將接送產品編號亦即在接送器§上戶斤 製作的晶片的產品編號設為〇(步驟S1302),將總測試時間 設為0(步驟S1303)。 接者’處理部15以D-D+1對接送產品編號的值加ι(步驟 S1304),參照記憶部14,將使用於該接送產品編號之晶片 的測試機作為使用於接送機S的測試機,判斷其是否已經被 登錄(步驟S1 305 )。 若測試機已被登錄(在步驟S1305中為Yes),不需要變 •更在接送器S中所使用之測試機的機種數,另外,若測試機 尚未登錄(在步驟S1 305中為No),處理部15該測試機作為在 接送器s中所使用的測試機,將之登錄於記憶部14(步驟 S1 306) ’藉此,在接送器s中所使用的測試機的機種數增加 了 1個,所以,可以M=M+1來更新Μ的值(步驟S1307)。 其次,處理部15從記憶部14讀取每一個樣本數的測試 =間t (步驟S1308),以T=T + tx樣本數(在該測試機上測試的 晶片的樣本數)來更新T的值(步驟S1309)。 •接著’處理部15判斷該晶片是否為最後產品(步驟 s1310),若不是最後產品(N〇),返回步驟si3〇4,反覆進行 二 右為最後產品(Yes),以TD=交換作業時間(1次的測 試機交換所需要的作業時間)χ (i)算出總作業交換時間 亦即交換測試機時所需要的總時間(步驟si3H)。 々其後,處理部15W(T+TD)X單價(每單位時間的測試成 本)^出整個接送器S的測試成本(步驟S131 2)。此外,單價 因各測試機而異,可算出整個接送器㈣測試成本。 2036-7232-ET 24 13153185 傷 如此,可在控制裝置10上算出在該接送 送器s的測試成本β π堅個接 然後,同一接送器内的各晶片中 類Γ棬鍤勃)合i 便用的測式機的種 、(機種數)愈>、,整個接送器3的測試成本愈低 例’各晶片的測試&太t合 ! 及第於是,藉由第a圖 及弟13圖的處理,苦雷政冰士士八 試得較低測試成本的測 ’〆@—接送器内所使用之測試機的種類,測 试公司1的作業也會更有效率。 “ 另外,藉由在接送器上進行測試,可提高晶片 (良品的比率)。 .手 所J二為流程圖,顯示製作不良閉極映射時的處理。 的:置:::映射’是指顯示在晶圓上的不良處(不良開極 的牛mi。在此所製作的不良間極映射是作為第8圖 的步驟S817中的測試結果的—部分來使用。 控制裝置10的處理部】^於水 P 5輸入來自第8圓的步驟58】6所 貫方也之測目式的不良資鮮 ^,nlAr^mc 科(不良圖樣),亦即,將其儲存於記 隐部14(步驟S1401)。 朽右Ϊ次’處理部15參照與儲存於記憶部14之晶圓上之間 極有關的資料庫亦即閘 閘極資枓庫(步驟S1402),找出不良閘 極(步驟514〇3)»此外,+031> 在閘極貝料庫中,不良資料和不良 閘極相關聯β Π朴纟理。Μ4取得儲存於記憶部14的晶圓的電路設 计-貝料(步驟襲4),找出不良開極的位置(步驟si4〇5)。 然後,處理部15判斷其是否為最後的不良資料,亦即,A. The data is transmitted to the pick-up window 28) of the test company's control chart 2, and the device 10 is manufactured. Next, in the control device 1 of the test company, the processing unit 15 inputs the sample number of the wafer (the shuttle s) from the terminal device 2 () of the circuit design company 2 as the wafer of the stomach image to the memory unit] 4 (step s〗 2〇l). Next, the processing unit 15 refers to the database related to the pick-up service stored in the storage unit 14, and determines whether or not the number of samples is vacated by the pick-up device S (S1202). - • # When the number of samples above the pick-up is not vacated ((4) in step S1202, the processing unit 15 ends the processing 'When the number of samples on the pick-up 5 is vacated (Yes in step 51 202), The crystal moon is registered in the database related to the pick-up service of the memory unit η (step sl2〇3). By this registration, the vacant area of the feeder S is reduced. Next, the processing unit 15 calculates the test of the entire pick-up unit 8. Cost (step SI 204··detailed in detail in Figure 13), divide the test cost of the entire pick-up S by the total number of samples, calculate the test cost per unit area, by multiplying the value by The number of samples of the t chip, # test cost of the wafer (step S1 205). The test cost can be calculated on the control device 10, and the 13th picture of the wafer on the shuttle s is a flow chart showing the calculation The processing of the test cost of the entire shuttle S in step 312 4 of Fig. 12. First, the processing unit 15 sets the number M of test machines used in the test of the shuttle S to U step S13G1). 2036-7232-PF 23 Ι3ύ3185 Next, the processing unit 15 sets the product number of the wafer, which is the pick-up product number, that is, the wafer produced by the pick-up device, to 〇 (step S1302), and sets the total test time to 0 (step S1303). . The receiver' processing unit 15 adds ι to the value of the pick-up product number by D-D+1 (step S1304), and refers to the memory unit 14, and uses the test machine for the wafer of the pick-up product number as the test for the pick-up machine S. The machine determines whether it has been logged in (step S1 305). If the test machine has been registered (Yes in step S1305), it is not necessary to change the number of models of the test machine used in the shuttle S, and if the test machine has not been registered (No in step S1 305) The processing unit 15 registers the test machine as a test machine used in the shuttle s in the storage unit 14 (step S1 306). Thereby, the number of test machines used in the shuttle s is increased. One, so the value of Μ can be updated by M=M+1 (step S1307). Next, the processing unit 15 reads the test=inter-t of each sample number from the memory unit 14 (step S1308), and updates the T by the number of T=T + tx samples (the number of samples of the wafer tested on the test machine). Value (step S1309). • Then the 'processing unit 15 determines whether the wafer is the last product (step s1310), if not the last product (N〇), returns to step si3〇4, and repeats the second right to the last product (Yes), and exchanges the working time with TD= (Operation time required for one-time test machine exchange) χ (i) Calculate the total operation exchange time, that is, the total time required to exchange the test machine (step si3H). Thereafter, the processing unit 15W (T + TD) X unit price (test cost per unit time) outputs the test cost of the entire shuttle S (step S131 2). In addition, the unit price varies depending on each test machine, and the test cost of the entire shuttle (4) can be calculated. 2036-7232-ET 24 13153185 In this case, the test cost of the shuttle s can be calculated on the control device 10 and then the same type of bobbin in the same shuttle can be used. The type of measuring machine used, the more (the number of models), the lower the test cost of the entire pick-up 3, the test of each wafer is too close! And then, with the processing of Figure a and Brother 13, the test of the lower test cost of the eight people of the ridiculous ice blaze 〆 — — — — — — — — — — — — — — — — — — — — — — — — — — — — The homework will also be more efficient. “In addition, by testing on the shuttle, the wafer can be increased (the ratio of good products). The hand J2 is a flowchart showing the processing when making a poor closed-pole mapping. The :::: mapping' means The defective portion displayed on the wafer (the defective open pole map is used as the test result in step S817 of Fig. 8). The processing unit of the control device 10] ^In the water P 5, the step 58 from the 8th circle is input, and the 6th part of the method is also used to detect the deficiencies, nlAr^mc (bad pattern), that is, to store it in the hidden portion 14 ( Step S1401) The processing unit 15 refers to the database related to the pole stored on the wafer of the memory unit 14, that is, the gate gate library (step S1402), and finds the bad gate (step 514〇3)» In addition, +031> In the gate of the gate library, the bad data and the bad gate are related to β. The circuit design of the wafer stored in the memory unit 14 is obtained. 4), find the position of the poor opening (step si4 〇 5). Then, the processing unit 15 determines whether it is the last Bad information, that is,

2036-7232-PF 25 王部的不复資粗0 不良資料二 畢(步驟叫⑹,若不是最後的 的不卜〇返回步㈣401,反覆進行處理,若為最後 的不良資料(Yes),結束處理。 測試八I作出不良閘極映射,在第8圖的步驟SS1 7中, 士十八5 1的控制裝置10將該不良閘極映射傳送至電路設 =司2的終端裝期,在终端裝置2Q的顯示部㈣示該不 ^極映射(參照第2圖的不良間極映射視窗29)。此外,此 =極映射可在各個晶片上製作’另外,亦可在整個 0上製作。 - 然後’電路設計公司2可藉由觀察此不良閘極映射,推 及:找出是在晶片製作程序中的某處產生問題的原因(異 物此入等)’藉由將此原因等傳達至晶圓製作公司3,可力 圖有效的改善’提高半導體積體電路製作中的良率(良品的 比率)。 以上結束了實施型態的說明,然而,本發明的實施型 態不在此限制。 例如,本發明不僅可應用於LSI等半導體積體電路,亦 可應用於記憶體等使用半導體技術的各種裝置。 另外,在DFT方面,不僅可使用SCAN技術,亦可使用 BIST(Built~In Self-Test)等其他技術。 再者’當各裝置透過網際網路4收發各種資料時,不僅 可以使用密碼等,亦可藉由編碼技術等來提高安全層級。 此外’有關硬體、淥程圖等具體結構,在不脫離本發 明之主旨的範圍内,可作適當變更。 2036-7232-PF 26 13153185 m 【圖式簡單說明】 半導體積體電路研發支援系統的整體構造圖。 圖為電路設計公司2的終端裝置2〇中的顯示部㈣ 董'面顯不例。 第3圖為顯示每個測試機的規格範例的表袼。 第4圖顯示在各種方式的測試機中所具有。的共通基本 構造。 第5圖顯示测試機語言TL的範例。 第6圖的(a)為顯示RTL形式的半 亍导體積體電路的模式 圖° (b)為模式圓顯示當在()中 尸汀不之丰導體積體電路L· 使用SCAN技術時,.在各種正反4少 你分裡止又斋bi肀附加多工器(Μυχ)的情 況。 第7圖顯示接送服務晶片的構造。 第8圖為流程圖,顯示關於半導體積體 統1_的整體動作流程。 研發支棱糸 第9圖為流程圖,顯示提取測試機資源時的處理。 第10圖為流程圖,顯示在第8圖中搜尋測試機和算出測 試成本時的處理。 第11圖為流程圖,顯示當算出使用SCAN技術之DFT的晶 片面積增加率時的處理。 第12圖為流程圖,顯示利用接送服務時的算出測試成 本的處理。 第13圖為流程圖’顯示算出第12圖的步驟S1204中的整 個接送機S的測試成本的處理。 2036-7232-PF 27 1563185 第14圖為流程圖,顯示製作不良閘極映射時的處理 【主要元件符號說明】 1測試公司 2電路設計公司 3晶圓製作公司 4網路 1 0控制裝置 1‘4記憶部 15處理部 20終端裝置2036-7232-PF 25 The king's non-re-investment is coarse. 0 Bad information is second (step is called (6). If it is not the last one, return step (4) 401, repeat the processing, if it is the last bad information (Yes), end The test VIII I makes a bad gate mapping, and in step SS1 7 of FIG. 8, the control device 10 of the singapore 401 transmits the bad gate map to the terminal of the circuit setting = division 2, at the terminal The display unit (4) of the device 2Q indicates the non-polar mapping (refer to the defective inter-pole mapping window 29 of Fig. 2). Further, the =-polar mapping can be created on each wafer, or it can be created over the entire 0. Then, Circuit Design Company 2 can observe the bad gate mapping and find out: it is the cause of the problem somewhere in the wafer fabrication process (foreign matter, etc.), by transmitting this reason to the crystal. In the production company 3, it is possible to effectively improve 'increasing the yield (the ratio of the good products) in the production of the semiconductor integrated circuit. The above description of the embodiment is completed, however, the embodiment of the present invention is not limited thereto. The present invention can be applied not only to LSI, etc. The semiconductor integrated circuit can also be applied to various devices using semiconductor technology such as memory. In addition, in the DFT, not only SCAN technology but also other technologies such as BIST (Built~In Self-Test) can be used. When each device transmits and receives various data through the Internet 4, not only a password or the like, but also a coding level can be used to improve the security level. Further, the specific structure of the hardware, the process diagram, and the like can be omitted from the present invention. 2036-7232-PF 26 13153185 m [Simple diagram of the diagram] The overall structure of the semiconductor integrated circuit development support system. The picture shows the display in the terminal device 2〇 of the circuit design company 2. Part (4) Dong's appearance is not typical. Figure 3 is a table showing the specifications of each test machine. Figure 4 shows the common basic structure of the test machine in various ways. Figure 5 shows the test An example of the machine language TL. Fig. 6(a) is a schematic diagram showing a semi-derivative volume circuit in the form of RTL. (b) is a mode circle showing when the corpus is not in the () · Make In the case of SCAN technology, in the case of various positive and negative 4, you can add additional multiplexers (Μυχ). Figure 7 shows the structure of the shuttle service wafer. Figure 8 is a flow chart showing the semiconductor product. The overall operation flow of the system 1_. The development diagram is shown in the flowchart, showing the processing when extracting the test machine resources. Figure 10 is a flow chart showing the search engine in Figure 8 and calculating the test cost. Fig. 11 is a flowchart showing the processing when calculating the wafer area increase rate of the DFT using the SCAN technique. Fig. 12 is a flowchart showing the processing for calculating the test cost when the shuttle service is used. Fig. 13 is a flowchart showing the process of calculating the test cost of the entire boarding machine S in step S1204 of Fig. 12. 2036-7232-PF 27 1563185 Figure 14 is a flow chart showing the processing when creating a bad gate map. [Main component symbol description] 1 Test company 2 circuit design company 3 wafer fabrication company 4 network 1 0 control device 1' 4 memory unit 15 processing unit 20 terminal device

2036—7232-PF 282036—7232-PF 28

Claims (1)

1363185 +、申請專利範圍: 種半導體積體電路 控制裝置’以测試機 包括. 者所擁有;及 #積體電路之測試的受託 終端裝置,上述_ 舆上述控制裝置 連線, 5所擁有且透過網際網路 其特徵在於: ^述控制裝置具有記憶部和處理部· 中 數 迷記憶部儲存測試機語言裎 測試機語言程式在上述測試中試機資料,其 並且使用於上述測試機上,測二轉換必要的參 機的動作用資料的容許範圍,並和複=料包含上述測試 且測試中含有必要的參數; 複數個上述測試機有關 述處理。P按照來自上述終端農置过 試機語言程式傳送至上述終端裝置,當從:述蚊將上述測 收根據上述測試機語言程式所製作的^1置接 測試程式的任意一部分並提取 ㈣H ’解析該 M , L ± J 口式機的動作用資料, 藉由比較該動作用資料和儲存於上 資料,提取出可使用之測試機。U的上述測試機 / 2.如申請專利範圍第丄項之半導體積體電路研發支援 系統’其中’上述記憶部進-步儲存各個測試機的單位時 間使用價格’上述處理部進__步將上述測試程式的任意— 部分轉換成上述測試機的動作用程式,根據該動作用^式 來推測上述每個可使用之測試機的測試需要時間,根據上 2036-7232-PF 29 13*63185 s =試需要時間和儲存於上述記憶部之上述使用單價算出 =母個可使用之測試機的測試成本,將上述可使用之測 试機及每個測試機的測試成本傳送 /手夕:申請專利範圍第2項之半導體積體::;發支援 & ψ其中,上述记憶部進一步館存運算式,該運算式可 在上述半導體積體電路中使用簡易化料時的上述 積體電路的增加面積,上述處理部根據儲存於上述 Π精上述運算式算出當使用上述測試簡易化設計時的 迷增加面積,使用上述測試簡易化設計時的上述可使用 之;則έ式機及其測試成太诉脾μ、+,说1 裝置。 W本並將上述增%面積傳送至上述終端 4. 如巾請專利範圍第3項之半導體龍電路研發支援 '、統’其令’上述運算式為對上述半導體積體電路中之 正反器每次附加3個閑極來算出上述增加面積的式子。 5. 如申請專利範圍第3項之半導體積體電路研發支援 这半中’上述處理部進—步從在上述測試機上測試上 路時所產生的資料找出不良閘極的位置, “.4不該不良間極之位置的不良閘極映射,將 極映射傳送至上述終端裝置。 ^ 6. —種半導體積體電路研發支援系統,包括. 控制裝置,以測試機作半導體積體電路之測試 者所擁有;及 w又託 終《置,上述測試之委託麵㈣且透 與上述控制裝置連線; 際,周路 2036-7232-PF 30 路,5固阳圓上製作複數個種類的半導體積體電 其特徵在於: 上述控制裝置具有記憶部和處理部; 上述記憶部儲存測試機纽古 式,其中,洌 σ尨式、測試機資料和運算 的參數,並 过川4争才曰定並轉換必要 使用於上述測試機上, 一 測試機.的動作用次杻& ^ ,、忒機貧料包含上述 >有關且測試中含有 〃、和複數個上述测試機 機的單位時間使 ^運^式根據上述各測試' 個使用測貝。在上述複數個各晶圓上涵蓋複數 半導 =2作業時間的測試需要時間來算出上述 等體積體電路的測試成本; 上述處理部按照來自上 試機語十π + # 扎,、知裝置的凊求,將上述測 頓-4切送至上述終端裝 收根據上述測試施纽丄扣』 田從上述終端裝置接 測試程式的任音一 収長式時,解析該 藉由比較# ^ a a忒機的動作用資料, …匕較該動作用資料和儲存於 資料,提取屮π # « 己隐邛的上述測試機 上述運糞々曾, f课上述儲存於記憶部之 上述各晶圓上==;電路的測試成本,將可在 裝置。 用之K機及其測試成本傳送至上述終端 2036-7232-pp 311363185 +, the scope of application for patents: a semiconductor integrated circuit control device 'to be tested by the machine; and the # terminal circuit test of the trusted terminal device, the above _ 舆 the above control device connection, 5 owned and Through the Internet, it is characterized in that: the control device has a memory unit and a processing unit, a middle memory system, a test machine language, a test machine language program, and a test machine data in the above test, which is used on the above test machine. The second test converts the permissible range of the necessary action data of the reference machine, and the test includes the above test and the test contains the necessary parameters; a plurality of the above test machines are related to the processing. P is transmitted to the terminal device in accordance with the language program from the terminal farm test machine, and the test is performed by the mosquito tester, and any part of the test program is extracted according to the test machine language program and extracted (4) H 'analysis The action data of the M, L ± J mouth machine is used to extract the usable test machine by comparing the action data and storing the data. The above-mentioned test machine of U / 2. The semiconductor integrated circuit research and development support system of the scope of the patent application of the second aspect of the invention, wherein the above memory unit further stores the unit time use price of each test machine. Any part of the above test program is converted into an action program of the above test machine, and according to the action, it is estimated that the test time of each of the test machines that can be used requires time, according to the above 2036-7232-PF 29 13*63185 s = test time and the above-mentioned unit price calculated in the above-mentioned memory unit = the test cost of the test machine that can be used by the parent, and the test cost of the above-mentioned test machine and each test machine can be transmitted. The semiconductor product of the second item of the second aspect of the invention, wherein: the memory unit further stores an arithmetic expression in which the integrated circuit is used when the simple material is used in the semiconductor integrated circuit. When the area is increased, the processing unit calculates the increased area when the test is simplified by using the above-described calculation formula stored in the above-described test, and the test is simple. Above may be used in the design; and testing the machine to the formula έ too v spleen μ, +, said device 1. W and transfer the above-mentioned increase % area to the above-mentioned terminal 4. As for the semiconductor dragon circuit research and development support of the third item of the patent scope, the above-mentioned arithmetic expression is the flip-flop in the above semiconductor integrated circuit. Each time three idle poles are added to calculate the above formula for increasing the area. 5. In the middle of the semiconductor integrated circuit research and development support in the third paragraph of the patent application, the above-mentioned processing unit proceeds to find the position of the bad gate from the data generated when testing the above-mentioned test machine, ".4 The poor gate mapping of the position of the defective pole is transmitted to the terminal device. ^ 6. A semiconductor integrated circuit development support system, including a control device, and a test machine for testing the semiconductor integrated circuit And the w and the end of the "set, the above-mentioned test of the commission (4) and through the control device;; Intersection, Zhou Road 2036-7232-PF 30, 5 Guyang round on the production of a variety of semiconductors The integrated electric machine is characterized in that: the control device has a memory unit and a processing unit; the memory unit stores a test machine, and the parameters of the test machine data and the calculation parameters are determined by The conversion must be used on the above test machine, the action of a test machine. The operation of the test machine is used to reduce the amount of the above-mentioned test, and the test contains 〃, and a plurality of the above test machines. The unit time of the machine is such that the test is performed according to each of the above tests. It takes time to test the plurality of semiconductors on the plurality of wafers, and the test cost of the above-mentioned equal volume circuit is calculated; The processing unit performs the test from the test device according to the request from the test machine, and the device is sent to the terminal for charging according to the test. When the program's tone is long, the data is analyzed by comparing the action information of the #^aa machine, ... compared to the action data and stored in the data, extracting the above test machine 屮π # « In the above-mentioned wafers stored in the memory section, the test cost of the circuit will be available in the device. The K machine and its test cost are transmitted to the above terminal 2036-7232-pp 31
TW094122698A 2005-07-05 2005-07-05 Semiconductor integrated circuit development support system TW200702692A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW094122698A TW200702692A (en) 2005-07-05 2005-07-05 Semiconductor integrated circuit development support system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW094122698A TW200702692A (en) 2005-07-05 2005-07-05 Semiconductor integrated circuit development support system

Publications (2)

Publication Number Publication Date
TW200702692A TW200702692A (en) 2007-01-16
TWI363185B true TWI363185B (en) 2012-05-01

Family

ID=46750616

Family Applications (1)

Application Number Title Priority Date Filing Date
TW094122698A TW200702692A (en) 2005-07-05 2005-07-05 Semiconductor integrated circuit development support system

Country Status (1)

Country Link
TW (1) TW200702692A (en)

Also Published As

Publication number Publication date
TW200702692A (en) 2007-01-16

Similar Documents

Publication Publication Date Title
Marinissen et al. On IEEE P1500’s standard for embedded core test
JP4074738B2 (en) Event-based semiconductor test system and LSI device design test system
US6961871B2 (en) Method, system and program product for testing and/or diagnosing circuits using embedded test controller access data
US6823497B2 (en) Method and user interface for debugging an electronic system
TWI221614B (en) Tester architecture construction data generating method, tester architecture constructing method and test circuit
TW201443463A (en) Test IP-based A.T.E. instrument architecture
WO2008064122A2 (en) Transaction co-validation across abstraction layers
WO2002001424A2 (en) System and method relating to verification of integrated circuit design
CN112444731B (en) Chip testing method and device, processor chip and server
TW200919246A (en) Enhancing speed of simulation of an IC design while testing scan circuitry
Zhang et al. Software-based self-testing of processors using expanded instructions
US20080120082A1 (en) Transaction Co-Validation Across Abstraction Layers
US20200184135A1 (en) Double glitch capture mode power integrity analysis
Lingappan et al. Satisfiability-based automatic test program generation and design for testability for microprocessors
US7409650B2 (en) Low power consumption designing method of semiconductor integrated circuit
TWI363257B (en)
TWI363185B (en)
Dey et al. Design for testability techniques at the behavioral and register-transfer levels
Das et al. Implementation of a testing environment for digital IP cores
Tang et al. Research and implementation of an automatic simulation tool
Das et al. System-on-chips design using ISCAS benchmark circuits-an approach to fault injection and simulation based on Verilog HDL
Choi et al. A debug scheme to improve the error identification in post-silicon validation
Portolan et al. A common language framework for next-generation embedded testing
JPH07319950A (en) Test system
JP2715928B2 (en) Probe information generator

Legal Events

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